STANDARDIZED ORGAN AT RISK VOLUME FOR EXTERNAL BEAM RADIATION THERAPY IN PROSTATE CANCER AND ITS RELATIONSHIP TO VOLUMES DEFINED BY CLINICAL PRACTICE

Abstract: Background/ Purpose: In radiation therapy (RT), the radiation dose calculated in tumours and normal tissue depends on how the volumes are defined. It is difficult to standardize how tumour volumes should be defined because their anatomical extension can differ greatly between individuals. Organs at risk (OARs) are more uniform, but opinions about how to define and delineate them diverge in the radiation oncology community. When OAR volumes are defined according to different guidelines, it is difficult to compare dose and volume measures between patients as well as between clinics and the understanding of the relationship between dose and side effects is also limited. The anatomical location of the prostate gland makes rectum one OAR in the treatment of prostate cancer. The purpose of this study was to investigate how rectal volumes defined according to upcoming Swedish national guidelines (STRÅNG) differ from rectal volumes defined in clinical practice for patients treated with external beam RT for prostate cancer. Other objectives were to determine potential dose differences between the two strategies to define rectal volumes and implications for treatment decisions based on associated rectal dose tolerance doses. Method: 121 patients with prostate cancer treated to 50 Gy (n=19), 66 Gy (n=71) or 70 Gy (n=31) in 2018 at the Sahlgrenska University Hospital in Göteborg, Sweden were studied with respect to planned rectum volumes and doses as calculated on pre treatment CT images in the Eclipse treatment planning system. For these patients, the rectum was defined previously both according to local clinical practice and according to STRÅNG. Volume and dose differences between the two rectal definitions were investigated for each patient and volume overlap was quantified using the Dice similarity coefficient. Comparisons were also made with respect to treatment planning criteria at the Sahlgrenska University Hospital (version 2013-2019 and from 2019) and tolerance thresholds by QUANTEC from 2010. Result: Conclusion: The clinical rectal volumes were on average larger than the volumes defined by STRÅNG (p<0.01). More patients had volume differences in caudal direction than in cranial direction, but the cranial volume differences were on average larger than the caudal volume differences (p<0.01). Dice ranged between 0.72 to 1.00 with no statistically significant average difference between the investigated treatment groups. There were small average differences in mean doses between clinically-defined rectum volumes and volumes according to STRÅNG (<1 Gy). Both rectal volumes generally satisfied the investigated criteria. However, there were 8 patients where the STRÅNG volume did not satisfy the Sahlgrenska University Hospital criteria used for the original treatment decisions although the clinical volume did. The result was similar for the QUANTEC criteria from 2010 that the clinically-defined rectum volumes satisfy the criteria better than volumes according to STRÅNG. This was a first comparison of how rectal volumes defined according to upcoming Swedish national guidelines differ from real-life rectal volumes in prostate cancer RT. The clinical rectal volumes were generally larger than the guideline volumes but overall the dose differences between them were small and resulted in a limited impact on clinical decision making except for individual cases. The presented results can guide future studies on what to expect when comparing different ways to define OAR volumes. Evaluation of compliance to guideline-directed OAR definitions could represent other future directions of this research.