The value of tumor paramaters and dosimetry on prognostic nomograms for cervical cancer.

Abstract

e17026 Background: A range of prognostic factors had been reported to be associated with clinical outcome of cervical cancer patients. However, most of these parameters were measured before the start of treatment but without consideration of tumor response to RT. We believed that it might be possible to provide locally advanced cervical cancer patients an opportunity to modify and guide the treatment strategies in the midway of treatment based on the early response evaluation during RT. To establish effective prognostic nomograms using clinical features including tumor volume and size mesured by MRI before treatment and after the completion of external beam radiotherapy (EBRT), and detailed dosimetry of brachytherapy dose for high-risk clinical target volume (HRCTV)D90 and Low-risk Clinical Target Volume(LR-CTV)D90. Methods: The nomogram for local control (LC) was based on a retrospective study of 316 patients who underwent IMRT at our hospital from 2010 to 2015. The predictive accuracy and discriminative ability of our nomogram models were determined by concordance index and calibration curve, and were compared with the nomogram models combining clinical features with FIGO stage. The results were validated using bootstrap resampling and a cohort study of 141 patients. The same data cohort was used to predict the progress-free survival (PFS) of cervical cancer with 3:1 training cohort (N = 310) and validation cohort (N = 155). Results: The following factors were assembled into our prognostic survival nomogram models: Age, tumor volume and size (TV & TS) before treatment, TV and TS after the completion of external beam radiotherapy (EBRT), brachytherapy dose of high-risk clinical target volume (HRCTV)D90 and Low-risk Clinical Target Volume(LR-CTV)D90. The calibration curves showed good agreement between nomogram-predicted and actual survival. Our nomogram models for LC and PFS, provided better results than the nomogram models combining clinical features with FIGO stage. Results were further confirmed in the validation set. Conclusions: Clinical features including tumor size and volume mesured before treatment and after EBRT, as well as detailed dosimetry of brachytherapy dose are able to improve the performance of prognostic nomograms for patients with cervical cancer.