Search

Purpose: The purpose of the present study was the optimization of dose calculation for intracranial stereotactic radiosurgery., Methods: We evaluated the Singular Value Decomposition (SVD) analysis as a novel optimization technique. Our approach to dose distribution optimization was to recover estimates of minibeams weights from well-defined provisional dose matrices. The dose delivered by a set of minibeams is formulated as a matrix equation. We studied the influence of dose matrix and voxel dimensions on the conditioning of stereotactic radiotherapy inverse problem. Dose matrix dimensions varied from 16 to 96 mm3, while voxel dimension was kept constant at 2 mm 3 . In the assessment of voxel dimension, matrix dimension was kept constant at 80 mm 3 while voxel dimensions varied from 1 to 8 mm 3 . The reconstruction of dose distributions was studied using a truncated SVD expansion in the calculation of approximation to the generalised matrix inverse., Results: The conditioning was deteriorated by either the decreasing of dose matrix dimensions or by the increasing of voxel size. The condition number was equal to 89×103 and 7×103 for the 16 mm3 and 96 mm3 dose matrix dimensions, respectively. The condition number was equal to 9.9×103 and to 2.7×106 for 1 mm3 and 8 mm3 voxel size, respectively. The reconstruction of dose distributions revealed that an ill-conditioned problem yields poor quality reconstruction., Conclusion: We considered that a good compromise between quality of dose distribution, time calculation and hard disk memory would be the use of a 64 mm 3 matrix dimension with a 2 mm 3 voxel size.

Pediatric posterior fossa tumor (PFT) survivors who have been treated with cranial radiation therapy often suffer from cognitive impairments that might relate to IQ decline. Radiotherapy (RT) distinctly affects brain regions involved in different cognitive functions. However, the relative contribution of regional irradiation to the different cognitive impairments still remains unclear. We investigated the relationships between the changes in different cognitive scores and radiation dose distribution in 30 children treated for a PFT. Our exploratory analysis was based on a principal component analysis (PCA) and an ordinary least square regression approach. The use of a PCA was an innovative way to cluster correlated irradiated regions due to similar radiation therapy protocols across patients. Our results suggest an association between working memory decline and a high dose (equivalent uniform dose, EUD) delivered to the orbitofrontal regions, whereas the decline of processing speed seemed more related to EUD in the temporal lobes and posterior fossa. To identify regional effects of RT on cognitive functions may help to propose a rehabilitation program adapted to the risk of cognitive impairment.

Purpose: To evaluate the usefulness of reporting the point A dose in patients with locally advanced cervical cancer treated with image-guided adaptive brachytherapy (IGABT)., Methods and Materials: Dosimetric data from patients treated with a combination of chemoradiation and intracavitary IGABT were examined in light of their outcomes. Prescribing followed the Groupe Européen de Curiethérapie-European Society for Radiation Oncology recommendations. All doses were converted in 2-Gy equivalent. The relationships between the D90 high-risk clinical target volume (CTV HR ) and intermediate-risk clinical target volume (CTV IR ) and point A doses were studied. Dose-effect relationships based on the probit model and log-rank test were assessed., Results: Two hundred twelve patients were included with a median followup of 53.0 months. A total of 28 local relapses were reported, resulting in a local control rate of 86.6% at 3 years. Mean D90 CTV HR , CTV IR , and point A doses were: 79.7 ± 10.4 Gy, 67.4 ± 5.8 Gy, and 66.4 ± 5.6 Gy, respectively. The mean D90 were significantly different and independent from the mean point A dose, even in bulky tumors at diagnosis or in large CTV HR lesions. Point A dose appeared correlated with TRAK, and finally with the D90 CTV HR through a complex formula including the CTV HR volume (R 2  = 0.55). Whereas significant relationships between the probability of achieving local control and the D90 CTV HR and CTV IR (p = 0.08 and 0.025) were observed, no similar relationship was found with point A dose except a trend of an inverse relation. After sorting patients according to three dose levels, highest local control rates were observed in patients with D90 CTV HR ≥85 Gy, whereas those with point A doses ≥70 Gy had the worst outcomes., Conclusions: In patients treated with IGABT, point A dose is not predictive of local control but a surrogate of the irradiated volume. Its relationships with the D90 CTV HR are indirect and complex rising the question of relevance of its reporting in routine., (Copyright © 2017 American Brachytherapy Society. Published by Elsevier Inc. All rights reserved.)

