The 21St Century Cures Act — Will It Take Us Back in Time? Jerry Avorn, M.D., and Aaron S

The NEW ENGLAND JOURNAL of MEDICINE

Perspective

The 21st Century Cures Act — Will It Take Us Back in Time? Jerry Avorn, M.D., and Aaron S. Kesselheim, M.D., J.D., M.P.H.

n May 2015, the 21st Century Cures Act was intro- tem. An underlying premise of duced in the U.S. House of Representatives, with the bill is the need to accelerate I approval for new products, but the goal of promoting the development and speeding this process is already quite effi- the approval of new drugs and devices.1 Championed cient. A third of new drugs are currently approved on the basis by the pharmaceutical, biotech- provide an additional $2 billion of a single pivotal trial; the me- nology, and device industries, the per year for 5 years to create an dian size for all pivotal trials is bill was approved unanimously “NIH Innovation Fund.” Together, just 760 patients. More than two (51 to 0) in committee and con- this support would help counter- thirds of new drugs are approved tinues to be debated. If enacted act the effects of sequestration on the basis of studies lasting into law, some of its provisions and budget cuts that have re- 6 months or less3 — a potential could have a profound effect on duced the purchasing power of problem for medications de- what is known about the safety the NIH to its lowest level in signed to be taken for a lifetime. and efficacy of medical products, years. Given the crucial role that Once the Food and Drug Admin- as well as which ones become NIH-funded research plays in istration (FDA) starts its review, it available for use. generating the findings on which approves new medications about Some aspects of the bill could so many new drugs are based,2 as quickly as any regulatory indeed enhance the development this boost would be a welcome agency in the world, evaluating of and access to new drugs. The development. Another useful pro- nearly all new drug applications legislation calls for annual in- vision could make deidentified within 6 to 10 months, an im- creases in the stagnating budget data from NIH-funded clinical tri- pressive turnaround for such for the National Institutes of als more available to researchers. complex assessments. Health (NIH) amounting to about Other proposed changes could Nonetheless, as introduced, the 3% per year for 3 years when ad- lead to less salutary outcomes for 21st Century Cures Act instructs justed for inflation. It would also patients and the health care sys- the FDA to consider nontraditional

n engl j med nejm.org 1 The New England Journal of Medicine Downloaded from nejm.org at PARTNERS HEALTHCARE on June 3, 2015. For personal use only. No other uses without permission. Copyright © 2015 Massachusetts Medical Society. All rights reserved. PERSPECTIVE The 21st Century Cures Act

