DUO-E Post Hoc Analysis Suggests Potential for Biomarker-Optimized pMMR Endometrial Cancer Management

Shannon Westin, MD, MPH, FACOG

Maintenance olaparib (Lynparza) added to durvalumab (Imfinzi) and chemotherapy improved progression-free survival (PFS) across biomarker-defined subgroups, regardless of PD-L1 expression, POLE or TP53 mutation status, homologous recombination repair (HRR)or BRCA mutation status, histologic subtype, and baseline circulating tumor DNA (ctDNA) detection, according to Shannon Westin, MD, MPH, FACOG.

She added that these findings reinforce the potential for a biomarker-driven treatment approach to optimize outcomes in mismatch repair–proficient (pMMR) recurrent endometrial cancer.

Findings from a post hoc exploratory subgroup analysis of relevant clinical markers among patients in the phase 3 DUO-E trial (NCT04269200) were presented at the 2025 SGO Annual Meeting on Women’s Cancer. In the pMMR subpopulation (n = 575), 84% of patients were positive for at least 1 biomarker. The most common alterations included PD-L1 positivity in 67% of patients and TP53 mutations in 59% of patients, with positivity concordance between the 2 markers observed in 44% of cases. Additionally, HRR mutations were identified in 21% of patients, and BRCA mutations were seen in 8% of patients.

In biomarker-positive subgroups, the addition of durvalumab to chemotherapy demonstrated a consistent PFS benefit compared with chemotherapy alone. Furthermore, the addition of olaparib to durvalumab and chemotherapy showed a potential additive PFS benefit, with these effects observed across molecular subtypes.

“The key takeaway is that, at this point, no single population can be isolated as definitively more likely to benefit from olaparib. The benefit appears broadly applicable across subgroups,” Westin stated in an interview with OncLive®. “To be clear, there may be a population that derives more benefit than others, but whether that can be determined from this study or will require further research remains to be seen.”

In addition to expanding on updated efficacy and biomarker data from DUO-E during the interview, Westin also shared the rationale for evaluating olaparib maintenance in the pMMR subgroup and highlighted the need for more homogeneous, biomarker-driven trials to optimize patient selection for PARP and checkpoint inhibitor combinations in endometrial cancer.

Westin is the director of Early Drug Development and the Phase I Trials Department and a professor in the Department of Gynecologic Oncology and Reproductive Medicine in the Division of Surgery at The University of Texas MD Anderson Cancer Center in Houston.

OncLive: What data have been previously reported from the DUO-E study?

Westin: The DUO-E trial was designed to answer 2 key questions: whether there is a role for the addition of checkpoint inhibition to chemotherapy in advanced and recurrent endometrial cancer, and whether there is additional benefit from adding a PARP inhibitor—olaparib—to the combination of the checkpoint inhibitor durvalumab and chemotherapy. The study enrolled an all-comers population, with subgroup analyses based on mismatch repair–deficiency [dMMR], PD-L1 positivity, and other molecular features.

Why was it important to further investigate outcomes with the addition of olaparib maintenance to this regimen in the pMMR subgroup?

What we found previously is that the addition of durvalumab to chemotherapy had a clear benefit across both [pMMR and dMMR] groups, but particularly in the dMMR group. This pattern has been seen across multiple studies involving checkpoint inhibition. Adding a checkpoint inhibitor to chemotherapy benefits the biomarker-positive group—namely, the dMMR group.

There remains an unmet need and opportunity in the pMMR group. In this group, checkpoint inhibition provides some improvement in PFS, but not to the extent observed in the dMMR population. What we observed was an additive benefit with the inclusion of the PARP inhibitor olaparib in the pMMR group. It was appropriate to evaluate whether there is a specific biomarker-defined or histologic subgroup within this population that derives the most benefit from the addition of the PARP inhibitor.

At this point, the treatment regimen includes 4 drugs—chemotherapy with durvalumab, followed by maintenance with durvalumab and olaparib. This raises concerns about additional, although manageable, toxicity. Therefore, it is important to determine which subgroup experiences the greatest benefit from this intensified approach.

What variables and patient populations were analyzed in the post hoc analysis?

This analysis included the entire group of patients [with pMMR disease], which comprised [575] patients. Of those, 486 had available biomarker data.

The biomarkers evaluated included histology, which has long been an important factor, [and the study] specifically evaluated serous type, endometrioid type, carcinosarcoma, etc. BRCA and HRR mutations were also analyzed, given their relevance to predicting benefit from PARP inhibition. PD-L1 positivity was examined as well, with data from this analysis previously presented and published in the Journal of Clinical Oncology. Additionally, p53 mutations and POLE mutations were evaluated. In the International Society of Gynecologic Oncology 2024 Annual Global Meeting presentation, those were the primary biomarker categories assessed.

In the most recent presentation [at the 2025 SGO Annual Meeting], ctDNA was added as an exploratory variable to assess its potential association with clinical benefit. For this analysis, baseline ctDNA levels were assessed at the time of enrollment. Although [data from] subsequent time points—after 3, 7, and 9 cycles—are available, [they] have not yet been presented.

What did the pMMR biomarker subgroup analysis reveal about the prevalence of these mutations and their association with clinical benefit?

