3206 - Context-Dependent Impact of Tumor-Infiltrating B Cells in High-Risk Prostate Cancer and Head and Neck Cancer

S. Kim¹, J. Murtadha², S. Miyauchi¹, S. T. Ryan³, J. Zhang⁴, E. Pittman⁵, K. Messer⁶, M. Karin², C. Kane⁷, A. Shabaik², S. B. Howell², C. Jamieson², and A. Sharabi¹; ¹UC San Diego, Moores Cancer Center, Department of Radiation Medicine and Applied Sciences, La Jolla, CA, ²University of California, San Diego, La Jolla, CA, ³MaineHealth, South Portland, ME, ⁴University of California, San Diego, La Jola, CA, ⁵Department of Family Medicine, Division of Biostatistics, University of California San Diego, La Jolla, CA, ⁶UC San Diego Moores Cancer Center, La Jolla, CA, ⁷University of California San Diego, San Diego, CA

Purpose/Objective(s): Tumor-infiltrating B cells have recently garnered significant interest; however, there are differing reports regarding the impact of B-cells on patient outcomes. In Head and Neck Cancer, B-cells have been reported to correlate with significant improvements in survival. On the other hand, in high-risk prostate cancer (HR-PCa), a high density of immunosuppressive B cells was found to correlate with biochemical failure. In this context, a single arm prospective study (PROTUX), was previously performed using neoadjuvant rituximab to deplete B cells in patients with HR-PCa prior to prostatectomy. Here we first report outcomes of deep TCR sequencing on PROTUX samples. We then perform a pan-cancer analysis of plasma cells and follow up with mechanistic studies using immunotherapies in murine tumor models to elucidate the impact of B cells and B-cell depletion on T cells and cancer outcomes. Materials/

Methods: A single arm study enrolled patients with HR-PCa to receive 4 cycles of neoadjuvant rituximab prior to radical prostatectomy (PROTUX, NCT: 01804712). A matched cohort of patients with HR-Pca who underwent prostatectomy without any prior therapy were used as controls. Genomic DNA specimens from 14 patients were isolated using a QIAamp DNA FFPE Advanced UNG kit. TCR sequencing was performed with Adaptive Biotechnologies using the Ultradeep TCRB v4b assay. Groups were compared using a Mann-Whitney U test. For murine models, a HNSCC line, AT84-E7-OVA, was injected into the buccal mucosa. In vivo therapeutics include anti-PD-L1 mAB (10F.9G2), anti-CD40 mAB (FGK4.5), and TLR9 agonist (ODN 1826). Statistical analyses were done in Prism 10 and R.
Results: We observed that patients treated with rituximab had higher TCRb productive clonality (0.080 vs 0.026, p = 0.03). Strikingly, the number of T cells was significantly greater in patients receiving rituximab compared to controls (1.18 cells/ng DNA input vs 0.10 cells/ng, p = 0.002). Maximum productive frequency and the cumulative frequency of the top 10 clones were not significantly different between the two cohorts. We found that the T cells from patients who received rituximab had longer TCRb CDR3 regions (proportion CDR3 =16 amino acids of 0.232 vs 0.197, p = 0.007), suggesting a less antigen-experienced phenotype. In mechanistic studies, we find that plasma cell- and B cell-directed immunotherapies, anti-CD40 and TLR9 agonist, in anti-PD-L1-resistant orthoptic models significantly improve tumor control.
Conclusion: Analysis of human samples from PROTUX trial demonstrated that B cell depletion impacts T cell infiltration, clonality, and phenotype. Furthermore, B cell-directed immunotherapies in tumor models improved local control. Given the increasing appreciation for crosstalk between T cells and B cells in mediating anti-tumor immune responses, these findings demonstrate a context and tumor-type dependent role for B-cells in cancer.