Variations in inter-observer contouring and its impact on dosimetric and radiobiological parameters for intensity-modulated radiotherapy planning in treatment of localised prostate cancer

Abstract

AbstractInter-observer variations in contouring and their impacts on dosimetric and radiobiological parameters in intensity-modulated radiotherapy (IMRT) treatment for localised prostate cancer patients were investigated. Four observers delineated the gross tumour volume (GTV) (prostate and seminal vesicles), bladder and rectum for nine patients. Contouring done by radiologist was considered as gold standard for comparison purposes and for IMRT plan optimisation. Maximum average variations in contoured prostate, bladder and rectum volumes were 3% (SD = 8.4), 2.5% (SD = 4.12) and 13.2% (SD = 6.77), respectively. The average conformity index for standard contouring set (observer A) was 0.85 (SD = 0.028) and statistically significant differences were observed for observers A–B (p = 0.008), A–C (p = 0.006) and A–D (p = 0.011). Average values of normal tissue complication probability for bladder and rectum for observer A were 0.361% (SD = 0.036) and 1.59% (SD = 0.14). Maximum average tumour control probability was 99.94% (SD = 0.035) and statistically significant difference was observed for observers A–B (p = 0.037) and observers A–C (p = 0.01). Inter-observer contouring variations have significant impact on dosimetric and radiobiological outcome in IMRT treatment planning. So accurate contouring of tumour and normal organs is a fundamental prerequisite to make good correlation between calculated and clinical observed results.