IS A DIAGNOSIS OF INFILTRATING VERSUS IN-SITU DUCTAL CARCINOMA OF THEBREAST POSSIBLE IN FINE-NEEDLE ASPIRATION SPECIMENS

BACKGROUND. With widespread use of mammographic screening, more cases of ductal carcinoma in situ (DCIS) are being detected. Fine-needle asp iration (FNA) is an established and reliable method for diagnosing bre ast carcinoma. However, its usefulness in distinguishing infiltrating ductal carcinoma (IDC) from DCIS is controversial. METHODS, The author s retrospectively reviewed 36 breast FNAs (23 palpable lesions and 13 nonpalpable lesions) that were confirmed histologically as being exclu sively DCIS. On review, 27 cases (75%) were categorized as carcinomas, and 9 cases (25%) were categorized as proliferative lesions with atyp ia. Findings of only these 27 carcinomas were compared with findings o f 42 breast FNAs that surgically were proven to be IDC (greater than o r equal to 85% invasive). Five key cytologic features compared include d the presence of fibroadipose tissue, stromal fragments, benign epith elial groups, angulated tumor cell clusters, and tubular structures of tumor cells; for the first three features, the proximity with tumor c ells also was noted. RESULTS. The presence of stromal fragments associ ated with tumor cells was significantly different between cases of DCI S (9 of 27; 33%) and IDC (29 of 42; 69%) (P = 0.006). Tubular structur es were present only in IDCs (10 of 42; 24%) (P = 0.02). Nine of ten I DCs that had tubular structures also contained stromal fragments. No o ther cytologic features were different statistically between IDC and D CIS on FNA specimens. CONCLUSIONS. Tubular structures of tumor cells a nd the presence of stromal fragments in breast FNA are significant ind icators of stromal invasion. However, the low occurrence rate of tubul ar structures (24% in this series) in IDC and the low specificity of s tromal fragments limit their utility in separating IDC from DCIS. None theless, if present, tubular structures in conjunction with stromal fr agments can be used to evaluate stromal invasion in patients whose dis ease is being managed surgically. (C) 1998 American Cancer Society.