PD-L1 Warrants Further Research as Biomarker for Endometrial Cancer Therapy Response

Additional biomarkers are needed to identify women with advanced endometrial carcinoma (EC) with mismatch repair (MMR) deficiency who would benefit from immune checkpoint inhibitors, and a recent study found potential significance in programmed death ligand-1 (PD-L1) analysis results. The findings were published in the journal Cancers.

For women with advanced EC with MMR deficiency, outcomes have historically been poor. Immune checkpoint inhibitors such as durvalumab have changed the treatment landscape and improved outcomes for some of these patients, but identifying individuals who are likely to respond is crucial to prevent overtreatment in patients who are unlikely to benefit from immune therapy.

The study aimed to determine the interobserver reliability of PD-L1 expression in tumors, the optimal cut points for PD-L1 expression, and whether PD-L1 has prognostic value among participants in the phase 2 PHAEDRA trial evaluating durvalumab, an anti–PD-L1 monoclonal antibody, in advanced EC with and without MMR deficiency.

Seventy-one women with advanced EC were included in the study: 36 with deficient MMR (dMMR) and 35 with proficient MMR (pMMR). The objective tumor response rates (OTRRs) for participants with dMMR and pMMR were 47% and 3%, respectively. Thirty-three women with pMMR and 34 women with dMMR had sufficient tumor for PD-L1 evaluation.

Scores for PD-L1, immunohistochemical staining of tumor (TC+), immune cells (IC+), and the presence of tumor-associated immune cells (ICP) were reported and evaluated for correlations with OTRR, progression-free survival (PFS), and overall survival (OS). Each of these aspects underwent receiver operating characteristic analyses, and area under the ROC curve, sensitivity, and specificity were calculated for each component to determine their prognostic values and ideal cut points. An algorithm based on the optimal cut points was also developed and tested.

The authors determined that the optimal cut points were a TC+ score of 1 or higher;, ICP 10 or higher; and IC+, 35 or higher. The highest sensitivity and specificity were seen with the ICP cut point, which showed 53% sensitivity and 82% specificity for PFS before adjusting for MMR. It was also prognostic for OTRR, but only before adjusting for MMR status in both cases. TC+ and IC+ were not found to be prognostic individually.

The optimal cut points algorithm was associated with OS before adjusting for MMR status, but the results were not significant after adjusting for MMR status. However, the algorithm did identify nonresponders with 88% sensitivity and 92% negative predictive value, with results still significant after adjusting for MMR status.

Overall, MMR status remained the most predictive factor for response to immune therapy in this cohort. However, the authors concluded that the significance of ICP of 10 or higher and the optimal cut point algorithm overall warrant further research.

Study limitations included the post-hoc nature of the analysis and the inclusion of samples from both primary tumor and metastatic tumor biopsies, although neither sample type had enough for subset analyses. There may also be unidentified molecular factors affecting PD-L1 expression, the authors noted. Additional research in a larger cohort is necessary to confirm these findings and identify additional markers of immune therapy response in patients with advanced EC.

Reference

Smith D, Robledo KP, Yip S, et al. Results of PD-L1 analysis of women treated with durvalumab in advanced endometrial carcinoma (PHAEDRA). Cancers (Basel). Published online December 30, 2022. doi:10.3390/cancers15010254