Regulatory ImpactRegulatory Impact
Regulatory InsightsRegulatory Intelligence

FDA-Approved Products for Nausea and Vomiting of Pregnancy: Efficacy, Safety, and Regulatory Precedent

Regulatory ImpactSeptember 4, 202610 min read
FDANausea and Vomiting of PregnancyPregnancy505(b)(2)DoxylaminePyridoxineClinical EfficacyDrug Safety

Executive Summary

The FDA-approved branded landscape for nausea and vomiting of pregnancy is centered on Diclegis and Bonjesta, two oral formulations containing the same active ingredients, doxylamine succinate and pyridoxine hydrochloride. Diclegis is a 10 mg/10 mg delayed-release formulation, while Bonjesta is a 20 mg/20 mg extended-release formulation. Generic versions of both dosage forms have subsequently received FDA approval.

The efficacy foundation is more nuanced than the presence of two branded approvals might suggest. Diclegis was supported by a modern randomized placebo-controlled study in which both treatment and placebo produced substantial symptom improvement, leaving a comparatively small placebo-adjusted difference on the primary symptom scale. FDA's statistical review explicitly characterized the effect as small and noted that sensitivity analyses were not uniformly supportive.

Historical evidence strengthened the efficacy case. An earlier factorial trial evaluated doxylamine, pyridoxine, dicyclomine, their combinations, and placebo. The results strongly supported the contribution of doxylamine, while evidence for the incremental contribution of pyridoxine to the combination was more limited.

Bonjesta did not generate a second independent efficacy estimate. Its NDA contained no new efficacy data. FDA instead accepted a pharmacokinetic bridge to Diclegis after reviewing single-dose and multiple-dose bioequivalence evidence. That bridge itself generated substantive review questions, including an initial failure to establish bioequivalence under one Day 1 comparison and concern that differences in pharmacokinetic profiles could potentially alter sedation or clinical response.

Safety is dominated by the pharmacology of doxylamine, particularly somnolence and anticholinergic effects. The pregnancy-specific evidence is notable because epidemiologic studies summarized in FDA labeling did not identify an increased congenital-malformation risk with first-trimester doxylamine-pyridoxine exposure. Neither labeled indication should be interpreted as a specific approval for hyperemesis gravidarum.

The Approved NVP Product Landscape

FDA approval materials substantiate two branded New Drug Applications (NDAs) specifically indicated for treatment of nausea and vomiting of pregnancy in women who do not respond to conservative management. Diclegis, NDA 021876, was approved April 8, 2013. Bonjesta, NDA 209661, was approved November 7, 2016. Both applications were submitted under section 505(b)(2) of the Federal Food, Drug, and Cosmetic Act. 1234

ProductApplicationApproval dateFormulationStrengthFDA-approved NVP indication
DiclegisNDA 021876April 8, 2013Delayed-release tablet10 mg doxylamine succinate / 10 mg pyridoxine hydrochlorideTreatment of nausea and vomiting of pregnancy in women who do not respond to conservative management
BonjestaNDA 209661November 7, 2016Extended-release tablet20 mg doxylamine succinate / 20 mg pyridoxine hydrochlorideTreatment of nausea and vomiting of pregnancy in women who do not respond to conservative management

FDA also identifies first generic approvals corresponding to each formulation. Abbreviated New Drug Application (ANDA) 205811 for the 10 mg/10 mg delayed-release formulation was approved August 19, 2016, and ANDA 212472 for the 20 mg/20 mg extended-release formulation was approved March 1, 2022. These FDA first-generic records establish generic entry for both formulations but do not constitute a complete current inventory of every subsequently approved ANDA. 56

Reference formulationFirst generic ANDA identified by FDAApplicantApproval date
Diclegis 10 mg/10 mg delayed-releaseANDA 205811Actavis Laboratories FL, Inc.August 19, 2016
Bonjesta 20 mg/20 mg extended-releaseANDA 212472Actavis Laboratories FL, Inc.March 1, 2022

Diclegis: A Statistically Significant but Modest Placebo-Adjusted Effect

The publicly confirmed Diclegis development and approval pathway can also be explored in Regulatory Designer, which maps disclosed clinical and regulatory milestones across the program.

