Association of ca27.29 and circulating tumor cells before and at different times after adjuvant chemotherapy in patients with early-stage breast cancer-the success trial

Abstract

Background: Evidence for the prognostic value of circulating tumor cells (CTCs) in early-stage breast cancer is swiftly increasing. An alternative approach for identifying patients at risk for recurrence is based on the detection of the mucin-1 (MUC1)-based tumor marker CA27.29. Here we report the association of these two prognostic markers before and immediately after chemotherapy (CHT), as well as after 2 and 5 years of follow-up. Patients and Methods: The SUCCESS trial compared fluorouracil, epirubicin and cyclophosphamide followed by docetaxel vs. FEC followed by docetaxel plus gemcitabine, and 2 vs. 5 years of treatment with zoledronic acid in 3,754 patients with node-positive or high-risk nodenegative early-stage breast cancer. CA27.29 was measured with the ST AIA-PACK CA27.29 reagent (Tosoh Bioscience, Belgium). The cutoff for CA27.29 positivity was >31 U/ml. CTCs were assessed with the CellSearch System (Veridex, USA). The cutoff for CTC positivity was 1 CTC/15 ml whole blood. The relationship between CTC positivity and CA27.29 positivity was assessed based on Chi-square statistics and Cramer's V, which varies from 0 (no association between the variables) to 1 (complete association). Samples for CA27.29 and CTC determinations during follow-up were only drawn from patients that had no relapse. Results: Both CA27.29 and CTC data were available for 1,981, 1,602, 1,159 and 707 patients before, immediately after and at 2 and 5 years after CHT, respectively. Positivity rates for CTC were 21.3%, 22.8%, 18.6% and 8.5%, respectively. CA27.29 was positive in 7.9%, 21.0%, 2.8%and 7.5%, respectively. Positivity for both CA27.29 and CTC was found in 2.4%, 4.2%, 0.7% and 1.8% of patients, respectively. The association between CA27.29 and CTC was significant but weak before CHT (p=0.0015; Cramer's V=0.063) and 5 years after CHT (p<0.001; Cramer's V=0.164), and not significant immediately after CHT (p=0.162; Cramer's V=0.035) and 2 years after (p=0.349; Cramer's V=0.028). Conclusion: We showed that CTC and CA27.29 positivity were significantly, but only weakly associated before CHT and 5 years after CHT, while no significant association was found immediately or 2 years after CHT during the course of early-stage breast cancer. It, therefore, seems reasonable to further evaluate the prognostic value of CTCs and CA27.29 as a combined prognostic test of two potentially independent markers that might provide complementary prognostic information.