Medullary carcinoma of the breast: a clinicopathologic study with appraisal of current diagnostic criteria.

Abstract

Fifty-three cases of mammary carcinoma originally diagnosed as medullary carcinoma (MC) or infiltrating duct carcinoma (IDC) with medullary features were reviewed and reclassified using the strictly defined histologic criteria applied a decade ago by Ridolfi et al. Our study interval (1961 to 1982) allowed for a minimum follow-up of 5 years for each patient, with a mean follow-up period of 7.2 years. When reclassified, 24 tumors fulfilled the criteria for MC, 16 tumors were determined to be atypical MC, and ten tumors were found to be IDC; the observed 5-year survival rates were 95%, 80%, and 70%, respectively. These findings confirmed those of other investigators, that when specific criteria are applied, MC proves to be a form of mammary carcinoma with a favorable prognosis. However, we also found that when tumors were excluded from the MC category solely on the basis of in situ carcinoma, focal marginal infiltration, or a sparse mononuclear infiltrate, the survival rate of these patients was similar to that of patients in the medullary category. Thus, we propose that one of these criteria alone should not suffice to exclude the diagnosis of MC. On the other hand, tumors with two or more of these atypical features, or with extensive marginal infiltration, no mononuclear cellular infiltrate, and/or less than 75% syncytial growth, should be classified as IDC with medullary features. Typical MC with bland nuclei or a focal microglandular growth pattern only were not observed in this series; however, these findings should probably also cause a tumor to be classified in the IDC category. By dividing our cases into two rather than three groups, we found a statistically significant difference between the survival rates of 94% and 64% for MC (34 tumors) and IDC (14 tumors), respectively. Although the latter figure probably exceeds the survival rate for IDC without medullary features, the difference does not appear great enough to warrant a separate diagnostic category.