The pivotal modern efficacy study, DIC-301, was a double-blind, randomized, multicenter, placebo-controlled study in women with nausea and vomiting of pregnancy (NVP). The primary endpoint was change from baseline to Day 15 in the Pregnancy-Unique Quantification of Emesis and Nausea (PUQE; a very fitting acronym) score, which combines duration of nausea, frequency of vomiting, and frequency of retching into a score ranging from 3 for no symptoms to 15 for the most severe symptoms. 27

In the intent-to-treat efficacy population using last observation carried forward, the Diclegis group included 131 women and the placebo group included 125. Baseline mean PUQE scores were 9.0 and 8.8, respectively. By Day 15, the mean change from baseline was approximately -4.7 with Diclegis and -3.9 with placebo. The placebo-adjusted treatment difference was -0.73 points, with a 95% confidence interval of approximately -1.24 to -0.22 and p=0.006. 28

DIC-301 efficacy measureDiclegisPlaceboTreatment differencep-value
Mean PUQE change from baseline-4.67-3.94-0.730.006
Change in hours of nausea-2.35-2.13-0.210.125
Change in number of vomiting episodes-0.95-0.72-0.230.008
Change in number of retching episodes-1.37-1.10-0.280.003

The magnitude of the placebo response is important for interpretation. Both groups improved substantially during the short trial, and the incremental benefit attributable to active treatment was less than one PUQE point. FDA's statistical reviewers described the treatment difference as small and stated that the clinical significance of that effect was a clinical judgment rather than a statistical conclusion. 8

The component-level findings were also uneven. Compared with placebo, Diclegis produced statistically significant improvements in vomiting frequency and retching frequency, but the difference in hours of nausea was not statistically significant. FDA's statistical review also noted that among the other secondary endpoints, only the global assessment of well-being showed a significant treatment effect. 28

Sensitivity analyses provided another qualification. FDA reported that analyses using alternative populations produced inconsistent efficacy results. A small treatment difference remained in the per-protocol population, but not in subjects who completed the study or in the completer population excluding major protocol violators. The primary intent-to-treat analysis remained the basis for the positive efficacy conclusion. 8

Diclegis Improved PUQE Scores, but the Incremental Effect Was Small

The Historical Factorial Trial Clarifies Which Components Drove Efficacy

The Diclegis clinical review also revisited an older randomized, double-blind, multicenter, placebo-controlled factorial study of Bendectin and its components. The study included 2,308 women and compared eight groups containing different combinations of doxylamine, pyridoxine, dicyclomine, or placebo. FDA's review states that 1,599 participants with nausea and/or vomiting took medication on each of six successive study days and provided the required diary cards. 2

For the doxylamine-pyridoxine group, the physician evaluation classified overall medication effectiveness as moderate or excellent in 78% of participants versus 57% with placebo. Physician-assessed nausea improvement was 75% versus 52%, with p<0.01. The corresponding vomiting comparison was 73% versus 66% and did not reach statistical significance at p=0.17. The historical analysis used one-sided comparisons of each active treatment with placebo. 2

Patient diaries showed a 64% reduction in nausea from pretreatment with doxylamine-pyridoxine compared with 31% with placebo, with p<0.01. Among participants evaluated for vomiting, 48% in the doxylamine-pyridoxine group reported no vomiting on at least five treatment days compared with 28% on placebo, also with p<0.01. 2

Historical doxylamine-pyridoxine comparisonDoxylamine + pyridoxinePlaceboReported p-value
Physician: medication rated moderate or excellent78%57%<0.01
Physician: nausea improved75%52%<0.01
Physician: vomiting improved73%66%0.17
Patient diary: reduction in nausea64%31%<0.01
Patient diary: no vomiting on at least 5 treatment days48%28%<0.01

The factorial design provides an unusually informative view of component contribution. FDA's historical reviewer concluded that doxylamine alone and all combinations containing doxylamine were consistently superior to placebo for control of nausea. The incremental evidence that pyridoxine improved efficacy beyond doxylamine alone was substantially weaker, with the review describing that comparison as marginal and reporting p-values of 0.12 and 0.26 in patient and physician records, respectively. Factorial comparisons nevertheless provided evidence supporting a pyridoxine contribution to nausea control. 2

Bonjesta Did Not Generate an Independent Efficacy Estimate

The central comparative point for Bonjesta is that NDA 209661 did not contain a new efficacy trial. FDA's clinical review repeatedly states that the application contained no efficacy data and that efficacy was not an objective of the two bioequivalence trials or the fed-versus-fasted bioavailability trial. Efficacy was instead considered through the pharmacokinetic bridge to Diclegis. 49

The bridge addressed both the starting dose and the maximum daily regimen. Trial 150336 compared one 20 mg/20 mg extended-release tablet with two 10 mg/10 mg delayed-release tablets after a single dose and established bioequivalence based on doxylamine and baseline-corrected pyridoxal 5'-phosphate. Trial 150033 supported bioequivalence at steady state on Day 11 for the higher daily regimen. FDA's clinical reviewers concluded that the two trials together adequately bridged Bonjesta's safety and efficacy to Diclegis. 410

