Background: The authors analyzed the radiation dose to the periphery of the cervix and area of the cervix in relation to local failure of radiotherapy for carcinoma of the cervix using computed tomography (CT) images.
Methods: Between 1981-1990, 127 consecutive patients were treated with definitive radiotherapy. Ninety-nine of these patients had CT images taken at the time of intracavitary therapy. Of these 99 patients, 80 were eligible for this analysis. After CT scanning, isodose curves relative to the point A dose were superimposed on the CT images. The minimum percent dose and minimum dose at the periphery of the cervix were estimated. The area of the cervix also was measured. These factors were examined in relation to the local tumor control rate.
Results: Histograms of both the minimum percent dose and the cervical area showed significant differences between the local control and local failure groups (P <0.001). The local control rates were related to both the minimum percent dose and the cervical area, and differed significantly over and below the values of 60% and 18 cm2 (P <0.001 each), respectively. The local control patients, over and below the line: Y = -0.220X + 21.2, in which X (gray [Gy]) and Y (Gy) are the whole pelvis dose and the minimum dose, respectively, could be well differentiated with significance (91.7% vs. 25.0%; P <0.001).
Conclusions: Computed tomography analysis indicated that the local tumor control rate was related strongly to the minimum percent dose, the cervical area, and the pair of whole pelvis and minimum dose values. These factors were found to be more useful than the point A dose in predicting local tumor control.