OBJECTIVETo analyze the relationship between primary tumor location and clinical response of chemotherapy in patients with metastatic colorectal cancer(mCRC).

METHODSClinical data of 721 mCRC patients who received first-line and second-line chemotherapy in Peking University Cancer Hospital between January 1996 and December 2011 were collected. All the patients were divided into 5 groups according to primary tumor location: ileocecum in 61 patients(8.5%), ascending colon or hepatic flexure in 126 patients (17.5%), transverse colon or splenic flexure in 26 patients (3.6%), descending or sigmoid colon in 172 patients (23.9%), rectum in 336 patients (46.6%). Outcomes of chemotherapy were evaluated by Response Evaluation Criteria in Solid Tumors (RECIST, version 1.1), including complete response (CR), partial response (PR), stable disease (SD) and progressive disease (PD). The overall response rate (ORR) was counted with the total number of patients divided by the number of CR+PR. Differences in first-line and second-line chemotherapy efficacy among different primary tumor sites in metastatic colorectal cancer were compared by using Chi-square test.

RESULTSOf the 571 patients receiving first-line chemotherapy, no one patient was classified as CR, while there were 190 as PR (33.3%), 277 as SD (48.5%) and 104 as PD (18.2%), with ORR 33.3% (190/571). The ORRs of patients with primary tumor located at ileocecum, ascending colon or hepatic flexure, transverse colon or splenic flexure, descending or sigmoid colon, rectum were 21.3% (10/47), 35.3% (36/102), 14.3% (3/21), 41.3% (57/138) and 31.9% (84/263), respectively, with statistically significant difference(P = 0.028). Difference of oxaliplatin-based first-line chemotherapy efficacy among different tumor sites was statistically significant(P = 0.009), while differences in irinotecan-based or single-agent 5-fluorouracil chemotherapy efficacy were not statistically significant (all P>0.05). In patients with primary tumor located at transverse colon or splenic flexure, irinotecan-based first-line chemotherapy had higher ORR than oxaliplatin-based or single-agent 5-fluorouracil chemotherapy, and the difference was statistically significant (P=0.042). There was no significant difference in the efficacy of different first-line chemotherapy regimens in patients with primary tumor located at other sites (all P>0.05). Of the 353 patients receiving second-line chemotherapy, no one patient was classified as CR, while there were 43 as PR (12.2%), 187 as SD (53.0%) and 123 as PD (34.8%), with ORR 12.2%(43/353). The ORRs of patients with primary tumor located at the ileocecum, the ascending colon or the hepatic flexure, the transverse colon or the splenic flexure, the descending or sigmoid colon, the rectum were 4.2%(1/24), 12.1%(8/66), 8.3%(1/12), 15.2%(12/79) and 12.3%(21/171) respectively, without statistically significant difference (P=0.686). Differences in second-line chemotherapy efficacy with the same regimen among different tumor sites were not statistically significant, and there were also no significant differences of efficacy of different second-line chemotherapy regimens in patients with the same tumor site (all P>0.05).

CONCLUSIONThere are differences in first-line chemotherapy efficacy among different primary tumor sites in metastatic colorectal cancer, while their second-line chemotherapy efficacy is equivalent.