We presented [data from] the group of 486 patients who had at least 1 of [the aforementioned] biomarkers. We found that 84% [of patients enrolled in DUO-E] had positivity for at least 1 of the biomarkers mentioned. Some patients were positive for multiple biomarkers. The most common aberrant biomarkers observed were PD-L1 positivity and p53 mutations, present in 67% and 59% of patients, respectively. There was concordance between PD-L1 positivity and p53 mutations in 44% of patients.

We then looked at smaller subsets.HRR gene mutations were seen in 21% of patients, and 8% of [patients had] BRCA mutations. Another important histologic feature evaluated was serous-type tumors. Historically, we relied heavily on pathology-based discrimination in endometrial cancer. In this cohort, only 27% of patients had serous tumors. Notably, p53 mutations are often associated with serous histology. However, this is not always the case—multiple other histologies can also harbor p53 mutations. This reinforces the need to assess molecular aberrations in addition to histology. [Although] pathologic assessments remain important, molecular profiling provides critical additional information.

When evaluating the population with at least 1 of the aforementioned molecular aberrations, there was a consistent benefit observed with the addition of durvalumab to chemotherapy. There also appeared to be an additive benefit with the addition of olaparib, and this trend was seen across all molecular subtypes.

Finally, regarding baseline ctDNA, approximately half of the patients with biomarker data had successful ctDNA testing. Among those, depending on the molecular or histologic subgroup, between 70% and [100%] had detectable ctDNA at baseline. This is an important finding.

This study represents one of the first large, randomized, prospective efforts to evaluate ctDNA in advanced and recurrent endometrial cancer. Prior to this, we had limited understanding of how commonly ctDNA is shed in this setting. The finding of high ctDNA positivity suggests that ctDNA could be a useful tool. In the ctDNA-positive population, there was a favorable hazard ratio [HR] for the addition of olaparib. However, it is still too early to [definitively identify this group] as the one that should receive olaparib.

What should be known about the regimen’s safety profile in this study?

The safety [profile] has remained consistent. When you add another drug, you see additional adverse effects [AEs]. The comforting aspect of the addition of the PARP inhibitor is that the additional grade 3 AEs we saw were as expected. We’ve been giving PARP inhibitors in ovarian cancer, breast cancer, and other cancer types, and we know that a PARP inhibitor is going to increase [the incidence of AEs like] anemia and bone marrow toxicity. Generally, those effects can be mitigated and managed. In the long term, we do not see a significantly higher discontinuation rate with the addition of the PARP inhibitor.

Another important point is that, as we continue to gather long-term data, we have not observed the kind of serious AEs that can potentially occur with prolonged therapy use, such as acute myeloid leukemia or myelodysplastic syndrome. We have seen zero of those cases so far. [We hope] this continues for our patients. Overall, to date, we have not seen any of the serious long-term AEs [with this regimen] that we might worry about.

What are the current limitations in biomarker stratification for patients receiving PARP inhibitor and checkpoint inhibitor combinations in endometrial cancer, and how might future trials address the heterogeneity observed across studies such as the phase 3 RUBY (NCT03981796) and DUO-E trials?

PARP inhibition in combination with checkpoint inhibition [is approved for patients with endometrial cancer] across Europe, Japan, many Asian countries, and Brazil. This combination of durvalumab and olaparib is available for patients with pMMR disease. Arguably, it is one of the most effective combinations for that population.

It’s probably not for all patients, but right now, based on the data we have and the approvals that are potentially available, it can be used for all patients. The next steps [for investigation] involve conducting trials in a more homogenous population, where we can interpret the data more clearly.

Even the designation of p53 mutation varied significantly across the RUBY study, DUO-E, and other studies that evaluated PARP inhibition in endometrial cancer. We can’t make a direct comparison [across these trials]. We can’t definitively say that p53-mutant tumors require PARP inhibition.

There was a lower HR [for PFS] in [the p53-mutant] population [vs the p53-mutant population], suggesting benefit, but the HR in the p53 wild-type population was 0.69—which historically would be considered meaningful. We used to be satisfied with a HR like that. Now, expectations are higher. We aim for HRs of 0.3 or 0.2, and I’m aligned with that goal. However, for now, there is not a clearly superior biomarker-defined population. It is our responsibility—as academics and in collaboration with industry partners—to design the right trials to answer these remaining questions.

What is your main message for colleagues regarding this research?

The bottom line is that we’ve evaluated some of the most important known biomarkers in advanced and recurrent endometrial cancer. To date, there does not appear to be a clear predictive marker associated with the additional benefit of PARP inhibition. Currently, we do not have an approval for the use of olaparib plus durvalumab in the United States [US]. Our European colleagues and colleagues across the world are potentially able to use this [combination]; it is to be determined what the regulatory situation will be in the US. However, consideration could be made for the use of this [regimen] in the population with pMMR disease once we get access to this regimen.

Reference

Moore K, Westin SN, Chon HS, et al. Durvalumab plus carboplatin/paclitaxel followed by durvalumab with/without olaparib in endometrial cancer: biomarkers, histological heterogeneity, baseline circulating tumor DNA, and efficacy in the DUO-E mismatch repair proficient subpopulation. Presented at: 2025 Society of Gynecologic Oncology Annual Meeting on Women’s Cancer (SGO); March 14-17, 2025; Seattle, WA.