Continuous low-dose-rate radiation of radionuclide phosphorus-32 for hemangiomas

The study goal was to clarify the therapeutic effect and the absorbed dose of radionuclide phosphorus-32 for skin hemangiomas and the consequent risk of side effects in these patients. Phosphorus-32 is an β emitter and is used for skin hemangioma treatment. In comparison with the few Gy per minute of the linear accelerators, the dose rate of phosphorus-32 for hemangiomas is much <1 Gy/hour; so, the latter is called low-dose-rate radiation. To achieve the therapeutic dose, continuous hours or days of radiation is necessary. For strawberry hemangiomas, the phosphorus-32 applicator was tightly placed on the lesion site for several hours until reaching therapeutic dose. The absorbed dose was estimated by radiochromic films. The absorbed dose of phosphorus-32 irradiation declined exponentially with a depth from 0 to 2.5 mm. Of the 316 patients with strawberry hemangiomas, the lesion disappeared completely within 3 months after one-time treatment in 259 cases (82%). For cavernous hemangiomas, 370KBq phosphorus-32 colloid was injected into the hemangioma each square centimeter, and the absorbed radiation was estimated by theoretical calculation. Forty-two of the 58 patients with cavernous hemangiomas (72%) had lesions that completely disappeared within 3 months after receiving one to six treatments. Thus, the phosphorus-32 for strawberry hemangiomas and the chromium phosphate-32 colloid for cavernous hemangiomas were clearly efficacious.     

Clinical quality assurance for endovascular brachytherapy devices.

BACKGROUND AND PURPOSE: Endovascular brachytherapy is still an important therapy modality with a high number of treated patients per year. Quality assurance of devices used has been addressed already in several publications (AAPM, DGMP, ESTRO, NCS). However, there are no clear recommendations given on test procedures and related equipment. Our experience with four different devices containing beta- ((32)P, (90)Sr/Y) and gamma-sources ((192)Ir), which were used in clinical routine during the last 3 years is described. PATIENTS AND METHODS: The incoming check includes leakage radiation, missing catheter interlock, positioning test, timer check, interrupt button check, power-off test and verification of the manual retraction facility. Dose profiles are measured using GafChromic film. Source strength verification is performed using well type chambers or air-kerma measurements. In addition, the proposed reference absorbed dose rate at 2 mm distance from the source centre is measured with a dedicated film dosimetry technique where two additional films are exposed to two known doses in a (60)Co field for calibration. RESULTS: Dosimetrical parameters (dose profiles, source strength) are found to be within +/-10% of the manufacturers specifications. The reference dose rate measured with film is on average +3.1% for 13 (90)Sr seed trains, +8.1% for three (32)P wire sources and -3.7% for one (192)Ir seed ribbon compared to the source certificate. The activity of 30 individual (32)P wire sources measured by using a calibrated well type chamber showed a deviation of mean -0.3%, the activity of 16 (192)Ir seed ribbons determined with air kerma measurements a deviation of mean 2.8%. CONCLUSIONS: The QA programme introduced in our department provides methods to verify all relevant parameters proposed by international recommendations. Film dosimetry can be used as independent verification of the reference dose rate within a 10% limit.     

Optimization of the Gafchromic™ EBT protocol for IMRT QA

Quality assurance of external beam (radio)therapy (EBT) requires tools with specific characteristics. A radiochromic film dubbed “Gafchromic? EBT” (G-EBT) that is particularly suited for external beam therapy because of its features was introduced in 2004. Its characteristics, especially the high spatial resolution, make it suitable for measurement of dose distributions in radiotherapy, especially intensity-modulated radiation therapy (IMRT). While several aspects of the film characteristics have been previously reported separately, we present a comprehensive evaluation centered on practical IMRT verification, leading to an optimized protocol. Therefore the constancy within one batch, the relationship between optical density (OD) and dose (dose range between 1.4Gy and 8.4Gy) and the dose rate dependence for four dose rates (55, 108, 217, 441 MU/min) were investigated. In addition to these characteristics, energy dependence between two energies (50 kV and 6 MV), tissue equivalency, post irradiation coloration over one month, pressure and temperature sensitivity were evaluated. We then optimized the protocol using the G-EBT films, in combination with an EPSON-Expression? 1680pro flatbed scanner, for IMRT QA, while either striving to keep the compound error as small as possible or trying to reduce evaluation time. As a basis for this protocol optimization, the characteristics of the scanner (such as inhomogeneity of the scanning field) and its software (such as consequences of extracting only the red color channel) had to be determined first. The interaction of film and scanner (variation of the OD depending on the scanning direction or the scanning resolution) was assessed as well. Using the optimized protocol for IMRT QA, the compound error could be reduced to approximately 2% for a quality-driven approach and maximum 5.5% for an approach attempting to reduce procedure time. While the quality-driven approach provides appropriate accuracy for individual patient QA, the procedure-time driven approach can only be used for preliminary measurements.     

