Letter from the DSMC regarding a clinical trial of lutein in patients with retinitis pigmentosa

prospective pathological study could better describe the in- cidence of this malformation and its clinical correlates. Tina Rutar, MD Susan Huang, MD Michele Bloomer, MD J. Brooks Crawford, MD Author Affiliations: Departments of Ophthalmology (Drs Rutar, Huang, Bloomer, and Crawford) and Pediatrics (Dr Rutar), University of California, San Francisco. Correspondence: Dr Rutar, Department of Ophthalmol- ogy, Division of Pediatric Ophthalmology and Strabis- mus, University of California, San Francisco, 10 Koret Way, K 301, San Francisco, CA 94143-0730 (rutart @vision.ucsf.edu). Financial Disclosure: None reported. 1. Spencer WH. Primary neoplasms of the optic nerve and its sheaths: clinical features and current concepts of pathogenetic mechanisms. Trans Am Oph- thalmol Soc. 1972;70:490-528. 2. Margo CE, Kincaid MC. Angiomatous malformation of the retrolaminar op- tic nerve. J Pediatr Ophthalmol Strabismus. 1988;25(1):37-39. Letter From the DSMC Regarding a Clinical Trial of Lutein in Patients With Retinitis Pigmentosa W e, the members of the Data Safety Monitor- ing Committee (DSMC) for Berson and col- league’s clinical trial of lutein in patients with retinitis pigmentosa who are receiving vitamin A, 1 share many of the concerns Massof and Fishman 2 expressed in their editorial. We served as the DSMC from 2002 through 2009. We reviewed the protocol, the statistical analysis plan, and the emerging data. We were im- pressed by the conduct of the trial, especially the excel- lent patient retention and adherence to the protocol. We reviewed and approved the manuscript before the authors submitted it for publication; however, we note some substantive changes made between the time of our review and the time of publication. For example, the ar- ticle’s new section on “Conclusions” is not consistent with our interpretation of the data, which emphasizes that the trial showed no effect of lutein on the primary outcome. We have carefully evaluated the data from the trial and view that the authors’ conclusion and the section on “Ap- plication to Clinical Practice” overstate the strength of evidence for the use of lutein. We wish to remind the clini- cal community that the evidence adduced for benefit comes from one of several secondary outcomes in a trial in which the primary outcome showed no evidence of benefit (the P value for the effect on Humphrey field ana- lyzer 30-2 field, dB/y was .66). Janet Wittes, PhD Michael B. Gorin, MD, PhD Susan T. Mayne, PhD Cynthia S. McCarthy, DHCE, MA Paul Sternberg Jr, MD Michael Wall, MD Author Affiliations: Statistics Collaborative, Inc, Wash- ington, DC (Dr Wittes); Jules Stein Eye Institute- University of California, Los Angeles (Dr Gorin); De- partment of Epidemiology and Public Health, Yale University, New Haven, Connecticut (Dr Mayne); Pri- vate consultant, Glenshaw, Pennsylvania (Ms McCarthy); Vanderbilt University Medical Center, Nashville, Ten- nessee (Dr Sternberg); and University of Iowa College of Medicine, Iowa City (Dr Wall). Correspondence: Dr Wittes, Statistics Collaborative, Inc, 1625 Massachusetts Ave NW, Ste 600, Washington, DC 20036 (janet@statcollab.com). Financial Disclosure: None reported. 1. Berson EL, Rosner B, Sandberg MA, et al. Clinical trial of lutein in patients with retinitis pigmentosa receiving vitamin A. Arch Ophthalmol. 2010;128 2. Massof RW, Fishman GA. How strong is the evidence that nutritional supple- ments slow the progression of retinitis pigmentosa [editorial]? Arch Ophthalmol. In reply The DSMC acknowledges approval of our draft manuscript. The publication 1 contained what we regard as minor adjust- ments requested by the journal, including a “Conclusion” sec- tion in the “Abstract” that restated results. The “Application to Clinical Practice” section was in the draft manuscript ap- proved by the DSMC. Throughout the publication we stated that the treatment effect of lutein was observed only on the secondary endpoint of midperipheral field sensitivity. The DSMC suggests that if significant differences be- tween the treatment groups are not seen with respect to the primary endpoint, then the results of the trial are negative and should have little or no affect on clinical practice. Pre- cedent exists for modifying practice based on results seen with secondary endpoints and subgroup analyses. 2,3 In the Physicians’ Health Study evaluating aspirin, the paucity of cardiovascular deaths led to revision of the primary end- point to include nonfatal myocardial infarction; aspirin was then found effective in preventing primary heart attacks. 2 The Women’s Health Study assessed aspirin’s efficacy in pre- venting heart attack in women older than 45 years. Al- though results of the analysis of the entire study cohort were negative, subgroup analyses showed that aspirin reduced the risk of major cardiovascular events, ischemic stroke, and myocardial infarction in women older than 65 years. 3 Similarly, results of the lutein trial should not be con- sidered negative simply because the beneficial effect was based on a secondary endpoint. A significant benefit of lutein on preserving midperipheral field sensitivity was observed in randomized comparisons using both parametric (P=.05) and nonparametric analyses (P = .03). Furthermore, observa- tional analyses showed that those with the highest serum lutein level and those with the highest increase in intrareti- nal macular pigment optical density (ie, a measure of in- traretinal lutein) had the least decline in midperipheral field sensitivity (P=.01 and P=.006, respectively). 1 Based on our results, 1,4,5 we reaffirm that most adults with typical retinitis pigmentosa should take 15 000 IU/d of vi- tamin A palmitate. They should avoid high-dose vitamin E supplementation. 4 Adults who start taking vitamin A for the first time should also take 1200 mg/d of docosahexaenoic acid (DHA) for 2 years; after 2 years, they should stop tak- ARCH OPHTHALMOL / VOL 129 (NO. 5), MAY 2011 WWW.ARCHOPHTHALMOL.COM ©2011 American Medical Association. All rights reserved. Downloaded From: http://jamanetwork.com/pdfaccess.ashx?url=/data/journals/ophth/10240/ by a University of California - Los Angeles User on 04/06/2017