Comparing log file to measurement-based patient-specific quality assurance.

Recent technological advances have allowed the possibility of performing patient-specific quality assurance (QA) without time-intensive measurements. The objectives of this study are to: (1) compare how well the log file-based Mobius QA system agrees with measurement-based QA methods (ArcCHECK and portal dosimetry, PD) in passing and failing plans, and; (2) evaluate their error sensitivities. To these ends, ten phantom plans and 100 patient plans were measured with ArcCHECK and PD on VitalBeam, while log files were sent to Mobius for dose recalculation. Gamma evaluation was performed using criteria 3%/2 mm, per TG218 recommendations, and non-inferiority of the Mobius recalculation was determined with statistical testing. Ten random plans were edited to include systematic errors, then subjected to QA. Receiver operating characteristic curves were constructed to compare error sensitivities across the QA systems, and clinical significance of the errors was determined by recalculating dose to patients. We found no significant difference between Mobius, ArcCHECK, and PD in passing plans at the TG218 action limit. Mobius showed good sensitivity to collimator and gantry errors but not MLC bank shift errors, but could flag discrepancies in treatment delivery. Systematic errors were clinically significant only at large magnitudes; such unacceptable plans did not pass QA checks at the TG218 tolerance limit. Our results show that Mobius is not inferior to existing measurement-based QA systems, and can supplement existing QA practice by detecting real-time delivery discrepancies. However, it is still important to maintain rigorous routine machine QA to ensure reliability of machine log files.

Evaluation of six TPS algorithms in computing entrance and exit doses.

Entrance and exit doses are commonly measured in in vivo dosimetry for comparison with expected values, usually generated by the treatment planning system (TPS), to verify accuracy of treatment delivery. This report aims to evaluate the accuracy of six TPS algorithms in computing entrance and exit doses for a 6 MV beam. The algorithms tested were: pencil beam convolution (Eclipse PBC), analytical anisotropic algorithm (Eclipse AAA), AcurosXB (Eclipse AXB), FFT convolution (XiO Convolution), multigrid superposition (XiO Superposition), and Monte Carlo photon (Monaco MC). Measurements with ionization chamber (IC) and diode detector in water phantoms were used as a reference. Comparisons were done in terms of central axis point dose, 1D relative profiles, and 2D absolute gamma analysis. Entrance doses computed by all TPS algorithms agreed to within 2% of the measured values. Exit doses computed by XiO Convolution, XiO Superposition, Eclipse AXB, and Monaco MC agreed with the IC measured doses to within 2%-3%. Meanwhile, Eclipse PBC and Eclipse AAA computed exit doses were higher than the IC measured doses by up to 5.3% and 4.8%, respectively. Both algorithms assume that full backscatter exists even at the exit level, leading to an overestimation of exit doses. Despite good agreements at the central axis for Eclipse AXB and Monaco MC, 1D relative comparisons showed profiles mismatched at depths beyond 11.5 cm. Overall, the 2D absolute gamma (3%/3 mm) pass rates were better for Monaco MC, while Eclipse AXB failed mostly at the outer 20% of the field area. The findings of this study serve as a useful baseline for the implementation of entrance and exit in vivo dosimetry in clinical departments utilizing any of these six common TPS algorithms for reference comparison.

Can radiation therapy treatment planning system accurately predict surface doses in postmastectomy radiation therapy patients?

Skin doses have been an important factor in the dose prescription for breast radiotherapy. Recent advances in radiotherapy treatment techniques, such as intensity-modulated radiation therapy (IMRT) and new treatment schemes such as hypofractionated breast therapy have made the precise determination of the surface dose necessary. Detailed information of the dose at various depths of the skin is also critical in designing new treatment strategies. The purpose of this work was to assess the accuracy of surface dose calculation by a clinically used treatment planning system and those measured by thermoluminescence dosimeters (TLDs) in a customized chest wall phantom. This study involved the construction of a chest wall phantom for skin dose assessment. Seven TLDs were distributed throughout each right chest wall phantom to give adequate representation of measured radiation doses. Point doses from the CMS Xio® treatment planning system (TPS) were calculated for each relevant TLD positions and results correlated. There were no significant difference between measured absorbed dose by TLD and calculated doses by the TPS (p > 0.05 (1-tailed). Dose accuracy of up to 2.21% was found. The deviations from the calculated absorbed doses were overall larger (3.4%) when wedges and bolus were used. 3D radiotherapy TPS is a useful and accurate tool to assess the accuracy of surface dose. Our studies have shown that radiation treatment accuracy expressed as a comparison between calculated doses (by TPS) and measured doses (by TLD dosimetry) can be accurately predicted for tangential treatment of the chest wall after mastectomy.

