Abstract 4101: Spatial heterogeneity of tumor mutational burden (TMB) counts in pulmonary adenocarcinoma: Separating biological effects from technical artifacts

Abstract

Introduction: Tumor mutational burden (TMB) is an emerging biomarker to identify patients more likely to benefit from immuno-oncologic therapy. Therefore, academia- and industry-driven endeavors are underway to establish mostly panel-based sequencing solutions. Aside from various unsettled technical aspects, tumor biology itself might play an important role in determining a tumor’s TMB. Intratumor heterogeneity (ITH) is frequent in pulmonary adenocarcinoma (ADC) and seems to be a key factor regarding response failure and resistance development during therapy. In addition, ITH might contribute to inconsistent distribution of a biomarker and therefore an assay’s readout might depend on the region of tumor which is submitted for testing. Therefore, the assessment of ITH is an essential step in the evaluation of new diagnostic applications. Experimental Set-up: In order to investigate potential spatial differences in the TMB of a tumor, we analyzed a comprehensively morpho-molecularly characterized cohort of pulmonary ADC by sequencing 3-4 sections per tumor including lymph node metastases. So far 10 tumors were included in the ongoing analysis. None of the selected tumors had an oncogenic EGFR mutation, six were KRAS positive and four had no known driver mutation. TMB was determined applying the TruSightTM Oncology 500 targeted sequencing panel (TSO 500), covering 500 genes, with a NextSeq 500 sequencing system (both Illumina Inc., San Diego, CA, USA). Results Summary: The current sequencing analysis using the TSO 500 panel covered an average genomic coding region of 1.26 Mbp with a mean read depth of x658. The average TMB of all samples was 8.55 mut/Mbp (range 0.79 - 18.95). Comparing different regions within a given tumor, we detected a considerable variance of TMB (± 5 mut/Mbp) in about half of the analyzed cases, with an overall mean absolute deviation of 2.2 and a maximum difference of 12.63 mut/Mbp. However, in some samples an in depth evaluation of morpho-molecular characteristics revealed several factors impairing TMB read out. Aside from technical issues like insufficient depth of coverage in combination with unadjusted automated variant calling, sample characteristics like tumor cell content had a prominent influence on TMB values. Often, tumor regions with lower TMB turned out to have lower tumor cell content, which led to decreased allele frequencies of tumor specific mutations and subsequently compromised mutation detection. Conclusions: Our data demonstrate spatial heterogeneity of TMB in pulmonary ADC, which can impact clinical decision making and therapy. However, spatial genetic heterogeneity reflecting tumor biology needs to be carefully dissected from technical conditions resulting in pseudo-heterogeneity. Citation Format: Volker Endris, Michael Allgauer, Jan Budczies, Jonas Leichsenring, Anna-Lena Volckmar, Martina Kirchner, Olaf Neumann, Regine Brandt, Eugen Rempel, Carolin Ploger, Moritz von Winterfeld, Alexander Harms, Mark Kriegsmann, Roland Penzel, Peter Schirmacher, Daniel Kazdal, Albrecht Stenzinger. Spatial heterogeneity of tumor mutational burden (TMB) counts in pulmonary adenocarcinoma: Separating biological effects from technical artifacts [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 4101.