Assessment of large single-fraction, low-energy X-ray dose with radiochromic film

PURPOSE: To investigate the accuracy of in vivo dosimetry using radiochromic film for large single-fraction, low-energy irradiations. METHODS AND MATERIALS: Gafchromic MD-55-2 radiochromic film and LiF thermoluminescent dosimeters (TLDs) were placed in vivo on 25 patients to ascertain their effectiveness for assessment of dose. All patients received 10 Gy single fractions at energies ranging from 100 kVp (half-value layer [HVL] = 3.5 mm Al) up to 250 kVp (HVL = 2.3 mm Cu). Effects of small air gaps were also investigated using LiF TLDs and radiochromic film. RESULTS: Radiochromic film adequately measured applied dose for 25 patients in vivo with a standard deviation of 5.5% from prescribed dose. LiF TLDs recorded a standard deviation of 4.1% from measured to applied dose. Small air gaps which can be created under the film or TLDs during in vivo dosimetry were shown to have a measurable but minimal effect on results for gaps less than 5 mm. CONCLUSIONS: Gafchromic film has adequately measured applied dose in vivo at low energy for large 10 Gy single-fraction irradiation.     

Optimal source position for irradiation of coronary bifurcations in endovascular brachytherapy with catheter based beta or iridium-192 sources.

BACKGROUND AND PURPOSE: Intracoronary brachytherapy after percutaneous transluminal coronary angioplasty (PTCA) is usually performed with catheter-based treatment techniques in a straight vessel segment. There is a growing interest for treatment of bifurcations, which requires consecutive positioning of the source in main vessel and side branch. MATERIALS AND METHODS: In-house developed software (IC-BT doseplan) is used to explore the optimal positioning of the source in modelled bifurcations with different shape for the source types available in our hospital, i.e. (90)Sr/(90)Y, (32)P and (192)Ir. The results were summarised in look-up tables. The usefulness of these look-up tables was tested on various clinical examples. RESULTS: Tabulated results for the modelled bifurcations yield an estimation of the distance between the sources (gap width) in relation to the geometry and source type: (90)Sr/(90)Y gap range 3-8.5 mm, (32)P gap range 2-7 mm and (192)Ir gap range 3.5-8 mm. The average dose relative to 2 mm from the source axis is: (90)Sr/(90)Y, (mean+/-SD) 120+/-40%; (32)P, 125+/-50% and (192)Ir, 120+/-22%. The look-up tables also provide the coarse location and value of maximum and minimum dose: (90)Sr/(90)Y, 220-60%, (32)P, 230-55% and (192)Ir, 170-85%. It appeared that the look-up tables provide a good approximation of the optimal gap width in the clinical examples. CONCLUSIONS: Tabulated optimal gap widths are very useful for quick estimation of the required gap width for a given bifurcation and source type, in case the prescribed dose in both vessels is the same. In unfavourable geometries there is a risk of local underdosage. Individual treatment planning using a program such as IC-BT doseplan is then recommended.     

^32P持续低剂量率辐射敷贴治疗的胶片剂量验证

目的：对^32P的敷贴治疗进行胶片剂量验证。方法：直线加速器建立25—575cGy的胶片辐射剂量曲线。放射性核素^32P溶液4．14MBq吸附在1．1cm×1．2cm的长方形滤纸上制成放射性敷贴器。将20张1．5cm×1．5cm的辐射直接显色（RC）胶片（胶片厚0．225mm）重叠放置（总厚度4．5mm）,把放射性敷贴器放置在最顶端,辐射0．5小时。通过辐射剂量曲线计算每张胶片辐射剂量,根据胶片厚度推算距离敷贴器不同距离处的辐射剂量,并绘制相应的剂量-距离曲线。结果：距离敷贴器0、0．225、0．45、0．675、0．9、1．125、1．35、1．8、2．025、2．25、2．475mm处的0．5小时辐射测量吸收剂量为265、175、120、98、68、55、39、25、19、16、12、5cGy,拟合曲线为Y：257．28e^-1.4521X。结论：放射性核素^32P敷贴器的辐射剂量随距离增加呈指数方式衰减,敷贴器表面和2．5mm处的辐射剂量相差约53倍。     