study designs and methods of even as it increased their risk of proven antibiotics, which could data analysis to further speed myocardial infarction. In 2013, encourage their more widespread approvals. Adaptive trial designs patients began to receive a new use. The bill gives the secretary and the use of Bayesian methods drug for tuberculosis approved of health and human services the hold promise in some kinds of on the basis of a randomized authority to expand this nontradi- evaluations, particularly in oncol- trial relying on a surrogate mea- tional approval pathway to other ogy. However, more problematic sure of bacterial counts in the drug categories as well, if “the proposals include encouraging sputum — even though patients public health would benefit from the use of “shorter or smaller given the drug in that trial had a expansion.” clinical trials” for devices and death rate four times that in the The 21st Century Cures Act the request that the FDA develop comparison group, mostly from goes still further in altering the criteria for relying on “evidence tuberculosis.4 These provisions in requirements for approving med- from clinical experience,” includ- the legislation would not imme- ical devices — an area long criti- ing “observational studies, regis- diately change FDA approval stan- cized for lack of rigor as com- tries, and therapeutic use” instead dards, but they would give the pared with drug evaluations,5 of randomized, controlled trials agency greater discretion, backed though regulatory oversight has for approving new uses for exist- by congressional support, to ap- improved in recent years. As pro- ing drugs. Although such data prove drugs on the basis of less posed, the new law would re can provide important informa- rigorous data. define the evidence on which tion about drug utilization and The proposed legislation would high-risk devices can be approved safety once a medication is in use, make immediate changes with to include case studies, registries, there is considerable evidence that respect to new antibiotics and and articles in the medical litera- these approaches are not as rigor- antifungals by enabling their ap- ture, rather than more rigorous ous or valid as randomized trials proval without conventional clin- clinical trials. Another section in assessing efficacy. ical trials, if needed to treat a would allow device makers to The bill would also encourage “serious or life-threatening infec- pay a third-party organization to the FDA to rely more on bio- tion” in patients with an “unmet determine whether the manufac- markers and other surrogate mea- medical need.” In place of proof turer can be relied on to assess sures rather than actual clinical that the antimicrobial actually the safety and effectiveness of end points in assessing the effi- decreases morbidity or mortality, changes it makes to its devices, cacy of both drugs and devices. the FDA would be empowered in place of submitting an appli- The FDA already uses surrogate to accept nontraditional efficacy cation to the FDA. Thus certified end points in about half of new measures drawn from small stud- by the external company, a de- drug approvals.3 Some biomark- ies as well as “preclinical, phar- vice maker would be authorized ers are accurate predictors of macologic, or pathophysiologic to continue to assess its own disease risk and can be useful evidence; nonclinical suscepti- products on an ongoing basis. measures of the efficacy of a bility and pharmacokinetic data, Informed consent by patients new drug (such as low-density data from phase 2 clinical trials; in drug trials has traditionally lipoprotein cholesterol for statins). and such other confirmatory evi- been sacrosanct, with exceptions But though a drug’s effect on a dence as the secretary [of health made only when consent is im- biomarker can make approval and human services] determines possible to obtain or contrary to quicker and less costly, especially appropriate to approve the drug.” a patient’s best interests. But an- if the comparator is placebo, it Antimicrobials approved in this other clause in the proposed law may not always predict the drug’s manner would carry disclaimers adds a new kind of exception: capacity to improve patient out- on their labeling, but there is no studies in which “the proposed comes. Bevacizumab (Avastin) de- evidence that such a precaution clinical testing poses no more layed tumor progression in ad- would restrict prescribing to only than minimal risk” — a major vanced breast cancer but was the most appropriate patients. If departure from current human shown not to benefit patients. passed in its current form, the subject protections. It is not clear Similarly, rosiglitazone (Avandia) bill would also provide hospitals who gets to determine whether a lowered glycated hemoglobin with a financial bonus for ad- given trial of a new drug poses levels in patients with diabetes ministering costly new but un- “minimal risk.”

2 n engl j med nejm.org The New England Journal of Medicine Downloaded from nejm.org at PARTNERS HEALTHCARE on June 3, 2015. For personal use only. No other uses without permission. Copyright © 2015 Massachusetts Medical Society. All rights reserved. PERSPECTIVE The 21st Century Cures Act

Embedded in the language of apeutic efficacy and risk. Patients 1. 21st Century Cures Act. May 19, 2015. (http://docs.house.gov/meetings/IF/IF00/ the 21st Century Cures Act are and physicians would not benefit 20150519/103516/BILLS-1146ih.pdf). some good ideas that could from legislation that instead of 2. Kesselheim AS, Tan YT, Avorn J. The roles streamline the development and catapulting us into the future, of academia, rare diseases, and repurposing in the development of the most transforma- evaluation of new drugs and de- could actually bring back some of tive drugs. Health Aff (Millwood) 2015;34: vices; its call for increased NIH the problems we thought we had 286-93. funding may prove to be its most left behind in the 20th century. 3. Downing NS, Aminawung JA, Shah ND, Krumholz HM, Ross JS. Clinical trial evi- Disclosure forms provided by the authors useful component. But political dence supporting FDA approval of novel are available with the full text of this article forces have also introduced other therapeutic agents, 2005-2012. JAMA 2014; at NEJM.org. provisions that could lead to the 311:368-77. 4. Avorn J. Approval of a tuberculosis drug From the Program on Regulation, Thera- approval of drugs and devices based on a paradoxical surrogate measure. peutics, and Law (PORTAL), Division of that are less safe or effective than JAMA 2013;309:1349-50. Pharmacoepidemiology and Pharmacoeco- 5. Dhruva SS, Bero LA, Redberg RF. Strength existing criteria would permit. nomics, Department of Medicine, Brigham of study evidence examined by the FDA in and Women’s Hospital and Harvard Medi- Over the past 80 years, this premarket approval of cardiovascular devic- cal School, Boston. country’s regulatory approach has es. JAMA 2009;302:2679-85. embraced steadily improving cri- This article was published on June 3, 2015, DOI: 10.1056/NEJMp1506964 teria for accurately assessing ther- at NEJM.org. Copyright © 2015 Massachusetts Medical Society.

n engl j med nejm.org 3 The New England Journal of Medicine Downloaded from nejm.org at PARTNERS HEALTHCARE on June 3, 2015. For personal use only. No other uses without permission. Copyright © 2015 Massachusetts Medical Society. All rights reserved. correspondence

