e15718 Background: The incidence of early-onset colorectal cancer (EOCRC) is increasing, yet its molecular and therapeutic landscape remains incompletely characterized. We compared somatic mutation profiles and the prevalence of clinically actionable, tissue-agnostic genomic alterations between EOCRC and late-onset colorectal cancer (LOCRC) to assess whether EOCRC represents a biologically distinct entity with differing therapeutic implications. Methods: We retrospectively analyzed next-generation sequencing data from patients with colorectal cancer who presented to Moffitt Cancer Center between 2011-2025, reflecting a large tertiary referral population in Western Florida. Patients were stratified by age at diagnosis as either EOCRC ( 65 years, n = 392). Frequency of somatic mutations and targetable genomic alterations were assessed, including biomarkers that inform eligibility for approved tissue-agnostic or molecularly directed therapies. Comparisons between cohorts were performed using chi-square or Fisher’s exact tests as appropriate, with significance defined as p < 0.05. Results: The median age of the EO cohort was 44 at time of NGS compared to 74 in the LO cohort. The EO group demonstrated higher frequencies of classic CRC driver mutations, including TP53 (81.4% vs 70.9%, p = 0.0077) and APC (79.3% vs 70.9%, p = 0.028). The EO cohort was significantly enriched in alterations affecting DNA damage repair including ATM (14.3% vs 7.7%, p = 0.036), BRCA2 (15.0% vs 5.2%, p = 0.0001), POLE/POLD1 (11.4% vs 2.3%, p = 0.0001) as well as ERBB2 mutations (10% vs 4.8%, p = 0.015). BRAF mutations were more frequent in the LOCRC cohort (11.2% vs 5.0%, p = 0.016) as were incidences of BRAF V600E in RAS wild-type disease (p < 0.0001), suggesting greater eligibility for BRAF-targeted therapy. MSI-high tumors were more common in LOCRC (6.6% vs 2.1%, p = 0.029). TMB-high tumors and KRAS G12C occurred at similar frequencies between cohorts. Patients from the EO cohort were significantly more likely to be enrolled in clinical trials (21.4% vs 4.04%, p < 0.0001). Conclusions: EOCRC and LOCRC exhibit distinct somatic mutation profiles and clinically actionable genomic landscapes. EOCRC is characterized by classic colorectal oncogenic drivers and alterations affecting DNA damage repair. LOCRC demonstrates a higher prevalence of alterations associated with immune responsiveness and pathway-directed therapies, including MSI-H, and BRAFV600E. These molecular differences underscore the importance of comprehensive genomic profiling and support the application of tissue-agnostic and molecularly targeted treatment strategies in colorectal cancer.
Huetteman et al. (Thu,) studied this question.