A Protocol for Accurate Radiochromic Film Dosimetry Using Radiochromic.com

BackgroundRadiochromic films have many applications in radiology and radiation therapy. Generally, the dosimetry system for radiochromic film dosimetry is composed of radiochromic films, flatbed scanner, and film analysis software. The purpose of this work is to present the effectiveness of a protocol for accurate radiochromic film dosimetry using Radiochromic.com as software for film analysis.Materials and methodsProcedures for image acquisition, lot calibration, and dose calculation are explained and analyzed. Radiochromic.com enables state-of-the-art models and corrections for radiochromic film dosimetry, such as the Multigaussian model for multichannel film dosimetry, and lateral, inter-scan, and re-calibration corrections of the response.ResultsThe protocol presented here provides accurate dose results by mitigating the sources of uncertainty that affect radiochromic film dosimetry.ConclusionsAppropriate procedures for film and scanner handling in combination with Radiochromic.com as software for film analysis make easy and accurate radiochromic film dosimetry feasible.Key words: radiochromic film, dosimetry, protocol, film analysis software     

In vivo dosimetry using radiochromic films during intraoperative electron beam radiation therapy in early-stage breast cancer.

BACKGROUND AND PURPOSE: To check the dose delivered to patients during intraoperative electron beam radiation therapy (IOERT) for early breast cancer and also to define appropriate action levels. PATIENTS AND METHODS: Between December 2000 and June 2001, 54 patients affected by early-stage breast cancer underwent exclusive IOERT to the tumour bed using a Novac7 mobile linac, after quadrantectomy. Electron beams (5, 7, 9 MeV) at high dose per pulse values (0.02-0.09 Gy/pulse) were used. The prescribed single dose was 21 Gy at the depth of 90% isodose (14-22 mm). In 35 cases, in vivo dosimetry was performed. The entrance dose was derived from the surface dose measured with thin and calibrated MD-55-2 radiochromic films, wrapped in sterile envelopes. Films were analysed 24-72 h after the irradiation using a charge-coupled-device imaging system. Field disturbance caused by the film envelope was negligible. RESULTS: The mean deviation between measured and expected doses was 1.8%, with one SD equal to 4.7%. Deviations larger than 7% were found in 23% of cases, never consecutively, not correlated with beam energy or field size and with no evidence of linac daily output variation or serious malfunctioning or human mistake. The estimated overall uncertainty of dose measurement was about 4%. In vivo dosimetry appeared both reliable and feasible. Two action levels, for unexplained observed deviations larger than 7 and 10%, were preliminary defined. CONCLUSIONS: Satisfactory agreement between measured and expected doses was found. The implementation of in vivo dosimetry in IOERT is suggested, particularly for patients enrolled in a clinical trial.     

Determination and application of the reference isodose length (RIL) for commercial endovascular brachytherapy devices.

BACKGROUND AND PURPOSE: During the last years endovascular brachytherapy has been established as a new field in radiotherapy. In a recent recommendation of the EndoVAscular, Groupe Européen de Curiethérapie, European Society for Therapeutic Radiation Oncology, Working Group the reference isodose length (RIL) has been introduced as a central parameter for treatment planning. It is defined as the vessel length at the reference depth enclosed by the 90% isodose. The dose is normalized to 100% at the reference depth (i.e. lumen radius plus 1 mm into the vessel wall) at the central plane. In order to avoid a geographic miss the clinical target length has to be encompassed by the RIL.MATERIAL AND METHODS: RILs are determined by Monte Carlo calculations and GafChromic film dosimetry for three endovascular brachytherapy devices currently in clinical use (192Ir seed ribbon, 32P wire source, 90Sr seed train). For all measurements, phantoms and devices the sensitive layer of GafChromic film is located in a plane at 2+/-0.1 mm parallel to the axis of the source delivery catheter. The EGSnrc code system is applied to calculate the dose profile at 2 and 3 mm distance from the source axis.RESULTS: For the ten seed 192Ir source calculated RIL at 2 mm radial distance is 30.2 mm whereas the measured RIL is 33.5 mm. In case of a 20 mm 32P wire source with two steps the calculated RIL of 36.6 mm shows excellent agreement compared with the measured value of 36.2 mm. The calculated RIL of a 40 mm 90Sr seed train was 35.8 mm compared to a measured value of 34.8 mm.CONCLUSION: As a compromise between calculated and measured RIL values at 2 and 3 mm radial distances we propose to use a RIL of 30 mm for the 192Ir ten seed ribbon, 36 mm for the32P wire source with two steps and 35 mm for the 40 mm 90Sr seed train. These parameter values can be used to define the maximum intervention length for endovascular sources.     

