Using daily diagnostic quality images to validate planning margins for prostate interfractional variations

• Published in: Journal of applied clinical medical physics 2016-05, Vol.17 (3), p.61-74
• Main Authors: Li, Wen, Vassil, Andrew, Godley, Andrew, Mossolly, Lama Muhieddine, Shang, Qingyang, Xia, Ping
• Format: Article
• Language: English
• Subjects: Bladder; Cancer therapies; Clinical medicine; diagnostic image; Dose Fractionation; Dosimetry; Follow-Up Studies; Humans; interfraction; Localization; Male; Patients; Planning; planning margins; Principal components analysis; Prostate cancer; Prostatic Neoplasms - diagnostic imaging; Prostatic Neoplasms - radiotherapy; Quality of Health Care; Radiation Oncology Physics; Radiation therapy; Radiotherapy Dosage; Radiotherapy Planning, Computer-Assisted - methods; Rectum - diagnostic imaging; Registration; Retrospective Studies; Studies; Tomography, X-Ray Computed - methods; Urinary Bladder - diagnostic imaging
• ISSN: 1526-9914; 1526-9914
• DOI: 10.1120/jacmp.v17i3.5923

The purpose of this study is to use the same diagnostic‐quality verification and planning CTs to validate planning margin account for residual interfractional variations with image‐guided soft tissue alignment of the prostate. For nine prostate cancer patients treated with IMRT to 78 Gy in 39 fractions, daily verification CT‐on‐rails images of the first seven and last seven fractions (n=126) were retrospectively selected for this study. On these images, prostate, bladder, and rectum were delineated by the same attending physician. Clinical plans were created with a margin of 8 mm except for 5 mm posteriorly, referred to as 8/5 mm. Three additional plans were created for each patient with the margins of 6/4 mm, 4/2 mm, and 2 mm uniform. These plans were subsequently applied to daily images and radiation doses were recalculated. The isocenters of these plans were placed according to clinical online shifts, which were based on soft tissue alignment to the prostate. Retrospective offline shifts by aligning prostate contours were compared to online shifts. The resultant daily target dose was analyzed using D99, the percentage of the prescription dose received by 99% of CTV. The percent of bladder volume receiving 65 Gy (V65Gy) and rectum V70Gy were also analyzed. After interfractional correction, using CTV D99>97%% criteria, 8/5 mm, 6/4 mm, 4/2 mm, and 2 mm planning margins met the CTV dose coverage in 95%, 91%, 65%, and 53% of the 126 fractions with online shifts, and 99%, 98%, 85%, and 68% with offline shifts. The rectum V70Gy and bladder V65Gy were significantly decreased at each level of margin reduction (p