Note: This post comes from one of our favorite readers (he has requested to remain anonymous), who has contributed posts in the past about FDA and clinical trial issues. Some of his previous contributions are here, here, here, and here.
He has broken down the latest finasteride side effect study for our readers and explains why interpreting the data is problematic:
This month an article appeared in a medical journal (Irwig MS. Journal of Sexual Medicine, 2012 Epub ahead of print) that relates to persistent sexual side effects of finasteride and has generated discussion in your blog from both posters and commenters. An abstract of the paper can be found at: Persistent Sexual Side Effects of Finasteride: Could They Be Permanent?
I am a physician scientist (MD-PhD) who spent over 20 years conducting clinical trials and interpreting clinical data. Because I believe this is an important issue, I wish to summarize this study (and discuss its strengths and limitations) for your readers. Two biases that I am disclosing: (1) I believe that persistent sexual side effects of finasteride exist, and (2) I also believe that well-done published scientific investigations on this phenomenon are lacking and readers (most who are unfamiliar with scientific investigations) historically come to form impressions from magazine and newspaper articles, lawyer-run web sites, and blogs… none of which are ideal to better define and understand this phenomenon.
Fifty-four men were recruited. Most men (undefined number) were recruited from an internet site that focuses on sexual side effects from Propecia. Other men (number undefined) were recruited from the author’s clinical practice. To participate in the study, all men were required to have had their finasteride use before age 40 and (by their self-report) and have no history of sexual dysfunction, medication use (other than antibiotics), or significant medical or psychiatric conditions at the time that they began taking finasteride. Initial information for the study was collected by telephone or Skype. Info on sexual dysfunction, medication use, significant medical or psychiatric conditions, etc before and after finasteride use was collected. Follow-up emails were sent to the participants 9-16 months after initial interview (average 14 months). Multiple valid questionnaire scales of sexual dysfunction were used.
The average age of the men at time of initial interviews was 31 years, and the average age when finasteride was begun was 26 years. Average duration of finasteride use was 23 months. Most of the subjects lived outside of the United States. At the time of the interviews, duration of persistent side effect ranged from 3-6 months (7% of subjects), 7-11 months (9%), 1-2 years (44%), 3-5 years (19%), and 6 or more years (20% of subjects). Most men had sexual dysfunction scale scores that showed significant greater dysfunction following initiation of finasteride.
This study confirms prior reports of persistent sexual dysfunction in men using finasteride. The study also has significant methodological limitations. These limitations do not invalidate the phenomenon, but make interpretation challenging: men (mostly from outside the US) were recruited by an internet site to self-report historical (and some prospective) data. While the author notes that some patients are recruited from his own practice, he does not report (or perhaps have access to) medical records for most of the recruited men and no information is available on their medical work-up. Thus, the reader is left wondering what sort of evaluation for differential diagnosis (and what medical evaluations) these men received.
The author doesn’t describe the countries that these men are from (which can affect their level of medical evaluation). This is not the author’s fault; this is a single author, single site, questionnaire/telephone/email study relying on the subject’s self-report and not presented as otherwise. The author should be commended for doing this study. As the author notes (page 5) “an important limitation to this study is selection bias, in that the most affected by finasteride are more likely to participate in a study such as this one.” I would also add that bias exists in that a study relying on “self-report” in the absence of medical records introduces potential error.
Finally, a prospective study in larger numbers of men would be able to better define the incidence of such events, obtain detailed medical evaluations that can rule out other causes of the dysfunction (psychologic, hormonal, vascular, etc), and perhaps identify factors at initiation of drug use that increase the probability of getting these effects. I am hopeful that such studies – considerably more expensive and involved – be conducted. I am also hopeful that the “pro-Propecia permanent sexual dysfunction” group understands that anecdotes and surveys pale in comparison to more robust and well done scientific investigations, which should be conducted, appear in top journals, and are ultimately in their best interest.