← Back to Home

Endometrial Cancer DUO-E Update: Longer Survival With Dual-Drug Maintenance, but Statistical Threshold Not Yet Met

An 11.5-month difference in median overall survival offers a treatment signal for patients with advanced or recurrent disease; however, this interim analysis is not yet sufficient to confirm a survival benefit, and the higher rates of adverse events must also be weighed.

By SURL BioNews

For patients with advanced or recurrent endometrial cancer, whether treatment can extend life beyond delaying disease progression is the longer-term test. An update from the phase 3 DUO-E trial presented at the 2026 European Society for Medical Oncology (ESMO) Congress showed median overall survival of 51.1 months among patients who received durvalumab with initial chemotherapy, followed by maintenance treatment with durvalumab plus olaparib, compared with 39.6 months in the chemotherapy control group. However, this 11.5-month difference remains a signal awaiting confirmation in the final analysis.

The key is the trial’s prespecified statistical threshold. According to reports from OncLive and OncoDaily, the hazard ratio for death with the dual-drug strategy versus the control group was 0.77, with a 95% confidence interval of 0.60 to 0.98; the nominal P value was 0.037, which did not meet the requirement of less than 0.0165 for this analysis. Interim analyses require a stricter threshold to control the chance of false conclusions arising from repeated testing, so a P value below the commonly used 0.05 alone does not establish that survival has been extended.

DUO-E enrolled 718 patients, randomly assigned to three groups: one received carboplatin and paclitaxel chemotherapy followed by placebo; another received the immunotherapy durvalumab with chemotherapy, followed by continued durvalumab monotherapy as maintenance; and the third received the same initial treatment, followed by maintenance with durvalumab and the PARP inhibitor olaparib. Median overall survival in the durvalumab monotherapy maintenance group was 40.7 months, and its comparison with the control group likewise did not meet the prespecified significance threshold. These results also cannot be used to directly conclude that dual-drug maintenance is superior to durvalumab monotherapy maintenance.

Tumor DNA mismatch repair status offers another layer of interpretation. In the mismatch repair-proficient (pMMR) population, median overall survival with the dual-drug strategy and in the control group was 45.8 and 37.8 months, respectively, but the 95% confidence interval for the hazard ratio crossed 1, leaving uncertainty about the benefit. In the mismatch repair-deficient (dMMR) population, median overall survival had not yet been reached in either durvalumab-containing group, compared with 50.2 months in the control group. These exploratory subgroup results can guide further research but are not sufficient on their own to establish survival conclusions for each population.

The treatment burden also increased with the combination. OncLive reported that grade 3 or higher adverse events occurred in 69.7% of patients in the dual-drug maintenance strategy group, compared with 58.3% in the durvalumab group and 57.2% in the control group; the proportions who discontinued any trial treatment because of adverse events were 29.0%, 23.4%, and 19.1%, respectively. These figures cover the entire course of trial treatment and cannot all be attributed to the maintenance phase, but they show a cost that cannot be overlooked when assessing potential benefits.

### Background

DUO-E previously met both primary endpoints, demonstrating that the two durvalumab-containing treatment strategies prolonged progression-free survival compared with the chemotherapy control—that is, the time during which patients remained alive without disease progression. This update asks a further question: can longer disease control translate into improved overall survival? These measure different outcomes, and success with the former does not guarantee success with the latter.

According to OncoDaily, at the February 2, 2026 data cutoff, overall survival data maturity was approximately 52%, meaning that approximately half of the participants had died; median follow-up among those without an observed death event was approximately 52 months. The trial will continue follow-up and formally retest overall survival in the final analysis. The current results provide another indication of potential long-term benefit with the dual-drug strategy while bringing the remaining questions into sharper focus: whether this difference can withstand the final statistical test, and whether the associated treatment burden is worthwhile.

References

  1. UT MD Anderson Cancer Center
  2. OncLive
  3. OncoDaily