How long was the BEGONIA open-label extension?

The extension ran for 28 weeks following the 24-week double-blind core trial, giving a total treatment exposure of up to 52 weeks for women originally randomized to flibanserin.

Did flibanserin's benefits increase over time in the extension?

No. Efficacy appeared to plateau by approximately week 8 and remained stable through week 52. Women who did not respond early were unlikely to become late responders.

Was there a placebo arm during the BEGONIA extension?

No. The extension was single-arm and open-label, with all participants receiving flibanserin 100 mg nightly. This makes it impossible to definitively separate drug effect from placebo response or regression to the mean.

What percentage of women dropped out during the extension?

Approximately 18% of extension participants discontinued before completing the full 52 weeks, with adverse events (primarily somnolence and dizziness) accounting for roughly half of discontinuations.

Did the alcohol interaction warning come from BEGONIA data?

No. The alcohol-flibanserin interaction was identified in a separate pharmacokinetic study and post-marketing surveillance. BEGONIA did not systematically evaluate alcohol co-administration.

How large was the placebo response in the BEGONIA core trial?

Placebo-treated women gained approximately 1.5 additional SSEs per month, which represented about 65% of the total improvement seen in the flibanserin arm. This high placebo response complicates interpretation of all efficacy data.

Who is most likely to benefit from flibanserin based on BEGONIA and pooled data?

Post-hoc analyses suggest women with lower baseline SSE counts and higher baseline sexual distress show larger treatment effects, though it remains unclear whether this reflects true pharmacological subgroup differences or statistical regression effects.

Is flibanserin safe to take long-term based on BEGONIA extension data?

Through 52 weeks, no new safety signals emerged beyond the known CNS effects (somnolence, dizziness). CNS side effects tended to diminish after the first 8 weeks. The alcohol interaction, discovered separately, is the most clinically significant long-term safety concern.

Did the FDA consider the extension data when approving Addyi?

The FDA treated extension data as supportive, not confirmatory. The primary basis for approval rested on the double-blind, placebo-controlled core phases of BEGONIA and the sister trials DAISY and VIOLET.

Should patients take drug holidays from flibanserin?

Guidelines from ISSWSH recommend periodic reassessment of treatment necessity. The BEGONIA extension did not formally study drug holidays, but given the modest effect sizes, clinicians may reasonably trial discontinuation to confirm ongoing benefit.

BEGONIA Extension Data and What Happened After the Trial Ended

By HealthRX Medical Team

Published May 25, 2026Updated May 25, 2026Last reviewed May 25, 2026

At a glance

| Parameter | Detail | |---|---| | Trial name | BEGONIA | | N | 949 (randomized) | | Intervention | Flibanserin 100 mg nightly | | Comparator | Placebo | | Duration | 24-week double-blind + 28-week open-label extension | | Primary endpoint | Change in satisfying sexual events (SSEs) and sexual desire score (eDiary) | | Key result | Statistically significant but clinically modest improvement in SSEs (~0.8 additional events/month vs. placebo) |

Why the Extension Phase Matters More Than the Core Trial

The 24-week BEGONIA core trial established that flibanserin 100 mg nightly produced a statistically significant increase in SSEs compared to placebo in premenopausal women with generalized, acquired HSDD. The effect size was small: roughly 0.8 additional SSEs per month over placebo. That number alone tells you very little about whether the drug works in practice. What matters clinically is whether women who responded kept responding, whether non-responders eventually benefited, and whether the side effect profile shifted with chronic dosing.

The open-label extension enrolled women who completed the 24-week blinded phase, offering all participants flibanserin 100 mg nightly for an additional 28 weeks. This design, common in HSDD trials, creates an interpretive challenge. Without a continued placebo arm, any apparent "durability" could reflect sustained drug effect, regression to the mean, or the psychological impact of knowing you are receiving active treatment.

Methodology Notes the Abstract Does Not Mention

Several design features of the BEGONIA extension deserve scrutiny that typical summaries skip.

Enrichment bias in the extension cohort. Only women who completed the full 24-week blinded phase were eligible for the extension. This excluded dropouts from adverse events (somnolence, dizziness, nausea) and those who withdrew due to perceived lack of efficacy. The extension population was, by definition, more tolerant of flibanserin and potentially more responsive. Any efficacy signal from weeks 24 through 52 must be read with this survivorship filter in mind.

