Abstract

Abstract Background: Initially developed as an alternative to tissue-based molecular assessment, LB is expanding its role in cancer care, with quantitative measure of blood ctDNA being tested for response prediction, response assessment and minimal residual disease detection. Its use for molecular classification has been extensively validated, but little is known about determinants of shedding parameters such as VAF. While it is correlated with tumor burden, characteristics of the tumor such as site of metastasis or particular mutations may have a role. Scope of this study is to elucidate these factors. Methods: Data from advanced NSCLC patients included in two prospective cohorts who received a LB with 36-genes InVisionFirst-Lung (INI) and the 324-gene FoundationOne Liquid CDx (FMI) assays were reviewed and patients with available 18F FDG positron emission tomography scan (PET) performed within 30 days from LB were included. Total Metabolic Tumor Volume was calculated for each PET with a 42% SUVmax threshold. The higher VAF (mVAF) was calculated for each LB, excluding germline mutations and clonal hematopoiesis-related variants. Patients should be either untreated or with progressive disease at the time of LB. Regression analysis was performed according to a rank-based estimation model. Results: Overall 301 LB were included, 119 with InVisionFirst-Lung and 182 with FoundationOne Liquid CDx. LB were contributive in 154 (85%) FMI and 102 (86%) INI. MaxVAF was correlated with tMTV in FMI ( rho 0.3571, p 0.0004) and IVI (rho 0.3387, p = 0.0003) cohort. In FMI cohort, median mVAF was higher for TP53 mutant vs wt (n= 111, mVAF 21.1 % vs 8.0%, p < 0.0001), RB1 mutant (n = 11, mVAF 26.3% vs 4.7%, p 0.0103) and in those with liver metastasis ( p 0.0089). LDH was correlated to mVAF (rho 0.280, p = 0.0037). In multivariate model, tMTV (p < 0.0001), TP53 mutation ( p 0.0039), RB1 mutation (p 0.0003) and liver metastasis ( p 0.0371) remained significantly associated with mVAF. When LDH was added to the model (LDH available for 119 pts), tMTV lost its significance (p 0.561). In INI cohort, median mVAF was also higher in patients with TP53 mutation ( 7.1% vs 3%, p 0.0038), in patients with liver metastasis ( 10.1% vs 4.6%, p 0.0299), and it was correlated with tMTV (rho 0.502, p < 0.0001). It was confirmed in a multivariate model (p < 0.0001 for tMTV and 0.0049 for TP53). Correlation between mVAF and LDH was also seen (LDH available for 54 pts, rho 0.591, p < 0.0001). Among patients with liver metastasis, mVAF was correlated with MTV of liver lesions in FMI (n = 30, rho = 0.212, p = 0.0069) and INI cohort (n= 26, rho 0.204, p 0.0313). Overall, in patients with LDH < ULN, 18/103 were non contributive, vs 4/70 in those with LDH > ULN. Conclusion: In patients with aNSCLC, mVAF depends on disease burden, on the molecular characteristics of the disease and on the presence and extent of liver disease. LDH levels above ULN predict the presence of ctDNA and percentage of mVAF. Citation Format: Filippo Gustavo Dall'Olio, Damien Vasseur, Camilo Garcia, Arianna Marinello, Mihaela Aldea, Laura Mezquita, Jordi Remon, Anas Gazzah, Marco Tagliamento, Kristi Beshiri, Pernelle Lavaud, Maud Ngocamus, David Planchard, Etienne Rouleau, Antoine Italiano, Benjamin Besse. Clinical and genomic correlates of liquid biopsy (LB) derived variant allele frequency (VAF) in advanced NSCLC patients [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 3361.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call