Background: Cancer stem cells are subpopulation of cancer cells characterized by the self-renewal ability, and primarily maintained by a network of pluripotency transcription factors (PTFs). Despite their chemo-resistant phenotype, recent clinical trials data show cetuximab (Cet) is more beneficial for a subset of metastatic colorectal cancer (mCRC) categorized in a mesenchymal/stem-like subtype based on transcriptomic classifications. We investigated whether single nucleotide polymorphisms (SNPs) in PTF encoding genes have predictive values for the efficacy of Cet in mCRC patients. Methods: Genomic and clinical data from three independent cohorts of patient treated with first-line chemotherapy (n = 451, in total) were tested: Cet cohort (FOLFIRI+Cet arm in FIRE-3 trial, n = 129); bevacizumab (Bev) cohort 1 (FOLFIRI+Bev arm in FIRE-3 trial, n = 107) and Bev cohort 2 (FOLFIRI+Bev arm in TRIBE trial, n = 215), as controls. Genomic DNA extracted from blood samples was genotyped using an OncoArray (Illumina, Inc., San Diego, CA, USA). Eight SNPs in PTF encoding genes (NANOG rs11055786, NANOG rs11055767, NANOGP8 rs2168958, NANOGP8 rs9944179, POU5F1 rs3130501, POU5F1 rs3130932, SOX2 rs11915160, and MYC rs3891248) were tested for association with progression-free survival (PFS) and overall survival (OS), using Cox proportional hazards model. Results: In the Cet cohort, three SNPs were significantly associated with PFS in univariate analysis:NANOG rs11055767 (C/C vs any A allele, hazard ratio [HR] = 0.62, 95% confidence interval [CI] = 0.42–0.94, log-rank p = 0.02), NANOGP8 rs2168958 (A/A vs any C allele, HR = 2.12, 95% CI = 1.36–3.29, log-rank p< 0.01), and NANOGP8 rs9944179 (G/G vs any A allele, HR = 3.68, 95% CI = 1.46–9.31, log-rank p< 0.01). Multivariate analysis confirmed the significance in NANOGP8 rs2168958 and NANOGP8 rs9944179. Furthermore, two SNPs were significantly associated with OS in multivariate analysis:POU5F1 rs313051 (G/G vs any A allele, HR = 0.46, 95% CI = 0.22–0.99, adjusted p = 0.04) and POU5F1 rs3130932 (A/A vs any C allele, HR = 0.46, 95% CI = 0.22–0.93, adjusted p = 0.03). Both in the Bev cohorts 1 and 2, no significant associations between the SNPs and clinical outcomes were observed. Conclusions: The current findings suggest that polymorphisms in the PTF genes could be predictive markers for Cet in patients with mCRC. Further studies are warranted to validate the predictive utility.