Key advances in Medicine

发布于:2021-07-22 00:53:23

January 2012


Access high quality clinical content written by international experts

January 2012 volume 9 no. 1

January 2012 volume 8 no. 1

January 2012 volume 9 no. 1



How best to train postgraduates and residents in robotic techniques?

Molecular biomarkers in bladder cancer
Stratifying response to targeted therapies

January 2012 volume 8 no. 1

nrurol_OFC_JAN12.indd 1

13/12/2011 10:31

Mucosal pathogenesis and new concepts in the role of mucosal inflammation

January 2012 volume 8 no. 1


Clinical applications of MRI in Crohn’s disease
Evaluating the extent of intestinal damage


Influence on cardiovascular risk parameters in RA

Systemic sclerosis

Cellular and molecular mechanisms of fibrosis

nrgastro_OFC_JAN12.indd 1

07/12/2011 16:34

nrrheum_OFC_JAN12.indd 1

08/12/2011 11:47

January 2012 volume 9 no. 1

Neuroendocrine neoplasms of the gut and pancreas
Diagnostic and treatment options
28/11/2011 15:31

January 2012 volume 8 no. 1


An official publication of

Relevance and potential applications in kidney disease

Metabolic regulation of puberty and fertility

Acute kidney injury in multiple myeloma

02/12/2011 12:49

Pathogenesis and diagnosis

nrendo_OFC_JAN12.indd 1

nrneph_OFC_JAN12.indd 1

January 2012 volume 9 no. 1


Using breast cancer gene signatures in the clinic
Knowledge, advances and controversies

Time for a rethink to avoid overtreatment of older men and those at low risk

New recommendations for MRI-based monitoring of immunomodulation

Postural tachycardia syndrome
Characteristics, neuroepidemiology and pathophysiological mechanisms

nrclinonc_OFC_JAN12.indd 1

07/12/2011 15:55

nrneurol_OFC_JAN12.indd 1

13/12/2011 11:42

An official publication of

Role of the autonomic nervous system

Transcatheter aortic valve implantation
Valves and approaches

nrcardio_OFC_JAN12.indd 1

24/11/2011 15:32

? Junior doctors keep up-to-date with Research Highlights ? Busy clinicians appreciate “Key point” summaries ? Medical Researchers value in-depth Reviews

The January issues are freely available on our website:

Follow us.
Download the free app for your phone at
Get the free mobile app for your phone

http:/ /

design: annette fenner & nicola hawes

Key Advances in Medicine
he articles included in Nature Reviews Key Advances in Medicine were originally published in the February 2012 issues of the eight clinical Nature Reviews journals. The journals’ editors commissioned international experts to write a short essay highlighting up to five key papers that made the biggest contribution to their field in 2011. Between them, the eight clinical Nature Reviews journals published 43 articles, which are collated in this eBook; if you choose to cite an article, please use the original journal citation rather than citing the eBook. We hope you enjoy reading Nature Reviews Key Advances in Medicine. If you would like to find out more about the Nature Reviews series, please visit:


Cover image: Hippocrates of Cos reflected in the style of the Roman god, Janus. Based on an engraving by Peter Paul Rubens (1638) from website.

Don’t miss out on next year’s Key Advances in Medicine
Sign up to free e-ALERTS and be notified when next year’s Key Advances in Medicine eBook is published. You won’t miss out on Nature Reviews collections and Focus issues either! Nature Reviews Cardiology
S1 acute coronary syndromes | Walking the tightrope between efficacy and bleeding Payal Kohli and Christopher P. Cannon S3 atrial fibrillation | Stroke prevention in AF Gregory Y. H. Lip S5 heart failure | Heart failure therapy—technology to the fore John J. V. McMurray S7 hypertension | New insights—from risk factors to treatment implications George L. Bakris S9 valvular disease | Breakthrough for intervention? Volkmar Falk

COPYRIGHT ? 2012 Nature Publishing Group.
No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form (electronic or otherwise) without prior permission from the copyright holder.

for internal or personal use, or internal or personal use of specific clients, is granted by Nature Publishing Group to libraries and others registered with the Copyright Clearance Center (CCC) Transactional Reporting Service, provided the relevant copyright fee is paid to CCC, 222 Rosewood Drive, Danvers, MA 01923, USA.

DISCLAIMER: Although every effort is made by the publishers to see that no inaccurate or misleading data, opinions or statements appear in this collection, they wish to make it clear that the data and opinions appearing in articles and advertisements herein are the responsibility of the contributor or advertiser concerned. The journal does include the personal opinions of the authors; therefore, it is not intended to be relied on solely as a guide to good practice or safe treatment. Accordingly, the publishers, employees, offices and agents accept no liability whatsoever for the consequences of any such inaccurate or misleading data, opinion or statement. Although every effort is made to ensure that drug doses and other quantities are presented accurately, readers are advised that the new methods and techniques involving drug usage and described within this journal should only be followed in conjunction with the drug manufacturer’s own published literature.

Nature Reviews Clinical Oncology
S11 ovarian cancer | Mutations and non-inferiority analyses show a way forward Maurie Markman S12 prostate cancer | Hitting old targets better and identifying new targets Yu Chen and Howard I. Scher S14 hematological cancer | New therapeutic targets and treatment strategies Paula Cramer and Michael Hallek S16 melanoma | A new paradigm tumor for drug development Alexander M. M. Eggermont and Caroline Robert S18 bone cancer | Prevention and treatment of bone metastases Robert E. Coleman

Nature Reviews Endocrinology
S21 thyroid disease in pregnancy | Thyroid function—effects on mother and baby unraveled Anthony P. Weetman S22 primary aldosteronism | Towards a better understanding of causation and consequences Michael Stowasser S24 polycystic ovary syndrome | Genes, aging and sleep apnea in polycystic ovary syndrome Andrea Dunaif S26 epigenetics and metabolism | Epigenetics, the life-course and metabolic disease Peter D. Gluckman S28 osteoporosis | Osteoporosis therapy—dawn of the postbisphosphonate era Roland Baron S30 type 1 diabetes mellitus | Heterogeneity of T1DM raises questions for therapy Paolo Pozzilli

Nature Reviews Neurology
S55 stroke | Major advances across the spectrum of stroke care Lee H. Schwamm S57 movement disorders | Translating new research findings into clinical practice Christine Klein and Dimitri Krainc S58 multiple sclerosis | Advances in therapy, imaging and risk factors in MS Bianca Weinstock-Guttman and Murali Ramanathan S60 dementia | Microbleeds in dementia—singing a different ARIA Philip Scheltens and Jeroen D. C. Goos S62 epilepsy | Insights into epilepsy treatments and biomarkers Fernando Cendes

Nature Reviews Gastroenterology & Hepatology
S33 hepatitis c | A new standard of care and the race towards IFN-free therapy Wolf Peter Hofmann and Stefan Zeuzem S35 hepatocellular carcinoma | Genomics in hepatocellular carcinoma—a big step forward Ryosuke Tateishi and Masao Omata S36 ibd | Advances in IBD management—towards a tailored approach Guillame P. Pineton de Chambrun and William J. Sandborn S38 the gut microbiota | Translating the microbiota to medicine Fergus Shanahan S40 neurogastroenterology | Emerging concepts in neurogastroenterology and motility Keith A. Sharkey and Gary M. Mawe

Nature Reviews Rheumatology
S65 rheumatoid arthritis | Advances in diagnosis, treatment and definition of remission Gerd R. Burmester S67 juvenile idiopathic arthritis | New takes on categorization and treatment Alberto Martini S68 systemic lupus erythematosus | Deciphering the role of NETs and networks in SLE Thomas D?rner S70 osteoarthritis | Age-related OA—a concept emerging from infancy? Thomas Aigner and Wiltrud Richter S72 systemic sclerosis | From mechanisms to medicines Luc Mouthon S74 vasculitis | The renaissance of granulomatous inflammation in AAV Stephan D. Gadola and Wolfgang L. Gross

Nature Reviews Nephrology
S43 glomerular disease | New clues to environmental influences in glomerular disease Peter J. Nelson and Charles E. Alpers S44 polycystic kidney disease | Connecting the dots toward a polycystic kidney disease therapy Vicente E. Torres and Peter C. Harris S46 acute kidney injury | Biomarkers are transforming our understanding of AKI Lakhmir S. Chawla and John A. Kellum S48 nondialysis chronic kidney disease | Progression, prediction, populations and possibilities Adeera Levin S50 dialysis | Can cardiovascular risk in dialysis patients be decreased? Peter Stenvinkel and Peter Bárány S52 transplantation | New agents, new ideas and new hope Titte R. Srinivas and Bruce Kaplan

Nature Reviews Urology
S77 prostate cancer | Redefining the therapeutic landscape for CRPC Carmel Pezaro and Gerhardt Attard S79 bladder cancer | The dawn of personalized medicine Thomas W. Flaig and Dan Theodorescu S80 sexual dysfunction | Advances in epidemiology, pathophysiology and treatment Eric Chung and Gerald B. Brock S82 male factor infertility | Semen quality, sperm selection and hematospermia Amichai Kilchevsky and Stanton Honig S84 kidney cancer | Objectifying risk for localized renal masses Marc C. Smaldone and Robert G. Uzzo


Walking the tightrope between efficacy and bleeding
Payal Kohli and Christopher P .?Cannon

Major advances in the diagnosis of acute coronary syndromes (ACS) have occurred in 2011, but physicians treating ACS still walk the tightrope between efficacy and bleeding. Key publications have shed light on this delicate balance and heralded a new era of novel oral anticoagulants for the treatment of ACS.
Kohli, P . & Cannon, C.?P . Nat. Rev. Cardiol. 9, 69–71 (2012); published online 20 December 2011; doi:10.1038/nrcardio.2011.206

The year 2011 has witnessed an evolution in all aspects of the management of acute coronary syndromes (ACS). The intro? duction of new-generation troponin assays has compli? cated the diagnostic dilemma in ACS by increasing the number false positive diagnoses. One of the most-interesting studies of 2011 addressed whether the reduced threshold for detection of myocardial injury translated into improved clinical outcomes. Mills et?al. undertook a prospective study of >1,000 patients who presented with possible ACS before and after the introduction of a lower diagnostic threshold (from 0.20?ng/ml to 0.05?ng/ml) for myocardial infarction (MI) using a high-sensitivity troponin assay with a low coefficient of variability (<10%).1 The study demonstrated that the lower threshold resulted in a 29% increase in detection of MI. Among patients with small increases in troponin (0.05–0.19?ng/ml), this diagnostic reclassification was also associated with improved risk stratification and better use of evidence-based therapies. Most importantly, the rate of death or recurrent MI was 52% lower (P?=?0.01) during the period in which the lower troponin thres? hold was used.1 Therefore, we can surmise that changing the diagnostic threshold for myocardial injury has identified a new, previously misclassified high-risk patient population that was and has led to improvements in clinical outcomes. Individualized medicine and tailored treatments have continued to be a dominant theme of this decade. The ELEVATETIMI?562 and GRAVITAS3 studies tested the effects of high-dose clopidogrel on platelet reactivity and clinical outcomes, respectively. In ELEVATE-TIMI?56,2 hetero?zygote carriers of the CYP2C19*2 allele needed 225?mg of clopidogrel to achieve the same

degree of platelet inhibition as noncarriers receiving 75?mg, suggesting that genotype could be important for choosing not only the type of antiplatelet agent, but also the dose. Homozygote carriers of CYP2C19*2 in ELEVATE-TIMI?56 were highly resistant to clopidogrel and maintained high degrees of platelet reactivity, even with 300?mg of the drug. Whereas ELEVATE-TIMI?56 confirmed the importance of pharmaco? genomic interactions in platelet reactivity, the GRAVITAS trial3 bridged the gap between laboratory findings and clinical outcomes and answered the question of whether increasing doses of clopidogrel would improve clinical outcomes. The investi? gators randomly assigned 2,214?patients with high on-? treatment platelet reactivity to high-dose (600?mg loading, 150?mg maintenance) or standard-dose (75?mg maintenance alone) clopidogrel.
Key advances
■■ Use of a sensitive troponin assay with a lower threshold for diagnosis of MI reduced death and recurrent MI in high-risk patients1 ■■ Increased doses of clopidogrel are necessary to achieve platelet inhibition in CYP2C19*2 heterozygotes 2 ■■ Apixaban 5?mg twice daily increases bleeding without decreasing the risk of cardiovascular death, MI, and stroke in patients with ACS5 ■■ Adding very low dose rivaroxaban to dual antiplatelet therapy reduces cardiovascular death, MI, and stroke with no increase in fatal hemorrhage, despite a dosedependent increase in major bleeding6 ■■ A bleeding risk score derived and validated on the basis of baseline characteristics demonstrated effective risk stratification for major bleeding in patients with ACS8

Despite a 22% absolute reduction in high on-treatment platelet reactivity at 30?days and 6?months with the high-dose strategy ( P ?<0.001), no difference was observed between the two groups in the incidence of cardiovascular death, MI, and stent thrombo? sis or in the rate of severe or moderate bleeding at 6?months (although the overall number of events in the trial was small).3 These results emphasize that, despite variation in drug metabolism and activity, no clinical impact of altering the dosing regimen of clopidogrel has yet been demon?strated. Furthermore, the reduction in platelet reactivity with second generation thienopyridines appears marginal as the dose is increased, and the effects in homozygote carriers of CYP2C19*2 are severely limited. With the introduction of more-potent agents, including prasugrel and ticagrelor, many of these pitfalls can potentially be avoided. With the emergence of novel oral anti? coagulants that require minimal monitoring, a new paradigm for treating ACS was ushered in—combining a low dose anticoagulant with dual antiplatelet therapy. With novel direct factor?Xa inhibitors, such as apixaban, rivaroxaban, and darexaban, straightforward outpatient anticoagulation finally became a realistic option for patients with ACS. However, the road was not without its stumbling blocks. In a phase?II trial of darexa? b an, increases in bleeding were demon?strated without an improvement in efficacy.4 The phase?III APPRAISE?2 trial5 of 7,392 patients with ACS was stopped early after an increase in major bleeding (HR?2.59, P?=?0.001) emerged for apixaban (5?mg twice daily) without an improvement in efficacy.5 Against these headwinds, the results of the ATLAS ACS 2?TIMI?51 trial6 were
JANUARY 2012? |? S1

15 – APPRAISE–2 ATLAS ACS 2-TIMI 56 Safety P <0.001 TRACER Safety P <0.001

10 – Events* per 1,000 patients

Safety P = 0.001


P = NS

P <0.001 P = 0.03 P = 0.04 P = NS


–5 –

P = NS Ef cacy P = 0.002

–10 – P = 0.02 Ef cacy CVD, MI, stroke CVD, MI, stroke, RI, UR Death

–15 –

P = NS Ef cacy TIMI major bleeding


Figure 1 | Comparison of the results of the APPRAISE?2,5 ATLAS ACS 2?TIMI 56,6 and TRACER7 trials. The data are for the study drug versus placebo. Bleeding was increased in all studies; only the ATLAS ACS 2?TIMI 56 trial demonstrated efficacy and decreased mortality (data for rivaroxaban 2.5?mg shown with modified intention-to-treat analysis). Follow-up times are as follows: APPRAISE?2 median 8?months; ATLAS ACS 2?TIMI 56 mean 13?months; TRACER median 16.5?months. *Events are reported on the basis of individual end points. Abbreviations: CVD, cardiovascular death; ICH,?intracranial hemorrhage; MI, myocardial infarction; RI, recurrent ischemia with rehospitalization; UR,?urgent coronary revascularization.

particu?larly favorably received. The investi? gators randomly assigned 15,526 patients with ACS to rivaroxaban 2.5?mg or 5.0?mg twice daily, or placebo on a background of aspirin alone (stratum?1) or aspirin plus a thienopyridine (stratum?2; 93% of patients). Notably, the doses of rivaroxaban used were much lower than that used for full anticoagulation, such as in atrial fibrillation (20?mg once daily). In the pooled cohort, a highly significant 16% relative risk reduction (8.9% vs 10.7%, P?=?0.008) in cardiovascular death, MI, and stroke was observed for rivaroxaban, with a similar reduction in all-cause mortality, MI, and stroke (9.2% vs 11.0%, P?=?0.006) and a 31% (2.3% vs 2.9%, P?=?0.02) reduction in stent thrombosis over a 2?year period. A dose-dependent increase in TIMI major bleeding occurred (placebo: 0.6% vs rivaroxa? ban 2.5?mg: 1.8% and rivaroxaban 5?mg: 2.4%, P?<0.001), but no increase in fatal bleeding was noted. Most importantly, a highly significant 34% reduction in cardiovascular mortality and a 32% reduction in total mortality was observed for rivaroxaban 2.5?mg twice daily.6 Despite its overwhelming efficacy, this trial reminded us that bleeding continues to be a limiting factor in studies of anticoagulant plus antiplatelet agents and that the window of efficacy needs to be carefully defined.
S2? |? JANUARY 2012

Another major trial involved the oral thrombin receptor antagonist vorapaxar. The TRACER trial7 faced issues with bleeding and was terminated early. In this study, 12,944 patients with ACS were randomly assigned to receive vorapaxar or placebo in addition to standard care. Vorapaxar failed to reduce cardio? vascular death, MI, stroke, recurrent ischemia, and urgent revasculariza? tion. This drug led to a 3.4-fold increased risk for intracranial hemorrhage (P ?<0.001) and a 35% increase in bleeding.7 The results of this trial highlighted the ever-present hazards of increased bleeding when adding new antithrombotic agents to standard therapy, especially in patients who are at increased risk for intracranial hemorrhage. A second trial (TRA 2°P-TIMI 50) of vorapaxar in patients with stable coronary artery disease will be reported next year; thus the final word on this drug is still to come. Side-by-side comparison of the three major oral anticoagulant trials shows that only the ATLAS ACS 2?TIMI?56 trial met its primary efficacy end point, reduced mortality, but all studies demon? strated significant increases in bleeding (Figure?1). Given the delicate balance between efficacy and bleeding, strategies to identify patients at high risk, and then triaging therapies accordingly, have generated much interest.

Mathews et?al. included 90,273 patients with ST-segment elevation or ST-segment elevation MI from the ACTION registry– GWTG? to derive (72,813 patients) and valid? ate (17,960 patients) a bleeding risk score using only baseline characteristics.8 Twelve patient characteristics and presenting factors emerged as predictors of major bleeding in the model. Notably, history of stroke or transient ischemic attack, which was present in only 8% of this cohort, did not emerge as risk factors for major bleeding in this model, although they have previously been identified as risk factors for intra? cranial hemorrhage.8 The risk stratification model by Mathews et?al. could guide clinical practice with respect to patient selection for antiplatelet and anticoagulant agents, but the extent to which it can be generalized remains unknown. Ideally, the score should be validated in the setting of randomized trials, various agents, and in high-risk and low-risk patient populations before it is applied?widely. The year 2011 was a historic landmark in the treatment of ACS. Major strides were made in diagnostic technology, tailored medical therapy, and the safe use of a very low-dose anticoagulant. Without doubt, the years to come will see many more trials to better educate physicians on how to walk the tightrope between efficacy and bleeding.
TIMI Study Group, Cardiovascular Division, Department of Medicine, Brigham and Women’s Hospital and Harvard Medical School, 350 Longwood Avenue, 1st floor office, Boston, MA 02115, USA (P . Kohli, C. P . Cannon). Correspondence to: C.?P . Cannon
Competing interests C. P. Cannon declares associations with the following companies: Accumetrics, Alnylam Pharmaceuticals, AstraZeneca, Automedics Medical Systems, BristolMyers Squibb/Sanofi, GlaxoSmithKline, Intekrin Therapeutics, Merck, Novartis, Pfizer, and Takeda. P.?Kohli declares no competing interests. 1. Mills, N.?L. et?al. Implementation of a sensitive troponin?I assay and risk of recurrent myocardial infarction and death in patients with suspected acute coronary syndrome. JAMA 305, 1210–1216 (2011). Mega, J.?L. et?al. Dosing clopidogrel based on CYP2C19 genotype and the effect on platelet reactivity in patients with stable cardiovascular disease. JAMA 306, 2221–2228 (2011). Price, M.?J. et?al. Standard- vs high-dose clopidogrel based on platelet function testing after percutaneous coronary intervention: the GRAVITAS randomized trial. JAMA 305, 1097–1105 (2011). Steg, P .?G. et?al. RUBY?1: a randomized, doubleblind, placebo-controlled trial of the safety and tolerability of the novel oral factor Xa inhibitor darexaban (YM150) following acute coronary syndrome. Eur. Heart J. 32, 2541–2554 (2011).




5. Alexander, J.?H. et?al. Apixaban with antiplatelet therapy after acute coronary syndrome. N. Engl. J. Med. 365, 699–708 (2011). Mega, J.?L. et?al. Rivaroxaban in patients with a recent acute coronary syndrome. N. Engl. J. Med. NEJMoa1112277. Tricoci, P . et?al. Thrombin-receptor antagonist vorapaxar in acute coronary syndromes. N. Engl. J. Med. NEJMoa1109719. Mathews, R. et?al. In-hospital major bleeding during ST?elevation and non?ST?elevation myocardial infarction care: derivation and validation of a model from the ACTION Registry?-GWTG?. Am. J. Cardiol. 107, 1136–1143 (2011).





