Tumor-stroma proportion on primary tumor as a prognostic biomarker in advanced ovarian cancer patients receiving chemo-immunotherapy as first-line therapy: analyses from the NeoPembrOV/GINECO phase II randomized trial

Abstract

Background: The efficacy of immune checkpoint inhibitors is limited in patients with high-grade serous ovarian cancer (HGSC). The predictive and prognostic value of tumor-stroma proportion (TSP) was assessed in patients with HGSC treated with platinum-based chemotherapy and pembrolizumab in the NeoPembrOV trial.

Materials and methods: TSP was quantified as the relative proportion of stromal tissue to tumor cells (low if <30% or high if ≥30%). RNA sequencing and multiplex immunofluorescence were conducted.

Results: TSP was assessed on 85 pre-treatment samples. Patients with a low TSP had a prolonged progression-free survival (PFS) compared with those with a high TSP (median PFS 23.4 versus 18.3 months, respectively, hazard ratio 0.51, 95% confidence interval 0.31-0.83, P = 0.010). The prognostic impact was higher when TSP was assessed on tubo-ovarian primary tumors (P = 0.01) and remained significant in the pembrolizumab arm (P = 0.01). Tumors with a low TSP were enriched in intratumoral CD8+PD1+ T cells and stromal proliferative CD8+Ki67+ T cells, while tumors with a high TSP exhibited an enrichment in M2 macrophages. A significant increase in intratumoral CD8+ T cells following pembrolizumab was observed only in low-TSP tumors (P = 0.02).

Conclusions: We confirmed the prognostic value of TSP in HGSC and demonstrated its significance in patients treated with chemo-immunotherapy. TSP is associated with an immunosuppressive tumor microenvironment and influences CD8+ T-cell enrichment with immunotherapy.

Keywords: high-grade serous ovarian cancer; immunotherapy; platinum-based chemotherapy; tumor microenvironment; tumor–stroma proportion.