WIN-R revisited: Response to editorial

Authors

  • Ira M. Jacobson M.D.,

    1. Division of Gastroenterology and Hepatology, Center for the Study of Hepatitis C, Weill Medical College of Cornell University, New York Presbyterian Hospital, New York, NY
    Search for more papers by this author
  • Robert S. Brown Jr M.D., M.P.H.

    1. Center for Liver Disease and Transplantation, Columbia University College of Physicians and Surgeons, New York Presbyterian Hospital, New York, NY
    Search for more papers by this author

  • Potential conflict of interest: Dr. Jacobson is a consultant for and advises Pfizer, GlaxoSmithKline, Human Genome Sciences, and Boehringer Ingelheim. He is a consultant for, advises, and received grants from Novartis, Coley Pharmaceuticals, and Vertex. He is a consultant for Intermune and Merck. He is also a consultant for, advises, is on the speakers' bureau of, and received grants from Gilead, Globelmmune, and Idenix. He is a consultant for, is on the speakers' bureau of, and received grants from Schering-Plough. He is a consultant for and received grants from Valeant. He is a consultant for, advises, and is on the speakers' bureau of Bristol-Myers Squibb.

To the Editor:

We thank Drs. Han and Smith for their thoughtful editorial1 that accompanied our 2 papers describing results of the WIN-R (Weight-Based Dosing of Peginterferon alfa-2b and Ribavirin) trial,2, 3 in which patients with chronic hepatitis C were treated with peginterferon (PEG-IFN) alfa-2b 1.5 μg/kg/week plus flat-dose (FD; 800 mg/day) or weight-based dose (WBD; 800–1400 mg/day) ribavirin (RBV). The authors appropriately focused on the importance and superiority of WBD RBV in hepatitis C virus genotype 1–infected patients and provided an excellent review of the treatment of African Americans. However, several points require correction or clarification.

The authors state that before 2001, the accepted dose of RBV was 800 mg. Actually, RBV was approved for use with standard, unmodified interferon alfa-2b at doses of 1000–1200 mg in the late 1990s. Later, RBV was approved in the United States for use with PEG-IFN alfa-2b at a dose of 800 mg; currently, the approved dose is 800–1200 mg, based on body weight, in the rest of the world. When using PEG-IFN alfa-2a, the approved RBV dose is 1000–1200 mg in the United States and abroad. Therefore, the statement “currently approved therapies recommend ‘standard’ WBD RBV in combination with PEG-IFN alfa-2a or 2b” is inaccurate. In fact, there is no “standard” WBD of RBV. A main objective of the WIN-R trial was to evaluate whether the regimen approved in many countries—PEG-IFN alfa-2b 1.5 μg/kg/week plus WBD RBV (800–1200 mg/day)—was superior to the U.S.-approved regimen of PEG-IFN alfa-2b 1.5 μg/kg/week plus RBV 800 mg/day. Further, because there are many heavy (>105 kg) patients in the United States, WIN-R also evaluated a novel, 1400-mg/day RBV dosage, which is not approved anywhere. Although not mentioned in the editorial, we reported equivalent sustained virologic response (SVR) rates in patients weighing >105 kg and receiving RBV 1400 mg/day compared with patients weighing ≤105 kg and receiving RBV 800–1200 mg/day, as well as equivalent hematologic safety across weight categories. We regard this as an important additional finding of the study that should lead to consideration of RBV 1400 mg in heavy patients weighing >105 kg.

The authors of the editorial ascribed to us a suggestion, which we did not make, that the WBD range of RBV used in our study was superior to that used in the study of Hadziyannis et al.,4 in which RBV 800 mg/day was compared with RBV 1000–1200 mg/day (both in combination with flat-dose PEG-IFN alfa-2a [180 μg/week]), based on a greater relative difference in SVR rates between the FD and WBD groups in our study. We made this comparison only with regard to African Americans with genotype 1 infection in the WIN-R trial,3 in whom the increment in SVR with WBD RBV was greater than in the overall patient cohorts of either study.2–4 The intent in our primary paper2 in citing the study by Hadziyannis et al.4 was to underscore the demonstration in the WIN-R trial that a broad WBD range of RBV, combined with weight-based PEG-IFN alfa-2b, produces equivalent SVR rates across a broad range of patient body weights, an issue not addressed in the Hadziyannis et al.4 study. Because this study is the only other one to compare a PEG-IFN alfa with WBD versus flat-dose RBV, we believed it was reasonable to examine the results of this study compared with ours. Further, Hadziyannis et al.4 reached similar conclusions concerning the difference between FD and WBD in both genotype 1 and 2/3 patients. Lastly, in the pivotal trial by Fried et al.,5 which also used flat-dose PEG-IFN alfa-2a and WBD RBV 1000–1200 mg, SVR rates were significantly affected by body weight, favoring patients weighing <75 kg. Additional studies are needed to address the comparative efficacy of PEG-IFN alfa-2a–based and PEG-IFN alfa-2b–based regimens in patients with various body weights. A large, multicenter, U.S.-based study comparing such regimens is nearing completion and may yield further insights into this issue.

The authors of the editorial suggested that use of erythropoiesis-stimulating agents led to the comparable discontinuation rates seen in WIN-R and the pivotal trial by Manns et al.6 In WIN-R, these agents were not permitted in lieu of dose reduction/discontinuation, and although our database admittedly was incomplete, their reported usage (9% in the WBD group and 5% in the FD group) does not support a substantial effect on discontinuation rates.

Finally, the authors of the editorial described the African American subanalysis as “post hoc.” Although WIN-R was not prospectively powered for this subanalysis, it was a protocol-specified secondary efficacy analysis and was appropriate to perform when the primary efficacy end point was met. Secondary data analyses are often not hypothesis-driven and the result of data “dredging”; as such, they can be prone to type 2 error. This analysis of WIN-R was an a priori, hypothesis-driven question and deserved individual consideration. Given the analytic strategy, the concordance with prior data, and the P value of 0.0006, the likelihood of type 2 error is extremely low. We suspect that analyses of important subgroups such as African Americans will and should continue to be nested in large clinical trials, because individual trials in African American patients are unlikely to be feasible or cost-effective.

Ira M. Jacobson M.D.*, Robert S. Brown Jr M.D., M.P.H.†, * Division of Gastroenterology and Hepatology, Center for the Study of Hepatitis C, Weill Medical College of Cornell University, New York Presbyterian Hospital, New York, NY, † Center for Liver Disease and Transplantation, Columbia University College of Physicians and Surgeons, New York Presbyterian Hospital, New York, NY.

Ancillary