Proper understanding of the risk of radiation-induced late effects for patients receiving external photon beam radiotherapy requires the determination of reliable dose-response relationships. Although significant efforts have been devoted to improving dose estimates for the study of late effects, the most often questioned explanatory variable is still the dose. In this work, based on a literature review, we provide an in-depth description of the radiotherapy dose reconstruction process for the study of late effects. In particular, we focus on the identification of the main sources of dose uncertainty involved in this process and summarise their impacts on the dose-response relationship for radiotherapy late effects. We provide a number of recommendations for making progress in estimating the uncertainties in current studies of radiotherapy late effects and reducing these uncertainties in future studies.

Purpose: Image-guided adaptive brachytherapy is a high precision technique that allows dose escalation and adaptation to tumor response. Two monocentric studies reported continuous dose-volume response relationships, however, burdened by large confidence intervals. The aim was to refine these estimations by performing a meta-regression analysis based on published series., Methods and Materials: Eligibility was limited to series reporting dosimetric parameters according to the Groupe Européen de Curiethérapie-European SocieTy for Radiation Oncology recommendations. The local control rates reported at 2-3 years were confronted to the mean D90 clinical target volume (CTV) in 2-Gy equivalent using the probit model. The impact of each series on the relationships was pondered according to the number of patients reported., Results: An exhaustive literature search retrieved 13 series reporting on 1299 patients. D90 high-risk CTV ranged from 70.9 to 93.1 Gy. The probit model showed a significant correlation between the D90 and the probability of achieving local control (p < 0.0001). The D90 associated to a 90% probability of achieving local control was 81.4 Gy (78.3-83.8 Gy). The planning aim of 90 Gy corresponded to a 95.0% probability (92.8-96.3%). For the intermediate-risk CTV, less data were available, with 873 patients from eight institutions. Reported mean D90 intermediate-risk CTV ranged from 61.7 to 69.1 Gy. A significant dose-volume effect was observed (p = 0.009). The D90 of 60 Gy was associated to a 79.4% (60.2-86.0%) local control probability., Conclusion: Based on published data from a high number of patients, significant dose-volume effect relationships were confirmed and refined between the D90 of both CTV and the probability of achieving local control. Further studies based on individual data are required to develop nomograms including nondosimetric prognostic criteria., (Copyright © 2016 American Brachytherapy Society. Published by Elsevier Inc. All rights reserved.)

Purpose: Although dose-volume parameters in image-guided brachytherapy have become a standard, the use of posterior-inferior border of the pubic symphysis (PIBS) points has been recently proposed in the reporting of vaginal doses. The aim was to evaluate their pertinence., Methods and Materials: Nineteen patients who received image-guided brachytherapy after concurrent radiochemotherapy were included. Per treatment, CT scans were performed at Days 2 and 3, with reporting of the initial dwell positions and times. Doses delivered to the PIBS points were evaluated on each plan, considering that they were representative of one-third of the treatment. The movements of the applicator according to the PIBS point were analysed., Results: Mean prescribed doses at PIBS -2, PIBS, PIBS +2 were, respectively, 2.23 ± 1.4, 6.39 ± 6.6, and 31.85 ± 36.06 Gy. Significant differences were observed between the 5 patients with vaginal involvement and the remaining 14 at the level of PIBS +2 and PIBS: +47.60 Gy and +7.46 Gy, respectively (p = 0.023 and 0.03). The variations between delivered and prescribed doses at PIBS points were not significant. However, at International commission on radiation units and measurements rectovaginal point, the delivered dose was decreased by 1.43 ± 2.49 Gy from the planned dose (p = 0.019). The delivered doses at the four points were strongly correlated with the prescribed doses with R(2) ranging from 0.93 to 0.95. The movements of the applicator in regard of the PIBS point assessed with the Digital Imaging and Communications in Medicine coordinates were insignificant., Conclusion: The doses evaluated at PIBS points are not impacted by intrafractional movements. PIBS and PIBS +2 dose points allow distinguishing the plans of patients with vaginal infiltration. Further studies are needed to correlate these parameters with vaginal morbidity., (Copyright © 2016 American Brachytherapy Society. Published by Elsevier Inc. All rights reserved.)

