Clinical Implementation of “Plan of the Day” Strategy in Definitive Radiation Therapy of Cervical Cancer: Online Adaptation to Address the Challenge of Organ Filling Reproducibility

Abstract

Background: Definitive pelvic intensity-modulated radiotherapy (IMRT) in cervical cancer is susceptible to geographic miss due to daily positional and volumetric variations in target and organs at risk. Hence, despite evidence of reduced acute and late treatment-related toxicities, implementation of image-guided IMRT (IG-IMRT) with a reasonable safety margin to encompass organ motion is challenging. Materials and Methods: In this prospective, non-randomized phase II study, patients with cervical cancer FIGO (2009) stage IB2-IIIB between the ages of 18 and 65 years were treated with definitive pelvic CT-RT with a pre-specified organ (bladder and rectum) filling protocol. Reproducibility of organ filling was assessed along with the implementation of daily comprehensive adaptive IGRT, with a library of 3 IMRT (VMAT) plans with incremental expansions of CTV to PTV (Primary) margins (small 0.7 cm, adequate 1 cm and large 1.5 cm) and a backup motion robust 3DCRT plan; the appropriate plan is chosen based on pre-treatment CBCT (‘Plan of the Day’ approach). Results: Fifty patients with a median age of 49 years (IQR 45-56 years), received definitive RT (45-46Gy in 23-25 fractions to pelvis; with simultaneous integrated boost to gross nodes in 15 patients) with the said IGRT protocol. In the analysis of 1171 CBCT images (in 1184 treatment sessions), the mean planning CT and CBCT bladder volumes were 417 ccs and 373 ccs respectively. Significant inter-fractional variation in bladder volume was noted with a mean absolute dispersion of 29.5% with respect to planning CT; significant influential random factors were post-chemotherapy sessions (p=<0.001), pre-CBCT protocol duration (p= 0.001), grades of CINV (p=0.001). Significantly higher variation in bladder filling noted in patients with higher age (p=0.014) and larger planning CT bladder volume (p=<0.001). Time trend analysis of fraction-wise bladder volume revealed an absolute systemic reduction of 16.3% in bladder volume means from 1st to 5th week. Variation in rectal diameter was much less pronounced, with 19.2% mean dispersion, without any significant factors affecting it.While in 19% and 2% of sessions ‘Large’ IMRT PTV and ‘3DCRT’ were necessary to cover the primary target respectively, reduction in treated volume was possible in 43% of sessions with ‘Small’ PTV selection instead of standard ‘Adequate’ PTV (36% sessions). ‘Plan of the day’ selection had a moderate to strong correlation with non-absolute dispersion of bladder filling (Spearman's R= 0.4; p= 0.001) and a weak (but significant) correlation with grades of acute toxicities. The planned protocol was well tolerated with no radiation-induced local grade III toxicity. Conclusion: Inter-fractional variation in organ filling (especially bladder) is inevitable despite fixed pre-treatment protocol in definitive settings (intact cervix). Despite the logistical challenges; adaptive IGRT in form of ‘plan of the day’ based on incremental CTV-to-PTV margins is a relatively simple and feasible strategy to minimise geometric uncertainties in radical IG-IMRT of cervical cancer.