Intensive post-operative follow-up of breast cancer patients with tumour markers: Accuracy of serum MCA-CA15.3 and CEA-TPA-CA15.3 tumour marker panels for early detection of relapse

Background: In breast cancer serum CEA-TPA-CA15.3 tumour marker panel is a cheap tool with high accuracy for the ″early″ detection of relapse (A Nicolini, Br J Cancer 1997); serum MCA and CA15.3 are also reported among the most useful serum tumour markers. However the most suitable cut-off of MCA has not yet been defined. The aims were to compare the MCA-CA15.3 with the CEA-TPA-CA15.3 tumour marker panel and to assess the effect of two different MCA cut-off values on this comparison. Methods: From May 2000 to February 2003, 264 breast cancer patients every 6 or 4 months, according to whether they were at low or high risk of relapse, were monitored with serial serum MCA, CEA, CA15.3, and TPA levels besides history and routine lab examinations. Cut-off value was 4.3 ng/mL (CEA), 32 U/mL (Ca15.3), 95 U/L (TPA), 11 and 16 U/mL (MCA). Tumour marker increase was dynamically evaluated according to a previously described method (A Nicolini, Br J Cancer 2000) Bone scintigraphy, liver echography and chest x-ray were carried out at the beginning of the study then at about 18 month interval. When a relapse was suspected by tumour markers, they were performed immediately. Results: So far 18 (7%) of the 264 patients relapsed. Accuracy of MCA-CA15.3 and CEA-TPA-CA15.3 was 46% and 69% respectively for MCA cut-off 11 U/mL and it was 77% and 71% for MCA cut-off 16 U/mL. For MCA cut-off 16 U/mL sensitivity, specificity and lead time were 58%, 78.5%, 7.4 months (mean; 0-22 range) for MCA-CA15.3 and 74%, 71%, 4.1 months (mean; 0-16 range) for CEA-TPA-CA15.3. Conclusions: 16 U/mL is MCA cut-off value more suitable than 11 U/mL. With MCA cut-off value of 16 U/mL, CEA-TPA-CA15.3 has similar accuracy but much higher sensitivity than MCA-CA15.3.