— ocular ischemia resulting from the intrave- As discussed by Silberberg, the recent reports of nous injection of drugs of abuse. A 26-year-old VZV in the temporal arteries of patients with woman presented to our emergency department giant-cell arteritis are interesting and open the with blurred tunnel vision after the intravenous door to innovative avenues for treatment. We injection of cocaine and methamphetamine into hope that future clinical trials that examine the her neck vessels.1,2 She indicated that her use of efficacy of antiviral agents for the treatment of this site of injection was longstanding and that giant-cell arteritis will guide us in treating these she had unintentionally punctured her carotid patients more efficiently and successfully. artery on several occasions. On presentation, her Although it is true that elevated intraocular visual acuity was 20/30 in both eyes. On exami- pressure may play a role in some cases of peri- nation, visual-field testing revealed constriction operative anterior ION, it seems unlikely that the of peripheral vision in both eyes and no pupillary medical reduction of intraocular pressure during defects. Optic neuropathy with pallor of both op- the perioperative period would play a role in pre- tic disks was identified. The retinal arteries were venting visual loss. Although Rubin presents an markedly narrowed and beaded in appearance, attractive idea, it would probably be impossible and the retinal veins were mildly dilated. Pin- to design a study that would test his hypothesis point erythematous marks were noted at the in- given the rarity of perioperative visual loss and jection sites, and ultrasonography of the vessels the many mechanisms that may be simultane- and soft tissues on both sides of the neck re- ously involved in this complication. vealed no abnormality. We suggest that clini- Diffuse ocular ischemia (most often retinal cians obtain a detailed history of both recent and arterial ischemia) is indeed a classic complica- past injection-drug use when evaluating patients tion of the use of systemic vasoconstricting with acute or chronic optic neuropathy. drugs, particularly when directly injected into an Rita G. McKeever, M.D. artery. The case described by McKeever et al. is Michael I. Greenberg, M.D., M.P.H. interesting and rather unusual. The narrowing Jason R. Lange, M.D. of the retinal arteries in association with optic- Drexel University College of Medicine nerve pallor suggests retinal ischemia that is Philadelphia, PA probably the result of drug use or the presence [email protected] of particulate emboli in the retinal arteries over No potential conflict of interest relevant to this letter was reported. a period of months rather than the result of acute isolated ION. As suggested by the authors, 1. De Giorgi A, Fabbian F, Pala M, et al. Cocaine and acute vascular diseases. Curr Drug Abuse Rev 2012;5:129-34. there are numerous causes of ION that were not 2. Kaye S, McKetin R, Duflou J, Darke S. Methamphetamine detailed in our review, and it is of course impor- and cardiovascular pathology: a review of the evidence. Addic- tant to inquire about drug use in relation to any tion 2007;102:1204-11. vascular event. DOI: 10.1056/NEJMc1509058 Valerie Biousse, M.D. Nancy J. Newman, M.D. The Authors Reply: The limited length and spe- Emory University School of Medicine cific focus of our review did not allow us to com- Atlanta, GA [email protected] ment on many aspects of ION or to go into great- Since publication of their article, the authors report no fur- er depth on a variety of important related topics, ther potential conflict of interest. such as giant-cell arteritis and perioperative ION. DOI: 10.1056/NEJMc1509058

The 21st Century Cures Act

To the Editor: In their Perspective article (June (FDA) by giving it greater discretion to approve 25 issue),1 Avorn and Kesselheim argue that the drugs on the basis of less rigorous data. In particu- 21st Century Cures Act, which is currently being lar, the authors argue that the legislation would debated in Congress, would lower the regulatory authorize the FDA to “rely” on observational standards of the Food and Drug Administration analyses, which are less rigorous than random-

n engl j med 373;17 nejm.org october 22, 2015 1677 The New England Journal of Medicine Downloaded from nejm.org on October 21, 2015. For personal use only. No other uses without permission. Copyright © 2015 Massachusetts Medical Society. All rights reserved. The new england journal of medicine

