While the role of CD8+ T cells in mediating response to cancer 1 immunotherapy is well established, the role of other cell types, 2 including B cells, remains more controversial. By conducting 3 the largest gene expression study of response to immune check-4 point blockade (ICB) to date, here we show that pre-treatment 5 expression of B cell genes is associated with ICB response inde-6 pendently of CD8+ T cells. Strikingly, we discovered that such 7 association is completely explained by a single gene expressed 8 in fibroblastic reticular cells (FRCs). We further validated this 9 finding in three independent cohorts of patients treated with 10 ICB, in both melanoma and renal cell carcinoma.
11The role of B cells in cancer immunotherapy is currently a 12 matter of debate (1-3). Recently, it has been reported that 13 expression of B cell specific genes inside tumours is asso-14 ciated with response to immune checkpoint blockade (ICB), 15 thus suggesting an unappreciated biological role for B cells in 16 promoting ICB response (4). However, association between 17 B cell gene expression and ICB response was not statisti-18 cally significant when other immune populations, including 19 CD8+ T cells, were taken into account (4). For this rea-20 son, it remains unclear whether B cells are associated with 21 ICB response independently of CD8+ T cells. Clarifying this 22 point is crucial for the biological interpretation of these re-23 sults. In fact, expression of B cell specific genes in the tu-24 mour microenvironment (TME) is often strongly correlated 25 with markers of CD8+ T cells (5), which express the ac-26 tual molecular targets of ICB treatment and are a validated 27 predictor of immunotherapy response (6-8). Two additional 28 recent studies on this topic did not address the issue, be-29 cause they described predictive signatures containing both B 30 cell and CD8+ T cell markers, as well as genes expressed 31 in other immune cell types (9, 10). In order to address this 32 problem, we analysed the entire transcriptome of a larger 33 number of melanoma samples compared to previous studies 34 (n = 366), and we then considered only samples collected 35 before commencing treatment with immune checkpoint in-36 82 nificant after FDCSP expression was included in the model 83 (Fig. 1D). Taken together, these results suggest that expres-84 sion of FDCSP might be an independent predictor of ICB 85 response. We tested this hypothesis in a completely inde-86 pendent cohort of melanoma patients treated with anti-PD1, 87 either alone or in combination with anti-CTLA4 (23). In this 88 validation cohort, patients with higher expression of FDCSP 89 at baseline experienced significantly longer overall survival 90 Biasci et al. | | 1-S22 (log-rank p < 0.0001) and progression-free survival (log-91 rank p < 0.0001) after commencing treatment with ICB (Fig-92 ures 1B and 1C). Moreover, we observed a significant asso-93 ciation between expression of FDCSP at baseline and subse-94 quent RECIST response (Cochran-Armitage test p < 0.0001; ...