Stroke prevention in AF
Gregory Y.?H. Lip

In 2011, key trials with oral factor?Xa inhibitors in patients with atrial fibrillation highlighted promising data on these novel anticoagulants. Patients with ≥1 stroke risk factors can be considered for oral anticoagulation. These novel, fixed-dose drugs are given without monitoring, so clinicians must learn to balance stroke and bleeding?risks.
Lip, G.?Y.?H. Nat. Rev. Cardiol. 9, 71–73 (2012); published online 20 December 2011; doi:10.1038/nrcardio.2011.203

The year 2011 saw the publication of three pivotal phase?III trials for two oral direct factor?Xa inhibitors, apixaban and rivaroxa? ban,1–3 as well as important articles on bleeding risk assessment 4 and the net clinical benefit of thromboprophylaxis.5 Novel oral anticoagulants that are viable alternatives to warfarin have clearly changed the landscape of stroke prevention in patients with atrial fibrillation (AF).6 Until recently, the recommended approach was artificially to stratify patients with AF into low, intermediate, and high risk strata—despite stroke risk being a continuum—so that those classed as being at high risk could be targeted for an inconvenient (and potentially dangerous) drug, warfarin. Many studies, however, have shown that the categorization of patients into low, intermediate, or high risk strata has poor correlation with actual warfarin prescribing, and that the predictive value of risk schemes such as the CHADS2 score (Box?1) to identify high-risk patients is suboptimal.7 Consequently, guidelines now recommended the use of the CHA2DS2?VASc score (Box?1) to complement the older (but simpler) CHADS2 score. Indeed, emphasis is now directed towards identification of ‘truly low-risk’ patients with AF by being more inclusive (rather than exclusive) of common risk factors for stroke, because these patients might not need any antithrombotic therapy. Meanwhile, patients with ≥1 risk factor for stroke should be considered for effective stroke prevention with oral anticoagulation, whether with (very) well-controlled warfarin or one of the

novel agents, either an oral direct thrombin inhibitor (such as dabigatran) or an oral direct factor?Xa inhibitor (for example, rivaroxaban or apixaban). 1 Indeed, the CHA2DS2?VASc score has consistently been shown to outperform the CHADS 2 score in identification of truly low-risk patients and is as good as—and possibly better than—the CHADS2 score in the identification of high-risk patients who subsequently suffer?thromboembolism.7,8
Key advances
■■ In the AVERROES trial, apixaban was superior to aspirin for stroke prevention in patients with atrial fibrillation (AF), with similar rates of major bleeding and improved tolerability1 ■■ In the ARISTOTLE trial, apixaban was superior to warfarin for prevention of stroke and systemic embolism in patients with AF, with significantly less major bleeding and improved survival2 ■■ In the ROCKET?AF trial, rivaroxaban was noninferior to warfarin for stroke prevention in a high-risk population of patients with AF, with similar rates of major bleeding3 ■■ The HAS?BLED score is well validated to predict major-bleeding events in patients receiving anticoagulation therapy, and outperforms other bleeding risk assessment schemes4 ■■ ‘Truly low-risk’ patients with a CHA2DS2?VASc score of 0 do not require thromboprophylaxis; net clinical benefit is greatest in patients with a high HAS?BLED score, where reduced ischemicstroke risk outweighs the increased intracranial-bleeding risk5

Attention has also been directed to assessment of bleeding risk. Common risk factors for bleeding (as well as potentially correctable risk factors, such as uncontrolled blood pressure and concomitant aspirin use in patients receiving anticoagulation therapy) can inform clinical decision-? making, especially with the novel oral anticoagulants that can come in high-dose and low-dose?regimens.9 Investigators in the AVERROES trial1 studied 5,599 patients with AF and ≥1 risk factor for stroke, and who had refused war? farin or been deemed unsuitable for war?farin by the investigators on the basis of the inclusion criteria. The trial was stopped early because of the clear superiority of apixaban over aspirin, with a 55% reduction in the primary end point of stroke and systemic embolism (HR?0.45, 95%?CI 0.32–0.62, P?<0.001), but with no significant difference between apixaban and aspirin for major bleeding (HR?1.13, 95%?CI 0.74–1.75, P?=?0.57) or intracranial hemorrhage. Furthermore, fewer drug discontinuations occurred with patients taking apixaban, which indicates that this drug is better tolerated than aspirin. Until now, patients who refused or were unsuitable for warfarin were treated with aspirin; therefore, this trial will certainly change clinical practice, especially given that the evidence for a beneficial (and safe) role for aspirin in patients with AF is?weak.10 Researchers in the ARISTOTLE trial 2 randomly allocated 18,201 patients with AF to receive either warfarin or apixaban, and reported a 21% reduction in the primary end point of stroke and systemic embolism with apixaban (HR?0.79, 95%?CI 0.66– 0.95, P?<0.001 for noninferiority, P?=?0.01 for superi? ority), with a 31% reduction in major bleeding (HR?0.69, 95%?CI 0.60–0.80, P?<0.001) and an 11% reduction in mortality (HR?0.89, 95%?CI 0.80–0.99, P?=?0.047). The primary end point was driven by the reduction in hemorrhagic stroke (HR?0.51, 95%?CI 0.35–0.75, P?<0.001), with no difference in ischemic stroke (HR?0.92, 95%?CI 0.74–1.13, P?=?0.42) between apixaban and warfarin. Another oral direct factor?Xa inhibitor, rivaroxaban, was compared with warfarin in the ROCKET?AF trial.3 A higher-risk cate? gory of patients (n?=?14,264, mean CHADS2 score?=?3.5) was enrolled in ROCKET?AF compared with the ARISTOTLE and AV E R ROE S t r i a l s ( m e an C HA D S 2 score?=?2.1 in both). Also, 55% of the ROCKET?AF population was a secondary prevention cohort (by contrast with 20% and 14% in the ARISTOTLE and AVERROES trials, respectively). Rivaroxaban was clearly
JANUARY 2012? |? S3

Box 1 | Abbreviations and definitions CHADS2 Congestive heart failure (1?point) Hypertension (1?point) Age ≥75?years (1?point) Diabetes mellitus (1?point) Stroke or thromboembolism (2?points) CHA2DS2?VASc Congestive heart failure (1?point) Hypertension (1?point) Age ≥75?years (2?points) Diabetes mellitus (1?point) Sex, female (1?point) Vascular disease (1?point) Age 65–74?years (1?point) Stroke or thromboembolism (2?points) HAS?BLED Hypertension (uncontrolled blood pressure) Abnormal renal or liver function Stroke Bleeding history or predisposition Labile international normalized ratio (INR) Elderly (age >65?years) Drugs concomitantly (for example, concomitant aspirin or nonsteroidal antiinflammatory drugs, or alcohol abuse) Net clinical benefit [ISR with no treatment – ISR on treatment] – 1.5*[IHR on treatment – IHR with no?treatment]
Abbreviations: IHR, intracranial-hemorrhage rate; ISR, ischemic-stroke rate.

noninferior to warfarin for the primary end point of stroke and systemic embolism (HR?0.79, 95%?CI 0.66–0.96, P?<0.001 for non? inferiority). In the more-conventional and conservative intention-to-treat analysis, rivaroxa? ban failed to achieve superiority (HR?0.88, 95%?CI 0.74–1.03, P?<0.001 for noninferiority, P ?=?0.12 for superiority), although a per-protocol, on-treatment analysis reported superiority for rivaro? xaban (HR?0.79, 95%?CI 0.65–0.95, P?=?0.02). The primary safety end point, major and non? m ajor clinically relevant bleeding, was not different (HR?1.03, 95%?CI 0.96– 1.11, P ?=?0.44), although hemorrhagic stroke, intracranial hemorrhage, and fatalbleeding events were lower with rivaroxaban than warfarin. Until the advent of novel oral anti? coagulants, patients with AF who were receiving oral anticoagulation often came under the care of a warfarin clinic, which was responsible for anticoagulation monitoring and doses, rather than cardiologists per?se. Now, clinicians will have to make informed decisions on doses of these novel drugs. Of the novel direct thrombin inhibitors, only dabigatran is approved for use for
S4? |? JANUARY 2012

stroke prevention in patients with AF, and this drug is licensed at doses of 110?mg or 150?mg twice daily (or 75?mg twice daily for patients with severe renal failure in the USA). Apart from the criteria of age (>80?years) and concomitant drug use (such as verapamil), how should the dose of dabigatran be decided on the basis of bleeding risk? To help clinical decision-making, a simple acronym of common risk factors for bleeding (the HAS?BLED score; Box?1) has been recommended in European and Canadian guidelines. Various validations of the HAS?BLED score were published in?2011.4,5 In a large, contemporary, anti? coagulated population of patients with AF, the HAS? BLED score performed best of all the tested bleeding-risk schemes, and more-accurately discriminated patients on the basis of their bleeding risk (whether assessed by the c?? s tatistic, or estimates of Net Reclassification Improvement and Integrated Discrimination Improvement), with a stepwise increase in rates of major bleeding with increasing HAS?BLED score (P?<0.0001 for trend).4 Multivariate analyses identified concurrent aspirin use (HR?2.10), renal impairment (HR?1.98), age ≥75?years (HR?1.63), diabetes mellitus (HR?1.47), and heart failure or left ventricular dysfunction (HR?1.32) as significant predictors of bleeding. What does this mean for clinicians in everyday practice, who have to balance the risks of stroke and bleeding? In a large, nationwide, ‘real-world’ cohort study from Denmark (n ?=?132,372), Olesen and colleagues balanced stroke risk (assessed by the CHADS2 and CHA2DS2?VASc scores) with bleeding risk (assessed using the HAS?BLED score), and the net clinical benefit (Box?1) of ischemic-stroke prevention versus intracranial-? hemorrhage risk was calculated.5 The net clinical benefit was not positive for aspirin at any stroke-risk or bleeding-risk strata, which led the investi? gators to conclude that “aspirin should not be used for thromboprophylaxis in any patient with AF”.5 Unsurprisingly, the combination of warfarin plus aspirin did not confer any additional benefit in preventing thromboembolism, but substantially increased bleeding?risk. In patients treated with warfarin, the only category where the net clinical benefit was negative was patients with a CHA2DS2?VASc score of 0, which reflects their truly low-risk status. The net clinical benefit was non-? negative (either neutral or positive) at a CHADS2 score?≥0 and a CHA2DS2?VASc score?≥1. In addition, the net clinical benefit

was clearly positive, in favor of warfarin, in patients with increased risk of stroke or thromboembolism, and was even greater at high HAS?BLED scores (≥3), which indicates that patients at high risk of bleeding are also at high risk of stroke or thrombo? embolism —and that the absolute benefit of warfarin in reducing the risk of ischemic stroke far outweighs the smaller absolute increase in the risk of serious bleeding. The practical use of the HAS?BLED score also encourages clinicians to consider bleeding risk factors that can be corrected, such as uncontrolled blood pressure, concomitant use of aspirin or nonsteroidal anti-? inflammatory drugs, and time in therapeutic range if the patient is on warfarin (the H, D, and L in HAS?BLED,?respectively). The year 2011 has seen great advances?in assessment of stroke and bleeding risks in?patients with AF, and exciting new data on the novel oral anticoagulants. Prospects are much improved for patients with?AF.
University of Birmingham Centre for Cardiovascular Sciences, City Hospital, Dudley?Road, Birmingham B18?7QH, UK
Competing interests The author declares associations with the following companies and organizations: Astellas, AstraZeneca, Bayer, Biotronik, Boehringer Ingelheim, Bristol-Myers Squibb/Pfizer, Daiichi?Sankyo, focusAF, Merck, Portola and Sanofi Aventis. See the article online for full details of the relationships. 1. Connolly, S.?J. et?al. Apixaban in patients with atrial fibrillation. N. Engl. J. Med. 364, 806–817 (2011). Granger, C.?B. et?al. Apixaban versus warfarin in patients with atrial fibrillation. N. Engl. J. Med. 365, 981–992 (2011). Patel, M.?R. et?al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N. Engl. J. Med. 365, 883–891 (2011). Lip, G.?Y., Frison, L., Halperin, J.?L. & Lane, D.?A. Comparative validation of a novel risk score for predicting bleeding risk in anticoagulated patients with atrial fibrillation: the HAS?BLED (Hypertension, Abnormal Renal/Liver Function, Stroke, Bleeding History or Predisposition, Labile INR, Elderly, Drugs/Alcohol Concomitantly) score. J. Am. Coll. Cardiol. 57, 173–180 (2011). Olesen, J.?B. et?al. Risks of thromboembolism and bleeding with thromboprophylaxis in patients with atrial fibrillation: a net clinical benefit analysis using a ‘real world’ nationwide cohort study. Thromb. Haemost. 106, 739–749 (2011). Ahrens, I., Lip, G.?Y. & Peter, K. New oral anticoagulant drugs in cardiovascular disease. Thromb. Haemost. 104, 49–60 (2010). Karthikeyan, G. & Eikelboom, J.?W. The CHADS2 score for stroke risk stratification in atrial fibrillation—friend or foe? Thromb. Haemost. 104, 45–48 (2010). Lip, G.?Y. Stroke in atrial fibrillation: epidemiology and thromboprophylaxis. J. Thromb. Haemost. 9?(Suppl.?1), 344–351 (2011).








9. Lip, G.?Y.?H. et?al. Bleeding risk assessment and management in atrial ?brillation patients: executive summary of a position document from the European Heart Rhythm Association [EHRA], endorsed by the European Society of Cardiology [ESC] Working Group on Thrombosis. Thromb. Haemostat. 106, 997–1011 (2011). 10. Lip, G.?Y. The role of aspirin for stroke prevention in atrial fibrillation. Nat. Rev. Cardiol. 8, 602–606 (2011).


Heart failure therapy—technology to the fore
John J. V. McMurray

Studies published in 2011 in the field of heart failure have reinforced the benefit of cardiac resynchronization therapy in patients with mild symptoms and confirmed the value of left ventricular assist devices and CABG surgery in selected patients. Conversely, the efficacy of nesiritide in acute heart failure has been questioned.
McMurray, J.?J. V. Nat. Rev. Cardiol. 9, 73–74 (2012); doi:10.1038/nrcardio.2011.212

Although many interesting studies on epidemio? logy, pathophysiology, and biomarkers in heart failure (HF) were published in 2011, my focus in this article is on treatments that are available for use and can make an impact in everyday practice. Five important studies, covering device and drug therapy, and surgery in patients with HF will be discussed. In August 2011, a follow-up report from MADIT-CRT1 reinforced the impressive effect of cardiac resynchronization therapy (CRT) on outcomes among patients with HF, even in those with mild symptoms. The investigators showed that CRT plus a defibrillator, compared with a defibrillator alone, reduced the risk of a first episode of worsening HF by 46% (P?<0.001) and, importantly, also reduced the risk of subsequent episodes by 38% (P?=?0.003). Because patients experiencing a first episode of worsen? ing HF were at a substantially elevated risk of a further event, the absolute reduction in risk of subsequent episodes with CRT was large (from 72 to 44 events per 100 patient years of follow-up). The benefit was greatest in patients with left bundle branch block. In addition, a 19-fold increase in mortality risk was reported in patients experiencing a second episode of worsening HF.1 These findings are important for several reasons. Clearly, worsening of HF is extremely unpleasant for patients and usually leads to hospital admission with major consequences (economic and otherwise) for health-care systems, and is associated with an increased risk of death.1 These findings also highlight the limitations of the conventional ‘time to first-event’ analysis of

trials, in which recurrent events that are frequent and of great clinical significance than first HF events are ignored. A report from INTERMACS 2 also re-? inforced the impressive technological progress made with left ventricular assist devices (LVADs). In a post-approval study required by the FDA, outcomes in the first 169 consecutive patients receiving a HeartMate?II? (Thoratec Corporation, Pleasanton, CA, USA) device as a bridge-to-transplant were compared with those in 169 patients receiving different LVADs for the same indication over the same time period. The proportion of patients transplanted, recovered, or receiving continuing LVAD support at 6?months was 91% in the HeartMate?II? group and 80% in the comparator group (85% and 70%,

respectively, at 1?year). These excellent outcomes from ‘real-world’ practice at 77 centers in the USA are consistent with the findings of the pivotal, randomized HeartMate?II? trial,3 which showed that this continuousflow device was superior to the comparator pulsatile-flow device. Monitoring devices are currently of immense interest in HF research and the findings of the CHAMPION study 4 were notable. This moderate-sized trial showed that a small, implantable hemodynamic monitor improved outcomes in patients with NYHA class?III HF who had been hospitali? zed for HF in the previous year and were optimally treated. The monitor was placed transvenously in a pulmonary artery branch and allowed wireless non? invasive measurement of pulmonary artery pressure. Physicians in the control group and patients in both groups did not have knowledge of pressure measurements; however, site investi?gators, the trial principal investi? gators, and sponsor (CardioMEMS, Atlanta, GA, USA) employees had access to pressure measurements and these were used to adjust therapy. Addition of nitrates and hydralazine and adjustment of diuretic therapy, including use of intravenous diure? tics, were more frequent in the inter? vention group. The primary outcome was the rate of HF hospitaliza? tion over 6?months; 83?hospitaliza? tions occurred among 55 of the 270 patients in the intervention group, compared with 120 among 80 of the 280 patients in the control group (rate 0.31 vs 0.44; HR?0.70, 95%?CI 0.60–0.84, P?<0.0001). In absolute terms, 12.5 fewer hospitaliza? tion events per 100 patients occurred over

? Newlight |

JANUARY 2012? |? S5

Key advances
■■ Cardiac resynchronization therapy reduced not only the first, but also subsequent, episodes of worsening heart failure (HF) in patients with systolic dysfunction and mild symptoms1 ■■ Use of the HeartMate II?(Thoratec Corporation, Pleasanton, CA, USA) continuous-flow left ventricular assist device was associated with excellent 6-month and 12-month survival in a ’real-world’ registry study2 ■■ Pharmacological intervention prompted by noninvasive pulmonary artery monitoring may reduce the risk of HF hospitalization in patients with moderately severe symptoms and previous admission for HF4 ■■ In selected patients with ischemic systolic HF and a life expectancy of more than 2?years, CABG surgery may reduce cardiovascular mortality and morbidity6 ■■ Nesiritide has a small and clinically insignificant effect on dyspnea and no effect on mortality or worsening HF in patients with acute HF7

6?months in the intervention group, meaning that eight patients needed to be treated to prevent one hospitalization. 4 Although these results are impressive, the trial has been criticized because of the active involvement of the sponsor in encouraging treatment changes in the intervention group, and whether the benefit would be as substantial when used in ordinary practice has been questioned. Nevertheless, the positive findings of the CHAMPION trial4 contrasted strikingly with those of other studies, such as the negative study DOT-HF5 in which an implantable intrathoracic impedance detection device was used. Another study that raised as many questions as answers was the STICH trial.6 In this study, 1,212 patients with a left ventric? ular ejection fraction <35%, coronary artery disease amenable to surgery, and no indica? tion for revascularization (that is, left main disease or Canadian Cardiovascular Class?III angina or greater) were randomly assigned to receive medical therapy or CABG surgery. The primary end point was death from any cause. Over a median follow-? up of 56?months, 218 of the 610 patients assigned to surgery and 244 of the 602? patients alloca? ted to medical therapy died (36% vs 41%; HR?0.86, 95%?CI 0.72–1.04, P?=?0.12). However, occurrence of the key secondary end points (cardiovascular death and death or cardiovascular hospitalization) was signifi? cantly less common in the surgical
S6? |? JANUARY 2012

group, as was all-cause mortality in an adjusted intention-to-treat analysis and in a per protocol analysis. As in prior trials of CABG surgery, there was an initial mortality excess in the surgery group and benefit from surgery was not apparent until after approximately 2?years of follow-up.6 On balance, the STICH trial probably was a positive study, although the patients were highly selected and atypically young (mean age 60?years) for the majority with HF; two-thirds had symptoms of angina. Unexpectedly, pre? operative evaluation of myocardial viability did not help to identify patients who would benefit from revascularization. The takehome message seems to be that, in carefully selected patients with systolic HF and coronary artery disease likely to survive for more than 2?years, CABG surgery leads to reduced long-term mortality and morbidity compared with medical therapy alone. The list of treatments that are ineffective in acute HF grew longer in 2011 with the publication of the results of ASCEND-HF,7 which showed that nesiritide (recombinant human B?type natriuretic peptide) had only a small effect on dyspnea at 6?h and 24?h and no effect on mortality or worsening HF at 30?days. This trial ended a long-running controversy about the safety and efficacy of nesiritide and highlighted the difficulty of demonstrating a benefit for novel therapies over and above that of conventional treatment with diuretics and nitrates, which is rapid and substantial.8 Despite the lack of success with treatments for acute HF to date, interesting new drugs continue to be developed. One of these, with intriguing proof-of-concept data in patients, is a ‘first-in-class’ cardiac myosin activator, omecamtiv mercarbil, which is thought to improve cardiac performance by increasing the rate of transition from the weak myosin-acting binding state to the strongly bound (that is, force-generating) state. This agent increases the duration of systole without increasing heart rate or myocardial oxygen?consumption.9 Finally, I would like to mention an important study relevant to the holistic care of patients with HF. Peterson and colleagues10 looked at health literacy in 1,494 patients with an emergency room visit or hospitaliza? tion with HF using three brief screening questions—how often do you have someone help you read hospital materials; how often do you have problems learning about your medical condition because of difficulty reading hospital materials; and how confident are you filling out forms by yourself?