Evidence questionDiclegisBonjesta
New placebo-controlled efficacy trial in NDAYes, DIC-301No
Direct placebo-adjusted efficacy estimate for marketed formulationYesNo independent estimate
Primary regulatory efficacy strategyClinical efficacy plus historical reliancePharmacokinetic bridge to Diclegis
Key bridge evidenceHistorical Bendectin findings and new clinical studySingle-dose and steady-state bioequivalence studies

This distinction matters when interpreting the apparent existence of two approved products. There is no FDA-reviewed head-to-head clinical trial demonstrating that the 20 mg/20 mg extended-release formulation provides greater symptom control than the 10 mg/10 mg delayed-release formulation. The efficacy section of the Bonjesta label instead describes the placebo-controlled study performed with the 10 mg/10 mg formulation and explicitly states that efficacy and safety trials were not conducted with Bonjesta itself. 104

The Bonjesta Bridge Was a Substantive Review Issue

The eventual bioequivalence conclusion obscures an important part of the review history. At filing, FDA's Office of Clinical Pharmacology determined that the submitted Day 1 data from Trial 150033 did not establish single-dose bioequivalence between the proposed 20 mg/20 mg extended-release formulation and Diclegis. FDA allowed the application to be filed but identified the failed Day 1 comparison as a review issue. 4

FDA also questioned the clinical relevance of differences in the pharmacokinetic profiles. The filing communication noted that the time to maximum doxylamine concentration for Diclegis occurred around early morning while the proposed formulation's peak occurred around noon in the relevant comparison. FDA identified the possibility of a different clinical response and more sedation as issues requiring evaluation, rather than treating total exposure alone as automatically sufficient. 4

Trial 150033 ultimately established bioequivalence at Day 11 for doxylamine and baseline-corrected pyridoxal 5'-phosphate. A separate single-dose study, Trial 150336, compared the same nominal dose, one 20 mg/20 mg tablet against two 10 mg/10 mg tablets, and met the prespecified bioequivalence criteria for doxylamine and baseline-corrected pyridoxal 5'-phosphate. FDA concluded that the totality of the single-dose and multiple-dose evidence established the required bridge. 4

The sequence illustrates a practical 505(b)(2) principle. Formulation bridging can substitute for a new efficacy trial when exposure relationships are adequately established, but a formulation that changes release kinetics or dosing architecture can create clinically relevant questions even when the active ingredients and total daily dose are unchanged. In this application, FDA had stated before submission that failure to establish an adequate pharmacokinetic bridge could necessitate clinical efficacy and safety trials with the new formulation. 4

Safety: Somnolence Is the Dominant Labeled Concern

In DIC-301, 74 of 133 Diclegis-treated women, or 55.6%, experienced at least one treatment-emergent adverse event, compared with 66 of 128 placebo-treated women, also reported as 55.6%. FDA's medical reviewer concluded that the overall number of subjects experiencing adverse events was not significantly different between treatment groups and that the study did not demonstrate a new safety or tolerability concern at doses ranging from two to four tablets daily. 2

Somnolence is nevertheless the adverse reaction most clearly associated with the product. Current Diclegis labeling reports somnolence in 14.3% of treated women versus 11.7% of placebo-treated women in the 15-day trial and identifies somnolence as the only adverse reaction occurring in at least 5% of treated participants at a rate exceeding placebo. 7

The warning has functional consequences. Diclegis labeling instructs women to avoid activities requiring complete mental alertness, including driving or operating heavy machinery, until cleared by a healthcare provider. Concurrent use with alcohol or other central nervous system (CNS) depressants is not recommended because severe drowsiness may lead to falls or accidents. 7

The doxylamine component also carries anticholinergic considerations. The Diclegis label advises caution in women with increased intraocular pressure, narrow-angle glaucoma, stenosing peptic ulcer, pyloroduodenal obstruction, or urinary bladder-neck obstruction, and use with monoamine oxidase inhibitors is contraindicated because these agents intensify and prolong antihistamine-related central nervous system and anticholinergic effects. 7

The current Diclegis label also describes reported false-positive urine immunoassay screens for methadone, opiates, and phencyclidine associated with doxylamine-pyridoxine use and recommends confirmatory testing when such a screening result occurs. 7

Bonjesta Safety Relied on a Small Pharmacology Program and the Existing Diclegis Experience

Bonjesta's safety database was fundamentally different from the Diclegis efficacy trial because the three supporting studies were clinical pharmacology studies in healthy non-pregnant women. FDA's Division Director review reports 115 randomized women and 103 completers across the three studies. There were no deaths or serious adverse events, and discontinuations attributed to adverse events were few. 94

