correspondence
these initiatives forward. However, insofar as innovations overcome important limitations such as intellectual property and privacy, two points remain relevant. First, implementing responsible participantlevel data sharing is a moral imperative to accelerate discovery with limited resources and to address the ominous quality of scientific output. Current siloed and fragmented science fails to answer important questions in an increasingly complex world.1 Conversely, participant-level data sharing shifts the paradigm to collaboration, which enables the pooling of skills, insights, and resources from different teams.2 Second, concurrently with this wave of data availability, computational tools can now tackle big data opportunities. It is possible to manage large-scale data sets, reformat them to link and integrate, and construct analytic algorithms in an effective and timely fashion.3 A perfect storm is exposing a large volume of data to new tools,
paving the way to groundbreaking collaboration across networks to increase productivity and boost the quality of medical science. Mauricio Ferri, M.D. University of Calgary Calgary, AB, Canada [email protected]
Kald Abdallah, M.D., Ph.D. Sanofi Bridgewater, NJ Dr. Abdallah reports being an employee of and holding equity in Sanofi. No other potential conflict of interest relevant to this letter was reported. 1. Unreliable research: trouble at the lab. The Economist.
October 18, 2013 (http://www.economist.com/news/briefing/ 21588057-scientists-think-science-self-correcting-alarming -degree-it-not-trouble). 2. Eichler HG, Pétavy F, Pignatti F, Rasi G. Access to patientlevel trial data — a boon to drug developers. N Engl J Med 2013; 369:1577-9. 3. Schadt EE, Linderman MD, Sorenson J, Lee L, Nolan GP. Computational solutions to large-scale data management and analysis. Nat Rev Genet 2010;11:647-57. DOI: 10.1056/NEJMc1314515
Access to Patient-Level Trial Data To the Editor: The Chief Medical Officers Roundtable (CMOR), of which we are members, welcomes the Perspective article by Eichler et al. (Oct. 24 issue),1 who are representatives of the European Medicines Agency (EMA), on access to patient-level trial data. CMOR formulates positions on medical topics, and its members include chief medical officers of major biopharmaceutical companies. CMOR supports a transparent, harmonized process for access to patient-level clinical trial data. Any approach to clinical trial data sharing must be in the interest of patients. Data sharing should be based on two tenets. First, principles should apply uniformly to all who generate clinical trial data — industry, academia, regulators, health systems, foundations, and others. There will be little benefit to patients if access does not occur across sectors. Second, the responsible release of data requires input from independent experts and agreements to respect confidentiality, promote scientific excellence, refrain from misleading conclusions (which can harm patients if they discontinue beneficial treatment),2 and safeguard future innovation by retaining incentives for investigators. Access should be determined after the sub-
mission of a proposal to a panel that includes independent experts. Evaluation should be based on scientific merit, relevance, researcher qualifications, potential conflicts of interest, and plans for dissemination of findings after peer review. To this end, the CMOR supports the consensus study launched by the Institute of Medicine. It is, to our knowledge, the only broadly inclusive initiative and has the international participation of academia, industry, the National Institutes of Health, the Food and Drug Administration, the EMA, journals, patient organizations, foundations, and others.2 Contrary to the assertion by Eichler et al. that industry opposes the sharing of patient-level data, many companies are creating processes like those mentioned above.2 Industry already shares results through ClinicalTrials.gov, public websites, and scientific publications.3 Furthermore, the European Federation of Pharmaceutical Industries and Associations and the Pharmaceutical Research and Manufacturers of America have moved beyond the status quo in adopting the Principles for Responsible Clinical Trial Data Sharing, which will be implemented in January 2014 and include the following major points.4 First, patient- and study-level clinical
n engl j med 370;5 nejm.org january 30, 2014
485
The New England Journal of Medicine Downloaded from nejm.org at UNIVERSITY OF NEW SOUTH WALES on August 11, 2015. For personal use only. No other uses without permission. Copyright © 2014 Massachusetts Medical Society. All rights reserved.
