Objective: To compare stages IB1 and IB2 cervical cancers treated with radical hysterectomy (RH) and to define predictors of nodal status and recurrence.
Methods: Patients with stage IB cervical cancers undergoing RH between 1990 and 2000 were evaluated and clinicopathological variables were abstracted. The perioperative complication rate, estimated blood loss (EBL), and OR time were also tabulated. Variables were analyzed using X(2) and t tests. Disease-free survival (DFS) was calculated by Kaplan-Meier method. Multivariate analysis was performed via stepwise logistic regression. Cox-proportional hazards were used to identify independent predictors of recurrence.
Results: RH was performed on 109 stage IB1 and 86 stage IB2 patients. Mean age, EBL, and perioperative complication rates were similar. Overall, 38 patients (14 IB1 vs. 24 IB2) had positive nodes (P = 0.01) including 9 patients with positive para-aortic nodes (2 IB1 and 7 IB2). Parametrial involvement (PI) and outer 2/3 depth of invasion (DOI) were significantly more common in the IB2 tumors as well. Patients with IB2 disease received adjuvant radiation more frequently than IB1 patients (52% vs. 37%, P = 0.04). Univariate predictors of nodal status included lymphovascular space involvement (LVSI) (P = 0.001), DOI (P = 0.011), PI (P = 0.001), and stage (P = 0.011). Multivariate analysis identified only LVSI (OR 6.4, CI 2.4-17, P = 0. 0002) and PI (OR 8, CI 3.1-20, P = 0. 0001) as independent predictors of positive nodes. With a median follow-up of 35 months, estimates of DFS revealed tumor size (P = 0.008), nodal status (P = 0.0004), LVSI (P = 0.002), PI (P = 0.004), and DOI (P = 0.0004) as significant univariate predictors. Neoadjuvant chemotherapy, age, grade, histology, and adjuvant radiation were not associated with recurrence. The significant independent predictors of DFS were LVSI (ROR 5.7, CI 2-16, P = 0.0064) and outer 2/3 DOI (OR 5.8, CI 2-20, P = 0.0029). Neither tumor size nor nodal status was a significant predictor of DFS.
Conclusions: The prognosis in stage IB cervical cancer seems to be most influenced by presence of LVSI and DOI and not by tumor size as the staging criteria would suggest. These factors are best determined pathologically after radical hysterectomy. This report contains the largest comparison of IB1 and IB2 patients managed by RH. Tumor size failed to predict recurrence or nodal status when stratified by LVSI, DOI, and PI. Treatment decisions based on tumor size alone should be reconsidered.