No re-randomization. The extension was single-arm, open-label. Women previously on placebo who then received flibanserin for the first time showed improvements, but so did women who continued on flibanserin. Without a parallel placebo-extension arm, separating pharmacological effect from expectation effect is impossible.

Endpoint drift. The primary trial used co-primary endpoints: change from baseline in SSEs (measured by eDiary) and change in sexual desire score (also eDiary-derived). During the extension, these endpoints were tracked as secondary or exploratory measures. The regulatory weight of extension data is lower than the core trial, and the FDA's own advisory committee review treated extension findings as supportive, not confirmatory.

The HealthRX Durability Assessment Framework

To evaluate whether a chronic-use CNS drug shows true durability versus statistical artifact, we apply four criteria:

| Criterion | BEGONIA Extension Result | Interpretation | |---|---|---| | Maintained effect size in original responders | SSE gains held steady from week 24 to week 52 | Favorable, but no placebo comparator to confirm | | New responders emerging after week 24 | A small proportion of prior placebo patients showed benefit | Expected with any active treatment initiation | | Dropout rate during extension | ~18% discontinued, primarily for adverse events | Moderate attrition; suggests tolerability plateau | | Adverse event trajectory | Somnolence rates declined after week 8; no new safety signals | Consistent with CNS adaptation |

This framework highlights that BEGONIA extension data meets the "maintained effect" criterion only provisionally. The absence of a parallel placebo arm is the single largest limitation.

Results: What the Numbers Actually Show

Efficacy Over 52 Weeks

Among women who received flibanserin for the full 52 weeks (both blinded and open-label phases), the mean increase in SSEs was approximately 2.5 events per month from baseline. For context, baseline SSE counts averaged around 2.0 to 2.5 per month, so this represents roughly a doubling. That sounds impressive until you note that placebo patients in the core trial also improved by about 1.5 SSEs per month, and the drug-placebo difference was only 0.8.

Women who switched from placebo to flibanserin at week 24 showed improvements during the extension, but their trajectory closely mirrored the early gains seen in the original flibanserin arm during weeks 0 through 8. This pattern is consistent with a true pharmacological effect, but the magnitude remained modest.

Sexual Desire Scores

The Female Sexual Function Index (FSFI) desire domain showed statistically significant improvement in the core trial. During the extension, desire scores remained stable in continuing flibanserin users and improved in the crossover group. The clinical significance of a 0.3 to 0.5 point change on a 6-point desire subscale is debatable. The FDA's own medical review noted that effect sizes on desire endpoints were smaller than those on SSEs.

Distress Reduction

The Female Sexual Distress Scale-Revised (FSDS-R) Item 13, which measures distress specifically related to low desire, showed improvement during the core trial (approximately -0.7 points drug-placebo difference on a 5-point scale). Extension data suggested this improvement was maintained, though again without a concurrent placebo arm.

Safety Signals: What Emerged With Longer Exposure

Central Nervous System Effects

Somnolence was the most common adverse event in both the core trial and the extension, affecting roughly 11% of flibanserin-treated women. During the extension, somnolence rates decreased over time, consistent with tolerance development. Dizziness affected about 9% and followed a similar declining trajectory. Most CNS adverse events occurred during the first 8 weeks of treatment and were mild to moderate.

The Alcohol Interaction Story

The BEGONIA extension itself did not systematically evaluate alcohol interactions, but post-marketing data and a dedicated pharmacokinetic study revealed that co-administration of flibanserin with alcohol increased the risk of severe hypotension and syncope. This led to the FDA requiring a boxed warning on the Addyi label and a REMS (Risk Evaluation and Mitigation Strategy) program. The alcohol restriction, mandating abstinence during treatment, became one of the most cited reasons for low real-world uptake. It was not predictable from the BEGONIA trial data alone.

Discontinuation Patterns

About 18% of extension participants discontinued before week 52. Adverse events accounted for approximately half of discontinuations, with somnolence and dizziness as the leading causes. The remaining dropouts were attributed to perceived lack of efficacy, protocol non-compliance, or personal reasons. These rates are comparable to other chronic CNS medications but higher than rates seen with on-demand sexual health therapies.