The purpose of this investigation was to evaluate the dose on megavoltage CT (MVCT) images required for tomotherapy. As imaging possibilities are often used before each treatment and usually used several times before the session, we tried to evaluate the dose delivered during the procedure. For each scanning mode (fine, normal, and coarse), we first established the relative variation of these doses according to different technical parameters (explored length, patient setup). These dose variations measured with the TomoPhant, also known as Cheese phantom, showed the expected variations (due to the variation of scattered radiation) of 15% according to the explored length and ± 5% according to the phantom setup (due to the variation of the point of measurement in the bore). In order to estimate patient doses, an anthropomorphic phantom was used for thermoluminescent and film dosimetry. The degree of agreement between the two methods was very satisfactory (the differences correspond to 5 mGy per imaging session) for the three sites studied (head & neck, thorax, and abdomen). These measurements allowed us to estimate the delivered dose of between 1 cGy and 4 cGy according to the site and imaging mode. Finally, we attempted to investigate a way to calculate this delivered dose in our patients from the study conducted on a cylindrical phantom and by taking into account data from the initial kV-CT scan. The results we obtained were close to our measurements, with discrepancies below 5 mGy per MVCT.

The purpose of this work was to experimentally investigate the out-of-field dose in a water phantom, with several high energy electron beams used in external beam radiotherapy (RT). The study was carried out for 6, 9, 12, and 18 MeV electron beams, on three different linear accelerators, each equipped with a specific applicator. Measurements were performed in a water phantom, at different depths, for different applicator sizes, and off-axis distances up to 70 cm from beam central axis (CAX). Thermoluminescent powder dosimeters (TLD-700) were used. For given cases, TLD measurements were compared to EBT3 films and parallel-plane ionization chamber measurements. Also, out-of-field doses at 10 cm depth, with and without applicator, were evaluated. With the Siemens applicators, a peak dose appears at about 12-15 cm out of the field edge, at 1 cm depth, for all field sizes and energies. For the Siemens Primus, with a 10 × 10 cm(²) applicator, this peak reaches 2.3%, 1%, 0.9% and 1.3% of the maximum central axis dose (Dmax) for 6, 9, 12 and 18 MeV electron beams, respectively. For the Siemens Oncor, with a 10 × 10 cm(²) applicator, this peak dose reaches 0.8%, 1%, 1.4%, and 1.6% of Dmax for 6, 9, 12, and 14 MeV, respectively, and these values increase with applicator size. For the Varian 2300C/D, the doses at 12.5 cm out of the field edge are 0.3%, 0.6%, 0.5%, and 1.1% of Dmax for 6, 9, 12, and 18 MeV, respectively, and increase with applicator size. No peak dose is evidenced for the Varian applicator for these energies. In summary, the out-of-field dose from electron beams increases with the beam energy and the applicator size, and decreases with the distance from the beam central axis and the depth in water. It also considerably depends on the applicator types. Our results can be of interest for the dose estimations delivered in healthy tissues outside the treatment field for the RT patient, as well as in studies exploring RT long-term effects.

