Host immune signatures as predictors of response to immunotherapy-based regimens in patients with metastatic renal cell carcinoma (mRCC)

BACKGROUND: Treatment options for mRCC have evolved to include VEGF targeted therapies (VEGF-TT), immune checkpoint inhibitors (ICIs), or combinations of both. However, clinical responses to systemic therapies in mRCC remain largely unpredictable and robust biomarkers are still lacking. The interaction between the tumor and its immune microenvironment has been shown to influence clinical outcomes in patients treated with ICI-based regimens. The aim of this study was to characterize the T-cell and B-cell immune repertoires in patients with mRCC treated with VEGF-TT, ICI or a combination of both, and evaluate their associations with clinical outcomes. METHODS: We identified patients with mRCC at Dana-Farber Cancer Institute treated with VEGF-TT, ICI or both, and for whom tumor and/or blood samples were available. T-cell receptor sequencing (TCR-seq) was performed on peripheral blood mononuclear cells (PBMCs), collected before and during therapy. Bulk RNA-sequencing (RNA-seq) was performed on available pre-treatment PBMCs and primary tumor samples. Immunoglobulin heavy chain (IgH) isotypes were inferred from bulk RNA-seq data using TRUST4. Parameters of the T-cell and B-cell repertoires were evaluated in responders vs. non-responders to systemic therapies, and between pre- and on-treatment samples within patient subgroups. RESULTS: In total, blood (PBMC) samples from 386 patients were available across all treatment cohorts (186 VEGF-TT, 126 ICI and 74 ICI+VEGF-TT). Following quality-control, TCR-seq data were available for 367 patients (228 pre-treatment and 139 on-treatment), while RNA-seq data were available for 105 PBMC and 17 tumor-derived pre-treatment samples. In the TCR-seq analysis, responders to ICI-based regimens (ICI or VEGF-TT+ICI) presented a trend towards an increased baseline (pre-treatment) TCR clonality as compared to non-responders (p=0.06) (Fig. 1A), corresponding to a less polyclonal T cell repertoire in responders at baseline. No significant changes in clonality were seen between pre- and on-treatment samples among responders to ICI regimens (p=0.14), as opposed to non-responders where a significant increase was identified (p=0.001). Therefore, responders to ICI-based regimens seem to have a more oligoclonal TCR repertoire (increased clonality) at baseline with no treatment-induced changes, whereas non-responders to ICI-based regimens seem to evolve from a more polyclonal to a more oligoclonal TCR repertoire in response to treatment. The analysis of IgH isotypes in baseline blood samples showed higher fraction of IgG1 in responders (vs. non-responders) to ICI regimens (p=0.01) (Fig. 1B). This was further confirmed in the analysis of IgH isotypes inferred from tumor samples (p=0.04). No significant differences in IgH isotype fractions were identified between responders and non-responders to VEGF-TT. Furthermore, while shared clonotypes were detected between blood and tumor samples, there were no differences in the Jaccard similarity index between responders and non-responders to ICI-based or VEGF-TT regimens. [Image: see text] Figure 1: (A) Evaluation of pre-treatment and post-treatment TCR clonality in responders vs. non-responders across treatment cohorts. (B) Differences in the fraction of IGH isotypes between responders and non-responders across treatment cohorts. CONCLUSIONS: We were successfully able to characterize T-cell receptor and B-cell IgH repertoires in a large cohort of patients with mRCC, and evaluate their associations with ICI response. Our results show that baseline TCR clonality and IgG1 antibody fraction are associated with the response to ICI regimens, suggesting a potential role for immune biomarker development. CDMRP DOD Funding: yes