Use of parallel-plate ionization chambers in reference dosimetry of NOVAC and LIAC® mobile electron linear accelerators for intraoperative radiotherapy: a multi-center survey.

PURPOSE: LIAC® and NOVAC are two mobile linear accelerators dedicated to intraoperative radiation therapy (IORT), generating electron beams in the energy range of 3-12 MeV. Due to high dose-per-pulse (up to 70 mGy per pulse), in 2003 the Italian National Institute of Health (ISS) stated that "for the measure of dose to water in reference conditions, ionization chambers cannot be employed and no published dosimetry protocol can be used". As a consequence, ferrous sulphate (or, alternatively alanine) dosimetry was recommended. Based on a retrospective multi-center survey, a comparison with ferrous sulphate dosimetry is now used to validate the parallel-plate ionization chambers for reference dosimetry of NOVAC and LIAC. METHODS: The IAEA TRS-398 dosimetry protocol was applied except for the reference irradiation setup and the determination of the ion-recombination correction factor ks . Moreover the depth of maximum dose (R100 ) instead of zref as measurement depth was chosen by the majority of centers, thus implying a renormalization of the beam-quality correction factor kQ,Qo , based on water-air stopping power ratios. Regarding the ks determination, a previously published method, independent of ferrous sulphate dosimetry, was adopted. All the centers participating in this study had used both ferrous sulphate dosimeters and ionization chambers in water phantoms for dosimetry under reference conditions. RESULTS: The mean percentage difference between ionization chambers and ferrous sulphate dosimetry was -0.5% with a dispersion of 3.9% (2σ). Moreover, the uncertainty analysis allowed the agreement between ionization chambers and ferrous sulphate dosimetry to be verified. These results did not show any significant dependence on electron energy, thus indirectly confirming kQ,Qo renormalization. The results from the centers using zref as the measurement depth were similar to the other data, but further focused studies could aim at investigating possible dependences of the dose differences on the chosen reference depth. CONCLUSION: The present study confirms that parallel-plate ionization chambers can properly and accurately substitute ferrous sulphate detectors in reference dosimetry of LIAC and NOVAC mobile linear accelerators. Therefore, we hope that the most commonly used protocols for reference dosimetry in external-beam radiotherapy will be updated in order to provide guidance in the calibration of electron beams from linear accelerators dedicated to IORT, so that users may benefit from specific, authoritative and up-to-date recommendations.

Thyroid disorders in patients treated with radiotherapy for head-and-neck cancer: a retrospective analysis of seventy-three patients.

PURPOSE: To evaluate the incidence of thyroid disorders and dose distribution to the thyroid in patients treated with radiotherapy for head-and-neck carcinomas. METHODS AND MATERIALS: A retrospective evaluation of data from 73 patients treated for head-and-neck cancers in our department was performed. Thyroid function was evaluated mainly by the measurement of thyrotropin (thyroid stimulating hormone [TSH]). A retrospective analysis of treatment plans was performed for 57 patients. Percentages of thyroid glandular volume absorbing 10, 30, and 50 Gy (V10, V30, and V50 respectively) were considered for statistical analysis. RESULTS: A majority of patients (61%) had a normal thyroid function whereas 19 patients (26%) had hypothyroidism. Mean thyroid volume was 30.39 cc. Point 3 (located at isthmus) absorbed lower doses compared with other points (p < 0.0001). Median values of V10, V30, and V50 were 92% (range, 57-100%), 75% (range, 28.5-100%), and 35% (range, 3-83%) respectively. Gender was associated with toxicity (presence of any kind of thyroid disorders) (p < 0.05), with females displaying higher levels of TSHr (relative TSH = patient's value/maximum value of the laboratory range) (p = 0.0005) and smaller thyroid volume (p = 0.0012) compared with male population. TSHr values were associated with thyroid volume, and the presence of midline shielding block in the anterior field was associated with relative free thyroxine (FT4r = patient's value/maximum value of the laboratory range) values. CONCLUSIONS: Gender and thyroid volume seem to play an important role in the occurrence of thyroid toxicity, but further studies on dose-effect relationship for radiotherapy-induced thyroid toxicity are needed.

Real-time in vivo dosimetry using micro-MOSFET detectors during intraoperative electron beam radiation therapy in early-stage breast cancer.

PURPOSE: In a previous paper we reported the results of off-line in vivo measurements using radiochromic films in IOERT. In the present study, a further step was made, aiming at the improvement of the effectiveness of in vivo dosimetry, based on a real-time check of the dose. MATERIALS AND METHODS: Entrance dose was determined using micro-MOSFET detectors placed inside a thin, sterile, transparent catheter. The epoxy side of the detector was faced towards the beam to minimize the anisotropy. Each detector was plugged into a bias supply (standard sensitivity) and calibrated at 5 Gy using 6 MeV electrons produced by a conventional linac. Detectors were characterized in terms of linearity, precision and dose per pulse dependence. No energy and temperature dependence was found. The sensitivity change of detectors was about 1% per 20 Gy accumulated dose. Correction factors to convert surface to entrance dose were determined for each combination of energy and applicator. From November 2004 to May 2005, in vivo dosimetry was performed on 45 patients affected by early-stage breast cancer, who underwent IOERT to the tumour bed. IOERT was delivered using electrons (4-10 MeV) at high dose per pulse, produced by either a Novac7 or a Liac mobile linac. RESULTS: The mean ratio between measured and expected dose was 1.006+/-0.035 (1 SD), in the range 0.92-1.1. The procedure uncertainty was 3.6%. Micro-MOSFETs appeared suitable for in vivo dosimetry in IOERT, although some unfavourable aspects, like the limited lifetime and the anisotropy with no build-up, were found. Prospectively, a real-time action level (+/-6%) on dose discrepancy was defined. CONCLUSIONS: Excellent agreement between measured and expected doses was found. Real-time in vivo dosimetry appeared feasible, reliable and more effective than the method previously published.