Search

Context: Polymerase epsilon (POLE) gene missense hotspot mutations can generate pathogenic (p) proofreading defects resulting in hypermutated genomic profiles. Aim: Determine the prevalence, genomic consequences and immunotherapy sensitivity of advanced POLE mutated tumors according to mutation site, primary tumor and tumor mutational burden (TMB). Results: Pan-Cancer TCGA & MSKCC databases genomic analyses found a prevalence of non-pathogenic POLE mutations (POLEnp) of 3.4% with median TMB of 11 mutations/Megabase (mt/Mb, IQR 3-34). Pathogenic POLE mutations (POLEp) prevalence was 0.4% with median TMB of 215 mt/Mb (IQR 107-324), predominantly in colorectal and endometrial cancers. Prevalence dropped to 0.1% in metastatic cancers. We assessed prospectively the efficacy of PD-1 blockade in mismatch repair proficient advanced solid tumors harboring POLE missense mutations (phase II ASCe Nivolumab trial; NCT03012581). Variants were categorized prospectively by a molecular board as POLEp, POLEnp or Variants with Unknown significance (VUS). The primary endpoint was the Overall Response Rate (ORR) at 12 weeks according to RECIST 1.1, and secondary endpoints included survival analyses according to POLE variants pathogenicity. Among 61 screened patients, 21 were eligible and 20 received Nivolumab and 19 were assessable for response (table 1). The 12-week ORR was 37% for patients harboring POLEp and VUS and resulted in major survival improvement compared to POLEnp patients (HR=0.1 ; CI95% 0.02-0.7); see results in Table 1. Among patients POLEp tumors, while higher TMB was not predictive of response, higher proportion of POLE-related mutational signature correlated with improved benefit. In silico exonucleasic POLE domain analyses confirmed that all POLEp and 2 VUS clustered in the DNA binding or the Catalytic site. Recategorizing the VUS according to the location within the exonucleasic domain improved the prediction of survival outcomes. Impact: This study gives new insights on how DNA repair defects, mutational burden and signatures sensitize to PD-1 blockade and may offer emerging tumor agnostic biomarkers for benefit to checkpoint blockade. POLE variant pathogenicity All(N=21) POLEnp(N=5) VUS(N=4) POLEp(N=12) Age, years ± SD 57 ± 16 64 ± 10 56 ± 16 54 ± 17 Sex, Male (%) 12 (57) 5 (100) 2 (50) 5 (42) PS (ECOG)=1 (%) 16 (75) 4 (80) 2 (50) 10 (83) Primary tumor Colorectal 9 (43) 2 (40) 2 (50) 5 (42) Endometrial 6 (29) 0 (0) 0 (0) 6 (50) Gastric 2 (9) 2 (40) 0 (0) 0 (0) Glial 1 (5) 0 (0) 0 (0) 1 (8) Biliary tract 1 (5) 0 (0) 1 (25) 0 (0) Pancreas 2 (9) 1 (20) 1 (25) 0 (0) Number of previous treatments 2.4 ± 2 5 ± 2 1.8 ± 1 1.5 ± 1 TMB (mt/Mb, Min-Max)(N=16) 36.2 (2-385) 5 (4-9) 3 (2-4) 114 (25-385) ORR at 12 weeks (CR+PR) 37%(N=7/19) 0%(N=0/5) 50%(N=2/4) 46%(5/10) DCR at 12 weeks (CR+PR+SD) 58%(N=11/19) 0%(N=0/5) 75%(N=3/4) 80%(8/10) Median Progresssion-Free survival (months) 5.6 2.3 10.3vs POLEnp: HR=0.2 IC95% 0.1-0.7 Median Overall Survival (months) 9.1 5.0 Not Reachedvs POLEnp:HR=0.1 IC95% 0.02-0.7 Citation Format: Benoît Rousseau, Ivan Bieche, Eric Pasmant, Nadim Hamzaoui, Nicolas Leulliot, Lucas Michon, Aurelien de Reynies, Mike Foote, Julien Masliah-Planchon, Magali Svrcek, Romain Cohen, Victor Simmet, Paule Augereau, David Malka, Antoine Hollebecque, Damien Pouessel, Carlos Gomez-Roca, Rosine Guimbaud, Amandine Bruyas, Marielle Guillet, Muriel