Quality control in interstitial brachytherapy of the breast using pulsed dose rate: treatment planning and dose delivery with an Ir-192 afterloading system.

BACKGROUND AND PURPOSE: In the Radiotherapy Department of Leuven, about 20% of all breast cancer patients treated with breast conserving surgery and external radiotherapy receive an additional boost with pulsed dose rate (PDR) Ir-192 brachytherapy. An investigation was performed to assess the accuracy of the delivered PDR brachytherapy treatment. Secondly, the feasibility of in vivo measurements during PDR dose delivery was investigated. MATERIALS AND METHODS: Two phantoms are manufactured to mimic a breast, one for thermoluminescent dosimetry (TLD) measurements, and one for dosimetry using radiochromic films. The TLD phantom allows measurements at 34 dose points in three planes including the basal dose points. The film phantom is designed in such a way that films can be positioned in a plane parallel and orthogonal to the needles. RESULTS: The dose distributions calculated with the TPS are in good agreement with both TLD and radiochromic film measurements (average deviations of point doses <+/-5%). However, close to the interface tissue-air the dose is overestimated by the TPS since it neglects the finite size of a breast and the associated lack of backscatter (average deviations of point doses -14%). CONCLUSION: Most deviations between measured and calculated doses, are in the order of magnitude of the uncertainty associated with the source strength specification, except for the point doses measured close to the skin. In vivo dosimetry during PDR brachytherapy treatment was found to be a valuable procedure to detect large errors, e.g. errors caused by an incorrect data transfer.     

Low dose out-of-field radiotherapy,part 1: Measurement of scattered doses

Purpose

To measure out-of-field doses in a phantom model to better quantify this radiation.

Modeling of dose to tumor and normal tissue from intraperitoneal radioimmunotherapy with alpha and beta emitters

Dose distributions for normal and tumor tissues from intraperitoneally administered radiolabeled antibodies have been calculated for 90-Yttrium (90Y), 131-Iodine (131I), and 211-Astatine (211At). The dose calculations use data on the activity of intraperitoneal fluid administered, the percent injected dose/gm uptake by tumor, biological half life, and a model for diffusion of antibody/radionuclide complex into peritoneal tissues. Calculations are performed for planar and hemispherical tumor shapes, ranging in size to establish the influence of geometry on dose distribution. Calculations for tumor geometry obtained from biopsies are also performed. When the activity is concentrated on or near the tumor surface, the maximum dose to a planar tumor for a 20 mci administration of 90Y is approximately 60 Gy, and falls rapidly to 50% of this value within 1 mm. However, for a hemispherical tumor, the dose is a maximum of 26 Gy, with an average of approximately 20 Gy. The surface dose from 131I (130 mci) is 240 Gy, and diminishes to 20 Gy in .05 cm in the planar case, whereas a hemispherical tumor receives a dose of 90 Gy over a large fraction of the volume, with the distal portions receiving 40 Gy. The surface dose for an administration of 70 mci of 211 At is 450 Gy and decreases to 50% of this value in 30 microns. Both surface geometry and tumor size are important determinants in the heterogeneity of tumor dose, as are the dose administered, antibody uptake, biodistribution, and residence time factors. These initial studies suggest that the size of disease which may be effectively treated is much less than the range of the particle emitted by radiolabeled antibodies. Furthermore, therapy is ultimately limited by the degree to which the antibody/radionuclide complex can diffuse and permeate the tumor.     

