Abstract 501P
Background
FDA has now approved pembrolizumab for all unresectable or metastatic TMB-H tumors, tissue based (invasive) Comprehensive Genomic Profiling (CGP) is considered as gold standard for the same. However, it has certain limitations because adequate tissue is often difficult to acquire. In contrast, ctDNA (non-invasive) CGP is now, increasingly redefining the molecular landscape in evaluating TMB in addition to establishing usefulness, combating tumor molecular heterogeneity. We hypothesise, in advanced cancer patients, complementary CGP (including TMB) of both, tissue and ctDNA should be done to identify eligible patients for immunotherapy and targeted therapy to achieve maximum beneficial outcomes.
Methods
We conducted a retrospective analysis of 194 advanced cancer patients with test results from commercial tissue (Guardant360 TissueNextTM) and/or ctDNA (Guardant360®) next-generation sequencing assays. For 110 patients, both tissue and ctDNA results were available (Arm T+L); 84 had only ctDNA results (Arm L). TMB-H was defined as >10 muts/Mb in tissue or >16 muts/Mb in ctDNA.
Results
TMB-H was detected in 38/194 patients (19.5%) with 19/38 (50% Arm T+L & T each) patients ranging in 17 different histologies. In Arm T+L, 5/19 (26.3%) had TMB-H in both, tissue and ctDNA; 8/19 (42.1%) in tissue only and 6/19 (31.5%) in ctDNA only. Average number of mutations detected of TMB-H patients in ctDNA was 15 muts/ sample while in tissue was 10 muts/ sample. TMB was not evaluable in 51/194 (26%). But, in Arm T+L, 39/51 (76.4%) had TMB evaluable in either tissue only 8/39 (20.5%) or ctDNA only 31/39 (79.4%) whereas in Arm L, 9/51 (17.6%) and 3/51 (6%) in neither of two samples.
Conclusions
Our study population had 19.5% prevalence of TMB-H. ctDNA demonstrated a significantly higher detection rate with much lower rate of failure of detection. The non-invasive nature with higher sensitivity of ctDNA analysis makes it an attractive alternative to tissue biopsies. In situations where tissue is available, Complementary CGP testing of both, tissue and ctDNA should be encouraged to adequately identify patients for immunotherapy and targeted therapy.
Clinical trial identification
Editorial acknowledgement
Legal entity responsible for the study
The authors.
Funding
Has not received any funding.
Disclosure
A. Bahl, N. Rohatgi: Financial Interests, Personal, Advisory Board: Guardant Health. All other authors have declared no conflicts of interest.