The extent of agreement between frozen and paraffin block data from axillary samples in patients with early-stage breast cancer

Introduction: Patients with early-stage breast cancer currently undergo sentinel lymph node dissection to evaluate the axillary region. Frozen tissue blocks are evaluated intra operatively and paraffin-embedded samples are studied postoperatively. We explored whether sentinel lymph node dissection adequately reflected axillary involvement (as revealed by the paraffin blocks) in patients with early-stage breast cancer; we sought to avoid axillary dissection.Methods: The agreement/non-agreement rates between the results of axillary ultrasonography and biopsy, sentinel lymph node and axillary dissections, and frozen and paraffin block results, were retrospectively analyzed for 200 patients with early-stage breast cancer. The positive predictive values and accuracies were recorded in those who were positive on both ultrasonography and biopsy. The negative predictive values were calculated for doubly negative cases. Results: The frozen and paraffin block results disagreed in 19 (9.5 %) cases and agreed in 181 (90.5 %). The frozen block and dissection results differed in five of 38 patients who underwent axillary dissection (AD) (one patient did not undergo AD); the results were in agreement in 32. Of the 19 block-disagreement cases, 16 were in the non-neoadjuvant chemotherapy (NAC) group and three in the NAC treatment group. Clinically, the negative predictive values of the frozen and paraffin block data were 80 % in patients lacking axillary involvement.Conclusion: Paraffin block evaluations only (thus, without frozen block examinations) of early-stage breast cancer lymph nodes seem to be sufficient to guide treatment. Also, a thorough clinical examination (with ultrasonog-raphy and axillary biopsy) reduces the dissection rate and the associated functional impairments.