Abstract P3-10-09: Relationship between tumor infiltrating lymphocytes (TILs) and response to pembrolizumab (Pembro)+chemotherapy (Chemo) as neoadjuvant treatment (NAT) for triple-negative breast cancer (TNBC): phase Ib KEYNOTE-173 trial

Abstract

Abstract Background:Increasing quantities of stromal TILs (sTILs) are associated with higher pathologic complete response (pCR) rates with conventional chemo in early-stage TNBC. We evaluated the association between sTILs and PD-L1 expression with response to pembro+chemo as NAT for TNBC in the KEYNOTE-173 trial (NCT02622074). Methods: sTILs were quantified using light microscopy of H&E-stained slides from pretreatment and on-treatment (during first 3 weeks of pembro monotherapy) tumor biopsies by a pathologist blind to response data. Pretreatment PD-L1 expression was assessed using the PD-L1 IHC 22C3 pharmDx assay and reported as combined positive score (CPS). Endpoints were pCR rate by ypT0 ypN0 and ypT0/Tis ypN0 and objective response rate (ORR; RECIST v1.1) after the first 4 cycles of NAT (taxane±carboplatin+pembro) by MRI. sTILs and PD-L1 CPS were evaluated as continuous variables. Association between sTILs and PD-L1 CPS with response was assessed using logistic regression and area under the reciever operating curve (AUROC) analyses, with a 1-sided alpha level of 0.10. Correlation between PD-L1 and sTILs was assessed by Spearman's rank correlation coefficient. Multivariate analysis included sTILs (pretreatment and on-treatment) and PD-L1 CPS. Likelihood ratio tests were used to evaluate the added value of factors in predicting pCR rate. Results: Of 60 total pts, 34 had tumors evaluated for pretreatment sTILs, 52 for PD-L1 CPS, and 33 for both sTILs and CPS. On-treatment sTILs were evaluated in 31 pts. Overall pCR rates were 56.7% and 60% by ypT0 ypN0 and ypT0/Tis ypN0, respectively; ORR was 78.3%. In pts evaluated for sTILs and CPS (individually), pCR rates and ORR were comparable with overall pCR rates and ORR. There was a significant correlation between pretreatment sTILs and PD-L1 CPS (ρ=0.65, P<0.001).Higher pretreatment sTILs were significantly associated with response: ypT0 ypN0 P= 0.011; ypT0/Tis ypN0 P=0.006; ORR P=0.061. On-treatment sTILs were also significantly associated with response: ypT0 ypN0 P=0.061; ypT0/Tis ypN0 P=0.041; ORR P=0.031. Pretreatment PD-L1 CPS was significantly associated with response: ypT0 ypN0 P=0.073; ypT0/is ypN0 P=0.030; and ORR P=0.021. AUROC of pretreatment sTIL association with pCR was numerically higher than with on-treatment sTILs and PD-L1 CPS (0.69 vs 0.61 vs 0.56 for ypT0ypN0 and 0.72 vs 0.67 vs 0.62 for ypT0/Tis ypN0). Responders had higher median pretreatment sTIL levels vs nonresponders: 45% [10, 75] vs 10% [5, 20] for pCR rate by ypT0 ypN0 and 52.5% [10, 73.8] vs 10% [5, 20] for pCR rate by ypT0/Tis ypN0; 25% [5, 70] vs 10% [6.3, 27.5] for ORR. In multivariate analysis, only pretreatment sTILs were significant for both pCR endpoints (ypT0 ypN0 P=0.031; ypT0/Tis ypN0 P=0.034). Likelihood ratio tests demonstrated that for both pCR endpoints, PD-L1 CPS (P=0.683/P=0.422) and on-treatment sTILs (P=0.984/P=0.568) did not add significantly more value to pretreatment sTILs when predicting pCR. Conclusions:Higher quantities of pretreatment sTILs and PD-L1 CPS and on-treatment sTILs were significantly associated with higher pCR rates and ORR in primary TNBC treated with pembro and NAT. Citation Format: Loi S, Schmid P, Cortés J, Park YH, Muñoz-Couselo E, Kim S-B, Sohn J, Im S-A, Holgado E, Foukakis T, Kuemmel S, Dent R, Wang A, Aktan G, Karantza V, Salgado R. Relationship between tumor infiltrating lymphocytes (TILs) and response to pembrolizumab (Pembro)+chemotherapy (Chemo) as neoadjuvant treatment (NAT) for triple-negative breast cancer (TNBC): phase Ib KEYNOTE-173 trial [abstract]. In: Proceedings of the 2018 San Antonio Breast Cancer Symposium; 2018 Dec 4-8; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2019;79(4 Suppl):Abstract nr P3-10-09.