Dynamics of the Tumor Immune Microenvironment during Neoadjuvant Chemotherapy of High-Grade Serous Ovarian Cancer

Abstract

The dynamic changes in the tumor immune microenvironment (TIME) triggered by neoadjuvant chemotherapy (NAC) have not been clearly defined in advanced-stage ovarian cancer. We analyzed the immunologic changes induced by NAC to correlate them with clinical outcomes. We compared the changes in the immune infiltration of high-grade serous carcinoma biopsies before and after NAC via immunohistochemistry (147 paired samples) and whole transcriptome sequencing (35 paired samples). Immunohistochemistry showed significantly increased PD-L1 levels and TIL levels after NAC. Whole transcriptome sequencing revealed that the stromal score, immune score, and cytolytic activity score significantly increased after NAC. An increased tumor-infiltrating lymphocyte (TIL) level in response to NAC was associated with shorter progression-free survival compared with decreased TIL level after NAC. In tumors with increased TIL levels after NAC, the relative fraction of CD8 T cells and regulatory T cells significantly increased with immunohistochemistry. Post-NAC tumors were enriched in gene sets associated with immune signaling pathways, such as regulatory T cell and JAK/STAT signaling pathways. NAC induced dynamic changes in the TIME that increased TIL levels, but their high abundance did not impart any survival benefit. Our data may provide therapeutic strategies to improve the survival benefit from immunotherapies in ovarian cancer.

Keywords: high grade serous ovarian cancer; immune checkpoint inhibitors; neoadjuvant chemotherapy; tumor immune microenvironment; tumor-infiltrating lymphocytes.

Figure 1

Changes in the tumor immune microenvironment in HGSOC during NAC. (a) Study schema. (b) Sankey plot of immune stage change from pre-NAC to post-NAC via immunohistochemical staining. (c) Representative images of low and high immunohistochemical expression for PD-L1, CD8, Foxp3, and TIL density distribution in ovarian cancer (original magnification; 40× objective). (d) The immune and stromal scores, and immune cytolytic activity were compared between pre- and post-NAC tumor samples. Asterisks indicate statistical significance based on multiple regression adjusting for subtype as a covariate (Stromal Score, p = 3.0 × 10⁻⁶; Immune Score, p = 5.3 × 10⁻⁶; CYT Score, p = 0.021, Wilcoxon test). HGSOC, high grade serous ovarian cancer; NAC, neoadjuvant chemotherapy; TIL, Tumor-infiltrating lymphocyte. * p < 0.05, *** p < 0.001.

Figure 2

Patterns in the pre-NAC tumor immune microenvironment and their association with treatment outcome. (a) Landscape of mutation profiles (Oncoplot). (b) Prognostic values of pre-NAC PD-L1 and TIL densities as indicated by the Kaplan–Meier curves of PFS and OS. (c) Distribution of tumors by pre-NAC PD-L1 and TIL densities, with the CRS. Lines indicate the baseline for each level, and proportions of CRS 3 for each quartile above/below the baseline are indicated in each corner. (d) Forest plot showing the association between immune cell fractions and the CRS. The x-axis shows the log odds ratio of% cell fractions in CRS 1 and 2 versus CRS 3. TMB, tumor mutation burden; NAC, neoadjuvant chemotherapy; TIL, tumor-infiltrating lymphocyte; PFS, progression-free survival; OS, overall survival; CRS, chemotherapy response score.

Figure 3

Comparison of changes in the tumor immune microenvironment between matched samples. (a) Comparison of PD-L1 and TIL densities in pre- and post-NAC samples based on immunohistochemical staining. Scatterplots with linear regression line and shaded 95% confidence region between PD-L1 and TIL changes. (b) Prognostic value of PD-L1 and TIL density change as indicated by the Kaplan–Meier curves of PFS and OS. (c) Representative images of pre- and post-NAC TILs, Foxp3 and CD8 immunostaining (original magnification; 40× objective). (d) Scatterplots with linear regression line and shaded 95% confidence region between TIL level and CD8, Foxp3 staining intensity. ** p < 0.01, *** p < 0.001.

Figure 4

Differentially expressed genes and gene set enrichment analysis in pre- and post-NAC samples. (a) Heatmap comparison of 6 significant genes (>two-fold change and adjusted p < 0.01) between pre- and post-NAC samples. (b) Enriched differentially expressed gene sets between pre- and post-NAC. (c) Gene set enrichment analysis of differentially expressed genes of regulatory T cells in pre- and post-NAC samples. NAC, neoadjuvant chemotherapy.