Does the benefit of sentinel node frozen section vary between patients with invasive duct, invasive lobular, and favorable histologic subtypes of breast cancer?

Background: Although many questions regarding sentinel lymph node (SLN) biopsy in breast cancer have been answered by observational studies and, increasingly, by prospective trials, the role of intraoperative SLN assessment remains a matter of debate. Here we report in detail the results of intraoperative SLN assessment by frozen section (FS), with particular attention to variations in sensitivity and yield by histologic subtype, by tumor size, and by other clinicopathologic parameters. Methods: Five thousand two hundred ninety-eight consecutive patients with clinical stage T1-3N0 invasive breast carcinoma had SLN biopsy with intraoperative FS at Memorial Sloan Kettering Cancer Center between 1996 and 2004. We report the results of FS by sensitivity (the proportion of all positive SLN detected by FS) and by yield (the proportion of all FS procedures in which the FS was positive). Results: The sensitivity of FS was 61% overall, was higher for invasive duct (ID) than for invasive lobular (IL) cancers (62% vs. 52%; P = 0.006), and was marginally lower for favorable subtypes (46%; P = 0.26). The yield of FS was 21% overall, with no difference between ID and IL cancers (22% vs. 21 %; P = 0.49), and with a substantially lower yield for favorable subtypes (3%; P < 0.001). The yield of FS increased with tumor size for ID and IL cancers (P < 0.001), but not for favorable subtypes. For both ID and IL cancers, the sensitivity and yield of FS were significantly higher with younger patient age, increasing tumor size, and lymphovascular invasion. The yield of FS was <10% for all patients with ID or IL tumors <= 1 cm in size who were older than 60 years of age. Among all FS-positive patients, only 45% were identified by the first FS, whereas 91% were cumulatively identified by the first, second, or third FS. Conclusions: For patients with ID and IL cancers, the overall sensitivity of FS is >50%, but the yield of FS is <10% for individuals >= 60 years of age with T1a/b tumors. Intraoperative FS may not be worthwhile for this low-yield subset, especially for patients with invasive breast cancer of favorable type.