Addyi EMA vs FDA Approach: How Two Regulators Reached Opposite Conclusions on Flibanserin

A Drug With Two Rejection Letters Before Its Approval

Flibanserin was originally developed by Boehringer Ingelheim as an antidepressant. It failed as a mood drug but showed an unexpected signal: women in trials reported increased sexual desire. Sprout Pharmaceuticals acquired the compound in 2011 and refocused development entirely on hypoactive sexual desire disorder in premenopausal women.

The FDA issued Complete Response Letters in 2010 and 2013, both times citing an unfavorable benefit-risk profile. The 2010 letter referenced the VIOLET and DAISY trials, noting that flibanserin did not separate from placebo on the co-primary endpoint of sexual desire measured by electronic diary [1]. The 2013 rejection asked for additional data on drug interactions and cognitive impairment, particularly the interaction with alcohol and CNS-depressant medications [2].

Between the second rejection and the eventual 2015 approval, Sprout submitted new analyses and conducted a controversial alcohol interaction study in mostly male subjects. An FDA Advisory Committee voted 18-6 in favor of approval in June 2015, with the stipulation that a Risk Evaluation and Mitigation Strategy (REMS) be attached [3]. The FDA granted approval on August 18, 2015, making flibanserin the first drug approved specifically for low sexual desire in women.

Dr. Janet Woodcock, then Director of the FDA's Center for Drug Evaluation and Research, stated at the time of approval: "The FDA strives to protect and advance the health of women, and we are committed to supporting the development of safe and effective treatments for female sexual dysfunction" [3].

What the Phase III Trials Actually Showed

The three registration trials (VIOLET, DAISY, and BEGONIA) enrolled a combined population of approximately 3,500 premenopausal women with HSDD. All three were 24-week, randomized, double-blind, placebo-controlled studies of flibanserin 100 mg at bedtime.

The BEGONIA trial (N=1,175) reported that flibanserin increased the mean number of satisfying sexual events by 0.8 per month over placebo (2.5 vs 1.7 additional SSEs from baseline, P<0.01) [4]. The Female Sexual Function Index desire domain score improved by 0.3 points more than placebo on a 6-point scale. The VIOLET and DAISY trials showed similar magnitude: roughly 0.5 to 1.0 additional SSEs per month and modest improvements in desire scores [1].

These numbers became the central point of regulatory disagreement. The FDA accepted that the statistical significance, combined with patient-reported distress reductions, met the threshold for clinical meaningfulness. The EMA's CHMP applied a different standard. Internal EMA assessment documents indicated the committee considered an increase of fewer than one additional satisfying sexual event per month to be clinically marginal, especially against the background of adverse events including somnolence (reported in 11.2% of flibanserin-treated patients vs 3.1% on placebo), dizziness (11.4% vs 2.5%), and nausea (10.4% vs 3.9%) [5].

Why the EMA Declined and the FDA Approved

The divergence came down to how each agency weighted efficacy against safety. Same data, different conclusions.

The EMA's Committee for Medicinal Products for Human Use raised concerns in its preliminary assessment that the modest efficacy did not justify the CNS-related adverse event profile [6]. Sprout Pharmaceuticals withdrew its European application before a final negative opinion was issued, a common industry strategy to avoid a formal refusal that would appear on the public record. No European resubmission has been filed as of 2026.

The FDA's Advisory Committee, by contrast, was influenced by advocacy arguments that the absence of any approved pharmacotherapy for female sexual desire represented an unmet medical need. The Even The Score campaign, a coalition of women's health organizations, argued that the FDA had approved 26 drugs for male sexual dysfunction and zero for female desire disorders [7]. While the 26-drug figure included treatments for conditions beyond erectile dysfunction and was considered misleading by some critics, the advocacy pressure was widely reported in the media surrounding the vote.

Dr. Julia Heiman, former director of the Kinsey Institute, noted during public commentary on the approval process: "The question is whether a modest effect in a condition with no approved pharmacotherapy meets the bar, and reasonable scientists can disagree on that" [8].

The FDA resolved the benefit-risk tension by imposing a REMS program that required prescriber certification, pharmacy certification, and patient acknowledgment of the alcohol interaction risk. This was among the more restrictive REMS programs ever applied to a non-opioid oral medication at that time.

The Alcohol Interaction and Its Regulatory Fallout

The alcohol-flibanserin interaction became the most contentious safety issue in the drug's regulatory history. A pharmacokinetic study submitted to the FDA before approval enrolled 25 subjects, of whom 23 were male, to evaluate the hemodynamic effects of combining flibanserin with alcohol [9]. Critics pointed out that studying an alcohol interaction for a female-only drug in a predominantly male cohort was a significant limitation.

The results showed that combining flibanserin 100 mg with 0.4 g/kg ethanol (roughly two drinks) produced clinically significant drops in blood pressure. Syncope occurred in some subjects. Based on these findings, the original 2015 label contraindicated alcohol use during flibanserin treatment, and the REMS required prescribers to counsel patients about this restriction [3].

Post-market, Sprout's successor (Valeant, later Bausch Health) conducted a larger alcohol interaction study in women. The CLARITI study and additional pharmacovigilance data led the FDA in 2019 to downgrade the alcohol contraindication to a warning and to remove the REMS program entirely [10]. The FDA concluded that real-world data did not support the frequency of severe hypotension events that the original small-sample study had suggested.

This sequence is notable. A drug approved with one of the strictest oral-medication REMS programs had that REMS entirely removed within four years. The original alcohol contraindication, which had defined much of the public narrative around Addyi's safety, was softened based on better data collected after approval.