ized controlled trials (RCTs). But the Cures Act Mark B. McClellan, M.D., Ph.D. does not diminish the FDA’s standards for requir- Brookings Institution ing that new medical products are safe and effec- Washington, DC tive. Rather, it recognizes that recent develop- Ellen V. Sigal, Ph.D. ments in genomics, systems biology, electronic Friends of Cancer Research Washington, DC data systems, and other fields can provide addi- [email protected] tional tools and resources to support better pre- Dr. McClellan reports serving on the board of directors of marketing and postmarketing regulation and Johnson & Johnson. No other potential conflict of interest rele- more efficient development of drugs and medical vant to this letter was reported. devices. 1. Avorn J, Kesselheim AS. The 21st Century Cures Act — will The authors note that the FDA now relies on it take us back in time? N Engl J Med 2015;372:2473-5. evidence beyond RCTs. Patients with coexisting DOI: 10.1056/NEJMc1509640 conditions or rare diseases are not studied much in traditional RCTs; further progress in precision To the Editor: Avorn and Kesselheim raise con- medicine is likely to make RCTs even more diffi- cerns about the 21st Century Cures Act, a bipar- cult. The Cures Act facilitates the use of new types tisan piece of legislation with provisions for bil- of evidence, enabling a more comprehensive lions in additional funding for the National understanding of risks and benefits for particular Institutes of Health (NIH), as well as a section on patients. The authors argue that such uses of the use of biomarkers as surrogate end points. It adaptive trials, Bayesian statistics, biomarkers is important to note that the FDA’s standards are and surrogate end points, and data from post- not altered by the bill. Rather, the bill will create marketing registries and surveillance systems will a solid scientific framework for the use of bio- adversely affect the FDA’s ability to approve safe markers for drug development. and effective drugs. However, as the authors state, Biomarkers can be precise and accurate mea- such tools have been valuable in many situations. sures of disease and efficacy. Phenylalanine, for For example, progress in therapies for the human example, is strongly associated with a decline in immunodeficiency virus and the hepatitis C virus IQ in patients with phenylketonuria. But IQ as reflects the use of validated biomarkers. The point an end point can take years to observe and can be of the legislation on biomarkers is to develop difficult to measure. Without the biomarker end better evidence on other markers that could be point, phenylketonuria treatments based on IQ valuable for evaluating treatments for currently would not be developed.1 unmet needs. Similarly, the point of developing The use of qualified biomarkers as surrogate better evidence on patient-reported outcomes, end points is crucial for quickly bringing thera- and better systems for studying clinical experi- pies to patients with rare diseases.2 The 21st ence, is to better assess the disease experience Century Cures Act provides hope to millions of of particular groups of patients. patients who for too long have surrendered their As such, the legislation’s provisions can in- lives to devastating rare diseases. crease the feasibility, efficiency, and influence of Emil Kakkis, M.D., Ph.D. RCTs by enabling them to be better targeted and Max G. Bronstein, M.P.P. more effectively designed — and perhaps to be EveryLife Foundation for Rare Diseases carried out in more routine clinical practice. The Novato, CA law empowers the FDA to use its expertise to [email protected] Dr. Kakkis reports being the chief executive officer of Ultra- guide the development of better science for the genyx, a biopharmaceutical company specializing in the devel- regulation of medical products. opment of treatments for rare diseases. No other potential con- Better evidence and up-to-date regulatory sci- flict of interest relevant to this letter was reported. ence are the best foundation for regulatory deci- 1. Kakkis ED, O’Donovan M, Cox G, et al. Recommendations sions and meaningful progress in biomedical for the development of rare disease drugs using the Accelerated Approval pathway and for qualifying biomarkers as primary innovation. They are also the best way to avoid endpoints. Orphanet J Rare Dis 2015;10:16. turning back the clock on new opportunities to 2. Fleming TR, DeMets DL. Surrogate end points in clinical develop safe and effective treatments for unmet trials: are we being misled? Ann Intern Med 1996;125:605-13. medical needs. DOI: 10.1056/NEJMc1509640

1678 n engl j med 373;17 nejm.org october 22, 2015 The New England Journal of Medicine Downloaded from nejm.org on October 21, 2015. For personal use only. No other uses without permission. Copyright © 2015 Massachusetts Medical Society. All rights reserved. correspondence

