P-73 the Role of Site of Metastatic Resection in Metastatic Colorectal Cancer Patients: A Mono-Institutional Cohort Study

Approximately 50%-60% of colorectal cancer (CRC) patients (pts) develop metastases, usually to liver and lung. When feasible, metastatic resection (MR) is the only potentially curative option in the multimodal management of mCRC pts. Few studies have compared survival outcomes based on different surgical sites with contrasting data. Hence, we retrospectively analyzed mCRC pts who underwent radical lung and/or liver resection at our institution, investigating the impact of resection site on overall survival (OS). mCRC pts who underwent radical liver (group 1), lung (group 2) or liver and lung (group 3) resection were included. The following variables were collected: age (> vs ≤ 65 years); gender (male vs female); primary tumor site (right vs left); synchronous vs metachronous; RAS/BRAF status; number (N) of MR (1, 2 or ≥3); chemotherapy treatment (No treatment vs Post-operative vs Peri-operative/pre-operative treatment) and chemotherapy regimen (5FU monotherapy, oxaliplatin-based, irinotecan-based regimen, FOLFOXIRI, bevacizumab, anti-EGFR). The association between MR site and OS was evaluated. Univariate and multivariate analyses for OS were performed. A total of 191 mCRC pts who underwent radical MR were included in the analysis: 112 (59%) pts in group 1, 38 pts (20 %) in group 2, 41 pts (21 %) in group 3. 145 (76%) pts had a left-sided tumor and 46 (24%) a right-sided tumor. Out of 156 evaluable pts, 73 (47%) pts harbored a RAS mutation, while out of 136 evaluable pts, 4 (3%) pts had a BRAF mutation. Regarding the N of MR, 125 pts (65%) underwent 1 radical MR, 43 (23%) pts 2 MR and 23 (12%) ≥3 MR. In the overall population, median OS was 77.2 months. According to MR site, median OS was 59.4, not reached (NR) and 99.1 months, in group 1, 2 and 3, respectively (p=0.075). At the multivariate analysis, no significant association with OS was shown for MR site, while the N of MR and RAS status were independently associated with OS. Median OS was 58.5, 97.7 months and NR in pts underwent 1, 2 and ≥3 MR, respectively (p=0.02). Median OS was 58.5 and 83.1 months in RAS mutated and RAS wild-type pts, respectively (p=0.12). Despite the limited number of pts and the retrospective nature of our study, these results confirmed that surgery represents the only option with curative intent for mCRC pts, independently of metastatic site (liver vs lung vs liver and lung). Based on our analysis, a higher number of MR is associated with long-term survival, and this could be explained with an accurate selection of patients that could benefit from multiple radical resections. Thus, a multidisciplinary approach is essential for the management of mCRC pts and surgery should be evaluated case by case and always performed when possible, even several times, independently of site of MR.