Estimated Limits of Dose Escalation for Localized Prostate Cancer Using Intensity Modulated Radiotherapy, Varied Planning Target Volume Margins and Established Dose Constraints: Correlation with Prostate Size and Target Volume Overlap with the Rectum

Abstract

Listen

Translate article

Purpose/Objective: Recent reports of dose escalation (DE) for prostate cancer (CaP) indicate minimal toxicity using established dose constraints (DC). The purpose of this study is to estimate the limits of DE using commonly employed 5 field intensity modulated radiotherapy (5FIMRT) as a function of planning target volume (PTV) margins and other factors. Materials/Methods: CT data was obtained from 18 patients with localized CaP treated with 5FIMRT between 9/03 and 2/05 to 75.6 Gy in 42 fractions. Uniform PTV margins of 10, 5, and 3mm were generated for each case. The maximum achievable dose (MAD) using identical 6MV 5FIMRT beam arrangements was determined through iterative plan optimizations until the RTOG protocol P0126 DC for the rectum, bladder, and PTV could no longer be met. The relationship between MAD and PTV margin, prostate volume, and PTV overlap with the rectum was examined. Results: Mean prostate volume is 56cc (range: 32–103cc). Mean conformity index is 1.09 (range: 1.04–1.14). The rectal V75 DC limits DE in 61% of these plans. Other limiting DC include PTV coverage (19%) and bladder V80 (19%). PTV margins of 10, 5, and 3mm yield a mean MAD of 83.0 Gy (range: 73.8–108.0 Gy), 113.1 Gy (range: 90.0–151.2 Gy), and 135.9 Gy (range: 102.6–189.0 Gy) respectively. All comparisons of the MAD between margin groups are statistically significant with one-sided p values <0.001 (paired t-test). Prostate volumes of 30–50cc (n=8) compared to those 50–70cc (n=7) and 70–105cc (n=3) show an inverse correlation with MAD (Fig. 1). Smaller prostates yield significantly higher MAD for PTV margins of 3 and 5mm (p<0.05). Decreasing the PTV margin significantly decreases the PTV overlap with the rectum (one-sided p values <0.001 between all margin comparisons). The volume of PTV overlapping the rectum demonstrates a substantial logarithmic correlation (R=0.87) with MAD (Fig. 2). MAD declines from 150 Gy to 100 Gy from 0 to 5cc of overlap, and thereafter drops slowly to 75 Gy with 5 to 20cc of overlap.Fig. 2Relationship between MAD and the component of PTV overlapping the rectum for varying prostate margins.View Large Image Figure ViewerDownload (PPT) Conclusions: By decreasing the PTV margin, while maintaining identical DC, doses well above those currently prescribed for treatment of localized CaP appear feasible. While RTOG protocol P0126 specifies a PTV margin of 5 to 10mm, the use of increasingly accurate target localization methods may reasonably permit use of a smaller margin, such as 3mm. Paired with the DC used here, it is suggested that IMRT doses could approach those routinely employed in permanent prostate brachytherapy. The inverse relationship demonstrated between the MAD and prostate size or PTV overlapping the rectum provides data which quantifies the potential benefit of prostate cytoreduction prior to radiation. However, the DE suggested here remains a theoretical estimate and DC in addition to those typically employed will likely be necessary to limit normal tissue toxicity.