More than one in six patients showed a low level of health literacy, which was an indepen? dent predictor of all-cause mortality, after taking account of other prognostic variables.10 These findings argue for testing of the effect of interventions thought to improve health literacy on outcomes in HF. In summary, 2011 has been largely a year of consolidation for HF therapeutics. We have an array of effective drugs, devices, and surgical procedures for patients with systolic HF, but substantial progress has yet to be made in the treatment of patients?with?preserved ejection fraction and those with acute?decompensation.
British Heart Foundation Cardiovascular Research Centre, University of Glasgow, 126?University Place, Glasgow G12 8TA, UK.
Competing interests The author declares associations with the following companies: Amgen and Johnson & Johnson/Scios. See the article online for full details of the relationships. 1. Goldenberg, I. et?al. Reduction of the risk of recurring heart failure events with cardiac resynchronization therapy: MADIT-CRT (Multicenter Automatic Defibrillator Implantation Trial With Cardiac Resynchronization Therapy). J. Am. Coll. Cardiol. 58, 729–737 (2011). 2. Starling, R.?C. et?al. Results of the post?US Food and Drug Administration-approval study with a continuous flow left ventricular assist device as a bridge to heart transplantation: a prospective study using the INTERMACS (Interagency Registry for Mechanically Assisted Circulatory Support). J. Am. Coll. Cardiol. 57, 1890–1898 (2011). 3. Slaughter, M.?S. et?al. Advanced heart failure treated with continuous-flow left ventricular assist device. N. Engl. J. Med. 361, 2241–2251 (2009). 4. Abraham, W.?T. et?al. Wireless pulmonary artery haemodynamic monitoring in chronic heart failure: a randomised controlled trial. Lancet 377, 658–666 (2011). 5. van Veldhuisen, D.?J. et?al. Intrathoracic impedance monitoring, audible patient alerts, and outcome in patients with heart failure. Circulation 124, 1719–1726 (2011). 6. Velazquez, E.?J. et?al. Coronary-artery bypass surgery in patients with left ventricular dysfunction. N. Engl. J. Med. 364, 1607–1616 (2011). 7. O’Connor, C.?M. et?al. Effect of nesiritide in patients with acute decompensated heart failure. N. Engl. J. Med. 365, 32–43 (2011). 8. Felker, G.?M. et?al. Diuretic strategies in patients with acute decompensated heart failure. N. Engl. J. Med. 364, 797–805 (2011). 9. Cleland, J.?G. et?al. The effects of the cardiac myosin activator, omecamtiv mecarbil, on cardiac function in systolic heart failure: a double-blind, placebo-controlled, crossover, dose-ranging phase?2 trial. Lancet 378, 676–683 (2011). 10. Peterson, P .?N. et?al. Health literacy and outcomes among patients with heart failure. JAMA 305, 1695–1701 (2011).


New insights—from risk factors to treatment implications
George L. Bakris

The results of several hypertension studies published in 2011 have contributed to our knowledge on the risks of and treatment for this condition, including the effects of slow-wave sleep, nocturnal dosing of medication, variability in post-stroke blood-pressure reduction, and the impacts of a low-sodium diet.
Bakris, G.?L. Nat. Rev. Cardiol. 9, 75–77 (2012); published online 13 December 2011; doi:10.1038/nrcardio.2011.202

Four key studies, completed and published in 2011, should affect the evaluation of patients with hypertension and practice patterns in this population. In one such trial, conducted in Spain, medications were dosed at night in order to improve the dipping status of blood pressure (BP). In this trial 661 patients with hypertension who were at various stages of chronic kidney disease (CKD) were randomly assigned to take all prescribed hypertension medications upon awakening or at least one drug before going to bed.1 At the end of follow-up (median 5.4?years), patients who took the nocturnal dose had an adjusted risk for total cardiovascular events that was approximately onethird that of patients who took medication upon awakening. 1 Patients who received nocturnal treatment also had a significantly lower mean BP during sleep and a greater proportion of these indivi? duals demonstrated controlled ambulatory BP compared with those taking medication upon awakening.1 These data are consistent with those from a shorter-term study of patients with stage 2–3 CKD and no signifi? cant proteinuria.2 However, a pilot study in the African American Study of Kidney Disease (AASK) cohort of ~100 patients with stage 4 CKD and proteinuria using similar methodo? logy as the Spanish group failed to show this benefit on BP dipping status or BP control in the morning after 1?year of dosing antihypertensive medications at night.3 One of the major differences between the studies was that 100% of the patients in the AASK cohort had proteinuria at various levels and stage 4 nephropathy, whereas only about 10% of the patients in the Spanish study fit this description. A nocturnal dosing strategy was also tested in the large CONVINCE trial, 4 but failed to show a reduction in cardio? vascular risk. However, the CKD and dipping status of the individuals enrolled in

the trial was unknown. Given the increased cardiovascular risk among patients with stage?3 CKD, ambulatory BP monitoring (ABPM) is required to assess dipping status and indivi?dualize dosing regimens in people who meet the criteria as nondippers. As yet no ABPM data are available from large cohorts with stage?4 CKD and so the recom? mendation cannot be extended to this group of patients. The second major advance in 2011 was the finding of an association between poor sleep quality and the development of hyper? tension.5 This relationship is unrelated to sleep apnea and has more to do with the amount and type of sleep. Fung et?al. evaluated whether incident hypertension is associ?ated with polysomnography measures of sleep-disordered breathing, sleep duration, and sleep architecture in older men (mean age 75?years).5 Participants included 784 community-dwelling, ambulatory men from the Outcomes of Sleep Disorders in Older Men Study who did not have
Key advances
■■ Among patients with moderate-stage nephropathy, dosing medications at night has the advantage of improving blood pressure (BP) goal attainment in the early morning when cardiovascular risk is?highest1 ■■ Failure to recognize insufficient quality of sleep secondary to frequent disturbances of awakening is a major contributing risk for hypertension development and poor?BP control5 ■■ Lowering BP by more than 15–20% in patients with hypertension early in the post-stroke period can result in long-term cognitive consequences6 ■■ Low-sodium diets clearly reduce BP, but tend only to reduce cardiovascular events in trials; in populations, low-sodium diets reduce?mortality8

hypertension at the time of their in-home sleep studies and who returned for followup?after 2–4?years. By the end of follow-up (mean 3.4?years), 243 men met the criteria?for incident hypertension. After adjustment for age, nonwhite race, study site, and BMI, the only sleep index to remain signifi? cantly associated with incident hyper?tension was decreased stage N3 (or slow-wave) sleep. No attenuation of this association was seen after accounting for sleep duration, sleep fragmentation, and indices of sleepdisordered breathing. Thus, failure to get an appropriate amount of slow-wave sleep each night is a risk factor for development of hyper? tension. Reduced amounts of slowwave sleep are seen commonly in elderly patients and could be related to increased frequency of waking to urinate. Physicians should be aware of decreased deep sleep as a risk factor for hypertension and ensure that patients get an extended rest period at night by decreasing stimuli, such as caffeine and liquid intake. This strategy is used in my practice with a fair amount of?success. A third important trial published in 2011 provides new information on the consequences of aggressive BP management in the setting of acute stroke. In SCAST,6 2,029?patients from northern Europe with acute ischemic or hemorrhagic stroke and systolic BP >140?mmHg were randomly assigned within 30?h of symptom onset to cande?sartan or placebo for 7?days. The dose was increased from 4?mg on day 1 to 16?mg on days 3–7. During the 7?day treatment period, BP was 5/2?mmHg lower in the candesartan group than in the placebo group (mean BP 147/82?mmHg vs 152/84?mmHg). At the end of the 6?month follow-up, the risk of the primary composite vascular end point did not differ between treatment groups. The co-primary end point of functional outcome (modified Rankin scale), however, was associ? ated with a higher risk of poor outcome in the candesartan group. The results were interpreted by the investigators to suggest some harm and no benefit for BP lowering after acute stroke.6 To understand this finding, clinicians should be aware that current BP guidelines for the post-stroke period differ according to whether the stroke is hemorrhagic or ischemic. 7 In addition, the approach to BP management for the prevention of stroke is totally different from its management post stroke. In hemorrhagic stroke, current guidelines suggest cautious lowering of BP by no more that 20% in the first 24?h.7 Agents with rapid, short-term
JANUARY 2012? |? S7

BP goals with pharmaco? therapy.10 In 2011, recruitment for the Symplicity HTN?3 trial started in the USA; with a protocol that is more stringent than Symplicity HTN?2 and involves more frequent ABPM. This new trial will include between-group differences in ABPM as a prespecified secondary end point. Hypertension management prior to study enrollment will be more intense and will include spironolactone. If the results of the Symplicity HTN?3 trial are as gratifying as that of Symplicity HTN?2, a true alternative to medications will exist for refractory hypertension. Many other ongoing clinical trials have been designed to address and, hopefully, answer important questions about specific approaches to reducing cardio? vascular risk and the progression of CKD in patients with hypertension.
Hypertensive Diseases Unit, Section of Endocrinology, Diabetes and Metabolism, Department of Medicine, University of Chicago Pritzker School of Medicine, 5841 South Maryland Avenue, MC 1027, Chicago, IL?60637, USA.
Competing interests The author declares associations with the following companies: Abbott, CVRx, Eli Lilly, FDA, Forest Laboratories, Johnson & Johnson, Medtronic, Novartis, Relypsa, Servier, and Takeda. See the article online for full details of the relationships. 1. Hermida, R.?C., Ayala, D.?E., Mojon, A. & Fernandez, J.?R. Bedtime dosing of antihypertensive medications reduces cardiovascular risk in CKD.?J.?Am. Soc. Nephrol. Minutolo, R. et?al. Changing the timing of antihypertensive therapy to reduce nocturnal blood pressure in CKD: an 8?week uncontrolled trial. Am. J.?Kidney Dis. 50, 908–917 (2007). Rahman, M. & Appel, L.?J. Should reducing nocturnal blood pressure be a therapeutic target in CKD? The time is ripe for a clinical outcomes trial. Am. J. Kidney Dis. 50, 901–903 (2007). Black, H.?R. et?al. Principal results of the Controlled Onset Verapamil Investigation of Cardiovascular End Points (CONVINCE) trial. JAMA 289, 2073–2082 (2003). Fung, M.?M. et?al. Decreased slow wave sleep increases risk of developing hypertension in elderly men. Hypertension 58, 596–603 (2011). Sandset, E.?C. et?al. The angiotensin-receptor blocker candesartan for treatment of acute stroke (SCAST): a randomised, placebocontrolled, double-blind trial. Lancet 377, 741–750 (2011). Aiyagari, V. & Gorelick, P .?B. Management of blood pressure for acute and recurrent stroke. Stroke 40, 2251–2256 (2009). Taylor, R.?S., Ashton, K.?E., Moxham, T., Hooper,?L. & Ebrahim, S. Reduced dietary salt for the prevention of cardiovascular disease: a meta-analysis of randomized controlled trials (Cochrane review). Am. J.?Hypertens. 24, 843–853 (2011).

? Elena Babushkina |

action such as labetalol, nicardipine, and fenoldapam are preferred as nitroglycerin and nitroprusside can increase intra? cranial pressure. For patients with ischemic stroke who are not receiving thrombolytic therapy, BP should be lowered if markedly elevated (>220/120?mmHg). 7 In SCAST, baseline BP was 171/90?mmHg,6 so a reduction to ~140?mmHg is excessive given the recommendation and, therefore, the lack of benefit is not surprising. The fourth notable study published in 2011 is a meta-analysis in which the investigators aimed to assess the long-term effects of dietary salt reduction on mortality and cardiovascular morbidity and whether BP reduction is an explanatory factor in any effect of such dietary interventions.8 A total of 7 studies were included, 3 in normo? tensive individuals, two in patients with hyper? tension, one in a mixed population, and one in patients with heart failure with end of trial follow-up of 7–36?months and longest observational follow up to 12.7?years. The data failed to show strong evidence for a reduced risk of all-cause mortality or cardio? vascular morbidity in normo? tensive indivi? duals and patients with hyper? tension on low-salt diets. The investigators suggest their review had insufficient power to exclude clinically important effects, since only 665?deaths occurred in 6,250 participants. They further state that their estimates of benefits from dietary salt restriction are consistent with the predicted small effects on clinical events attributable to the small BP reduction?achieved. This information puts into perspective the limitations of clinical trials. The evidence that salt reduction lowers BP is clear and consistent. Since BP reduction is critical for reducing the risk of stroke and progression of CKD, low-salt diets should also be of benefit.
S8? |? JANUARY 2012

However, most people do not consistently maintain a low sodium intake leading to fluctu?ation in BP, which is a major risk factor for stroke. The real impact of low sodium intake on mortality has been demonstrated in Japan and Finland where sodium reduction policies have been instituted for all foods sold in stores. In the USA, reducing dietary salt by 3?g per day is projected to reduce the annual number of all-cause deaths from 44,000 to 92,000.9 Hence, large numbers of indivi? duals and years of?follow-up are required to see the effect?of consistent reduced sodium intake. Not surprisingly, therefore, trial data and even meta-? analyses cannot adequately demon? strate these effects. Clinicians should focus on the established fact that BP reduction is consistently associ?ated with reductions in cardio?vascular event rates and continue to stress the importance of a low-sodium diet to patients with?hypertension. Finally, the multicenter, prospective, randomized Symplicity HTN?2 trial, which was published in late 2010, 10 provided ground-breaking results for a new concept of treatment for refractory hypertension. Participants with systolic BP ≥160?mmHg (or ≥150?mmHg for those with type?2 diabetes mellitus) were randomly allocated to undergo renal denervation together with continuation of previous treatment or maintain previous treatment alone. Office-based BP measurements were reduced by 32/12?mmHg (baseline 178/96?mmHg) in the renal denervation group, control group BP values did not differ from the baseline of 178/97?mmHg, and between-group differences in BP at 6?months were 33/11?mmHg. No serious procedure or device-related complications occurred and adverse events did not differ between groups. The investi? gators concluded that catheterbased renal denervation is a safe and viable treatment for patients who cannot achieve








9. Bibbins-Domingo, K. et?al. Projected effect of dietary salt reductions on future cardiovascular disease. N.?Engl. J.?Med. 362, 590–599 (2010). 10. Esler, M.?D. et?al. Renal sympathetic denervation in patients with treatment-resistant hypertension (the Symplicity HTN?2 trial): a randomised controlled trial. Lancet 376, 1903–1909 (2010).


Breakthrough for intervention?
Volkmar Falk

In 2011, both the PARTNER?A trial, in high-risk patients with severe aortic stenosis, and EVEREST?II, in patients with mitral insufficiency, showed noninferiority of transcatheter interventions compared with surgery for the chosen primary end points. However, both of the trials, and important registry data, identified limitations of transcatheter valve interventions.
Falk, V. Nat. Rev. Cardiol. 9, 77–78 (2012); published online 20 December 2011; doi:10.1038.nrcardio.2011.204

Data from two major randomized, controlled trials in which interventional therapy was compared with conventional surgery for the treatment of aortic stenosis in high-risk patients1 and for selected patients with mitral insufficiency 2,3 were published in 2011. Both trials showed noninferiority for their respective primary end points, but (together with data from national registries4,5) also highlighted limitations of interventional?therapy. The most-important trial in the field of valvular heart disease that was published in 2011 was the PARTNER?A trial,1 in which transcatheter aortic-valve implantation (TAVI) was compared with surgical aorticvalve replacement (SAVR) in a high-risk population of patients with severe aortic stenosis. From a total of 3,105?screened patients, 699 individuals (23%) were randomly allocated to either SAVR (n?=?351) or TAVI (n?=?348, transfemoral or trans? apical) using the Edwards SAPIEN ? (Edwards Lifesciences Corporation, Irvine, CA, USA) balloon-expandable valve. Patients were considered high risk on the basis of coexisting conditions associated with a risk of death ≥15% at 30?days. In the intention-to-treat analysis, death from any cause (the primary end point) was not different between the TAVI and SAVR groups at 30?days (3.4% vs 6.5%,?respectively) or 1?year (24.2% vs 26.8%, respectively), which led to the conclusion that TAVI was non? inferior to SAVR. Patients in the TAVI group had a shorter length of stay in the intensive-care unit and a shorter index hospitalization compared with patients undergoing SAVR. Bleeding and new onset of atrial fibrillation were more frequent in the surgical group, whereas rates of vascular and neuro? logical complications were higher in

the TAVI group. The rates of major stroke were 3.8% and 2.1% at 30?days, and 5.1% and 2.4% at 1?year, in the TAVI and surgical groups, respectively. At 1?year, TAVI was associated with a lower mean aorticvalve gradient compared with the surgical group, but with a higher rate of moderate or severe paravalvular regurgitation at 30?days (12.2% vs 0.9%) and 1?year (6.8% vs 1.9%). Functional improvement was significant for both groups after 1?year and not different between the groups. The results of the PARTNER?A trial 1 demonstrate that TAVI is not only superior to medical therapy in non?surgical candidates, but also a true alternative to surgical replacement for a selected, high-risk subgroup of patients with aortic stenosis. Data from a number of national TAVI registries were published in 2011. The FRANCE registry 6 included data on 244 patients who received an Edwards SAPIEN? (68%) or CoreValve? (Medtronic?CV Luxembourg S.a.r.l., Luxembourg; 32%) in 2009. Device success rate was 98.3% and 30?day mortal?ity
Key advances
■■ In high-risk patients with severe aortic stenosis, transcatheter aortic-valve implantation (TAVI) can be performed with similar 1?year survival and functional outcomes as conventional surgery1 ■■ Moderate and severe paravalvular leakage after TAVI is an independent predictor of late mortality4,5 ■■ Percutaneous edge-to-edge mitral-valve repair can be safely performed with low periprocedural complications2 ■■ Residual moderate mitral regurgitation is more frequent after percutaneous edge-to-edge repair than surgery and might impair long-term results3

was 12.7%. Stroke and vascular complications occurred in 3.6% and 7.3% of patients, respectively. Data from the FRANCE?2 registry,7 which included 2,419?patients treated between January?2010 and July?2011, were presented at the ESC Congress in 2011. Device success rate was 97.1%, and 30?day and 6?month mortality were 9.9% and 17.2%, respectively. The stroke rate of 4.0% was similar to the rates reported in the PARTNER trials.1 The high incidence of stroke after TAVI is concerning, and might result from a substantial embolic load during the TAVI procedure, as has been demonstrated using transcranial Doppler imaging during implantation and cerebral MRI after the?procedure. The UK TAVI registry 8 included 890 patients. Procedural success (97.2%) and stroke rate (4.1%) were similar to the French experience. Mortality at 30?days was 7.1%, and 1?year and 2?year survival were 78.6% and 73.7%, respectively. A degree of paravalvular aortic regurgitation (AR) regarded as suboptimal or unacceptable occurred in 61% of patients. The presence of moderate or severe AR (13% of patients) was an independent predictor of mortality at 1?year. In the German TAVI registry, 4 which comprised 690 patients (84% Medtronic CoreValve ? system), the rate of AR ≥2 was 17.2% and its presence was a strong, in? dependent predictor of in-hospital death (adjusted OR?2.43). In the Italian, multicenter registry 5 of 663 patients who underwent transfemoral TAVI with the third-generation 18?Fr?CoreValve? device, the rate of paravalvular leak ≥2 after implantation was 21%, and independently associated with mortality between 30?days and 1?year (HR?3.79). Therefore, data from three national registries have shown that paravalvular leakage is an independent risk factor for mortality after TAVI.4,5,8 The development of improved prosthetic implants and implantation techniques is required. In the absence of long-term follow-up data after TAVI, the results of the PARTNER?A trial1 and the data from the national registries4–8 must be interpreted with caution. Recommendations to individual patients must balance the obvious advantages of a less-invasive transcatheter approach compared with SAVR, with the increased risk of stroke and unfavorable longterm consequences of paravalvular leakage after?TAVI. The landmark mitral-valve trial published in 2011 was EVEREST?II.2 In total, 279?patients with severe mitral regurgitation
JANUARY 2012? |? S9




Figure 1 | Images from a patient in NYHA class?IV because of ischemic cardiomyopathy with poor LV ejection fraction (<20%) and mitral-valve insufficiency?III–IV, who underwent a MitraClip? (Abbott Vascular, Abbott Park, IL, USA) procedure. a | The preprocedural LV end-diastolic volume was assessed as 550?ml. b | At 1?year follow-up after the procedure, LV end-diastolic volume was reduced to 480?ml, the patient had improved to NYHA class?II, and residual mitral-valve regurgitation was <1+. Abbreviation: LV, left ventricular.

in similar groups of patients. The NYHA functional class had already improved at discharge in 73% of patients. Reverse remodel? ing was observed to continue at 6?months and 1?year (Figure?1). A randomized trial with larger patient numbers than in the PERMIT trial10 will be needed to confirm these findings and determine the role of the MitraClip? in patients with functional MR. The treatment of valvular heart disease continues to progress with the evolution of percutaneous approaches as alternatives to surgical therapy in selected patients.?To provide patients with the best-possible treatment, short-term effectiveness must be balanced with long-term outcomes.
Division of Cardiovascular Surgery, University?of?Zurich, R?mistrasse?100, Zurich?CH?8091, Switzerland.
Competing interests The author declares associations with following companies: Edwards Lifesciences, Medtronic, Philips, St Jude Medical, and Valtech. See the article online for full details of the relationships. 1. Smith, C.?R. et?al. Transcatheter versus surgical aortic-valve replacement in high-risk patients. N. Engl. J. Med. 364, 2187–2198 (2011). 2. Feldman, T. et?al. Percutaneous repair or surgery for mitral regurgitation. N. Engl. J. Med. 364, 1395–1406 (2011). 3. George, J.?C., Varghese, V., Dangas, G. & Feldman,?T.?E. Percutaneous mitral valve repair: lessons from the EVEREST?II (Endovascular Valve Edge-to-Edge REpair Study) and beyond. JACC Cardiovasc. Interv. 4, 825–827 (2011). 4. Abdel?Wahab, M. et?al. Aortic regurgitation after transcatheter aortic valve implantation: incidence and early outcome. Results from the German transcatheter aortic valve interventions registry. Heart 97, 899–906 (2011). 5. Tamburino, C. et?al. Incidence and predictors of early and late mortality after transcatheter aortic valve implantation in 663 patients with severe aortic stenosis. Circulation 123, 299–308 (2011). 6. Eltchaninoff, H. et?al. Transcatheter aortic valve implantation: early results of the FRANCE (FRench Aortic National CoreValve and Edwards) registry. Eur. Heart J. 32, 191–197 (2011). 7. Gilard, M. et?al. FRANCE?2: FRench Aortic National CoreValve and Edwards Registry view.aspx?eevtid=48&fp=3126 (2011). 8. Moat, N.?E. et?al. Long-term outcomes after transcatheter aortic valve implantation in high-risk patients with severe aortic stenosis: The U.K. TAVI (United Kingdom Transcatheter Aortic Valve Implantation) Registry. J. Am. Coll. Cardiol. 58, 2130–2138 (2011). 9. Mihaljevic, T. et?al. Robotic repair of posterior mitral valve prolapse versus conventional approaches: potential realized. J. Thorac. Cardiovasc. Surg. 141, 72–80 (2011). 10. Auricchio, A. et?al. Correction of mitral regurgitation in nonresponders to cardiac resynchronization therapy by MitraClip improves symptoms and promotes reverse remodeling. J. Am. Coll. Cardiol. 58, 2183–2189 (2011).

