Previous studies have shown that basal-type cytokeratins (CKs) can distinguish usual ductal hyperplasia (UDH) from the spectrum of atypical ductal hyperplasia (ADH), ductal carcinoma in situ (DCIS) and lobular carcinoma in situ (LCIS). Indeed, expression of these CKs is weak or absent in ADH, DCIS and LCIS. However, the diagnostic usefulness of D5/16B4 antibody (anti-CK5/6) has never been compared with that of 34betaE12 antibody (anti-CK1/5/10/14). We performed immunostaining of CK 5/6 and CK1/5/10/14 on 100 breast lesions, including UDH ( n=31), ADH ( n=5), DCIS ( n=54) and LCIS ( n=10). Abundant immunostaining was observed in all UDH using both antibodies. Four of five of the ADH cases showed less than 5% of CK5/6 stained cells, the remaining case showed 30% of labeled cells. With 34betaE12 antibody, three of five of the ADH cases showed less than 5% labeled cells, while two cases showed more than 30% of stained cells. None of the 54 DCIS or the 10 LCIS was labeled by D5/16B4, while a lack of 34betaE12 immunostaining was observed in only 15 of 54 DCIS and 2 of 10 LCIS. We confirmed that D5/16B4 antibody directed against CK5/6 is useful in distinguishing UDH from the spectrum of ADH/DCIS/LCIS. We also demonstrated that D5/16B4 is far a more specific marker than 34betaE12 antibody.