The use of BED and EUD concepts in heterogeneous radioactivity distributions on a multicellular scale for targeted radionuclide therapy

There is evidence that nonuniform activity distributions within tumors might cause targeted radionuclide therapy (TRT) to fail. The aim of this study was to investigate the effects of the temporal and spatial behavior of the radioactivity in TRT, focusing on heterogeneous radiopharmaceutical distributions at a multicellular scale. Various activity distributions at the multicellular level from three radionuclides ((32)P, (90)Y, and (131)I) were simulated in cubic matrices (1- and 3-mm side). The in-house software package DOVE was used to calculate dose-rate maps, and survival fractions were calculated taking into account an up-take and a clearance phase. The effect from nonuniform activity distributions was analyzed in terms of dose volume histograms (DVHs), biologically effective dose (BED), and the effective uniform dose (EUD). The fraction of the absorbed dose that is "wasted," without producing a biological effect to the treatment, reaches 60% in the highly nonuniform distributions. For (32)P and (90)Y, the loss of therapeutic effectiveness was shown to be less than for (131)I. However, (90)Y, owing to its shorter physical half-life, presented lower mean BED values in almost every geometry, compared to (32)P and (131)I, and thus was less effective. (131)I, among all geometries, appeared to be more effective in more homogeneous activity distributions and in the 1-mm volume of interest, whereas it was the least effective radionuclide in the more heterogeneous activity distributions. (32)P presented the highest values of EUD, compared to (90)Y and (131)I. The EUD is a unique value that facilitates comparisons between different activity distributions in terms of treatment outcome. This study showed that as the degree of the heterogeneity in the dose distributions increases, the therapy effectiveness worsens. Nonuniform absorbed dose distributions can create a situation in which a fraction of cells are underirradiated, while another fraction of cells is "over-killed."     

Évaluation des modifications radio-induites tardives de la microcirculation superficielle après irradiation aiguë β.II. Intérêt pronostique du laser doppler cutané

OBJECTIVE: The changes that occur in the tissular microcirculation after accidental acute irradiation account for some of the early effects of such irradiation, especially at the cutaneous level. The prognostic importance of the cutaneous laser doppler was tested in an experimental model of acute beta-irradiation. METHODS: Ten pigs were given beta-irradiation with a high single localized dose of 90Sr/90Y (32 or 64 Gy, 7 mg/cm2) delivered to the flank, and were evaluated 2, 7, 14, 21 and 28 days thereafter. Each individual was its own control. The local microcirculation was measured in the resting state and during thermal stimulation at 42 degrees C, using a Periflux cutaneous Doppler laser with p413 probes. Three periods of six minutes each were continuously recorded: period 1 (P1) represented basal resting cutaneous perfusion, with the slope p corresponding to the increase in perfusion when two minutes of thermal stimulation at 42 degrees C began; P2 to plateau perfusion during this stimulation; and P3 to perfusion on the return to equilibrium. RESULTS: After acute beta-irradiation in the pig, all the cutaneous microcirculation parameters measured (P1, p, P2 and P3) had risen at day 2 in the irradiated area by a factor of 2 to 4, depending on the dose (p < 0.001), compared to the adjacent control area. On the other hand, as from day 7, the resting and the stimulated microcirculation varied little, except for a reduction of the slope p by a factor of 2 (p < 0.05) after the strongest radiation dose. CONCLUSION: After acute irradiation, the increase in the resting cutaneous microcirculation may correspond to immediate but transitory capillary vasodilatation that accompanies the initial erythema in accidental irradiation. The absence of vascular response to thermal stimulation seems to be a good means of reaching an early diagnosis of delayed cutaneous radiation necrosis.     

Radiation dose to abdominal organs of the mouse due to 90Y in the urinary bladder