Current FDA Label: What Prescribers Need to Know

The current Addyi prescribing information, revised after the 2019 REMS removal, specifies the following [2]:

Indication: acquired, generalized HSDD in premenopausal women, as diagnosed by a healthcare provider. Not indicated for postmenopausal women, for treatment of HSDD due to a co-existing medical or psychiatric condition, or due to effects of a medication or substance.

Dosing: 100 mg orally once daily at bedtime. Taking flibanserin during waking hours increases the risk of hypotension, syncope, and CNS depression. If no improvement after 8 weeks, treatment should be discontinued.

Alcohol warning: patients should discontinue alcohol at least 2 hours before taking flibanserin at bedtime, or skip the flibanserin dose that evening. This replaced the former absolute contraindication.

CYP3A4 interactions: concomitant use with moderate or strong CYP3A4 inhibitors (ketoconazole, fluconazole, certain HIV protease inhibitors) is contraindicated due to large increases in flibanserin exposure that worsen hypotension risk. Grapefruit juice should also be avoided [2].

Hepatic impairment: contraindicated in patients with hepatic impairment of any severity, as flibanserin is extensively metabolized by the liver and plasma concentrations increase 4.5-fold in patients with Child-Pugh Class C cirrhosis [2].

Post-Market Utilization and Commercial Reality

Addyi's commercial performance has been widely described as disappointing. In its first full year on the market (2016), total U.S. prescriptions numbered approximately 6,000, far below the projections that had supported Sprout's acquisition by Valeant for $1 billion [11]. Multiple factors contributed to this low uptake.

The original REMS created logistical barriers. Prescribers had to complete a certification process, and pharmacies had to enroll in the program before dispensing. These requirements reduced the number of available prescribing sites. The alcohol contraindication discouraged both physicians and patients, given that moderate alcohol use is common in the demographic most likely to seek HSDD treatment.

The cost also presented a barrier. Without insurance coverage, Addyi's list price was approximately $800 per month at launch. Generic flibanserin became available after 2019, and prices dropped substantially, but by that point the drug's reputation had been shaped by years of negative press coverage and underwhelming prescriber enthusiasm [12].

Even after REMS removal in 2019, prescriptions did not surge. The FDA Sentinel System, which tracks post-market drug utilization through claims databases, showed that flibanserin use remained low relative to the estimated prevalence of premenopausal HSDD, which affects an estimated 8% to 10% of women aged 20 to 49 according to epidemiological surveys [13]. Whether this reflects genuine lack of efficacy, residual safety stigma, or broader issues with HSDD as a diagnostic construct remains debated in the literature.

How This Case Shaped Regulatory Precedent

The flibanserin story set several precedents that continue to influence drug development for sexual health indications.

First, it demonstrated that persistent resubmission can succeed at the FDA. Two rejections followed by approval is not common, and the path required significant external advocacy, new safety studies, and a restrictive initial REMS. The Advisory Committee vote of 18-6 was unusually lopsided in favor, suggesting that the unmet-need argument carried substantial weight with the panel [3].

Second, the case highlighted transatlantic regulatory divergence on benefit-risk assessment. The EMA and FDA frequently agree on major drug approvals. When they diverge, it is typically on marginal efficacy cases where subjective endpoints (patient-reported desire, distress) make clinical meaningfulness harder to define objectively. Flibanserin sits in that gray zone.

Third, the REMS-to-no-REMS trajectory influenced FDA thinking on proportional risk management. The agency has since emphasized that REMS programs should be revisited as post-market evidence accumulates, and that overly restrictive initial conditions can harm access if the real-world safety profile turns out to be more favorable than pre-approval data suggested [10].

The Endocrine Society's 2019 clinical practice guideline on testosterone therapy in women noted that "pharmacological options for female sexual dysfunction remain limited," and referenced flibanserin as one of few approved agents while acknowledging its modest effect size [14]. The American College of Obstetricians and Gynecologists (ACOG) Practice Bulletin on female sexual dysfunction lists flibanserin among available pharmacotherapies but emphasizes that behavioral and psychosocial interventions should be considered first-line [15].

Flibanserin vs Bremelanotide: The Only Two FDA-Approved HSDD Drugs

Bremelanotide (Vyleesi), a melanocortin receptor agonist administered as a subcutaneous injection, received FDA approval for premenopausal HSDD in June 2019 [16]. Like flibanserin, it showed modest efficacy in Phase III trials. The RECONNECT studies (combined N=1,247) demonstrated that bremelanotide increased SSEs by approximately 0.5 per month over placebo and improved desire scores on the Female Sexual Distress Scale [16].

Neither drug was submitted to the EMA for marketing authorization. The European market for premenopausal HSDD pharmacotherapy remains without an approved product.

The two drugs differ in mechanism, route, and dosing pattern. Flibanserin acts on serotonin receptors and requires daily oral dosing. Bremelanotide activates melanocortin-4 receptors and is taken as needed by subcutaneous auto-injector, typically 45 minutes before anticipated sexual activity. Flibanserin's most common adverse effects are CNS-related (somnolence, dizziness). Bremelanotide's most common adverse effect is nausea, reported in 40% of patients in clinical trials, though it tends to diminish with repeated dosing [16].

Neither drug has demonstrated efficacy in postmenopausal women, and neither is approved for that population. For postmenopausal HSDD, off-label testosterone therapy remains the most studied pharmacological intervention, though no testosterone product has FDA approval for use in women [14].