An ultrasonographic evaluation of skin thickness in breast cancer patients after postmastectomy radiation therapy.

BACKGROUND: To determine the usefulness of ultrasonography in the assessment of post radiotherapy skin changes in postmastectomy breast cancer patients. METHODS: Patients treated for postmastectomy radiotherapy in National University Hospital (NUH) and Tan Tock Seng Hospital (TTSH), Singapore between January 2004- December 2005 was recruited retrospectively. Ultrasound scan was performed on these Asian patients who had been treated to a total dose of 46-50 Gy with 1 cm bolus placed on the skin. The ultrasound scans were performed blinded to the RTOG scores, and the skin thickness of the individually marked points on the irradiated chest wall was compared to the corresponding points on the non-irradiated breast. RESULTS: The mean total skin thickness inclusive of the epidermis and the dermis of the right irradiated chest wall was 0.1712 mm (± 0.03392 mm) compared with the contra-lateral non-irradiated breast which was 0.1845 mm (± 0.04089 mm; p = 0.007). The left irradiated chest wall had a mean skin thickness of 0.1764 mm (± 0.03184 mm) compared with the right non-irradiated breast which was 0.1835 mm (± 0.02584 mm; p = 0.025). These independent t-tests produced a significant difference of reduced skin thickness on the right irradiated chest wall, p = 0.007 (p < 0.05) and left irradiated chest wall p = 0.025 (p < 0.025) in comparison to the non-irradiated skin thickness investigating chronic skin reactions. Patients with grade 2 acute skin toxicity presented with thinner skin as compared to patients with grade 1 (p = 0.006). CONCLUSIONS: This study has shown that there is a statistically significant difference between the skin thicknesses of the irradiated chest wall and the contra-lateral non-irradiated breast and a predisposition to chronic reactions was found in patients with acute RTOG scoring of grade1 and grade 2.

Optimising radiation therapy techniques for tumours of the central nervous system.

INTRODUCTION: This study aims to assess the early tumour outcome and morbidity associated with radiation therapy (RT) in tumours of the central nervous system (CNS). MATERIALS AND METHODS: Patients receiving RT with radical intent were entered on a prospective database. Tumour types were categorised into glioma, base of skull, pituitary, germ cell or primitive neuroectodermal tumour (PNET) and other malignant CNS tumours. Study endpoints were overall survival and progression free survival. Acute and late toxicity endpoints included Common Terminology Criteria version 3.0 (CTC) grade 3 or 4 events, need for admission during RT and change in performance status at 12 months. RESULTS: One hundred and fifty-two patients with CNS tumours were managed with radical intent over the 4-year period. The median age was 49 years and 68.4% were Eastern Co-operative Group (ECOG) 0-1 performance status. The major pathology groups were glioma (59.9%) and base of skull tumours (17.1%). Gross total resection was performed in 28.3% only and RT was delayed after diagnosis until time of progression in 19.7%. For the 91 patients with glioma, the median survival and 2-year survival rate was 19.1 months and 44.1%, respectively. The 2-year survival rates for the subgroups of WHO Grade I or II, III and IV were 100%, 52% and 35%, respectively. For the non-glioma tumour groups, the relapse varied with pathology. Toxicity was minimal with only 3 acute and 3 late CTC grade 3 or 4 events occurring. Overall, 47 or 31% of patients required some inpatient hospitalisation during RT, although this was determined to have some causative relationship to RT in only 12 or 8% of patients. In the 12 months post-RT, performance status was stable or improved in 76.2% of patients, and most deterioration was associated with tumour relapse. CONCLUSIONS: RT for CNS tumours using modern techniques was well-tolerated with good tumour outcome and minimal morbidity.