INTRODUCTION: Inconsistencies in contouring target volumes for partial breast irradiation (PBI) may result in geographical misses and compromise treatment outcomes. The present study aimed to (1) determine the variability of the target volumes contoured and treatment plans generated by participating centres in credentialing for participation in a multicentre PBI trial; and (2) assess dosimetric changes when standardized target volumes were used. METHODS: The CT image sets of two de-identified patients post-breast conserving surgery were used. Contouring of the target volumes for the two cases was performed and a treatment plan as per protocol specifications was generated for each case by the seven participating centres. Planning of both cases was repeated by five centres using a set of standardized target volumes to evaluate resulting dosimetric changes in the treatment plans. RESULTS: The surgical cavity, the part of the planning target volume used for dose evaluation and ipsilateral whole breast volumes contoured by the centres varied by 25%, 16% and 21% (1 standard deviation), respectively. The dosimetric variations found when the standardized target volumes were used were smaller than those noted when centre-specific volumes were used. The volumes of the ipsilateral lungs receiving 30% of the prescribed dose and the volumes of the ipsilateral whole breasts receiving 95% and 50% of the prescribed dose were reduced in the treatment plans developed using the standardized target volumes. CONCLUSIONS: Given the impact of contouring on dose distributions, quality assurance procedures in clinical trials of PBI need to take into account both the technical approaches and the contouring.
INTRODUCTION: Inconsistencies in contouring target volumes for partial breast irradiation (PBI) may result in geographical misses and compromise treatment outcomes. The present study aimed to (1) determine the variability of the target volumes contoured and treatment plans generated by participating centres in credentialing for participation in a multicentre PBI trial; and (2) assess dosimetric changes when standardized target volumes were used. METHODS: The CT image sets of two de-identified patients post-breast conserving surgery were used. Contouring of the target volumes for the two cases was performed and a treatment plan as per protocol specifications was generated for each case by the seven participating centres. Planning of both cases was repeated by five centres using a set of standardized target volumes to evaluate resulting dosimetric changes in the treatment plans. RESULTS: The surgical cavity, the part of the planning target volume used for dose evaluation and ipsilateral whole breast volumes contoured by the centres varied by 25%, 16% and 21% (1 standard deviation), respectively. The dosimetric variations found when the standardized target volumes were used were smaller than those noted when centre-specific volumes were used. The volumes of the ipsilateral lungs receiving 30% of the prescribed dose and the volumes of the ipsilateral whole breasts receiving 95% and 50% of the prescribed dose were reduced in the treatment plans developed using the standardized target volumes. CONCLUSIONS: Given the impact of contouring on dose distributions, quality assurance procedures in clinical trials of PBI need to take into account both the technical approaches and the contouring.