A new meta-study finds the following:
“Of 34 clinical trials, none had adequate safety reporting, 19 were partially adequate, 12 were inadequate, and 3 reported no adverse events. Funnel plots were asymmetric with a bias toward lower odds ratio for sexual adverse effects, suggesting systematic under detection…”
From Wiki definition: A meta-analysis comprises statistical methods for contrasting and combining results from different studies in the hope of identifying patterns among study results, sources of disagreement among those results, or other interesting relationships that may come to light in the context of multiple studies.[1] Meta-analysis can be thought of as “conducting research about previous research.”
The report is interesting but a Meta Analysis is only as good as the clinical trials it is analyzing. All studies have its advantages and pitfalls. If adverse or sexual side effects were being examined, I question why other (much larger and comprehensive clinical trials) of finasteride 5mg was excluded.
This seems a little concerning…
As someone who has been in the biotech industry for over 30 years(overseeing clinical trials), the above post is lacking detail (andunreferenced) that would make it easier to understand. But, I dobelieve the poster’s observation, with some very important caveats. AsDr R notes, a meta-analysis is only as good as a trial it isevaluating. I have observed poorly conducted trials, most often“investigator-initiated” (university professor) rather than anindustry-sponsored trial for drugs where marketing approval is sought.With thousands of clinical trials, and many more supporting drugapproval, I suspect the above clinical trials are early-stage andpoorly done and not ones that are considered ‘pivotal” phase 3trials supporting drug approval. The latter studies undergo extremescrutiny by the regulators evaluating the trials (as evidenced by thesubmission packages that are freely available on the FDA web site).Even in an early-stage study I have never heard of a trial with noadverse events, given that an adverse event is any untoward changeregardless of causality (ie, someone can hit their leg on a table andhave pain for 2 seconds and that is an adverse event). Finally, if theabove quotes refer to sexual adverse events only, the key piece ofinfo (how many patients were in the meta-analysis) is missing. With a2-5% rate of an event, an uncontrolled trial of 30 patients is goingto miss certain events on average that a controlled trial of 4000patients would not. Finally, how an investigator asks about adverseevents (open-ended vs. checklist) has been shown to influence thenumber of events reported and is another critical feature that shouldbe considered in a meta-analysis evaluating rates of adverse events(assuming that the study populations are similar, which alsoinfluences reporting. If the “take away” message from the post aboveis that finasteride trials may underrepresent sexual advrese events, Iwould suspect that the above met-analysis did not include well-sonepivotal trials (by the nottaion that “none had adequate” safetyreporting.
I was able to locate the article that forms the basis of the post; itis published in the April edition of JAMA Dermatology and is freelyavailable to readers at:
http://www.ncbi.nlm.nih.gov/pubmed/?term=belknap+finasteride
A well-written article but needs proper interpretation. As I suspected(and noted in my earlier comment), most of the trials were smallstudies conducted by academic investigators (at universities) notseeking drug approval (rather than large, pivotal studies by drugcompanies). The “funding” column in Table 1 does not allow thisdistinction as many academic investigators doing their“investigator-initiated trials” may still receive funding from adrug company. But, academic-initiated drug studies – which are notsubmitted to regulators – are not overseen by drug companies,conducted completely indepently, and often do vary in safety reportingmethods. In the case of finasteride and alopecia, I would consider thestudies in the finasteride label to be more informative than, forexample, an academic center’s 33-patient open-label study (Example 3in Table 1).