Effect of FOLFOX on minimal residual disease in Stage III colon cancer and risk of relapse

Abstract

Introduction: 25% of Stage III colon cancer patients relapse within 5 years due to minimal residual disease (MRD) not eliminated by surgery and chemotherapy. We hypothesise that sub-types of MRD, defined by circulating tumour cells (CTCs) and bone marrow micro-metastasis (mM) have different types and kinetics of relapse. Patients and Methods: One month of curative surgery and 1 month after completing six cycles of FOLFOX chemotherapy blood and bone marrow samples were taken to detect CTCs and mM using immunocytochemistry with anti-carcino-embryonic antigen (CEA). Follow up was up to 5 years or disease progression defined as new images on CT scanning. Survival curves using Kaplan-Meier (KM) and Restricted Mean Survival Time (RMST) were calculated for three prognostic groups: CTC and mM negative, CTC negative mM positive, and CTC positive. Results: 76 patients (39 men) participated, mean age 67 years, median follow-up 3.6 years. The response to chemotherapy was heterogeneous and MRD pre-treatment did not predict response to therapy. Of 21 patients MRD (-), 20 remained MRD negative and one patient became mM (+); of 21 patients mM (+), 10 became MRD (-), 8 remained the same and 3 became CTC (+); of the 34 CTC positive, 8 became MRD (-), 8 with only mM, and 18 remained positive. After chemotherapy, 38 patients were negative for CTC and mM, 17 were positive for only mM, and 21 for CTCs. For the whole cohort, the 5 year KM was 58%, the median survival was not reached. For the three prognostic groups, the KM 5-year survivals were 87%, 58%, and 4%, respectively, the median survival for patients MRD negative and mM only was not reached. RMST for the whole cohort was 3.6 years, for the three prognostic groups the RMST was 4.6 years, 4.0 years, and 1.5 years, respectively. Serum CEA was significantly higher pre-surgery in the CTC positive group. There were no significant differences with respect to age or sex between the three groups. Conclusions: MRD subtypes pre-chemotherapy did not predict treatment response. Post-chemotherapy MRD subtypes were associated with the pattern of failure and time to failure. MRD negative patients had an excellent prognosis with 87% disease-free survival at 5 years. Those with only mM had a similar outcome up to 2 years and then were at increasing risk of late failure. Patients who were CTC positive had a high risk of early failure. MRD subclassification may be useful to define the risk of relapse in Stage III colon cancer patients and warrants further studies with a larger number of patients.

Introduction: 25% of Stage III colon cancer patients relapse within 5 years due to minimal residual disease (MRD) not eliminated by surgery and chemotherapy. We hypothesise that sub-types of MRD, defined by circulating tumour cells (CTCs) and bone marrow micro-metastasis (mM) have different types and kinetics of relapse. Patients and Methods: One month of curative surgery and 1 month after completing six cycles of FOLFOX chemotherapy blood and bone marrow samples were taken to detect CTCs and mM using immunocytochemistry with anti-carcino-embryonic antigen (CEA). Follow up was up to 5 years or disease progression defined as new images on CT scanning. Survival curves using Kaplan-Meier (KM) and Restricted Mean Survival Time (RMST) were calculated for three prognostic groups: CTC and mM negative, CTC negative mM positive, and CTC positive. Results: 76 patients (39 men) participated, mean age 67 years, median follow-up 3.6 years. The response to chemotherapy was heterogeneous and MRD pre-treatment did not predict response to therapy. Of 21 patients MRD (-), 20 remained MRD negative and one patient became mM (+); of 21 patients mM (+), 10 became MRD (-), 8 remained the same and 3 became CTC (+); of the 34 CTC positive, 8 became MRD (-), 8 with only mM, and 18 remained positive. After chemotherapy, 38 patients were negative for CTC and mM, 17 were positive for only mM, and 21 for CTCs. For the whole cohort, the 5 year KM was 58%, the median survival was not reached. For the three prognostic groups, the KM 5-year survivals were 87%, 58%, and 4%, respectively, the median survival for patients MRD negative and mM only was not reached. RMST for the whole cohort was 3.6 years, for the three prognostic groups the RMST was 4.6 years, 4.0 years, and 1.5 years, respectively. Serum CEA was significantly higher pre-surgery in the CTC positive group. There were no significant differences with respect to age or sex between the three groups. Conclusions: MRD subtypes pre-chemotherapy did not predict treatment response. Post-chemotherapy MRD subtypes were associated with the pattern of failure and time to failure. MRD negative patients had an excellent prognosis with 87% disease-free survival at 5 years. Those with only mM had a similar outcome up to 2 years and then were at increasing risk of late failure. Patients who were CTC positive had a high risk of early failure. MRD subclassification may be useful to define the risk of relapse in Stage III colon cancer patients and warrants further studies with a larger number of patients.