Radiation dosimetry estimates in mice have proven useful in evaluating therapeutic radiopharmaceuticals. Current models for mice do not take into account the dose to abdominal organs from radioactivity in the urinary bladder. Although the dose from this source is probably low for slowly clearing compounds such as antibodies, it may be considerable for small molecule (90)Y conjugates undergoing rapid renal clearance. To evaluate this possibility, we modeled the mouse bladder as a 6 mm sphere, surrounded by a 0.5 mm thick shell. We then calculated the radiation dose that might be received by the shell and by more distant points, using the point kernel method with the Loevinger analytical point kernel. A Monte Carlo calculation using EGS4 was also performed. Surface dose calculations were compared with in vitro experimental data. LiF TLD dosimeters were placed directly under five separated, flat-bottomed, 6-mm diameter wells containing (90)Y on a 96-well plate. Dose versus distance from the mouse urinary bladder was calculated using kinetic data from imaging studies of a renally cleared (111)In analog compound currently under investigation. From this, it was estimated that whole body administration of 34.8 MBq of the (90)Y analog compound would yield a bladder wall dose estimate of approximately 98 Gy. Structures within 2 mm of the bladder would receive additional estimated doses of at least 15 Gy. This radiation dose approaches that which is known from external beam data to cause fibrosis in mice. Because of the greater size of the human bladder compared with that of the mouse relative to the range of (90)Y beta particles, the radiation exposure from the same residence time in man was estimated to be considerably lower. This highlights a potential practical limitation of extrapolating radiotoxicity findings in the mouse to human subjects.     

The correlation between D90 and outcome for I-125 seed implant monotherapy for localised prostate cancer.

BACKGROUND AND PURPOSE: In 1998 Stock and Stone demonstrated a dose response relationship correlating D90 with probability of biochemical control and showed that a D90 of 140 Gy is a highly significant factor in predicting PSA relapse free survival (PSA-RFS). Although, a mean D90 of over 140 Gy was achieved in our series, there is nevertheless a normal distribution with 20% of patients achieving a D90 of less than 120 Gy. We have analysed the possible causes for the low D90 and the impact on outcome. PATIENTS AND METHODS: Prospective data from 667 patients treated between 1995 and 2001 by I-125 seeds prostate implant as monotherapy were analysed. Post-implant dosimetry was performed on 413 patients. D90 and other indices were calculated for each patient. Statistical analysis was performed on D90 dose to identify the correlation that would predict the 8.2 years PSA relapse free survival as defined by the American Society for Therapeutic Radiology and Oncology (ASTRO). RESULTS: Correlation between D90 and outcome shows no significant difference for the whole population between those who receive greater or less than 140 Gy (P=0.43) and there was also no difference for those receiving more or less than 130 Gy (P=0.14). Subgroup analysis by risk group, however, showed that for low risk patients there was a significant correlation between D90 and PSA control (P<0.01). Although, post-implant dosimetry was performed 6-8 weeks after brachytherapy, post-implant CT still showed variable levels of oedema compared with the pre-implant ultrasound. A statistically significant relationship was shown between D90 and the ratio between CT and ultrasound volume (P<0.01) which suggests that some low D90s may be related to persistent oedema at the time of calculation. Segmental analysis of a subgroup of 32 patients showed that the dose was most often deficient in the anterior basal segment of the gland. CONCLUSIONS: D90 was found to be a good discriminator for those with low risk where failure to achieve local control is likely to be the dominant cause of PSA failure. No significant dose response relationship between D90 and PSA was found in the intermediate and high-risk population of patients. This could be due to (1) the presence of oedema or discrepancy between pre- and post-implant volumes causing a low D90, (2) the possibility that the underdosed area could be situated where there is unlikely to be tumour, (3) the fact that biochemical control does not equate to local control because some patients fail outside the prostate, particularly in the high and intermediate risk patients, (4) if D90 is a good discriminator only for low risk patients, the absence of a dose response correlation in this series which contained 53.8% intermediate and high risk patients could be related to case mix.     

Management of low-grade gliomas of the optic nerve and chiasm

Thirty-six patients were evaluated between 1965 and 1983 for glioma of the optic nerves and/or chiasm. Median follow-up was 10.2 years. Pathologic verification was obtained in 32 patients. Tumor initially confined to the optic nerve recurred in one of five patients after complete resection. The actuarial survival for 25 patients irradiated for biopsy-proven glioma of the optic chiasm was 96%, 90%, and 90% at 5, 10, and 15 years, respectively, and the progression-free survival was 87% at 5, 10, and 15 years. Vision stabilized or improved in 86% of patients after radiotherapy. Patients irradiated to a dose greater than a NSD of 1385 ret had a significantly improved progression-free survival (P = 0.015). One serious complication occurred after a dose of 1533 ret. The recommended radiation dose for optic glioma is 45 to 50 Gy with 1.8 Gy fractions.