(MR; grade?3+ or 4+) were randomly allocated to either surgery or percutaneous edgeto-edge clipping using the MitraClip? device (Abbott Vascular, Abbott Park, IL, USA). Acute procedural success was not achieved with the MitraClip? in 23% of patients, the majority of whom underwent subsequent mitral-valve surgery. The transvenous–? transseptal approach was associated with fewer adverse events at 30?days compared with surgery. The composite end point of major adverse events was reached in 9.6% and 57% of patients in the MitraClip? and surgical groups, respectively. This striking difference was almost exclusively driven by the need for blood transfusion (8% vs 53%, respectively). Noninferiority for the clinicaleffectiveness 1?year end point (freedom from death, mitral-valve surgery, mitral-valve dysfunction >90?days after the index procedure, and MR >2+ at 12?months) was met, despite the fact that 18.5% and only 3% of patients had residual mitral insufficiency >2+ at 1?year in the MitraClip? and surgical groups, respectively. Another 33% of patients in the MitraClip? group had residual mitral insufficiency?=?2+, and more than 15% of patients in the device arm subsequently underwent mitral-valve surgery because of persistent, severe MR. The arbitrary end point of freedom from mitral insufficiency >2+ has triggered lively discussion. Residual mitral insufficiency >1+ after mitral repair is clearly unacceptable, and is an independent predictor of late mortality. If the EVEREST?II investi? gators had chosen presence of mitral insufficiency?<2+ instead of absence of mitral
S10? |? JANUARY 2012

insufficiency?>2+ as part of the effectiveness end point, fewer than half of the patients (47.9%) treated with the MitraClip? would have fulfilled the combined criteria. Statistical noninferiority for effectiveness at 1?year was reached by definition, but the debate whether this result can be regarded as noninferior from a clinical perspective is likely to continue—particularly because the intention-to-treat analysis revealed that the primary composite end points at 2?years were still significantly better with surgery than percutaneous intervention.3 Longterm data are needed before the indication for this therapy can be extended to patients with degenerative mitral-valve disease, in whom the success rate of endoscopic surgical mitral-valve repair is >95%, with excellent short-term and long-term results.9 Mitralvalve repair after previous clip placement is often not possible because the device is encapsulated over time by fibrous tissue; the use of the MitraClip? should, therefore, be restricted to patients who are at high surgical risk. The unconvincing long-term results of downsizing annuloplasty in patients with severe leaflet tethering make the MitraClip? an attractive option in patients with functional MR. In the PERMIT trial,10 51 severely symptomatic nonresponders to cardiac resynchronization therapy with severe left ventricular dysfunction and functional MR underwent treatment with the MitraClip?. Residual MR?≥2+ was present in <20% of patients at discharge, and in only about 10% at 1?year follow-up. Periprocedural mortality was 5.8%, but was not less than surgery


Mutations and non-inferiority analyses show a way forward
Maurie Markman

Highly clinically relevant ovarian cancer clinical research in 2011 focused on an increased understanding of the biology of the malignancy, limitations of strategies for early detection and screening, and the provocative reports of alternative primary and second-line management strategies.
Markman, M. Nat. Rev. Clin. Oncol. 9, 69–70 (2012); published online 20 December 2011; doi:10.1038/nrclinonc.2011.200

Although there were a number of very interesting preliminary reports of thera? peutic advances in ovarian cancer in 2011 (for example, bevacizumab in the first-line and second-line management of the malignancy, and olaparib as maintenance therapy for high-grade serous cancers), as of the writing of this commentary these studies have not appeared in the peer-reviewed oncology literature. Despite the absence of major advances in the realm of treatment, several papers published in 2011 provide highly clinically relevant insight into the management and unique biology of ovarian cancer. Perhaps the most important paper in 2011 was the long-awaited final report from the Prostate, Lung, Colorectal and Ovarian (PLCO) cancer screening randomized controlled trial dealing specifically with ovarian cancer.1 The study, involving 78,000 women with ages ranging from 55 to 74?years, randomly assigned participants to what was classified as ‘usual care’ or a rather-intensive screening strategy that included annual serum CA125 determinations and trans? vaginal ultrasounds. It should be noted that this study did not specifically target women identified as being at ‘high risk’ for the develop?ment of ovarian cancer (for example, those with a family history of ovarian cancer). The screening protocol, undertaken from November 1993 to July 2001, was performed at one of 10 centers in the USA and the median follow-up period for the population was 12.4?years, with patients being followed until death or for a maximum of 13?years. The primary study end point was the rate of mortality from ovarian cancer (including primary peritoneal or fallopian tube cancer), with secondary end points of disease incidence and complications associ? ated with

Table 1 | Influence of BRCA1 or BRCA2 on outcome in high-grade serous ovarian cancer6 Mutational status
BRCA1 mutation BRCA2 mutation BRCA WT

3-year PFS (%)
22 44 16

5-year PFS (%)
13 39 10

HR compared with WT patients
0.81 (P?=?0.44) 0.40 (P?=?0.004) –

3-year OS (%)
64 83 58

5-year OS (%)
44 61 25

HR compared with WT patients
0.76 (P?=?0.35) 0.33 (P?=?0.003) –

Abbreviations: HR, hazard ratio; OS, overall survival; PFS, progression-free survival; WT, wild type.

the screening strategy. The study revealed no difference in deaths from ovarian cancer or other causes between the screening and usual-care groups. Specifically, there were 212 cases of ovarian cancer and 118 deaths from the malignancy in the screening group versus 176 cases and 100 deaths in the routine-care population. Furthermore, and of considerable relevance, 1,080 surgeries were performed among the 3,285 women with false-positive screening tests, and 163 of these individuals developed at least a single ‘serious complication’.1 The important data from this screening trial provide no support for the routine use of annual CA125 determinations or vaginal ultrasounds for completely asymptomatic women as a screening strategy to detect ovarian cancer. However, it is again rele? vant to acknowledge that this trial did not specifically address the issue of screening a more high-risk population, nor did it attempt to define the utility of these tests in the detection of ovarian cancer in indivi?duals presenting with symptoms (for example, several weeks of persistent mild abdominal pain). A study that is somewhat related to the PLCO trial attempted to address the important question of whether an earlier diagnosis of ovarian cancer in a sympto? matic individual might be associated with an improved outcome.2 Australian investigators

retrospectively examined the survival of 1,300 patients with ovarian cancer seen by a physician within varying time intervals (55%, 70% and 92% of the women presented within 1, 2 and 6?months, respectively) from the onset of their initial symptoms. The investigators were unable to find any difference in survival based on the duration of symptoms before the time of diagnosis. Importantly, these data do not support the hypothesis that a somewhat earlier diagnosis (timeline measured in ‘months’) will be associ? ated with superior ovarian cancer-specific survival,3 although the more-timely recognition of the correct diagnosis will likely result in more-rapid initiation of a management plan that will hopefully favorably impact serious symptoms and the individual’s overall quality of life. It is well established that mutations in BRCA1 and BRCA2 are associated with an increased lifetime risk for the development of ovarian cancer. 4 Furthermore, highly provocative data from a number of ovarian cancer investigators have suggested that within the population of women with docu? mented advanced-stage ovarian cancer, patients with BRCA mutations experience an overall superior survival (compared with patients with wild-type BRCA), a difference that is possibly related to increased sensiti? vity to platinum-based chemotherapy.5 In a most-provocative report, investigators
JANUARY 2012? |? S11

Key advances
■■ There are currently no evidence-based data supporting the clinical utility of any ovarian cancer screening strategy in non?high-risk populations1 ■■ Provocative data suggest there may be a clinically meaningful difference between the presence of a BRCA1 or a BRCA2 mutation in influencing outcome in ovarian cancer6 ■■ Under specific circumstances (for example, neuropathy) it might be reasonable to substitute pegylated liposomal doxorubicin for paclitaxel in the front-line chemotherapy management of ovarian cancer7

conducted an observational study involving 316 women with high-grade serous ovarian cancer whose BRCA mutational status was established, to evaluate the impact of such mutations on outcome.6 Although the total sample size was limited (37 and 29 patients with BRCA1 and BRCA2 mutations, respectively), there was a rather striking difference in both the inherent chemosensitivity and survival between the two mutation groups, and compared with women without mutations (Table?1). Patients with BRCA2 mutations experienced a statistically significant improvement in 5?year survival compared with BRCA1 deficient and BRCA ‘wild-type’ cases. Although these results will need to be confirmed by other investigative groups with larger sample sizes, they suggest the potential that these two genetic abnormalities might exert quite different influences on outcome in high-grade serous ovarian cancer. At the present time, there are no known thera?peutic implications associated with these findings, but it is possible that future research may result in recommendations for different management strategies for patients with BRCA1 or BRCA2 mutations. Finally, it is relevant to note an interesting (although unfortunately flawed) phase?III randomized trial that directly compared a regimen of carboplatin plus paclitaxel to carboplatin plus pegylated liposomal doxorubicin as primary treatment for epithelial ovarian cancer.7 Although this potentially paradigm changing study included a total of 820 patients, it was specifically designed to demonstrate the therapeutic superiority (rather than non-inferiority with possibility an improved toxicity profile) of the investi? gative pegylated liposomal doxorubicincontaining regimen. In fact, although the study demonstrated similar median progression-free survival times (19.0?months and 16.8?months) and median overall survival times (61.6?months and 53.2?months)
S12? |? JANUARY 2012

for the carboplatin–pegylated liposomal doxorubicin and carboplatin–paclitaxel study arms, respectively, the trial failed to achieve its primary end point (the documentation of ‘superiority’ for the experimental arm over the established therapy arm). As a result of this failure, the investigators appropriately concluded that the investigative regimen “was not superior to carboplatin– paclitaxel which remains the standard firstline chemo? t herapy for advanced ovarian cancer.”7 However, it is reasonable to suggest that for patients unable to tolerate the taxane (for example, patients who develop neuro? pathy early in the treatment course or severe paclitaxel-associated hypersensitivity reaction) the substitution of pegylated liposomal doxorubicin is not an unreasonable option. Thus, it is reasonable to characterize the ovarian cancer peer-reviewed literature in 2011 as providing modestly useful information regarding disease management and including important discussions of the limitations in our ability to modify the relatively early natural history of the malignancy. It is rather striking that in a clinical research world where unique targets have been discovered in multiple cancers (for example, HER2 overexpression in breast cancer, EGFR mutations in lung cancer, and BRAF mutations in melanoma) leading to exciting new treatment strategies, there remain no such molecular targets in ovarian cancer, the presence of which would lead to specific management paradigms to favorably impact outcome. It is clear that much work needs

to be done to improve our understanding of the fundamental biology of ovarian cancer to change this current state-of-affairs.
Cancer Treatment Centers of America, Eastern Regional Medical Center, 1331 East Wyoming Avenue, Philadelphia, PA 19124, USA.
Competing interests The author declares an association with the following company: Genentech. See the article online for full details of the relationship. 1. Buys, S.?S. et?al. Effect of screening on ovarian cancer mortality: the Prostate, Lung, Colorectal and Ovarian (PLCO) cancer screening randomized controlled trial. JAMA 305, 2295–2303 (2011). Nagle, C.?M. et?al. Reducing time to diagnosis does not improve outcomes for women with symptomatic ovarian cancer: a report from the Australian Ovarian Cancer Study Group. J.?Clin. Oncol. 29, 2253–2258 (2011). Anderson, M.?R. et?al. Combining a symptoms index with CA125 to improve detection of ovarian cancer. Cancer 113, 484–489 (2008). King, M.?C. et?al. Breast and ovarian cancer risks due to inherited mutations in BRCA1 and BRCA2. Science 302, 643–646 (2003). Tan, D.?S. et?al. “BRCA-ness” syndrome in ovarian cancer: a case-control study describing the clinical features and outcome of patients with epithelial ovarian cancer associated with BRCA1 and BRCA2 mutations. J.?Clin. Oncol. 26, 5530–5536 (2008). Yang, D. et?al. Associations of BRCA1 and BRCA2 mutations with survival, chemotherapy sensitivity, and gene mutator phenotype in patients with ovarian cancer. JAMA 306, 1557–1565 (2011). Pignata, S. et?al. Carboplatin plus paclitaxel versus carboplatin plus pegylated liposomal doxorubicin as first-line treatment for patients with ovarian cancer: The MITO?2 randomized phase?III trial. J.?Clin. Oncol. 29, 3628–3635 (2011).







Hitting old targets better and identifying new?targets
Yu Chen and Howard I. Scher

Options to treat late-stage castration-resistant prostate cancer continued to increase in 2011, as three agents with different mechanisms of action prolonged life and a fourth reduced the morbidity of skeletal metastases. These outcomes contrasted with the heightened controversy generated by the recommendation against PSA screening and other early detection?strategies.
Chen, Y. & Scher, H.?I. Nat. Rev. Clin. Oncol. 9, 70–72 (2012); published online 10 January 2012; doi:10.1038/nrclinonc.2011.213

In the Western world, one in six men will be diagnosed with prostate cancer and, of these, one in six will die of metastatic disease. Improving outcomes for men with prostate cancer depends on the one hand on developing more-effective systemic

therapies and, on the other hand, on early diagnosis and treatment, before the disease has?metastasized. 2011 is the 70 th anniversary of when Charles Huggins established that prostate cancer is an androgen-dependent

malignancy and, until recently, docetaxel was the only non-hormonal therapy to prolong life. In 2010, sipuleucel?T and cabazitaxel were shown to confer a survival benefit and were subsequently approved by the FDA. This trend continued in 2011, as three therapies, CYP17 inhibitor abiraterone acetate,1 bone-targeting agent radium?223, 2 and androgen-signaling inhibitor MDV3100,3 were shown to prolong life in definitive phase?III clinical trials (Figure?1), and the RANKL inhibitor, denosumab, was shown to be superior to zoledronic acid in reducing the morbidity associated with bone metastases.4 The demonstration that agents targeting unique aspects of the malignant process associated with tumor cell growth and survival could provide meaningful clinical benefits has highlighted the importance of understanding the biology of castrationresistant prostate cancer (CRPC). The trials established important principles, which will be discussed. The first principle is that CRPCs are not hormone refractory. CRPCs acquire diverse mechanisms to reactivate the androgen receptor signaling pathway in the environment of castrate levels of androgens, including an increase in the androgen biosynthetic machinery and overexpression of androgen receptor. Thus, further decreasing androgen levels by inhibiting steroid synthesis enzymes in the adrenal glands and tumor may be of benefit. CYP17 is a cytochrome P450 enzyme that catalyzes the rate-limiting step of androgen synthesis; abiraterone acetate is a structural analog of the CYP17 substrate pregnenolone that is rationally designed to be a specific and irreversible inhibitor of CYP17. A large international phase?III trial (Cougar AA?301) that compared abiraterone acetate plus prednisone (to prevent mineralocorticoid excess) and placebo plus prednisone in patients with CRPC who had received docetaxel showed an increase in median overall survival from 10.9 to 14.8?months (hazard ratio?=?0.65; P?<0.001).1 All major subgroups of patients benefitted and treatment was well tolerated. A separate trial assessing abiraterone acetate in chemotherapy-naive patients has completed accrual and the results are awaited?(NCT00887198). Another drug supporting the principle that CRPCs are not hormone refractory is MDV3100, a next-generation anti-androgen that was rationally designed to overcome several deficiencies of available agents including modest receptor binding and agonist properties that can promote tumor
Clinical metastasis Non-castrate Rising PSA Castrate Clinical metastases Castrate Pre-docetaxel Sipuleucel-T Clinical metastases Castrate First-line chemotherapy Docetaxel Radium-223 MDV3100 Clinical metastases Castrate Post-docetaxel Cabazitexel Abiraterone

Localized disease

Rising PSA

Non-castrate prostate cancer Castrate prostate cancer Trial drug shown to improve survival in phase III trials

Figure 1 | Clinical states model of prostate cancer progression and systemic therapies shown to improve survival for castrate disease. Blue text represents FDA approved therapies.

growth. The drug displays no androgen receptor activation, blocks nuclear trans? location of the receptor and completely inhibits the ability of androgen receptor to bind to DNA. 3 Promising activity of MDV3100 was demonstrated in a phase?I– II trial in 140 patients, which led to the phase?III AFFIRM trial in patients with chemotherapy-treated CRPC that compared MDV3100 with placebo. 3 This trial was stopped by the Data Safety and Monitoring Board in November 2011 when the first planned interim analysis showed a 37% reduction in mortality, and superior overall survival (median 18.4 versus 13.6?months) in favor of MDV3100.3 The second principle was that targeting the bone environment can confer benefits, independent of an effect on prostate-specific antigen (PSA). Bone is the most common site of prostate cancer spread and is responsible for the highest morbidity burden from the disease. Therapies directed at the bone microenvironment can be divided into three classes: osteoclast inhibitors of bone resorption, radiopharmaceuticals that target bone, and kinase inhibitors. Despite the radiographic appearance of osteoblastic lesions, there is heightened osteoclastic activity and bone turnover in metastatic prostate cancer lesions. Osteoclast inhibitors, including zoledronic acid, are used as an adjuvant therapy to maintain bone density and to decrease skeletal-related events (SRE) in patients with CRPC. Denosumab is a humanized monoclonal antibody that blocks RANKL, which is required for osteoclast activation and survival. In 1,904 patients with CRPC that had meta? stasized to the bone, denosumab delayed the median time to first SRE from 17.1 to 20.7?months (P?=?0.008) compared to zoledronic acid, leading to FDA approval.4 The beta-emitting bone-seeking radioisotopes strontium?89 and samarium?153 lexidronam are approved for the palliation of pain, but neither has been shown to prolong life. Radium?223 is a unique bone-directed radioisotope that emits high-energy alpha

particles that penetrate only several cell layers in tissue, significantly improving the radiation delivery to areas of sclero? tic bone where the tumor resides while minimizing radiation to hematopoietic tissue. In a phase?III trial in 922 patients with CRPC and bone pain, 6-monthly doses of radium?223 conferred a significant overall survival benefit over placebo (14.0 versus 11.2?months; P?=?0.00185);2 it is the first bone-directed agent to prolong overall?survival. Multiple signaling cascades are critical for bone homeostasis and turnover. Two promising investigational agents targeting this axis deserve mention. Dasatinib inhibits kinases including Src family tyrosine kinases that are important for bone turnover; a single-agent phase?II trial in patients with CRPC showed a marked decrease in bone turnover, disease stabilization5 and additive effects with docetaxel. A phase?III trial comparing dasatinib with placebo in combination with docetaxel has completed accrual. Cabozantinib inhibits kinases including Ret, Met, and VEGFR2; in a randomized phase?III discontinuation trial in patients with CRPC, treatment with cabozantinib was associated with normalization of radionuclide bone scans in 86% of patients, improvement of bone pain in 64% of patients, decline in bone marker levels, improvement of anemia, and a marked reduction in bone pain independent of an effect on PSA.6 Two phase?III trials will be opened for recruitment soon: the first for an indication for the palliation of pain and a second for an improvement in overall survival in patients with CRPC. Limiting the use of agents shown to prolong life to men with advanced metastatic CRPC will not considerably decrease the number of men who die from prostate cancer. To do so requires the identification and treatment of those cancers that are destined to metastasize, produce symptoms and ultimately shorten a patient’s anticipated life expectancy when tumor burdens
JANUARY 2012? |? S13

