The correlation of cone biopsy with findings at radical hysterectomy and the need for adjuvant radiation therapy.

Abstract

5122 Background: Our objective is to determine if pathologic findings in cone biopsy specimens correlate with residual invasion in radical hysterectomy (RH) specimens and the need for adjuvant radiation therapy (XRT). Methods: We identified 64 patients who underwent a cone biopsy and subsequent RH for the treatment of stage I cervical cancer from 2000-2009. Lymph-vascular space invasion (LVSI), margin status, endocervical curettage (ECC) status, and depth of invasion (< 3 mm or ≥ 3 mm) on the cone biopsy specimens were correlated with the presence of residual invasion in the RH specimens and the need for adjuvant XRT. Univariate comparisons were performed using chi-square analysis. Logistic regression was used to develop predictive models and determine independent predictors of residual invasion and adjuvant XRT. Results: Chi-square analysis revealed that positive deep margin (p < 0.009), ≥ 3 mm of invasion (p < 0.003), and positive endocervical margin (p < 0.001) significantly correlated with the presence of residual disease in the RH specimens. Logistic regression revealed that only ≥ 3 mm of invasion (p < 0.018) correlated independently with residual invasion, while positive ECC (p < 0.052) and positive endocervical margin (p < 0.053) approached significance. Positive ECC (p < 0.001), positive LVSI (p < 0.001), and positive deep margin (p < 0.003) significantly correlated with the need for adjuvant XRT by both chi-square and logistic regression analyses. A model using positive ECC, positive endocervical margin, and ≥ 3 mm of invasion on the cone specimen correlated with residual invasion (r2 = 0.456) and had a positive predictive value of 86.7%. A model using positive LVSI, positive ECC, and positive deep cone margin correlated with the need for adjuvant XRT (r2 = 0.703) and had a positive predictive value of 91.1%. Conclusions: Patients with ≥ 3 mm of invasion, positive ECC, and positive endocervical margin may be at risk for residual invasion in the RH specimen. Patients with positive LVSI, positive ECC, and positive deep cone margin may require adjuvant XRT. These factors should be considered during primary treatment planning in order to avoid the morbidity of multimodal therapy. No significant financial relationships to disclose.