The
n e w e ng l a n d j o u r na l
of
m e dic i n e
data, protocols, and clinical-study reports should Michael Rosenblatt, M.D. be made available on request to qualified re- Merck searchers pursuing legitimate research projects Whitehouse Station, NJ [email protected] for medicines and indications approved in the United States and the European Union. Second, for the Chief Medical Officers Roundtable Barron reports being an employee of Hoffmann–La Roche there should be public disclosure of synopses of andDr.Genentech and holding stock and stock options in Roche; clinical-study reports after approval in the United and Dr. Rosenblatt, being an employee of and holding stock and States and the European Union. Third, summa- stock options in Merck. No other potential conflict of interest ries that are written in lay language should be relevant to this letter was reported. provided to patients involved in clinical trials. 1. Eichler H-G, Pétavy F, Pignatti F, Rasi G. Access to patientlevel trial data — a boon to drug developers. N Engl J Med 2013; Fourth, procedures should be adopted to imple- 369:1577-9. ment these principles. And fifth, researchers need 2. Nisen P, Rockhold F. Access to patient-level data from to make a commitment to publish study results, GlaxoSmithKline clinical trials. N Engl J Med 2013;369:475-8. Rawal B, Deane BR. Clinical trial transparency: an assesswhether negative or positive. The CMOR fully 3. ment of the disclosure of results of company-sponsored trials supports the sharing of patient-level clinical trial associated with new medicines approved recently in Europe. data, by all who generate such data, in the inter- Curr Med Res Opin 2013 November 11 (Epub ahead of print). 4. EFPIA and PhRMA. Principles for responsible clinical trial est of patients. data sharing: our commitment to patients and researchers (http:// phrma.org/sites/default/files/pdf/PhRMAPrinciplesForResponsible Hal Barron, M.D. ClinicalTrialDataSharing.pdf).
Hoffmann–La Roche South San Francisco, CA
DOI: 10.1056/NEJMc1315673
Maraviroc and JC Virus–Associated Immune Reconstitution Inflammatory Syndrome To the Editor: The immune reconstitution inflammatory syndrome (IRIS) can be a serious complication in immunocompromised patients in whom progressive multifocal leukoencephalopathy (PML) has developed. IRIS occurs in such patients when their immunocompromised state reverses.1 Although restored immune surveillance is vital for clearing the JC virus that causes PML, improved management strategies for IRIS are needed because of its potential for considerable complications and death. The common, though unproven, use of glucocorticoids to treat IRIS may limit JC viral clearance.2 Because of the indirect implication of chemokine receptor 5– positive (CCR5+) T cells in IRIS pathophysiology,3 we used the small-molecule CCR5 antagonist maraviroc, which has been approved for the treatment of human immunodeficiency virus (HIV) infection to inhibit CCR5-dependent viral docking, in a patient with multiple sclerosis who presented with natalizumab-associated PML and features that were expected to predict the development of IRIS.4,5 A 49-year-old, HIV-negative woman with relapsing–remitting multiple sclerosis and a score of 2.5 on the Expanded Disability Status Scale 486
(EDSS, which ranges from 0 to 10, with higher scores indicating more severe disability) had received 43 infusions of natalizumab monotherapy over a period of 3.6 years. She received a diagnosis of PML after presenting with an episode of aphasia and seizure. She was considered at risk for severe IRIS because of the multilobar PML (Fig. 1A and 1B) and an elevated JC viral load in the cerebrospinal fluid (130,000 copies per milliliter) as measured by means of a polymerasechain-reaction assay (Focus Diagnostics). Oral maraviroc at a dose of 300 mg twice daily was initiated shortly after plasmapheresis. No glucocorticoids or other immunomodulating therapy was given throughout the course of the PML illness. The patient was monitored closely with frequent clinical and imaging evaluations. Serial cerebrospinal fluid analyses showed that maraviroc treatment was associated with a selective decrease in CCR5+ immune cells in the cerebrospinal fluid (see the Figure in the Supplementary Appendix, available with the full text of this letter at NEJM.org). These findings suggest that maraviroc functionally inhibits CCR5-dependent immune-cell trafficking into the central nervous system (CNS). The patient’s condition remained
n engl j med 370;5 nejm.org january 30, 2014
The New England Journal of Medicine Downloaded from nejm.org at UNIVERSITY OF NEW SOUTH WALES on August 11, 2015. For personal use only. No other uses without permission. Copyright © 2014 Massachusetts Medical Society. All rights reserved.