Key advances
■■ Abiraterone acetate and MDV3100 prolong overall survival in patients with docetaxel-treated castration-resistant prostate cancer (CRPC)1,3 ■■ Radium?223 prolongs overall survival in patients with CRPC, bone pain from disease, and who are docetaxel-treated or ineligible for docetaxel treatment2 ■■ Denosumab delays skeletal-related events compared with zoledronic acid in patients with CRPC and bone disease4

are minimal. With this goal but without firm evidence, the advent of the PSA test led to widespread screening that has been accompanied by a marked increase in men diagnosed with localized disease who undergo radical prostatectomy and definitive radiotherapy with the associated morbidities. Over the past 2?years, the US PLCO trial showed no reduction in prostate cancer-specific mortality (PCSM) at 7?years after the initiation of PSA screening; the European Randomized Study of Screening for Prostate Cancer showed a 20% decrease in PCSM at 9?years after the initiation of PSA screening; and the G?teborg randomized screening trial showed a 44% decrease in PCSM at 14?years after the initiation of screening.7–9 These results, along with an analysis of other early detection studies led the USPSTF to recommend against PSA screening because of “moderate or high certainty that the service has no net benefit or that the harms outweigh the bene?fits.”10 The recommendation triggered a firestorm of debate, proponents of screening highlighting flaws in methodology and interpretation and opponents the limited benefits, anxiety and morbidity a diagnosis produces, and the high societal costs. There is strong evidence prostatectomy and radiotherapy can decrease PCSM by approximately 50% in clinically localized disease. However, in low-risk disease where the 15-year PCSM is <2%, these benefits become tiny, especially in those with limited longevity. Despite this, more than half of the men treated with prostatectomy and radiotherapy in the USA have low-risk disease. It is important to note that the harms of screening come not from the PSA test, but from the chain of actions following PSA measurement. It has been ingrained into the psyche of physicians and patients that all cancers are rapidly lethal unless treated aggressively. The USPSTF recommendation is an appropriate pushback against nondiscriminate screening and subsequent
S14? |? JANUARY 2012

treatment and argues that early detection strategies focus on diagnosing cancers that need treatment to reduce morbidity and mortality, and avoid finding highly prevalent indolent cancers for which any treatment is overtreatment. The two overarching objectives in the field are to decrease death and suffering but, at the same time, decrease those who are subjected to unnecessary treatment that can produce significant morbidity. It is clear that neither our current practice pattern nor the blanket discontinuation of PSA screening can adequately address these goals. The benefit-to-harm ratio can be shifted if both screening and treatment are targeted to high-risk populations and surveillance becomes more widely adopted for non-aggressive localized tumors.
Human Oncology and Pathogenesis Program and Department of Medicine, Memorial Sloan– Kettering Cancer Center, 1275 York Avenue, New York, NY 10065, USA (Y. Chen, H.?I. Scher). Correspondence to: H.?I. Scher
Competing interests H.?I. Scher declares associations with the following companies: Amgen, Dendreon, Exelixis, Medivation, Ortho Biotech Oncology Research & Development, Sanofi-Aventis. See the article online for full details of the relationships. Y. Chen declares no competing interests. 1. de Bono, J.?S. et?al. Abiraterone and increased survival in metastatic prostate cancer. N. Engl. J. Med. 364, 1995–2005 (2011).

Parker, C. et?al. Overall survival benefit of radium?223 chloride (Alpharadin?) in the treatment of patients with symptomatic bone metastases in castration-resistant prostate cancer (CRPC): a phase?III randomized trial (ALSYMPCA) [abstract]. Eur. J. Cancer 47, a7003 (2011). 3. Medivation. Medivation and Astellas announce positive survival data from interim analysis of phase?3 AFFIRM trial of MDV3100 in men with advanced prostate cancer [online], http:// releaseid=620500 (2011). 4. Fizazi, K. et?al. Denosumab versus zoledronic acid for treatment of bone metastases in men with castration-resistant prostate cancer: a randomised, double-blind study. Lancet 377, 813–822 (2011). 5. Yu, E.?Y. et?al. Phase?II study of dasatinib in patients with metastatic castration-resistant prostate cancer. Clin. Cancer Res. 15, 7421–7428 (2009). 6. Hussain, M. et?al. Cabozantinib (XL184) in metastatic castration-resistant prostate cancer (mCRPC): Results from a phase?II randomized discontinuation trial [abstract]. J. Clin. Oncol. 29?(Suppl.), a4516 (2011). 7. Hugosson, J. et?al. Mortality results from the Goteborg randomised population-based prostate-cancer screening trial. Lancet Oncol. 11, 725–732 (2010). 8. Schr?der, F.?H. et?al. Screening and prostatecancer mortality in a randomized European study. N. Engl. J. Med. 360, 1320–1328 (2009). 9. Andriole, G.?L. et?al. Mortality results from a randomized prostate-cancer screening trial. N.?Engl. J. Med. 360, 1310–1319 (2009). 10. Chou, R. et?al. Screening for prostate cancer: a review of the evidence for the U.?S. Preventive Services Task Force. Ann. Intern. Med. 155, 762–771 (2011).



New therapeutic targets and treatment strategies
Paula Cramer and Michael Hallek

2011 saw improvements in our understanding of B?cell malignancies: insights into the genomic basis of chronic lymphocytic leukemia were achieved; reduced treatment intensity caused fewer toxic effects in early-stage Hodgkin lymphoma; first-line rituximab maintenance therapy improved outcome in follicular lymphoma; and selected patients with diffuse large-cell lymphoma benefited from the addition of bortezomib.
Cramer, P . & Hallek, M. Nat. Rev. Clin. Oncol. 9, 72–74 (2012); published online 10 January 2012; doi:10.1038/nrclinonc.2011.212

Chronic lymphocytic leukemia (CLL) is a clonal B?cell disorder initiated by genetic aberrations in B?cells that increase resistance to apoptosis. The survival of CLL cells is strongly dependent on inter? actions with surrounding cells (macro? phages and dendritic cells), biochemical cues (chemokines) and specific receptors expressed on CLL cells. 2011 saw publication of three important papers providing new insights into

the pathogenesis of CLL. Kikushige et?al.1 reported that the capacity to generate clonal B?cells is acquired at the hematopoietic stem cell (HSC) stage in CLL, suggesting multi? potent, self-renewing HSCs are involved in the initiation of CLL. These findings can lead to the development of new therapies that target these progenitor CLL cells. The study by Puente et?al.2 reported the first data on whole-genome sequencing

in CLL. Four cases were analyzed and 46 somatic mutations were identified.2 Four genes with recurrent mutations were confirmed in 363 patients with CLL: NOTCH1, XPO1, MYD88 and KLHL6. Mutations in MYD88 and KLHL6 were predominantly found in patients with CLL who had a high number of somatic hypermutations in the variable region of IGHV, whereas mutations in NOTCH1 and XPO1 were mostly detected in patients who did not have IGHV mutations. These findings indicate a role for mutations in NOTCH1 , MYD88 and XPO1 in the clinical evolution of CLL. In a similar study on 91 patients with CLL, Wang et al.3 performed massively parallel sequencing of 88 whole exomes and genomes. They found nine genes that are mutated at significant frequencies, including four with established roles (TP53 in 15% of patients, ATM in 9%, MYD88 in 10%, and NOTCH1 in 4%) and five genes without established roles (SF3B1, ZMYM3, MAPK1, FBXW7, and DDX3X). SF3B1, which functions at the catalytic core of the spliceosome, was the second most frequently mutated gene (15% of patients). SF3B1 mutations occurred primarily in tumors with deletions in chromosome 11q, which are associated with a poor prognosis. The authors also showed that mutations in SF3B1 induced alterations in pre–mRNA splicing. All three studies have considerable clinical implications for CLL. Firstly, the stem cell origin of CLL offers a plausible explanation for the fact that conventional chemo? immunotherapies regularly fail to cure CLL. This type of therapy may not effectively eradi?cate the HSC pool that regenerates CLL cells. Secondly, our understanding of the genetic basis of CLL has improved. These advances will facilitate the development of better-targeted therapies for this disease. In the past 10?years, a remarkable improvement in overall survival was achieved for several types of B?cell lymphomas. This improvement was largely a result of the use of rituximab, an antibody targeting CD20. In follicular lymphoma, several randomized trials have demonstrated that addition of rituximab to different chemotherapeutic regimens can prolong survival; therefore, rituximab-based chemoimmuno? therapy is the current standard first-line treatment. Disease progression usually occurs 3–5 years after initial therapy and rituximab maintenance therapy was previously shown to have a clinical benefit in patients with relapsed disease. The PRIMA study 4 evaluated the benefit of rituximab maintenance therapy

after first-line treatment for follicular lymphoma. In this study, 1,019 patients who achieved a response after different chemo? immunotherapies (rituximab plus cyclophosphamide, vincristine, doxo? r ubicin and prednisone [R-CHOP]; rituximab plus cyclophosphamide, vincristine and pred? nisone; or rituximab plus fludarabine, cyclo? phosphamide and mitoxantrone) were randomly assigned to receive 2?years of maintenance therapy with rituximab (375?mg/m? every 8?weeks) or no further therapy (observation). The maintenance treatment was well tolerated. Infectious events were more common in the maintenance arm than the observation arm, but were mostly mild or moderate and only a few patients withdrew from the study because of toxicity. No signifi? cant decrease in immunoglobulin levels was observed and patient quality of life was not affected by the repeated rituximab infusions. Maintenance therapy improved the quality of response and prolonged progression-free survival and event-free survival. No effect on overall survival was shown for rituximab maintenance therapy, which might be explained by a short follow-up period and the efficacy of salvage therapies. However, rituximab should be considered as maintenance therapy after first-line chemo? immunotherapy in follicular lymphoma, as the duration of response to first-line therapy has prognostic value.5 The outstanding outcome observed in patients with Hodgkin lymphoma allowed researchers to consider a reduction in treatment intensity to reduce both acute and long-term adverse events, such as infertility and secondary neoplasias. The
Key advances
■■ Genomic analyses of patients with chronic lymphocytic leukemia identified mutations in genes such as Notch1 and SF3B1 that have prognostic impact2,3 ■■ Rituximab maintenance therapy after first-line treatment with rituximab-based chemoimmunotherapy significantly improves the quality of response and prolongs progression-free survival in patients with follicular lymphoma4 ■■ In advanced-stage, high-risk Hodgkin lymphoma, reduction in treatment intensity has not improved outcome and further analyses are needed to define the optimal treatment intensity7 ■■ Patients with diffuse large B?cell lymphoma, in particular, those with a non-germinal center B?cell-like subtype, benefit from the addition of bortezomib to first-line chemoimmunotherapy9

HD10 trial of the German Hodgkin Study Group (GHSG)6 showed that two cycles of ABVD (doxo? rubicin, bleomycin, vinblastine and dacarbazine) therapy followed by 20?Gy of involved–field radio? t herapy was as efficacious as four cycles of ABVD and 30?Gy of involved-field radiotherapy in patients with stage I or II disease and favorable risk factors. Reduced treatment intensity may be considered as the new standard treatment for early-stage, lowrisk Hodgkin lymphoma. By contrast, the reduction of treatment intensity is debated in advanced-stage Hodgkin lymphoma (stage II–IV or IIB with extranodal lesions or mediastinal?mass). The results of the HD12 trial by Borchmann et?al.7 were slightly disappointing, because the reduction of the intensity of chemotherapy from eight cycles of high-dose BEACOPP (bleomycin, etoposide, doxo? rubicin, cyclophosphamide, vincristine, procarbazine and prednisolone) to four cycles of high-dose BEACOPP followed by four cycles of BEACOPP at the baseline dose did not reduce the severity of toxicity or treatment-related mortality, but could possibly reduce efficacy. Thus, the GHSG considers eight cycles of high-dose BEACOPP as the standard treatment for advancedstage Hodgkin lymphoma and is evaluating other individualized, PET-controlled dose reduction strategies in ongoing trials. Viviani et?al.8 compared BEACOPP and ABVD in advanced-stage Hodgkin lymphoma. In this trial, 331 patients received either four cycles of high-dose BEACOPP plus four cycles of BEACOPP at baseline dose or six to eight cycles of ABVD; each treatment was followed by involved-field radiotherapy. The estimated 7?year rate of freedom from first progression was 85% in patients treated with BEACOPP and 73% in patients treated with ABVD. However, all patients with a relapse or less than complete response after initial therapy were treated with multiple cycles of ifosfamidecontaining chemotherapy followed by
JANUARY 2012? |? S15

? Eraxion |

high-dose chemo? t herapy with carmustine, etoposide, cytarabine, and melph? alan plus autologous hemato? p oietic stem-cell support. After completion of treatment including the salvage therapy, the 7?year rate of overall survival was 89% in patients initially treated with BEACOPP compared to 84% in patients who had received ABVD. As this difference in overall survival was not significant, the authors concluded that treatment with BEACOPP resulted in a more effective disease control, but not in a better long-term outcome, compared with ABVD. They argued that while a major proportion of patients could be cured by the initial BEACOPP therapy, this clinical benefit might be neutralized by a high rate of longterm adverse events, in particular myelo? toxicity, infections and secondary neoplasias. Viviani et?al.8 deduced that the BEACOPP regimen presents an over? treatment strategy for a large number of patients and recommended ABVD for patients with advanced-stage or unfavorable Hodgkin lymphoma. This study has triggered an intense discussion about the need for an intensive BEACOPP therapy in advancedstage Hodgkin lymphoma, as few patients who relapse can be rescued with a high-dose salvage therapy. However, this trial8 was not powered to demonstrate a survival difference between the two treatments and the 7?year overall survival differed by 5% in favor of the BEACOPP regimen. Moreover, median observation time (61? months) and the number of patients were low. As a result, an appropriate treatment intensity in advancedstage or unfavorable Hodgkin lymphoma is still debated. The improvements in our understanding of the molecular pathogenesis of non-Hodgkin lymphoma have led to the develop? ment of novel therapeutic agents that interfere with signaling pathways involved in lympho? magenesis. For instance, the addition of the proteasome inhibitor bortezomib enhanced the activity of chemo? immunotherapy with R?CHOP as first-line treatment for diffuse large B?cell lymphoma (DLBCL) and mantlecell lymphoma.9 Ruan et?al.9 demon?strated that, in particular, patients with a nongerminal center B?cell-like (GCB) subtype, which is usually associ? ated with a worse outcome, benefit from the addition of borte? zomib. This increased clinical benefit can be explained by the fact that the non-GCB subtype is characterized by a deletion in a tumor-suppressor gene that activates NF-κB and this activation is blocked by bortezomib. Thus, a worse prognosis in this subgroup of
S16? |? JANUARY 2012

patients with DLBCL could be overcome by bortezomib treatment. Overall, 2011 saw exciting developments in the understanding and treatment of B?cell lymphomas. Whereas research in Hodgkin lymphoma aims to reduce the adverse effects of first-line treatment, in most other lymphomas more-effective therapies are needed. The growing understanding of the pathogenesis of these lymphomas will facilitate the develop?ment of novel therapeutic agents.
Department of Internal Medicine I and Center for Integrated Oncology K?ln-Bonn, University of Cologne, Joseph-Stelzmann-Stra?e. 9, 50924 Cologne, Germany (P . Cramer, M.?Hallek). Correspondence to: M. Hallek
Competing interests M. Hallek declares associations with the following companies: Celgene, Mundipharma, Roche. See the article online for full details of the relationships. P .?Cramer declares no competing interests. 1. Kikushige, Y. et?al. Self-renewing hematopoietic stem cell is the primary target in pathogenesis of human chronic lymphocytic leukemia. Cancer Cell 20, 246–259 (2011). Puente, X.?S. et?al. Whole-genome sequencing identifies recurrent mutations in chronic lymphocytic leukaemia. Nature 475, 101–105 (2011).









Wang, L. et al. SF3B1 and other novel cancer genes in chronic lymphocytic leukemia. N. Engl. J. Med. NEJMoa1109016. Salles, G. et?al. Rituximab maintenance for 2?years in patients with high tumor burden follicular lymphoma responding to rituximab plus chemotherapy (PRIMA): a phase?3, randomised controlled trial. Lancet 377, 42–51 (2011). Bachy, E. et?al. Long-term follow-up of patients with newly diagnosed follicular lymphoma in the prerituximab era: effect of response quality on survival - a study from the Groupe d’Etude des Lymphomes de l’Adulte. J. Clin. Oncol. 28, 822–829 (2010). Engert, A. et?al. Reduced treatment intensity in patients with early-stage Hodgkin’s lymphoma N.?Engl. J. Med. 363, 640–652 (2010). Borchmann, P . et?al. Eight cycles of escalateddose BEACOPP compared with four cycles of escalated-dose BEACOPP followed by four cycles of baseline-dose BEACOPP with or without radiotherapy in patients with advancedstage Hodgkin’s lymphoma: final analysis of the HD12 trial of the German Hodgkin Study Group. J. Clin. Oncol. 29, 4234–4242 (2011). Viviani, S. et?al. ABVD versus BEACOPP for Hodgkin’s lymphoma when high-dose salvage is planned. N. Engl. J. Med. 365, 203–212 (2011). Ruan, J. et?al. Bortezomib plus CHOP-rituximab for previously untreated diffuse large B?cell lymphoma and mantle cell lymphoma. J. Clin. Oncol. 29, 690–697 (2011).


