Determinants of clinical outcomes: A pooled analysis of 71 phase I clinical trials evaluating metastatic colorectal cancer

Background: Minority patients are historically underrepresented in clinical trials. Their clinical outcomes are often reviewed retrospectively after drug approval, leaving uncertainty about efficacy in the broader population, including patients with metastatic colorectal cancer (mCRC). In phase I trials for mCRC, patient selection criteria remain challenging, and objective clinical and laboratory markers may improve uniform screening and outcome prediction. Methods: We retrospectively reviewed medical records of patients with mCRC enrolled in phase I trials at two institutions between 1999-2018 and 2021-2025. Data collected included demographics, clinical and laboratory findings, treatment responses, and overall survival (OS). Variables were dichotomized using institutional or clinically relevant cutoffs. OS was defined from the date of first dose to death or last follow-up (censored if alive). Univariate analyses were performed using the Mantel-Cox test (Prism GraphPad v10), with hazard ratios (HRs) and 95% confidence intervals (CIs) reported. Results: We analyzed 295 patients across 71 trials. Median age was 59 years (range, 29-83 years). 50.5% of patients were female. 24.4% were Non-Hispanic Black (NHB), 24.4% Hispanic, 6.1% Asian, 43.0% Non-Hispanic White (NHW). OS was 9 months and did not differ significantly among the different races/ethnicities (p = 0.71). The median duration on the trial was 59 days (range, 54-65 days). Median number of prior lines of therapy was 3 (range, 0-11), and the median number of sites of metastasis was 3 (range, 0-10). In a univariate model, poor performance status (PS ≥2; HR 2.665, p=0.0026), ≥3 metastatic sites (HR 1.497, p=0.0049), and > 3 prior lines of therapy (HR 1.358, p=0.0167) were associated with decreased OS. Laboratory markers including low albumin (<3.9 g/dL; HR 1.596, p=0.0002), high LDH (>281.5 U/L; HR 1.751, p<0.0001), high CEA (>119.5 ng/mL; HR 1.513, p=0.0031), elevated liver enzymes (AST>40 U/L, HR 1.696, p=0.0004; ALP>150 U/L, HR 2.102, p<0.0001), elevated bilirubin (≥1 mg/dL; HR 1.852, p=0.0204), and cytopenias (WBC<6 K/µL, HR 0.66, p=0.0011; ANC<4 K/µL, HR 0.63, p=0.0004; hemoglobin <11.5 g/dL, HR 1.435, p=0.0043) also predicted worse OS. Conclusion: Patients with mCRC, regardless of age, sex, or race/ethnicity, had similar outcomes in phase I studies. The OS was at par with published data with similar exposure to prior therapies. Heavily pre-treated patients, low performance status, and spread to >3 sites were all associated with decreased survival. High LDH and high CEA, along with liver function tests, are key parameters and may refine eligibility and stratification in early-phase mCRC studies. A multivariate analysis will be presented, looking into the interplay of these key clinical findings.