these initiatives forward. However, insofar as innovations overcome important limitations such as intellectual property and privacy, two points remain relevant. First, implementing responsible participantlevel data sharing is a moral imperative to accelerate discovery with limited resources and to address the ominous quality of scientific output. Current siloed and fragmented science fails to answer important questions in an increasingly complex world.1 Conversely, participant-level data sharing shifts the paradigm to collaboration, which enables the pooling of skills, insights, and resources from different teams.2 Second, concurrently with this wave of data availability, computational tools can now tackle big data opportunities. It is possible to manage large-scale data sets, reformat them to link and integrate, and construct analytic algorithms in an effective and timely fashion.3 A perfect storm is exposing a large volume of data to new tools,
paving the way to groundbreaking collaboration across networks to increase productivity and boost the quality of medical science. Mauricio Ferri, M.D. University of Calgary Calgary, AB, Canada [email protected]
Kald Abdallah, M.D., Ph.D. Sanofi Bridgewater, NJ Dr. Abdallah reports being an employee of and holding equity in Sanofi. No other potential conflict of interest relevant to this letter was reported. 1. Unreliable research: trouble at the lab. The Economist.
October 18, 2013 (http://www.economist.com/news/briefing/ 21588057-scientists-think-science-self-correcting-alarming -degree-it-not-trouble). 2. Eichler HG, Pétavy F, Pignatti F, Rasi G. Access to patientlevel trial data — a boon to drug developers. N Engl J Med 2013; 369:1577-9. 3. Schadt EE, Linderman MD, Sorenson J, Lee L, Nolan GP. Computational solutions to large-scale data management and analysis. Nat Rev Genet 2010;11:647-57. DOI: 10.1056/NEJMc1314515
Access to Patient-Level Trial Data To the Editor: The Chief Medical Officers Roundtable (CMOR), of which we are members, welcomes the Perspective article by Eichler et al. (Oct. 24 issue),1 who are representatives of the European Medicines Agency (EMA), on access to patient-level trial data. CMOR formulates positions on medical topics, and its members include chief medical officers of major biopharmaceutical companies. CMOR supports a transparent, harmonized process for access to patient-level clinical trial data. Any approach to clinical trial data sharing must be in the interest of patients. Data sharing should be based on two tenets. First, principles should apply uniformly to all who generate clinical trial data — industry, academia, regulators, health systems, foundations, and others. There will be little benefit to patients if access does not occur across sectors. Second, the responsible release of data requires input from independent experts and agreements to respect confidentiality, promote scientific excellence, refrain from misleading conclusions (which can harm patients if they discontinue beneficial treatment),2 and safeguard future innovation by retaining incentives for investigators. Access should be determined after the sub-
mission of a proposal to a panel that includes independent experts. Evaluation should be based on scientific merit, relevance, researcher qualifications, potential conflicts of interest, and plans for dissemination of findings after peer review. To this end, the CMOR supports the consensus study launched by the Institute of Medicine. It is, to our knowledge, the only broadly inclusive initiative and has the international participation of academia, industry, the National Institutes of Health, the Food and Drug Administration, the EMA, journals, patient organizations, foundations, and others.2 Contrary to the assertion by Eichler et al. that industry opposes the sharing of patient-level data, many companies are creating processes like those mentioned above.2 Industry already shares results through ClinicalTrials.gov, public websites, and scientific publications.3 Furthermore, the European Federation of Pharmaceutical Industries and Associations and the Pharmaceutical Research and Manufacturers of America have moved beyond the status quo in adopting the Principles for Responsible Clinical Trial Data Sharing, which will be implemented in January 2014 and include the following major points.4 First, patient- and study-level clinical
n engl j med 370;5 nejm.org january 30, 2014
485
The New England Journal of Medicine Downloaded from nejm.org at UNIVERSITY OF NEW SOUTH WALES on August 11, 2015. For personal use only. No other uses without permission. Copyright © 2014 Massachusetts Medical Society. All rights reserved.