To the Editor: We agree with Avorn and Kes- Technology, September 2014 (https://www.whitehouse.gov/sites/ default/files/microsites/ostp/PCAST/pcast_carb_report_sept2014 selheim that increased funding for the NIH in the .pdf). 21st Century Cures initiative is a highly needed 3. Letter of support to Rep. Phil Gingrey and Rep. Gene Greene for step. However, we disagree with their opposition LPAD, February 26, 2014 (http://www.idsociety.org/uploadedFiles/ IDSA/Policy_and_Advocacy/Current_Topics_and_Issues/Anti to the Limited Population Antibacterial Drug microbial_Resistance/10x20/Letters/To_Congress/ADAPT%20 (LPAD) approval pathway, also in the legislation, group%20sign%20on%20letter%20FINAL.pdf). which facilitates the evaluation of new antibiot- DOI: 10.1056/NEJMc1509640 ics for serious infections for which current therapies are inadequate. Every year, at least 2 million Americans contract antibiotic-resistant infections, The Authors Reply: McClellan and Sigal contend and 23,000 die.1 For many of these infections, that 21st Century Cures “does not diminish the the limited number of patients and lack of ap- FDA’s standards,” but that is not true. Several pro- propriate comparator therapies make standard visions codify lower approval standards: section clinical trials impractical. The LPAD pathway was 2062 instructs the FDA to develop a process to recommended by the President’s Council of Advi- approve new uses for existing drugs on the basis sors on Science and Technology.2 This pathway is of lower-quality evidence than that provided by also supported by the Infectious Diseases Soci- clinical trials, including “experience,” “observa- ety of America (IDSA), representing more than tional studies,” and “registries”; section 2222 10,000 physicians and scientists.3 Antibiotics permits approval of high-risk devices on the ba- studied as proposed in the Cures bill would be sis of case studies of patients or poorly conducted approved only for that specific, limited popula- studies, as long as they are published in a jour- tion, similar to drugs for orphan diseases. Con- nal; and section 2121 appears to encourage the cerns expressed about approving potentially risk- approval of antimicrobials and antifungals on the ier drugs must be balanced against the greater basis of effects observed in laboratory tests or risk of not being able to provide the antibiotics preliminary studies involving small numbers of desperately needed by patients. patients.1 Section 2121 also permits the secretary Stephen B. Calderwood, M.D. of health and human services to apply this by- pass track to other drug categories if “public Massachusetts General Hospital Boston, MA health would benefit,” language open to abuse by future administrations inclined to further reduce Barbara E. Murray, M.D. FDA authority. Of course regulatory flexibility is University of Texas Medical School Houston, TX warranted to address urgent unmet needs, but the FDA already has this authority2 and uses it fre- Henry F. Chambers, M.D. quently.3 Once Congress starts providing de- University of California San Francisco School of Medicine tailed instructions for altering the FDA processes San Francisco, CA used to scientifically evaluate products (an odd Dr. Calderwood reports serving as the current president of the IDSA, having served on the scientific advisory boards of Pulma- proposition at best), such guidance could become trix and PharmAthene and holding stock in Pulmatrix and stock standard practice beyond the uncommon instanc- options in PharmAthene. Dr. Murray reports serving as the im- es in which these approaches’ benefits may out- mediate past president of the IDSA, receiving consulting fees from Rib-X Pharmaceuticals, GlaxoSmithKline, Theravance, and weigh their risks. the Innovative Medicines Initiative, and serving as a coinvestiga- We agree with Kakkis and Bronstein that high- tor on research funding provided to her institution by Astellas quality biomarkers can be essential for approving Pharma, Cubist Pharmaceuticals, Forest Laboratories, and Ther- avance. Dr. Chambers reports serving as the chair of the Anti- new drugs, particularly for rare diseases. How- microbial Resistance Committee of the IDSA, receiving consult- ever, the FDA can already approve new drugs on ing fees from Cubist Pharmaceuticals, AstraZeneca, Theravance, this basis, as it approved sapropterin (Kuvan) for and Pfizer, holding stock from Merck, and serving as an investi- gator in research funded by Cubist Pharmaceuticals. No other phenylketonuria on the basis of its effect on blood potential conflict of interest relevant to this letter was reported. phenylalanine levels. What’s needed is research to 1. Centers for Disease Control and Prevention. Antibiotic resis- discover and validate more such biomarkers, a tance threats in the United States, 2013 (http://www.cdc.gov/ prospect the bill advances only marginally, with drugresistance/pdf/ar-threats-2013-508.pdf). 2. Report to the President on combating antibiotic resistance. its modest proposed increase in NIH funding. Washington, DC: President’s Council of Advisors on Science and By contrast, pushing the FDA to approve drugs

n engl j med 373;17 nejm.org october 22, 2015 1679 The New England Journal of Medicine Downloaded from nejm.org on October 21, 2015. For personal use only. No other uses without permission. Copyright © 2015 Massachusetts Medical Society. All rights reserved. The new england journal of medicine

