The research, conducted by a team at Anglia Ruskin University’s School of Medicine in Chelmsford, United Kingdom, pooled data from eight randomised controlled trials involving a total of 1,741 patients. Both palonosetron and ondansetron belong to a class of drugs known as 5-hydroxytryptamine type 3 receptor antagonists, or 5HT3RAs, which block serotonin receptors in the gut and brainstem that trigger the vomiting reflex. Ondansetron, a first-generation agent introduced in the early 1990s, transformed cancer care by dramatically reducing acute vomiting, while palonosetron, approved later, binds to the same receptor with far higher affinity and a much longer half-life of roughly 40 hours compared with ondansetron’s three to six hours.

The researchers set out to address a gap in the evidence base. Although earlier meta-analyses had suggested palonosetron was superior to first-generation 5HT3RAs as a class, direct head-to-head comparisons with ondansetron specifically remained limited, and prior analyses may not have captured the most recent trial data. To fill that void, the team systematically searched PubMed, the Cochrane Library and Scopus, identifying randomised controlled trials that compared the two drugs directly in patients receiving chemotherapy.

The primary outcome was complete response, defined as the absence of vomiting or retching with no use of rescue antiemetic medication. Secondary outcomes included complete control, which additionally requires no significant nausea, and the incidence of adverse events. The meta-analysis employed a statistically rigorous approach, using the Restricted Maximum Likelihood estimator combined with the Hartung-Knapp-Sidik-Jonkman correction, a method designed to produce more reliable confidence intervals in meta-analyses with a small number of trials, a situation where conventional methods can be overly optimistic.

The results were striking in their consistency. Palonosetron was associated with significantly higher complete response rates than ondansetron in all three clinically defined phases of CINV. During the acute phase, the first 24 hours after chemotherapy, the risk ratio was 1.10 with a 95 percent confidence interval of 1.04 to 1.15. In the delayed phase, spanning 24 to 120 hours after treatment, the advantage grew, with a risk ratio of 1.22 and a confidence interval of 1.13 to 1.32. Across the overall phase, the risk ratio reached 1.28 with a confidence interval of 1.20 to 1.36, meaning patients receiving palonosetron were substantially more likely to avoid emesis entirely throughout the full risk window.

The delayed and overall findings are particularly meaningful because delayed CINV has historically proven harder to control than the acute phase. Ondansetron’s short duration of action means its protective effect wanes within hours of administration, whereas palonosetron’s prolonged receptor binding is thought to sustain serotonin blockade through the delayed period. The pharmacology aligns neatly with the pooled clinical data: the largest effect sizes appeared precisely where the drug’s longer half-life would be expected to matter most.

However, the picture changes when the outcome shifts from complete response to complete control. For this stricter measure, which demands freedom from both emesis and significant nausea, the meta-analysis found no statistically significant differences between the two drugs in any phase. The risk ratios were 1.09 in the acute phase, 1.17 in the delayed phase and 1.08 overall, with confidence intervals in each case crossing the threshold of no effect. This divergence highlights a well-recognised challenge in antiemetic research: nausea, a subjective sensation, is considerably harder to suppress than vomiting, and no 5HT3 receptor antagonist has shown robust superiority against it. Nausea is now understood to involve multiple neurotransmitter pathways beyond serotonin, including neurokinin-1 and substance P signalling, which explains why combination regimens incorporating NK1 receptor antagonists and dexamethasone have become standard for highly emetogenic chemotherapy.

On safety, the analysis found no significant differences between the two agents in the most commonly reported adverse events. Headache showed a risk ratio of 0.88 with a confidence interval of 0.68 to 1.14, constipation a risk ratio of 1.47 with a confidence interval of 0.67 to 3.25, and dizziness a risk ratio of 0.71 with a confidence interval of 0.32 to 1.59. Yet the authors are careful to caution against overinterpreting these null findings. The available safety data were limited, and the absence of statistically significant differences should not, they emphasise, be read as proof that the two drugs are equally safe. Underpowered safety comparisons are a persistent weakness of meta-analyses built on trials designed primarily to measure efficacy.

Methodological transparency was a strength of the review. The team assessed risk of bias in the included trials using the Cochrane Risk of Bias 2.0 tool, the current standard for judging the internal validity of randomised studies, and evaluated the certainty of the evidence using the GRADE framework, which grades confidence in effect estimates from high down to very low based on study limitations, inconsistency, indirectness and imprecision. The review was prospectively registered under CRD420251108042, and the authors declared no funding and no competing interests, factors that reduce the risk of bias in the synthesis itself.

For clinicians and patients, the findings carry practical weight. Ondansetron remains one of the most prescribed antiemetics worldwide, largely because it is inexpensive and available in generic formulations, while palonosetron typically costs more. The evidence that palonosetron delivers significantly better protection against vomiting, particularly in the delayed phase, may prompt oncology teams to reconsider drug selection, especially for moderately emetogenic regimens where the choice between the two agents has often been driven by cost rather than comparative data. At the same time, the authors’ caution about safety data and the lack of a complete-control advantage suggest that palonosetron is not a wholesale replacement but rather a better-performing option within a broader antiemetic strategy. As CINV management continues to evolve toward guideline-driven, risk-adapted combinations, this analysis provides one of the clearest head-to-head syntheses to date of the two most familiar 5HT3 receptor antagonists, and a reminder that even decades-old drug classes still hold questions worth answering.

Subject of Research: Comparative efficacy and safety of palonosetron versus ondansetron for preventing chemotherapy-induced nausea and vomiting

Article Title: Efficacy and safety of palonosetron versus ondansetron for preventing chemotherapy-induced nausea and vomiting: a systematic review and meta-analysis of randomised controlled trials

Article References: Efficacy and safety of palonosetron versus ondansetron for preventing chemotherapy-induced nausea and vomiting: a systematic review and meta-analysis of randomised controlled trials. (n.d.). https://doi.org/10.1186/s40360-026-01206-z

Image Credits: AI Generated

DOI: 10.1186/s40360-026-01206-z

Keywords: palonosetron, ondansetron, chemotherapy-induced nausea and vomiting, 5HT3 receptor antagonists, meta-analysis, systematic review, antiemetics, complete response, drug safety, oncology, randomised controlled trials, pharmacology