The influence of age and extensive intraductal component histology upon breast lumpectomy margin assessment as a predictor of residual tumor

Purpose: Young age and extensive intraductal component (EIC) histology have been shown to be associated with increased local recurrence in women treat ed with breast conservation therapy. This study was conducted to determine if the status of the lumpectomy specimen margin consistently predicted for residual tumor burden risk irrespective of these variables. Methods and Materials: As part of an institutional prospective approach for breast conservation therapy (BCT), 265 cases with AJCC Stage I/II carcinom a with an initial excision margin that was less than or equal to 2 mm or in determinate were subjected to reexcision. The probability of residual tumor (+RE) was evaluated with respect to tumor size, histopathologic subtype (i nvasive ductal carcinoma, invasive ductal carcinoma with an EIC, and invasi ve lobular carcinoma), relative closeness of the measured margin, and the e xtent of margin positivity graded as focal, minimal, moderate, or extensive . The amount of residual tumor was graded as microscopic, small, medium, or large. All variables were analyzed for patient age less than or equal to 4 5 or >45 years. Results: There was no significant difference in the incidence of a +RE acco rding to age less than or equal to 45 versus >45 years when the margin was >0 less than or equal to 2 mm. Of the patients aged less than or equal to 4 5 years, the incidence of a +RE with a margin that was positive as compared to >0 less than or equal to 2 mm was 71% vs. 23%, respectively (p = 0.002) . for women >45 years old, the difference in the incidence of +RE comparing margins that were positive or >0 less than or equal to 2 mm was not signif icant at 50% vs. 40%, respectively (p = 0.23). For all cases in aggregate, age less than or equal to 45 years was associated with a greater incidence of +RE as compared to patients aged >45 Sears with the discrepant incidence of a +RE by age strata most pronounced for focally positive margins (60% v s. 18%; p less than or equal to 0.05). In a logistic regression analysis, a ge (per year, as a continuous variable) and an EIC histology were significa ntly associated with the probability of a +RE (odds ratio [OR] = 0.80,p = 0 .05 and OR = 1.9, p = 0.01, respectively). Tumor size was not significant ( p = 0.23). In patients with an EIC histology, margin status is generally le ss predictive for differences in the incidence of a +RE. further, the overa ll magnitude of difference in the incidence of a +RE related to age appears to be minimized when an EIC histology is present. In contrast, for cases c lassified as having non-ETC histology, there is a near-linear relationship for both age strata with respect to margin status and the incidence of a +R E. When histology is classified as non-ETC, age less than or equal to 45 ye ars is consistently associated with a greater risk of residual tumor for al l margin status categories. When the extent of margin positivity was graded as focal or minimal, residual tumor was semiquantitatively estimated as a medium/large amount in 33% versus 26% of cases aged less than or equal to 4 5 or >45 years, respectively (p = 0.62). Conclusion: For positive lumpectomy specimen margins, younger age is associ ated with an increased residual tumor risk. An EIC histology appears to be associated,vith an elevated risk of residual tumor irrespective of age and may undermine the predictive utility of margin status. Therefore, age and a n EIC histology should be factored into risk assessments for residual tumor that rely upon margin assessment. (C) 1999 Elsevier Science Inc.