on the basis of biomarkers that are not rigor- treatment even as we advance the understanding ously linked to patient outcomes, as other proof the medications. visions do, can hurt patients by exposing them Aaron S. Kesselheim, M.D., J.D. to ineffective or unsafe treatments,4 wasting re- Jerry Avorn, M.D. sources, and giving false hope. Brigham and Women’s Hospital Calderwood et al. do not adequately consider Boston, MA the likelihood of substantial off-label use of anti- Since publication of their article, the authors report no further potential conflict of interest. biotics approved through the proposed “limited population” pathway, already common practice 1. 21st Century Cures Act (H.R. 6) (10 Jul 2015) (https://www .congress.gov/114/bills/hr6/BILLS-114hr6rfs.pdf). with certain drugs approved only for limited 2. Kesselheim AS, Tan YT, Darrow JJ, Avorn J. Existing FDA populations.5 Few clinical situations warrant pathways have potential to ensure early access to, and appropriate authorizing physicians to prescribe antibiotics use of, specialty drugs. Health Aff (Millwood) 2014;33:1770-8. 3. Darrow JJ, Kesselheim AS. Drug development and FDA ap- not known to improve clinical outcomes. Such a proval, 1938-2013. N Engl J Med 2014;370(26):e39. policy for antibiotics would be particularly self- 4. Kesselheim AS, Myers JA, Avorn J. Characteristics of clinical defeating, since it would induce resistance to trials to support approval of orphan vs nonorphan drugs for cancer. JAMA 2011;305:2320-6. those drugs and to other, related drugs. A better 5. Kesselheim AS, Myers JA, Solomon DH, Winkelmayer WC, solution would be to encourage the enrollment Levin R, Avorn J. The prevalence and cost of unapproved uses of of patients with serious infections in trials of top-selling orphan drugs. PLoS One 2012;7(2):e31894. investigational antibiotics to offer the chance of DOI: 10.1056/NEJMc1509640

Intravenous Immune Globulin for Statin-Triggered Autoimmune Myopathy

To the Editor: Although treatment with stat- immune myopathy evaluated at the Johns Hopkins ins may cause muscle-related symptoms in 10 to Myositis Center, 3 patients with diabetes declined 20% of patients, these symptoms usually resolve glucocorticoids because of concerns about poten- within weeks after the medication is stopped. tial side effects but agreed to try monotherapy with In rare instances, however, the medication causes IVIG, administered at a rate of 2 g per kilogram statin-triggered autoimmune myopathy, a condi- of body weight per month. Detailed clinical char- tion characterized by proximal muscle weakness, acteristics of these patients are shown in Table 1. prominent necrosis of muscle fibers (detected Immediately before IVIG, the mean (±SD) creatine on biopsy), elevated serum levels of creatine ki- kinase level for these patients was 4919±3523 IU nase, and the presence of autoantibodies that per liter, and all 3 patients had documented weak- recognize 3-hydroxy-3-methylglutaryl coenzyme ness in the proximal arms and legs. No infusion A (HMG-CoA) reductase, the pharmacologic tar- reactions occurred in any of the patients during get of statins.1-3 Moreover, statin-triggered auto- treatment. After two or three rounds of IVIG, the immune myopathy progresses despite the dis- mean creatine kinase level declined to 1125±1101 continuation of statins and requires control with IU per liter, quantitative dynamometry showed an immunosuppressive therapy. increase in the mean strength of arm abduction No clinical trials have been conducted to es- from 3.5 to 6.2 kg, and hip-flexion strength im- tablish effective treatments for statin-triggered proved or normalized. These gains persisted with- autoimmune myopathy. However, most clinicians out the addition of another agent. Between 9 and use glucocorticoids as first-line therapy. Statin- 19 months after starting IVIG, 2 patients had no triggered autoimmune myopathy can be especially subjective muscle-related symptoms and had difficult to treat; achieving remission frequently normal strength on examination. Patient 1 con- requires the addition of not only a second oral tinued to have mild hip-flexor weakness but agent (e.g., methotrexate) but also intravenous declined our advice to add another agent. immune globulin (IVIG).1,3,4 The mechanisms underlying the effects of Among 82 patients with statin-triggered auto- IVIG in statin-triggered autoimmune myopathy

1680 n engl j med 373;17 nejm.org october 22, 2015 The New England Journal of Medicine Downloaded from nejm.org on October 21, 2015. For personal use only. No other uses without permission. Copyright © 2015 Massachusetts Medical Society. All rights reserved.