A new paradigm tumor for drug development
Alexander M. M. Eggermont and Caroline Robert

Melanoma has emerged as the paradigm tumor for drug development through mutation-targeted therapies (inhibitors targeting BRAF, MEK, and c?KIT) and immunotherapy. Exploring the combinations of both approaches is a challenge that will require scientific rationale and the cooperation of the pharmaceutical industry. But, with these challenges comes another opportunity to change the paradigms in drug development.
Eggermont, A. M. M. & Robert, C. Nat. Rev. Clin. Oncol. 9, 74–76 (2012); published online 10 January 2012; doi:10.1038/nrclinonc.2011.201

For the past 40?years no treatments developed for melanoma have significantly improved survival over dacarbazine, a drug with a response rate of around 10%. In the past decade, we have witnessed—in the fields of mutation-driven drug development and immuno? modulation—the establishment of treatments actively impacting survival. This is just the beginning of a promising journey to develop rational and biology-driven treatments for patients with melanoma with clini? cally significant impact to greatly change the thera?peutic landscape. The treatment of melanoma has evolved from using non-selective inhibitors to more

selective agents, with BRAF, MEK and c?KIT inhibitors leading the field. BRAF mutations are the most frequently occurring and most important mutations that provide drugable targets in melanoma. The first selective BRAF inhibitor developed in the clinical setting was vemurafenib. 1 In a randomized phase?III trial published in 2011, this drug led to major tumor responses in 50% of patients, and minor responses in >30% of patients.2 The most common adverse events were arthralgia, fatigue, and cutaneous mani?festations such as rash, photo?sensitivity, pruritis, and squamous-cell carcinoma of the keratoacanthoma-type in around 20%

of patients.1,2 The phase?III BRIM?3 trial, in which vemurafenib was compared with dacarbazine in treatment-naive patients, showed an improvement in progressionfree survival (PFS) of 5.3?months for vemurafenib. The impact on overall survival has not yet been established and is probably similar because, as the trial was unblinded at the first interim analysis and cross over was allowed, changes in overall survival will be somewhat compromised by this cross over. Responses to vemurafenib are immedi? ate and evident on PET scans within 1–2?weeks after treatment commencement. Disappointing, however, is the short median duration of these responses, and essentially all patients relapse, 50% in the first 5–6?months, 50% any time after 6?months. Various mechanisms of drug resistance have been described, mostly reflecting primary resistance (present from the beginning) and heterogeneity of the tumors. As BRAF inhibition leads to reactivation of the MAPK pathway and enhances cell growth through CRAF, combining a BRAF inhibitor with a MEK inhibitor may both prolong PFS and prevent the appearance of squamous-cell carcinomas. The report in 2011 of the use of a BRAF inhibitor (GSK436) combined with a MEK inhibitor (GSK212), clearly indicate these beneficial effects.3 NRAS mutations occur in about 20% of melanomas but RAS remains a rather elusive target in cancer, with no available drugs that can directly antagonize its signaling activity. Dual targeting of the MAPK and PI3K pathways might abrogate the effect of NRAS mutation (Figure?1). Mutations in c?KIT are also important in melanoma, especially in mucosal and acral melanomas, in which mutations in c?KIT are found in 20–30% of the cases. These types of melanoma represent less than 20% of all melanomas in the white population in the Western world, but incidence in the
Key advances
■■ BRAF and MEK inhibitors are key agents in treating BRAF-mutated melanomas, and their combination is essential to optimize results2,3 ■■ Immunomodulators, such as antibodies blocking CTLA?4 and PD?1, establish immunotherapy as a key component of anti-tumor strategies, required for long-term responses and potential cure6 ■■ Ulcerated primary melanoma is a distinct biologic entity that indicates sensitivity to adjuvant IFN therapy determining when such treatment should be indicated10

Asian population is >70%.4 Response rates and PFS rates with imatinib are in the range of 15–20%, which is rather disappointing compared to the activity of this inhibitor in GIST tumors, and a reflection of a different pattern of mutations.4 These mutations make melanoma a tumor that often presents with multiple perturbed pathways and with many routes to escape from single or multiple drug exposure. Therefore, although combinations of drugs targeting two proteins in the same signaling pathway or in different pathways are currently being explored, it will be critical to combine targeted drugs with drugs that modulate the immune system. The cytotoxic T?lymphocyte–associated antigen 4 (CTLA?4) ligand is a negative regulator of T?cells; therefore, blocking CTLA?4 action augments T?cell activation and proliferation, and inhibits immune system tolerance. The anti-CTLA?4 monoclonal antibody ipilimumab has recently been assessed in randomized phase?III trials. Improved overall survival (4?months) was demonstrated for ipilimumab alone or in combination with the gp100 peptide vaccine compared with the vaccine alone in patients who did not respond to prior treatment.5 Improved overall survival was also demonstrated for ipilimumab in the firstline setting when combined with dacarbazine (2.1?months).6 Thus, in 2011, the FDA approved ipilimumab for advanced melanoma at a dose of 3?mg/kg, administered every 3?weeks for a total of four doses for all patients with advanced melanoma in the first-line and second-line and settings. In the pivotal trial assessing the combination of dacarbazine with ipilimumab as first-line treatment at a dose of 10?mg/kg, a high percentage of liver enzyme alterations was observed.6 This adverse effect resulted in significant percentage of patients abandoning the treatment, which is believed to have reduced the impact of ipilimumab. Because ipilimumab reactivates immune responses that have been silenced, there is a risk that it may reactivate lingering subclinical auto? immune responses in certain patients, resulting in immune-related adverse events. Some of the most prominent were colitis, dermatitis, and hypo? physitis, all of which must be closely monitored as patients may need the immediate use of corticosteroids. Ipilimumab has a modest response rate of around 10%, but responses are almost always durable with a median of 20?months. Moreover, there was a clear separation of the survival curves in both trials, indicating durable responses that last


PI3K pathway PI3K


MAPK pathway BRAF GSK436


AKT Inhibition mTOR


Cell growth, proliferation, survival

Figure 1 | Dual targeting of the MAPK and PI3K pathways in melanoma treatment. This figure illustrates how combining inhibition of BRAF and MEK (intra-pathway combined blocking) with AKT (inter-pathway combined?blocking) might abrogate the effect of NRAS?mutation.

for >3?years—with a significant number of patients that have not relapsed after 5?years—which may indicate potential cure. A reason why patients might have stopped taking the treatment is that it takes time for the responses to ipilimumab to develop. Responses may even be preceded by progression of disease at 6 or 8?weeks, which might reflect lymphocyte infiltration of the lesions that will subsequently regress. These observations have led to the development of novel, immune-related response criteria that might more accurately describe response to immunotherapy and avoid premature treatment cessation in patients with disease progression before response.7 Studies that combine BRAF inhibitors with cytokines such as IFN, IL2, GM-CSF and anti-angiogenic agents are currently under clinical development. Although early reports indicate improved response rates to these combinations, the interactions between these drugs are poorly understood, and in the absence of clear biomarkers to provide rational guidance as to which combinations should be used and how, serious efforts should be made in this area to allow for rational plans and progress. Green et al.8 have proposed that various chemotherapeutics may exert their action through immunogenic cell death (leading to antigen presentation and stimulating the immune system) or tolerogenic cell death (necrosis without immune-system priming but instead leading to immuno? logical tolerance).9 This approach could provide rational guidance to explore combination treatments.
JANUARY 2012? |? S17


New immunomodulators are being tested in early-phase studies. Most promising are the results seen with the monoclonal antibody that acts against the programmed death?1 receptor (PD-1R), the ligand of which (PD-1L) can be expressed directly on melanoma cells. Good responses have been observed in phase?I studies with possibly fewer accompanying immune-related adverse events than observed with ipili? mumab treatment. 9 Another immuno? modulator, pegylated IFN?α-2b, was approved in 2011?by the FDA as adjuvant therapy for patients with stage III lymphnode positive disease after lymph-node dissection, on the basis of a significant improvement of relapse-free survival (RFS).10 This EORTC 18991 trial (n?=?1,256 patients) had been preceded by the EORTC 18952 trial, which evaluated intermediate doses of regular IFN?α-2b in 1,388 patients.10 In both trials, patients were stratified according to nodal status and presence or absence of ulceration in the primary tumor. In both trials, low tumor-stage and the presence of ulceration significantly correlated with improved outcome in the IFN-treatment arms, as reported by the recently published meta-analysis of these trials.10 In patients with an ulcerated primary tumor and stage IIB or limited stage III disease (positive sentinel node) there was a highly significant reduction (30–40%) of the relative risk of RFS, distant metastasis-free survival and overall survival. In patients with stage?III–N2 disease, these benefits were progressively lost. This finding strongly indicates that ulcerated melanoma represents a distinct biologic entity that will be assesed in the EORTC 18081 trial, which will evaluate pegylated-IFN-α-2b versus observation in node-negative patients with ulcerated primary melanomas. Adjuvant treatment with ipilimumab in patients with advanced stage IIIb/IIIc is under evalu? ation in the EORTC 18071 trial that has now completed patient accrual. There has never been a more exciting and dynamic time in the treatment of melanoma. To explore all potential combinations is impossible. Thus, we need to stick to a rational approach—guided by principles such as immunogenic cell death—and translational research programs to identify biomarkers predictive for outcomes of both the immuno? therapeutic and the mutation-driven drug development approach. Without such an approach we could revert to the empiric ways that characterized clinical drug development of the past, and that is no longer acceptable in the 21st century.
S18? |? JANUARY 2012

Institut de Cancérologie Gustave Roussy, 114 Rue Edouard Vaillant, 94805 Villejuif, ParisSud, France (A. M. M. Eggermont, C. Robert) Correspondence to: A. M. M. Eggermont
Competing interests The authors declare associations with the following companies: Bristol-Myers Squibb, GlaxoSmithKline, Merck, Roche. See the article online for full details of the relationships. 1. Flaherty, K.?T. et?al. Inhibition of mutated, activated BRAF in metastatic melanoma. N.?Engl. J. Med. 363, 809–819 (2010). Chapman, P .?B. et?al. Improved survival with vemurafenib in melanoma with BRAF V600E mutation. N. Engl. J. Med. 364, 2507–2516 (2011). Infante, J.?R. et?al. Phase?I/II study to assess safety, pharmacokinetics, and efficacy of the oral MEK 1/2 inhibitor GSK1120212 (GSK212) dosed in combination with the oral BRAF inhibitor GSK2118436 (GSK436) [abstract]. J.?Clin. Oncol. 29 (Suppl.), CRA8503J (2011).




Carvajal, R.?D. et?al. KIT as a therapeutic target in metastatic melanoma. JAMA. 305, 2327–2334 (2011). 5. Hodi, F.?S. et?al. Improved survival with ipilimumab in patients with metastatic melanoma. N. Engl. J. Med. 363, 711–723 (2010). 6. Robert, C.? et?al. Ipilimumab plus dacarbazine for previously untreated metastatic melanoma. N. Engl. J. Med. 364, 2517–2526 (2011). 7. Hoos, A.? et?al. Improved endpoints for cancer immunotherapy trials. J. Natl Cancer Inst. 102, 1388–1397 (2010). 8. Green, D.?R., Ferguson, T., Zitvogel, L. & Kroemer, G. Immunogenic and tolerogenic cell death. Nat. Rev. Immunol. 9, 353–363 (2009). 9. Brahmer, J.?R. et?al. Phase?I study of single-agent anti-programmed death?1 (MDX?1106) in refractory solid tumors: safety, clinical activity, pharmacodynamics, and immunologic correlates. J. Clin. Oncol. 28, 3167–3175 (2010). 10. Eggermont, A.?M. et?al. Ulceration and stage are predictive of interferon efficacy in melanoma: results of the phase?III adjuvant trials EORTC 18952 and EORTC 18991. Eur. J. Cancer. doi:10.1016/j.ejca.2011.09.028?


Prevention and treatment of bone metastases
Robert E. Coleman

Bone-targeting treatments have transformed the quality of life of patients with metastatic bone disease. 2011 saw the emergence of denosumab—a RANK ligand-specific antibody—as a more-effective alternative treatment to bisphosphonates and of data on the use of bone-targeting treatments to prevent metastasis from breast and prostate cancers.
Coleman, R.?E. Nat. Rev. Clin. Oncol. 9, 76–78 (2012); published online 20 December 2011; doi:10.1038/nrclinonc.2011.198

Bone disease accounts for substantial morbidity and mortality in patients with cancer. Advanced-stage solid tumors, notably those arising from the breast and prostate, and multiple myeloma are associated with a heavy burden of skeletal disease, with potentially debilitating or life-limiting skeletal-related events (SREs). As a result, bonetargeting treatments have been introduced alongside specific anticancer treatments to minimize skeletal morbidity and preserve the patients’ quality of life and physical function. An increased understanding of the interplay among the disseminated tumor cells, the bone marrow and both hostderived and tumor-derived growth factors has shown that the complex interactions in the bone marrow micro?environment provide candidate therapeutic targets. Modification of these inter? a ctions is emerging as an important anticancer treatment strategy and, in 2011, considerable developments have

Key advances
■■ Denosumab provides a more effective, convenient and well-tolerated alternative treatment to zoledronic acid for the prevention of skeletal morbidity2,3 ■■ The adjuvant use of zoledronic acid in early-stage breast cancer should be restricted to postmenopausal women, as women with residual ovarian function do not benefit from this treatment7 ■■ Denosumab delays the development of bone metastases in men with castration?resistant prostate cancer10

occurred in the treatment and prevention of bone metastases. In the past 20?years, bisphosphonates have been the choice of treatment for preventing skeletal morbidity associated with malignant bone disease. However, recognition of the importance of the receptor activator of nuclear factor κ?B ligand (RANKL) in bone

metastasis has provided a new specific therapeutic target. Denosumab is a fully human, synthetic IgG2 antibody that is administered via subcutaneous injection and binds with high affinity to RANKL, preventing the interaction of RANKL and RANK and inhibiting osteoclast function. In 2011, denosumab became available for clinical use after the publication of three large randomized double-blind registration studies. These trials included 5,723 patients with bone metastases secondary to breast cancer (n ?=?2,046), 1 castration-resistant prostate cancer (n ?=?1,901) 2 and other advanced-stage solid tumors including non-small-cell lung cancer (n?=?702), other types of solid tumors (n?=?894) and myeloma (n?=?180).3 In these studies, patients were randomly assigned to receive denosumab or zoledronic acid every 3–4?weeks, plus calcium and vitamin D supplements. The primary end point was time to first onstudy SRE. Stopeck and colleagues1 showed that denosumab was statistically superior to zoledronic acid in delaying the first SRE in patients with bone meta? stasis secondary to breast cancer; the median time to first on-study SRE was 26.4?months in the zoledronic acid-treated group and has not been reached in the denosumab-treated group. Similarly, Fizazi and colleagues2 reported that denosumab was superior to zoledronic acid in delaying the first SRE in patients with bone meta? stasis secondary to prostate cancer; the median time to first on-study SRE was 17.1?months in the zoledronic acid-treated group and 20.7?months in the denosumab-treated group. In patients with bone metastasis secondary to advanced solid tumors and myeloma, Henry and colleagues3 showed that denosumab was not inferior to zoledronic acid in delaying the first SRE, but was not superior either. In an analysis of multiple events of SREs, denosumab was superior to zoledronic acid in patients with breast cancer 1 and prostate cancer 2 (Table?1). In all three studies, denosumab was well tolerated, and a key difference in the safety profiles of denosumab and zoledronic acid was the lack of an effect on kidney function with denosumab, which obviates the need for monitoring renal function that is mandated when treating patients with intravenous bisphosphonates. Similar to the use of intravenous bisphosphonates used in oncology settings, osteonecrosis of the jaw (ONJ) was the most important adverse event associated with denosumab, but it was uncommon. The cumulative incidence rates of ONJ were

Table 1 | Phase?III registration trials of denosumab in advanced-stage cancer Disease setting
Breast cancer1 CRPC2 OST and MM3 Total4

End point
Time to 1st SRE Time to 1st and subsequent SRE Time to 1st SRE Time to 1st and subsequent SRE Time to 1st SRE Time to 1st and subsequent SRE Time to 1st SRE Time to 1st and subsequent SRE

Hazard ratio
0.82 0.77 0.82 0.82 0.84 0.90 0.83 0.82

95% CI
0.71–0.95 0.66–0.89 0.71–0.95 0.71–0.94 0.71–0.98 0.77–1.04 0.76–0.90 0.77–1.04

P (relative to ZA)
P?=?0.01 P?=?0.001 P?=?0.008 P?=?0.008 P?=?0.03 P?=?0.14 P?<0.001 P?<0.001

Abbreviations: CI, confidence interval; CRPC, castration-resistant prostate cancer; MM, multiple myeloma; OST, solid tumors other than breast or prostate cancer; SRE, skeletal-related event; ZA, zoledronic acid.

similar for both treatments: 0.5% and 0.8% in year 1, 1% and 1.8% in year 2, and 1.3% and 1.8% in year 3 for zoledronic acid and denosumab, respectively.4 The use of denosumab in patients with advanced-stage cancer is likely to be influenced by cost, in particular, because denosumab did not improve either progression-free survival or overall survival compared with zoledronic acid. However, collectively these trials show that denosumab is slightly more effective and convenient to administer, has some advantages in terms of safety over the current gold standard of treatment and represents an important advancement in the quality of life of patients with advanced-stage cancer. Other agents targeting the bone meta? stasis pathway are in development. The most interesting novel strategy for bone metastasis treatment during 2011 was reported by Parker and colleagues. 5 In this study, 922 men with advanced-stage castration-resistant prostate cancer who had not responded or were considered unsuitable for chemotherapy were randomly assigned (2:1) to receive a bone-localizing α?particle-emitting agent, 223RaCl 2 (also called alpharadin) or placebo. Median overall survival increased from 11.2?months to 14?months in patients receiving alpharadin (hazard ratio [HR] 0.695; 95% CI 0.552–0.875, P?=?0.002).5 In addition to the effects of bone-targeting treatments on the risk of skeletal complications in patients with cancer, these agents could modify the course of the disease via inhibitory effects on the ‘vicious cycle’ of growth factor and cytokine signaling between tumor and bone cells within the bone marrow microenvironment. In patients with early-stage breast cancer, inconsistent effects on disease outcomes were previously reported in clinical trials of adjuvant bisphosphonate treatment. However, in 2009, the results of the

ABCSG?12 trial 6 stimulated a considerable amount of interest in the field of targeting bone micro? environment to modify the course of early-stage breast cancer. Zoledronic acid treatment every 6?months for 3?years significantly improved diseasefree survival (DFS) in premenopausal women with endocrine-sensitive earlystage breast cancer previously treated with ovarian suppression therapy (goserelin) and tamoxi? fen or anastrozole.6 In 2011, the updated results of this trial7,8 showed that DFS was still improved with adjuvant zoledronic acid treatment and overall survival was also increased. Notably, these bene? ficial effects of zoledronic acid were only seen in patients over 40?years of age; goserelin treatment induced post? menopausal levels of circulating reproductive hormones consistently in these patients. Improvements in DFS and overall survival was also reported in postmenopausal women in the AZURE trial, a large randomized adjuvant trial that was designed to determine whether treatment with zoledronic acid in addition to adjuvant therapy improved disease outcomes in most women with early-stage breast cancer.9 In this study, 3,360 patients were randomly assigned to receive standard adjuvant systemic therapy with or without zoledronic acid every 3–4?weeks for six doses, then every 3–6?months thereafter, for a total of 5?years.9 After a median follow-up of 59?months, no significant difference was seen in DFS or overall survival. However, in a prespecified subgroup analysis, a significant increase in DFS was detected with zoledronic acid in 1,041 postmenopausal women (who were postmenopausal for >5?years at the time of diagnosis): 5?year invasive DFS was 71% in the control arm and 78.2% in those treated with zoledronic acid (adjusted HR 0.75; 95% CI 0.59–0.96, P?=?0.02).9 In addition, zoledronic acid improved overall survival in this group of women, with 5?year overall
JANUARY 2012? |? S19

survival rates of 79% in the control group and 85% in patients treated with zoledronic acid (adjusted HR 0.74; 95% CI 0.55–0.98, P?=?0.04). The data from both ABCSG?12 and AZURE6,9 trials suggest that the bene? ficial effects of adjuvant zoledronic acid treatment are dependent on an environment with low reproductive hormone levels. The mechanisms underlying this observation are under investigation. Prostate cancer has the propensity to metastasize almost exclusively to the bone and, therefore, provides the ideal clinical setting for the evaluation of bone-targeting treatments to modify the disease course. In a study by Smith and colleagues,10 1,432 men with non-metastatic castration-resistant prostate cancer at high risk for bone meta? stasis (determined by either a PSA level ≥8.0?ng/ml and/or PSA doub? ling time of ≤10?months) were randomly assigned to receive a monthly treatment of denosumab (120?mg) or placebo. Bone metastasis-free survival was significantly increased with denosumab by a median of 4.2?months (HR 0.85; 95%CI 0.73–0.98, P?=?0.028), and the onset of first symptomatic bone metastases was delayed. This effect, however, did not translate into any improvement in overall survival, and as the cumulative incidence of ONJ was high (4% after 3?years), whether the efficacy of denosumab to reduce bone meta? stasis is sufficient to change clinical practice is unclear. Overall, 2011 saw important developments in the prevention and management of bone metastasis. Integrating the current bone-targeting treatments in routine clinical practice and achieving a better understanding of the cellular processes involved in bone metastasis are the challenges for the?future.
Sheffield Cancer Research Center, Academic Unit of Clinical Oncology, Weston Park Hospital, Whitham Road, Sheffield S10 2SJ, UK.
Competing interests The author declares associations with the following companies: Amgen and Novartis. See the article online for full details of the relationships. 1. Stopeck, A.?T. et?al. Denosumab compared with zoledronic acid for the treatment of bone metastases in patients with advanced breast cancer: a randomized, double-blind study. J.?Clin. Oncol. 28, 5132–5139 (2010). Fizazi, K. et?al. Denosumab versus zoledronic acid for treatment of bone metastases in men with castration-resistant prostate cancer: a randomised, double-blind study. Lancet 377, 813–822 (2011). Henry, D.?H. et?al. Randomized, double-blind study of denosumab versus zoledronic acid in the treatment of bone metastases in patients with advanced cancer (excluding breast and prostate cancer) or multiple myeloma. J. Clin. Oncol. 29, 1125–1132 (2011). 4. Saad, F. et?al. Incidence, risk factors, and outcomes of osteonecrosis of the jaw: integrated analysis from three blinded activecontrolled phase?III trials in cancer patients with bone metastases. Ann. Oncol. http:// 5. Parker, C. et?al. Overall survival benefit of radium?223 chloride (alpharadin) in the treatment of patients with symptomatic bone metastases in castration-resistant prostate cancer (CRPC): a phase?III randomised trial (ALSYMPCA) [abstract]. Eur. J. Cancer 47?(Suppl. 2), a3 (2011). 6. Gnant, M. et?al. Endocrine therapy plus zoledronic acid in premenopausal breast cancer. N. Engl. J. Med. 360, 679–691 (2009). 7. Gnant, M. et?al. Adjuvant endocrine therapy plus zoledronic acid in premenopausal women with early-stage breast cancer: 62-month follow-up from the ABCSG?12 randomised trial. Lancet Oncol. 12, 631–641 (2011). 8. Gnant, M. et?al. Overall survival with adjuvant zoledronic acid in patients with premenopausal breast cancer with complete endocrine blockade: Long-term results from ABCSG?12 [abstract]. J. Clin. Oncol. 29 (Suppl. 15), a520 (2011). 9. Coleman, R.?E. et?al. Breast-cancer adjuvant therapy with and without zoledronic acid. N.?Engl. J. Med. 365, 1396–1405 (2011). 10. Smith, M.?R. et?al. Denosumab and bone metastasis-free survival in men with castrationresistant prostate cancer: results of a global phase?3, randomised, placebo-controlled trial. Lancet S0140-6736(11)61226-9.