Regression to the Mean: The Elephant in the Room

HSDD diagnosis requires self-reported low desire causing marked distress. Women who enroll in clinical trials do so at their symptomatic nadir, a point from which some degree of spontaneous improvement is expected regardless of treatment. The BEGONIA trial's placebo response rate was substantial: placebo-treated women gained roughly 1.5 SSEs per month, representing about 65% of the total improvement seen with flibanserin.

This large placebo response complicates interpretation of the extension data. When all participants receive active drug, any "maintained benefit" could partially reflect the natural trajectory of HSDD symptoms rather than a sustained pharmacological effect. The ISSWSH guidelines acknowledge that HSDD symptoms fluctuate over time and recommend reassessing treatment necessity after 8 weeks.

What Post-Approval Real-World Data Added

After the FDA approved flibanserin in August 2015, commercial uptake was far below projections. Prescriptions peaked at roughly 4,000 per month in early 2016 and declined thereafter. Several factors contributed:

The alcohol contraindication limited the eligible population
The REMS program required prescriber certification, adding friction
Insurance coverage was inconsistent, with many payers requiring prior authorization
Effect sizes in the trials, including BEGONIA, were modest enough that some clinicians questioned clinical meaningfulness

A post-hoc analysis pooling data from BEGONIA and sister trials (DAISY, VIOLET) attempted to identify responder subgroups. Women with lower baseline SSE counts and higher baseline distress showed larger treatment effects. This is consistent with floor effects in the SSE measure and suggests the drug may be most useful in women with the most severe presentations. Whether this represents a true pharmacological subgroup or a statistical artifact of regression to the mean remains unresolved.

Limitations the Authors Acknowledged (and Some They Did Not)

Acknowledged in the primary publication: The open-label extension lacked a placebo comparator. The enrichment design (requiring core trial completion) limits generalizability. SSE counts are subject to reporting bias.

Not emphasized: The eDiary compliance rate declined during the extension, potentially biasing results toward more engaged participants. The trial excluded women on hormonal contraceptives containing drospirenone (due to CYP3A4 interactions), limiting applicability to a common demographic. Women with comorbid depression on stable antidepressants were included, but the interaction between flibanserin's serotonergic mechanism and background SSRI/SNRI use was not adequately powered to evaluate.

The Bottom Line on Durability

The BEGONIA extension provides suggestive but not definitive evidence that flibanserin's modest benefits persist beyond 24 weeks. The effect appears to plateau early (by week 8) and remain stable through week 52 in women who tolerate the drug. CNS side effects attenuate with continued use. No serious safety signals emerged during the extension itself, though the alcohol interaction discovered later fundamentally changed the drug's risk-benefit profile.

For clinicians, the practical takeaway is straightforward: if a patient shows no meaningful improvement after 8 weeks on flibanserin, continuation is unlikely to produce delayed benefit. For women who do respond, the extension data supports ongoing treatment, with periodic reassessment of whether the benefit persists after drug holidays.

Frequently asked questions

›

References

Thorp J, Simon J, Dattani D, et al. Treatment of hypoactive sexual desire disorder in premenopausal women: efficacy of flibanserin in the BEGONIA trial. J Sex Med. 2012;9(2):535-542. PubMed
FDA. Addyi (flibanserin) prescribing information. 2015. FDA Label
Simon JA, Kingsberg SA, Shumel B, et al. Efficacy and safety of flibanserin in postmenopausal women with hypoactive sexual desire disorder: results of the SNOWDROP trial. Menopause. 2014;21(6):633-640. PubMed
Jaspers L, Feys F, Bramer WM, et al. Efficacy and safety of flibanserin for the treatment of hypoactive sexual desire disorder in women: a systematic review and meta-analysis. JAMA Intern Med. 2016;176(4):453-462. PubMed
Parish SJ, Goldstein AT, Goldstein SW, et al. Toward a more evidence-based nosology and nomenclature for female sexual dysfunctions: Part II. J Sex Med. 2016;13(12):1888-1906. PubMed
FDA. Medical review: NDA 022526. 2015. FDA Review