FDA reported that no new safety signals were identified when the comparative bioavailability studies were reviewed separately or collectively and that the observed adverse-event profiles were consistent with the established doxylamine-pyridoxine safety profile. The review nevertheless cautioned that comparisons of somnolence-related events were constrained by small sample sizes, low event counts, short exposure durations, and differences in comparator exposure. 9

The final Bonjesta labeling therefore carries the same central clinical concern: somnolence caused by the anticholinergic properties of doxylamine. Women are instructed to avoid driving or other activities requiring full mental alertness until cleared, and use with alcohol or other CNS depressants is not recommended because of the risk of severe drowsiness, falls, or accidents. 10

The review history is particularly relevant because FDA explicitly considered whether the altered pharmacokinetic profile of the extended-release formulation could shift sedation into a different part of the day. That issue was resolved sufficiently for approval, and FDA ultimately identified no new safety signal, but the concern illustrates why bioequivalence assessments for modified-release products may involve more than a simple comparison of total exposure. 49

Pregnancy Safety Evidence Is Unusually Central to the Benefit-Risk Assessment

Diclegis labeling states that no increased risk of congenital malformations has been reported in epidemiologic studies of pregnant women exposed to doxylamine succinate and pyridoxine hydrochloride. The label summarizes a meta-analysis of 16 cohort studies and 11 case-control studies published between 1963 and 1991 that found no increased risk of malformations after first-trimester exposure to the combination, with or without dicyclomine. 7

The label also describes a second meta-analysis incorporating 12 cohort studies and five case-control studies published between 1963 and 1985. That analysis found no statistically significant relationship between fetal abnormalities and first-trimester exposure to doxylamine-pyridoxine, with or without dicyclomine. These observational findings support the pregnancy risk summary in labeling, although they are not randomized evidence of fetal safety. 7

Both approved products are specifically intended for use during pregnancy, but both labels limit the scope of the evidence with respect to more severe disease. Diclegis has not been studied in women with hyperemesis gravidarum, and Bonjesta labeling likewise states that its safety and effectiveness are not established for that condition. The FDA-approved NVP indication should therefore not be interpreted as a specific hyperemesis gravidarum approval. 710

Regulatory Implications

The efficacy precedent is narrower than a simple count of approvals suggests. The modern direct placebo-controlled evidence comes from the Diclegis program, where FDA accepted a statistically significant primary result despite a placebo-adjusted PUQE difference of less than one point and sensitivity analyses that were not uniformly supportive. Historical controlled evidence supplied additional context supporting the combination and the contribution of its components. 28

Bonjesta demonstrates a different development pathway. FDA had already accepted the efficacy of the 10 mg/10 mg formulation, so the later program could focus on establishing that the new release characteristics and dosing regimen produced exposures sufficient to bridge to those findings. The absence of new Bonjesta efficacy data is therefore not a gap within its approved 505(b)(2) strategy, but it does mean that the two branded products should not be interpreted as two independent confirmations of clinical effect. 49

The review also shows that modified-release bridging can become a clinically substantive issue. FDA scrutinized both total exposure and the shape and timing of pharmacokinetic profiles because differences could theoretically alter symptom coverage or sedation. An additional single-dose bioequivalence study ultimately helped resolve the initial Day 1 concern. 4

For a new NVP development program, these approvals provide precedent for the indication, the PUQE endpoint, a doxylamine-pyridoxine benefit-risk framework, and a formulation-bridging strategy. They do not establish that the same evidence package would be sufficient for a novel active ingredient, a materially different mechanism, a different disease-severity population, or a specific claim in hyperemesis gravidarum. 284710

Conclusion

The FDA-approved NVP landscape is concentrated around one established pharmacologic combination rather than multiple independent therapeutic mechanisms. Diclegis supplies the principal modern placebo-controlled efficacy evidence, supported by a much larger historical factorial trial. Bonjesta extends the same evidence base through a different formulation and dosing architecture rather than through a second efficacy trial.

The details of the reviews are important. Diclegis produced statistically significant improvement, but the incremental effect over placebo was modest and some sensitivity analyses were inconsistent. Historical data strongly supported the contribution of doxylamine while providing a more qualified picture of the incremental contribution of pyridoxine. Bonjesta then demonstrated how an established efficacy finding can be leveraged through 505(b)(2), provided that the pharmacokinetic bridge is sufficiently rigorous.

Safety considerations are similarly coherent across the two products. Somnolence and anticholinergic effects are central to the doxylamine component, while the pregnancy-specific epidemiologic record provides substantial historical reassurance regarding congenital malformations. The remaining regulatory boundary is equally important: these NVP approvals do not themselves establish an FDA-approved treatment claim for hyperemesis gravidarum.