Analysis of response rate (RR), liver resection (LR), and toxicity (T) in 129 advanced colorectal cancer (CRC) patients (pts) treated with chronomodulated infusion of ironotecan (I), 5-fluorouracil/leucovorin (F/L), and oxaliplatin (O) (CHRONO-IFLO) with or without cetuximab (Cmab).

Abstract

e14099 Background: Advanced CRC pts received chrono-IFLO in three sequential studies: (A) EORTC 05011: 54 pts (Garufi et ASCO 2007); (B) selected liver metastatic 43 pts (Garufi et al, POCHER study, BJC 2010) and (C) unselected CRC 32 pts (Garufi et al ASCO 2010). Cmab was added in groups B+C. Methods: Aim of the study was to retrospectively analyze activity, tolerability and outcomes. Results: Median (m) age 60 y (range 29-78), M/F:78/51 (60.5/39.5%), PS 0: 112 (87%) pts, 1: 16 pts, 2: 1 pt; synchro vs metachronous (s/m) metastases 105/24 (81/19%), 110 pts were treated as first with only 19 as second-line (15%). Groups A and B+C did not differ for m-age, sex, PS, s/m . RR was obtained in 71/129 pts (55%) while 21 pts progressed; LR was obtained in 52/129 pts (40.3%). G3-4 gastrointestinal T affected 67 pts (52%); dose-intensity (DI) >80% at c5: 63.4% of pts. m-Follow-up was 19.6 months (ms) (1-100), mPFS was 14 ms (CI95% 12-16), with 1-2-3 y PFS of 55%-23.8%-13.0, respectively; mOS was 31 ms (25-37 CI 95%), with 1-2-3 y OS of 84.1%-60.2%-36.8% respectively. PFS at 1-2 y was significantly longer in LR 77.5-37.9% vs non-LR pts 40.1-13.6% (p=0.0001). mPFS and mOS did not differ between pts treated ± Cmab, also for LR pts. There was neither difference in m-DI for I, F/L or O, nor 20%-dose-reduction due to T criteria in M/F in the entire population, in subgroups according to ± Cmab, or in group A alone but F-pts treated with Cmab experienced significant more T than M-pts in B+C, 32.3 vs 10.0% (p=0.002). In groups B+C vs A there was a RR of 64.0% vs 40.7% (p=0.009), LR 48.0% vs 29.6% (p=0.03), but Gr 3-4 GI-T was 57.3% vs 44.4.1% (p=0.15), There was no relationship between RR or LR with Gr 3-4 GI toxicity, m-DI, or 20% dose reduction in the entire population and in group A vs B+C. Predictive factors for response were PS 0 vs 1-2, first vs second line and Groups B+C vs A. RR was the only significant predictor factor for LR. At Cox multivariate analysis LR predicted PFS with HR 2.65 (1.68-4.19). Conclusions: LR is the most relevant prognostic factor for CRC pts treated with chrono-IFLO ± Cmab. LR were increased with Cmab.