84MO - Tumour-only sequencing led to inflated tumour mutational burden estimation especially in under-represented ethnic groups

Background

Immune checkpoint inhibitors (ICIs) have dramatically improved the survival of cancer patients. Since the auto-immune toxicities with ICIs can be fatal, it is critical to optimize patient selection criteria. The current use of PD-L1 level and mismatch-repair/microsatellite-instability status has limitations. Recently tumor mutational burden (TMB) has emerged as a promising biomarker for ICI response. Accurate TMB estimation requires patient germline sequencing to filter out non-somatic variants; however, TMB is often calculated from tumor-only sequencing, relying on public databases (DB) to filter out non-somatic polymorphisms. We previously reported that filtering based on public DB significantly inflated TMBs. In this study, we investigated the impact of under-representation of minority ethnic groups in these DB and hypothesized that the inflation of TMBs will be racially disparate with falsely higher TMB in Black vs. White individuals.

Methods

Variants were identified in individual tumor and germline exomes of 157 Black and 713 White patients sequenced by Multiple Myeloma Research Foundation (MMRF). TMBs for each patient were calculated using 4 filtering criteria: (1) excluding variants in paired germline exomes; (2) excluding variants with minor allele frequency (MAF) ≥ 1% in 1000G/ExAc DB; (3) excluding variants with MAF ≥ 0.1% in 1000G/ExAc; and (4) excluding all variants in 1000G/ExAc.

Results

Compared to the gold standard of filtering germline variants by patient-paired germline sequencing data, TMBs were significantly higher in both Black and White patients using public DB for filtering regardless of the MAF thresholds. Furthermore, the TMBs were more significantly overestimated in Blacks compared to Whites (Table). Table: 84MO

Conclusions

TMB as a biomarker for selecting patients to receive ICIs without patient-paired germline sequencing may introduce racial bias.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

M. Borad: Research grant/Funding (institution): Senhwa Pharmaceuticals, Adaptimmune, Agios Pharmaceuticals, Celgene Pharmaceuticals, EMD Merck Serono, Toray, Dicerna, Taiho Pharmaceuticals, Sun Biopharma, Isis Pharmaceuticals, Boston Biomed, Basilea, Incyte Pharmaceuticals, Mirna Phamraceuticals, Medim; Advisory/Consultancy, To self: ADC Therapeutics, G1 Therapeutics, Immunovative Therapies, OncBioMune Pharmaceuticals, Western Oncolytics, Lynx, Genentech, Merck, Huya, ; Travel/Accommodation/Expenses: AstraZeneca. A.S. Mansfield: Research grant/Funding (institution): Novartis & Verily; Honoraria (institution), Advisory Boards: AbbVie, AstraZeneca, BMS & Genentech; Travel/Accommodation/Expenses: Roche; Non-remunerated activity/ies, Director: Mesothelioma Applied Research Foundation. All other authors have declared no conflicts of interest.