Early-Onset Colorectal Cancer With Liver-Only Metastases: A Retrospective Cohort Study Integrating Prospectively Collected Real-World Clinical and Molecular Data From an Australian National Database (2009–2024) to Guide Treatment Planning

Abstract

Objective

To leverage the Treatment of Recurrent and Advanced Colorectal Cancer (TRACC) registry (an Australian cancer database) to explore the ideal timing and sequence of therapies and the factors influencing these decisions in colorectal cancer (CRC) patients with liver-only metastases to inform contemporary decision-making and future trials.

Study Type

Retrospective registry-based cohort study using the TRACC registry.

Setting and Participants

Consecutive patients with liver-only metastatic CRC enrolled in the TRACC registry.

Main Outcome Measures

To explore cancer biology, intended treatment at presentation, actual treatment received and the resultant outcomes for early-onset CRC (EOCRC) (≤ 50 years) and late-onset CRC (LOCRC) (> 50 years) patients with liver-only metastases from a real-world perspective.

Results

Between 14 January 2009 and 2 September 2024, 1691 patients with liver-only metastatic CRC were enrolled in TRACC. These included 276 EOCRC patients (16.3%) and 1415 LOCRC patients (83.7%). In the EOCRC subset, there were more females (48.2% vs. 34.5%, p < 0.001), less comorbidity (Charlson comorbidity index score 0, 90% vs. 59%, p < 0.001), more left-sided primaries (76.1% vs. 65.7%, p < 0.001), more synchronous disease (53.3% vs. 42.1%, p < 0.001) and BRAF V600E mutations (13.9% vs. 8.1%; p = 0.010). Overall, EOCRC patients had a longer median survival compared with LOCRC patients (3.20 vs. 2.38 years, p < 0.001). For the 662 patients (39.1%) undergoing liver resection, median survival was 5.99 years in EOCRC patients and 5.88 years in LOCRC patients. For all patients and for those undergoing resection, respectively, B-Raf proto-oncogene, serine/threonine kinase (BRAF) (hazard ratio, 1.97 [p < 0.001] and hazard ratio, 2.25 [p < 0.001]) and Kirsten rat sarcoma viral oncogene homologue (KRAS) mutations were associated with worse outcomes (hazard ratio, 1.29 [p < 0.001] and hazard ratio, 1.34 [p = 0.003]).

Conclusion

Differences in sex distribution, BRAF mutation rates, primary tumour site and overall survival suggest biological differences between EOCRC and LOCRC. Liver resection was associated with improved survival in LOCRC, with the benefits of all therapies varying depending on age, primary tumour site and whether patients presented with synchronous or metachronous liver-only metastases.