© 2000 by Oxford University Press
Journal of the National Cancer Institute, Vol. 92, No. 3, 274B-275,
February 2, 2000
© 2000 Oxford University Press
CORRESPONDENCE |
RESPONSE: Re: Vitamin A Analogue for Breast Cancer Prevention: a Grade of F or Incomplete?
Correspondence to: Steven Piantadosi, M.D., Ph.D., Oncology Biostatistics, The Johns Hopkins Oncology Center, The Johns Hopkins University School of Medicine, 550 N. Broadway, Suite 1103, Baltimore, MD 21205-2013.
It's not often that a favorable editorial gets taken to task by the recipient. I hope that other readers did not miss my points or get them backwards, as Veronesi et al. did. Careful reading, time to digest, and perhaps even e-mail might have prevented their perversion. In any case, I will ignore the ad hominem in their letter in favor of the following comments. Dr. Meyskens objects to the title alone, but oddly didn't reconsider it, even after finding the content "thoughtful." His concerns are obviated below.
First, the title. It clearly interrogates "vitamin A analogue" (fenretinide) as a drug, not the fenretinide trial or the work of the authors. Since when is querying the efficacy of a drug considered "sensationalist" and "yellow journalism"? It is only because the trial was well done, as I said explicitly, that one can make inferences about the drug. The reasonableness of this title is further emphasized below.
Second, the methodologic danger as I see it is that the findings of the trial with respect to the treatment-covariate interaction will be dismissed outright by many. This dismissal is because of the small magnitude, marginal significance, and post hoc pedigree of the test. Those who take such a harsh view would rate fenretinide as a proven failure. The editors of the Journal correctly saw this as an issue, which is why a trial statistician was asked to comment. My editorial argued against dismissal, again explicitly.
Third, are Veronesi et al. (or Meyskens) more heavily invested in the reality of a beneficial fenretinide effect than the manuscript suggests? It is hardly proven by the putative biologic mechanism—but it is given support. It is my opinion, a statement of which is the right and purpose of an editorial, that additional empirical data will be required to establish the truth of the observation. Hence, my assessment that the tale of fenretinide is incomplete. Even Veronesi et al. say in their complaint that we need further studies to address the new hypothesis. So, asking if fenretinide is a failure or if the information is merely incomplete captures the essence of the issue.
Finally, I am not responsible for the poor reading habits of journalists.
Perhaps with these points emphasized, Veronesi et al. can regain their composure, recognize a serious inferential issue, appreciate a favorable review of their study, and continue with the important work of developing prevention strategies for breast cancer.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||