Local control prediction by histologic grading or radiation effects with drill biopsy specimens for uterine cervical cancer.

Abstract

Local control predictability by histologic grading for radiation effects was analyzed on 224 patients with cervical squamous cell carcinomas obtaining complete regression. The cervical specimens after radiation therapy were obtained by drill biopsy technic. Radiation effect was assessed with Shimosato's histologic classification for radiation effect. The cancer tissues were altered by radiation to various degree from destruction of cancer nests with numerous viable cancer cells indicating grade 2a to scar tissues without viable or non-viable cancer cells indicating grade 4 at the end of treatment. Forty five percent of the patients had viable cancer cells remaining on the specimens at that time. The numbers of the grades 2a, 2b, 3, and 4 in biopsy specimens were 36, 65, 39, and 84, respectively. The local control rates and 5-year survival rates were 53% and 28% for grades 2a, 75% and 51% for grade 2b, 95% and 74% for grade 3, and 92% and 76% for grade 4, respectively. The histologic grading was significantly associated with the local control and 5-year survival. Of 41 patients having remaining viable cancer cells, residual tumors disappeared in 33 patients (80%) and 8 (20%) still persisted to show grades 2a-b at 1 to 2 months after radiotherapy. Of the 8, all four patients without any salvage treatment, died of local recurrence. These results suggested that the remaining viable cancer cells in the specimens at 1 to 2 months after radiation therapy were highly associated with recurrence and that the histologic grading system was an indicator predictive for prognosis.