Biomarkers for local recurrence after breast-conservation - a nested case-control study

Purpose: A previous cohort study of 759 women with invasive T1-T2 breast ca ncer operated on with breast-conserving surgery in Stockholm between 1976 a nd 1986 indicated that age < 50 years, no postoperative irradiation, and no dal involvement were independent risk factors for ipsilateral breast tumor recurrences (IBTR). The aim of the current study was to analyse if selected biological markers assayed in tumor specimens from these patients could ad d prognostic information, thereby helping to identify groups of patients at high versus low risk of IBTR. Methods: The study was designed as a case-control study 'nested' within the cohort. The cohort was stratified according to nodal status and the use of postoperative irradiation. In these four strata, the cases were those 80 w omen who developed IBTR between 1977 and 1994. In each stratum, women witho ut IBTR were randomly selected as controls (n=159). Median time at risk was 12 (8-18) years. The following factors were analysed: histopathological tu mor grade according to Elston-Ellis, DNA ploidy, immunohistochemical staini ng for apoptosis, angiogenesis, Ki-67 (MIB-1), c-erbB-2, p-53, waf-1, and b cl-2. The prognostic role of each factor was assessed using linear logistic regression methods. Results: In univariate analyses only age < 50 years was identified as a sig nificant risk factor for IBTR, whereas none of the studied biomarkers yield ed statistically significant information. However, in a multivariate model, age, MIB-1-index, and tumor grade significantly influenced the risk for IB TR: the odds-ratio (OR) for age greater than or equal to 50 years was 0.4, 95% confidence interval (CI) = 0.2-0.9; for medium or high grade tumors it was 0.4 (CI = 09-0.9); and for MIB-1-index > 30%, 2.1 (CI = 1.0-4.4). In wo men greater than or equal to 50 years, MIB-1-index > 30% was associated wit h an OR of 3.5 (CI = 1.4-8.8) compared to those who were younger. Patients greater than or equal to 50 years with MIB-1-index less than or equal to 30 % were thus identified as a low-risk group with an OR of 0.2 (CI = 0.1-0.5) . A possible high-risk group was patients < 50 years with tumors showing a combination of c-erbB-2 and waf-1 immunoreactivity, with an OR of 6.7 (CI = 1.3-34.7). Conclusion: Women greater than or equal to 50 years with MIB-1-index less t han or equal to 30% constituted a subgroup with a low risk of IBTR. This ob servation raises the issue whether this group of patients might be spared p ostoperative irradiation following breast-conserving surgery. However, due to the methodology of the study, including the large number of comparisions , the presented results warrant cautious interpretation and should be regar ded as tentative.