Survival, prognostic factors, and relapse patterns in uterine cervical carcinoma

Between 1965 and 1979, 295 evaluable patients were treated with radiotherapy and/or surgery for invasive cervical carcinoma at the Stanford University Medical Center. The records of these patients were analyzed retrospectively with regard to survival, possible prognostic factors of freedom-from-relapse, and relapse patterns. Survival for the entire population at 5 years was 67%, with a median follow-up of 5 years. Five-year freedom-from-relapse (FFR) rates for stages I-IV were 88%, 61%, 33%, and 11%, respectively. Eighty-eight percent of all relapses occurred within 5 years. Several possible prognostic factors of FFR--including stage, histology, tumor differentiation, lymphangiogram status, treatment modality, age, presence of cervical stump, and initial serum hemoglobin value--were evaluated with a multivariate analysis. Stage was the best predictor of FFR, and lymphangiogram (LAG) status was the next most important covariate: 13 of 19 patients who showed involvement of paraaortic nodes on LAG relapsed within 1 year. There was no difference in survival or FFR between adenocarcinoma or squamous cell histologies. FFR in stage IB patients was independent of treatment modality; i.e., radiotherapy alone, surgery alone, or combined therapy. Initial serum hemoglobin (less than 12 vs. greater than 12) was not predictive of FFR or survival. Stage for stage, patients over 57 years of age tended to have better FFR than younger patients. The majority of relapses occurred centrally or at the pelvic sidewalls (58%), and future attempts to improve local control rates will rely more heavily on the individualized use of interstitial template applicators.