1887P - The consistency of mutation profiles between treatment-naïve circulating tumor DNA and tumor tissue mutations in a multi-cancer monitoring cohort

Background

Next-generation sequencing (NGS) using circulating DNA (ctDNA) has been used for a non-invasive evaluation of solid tumor loading perioperatively. The genomic profiles between plasma ctDNA and tumor tissue DNA (tDNA) often vary and choosing right targets for an accurate monitoring has not been assessed widely. Herein, we compared the consistency between treatment-naïve ctDNA and tDNA in a multi-cancer monitoring cohort.

Methods

316 patients scheduled for surgical excision were enrolled from Dec. 2018 to Apr. 2020. Surgically resected tumor tissues and plasma samples before treatment were collected and subjected to NGS using Onco-Panscan^TM 825-gene panel. Somatic mutations in matched tDNA and ctDNA were analyzed with 0.1% mutation frequency as the cutoff.

Results

316 patients diagnosed with 7 solid tumor types were accrued. Mutations were identified in 84.2% (266/316) of ctDNA from plasma and in 96.2% (304/316) of tDNA from tissues. The median mutation count of ctDNA and tDNA was 3 (1-110) and 7 (1-659). The median incidence of tDNA detected in ctDNA was 52.9% (22.2% -67.5%), with 67.5% (110/163) lung cancer patients and only 22.2% (2/9) skin cancer patients with tDNA mutations observed in ctDNA. The most frequently mutated genes were TP53 (tDNA/ctDNA, same below. 58.5%/43.2%), EGFR (18.0%/14.3%) and LRP1B (11.4%/11.4%). TP53 mutations demonstrated both high prevalence (58.5%, 185/316) in tDNA and high consistency (consistency, cases with same mutation in ctDNA/cases with mutation in tDNA, same below. 60.4%, 113/185) between tDNA and ctDNA. Furthermore, genes with highest mutation consistency for each tumor types are TP53 (69/106) and EGFR (35/55) in lung cancer, TP53 (12/22), ARID1A (6/11) and KRAS (7/10) in liver cancer, APC (12/22) and TP53 (12/23) in intestinal cancer, PIK3CA (6/7) and TP53 (4/6) in breast cancer, KRAS (6/17) and TP53 (5/14) in pancreas cancer and TP53 (11/13) in stomach cancer.

Conclusions

Our study has demonstrated variations of genomic features in different solid tumor types. This information offers a strategy for selecting targets for an accurate monitoring of tumor loading perioperatively.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.