Risk predictors of underestimation and the need for sentinel node biopsy in patients diagnosed with ductal carcinoma in situ by preoperative needle biopsy

Background Diagnosis of ductal carcinoma in situ (DCIS) by core needle biopsy showed a high rate of underestimation of invasiveness, and performing sentinel lymph node biopsy (SLNB) in DCIS patients was controversial. Methods We analyzed 340 DCIS patients who were diagnosed by needle biopsies. Final pathology and clinicopathological features were reviewed. Predictors were accessed using the Chi-square test and a binary logistic regression model. Results The overall DCIS underestimation rate was 42.6%. The underestimation was significantly related to the palpability, mass or calcification by ultrasonography, grade, suspicious microinvasion, and biopsy method in univariate analysis. In multivariate analysis, palpability, ultrasonographic calcification or mass, suspicious microinvasion, and core needle biopsy were independent predictors of underestimation of invasive cancer. In cases with one or no risk predictors, the underestimation rate was 14.3%, whereas, in those with five predictors, it increased to 90.9%. Among 144 invasive cancer patients who underwent axillary staging, 15.4% had node metastasis. Conclusions DCIS diagnosed by preoperative needle biopsy has a high probability of underestimation, and 15% of invasive cancer patients have node metastasis. SLNB may be justified in DCIS patients undergoing needle biopsies, and caution should be exercised in omitting SLNB in patients with one or no risk predictors.