A pre-optimised dosimetry system using a rigid applicator for intracavitary treatment of cervical carcinoma

Tumour control and complication risk have been major concerns in the treatment of cervical carcinoma. A review of dose distribution for intracavitary treatment of cervical carcinoma revealed that modification of the Manchester dosimetry system is necessary for cases of narrow-sized vagina. A revised dosimetry system was introduced in the present study, with the objective of optimising the dose coverage for the parametrium while minimising the bladder and rectum dosage by restricting the rectal dose so as not to exceed 75% of the brachytherapy prescription dose.

A suitable-sized applicator was selected according to the patient's anatomy. The revised system is optimised based on the fixed geometry of the applicator. The system was therefore predefined and the distribution of the treatment dose already determined before application. The revised system was applied to 135 cases, involving 540 applications. The clinical outcome in terms of local tumour control and complication rates is reported. The differences between the revised system and the Manchester system in terms of dose coverage for the parametrium and the rectum dose were compared.

The results showed that higher rectal and parametrial dosages were obtained with the Manchester system as compared with the revised system. Our study showed that over 50% of our patients would have received a rectal dose close to 100% of the point A dose if the Manchester system was applied, whereas it was restricted to below 75% using the revised system. Using the revised system, the significance of the parametrial dosage coverage in relation to local control was assessed: the mean dose to the rectum and the bladder as a percentage of point A was 65.7 +/- 5% (range 50-85%) and 66.4 +/- 14% (range 29-116%), respectively. The 5-year actuarial local failure-free survival rates were 90, 92.9, 86.8, 100, 69.7 and 0% for stages IB, IIA, IIB, IIIA, IIIB and IV (P < 0.0001), respectively. The 3-year actuarial complication rates (grade 3/4) for proctitis and cystitis were 1.4 and 0.5%, respectively. The dosage coverage for the parametrium was found to be significant (P = 0.029) in relation to local control for early-stage disease.

The favourable local tumour control and low complication rates shown by our results indicate that the revised system presents an optimal dose distribution, particularly for the application of small ovoids, whereas morbidity was reduced to a lower level without compromising local control.