Gemcitabine(G)/Erlotinib(E) Versus Gemcitabine/Erlotinib/Capecitabine(C) In The First-Line Treatment Of Patients With Metastatic Pancreatic Cancer (Mpc): Efficacy And Safety Results Of A Phase Ilb Randomized Study From The Spanish Ttd Collaborative Group.

Introduction: G and E have shown a survival benefit in the 1st-line setting in mPC. The aim of this study was to assess whether combining C with G + E was safe and effective as compared with G-E in patients (pts) with mPC. Methods: Previously untreated pts with mPC were randomized in a 1:1 ratio to receive G(1000 mg/m2, d1,8,15) + E(100 mg, d1–28) + C(1660 mg/m2, d1-21) (GEC arm) or G + E (GE arm), q4-wk, until progression or unacceptable toxicity. Primary endpoint was progression-free survival (PFS); secondary endpoints included overall survival (OS), response rate (RR), relationship of rash with PFS/OS, and safety. A sample size of 118 pts was required for a hazard ratio (HR) of 0.63, assuming a median PFS of 6 months (m) in pts treated with GEC; unilateral &agr; = 0.05; &bgr; = 0.8. Results: 120 pts were randomized: 60 GEC and 60 GE. No significant differences in demographic characteristics: median age: 63 years (range 29-78); ECOG status 0/1/2 (%), 33/58/8. Median follow-up was 16.5 m (95%CI: 13.7-26.2). Efficacy results are presented in the Table. There were not statistically significant differences in RR, tumor control rates (TCR: RR + stable disease), PFS and OS between the 2 arms. Analysis of safety shows that tolerability was acceptable in the 2 arms. Neutropenia grade 3/4 (GEC 43% vs GE 15%; p = 0.0008) and mucositis (GEC 9% vs GE 0%; p = 0.03), were the only statistically significant differences in grade 3/4 adverse events. PFS and OS were significantly longer in pts with rash (grade ≥1) vs no rash (grade = 0): PFS 5.5 vs 2.0 m; HR = 0.39 95%CI: 0.26-0.6; p < 0.0001 and OS: 9.5 vs 4.0 m; HR = 0.51 95%CI: 0.33-0.77; p = 0.0014). Conclusion: The results from this randomize phase II multicenter study indicate that PFS with GEC was not different to that with GE as it did not meet the criterion for statistical significance. Skin rash strongly predicted erlotinib efficacy, deserving further investigation for patient selection.