Impact of adjuvant chemotherapy and patient age on patterns of metastatic recurrence in stage III colorectal cancer.

Abstract

122 Background: The patterns of metastatic relapse for Stage III colorectal cancer (CRC) are known to be influenced by the primary site and mutation status. Whether the use of adjuvant therapy also influences recurrence patterns, due to a variable chemotherapy impact by metastatic site, has yet to be explored. Patient age 70 years has consistently been associated with a lack of benefit from the addition of oxaliplatin to a fluoropyrimidine (FP) chemotherapy. Methods: Data from a multi-site CRC database that prospectively captures, patient, tumour, treatment and outcome data was examined. Patients with de novo metastatic or multiple primary tumours were excluded. Patterns of recurrence were analysed by receipt of surgery and adjuvant chemotherapy (S + AC) or surgery alone (S alone) and by age 70 years and <70 years old. Results: Of 701 patients with stage III colon cancer 397 were male, median age was 68 years (range 33-83) and 574 (81%) were ECOG 0-1. S + AC was administered in 555 (79%), including 228 treated with a FP alone and 324 treated with an oxaliplatin doublet. Metastatic relapse was delayed in S + AC patients, occurring at a median of 17.81 vs 9.43 months (p< 001). Liver metastases were the most common site of relapse overall (40%) and numerically less common in the S + AC group (38% vs 46%, p = 0.06). Brain metastases were less common in S + AC patients (1% vs 3.5%, p = 0.04). When analysed by primary tumour side the only significant finding was that for patients with a right-side primary (n= 208, 76% had S + AC), S+ AC was associated with a lower rate of lung metastases (28% vs 41% p = 0.046). 318 of the 701 (45%) of this relapsed cohort were 70 years old, 220 (69%) of these patients had S + AC. In both the older and younger cohort metastatic relapse was delayed with S + AC (18.7 vs 11.3 months, p < 0.001, 17.08 vs. 6.01 months, p < 0.001) and liver metastases were the most common site of relapse (41% vs 38%, p=0.56). Similar rates of lung metastases (40% vs 34%, p=0.1), peritoneal disease (20% vs 22%, p=0.57) and brain metastases (2.2% vs 0.8%, p=0.13) were seen in the 70 years old and <70 years old patients, with no evident interaction between chemotherapy use and age with respect to recurrence sites. Conclusions: S + AC for stage III colon cancer in this community series delayed time to recurrence and resulted in numerically lower recurrence rates at all disease sites. Brain metastases although uncommon were less frequent in S + AC patients. Patients with a right sided primary had a lower incidence of lung metastases following S + AC patient cohort. There was no significant impact of age on recurrence patterns.