Duluc, Sophie Cousin, Christelle de la Fourchardiere, Frederic Rolland, Sandrine Hiret, Esma Saada-Bouzid, Olivier Bouche, Thierry Andre, Diane Pannier, Farid El Hajbi, Stephane Oudard, Christophe Tournigand, Jean-Charles Soria, Drew Gerber, Dennis Stephens, Michelle Lamandola-Essel, Steven B Maron, Bill Diplas, Guillem Argiles, Asha Krishnan, Neil Segal, Andrea Cercek, Nathalie Hoog-Labouret, Frederic Legrand, Clotide Simon, Assia Lamrani-Ghaouti, Luis A. Diaz, Pierre Saintigny, Sylvie Chevret, Aurelien Marabelle. PD-1 blockade in solid tumors with defects in polymerase epsilon [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr CT021.

Colon tumors with deficient DNA mismatch repair (dMMR) are infiltrated more densely by T cells than tumors with proficient mismatch repair (pMMR). However, high tumor-infiltrating lymphocytes (TIL) are found in a subset of pMMR tumors and low TIL are seen in a subset of dMMR colon cancer. In this study, we compared T-cell repertoires in 20 pMMR and 27 dMMR colon tumors with high and low TIL counts. CDR3 regions of the TCR beta chain were sequenced by ImmunoSEQ Assay (Adaptive Biotechnologies, Seattle, WA). The fraction T cells was calculated by normalizing TCR beta template counts to the total DNA. Sample richness was calculated as the number of unique productive rearrangements in a sample after computationally down-sampling to a common number of T cells to control for variation in sample depth. Simpson clonality was computed as a measure of how evenly productive rearrangements are distributed amongst the total number of T cells. Statistical analyses were performed using Prism 8.0 (GraphPad, La Jolla, CA). The mean TIL counts per high power field (HPF) for TIL-high and TIL-low dMMR tumors were 17.4 (range, 5-37) and 1.8 (range, 0-4). In pMMR tumors, they were 12.8 (range, 6-37) for TIL-high and 0.3 (range, 0-1) for TIL-low groups. The number of T cells measured by TCR sequencing correlated positively with the TIL count (Spearman rho = 0.391; P = 0.007). When comparing by MMR status, we found that dMMR tumors exhibited a trend towards higher Simpson clonality index (P = 0.09), with significantly lower repertoire richness compared to pMMR tumors (P = 0.030). Within dMMR tumors, TIL-high dMMR tumors had higher density (P = 0.017), higher clonality (P = 0.003), and a less rich repertoire (P = 0.003) compared to TIL-low dMMR tumors. Within pMMR tumors, while the T cell density was higher in the TIL-high tumors, we did not observe differences in T cell clonality or repertoire richness between TIL-high vs TIL-low pMMR tumors. Our study found that T cells in dMMR tumors are more clonal and their repertoire is less rich compared with T cells in pMMR tumors. In the dMMR group, T cells in TIL-high tumors were more clonal and their repertoire was less rich compared with T cells in TIL-low tumors, but in the pMMR group, T-cell diversity in TIL-high tumors was comparable to T-cell diversity in TIL-low tumors. These findings suggest that T cells clonally expand in dMMR tumors, possibly in response to MMR deficiency-induced tumor neoantigens. Citation Format: Jin K. Kim, Chin-Tung Chen, Asama Khan, Neil Segal, Zsofia Stadler, Jinru Shia, Vinod P. Balachandran. Comparison of intratumoral T-cell repertoires in DNA mismatch repair proficient and deficient colon adenocarcinoma [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1382.