Association of KRAS G12D vs. G12V circulating tumor DNA variant allele fraction and real-world overall survival in metastatic non-small cell lung and colorectal cancers.

Jacob Edward Till,Leslie A Bucheit,Nicole Zhang,Keelia M. Clemens,Erica L. Carpenter
DOI: https://doi.org/10.1200/jco.2024.42.16_suppl.3041
IF: 45.3
2024-06-01
Journal of Clinical Oncology
Abstract:3041 Background: We previously reported KRAS G12D but not G12V variant allele fraction (VAF), as measured in circulating tumor DNA (ctDNA), was associated with overall survival (OS) in metastatic treatment-naïve pancreatic cancer (mPDAC). However, this has not been examined in other RAS-driven cancers. Here we assessed association of real-world OS (rwOS) with KRAS G12D and G12V VAF in metastatic colorectal cancer (mCRC) and non-small cell lung cancer (mNSCLC). Methods: We queriedGuardantINFORM, a real-world evidence database of aggregated commercial payer health claims and de-identified results from Guardant360 (G360) ctDNA testing from 2014 to 2023, to identify mNSCLC/mCRC patients who were treatment-naive and had KRAS G12D or G12V mutations identified < 30 days prior to first-line therapy initiation. Those with co-occurring KRAS alterations, KRAS amplifications, and/or MSI-H were excluded. Outcomes were assessed as above vs. below median VAF and stratified by treatment type. The Cox regression model was run for continuous VAF, adjusting for age, gender, summary comorbidity score [Elixhauser Comorbidity Index (ECI)], and year of G360. Results: For mCRC, KRAS VAF above median was significantly associated with shorter rwOS for both G12D and G12V (Table 1); subgroup analysis of the predominant standard-of-care chemotherapy, FOLFOX, showed similar findings. For mNSCLC, KRAS VAF above median was associated with shorter rwOS for G12V, but not G12D (Table). Subgroup analysis by first-line treatment type revealed this association to be similar for chemotherapy-containing regimens but not immunotherapy (IO) alone. Conclusions: While a variant-specific association of KRAS VAF with rwOS was observed in mNSCLC overall (G12V associated, G12D not), above vs below median VAF was not associated with rwOS for the subset of patients receiving IO only. In contrast, for mCRC both G12D and G12V VAF were associated with rwOS, including for the subset of patients who received chemotherapy. These data suggest a complex interplay between KRAS variant, ctDNA shedding, first-line treatment, and OS that will be important in the interpretation of the prognostic implications of ctDNA-identified KRAS alterations. [Table: see text]
oncology
What problem does this paper attempt to address?