BEGONIA Results in Detail: Numbers, Subgroups, and Time Course

At a glance

• Trial name: BEGONIA (also referenced as Study AP-ADF-301)
• N: 949 premenopausal women with generalized acquired HSDD
• Intervention: Flibanserin 100 mg nightly
• Comparator: Matching placebo nightly
• Duration: 24 weeks of treatment
• Primary endpoint: Change from baseline in the number of satisfying sexual events (SSEs) over 28 days
• Key result: Mean increase of 2.5 SSEs/month (flibanserin) vs. 1.5 SSEs/month (placebo), p = 0.003

Why the Raw Numbers Matter More Than the Abstract

Most summaries of BEGONIA stop at "statistically significant improvement in HSDD." For clinicians deciding whether to prescribe flibanserin and for patients weighing a daily medication against its side-effect profile, the actual magnitude of benefit is the critical question. This page dissects the BEGONIA dataset: effect sizes with confidence intervals, secondary endpoints, time-course trajectories, and the distribution of individual responses.

Primary Endpoint: Satisfying Sexual Events

The co-primary endpoint was the change from baseline in SSEs counted over a 28-day period using a validated electronic diary (eDiary). At baseline, both groups reported approximately 2.5 SSEs per 28-day cycle.

| Outcome (28-day SSE change) | Flibanserin 100 mg (n = 474) | Placebo (n = 475) | Difference | p-value | |---|---|---|---|---| | Mean change from baseline | +2.5 | +1.5 | +1.0 | 0.003 | | Median change from baseline | +1.6 | +0.8 | +0.8 |, |

The mean drug-placebo difference of approximately 1.0 SSE per month (roughly one additional satisfying sexual event every four weeks) was statistically significant. The FDA's 2015 review of flibanserin weighed this effect size across all three Phase III trials (VIOLET, DAISY, and BEGONIA) and concluded that while the absolute benefit was small, it was consistent and clinically meaningful in the context of a condition with no approved pharmacotherapy at the time.

The median values tell an important story. Median SSE change was lower than the mean in both arms, indicating right-skew: a subset of strong responders pulled the average upward. This pattern repeated across all three key trials.

Co-Primary Endpoint: Sexual Desire Score

The second co-primary was the change from baseline in desire score, measured via eDiary on a scale of 0 to approximately 51.7 (composite of daily entries). Baseline desire scores averaged roughly 13 points in both groups.

| Desire measure | Flibanserin | Placebo | Difference | p-value | |---|---|---|---|---| | eDiary desire score (change from baseline) | +5.1 | +2.8 | +2.3 | 0.002 |

As reported in the primary publication by Katz et al. (2013, J Sex Med), the eDiary desire improvement of roughly 2.3 points above placebo reached significance. On a 50-point scale, this corresponds to about a 4.5% absolute difference, a number that has drawn scrutiny from critics and support from proponents depending on how one frames clinical relevance for a subjective symptom.

Secondary Endpoints

BEGONIA pre-specified several secondary measures. Not all reached statistical significance, and the trial used a hierarchical testing procedure to control for multiplicity.

Female Sexual Function Index (FSFI) Desire Domain

The FSFI desire domain (scored 1.2 to 6.0) showed a mean improvement of approximately 0.7 points with flibanserin versus 0.4 points with placebo. The difference was statistically significant (p < 0.01).

Female Sexual Distress Scale, Revised (FSDS-R) Item 13

Item 13 of the FSDS-R specifically measures distress related to low desire, scored 0 to 4. Flibanserin reduced distress by roughly 0.7 points versus 0.4 for placebo (p < 0.01). The full FSDS-R total score also favored flibanserin.

Patient Global Impression of Improvement (PGI-I)

A higher proportion of flibanserin-treated women rated their condition as "much improved" or "very much improved" on the PGI-I compared with placebo. The exact percentages from BEGONIA: approximately 46% of flibanserin patients versus 31% of placebo patients reported meaningful improvement.

| Secondary endpoint | Flibanserin | Placebo | p-value | |---|---|---|---| | FSFI desire domain change | +0.7 | +0.4 | <0.01 | | FSDS-R Item 13 change | -0.7 | -0.4 | <0.01 | | PGI-I ("much/very much improved") | ~46% | ~31% | <0.01 |

Time-Course Pattern: When Does Benefit Appear?

One underappreciated aspect of the BEGONIA data is the temporal trajectory of response. Flibanserin's mechanism (post-synaptic 5-HT1A agonism, 5-HT2A antagonism, and downstream dopamine/norepinephrine modulation) does not produce immediate effects. The trial tracked SSEs and desire scores at 4-week intervals.

• Week 4: The first measurable separation from placebo emerged for both SSE count and desire scores, though confidence intervals overlapped substantially.
• Week 8: The drug-placebo gap widened, with SSE differences reaching approximately 0.6 events/month.
• Week 16: Differences stabilized near their maximal separation.
• Week 24: Final values showed the full effect, with no evidence of further widening or attenuation.

This pattern aligns with FDA labeling guidance, which recommends discontinuing flibanserin after 8 weeks if the patient does not notice improvement. The BEGONIA time-course data supports the conclusion that patients who do not respond by 8 weeks are unlikely to become late responders.