The
n e w e ng l a n d j o u r na l
of
m e dic i n e
data, protocols, and clinical-study reports should Michael Rosenblatt, M.D. be made available on request to qualified re- Merck searchers pursuing legitimate research projects Whitehouse Station, NJ [email protected] for medicines and indications approved in the United States and the European Union. Second, for the Chief Medical Officers Roundtable Barron reports being an employee of Hoffmann–La Roche there should be public disclosure of synopses of andDr.Genentech and holding stock and stock options in Roche; clinical-study reports after approval in the United and Dr. Rosenblatt, being an employee of and holding stock and States and the European Union. Third, summa- stock options in Merck. No other potential conflict of interest ries that are written in lay language should be relevant to this letter was reported. provided to patients involved in clinical trials. 1. Eichler H-G, Pétavy F, Pignatti F, Rasi G. Access to patientlevel trial data — a boon to drug developers. N Engl J Med 2013; Fourth, procedures should be adopted to imple- 369:1577-9. ment these principles. And fifth, researchers need 2. Nisen P, Rockhold F. Access to patient-level data from to make a commitment to publish study results, GlaxoSmithKline clinical trials. N Engl J Med 2013;369:475-8. Rawal B, Deane BR. Clinical trial transparency: an assesswhether negative or positive. The CMOR fully 3. ment of the disclosure of results of company-sponsored trials supports the sharing of patient-level clinical trial associated with new medicines approved recently in Europe. data, by all who generate such data, in the inter- Curr Med Res Opin 2013 November 11 (Epub ahead of print). 4. EFPIA and PhRMA. Principles for responsible clinical trial est of patients. data sharing: our commitment to patients and researchers (http:// phrma.org/sites/default/files/pdf/PhRMAPrinciplesForResponsible Hal Barron, M.D. ClinicalTrialDataSharing.pdf).
Hoffmann–La Roche South San Francisco, CA
DOI: 10.1056/NEJMc1315673
Maraviroc and JC Virus–Associated Immune Reconstitution Inflammatory Syndrome To the Editor: The immune reconstitution inflammatory syndrome (IRIS) can be a serious complication in immunocompromised patients in whom progressive multifocal leukoencephalopathy (PML) has developed. IRIS occurs in such patients when their immunocompromised state reverses.1 Although restored immune surveillance is vital for clearing the JC virus that causes PML, improved management strategies for IRIS are needed because of its potential for considerable complications and death. The common, though unproven, use of glucocorticoids to treat IRIS may limit JC viral clearance.2 Because of the indirect implication of chemokine receptor 5– positive (CCR5+) T cells in IRIS pathophysiology,3 we used the small-molecule CCR5 antagonist maraviroc, which has been approved for the treatment of human immunodeficiency virus (HIV) infection to inhibit CCR5-dependent viral docking, in a patient with multiple sclerosis who presented with natalizumab-associated PML and features that were expected to predict the development of IRIS.4,5 A 49-year-old, HIV-negative woman with relapsing–remitting multiple sclerosis and a score of 2.5 on the Expanded Disability Status Scale 486
(EDSS, which ranges from 0 to 10, with higher scores indicating more severe disability) had received 43 infusions of natalizumab monotherapy over a period of 3.6 years. She received a diagnosis of PML after presenting with an episode of aphasia and seizure. She was considered at risk for severe IRIS because of the multilobar PML (Fig. 1A and 1B) and an elevated JC viral load in the cerebrospinal fluid (130,000 copies per milliliter) as measured by means of a polymerasechain-reaction assay (Focus Diagnostics). Oral maraviroc at a dose of 300 mg twice daily was initiated shortly after plasmapheresis. No glucocorticoids or other immunomodulating therapy was given throughout the course of the PML illness. The patient was monitored closely with frequent clinical and imaging evaluations. Serial cerebrospinal fluid analyses showed that maraviroc treatment was associated with a selective decrease in CCR5+ immune cells in the cerebrospinal fluid (see the Figure in the Supplementary Appendix, available with the full text of this letter at NEJM.org). These findings suggest that maraviroc functionally inhibits CCR5-dependent immune-cell trafficking into the central nervous system (CNS). The patient’s condition remained
n engl j med 370;5 nejm.org january 30, 2014
The New England Journal of Medicine Downloaded from nejm.org at UNIVERSITY OF NEW SOUTH WALES on August 11, 2015. For personal use only. No other uses without permission. Copyright © 2014 Massachusetts Medical Society. All rights reserved.
