Background: Genomics-based precision medicine has greatly improved how patients with cancer are being treated with targeted agents, but clinical-grade genomic biomarkers for chemotherapies are currently lacking. The chemotherapeutic trifluridine/tipiracil (FTD/TPI) is approved for the treatment of late-stage metastatic colorectal cancer (mCRC). We aimed to find genomic biomarkers to improve patient selection for FTD/TPI treatment in mCRC. Methods: In a discovery cohort of FTD/TPI-treated mCRC patients (n = 37), genome-wide somatic variants were tested for association with treatment duration and overall survival (OS). In vitro drug testing on isogenic cell lines and patient-derived mCRC organoids, as well as a re-analysis of the double-blind, placebo-controlled, phase 3 RECOURSE trial (n = 800) were performed to support our findings. Results: In the discovery cohort, KRAS codon G12 (KRAS^G12) mutation status was the only significant genomic determinant of poor outcome of FTD/TPI treatment, which could be replicated in vitro by drug testing on isogenic cell lines and PDOs. In these models, KRAS^G12 mutations were associated with increased resistance to FTD-induced (geno)toxicity in vitro. KRAS^G12-based resistance was absent for the closely related chemotherapeutic 5-FU. In the RECOURSE study, KRAS^G12 mutations were predictive biomarkers for reduced OS benefit of FTD/TPI vs placebo (unadjusted interaction P= 0.0017, adjusted interaction P= 0.017). For patients with KRAS^G12 mutations, OS was not significantly prolonged with FTD/TPI vs placebo (n = 279; HR, 0.97; 95% CI, 0.73−1.20; P= 0.85). An exploratory analysis showed that the KRAS^G13 mutant subgroup demonstrated clearly prolonged OS with FTD/TPI vs placebo (n = 60; unadjusted HR, 0.29; 95% CI, 0.15−0.55; P< 0.001; adjusted HR, 0.20; 95% CI, 0.092−0.45; P< 0.001), which was significantly more pronounced as compared to the KRAS^G12 mutant and KRAS^WT populations (adjusted interaction P< 0.001 and P= 0.036, respectively). Conclusions: Together, KRAS^G12 mutations were associated with reduced OS benefit of FTD/TPI treatment, with potential implications for ̃28% of patients with metastatic colorectal cancer now considered for treatment with FTD/TPI. Furthermore, our data show that genomics-based precision medicine may be possible for a subset of chemotherapies.