How routine pharmacovigilance failed to identify finasteride's persistent sexual side effects
Michael S. Irwig
Andrology · 10:207-2082022
In plain terms
This short opinion article, by a US hormone specialist, looks at how the maker of finasteride and the US drug regulator handled early signs that the drug's sexual side effects could last after people stopped taking it. Company documents made public in 2021 show that by 2009 the maker knew of 278 reports of erection problems that had not gone away, but chose to keep relying on routine side-effect reporting for two more years instead of running new studies. The author argues this was a missed chance, and that the original trials were not set up to pick up rare sexual side effects.
What it doesn’t show: It can't tell us how often these lasting problems happen or whether the drug causes them: it is one doctor's opinion, based on company documents and earlier reports, and it adds no new patient data.
Summary (paraphrased)
This opinion article examined how Merck and the US FDA handled signals of persistent sexual adverse effects of finasteride, noting the first regulatory warnings in Sweden (2008) and the UK (2009) and two independent groups' peer-reviewed reports of persistent effects in 2011. Drawing on Merck documents unsealed in 2021, it described a 2009 Risk Management Plan that identified 278 cases of erectile dysfunction that had not recovered, after which Merck's safety team planned routine pharmacovigilance for another two years rather than further prospective studies. The FDA approved a label warning of erectile dysfunction continuing after discontinuation in 2011, effective 2012, but the author held it partly responsible for not requiring further safety studies. He argued that the original trials could not properly assess sexual adverse effects, as they used no validated sexual-function instruments, were not powered to detect adverse effects rarer than 1%, and usually did not disclose details of participants who had adverse effects or withdrew. He noted that the incidence of persistent effects was unknown and that adverse events were heavily under-reported.
Evidence
Extracted from the full text; page numbers refer to the paper. The library’s own notes are labelled as such.
- Design
- CommentaryOpinion article; single-author commentary on unsealed Merck documents, FDA labelling history and published literature
- Setting
- Merck and the US FDA, with reference to earlier Swedish and UK warnings; Merck documents unsealed in January 2021
- Population
- Not applicable; concerns men who used finasteride, mainly 1 mg/day for male pattern hair loss (Propecia)
- Size
- Not applicable (opinion article, no new data). Figures cited: 278 unrecovered erectile dysfunction reports in Merck's safety database; 3,210 patients in Merck's phase III trials of 1 mg/day; about 4.6 million men given finasteride 1 mg from 1998 to 2008; 10,295 serious adverse event reports for finasteride in FAERS as of June 2021
- Exposure
- Finasteride (Propecia 1 mg; Proscar also mentioned)
- Compared with
- None
- Outcome
- How persistent sexual side effects were detected, assessed and acted on by Merck and the FDA
- Follow-up
- Not applicable; covers events from 1998 to June 2021
Key results
- The first regulatory warnings of persistent sexual effects came from Sweden (2008) and the UK (2009); two independent groups published peer-reviewed reports of persistent effects in 2011 · p. 207
- Merck's Risk Management Plan (version 1.0, February 2009), discussed by its Risk Management Safety Team in June 2009, listed three potential risks (persistent erectile dysfunction, male infertility, depressive disorders) and identified 278 cases of erectile dysfunction that had not recovered; the plan noted that critical data were missing in most of them, while also describing several representative cases without apparent confounders · p. 207
- In March 2011 the FDA held a teleconference with Merck and approved adding erectile dysfunction continuing after discontinuation to the label; the labelling changes to Propecia and Proscar took effect in April 2012 · p. 207
- Merck's phase III trials of 1 mg/day included 3,210 patients with a mean exposure of 749 days; citing a meta-analysis, the author notes the trials used no validated sexual-function instruments, were not powered to detect adverse effects rarer than 1%, and usually did not disclose details of participants who had adverse effects or withdrew · pp. 207–208
- About 4.6 million men received finasteride 1 mg from 1998 to 2008 (Merck's estimate); the FDA's FAERS dashboard listed 10,295 serious adverse events for finasteride as of June 2021, and three of the four most common reaction types were sexual (erectile dysfunction, sexual dysfunction, decreased libido) · p. 208
- Merck's team planned routine pharmacovigilance for another two years rather than prospective studies with validated instruments; the author argues the FDA also bears some responsibility for not requiring further safety studies · p. 208
Limitations the authors note
- Most of Merck's 278 cases lacked critical data (time from stopping to report, other medicines, medical history), as Merck's own report said
- Much of the controversy comes from the inherent limitations of post-marketing data
- The incidence of persistent sexual effects is unknown, and adverse drug events are grossly under-reported
Also worth weighing (library’s note)
- A two-page opinion piece by one author, with no methods section and no systematic search of sources
- Its main source is litigation documents cited as a Merck supplementary file (ref. 5); the article quotes and summarises them, so their full context cannot be judged from the article alone
- The author co-wrote one of the two 2011 studies it cites as the first reports of persistent effects (ref. 2)
- FAERS figures are counts of reports, not rates; the author himself says the incidence is unknown
Funding: None declared Interests: None declared
What it can support (library’s note): An account of what Merck's 2009 risk management plan recorded about unrecovered erectile dysfunction and how Merck and the FDA responded. It provides no new data on how often persistent effects occur or whether finasteride causes them.
Why it’s in the corpus
Argues that Merck had a signal of persistent erectile dysfunction by 2009 but chose routine pharmacovigilance over further studies, and that the FDA did not require them. Essential context for the corpus on how post-drug syndromes can go unaddressed by standard safety surveillance.
Related records
- Peer-reviewed paper
Persistent Sexual Dysfunction and Suicidal Ideation in Young Men Treated with Low‐Dose Finasteride: A Pharmacovigilance Study
Ayad K. Ali, Balraj S Heran, Mahyar Etminan · Pharmacotherapy
Mining US FDA adverse-event reports (1998–2013) for men aged 18–45 on low-dose (1 mg) finasteride, the authors applied disproportionality analysis to detect signals for persistent sexual dysfunction and suicidal ideation. Of 4,910 reports, 577 described…
LIT-0172015PFSLiterature sweep · Oct 2026 - Regulatory document
Finasteride-dutasteride Article 31 referral — PRAC assessment report (EMA/PRAC/55203/2025)
European Medicines Agency, Pharmacovigilance Risk Assessment Committee (PRAC) · EMA (ema.europa.eu)
The full PRAC assessment behind the 2025 EU-wide review of finasteride and dutasteride, triggered by France's ANSM over suicidal ideation. The report evaluates 325 EudraVigilance cases of suicidal ideation (313 finasteride, 13 dutasteride) against roughly 270…
LIT-0652025-05PFSLiterature sweep · Oct 2026 - Regulatory document
Re: Docket No. FDA-2017-P-5787 — Response to the Post-Finasteride Syndrome Foundation citizen petition
U.S. Food and Drug Administration (FDA) · FDA (letter to PFS Foundation, hosted as PDF on citizen.org)
FDA's 2022 decision letter on the PFS Foundation's 2017 citizen petition asking for withdrawal of Propecia 1 mg or stronger labelling. The agency denied the petition in part and granted it in part. It recounts the 2010–2012 labelling history (depression added…
LIT-0722022-06-08PFSLiterature sweep · Oct 2026