Don’t miss out!

Sign up to free e-ALERTS and be notified when new articles are published in Nature Reviews Clinical Oncology. Research Highlights, News & Views, Reviews, and Perspectives—you won’t miss a thing!

S20? |? JANUARY 2012


Thyroid function—effects on mother and baby unraveled
Anthony P . Weetman

The complex relationship between pregnancy and thyroid function, and its clinical effect on mother and baby, continued to stimulate research in 2011. Key advances were made on three important issues: how long maternal thyroid function affects fetal thyroid hormone levels; whether thyroid autoimmunity affects pregnancy outcome; and the prevalence of permanent hypothyroidism after postpartum thyroiditis.
Weetman, A.?P . Nat. Rev. Endocrinol. 8, 69–70 (2012); published online 6 December 2011; doi:10.1038/nrendo.2011.217

The fetal thyroid gland starts to develop at 12?weeks of gestation, and for the first half of pregnancy, transplacental passage of maternal thyroid hormone is essential for normal fetal development. Failure to deliver sufficient thyroid hormone to the fetus causes impaired neurological development, although whether other elements of fetal development are also affected remains unclear. The essentially normal outcome in neonates with congenital hypothyroidism promptly treated with levothyroxine after birth suggests that maternal thyroid hormones also have a fetal role in the second half of pregnancy. In a prospective survey of 886 pregnant Dutch women published in 2011, neonates born to mothers who had high-normal TSH levels (above the 97.5th percentile) one or more times during pregnancy had lower total T4 levels when routine screening was performed for congenital hypothyroidism than neonates born to mothers whose TSH levels had remained below the 97.5th percentile.1 This result contrasts with that of a previous study that showed no relationship between maternal and neonatal thyroid function,2 a difference possibly related to the use of pregnancy-specific refer? ence ranges for TSH measurement in the 2011?study. Although the women with high-normal TSH levels had normal free T4 levels, the TSH levels might reflect a resetting of the pituitary–thyroid axis, which resulted in ? less T4 being available to cross the placenta. Alternatively, an unsuspected increase in placental deiodinase activity might have been responsible for low T4 availability to the fetus. Low gestational age was associated with low neonatal thyroxine levels in this and previous studies, which raises the

possibility that maternal thyroid function throughout pregnancy, rather than just the first trimester, is an important determinant of gestational outcome. Detailed sampling of maternal thyroid function is required to investigate this hypo? thesis further. In a related study from 2011, maternal thy? roid function variables measured at 28?weeks gestation were compared with those measured in the umbilical cord at birth in 616 preg?nancies in the UK.3 Maternal free T4 and cord free T4 values showed a positive correlation, and maternal TSH levels were inversely corre? lated with cord free T4 levels. This evi? dence by Shields and co-investigators is con? sistent with that of the Dutch study 1 and sug? gests that maternal thyroid function has a measurable effect on fetal thyroid hormone levels throughout pregnancy. Shields et al. also examined the effect of thyroid hormone levels on various para? meters of fetal growth. Cord free T4 levels were associated with birth weight, length and skin-fold thickness. These results imply that fetal thyroid hormone levels have an important role in regulating fetal growth, especially when taken in conjunction with the results of animal experiments. 4 An unexplained finding was that birth weight was negatively associated with maternal free T4 levels, given that maternal and cord free T4 levels were positively associated. Together, these two studies highlight the intricacy of the maternal–fetal relationship with regard to thyroid function and its clinically important effects on fetal growth and ges? tational age. Studies with detailed assessment of maternal thyroid function throughout the third trimester that incorpo? r ate assess? ment of placental deiodinase activity

are needed. Variable iodine intake is also a con? founding factor that needs addressing, given the surprising discovery in 2011 of inade? quate dietary supplies in the UK and in other countries.5 Multiple studies have shown that euthyroid pregnant woman with positive thyroid peroxidase (TPO) antibodies have an increased risk of spontaneous mis?carriage and possibly preterm delivery. Two mechanisms might be responsible. Firstly, even though ‘euthyroid’, these women may have subtle impairment of thyroid function, and the effects of maternal thyroid function on the fetus, as discussed above, are responsible. Secondly, thyroid auto? immunity dis? torts the immunological balance required to maintain tolerance to the fetal semi-allograft.6 In 2011, a study of 245 TPO-antibodypositive women with TSH levels <2.5?mU/l has shown that very preterm delivery (<34?weeks of gestation) and respiratory dis? tress at birth were more common than

JANUARY 2012? |? S21

? Gabriel Blaj |

in pregnancies of women who were negative for TPO antibodies.7 No association with TPO-antibody positivity was observed for 12 other adverse outcomes, including preterm delivery between 34–37?weeks. Although corrected for multiple comparisons, the statistical tests applied were onetailed rather than two-tailed. Given that the association with respiratory distress is novel, this finding could be a type?1 statis?tical error. The key feature of this study was that all the par?ticipants were chosen to have low-normal TSH levels, and thus the results appear to show that autoimmunity per?se is responsible for these adverse out? comes. How? ever, miscarriage rates, pre? viously shown to be increased in TPO-antibody-positive women, were not specifically increased and although below 2.5?mU/l, TSH levels were significantly higher in the TPO-antibody-positive women than in control women. Although the results clearly show that a trial of levothyroxine during pregnancy is warranted in this subset of TPO-antibody-positive women with lownormal TSH levels, the pathogenic mechanism behind the adverse outcomes of such pregnancies is still in question. Postpartum thyroiditis is a transient flaring of pre-existing autoimmune thyroidi? tis in the year after delivery, which seems to be a consequence of the restoration of nor? mal immune function after a period of altered immuno? r egulation during pregnancy. 6 Tran? sient abnormalities of thyroid dysfunction are common in postpartum thyroiditis; however, as has been clear for over two decades, 2–21% of such patients have permanent hypo? thyroidism at the end of the year after delivery, and this figure rises during follow-up over subsequent years. A survey of 4,394 women from Southern Italy reported in 2011 that 3.9% had postpartum thyroiditis and 54% of these were hypo? thyroid at 1?year after delivery.8 This figure for the prevalence of post? partum thyroiditis is rather low compared with that in many other studies, a difference that could be related to the exclusion of women with any thyroid dysfunction during the first trimester of pregnancy in the 2011 study. In addition, the women in this Italian study were only screened twice postpartum, so some cases of dysfunction might have been missed. By contrast, the figure for the prevalence of hypothyroidism at 1?year is strikingly high. This finding might also be related to the study design—if women with milder forms of postpartum thyroiditis were excluded because of limited sampling, this exclusion could have distorted the overall prevalence results, given that such women are
S22? |? JANUARY 2012

Key advances
■■ Maternal thyroid function has a measurable effect on fetal thyroid hormone levels throughout pregnancy1,3 ■■ Fetal thyroid hormone levels have an important role in regulating fetal growth, as measured by birth weight, length and skin-fold thickness3 ■■ Very preterm delivery (<34?weeks of gestation) and respiratory distress may be associated with maternal thyroid peroxidase antibody positivity7 ■■ Hypothyroidism following an episode of postpartum thyroiditis could be more common in some populations than previously described, and the risk of this outcome correlates with the severity of the preceding destructive thyroiditis8,9

fully established. However, no doubt exists that changes to the maternal autoimmune response in the postpartum period have considerable clinical effects on the mother’s thyroid reserve.
Department of Human Metabolism, University of Sheffield, The Medical School, Beech Hill Road, Sheffield S10 2RX, UK.
Competing interests The author declares no competing interests. 1. Kuppens, S. M. et al. Neonatal thyroid screening results are related to gestational maternal thyroid function. Clin. Endocrinol. (Oxf.) 75, 382–387 (2011). Oken, E. et al. Neonatal thyroxine, maternal thyroid function, and child cognition. J. Clin. Endocrinol. Metab. 94, 497–503 (2009). Shields, B.?M., Knight, B.?A., Hill, A., Hattersley,?A.?T. & Vaidya, B. Fetal thyroid hormone level at birth is associated with fetal growth. J. Clin. Endocrinol. Metab. 96, E934–E938 (2011). Bassett, J.?H. et al. A lack of thyroid hormones rather than excess thyrotropin causes abnormal skeletal development in hypothyroidism. Mol. Endocrinol. 22, 501–512 (2008). Vanderpump, M.?P . et al. Iodine status of UK schoolgirls: a cross-sectional survey. Lancet 377, 2007–2012 (2011). Weetman, A.?P . Immunity, thyroid function and pregnancy: molecular mechanisms. Nat. Rev. Endocrinol. 6, 311–318 (2010). Negro, R. et al. Thyroid antibody positivity in the first trimester of pregnancy is associated with negative pregnancy outcomes. J. Clin. Endocrinol. Metab. 96, E920–E924 (2011). Stagnaro-Green, A. et al. High rate of persistent hypothyroidism in a large-scale prospective study of postpartum thyroiditis in southern Italy. J. Clin. Endocrinol. Metab. 96, 652–657 (2011). Stuckey, B.?G. et al. Low urinary iodine postpartum is associated with hypothyroid postpartum thyroid dysfunction and predicts long-term hypothyroidism. Clin. Endocrinol. (Oxf.) 74, 631–635 (2011).



probably less likely to become hypothyroid. In this context, a related study has shown that low urinary iodine levels postpartum are associated with long-term hypothyroidism, probably because the low iodine excretion reflects more severe destructive thyroiditis that causes a discharge of thyroid iodine content.9 Together, these results reinforce the need for careful clinical evaluation of all women who have postpartum thyroiditis at 1?year after delivery and annually thereafter. In summary, five studies from 2011 have shown a pervasive effect of maternal thyroid function on the fetus throughout pregnancy. Abnormalities in maternal or fetal thyroid function are associated with an increasing number of adverse pregnancy outcomes. The additional role of autoimmunity per?se on pregnancy, rather than through its effect on maternal thyroid function, remains to be








Towards a better understanding of causation and consequences
Michael Stowasser

With primary aldosteronism now widely acknowledged as common and associated with both hypertension-related and non-hypertension-related pathology, research interest into its causes and consequences continues to grow. In 2011, major breakthroughs occurred in understanding the role and nature of underlying genetic disturbances and elucidating the?pathophysiology of its cardiovascular sequelae.
Stowasser, M. Nat. Rev. Endocrinol. 8, 70–72 (2012); published online 13 December 2011; doi:10.1038/nrendo.2011.223

The concept that primary aldo?steronism has a genetic basis is not new. Origi?nally reported in 1966,1 the rare, glucocorticoid-remediable

familial hyperaldosteronism type?1 (FH?I) is due to inherit? ance of a hybrid gene containing the regulatory elements of CYP11B1—a

gene that encodes steroid 11β-hydroxylase and is regulated by adrenocorticotropic hor? mone (ACTH)—and coding sequences from CYP11B2—a gene that encodes aldosterone synthase and is primarily regulated by angiotensin II. Expression of this hybrid gene leads to ACTH-regulated aldo? sterone overproduction.2 Glucocorticoids, in small doses, suppress hybrid gene expression and ameliorate primary aldosteronism and hypertension. Because wild-type CYP11B1 is expressed in the zona fasciculata (ZF) of the adrenal cortex, so is the hybrid gene (Figure?1). Thus, aberrant aldosterone synthase activity, normally restricted to the zona glomerulosa (ZG), occurs in the ZF. In contrast to the ZG, which lacks the steroid 17α-hydroxylase enzyme necessary for the formation of cortisol precursors, cortisol is available as a substrate for aldosterone synthase in the ZF. Hence, cortisol is converted into the ‘hybrid steroids’ 18-hydroxycortisol and 18-oxocortisol, the levels of which are elevated in patients with FH?I.2 The second recognized form of fami? lial hyperaldosteronism (FH-II) is neither gluco?corticoid-suppressible nor associ?ated with the hybrid gene mutation.3 Approxi? mately 30% of patients with FH?II have uni? lateral forms of primary aldosteronism (mostly due to an aldosterone-producing ade? n oma [APA]), with the remainder being bilateral (due to bilateral adrenal hyper? plasia [BAH]). The genetic basis of FH?II remains uncertain and is almost certainly hetero? geneous; several families have demon?strated linkage of FH?II with a locus at chromosome?7p22.3 In 2011, Mulatero and colleagues assessed prevalence rates of familial hyper? aldo? stero? nism among 300 patients consecutively diag? nosed with primary aldosteronism, after exclud? ing those suspected or diagnosed as having a form of familial primary aldo?steronism.4 Hybrid gene testing yielded only two (0.7%) patients with FH?I. Of the remaining 298 study participants, 199 had relatives available for and consenting to bio? chemical screening by aldosterone– renin ratio (ARR) testing and, where positive, definitive confirmation or exclusion of primary aldo? steronism by saline infusion testing. This analysis revealed that 12 patients had at least one affected relative, result? ing in a preva? lence of FH?II of at least 6% (almost 10-fold that of FH?I). The prevalence could be higher, as another 54 patients were classified as having ‘uncertain’ status, either because not all family members with hypertension underwent ARR testing

or because testing of family members was?inconclusive.4 In a major breakthrough towards further understanding the genetic basis of pri? mary aldosteronism, Choi and col? leagues reported somatic mutations (Gly151Arg and Leu168Arg) in KCNJ5, which encodes an inwardly-rectifying K+ channel, in eight of 22 patients with APA.5 A third, germline mutation in KCNJ5 (Thr158Ala) had pre? viously been identified in affected members of a US family with a new familial form of pri? mary aldosteronism (FH-III).6 The affected father and two daughters demon? strated very severe primary aldo? steronism, with childhood-onset hyper?tension, hypo?kalemia and markedly elevated aldosterone and sup? pressed renin levels. As in patients with FH?I, 18-hydroxycortisol and 18-?oxocortisol levels were elevated, albeit to a greater degree. How?ever, the primary aldosteronism was not ? glucocorticoid-suppressible, and the hybrid gene mutation was not present. Resected adrenal glands showed marked, diffuse hyperplasia of the ZF, with the combined weight of both adrenal glands in one daughter reaching 81?g (normal <12?g).6 None of the three KCNJ5 mutations had been reported in genetic databases. Choi et?al. discovered that all three amino acid sub? stitutions were in close proximity to the channel’s selectivity filter and, when expressed in human embryonic kidney (HEK293T) cells, were associated with the loss in channel selectivity for K+.5 These findings suggest that KCNJ5 mutations pre? dispose indivi? duals to chronic adreno? cortical cell membrane depolarization via increased channel Na+ permeability, the extra? cellular concentrations of which are much higher than those of K+. The resulting influx of Ca2+, in turn, leads to upregulation of enzymes involved in al? dosterone synthesis and cell?proliferation.5 In normal adrenal glands, expression of wild-type KCNJ5 is restricted to the ZG,5 but the massive hyperplasia seen in patients with FH-III involved the ZF.6 The fact that these patients had markedly elevated hybrid steroid levels also suggests that, as in FH?I, the adrenal cells express both aldosterone synthase and 17α-hydroxylase and, hence, have molecular characteristics of both the ZG and the ZF. Neither hybrid steroid levels nor the cellular composition of the tumors were reported for patients with APA. However, on the basis of findings in patients with FH?III, those with somatic KCNJ5 mutations are likely to have elevated levels of hybrid s ? teroids, and their APAs probably consist of mostly ZF-like cells.

Capsule Aldosterone ZG Aldosterone synthase



Cortisol Steroid 17αhydroxylase Steroid 11βhydroxylase


Figure 1 | Cellular and functional zones of steroid synthesis in the adrenal cortex. Normally, production of cortisol is restricted to ZF and ZR, owing to the enzymatic activity of steroid 17α-hydroxylase (encoded by CYP17A1) and 11β-hydroxylase (encoded by CYP11B1). Similarly, production of aldosterone is restricted to ZG, as enzymatic activity of aldosterone synthase (encoded by CYP11B2) is limited to this zone. In familial hyperaldosteronism type?I, aldosterone synthase is active in ZF and ZR, as the hybrid CYP11B1/CYP11B2 gene has regulatory elements derived from CYP11B1. Hence, aldosterone synthase has access to cortisol as a substrate, leading to the production of 18-hydroxycortisol and 18-oxocortisol. Abbreviations: ZF, zona fasciculata; ZG, zona glomerulosa; ZR, zona reticularis.

Gordon et al. previously described two dis? tinct forms of APA.7,8 Angiotensin-II-? unresponsive APA are composed predominantly of ZF-like cells; patients with this form of APA have elevated hybrid steroid levels and lack responsiveness of aldo? sterone to upright posture or angiotensin-II infusion. Angiotensin-II-responsive APA contain predominantly non-ZF-like morphology (ZG or hybrid ZF/ZG), and patients with this form usually have normal hybrid steroid levels and are aldosterone-responsive to upright posture or angiotensin-II infusion. Taking the above findings into account, APAs bear?ing KCNJ5 mutations are likely to be of the angiotensin-II-unresponsive subtype. Further studies comparing patients with APAs with or without mutations in terms of histology, steroid hydroxylase expression, aldosterone responsiveness to posture and/or ?angiotensin-II, and peripheral hybrid steroid levels are awaited with great interest. If KCNJ5 mutation-bearing cells have both ZG and ZF characteristics, what is
JANUARY 2012? |? S23

Key advances
■■ Familial hyperaldosteronism type?II (FH-II) may account for at least 6% of primary aldosteronism cases and is approximately 5–10 times more common than FH?I4 ■■ A germline mutation in KCNJ5 is associated with severe, early-onset familial primary aldosteronism, and somatic KCNJ5 mutations are common in aldosterone-producing adenoma5 ■■ Patients with primary aldosteronism have reduced circulating levels of endothelial progenitor cells, which might contribute to the development of vasculopathy in these individuals9 ■■ The degree of left ventricular enlargement in primary aldosteronism is largely determined by dietary salt; dietary salt restriction might reduce cardiovascular risk in this condition10

their origin? Do KCNJ5 mutations bestow ZF-like characteristics (including his? tol? ogy, 17α-hydroxylase expression, hybrid steroid formation, and loss of aldosterone responsive? ness to angiotensin-II) to ZG cells, or are ZG-like characteristics (aldo? sterone synthase expression) bestowed onto ZF cells? Further studies should shed light on this fascinating issue. Other studies in 2011 have focused on the mechanisms underlying adverse cardio? vascular effects of aldosterone excess. In a novel study by Wu et al.,9 levels of endothelial progenitor cells (EPCs), which are thought to protect against cardiovascular disease by repairing endothelial injury, were lower in 113 patients with primary aldo? steronism (APA n?=?87; BAH n?=?26) than in 55 patients with essential hypertension.9 Differences in pulse-wave velocity, a marker of arterial stiffness, and high-? s ensitivity C?reactive protein (hsCRP) levels, a marker of cardiovascular inflammation, and EPC counts were attenuated following uni? lateral adrenalectomy or during treatment with aldosterone antagonists. Overall, the findings suggest that increased circulating aldosterone levels in patients with primary aldosteronism contribute to vasculopathy by reducing EPC numbers, partly by activating EPC mineralocorticoid receptors and possibly indirectly by raising hsCRP levels.9 Animal studies have demonstrated a cri? tical role for salt in the development of aldosterone-induced cardiovascular damage; how?ever, corroborative data in humans were lack?ing. In a case–control study of 21 patients with primary aldosteronism and 21 matched individuals with essential hyperten? sion, Pimenta and co-workers reported in 2011
S24? |? JANUARY 2012

that patients with primary aldosteronism had greater thickness of the left ventricular wall, end-diastolic dia?meter and mass. More? over, urinary sodium excretion, a mar? ker of dietary salt intake, positively correlated with and was an indepen? dent predictor of left ventricular wall thickness and mass in patients with pri? mary aldosteronism, but not in those with essential hypertension.10 Hence, as in animal studies, salt appears to interact with auto? nomous aldosterone excess to bring about cardio?vascular damage in patients with primary aldosteronism. The lack of a posi?tive correlation between plasma aldoster? one and left ventricular dimen? sions in patients with primary aldosteronism raises the possibility that, above a certain thres? hold, aldosterone plays a more permissive part, whereas salt has a more graduated effect. Either way, these results argue for a role of dietary salt restriction to reduce the risk of cardiovascu?lar disease in patients with primary aldosteronism.
Endocrine Hypertension Research Center, University of Queensland School of Medicine, Greenslopes and Princess Alexandra Hospitals, Ipswich Road, Woolloongabba, Brisbane 4102, Australia.
Competing interests The author declares no competing interests. 1. Sutherland, D.?J., Ruse, J.?L. & Laidlaw, J.?C. Hypertension, increased aldosterone secretion and low plasma renin activity relieved by

dexamethasone. Can. Med. Assoc. J. 95, 1109–1119 (1966). 2. Lifton, R.?P . et al. Hereditary hypertension caused by chimaeric gene duplications and ectopic expression of aldosterone synthase. Nat. Genet. 2, 66–74 (1992). 3. Stowasser, M., Pimenta, E. & Gordon, R.?D. Familial or genetic primary aldosteronism and Gordon syndrome. Endocrinol. Metab. Clin. North Am. 40, 343–368, viii (2011). 4. Mulatero, P . et al. Prevalence and characteristics of familial hyperaldosteronism: the PATOGEN study (Primary Aldosteronism in TOrino-GENetic forms). Hypertension 58, 797–803 (2011). 5. Choi, M. et al. K+ channel mutations in adrenal aldosterone-producing adenomas and hereditary hypertension. Science 331, 768–772 (2011). 6. Geller, D.?S. et al. A novel form of human mendelian hypertension featuring nonglucocorticoid-remediable aldosteronism. J. Clin. Endocrinol. Metab. 93, 3117–3123 (2008). 7. Gordon, R.?D., Hamlet, S.?M., Tunny, T.?J. & Klemm, S.?A. Aldosterone-producing adenomas responsive to angiotensin pose problems in diagnosis. Clin. Exp. Pharmacol. Physiol. 14, 175–179 (1987). 8. Tunny, T.?J., Gordon, R.?D., Klemm, S.?A. & Cohn,?D. Histological and biochemical distinctiveness of atypical aldosteroneproducing adenomas responsive to upright posture and angiotensin. Clin. Endocrinol. (Oxf.) 34, 363–369 (1991). 9. Wu, V.?C. et al. Endothelial progenitor cells in primary aldosteronism: a biomarker of severity for aldosterone vasculopathy and prognosis. J.?Clin. Endocrinol. Metab. 96, 3175–3183 (2011). 10. Pimenta, E. et al. Cardiac dimensions are largely determined by dietary salt in patients with primary aldosteronism: results of a case– control study. J. Clin. Endocrinol. Metab. 96, 2813–2820 (2011).


