Premium
MO‐F‐16A‐05: Spot‐Scanning Proton Therapy Patient‐Specific Quality Assurance: Our Methodology and Results From 295 Treatment Plans
Author(s) -
Mackin D,
Zhang X,
Li Y,
Li H,
Wu R,
Poenisch F,
Suzuki K,
Kerr M,
Holmes C,
Zhu X,
Sahoo N,
Gillin M
Publication year - 2014
Publication title -
medical physics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.473
H-Index - 180
eISSN - 2473-4209
pISSN - 0094-2405
DOI - 10.1118/1.4889176
Subject(s) - quality assurance , proton therapy , nuclear medicine , radiation treatment planning , dose rate , dosimetry , proton , range (aeronautics) , head and neck , medical physics , medicine , radiation therapy , materials science , physics , radiology , surgery , nuclear physics , external quality assessment , pathology , composite material
Purpose: To report on the patient‐specific quality assurance (PSQA) results for 295 spot‐scanning proton therapy treatment plans from the MD Anderson PTC‐Houston. We show how the results differed by treatment site and how they were affected by the treatment plan optimization method and by a range shifter in the treatment field. We also discuss some causes of PSQA problems. Methods: The PSQA procedure, which is designed to verify both the accuracy of the treatment planning system's (Eclipse™ v8.9) dose calculations and the dose delivery of the Hitachi PROBEAT synchrotron, consists of (1) an end‐to‐end test in which the beam is delivered and measured at the prescribed gantry angle, and (2) additional dose plane measurements made from gantry angle 270°. HPlusQA™ software automatically performs the gamma analysis with criteria 3% (dose tolerance), 3 mm (distance‐to‐agreement, DTA) and 2%, 2 mm. Passing is defined as at least 90% of the pixels having a gamma score less than 1. Results: The PSQA gamma passing rate was 96.2% for 3%, 3 mm, and 85.3% for 2%, 2 mm. The rate depended on the treatment site. For example, the 3%, 3 mm passing rate was 95% for head and neck plans, vs 100% for prostate plans. The passing rates of multi‐ vs. single‐field optimization plans did not significantly differ. However, the rate for fields with range shifters was 94.8±0.6%, vs 99.0±0.6% for those without (p = 0.002). Longitudinal dose gradients caused most of the low scores. Overestimation of the calculated dose proximal to the spread‐out Bragg peak (SOBP) caused many of the others. Conclusion: The planned and delivered doses consistently agreed within tolerance levels. Minor dose modeling deficiencies remain proximal to the SOBP. The 3% dose tolerance, 3 mm DTA, with 90% pixel passing rate is a reasonable action level for 2D gamma comparisons.