Abstract
Introduction: Epigenetic dysregulation, driven by alterations in the lysine methyltransferase 2C (KMT2C) gene, contributes to tumor growth. KMT2C is a putative tumor suppressor mutated in 7%-10% of solid tumors. Loss of KMT2C function impairs homologous recombination DNA repair, making tumor cells more susceptible to PARP inhibition. The study used a real-world dataset to assess if KMT2C mutations predict platinum-based chemotherapy (PBC) responsiveness and clinical outcomes in solid cancers.
Methods: The Tempus Lens Platform (Tempus AI, Inc., Chicago, IL) was used to query the Tempus multimodal de-identified database and establish and subsequently analyze a cohort of patients (pts) with solid cancers and Tempus xT (DNA) testing. Pts characteristics were compared using Chi-squared/Fisher’s exact tests and/or Wilcoxon sum tests. Real-world overall survival (rwOS) and progression-free survival (rwPFS) were compared between KMT2Cmut vs KMT2Cwt and defined as the time from PBC initiation to death or loss to follow up (FU) and as the time from PBC initiation to first progression, death, or loss to FU, respectively. Median rwOS and rwPFS were estimated by Kaplan-Meier and compared using log-rank tests. Risk set adjustment was used to avoid immortal time bias due to left truncation. Findings were validated in the AACR Genie dataset for rwOS after PBC and non-PBC.
Results: A total of 143,961 patients were included, and stage IV disease accounted for 76%. GI cancers comprised 36% of the cohort, followed by lung (25%) and breast (11%). KMT2C mutations occurred in 5.3% overall, most frequent in cervical (14%), endometrial (11%), urinary tract (8.6%) and breast cancer (8.4%). African American patients showed slightly higher prevalence versus non-AA both overall (5.6% vs 5.2%, p=0.089) and in GI cancers (5.4% vs 4.6%, p=0.013). In PBC-treated patients, median rwOS was longer in KMT2Cmut vs KMT2Cwt across all cancers (19.6 vs 16.7 mo, p=0.038), endometrial (42.3 vs 31.9 mo, p=0.03), head and neck (28.4 vs 13.3 mo, p=0.02), and GI (18.9 vs 15.7 mo, p=0.006). Within the GI group, colorectal cancers (CRC) had the largest impact (median rwOS of 51.0 vs 25.3 months, p=0.01). In CRC, rwPFS for KMT2Cmut vs KMT2Cwt were Not Reached vs 16.7 months (p=0.001), respectively. AACR Genie was analyzed to validate these results. A total of 1,551 CRC cases were included, with 8% harboring KMT2C mutations. CRC KMT2Cmut had improved rwOS compared to KMT2Cwt following PBC (Not Reached vs 57.5 months, p=0.005). This effect appeared specific to PBC, with no difference after irinotecan-based therapy.
Conclusion: KMT2C mutations are linked to improved survival after platinum chemotherapy, particularly in CRC and other GI cancers, supporting KMT2C as a predictive biomarker. Laboratory studies to define the mechanistic basis of this association are underway
VIEW THE PUBLICATION