Genes, aging and sleep apnea in polycystic ovary syndrome
Andrea Dunaif

Polycystic ovary syndrome (PCOS) is a complex genetic disease that affects approximately 7% of women of reproductive age worldwide. From novel pathways implicated in the etiology of PCOS through genome-wide association to characterization of the reproductive and metabolic changes that occur in ageing women with PCOS, the year 2011 has seen a number of studies published that highlight the intricacies of this condition.
Dunaif, A. Nat. Rev. Endocrinol. 8, 72–74 (2012); published online 20 December 2011; doi:10.1038/nrendo.2011.227

The year 2011 saw the advent of the first genome-wide association study (GWAS) in polycystic ovary syndrome (PCOS).1 GWAS have been widely used, since the publication of the human haplotype map in 2005, to localize susceptibility genes for com? plex traits, such as obesity and type?2 dia? betes

mellitus (T2DM).2 This analysis per? mits an unbiased examination of the entire genome for novel disease susceptibility loci and, unlike candidate gene approaches, is ?hypothesis-generating. 2 The first GWAS of PCOS was conducted in Han Chinese women with PCOS, who were diagnosed

using the Rotterdam cri? teria (two of three of the following findings: oligomenor? rhea, hyperandrogen?ism and/or poly?cystic ovaries detected on ultrasono?graphy).1 Asso?cia?tions achiev? ing genome-wide levels of significance3 between PCOS and loci on chromosomes 2p16.3 (OR?0.71), 2p21 (OR?0.67) and 9q33.3 (OR 1.34) were observed in a dis? covery sample of 744 PCOS cases and 895 con?trols and in two indepen?dent replication cohorts of 2,840 PCOS cases and 5,012 controls, and 498 PCOS and 780?controls.1 ? Several known genes are located near the locus at 2p16.3, which showed the great? est association with PCOS, includ? ing GTF2A1L and LHCGR. GTF2A1L, which encodes TFIIAα/β?like factor (ALF), is a germ-cell-specific transcription factor that is highly expressed in adult testis and may play a part in spermato? genesis. How variation in GTF2A1L might contribute to PCOS is unclear. LHCGR encodes the lutropin–? choriogonadotropic hormone recep? t or (LH/CG?R) and is a highly plau?? si? ble PCOS candidate gene. The strongest asso?ci?ation at the locus on chromo? some?2p21 was with THADA, a gene origi? nally iden? tified in thy?? roid adenomas. A GWAS in white indivi? duals of European ances? t ry reported an associa?tion of THADA with T2DM.4 The region on chromosome?9q33.3 associ? ated with PCOS was located within DENND1A, which encodes a domain differen? t ially expressed in normal and neoplastic cells (DENN) that can bind to and negatively regulate endo? plasmic reticulum amino? peptidase?1. Elevations of circulating endoplasmic reti? culum amino? peptidase 1 have been reported in obese women with PCOS. Thus, it is possible that varia? tion in THADA and DENND1A contribute to T2DM and obesity risk in women with PCOS. Confirmation that the GWAS signals reflect a variation in these genes, rather than in other genes that are in linkage dis? equilibrium with these loci, requires fur? ther genetic analyses. In addition, the roles of these loci in PCOS suscepti? bility in other racial or eth? nic groups and in other pheno? types, for instance, in women with PCOS diag? nosed using the National Insti? tute of Child Health and Human Develop? ment (NICHD) criteria (history of oligo?menorrhea and clinical and/or biochemi? cal evidence of hyper? androgenism), will require further investigation. Studies in monozygotic twins indicate that the herit? ability of PCOS is close to 80%. However, the loci identified in the PCOS GWAS confer modest (~30%) changes in disease risk and do not account

Key advances
■■ Three genetic susceptibility loci have been mapped by genome-wide association study in Han Chinese women with PCOS1 ■■ Increased ovarian and adrenal androgen production and insulin resistance persist in postmenopausal women with PCOS6,7,8 ■■ Obstructive sleep apnea contributes to insulin resistance and continuous positive airway pressure improves insulin sensitivity in women with PCOS10

for the observed herit? ability of PCOS.5 This so-called ‘missing heritability’ has been seen in other common complex traits, such as T2DM, and might reflect the fact that rare rather than common variants contribute to complex diseases.5 Nevertheless, GWAS has been important for implicating novel b? iologic pathways in disease pathogenesis. Little was known about the reproductive and metabolic changes that occur with age in women with PCOS. In 2011, two crosssectional studies compared postmenopausal women in the sixth decade of life with and without PCOS.6,7 In both studies, the diagnosis of PCOS was made on the basis of the NICHD diagnostic criteria, although the study by Puurunen et?al. 7 also contained premenopausal women with PCOS diagnosed by the Rotterdam criteria and a control group consisting of reproductively normal, premenopausal women. The study by Markopoulos et?al. 6 found that postmenopausal women with PCOS had significantly elevated levels of circulating total testosterone, androstenedione, dehydro? epiandrosterone sulfate (DHEAS) and 17-hydroxyprogesterone and an increased free androgen index (FAI). Sex hormonebinding globulin (SHBG) levels, how? ever, were significantly decreased compared with the control group. No evidence of increased sensitivity to corticoliberin (also known as corticotropin-?releasing hormone) or adreno? cortico?t ro?pic hormone was reported in women with PCOS, which suggests that increases in adrenal androgen production did not result from changes in responsiveness to tropic hor? mones. Dexamethasone suppression sug? gested that elevated total testosterone and DHEAS levels were partly adrenal in origin. Puurunen et?al.7 also found increased androstenedione levels, basally and in response to chorio? gonado? t ropin,

in postmenopausal women with PCOS compared with postmenopausal control women. Furthermore, they found persistent evidence for insulin resistance and increased inflammation, independent of BMI, in post? menopausal women with PCOS compared to those without the disease. In a 2011 prospective study, Schmidt et?al.8 diagnosed PCOS on the basis of ovarian histology from ovarian wedge resec? tion or unilateral oophorectomy per?formed in 1956– 1965. These women were examined in 1987 and compared with control women from a population-based study. Both groups were recalled for repeat examination in 2008, in their eighth decade of life. The PCOS group had higher FAI and lower follitropin (also known as follicle-stimulating hormone; FSH) and SHBG levels than the control group, as was the case in 1987. DHEAS, total testo?sterone and androstenedione levels were significantly increased in premenopausal women with PCOS in 1987, but were similar to those of the control group after menopause. The prevalence of hypo? thyroidism was lower in post? menopausal women with PCOS than in control individuals. Taken together, these three studies6,7,8 sug? gest that hyperandrogenism of both ova? rian and adrenal origin, as well as insulin resistance, persist in postmenopausal women with PCOS. Increased androgen levels declined with age in older post? menopausal women with PCOS such that only differences in FAI were present in women aged >70?years.8 By contrast, decreases in SHBG and FSH levels persist into old age. All of the studies are limited by small sample sizes, which might account for the differences in the prevalence of hypothyroidism. In

JANUARY 2012? |? S25

the cross-sectional studies, the accuracy of diagnosis of PCOS on the basis of medical records is a potential limitation, although oligo?menorrhea and symptoms of androgen excess, such as hirsutism, have been shown to be excellent proxies for PCOS.9 The prevalence of obstructive sleep apnea (OSA) is significantly increased in women with PCOS, independent of obesity—a wellknown risk factor for OSA. Androgen levels and insulin resistance are positively associated with OSA in PCOS. In 2011, Tasali and colleagues proposed that OSA contributes to insulin resistance in PCOS, as it does in other OSA populations.10 The investigators directly tested this hypothesis by treating affected women with continuous positive airway pressure (CPAP), which resulted in a modest but significant improvement in insulin sensitivity.10 A dose–response effect of CPAP on insulin sensitivity was observed, with greater improvements in sensitivity with longer duration of CPAP. A significant decrease in circulating norepinephrine levels was also reported, without changes in epinephrine, cortisol or leptin levels. This observation suggests that the decrease in insulin sensitivity is mediated by sympathetic nervous system activation. Diastolic blood pressure, which was not elevated in women with PCOS who had OSA, decreased significantly with CPAP. The limitations of this study were the small number of women (n?=?9) who completed the CPAP intervention with acceptable compliance (≥4?h of use per night).10 Further? more, the study population was extremely obese, with a mean BMI of 48.4?kg/m 2. The improvements in insulin sensitivity, although significant, were very modest (on average 7%); the study par?ticipants remained profoundly insulin resistant after CPAP treatment. Modeling of the data suggested that the beneficial effect of CPAP would be greater in overweight or mildly obese patients with PCOS. How? ever, the robustness of the model is questionable given the small sample size. Never? the? less, this study suggests that OSA con? tributes to insulin resistance in PCOS and that CPAP improves insulin sensitivity in affected?women. In conclusion, these 2011 publications will have a major effect on the field. It is now clear from an adequately powered and replicated study that genetic variation contri? butes to the etiology of PCOS. Furthermore, reproductive and metabolic features of the dis? order persist beyond the reproductive years, increasing the impact of PCOS as a leading women’s health issue. Finally, a
S26? |? JANUARY 2012

component of the insulin resistance associated with PCOS is secondary to OSA and improves with CPAP treatment.
Division of Endocrinology, Metabolism and Molecular Medicine, Northwestern University Feinberg School of Medicine, 303 East Chicago Avenue, Tarry 15?745, Chicago, IL 60611, USA.
Competing interests The author declares no competing interests. 1. Chen, Z.?J. et al. Genome-wide association study identifies susceptibility loci for polycystic ovary syndrome on chromosome?2p16.3, 2p21 and 9q33.3. Nat.?Genet. 43, 55–59 (2011). Hirschhorn, J.?N. & Daly, M.?J. Genome-wide association studies for common diseases and complex traits. Nat. Rev. Genet. 6, 95–108 (2005). Hoggart, C.?J., Clark, T.?G., De Iorio, M., Whittaker, J.?C. & Balding, D.?J. Genome-wide significance for dense SNP and resequencing data. Genet. Epidemiol. 32, 179–185 (2008). Zeggini, E. et al. Meta-analysis of genome-wide association data and large-scale replication identifies additional susceptibility loci for type?2 diabetes. Nat. Genet. 40, 638–645 (2008).





Manolio, T.?A. et al. Finding the missing heritability of complex diseases. Nature 461, 747–753 (2009). 6. Markopoulos, M.?C. et al. Hyperandrogenism in women with polycystic ovary syndrome persists after menopause. J. Clin. Endocrinol. Metab. 96, 623–631 (2011). 7. Puurunen, J. et al. Unfavorable hormonal, metabolic, and Inflammatory alterations persist after menopause in women with PCOS. J. Clin. Endocrinol. Metab. 96, 1827–1834 (2011). 8. Schmidt, J., Br?nnstr?m, M., LandinWilhelmsen, K. & Dahlgren, E. Reproductive hormone levels and anthropometry in postmenopausal women with polycystic ovary syndrome (PCOS): a 21-year follow-up study of women diagnosed with PCOS around 50?years ago and their age-matched controls. J. Clin. Endocrinol. Metab. 96, 2178–2185 (2011). 9. Taponen, S. et al. Hormonal profile of women with self-reported symptoms of oligomenorrhea and/or hirsutism: Northern Finland Birth Cohort 1966 Study. J. Clin. Endocrinol. Metab. 88, 141–147 (2003). 10. Tasali, E., Chapotot, F., Leproult, R., Whitmore,?H. & Ehrmann, D.?A. Treatment of obstructive sleep apnea improves cardiometabolic function in young obese women with polycystic ovary syndrome. J. Clin. Endocrinol. Metab. 96, 365–374 (2011).


Epigenetics, the life-course and metabolic disease
Peter D. Gluckman

Clinical and experimental studies suggest that early life experiences, perhaps spanning multiple generations, affect lifelong risk of metabolic dysfunction through epigenetic mechanisms. Data published in 2011 suggest that epigenetic analysis could potentially have utility as a marker of early metabolic pathology and might enable early life prophylaxis.
Gluckman, P .?D. Nat. Rev. Endocrinol. 8, 74–76 (2012); published online 20 December 2011; doi:10.1038/nrendo.2011.226

Both population and individual variation in sus? ceptibility to an obesogenic environ? ment exist. Genome-wide association studies have been disappointing in explaining this variation, and there is a growing focus on possible epigenetic explanations. Epigenetic variation is most likely to arise early in the life-course. An extensive body of experimental, clini? cal and epidemiological evidence demon? strates that early life develop? mental factors—? including the maternal diet, but operating within the normative range of develop? mental exposures—affect the risk of developing meta?bolic disease later in life.1 This phe? nomenon is sometimes termed developmental programming. The early observations were largely ignored because of the absence of a plausible biological

mechanism, but explana? t ions developed over the past decade have been framed in terms of developmental p ? lasticity—the adaptive processes enabling an organism to respond to environmental cues acting in early life and adjust its develop? mental trajectory. Developmental plasticity is at least partly underpinned by epigenetic mecha? nisms, including DNA methylation and histone modifications. A number of papers published during 2011 have re? inforced the validity of these conclusions. Extensive previous work has shown that rats whose mothers are underfed during preg? nancy and lactation develop an insulin resis? t ant and obese phenotype in adulthood. Moreover, mice lacking the transcription factor hepatocyte nuclear factor 4α

(HNF?4α), a crucial regulator of gene expression in pancreatic islet β cells, have impaired glucose tolerance. In 2011, Sandovici et?al. reported how maternal diet in rats affects the epigenetic regulation of the Hnf4a locus in pancreatic islets of offspring.2 The researchers found that a low-protein maternal diet led to underexpression of Hnf4a in the islets of the offspring. Mechanistically, the under? expression could be attributed in part to his? tone modifications at the gene enhancer region of Hnf4a, which led to changes in chromatin looping that ultimately weakened the interaction between the enhancer region and a specific promoter region. The normative age-dependent epigenetic silencing of the Hnf4a locus was exacerbated in rats exposed in?utero to a low-protein diet. These data add to the growing experimental literature showing epigenetic consequences from the manipulation of early life undernutrition in metabolically relevant tissues. Previous studies in rats have shown that neonatal administration of the adipokine leptin can reverse developmental programming leading to metabolic compromise as well as some of the associated epi? genetic changes. 3 Such studies are provocative in that they suggest that develop? mental program? ming in humans might be reversi? ble. Pinney et?al. examined how adminis? tration of exenatide, an analogue of the incretin glucagon-like peptide 1, acts to prevent the development of glucose intolerance in adulthood in rats subjected to intrauterine growth restriction.4 Rats subjected to this abnormal intrauterine environ? ment develop glucose intolerance and are charac? terized by progressive epi? genetic silencing of pancreatic islet transcription factor Pdx1. The 2011 paper found that injection of exenatide during postnatal day 1–6 in rats after intrauterine growth restriction increased Pdx1 transcription by normalizing post-translational and epigenetic changes
Key advances
■■ Changes in promoter–enhancer interactions underlie the epigenetic regulation of a transcription factor in rat pancreatic islets induced by poor maternal diet 2 ■■ The trajectory towards onset of type?2 diabetes mellitus (T2DM) in the growth-compromised newborn rat can be reversed by an incretin analogue that normalizes epigenetic modifications associated with pancreatic β?cell dysfunction4 ■■ Strong relationships exist between epigenotype at birth and later adiposity, and between maternal nutrition and offspring epigenotype in humans6 ■■ The epigenetic states of genes associated with T2DM and obesity might be a presymptomatic marker of the lifetime risk of T2DM in humans7 ■■ The phenotypic effects of early life overnutrition can be transgenerationally transmitted through the paternal line in mice9

related to this gene, including Pdx1 activator phosphorylation, histone acetylation and trimethylation, and DNA methylation at the CpG island in the Pdx1 promoter. Exenatide is already in use in the treatment of adults with type 2 diabetes mellitus (T2DM), and this study provides an experimental proof of concept for its usage early in life prior to disease onset. Prophylactic treatment of at-risk indivi? duals early in life is generally more effica? cious and provides greater health and cost benefits than treatment after a disease de? velops. This is particularly the case for diseases such as diabetes mellitus, where the cost of treat? ing associated morbidities and complica? tions compounds over many years. How??ever, beyond the obvious research program that would be needed to validate such interventions, a challenge remains: how do we identify infants whose particular developmental programming has placed them at increased metabolic risk? Birth weight is not a satisfactory proxy, given that metabolically adverse developmental programming can occur independent of birth weight.5 A recent paper by Godfrey et?al. —to which I contributed—has suggested that the epigenetic state at birth can both predict childhood adiposity and be a measure of prenatal nutritional exposures.6 Umbilical cord tissue-derived DNA was sourced from two independent large-scale prospective cohorts, and the methylation status at specific CpGs in the promoter region of RXRα , which encodes a transcription factor involved in fat metabolism and insulin sensitivity, was measured after a preliminary genome-wide scan. A positive correlation between degree of RXRα methyla? tion at one specific site in the promoter and childhood body adipo? sity at age 6?years or 9?years was found in both within-cohort and between-cohort replicates. Additionally, a negative association was observed between mater? nal carbo? hydrate intake during the first tri?mes?ter of pregnancy and the degree of methylation at

this site. In contrast to the small effect sizes typically seen with genetic polymorp? hisms associated with metabolic disease risk, this single site epigenotype could account for at least 25% of the variance in child? hood adipo? sity. This association was obser? ved across a healthy population, indicating that develop? mental programming of metabolic risk and associated epigenetic variation occurs across the normative range of developmental?experiences. Genes for which polymorphisms have strong associations with disease risk might be anticipated to have epimutations (aberrant epigenetic marks). One such gene is FTO, which has been identified by genomewide association studies as being involved in obesity risk. In a prospective study of nonsymptomatic indivi? duals of a mean age of 30?years by Toperoff et?al.,7 those who showed hypomethylation at an intronic CpG site of FTO in blood samples had an increased risk of develop? ing impaired glucose metabolism by a mean age of 43?years. This pattern was also seen in a sepa?rate case–control analysis, although the level of hypomethylation in cases relative to controls was small, as was the effect size.7 These findings were made in a tissue not specifically related to metabolic function, and the differential methylation levels may have been established in early develop? ment before major cell differentiation had occurred, although this inference is contingent on the long-term stability of the FTO epigenotype after its establishment. The methylation levels could not be explained by the presence or absence of the risk allele, indicating that their association with disease risk was independent of DNA sequence. Global prevalence of maternal obesity and ges? tational diabetes mellitus is rising, and both are risk factors for the develop? ment of meta?? bolic disorder in offspring;8 there? fore, the potential for an escalating prevalence of intergenerationally induced T2DM is a par? ticular concern. Epigenetic
JANUARY 2012? |? S27


? Paul Hakimata |

mechanisms could underlie the inter? generational transmis? sion of predisposition to T2DM, but unraveling the mechanisms involved is highly complex owing to the multiple pathways by which nongenomic inheritance, involving either direct or indirect epi? genetic mechanisms, could occur. Dunn and Bale utilized a maternal high-fat diet mouse model and sought to eliminate maternal physiological or behavioral effects and grand-maternal effects by studying pheno? typic outcomes up to the F3 generation through the paternal line.9 They found that female-specific increased body weight and length was inherited through the pater? nal, not the maternal, lineage. Taken together with other research show? ing that specific histone marks located at developmental loci in sperm are maintained beyond spermio?genesis,10 this research suggests that the phenotype was transmitted through the male germline via stable epigenetic marks in sperm. Our knowledge of developmental path? ways to metabolic disease is rapidly increas? ing. It is becoming apparent that epi? genetic variations induced early in life or that are present later in life are associated with metabolic risk. How far environ?ment