Abstract
BACKGROUND: Characteristics of patients significantly differ between registrational clinical trials (CTs) and Italian real-world practice, with older median age, higher elderly (≥ 65) rate and worse performance status (PS) in the latter, without imbalance in female rate. We compared the same characteristics between registrational phase 2/3 and phase 1 CTs. METHODS: Data on age, sex and PS were extracted from European Public Assessment Reports of European Medicines Agency. Weighted means and standard deviations were calculated in both groups and differences were described overall, by cancer type and drug class. RESULTS: We collected 103 phase 2/3 and 111 phase 1 CTs, supporting 97 therapeutic indications. Age and sex were compared in 59 indications. Mean median age (SD) was 60.7 (5.1) years in phase 2/3 and 59.7 (5.6) years in phase 1 (p = 0.051). Age difference was greater for skin and breast cancer; no heterogeneity emerged among drug classes. Mean female rate was not statistically significantly lower in phase 2/3 than phase 1 CTs overall, (mean difference - 4.9%, p = 0.999); difference was greater for skin and upper-gastrointestinal cancers and for cytotoxic agents. Mean PS > 1 rate, compared in 47 indications, was similar in phase 2/3 [2.3% (4.7)] and phase 1 [1.8% (3.5)] (p = 0.374); difference was greater for colorectal cancer and cytotoxic agents. CONCLUSIONS: We found no statistically significant difference in age, sex and PS between patients in phase 2/3 and corresponding phase 1 CTs for anticancer treatments. Therefore, patient selection in phase 1 trials appears crucial, considering its potential impact in later development phases.