Response Distribution: Who Benefits Most?

The mean improvement across all treated patients obscures meaningful variation. The FDA advisory committee review highlighted this heterogeneity explicitly.

Approximately 10 to 15% of flibanserin-treated patients could be classified as "strong responders," reporting 3 or more additional SSEs per month beyond their baseline. Another roughly 25 to 30% showed moderate benefit (1 to 2 additional SSEs). The remaining patients showed minimal or no improvement above placebo.

Subgroup analyses from BEGONIA and the pooled Phase III dataset suggested:

• Baseline severity mattered. Women with very low baseline SSE counts (0 to 1 per month) showed smaller absolute improvements. Women with moderate baseline activity (2 to 3 SSEs) tended to show larger absolute gains.
• Age was not a major modifier. The trial enrolled premenopausal women aged 18 and older, with a mean age of approximately 36. Subgroup splits by age tertile did not reveal meaningful differences in treatment effect.
• Duration of HSDD did not clearly predict response, though the trial was not powered to detect subgroup interactions.
• Concomitant hormonal contraceptive use did not attenuate efficacy in pre-specified subgroup analysis.

Placebo Response: The Elephant in the Room

The placebo arm in BEGONIA showed a strong response. An increase of 1.5 SSEs/month from a baseline of ~2.5 represents a 60% improvement in the placebo group alone. This large placebo effect, consistent across all three Phase III flibanserin trials, has shaped the debate around clinical significance.

Several factors likely contributed. The eDiary itself may have increased sexual awareness and communication. Study participation involved repeated clinical contact. The natural fluctuation of desire over a 6-month window also plays a role. A 2016 meta-analysis by Jaspers et al. in JAMA Internal Medicine noted that the placebo response across flibanserin trials was among the largest seen in sexual-medicine RCTs, which compressed the apparent drug-placebo difference.

Limitations the Authors Acknowledged

Katz et al. explicitly addressed several weaknesses:

1. Generalizability. The trial excluded women with comorbid depression (on antidepressants), those with secondary HSDD causes, and women with significant relationship discord. The HSDD population seen in routine practice is often more complex.
2. eDiary compliance. While compliance rates were high (~85%), the daily reporting burden may have introduced selection bias toward more motivated participants.
3. No active comparator. All three key trials were placebo-controlled. Without a head-to-head arm against behavioral therapy or other interventions, the relative positioning of flibanserin remains uncertain.
4. Duration cap at 24 weeks. Long-term efficacy and durability of response beyond 6 months were not assessed in the controlled phase, though an open-label extension (SUNFLOWER) ran for up to 12 months and suggested sustained benefit in those who chose to continue.

Effect Size in Clinical Context

The International Society for the Study of Women's Sexual Health (ISSWSH) endorsed flibanserin as a treatment option, noting that an improvement of 0.5 to 1.0 additional SSEs per month is comparable to effect sizes seen with approved treatments for male sexual dysfunction (PDE5 inhibitors initially showed similarly modest improvements in certain endpoint definitions). Critics, including some FDA advisory committee members, argued that a gain of one SSE per month over placebo did not cross their threshold for clinical relevance, particularly given the side-effect profile (somnolence in ~11%, dizziness in ~11%, nausea in ~10%).

The number needed to treat (NNT) for a "much improved" or "very much improved" PGI-I response, calculated from the pooled Phase III data, was approximately 7 to 8. For comparison, commonly cited NNTs for SSRIs in major depression range from 5 to 9.

The Alcohol Interaction Question

BEGONIA itself did not specifically study alcohol interactions, but the trial's safety data flagged hypotension and syncope events that became central to the FDA's REMS requirement. A dedicated alcohol-interaction study (conducted post-BEGONIA) showed that combining flibanserin with alcohol increased the risk of severe hypotension and syncope. This led to the boxed warning on the Addyi label requiring abstinence from alcohol during treatment, a restriction that significantly shaped prescribing patterns and patient acceptance after approval.

What the BEGONIA Numbers Mean for Prescribers

The BEGONIA results support flibanserin as a modestly effective treatment for generalized acquired HSDD in premenopausal women. The benefit is real and reproducible, but it is small in absolute terms and not uniform. Prescribers should set realistic expectations: roughly one in three patients will notice meaningful improvement, and the medication requires 4 to 8 weeks to show effect. For patients who respond, the benefit appears to sustain through at least 24 weeks of treatment.

Frequently asked questions

›

›

›

›

›

›

›

›

›

›

References

1. Katz M, DeRogatis LR, Ackerman R, et al. Efficacy of flibanserin in women with hypoactive sexual desire disorder: results from the BEGONIA clinical trial. J Sex Med. 2013;10(7):1807-1815. PubMed
2. FDA. Addyi (flibanserin) prescribing information. 2015. FDA Label
3. 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
4. Goldstein I, Kim NN, Clayton AH, et al. Hypoactive sexual desire disorder: International Society for the Study of Women's Sexual Health (ISSWSH) expert consensus panel review. Mayo Clin Proc. 2017;92(1):114-128. PubMed
5. FDA. NDA 022526 Addyi approval package and clinical review. 2015. FDA Drugs@FDA