Context: Thyroid carcinoma is a frequent complication of childhood cancer radiotherapy. The dose response to thyroid radiation dose is now well established, but the potential modifier effect of other factors requires additional investigation., Objective: This study aimed to investigate the role of potential modifiers of the dose response., Design: We followed a cohort of 4338 5-year survivors of solid childhood cancer treated before 1986 over an average of 27 years. The dose received by the thyroid gland and some other anatomical sites during radiotherapy was estimated after reconstruction of the actual conditions in which irradiation was delivered., Results: Fifty-five patients developed thyroid carcinoma. The risk of thyroid carcinoma increased with a radiation dose to the thyroid of up to two tenths of Gy, then leveled off for higher doses. When taking into account the thyroid radiation dose, a surgical or radiological splenectomy (>20 Gy to the spleen) increased thyroid cancer risk (relative risk [RR] = 2.3; 95% confidence interval [CI], 1.3-4.0), high radiation doses (>5 Gy) to pituitary gland lowered this risk (RR = 0.2; 95% CI, 0.1-0.6). Patients who received nitrosourea chemotherapy had a 6.6-fold (95% CI, 2.5-15.7) higher risk than those who did not. The excess RR per Gy of radiation to the thyroid was 4.7 (95% CI, 1.7-22.6). It was 7.6 (95% CI, 1.6-33.3) if body mass index at time of interview was equal or higher than 25 kg/m(2), and 4.1 (95% CI, 0.9-17.7) if not (P for interaction = .1)., Conclusion: Predicting thyroid cancer risk following childhood cancer radiation therapy probably requires the assessment of more than just the radiation dose to the thyroid. Chemotherapy, splenectomy, radiation dose to pituitary gland, and obesity also play a role.

Objective: To report the outcomes and late toxicities of patients with locally advanced cervical cancer treated with concomitant chemoradiation (CRT) followed by intracavitary image-guided adaptive brachytherapy (IGABT)., Methods: Data from consecutive patients with histologically proven stage IB-IVA cervical cancer treated with curative intent in a single institution were analyzed. After pelvic +/- para-aortic external-beam radiation therapy, they received pulsed-dose rate IGABT following GEC-ESTRO recommendations., Results: Two hundred and twenty-five patients were enrolled. Sixty-five percent were stage≥IIB according to FIGO classification. Ninety-five percent received CRT. Mean D90 to HR and IR-CTV were 80.4+/-10.3Gy and 67.7+/-6.1Gy. After a median follow-up of 38.8months, 3-year local control and overall survival rates were 86.4% and 76.1%, respectively. A trend for a detrimental effect of tumor stage on local control rates was observed with 3-year local control rates of 100% for stages IB1 and IIA, 90.5 for IB2, 85.8% for IIB, 50% for IIIA, 77.1 for IIIB, and 66.7% for IVA tumors (p=0.06). Local control rates at 3years were 95.6% in the group of patients with D90 of HR-CTV≥85Gy, 88.8% in those with D90 between 80 and 85Gy, and 80% when D90<80Gy (p=0.018). Eighteen severe late gastrointestinal and urinary effects affecting 14 patients were reported corresponding with a crude incidence of 6.6%., Conclusions: CRT followed by IGABT provides high local control rates with limited toxicity. Reaching high doses is mandatory to achieve local control and interstitial brachytherapy is necessary in advanced diseases., (Copyright © 2015 Elsevier Inc. All rights reserved.)

Purpose: To investigate the roles of radiation therapy and chemotherapy in the occurrence of subsequent leukemia after childhood cancer., Methods and Materials: We analyzed data from a case-control study with 35 cases and 140 controls. The active bone marrow (ABM) was segmented into 19 compartments, and the radiation dose was estimated in each. The chemotherapy drug doses were also estimated to enable adjustments. Models capable of accounting for radiation dose heterogeneity were implemented for analysis., Results: Univariate analysis showed a significant trend in the increase of secondary leukemia risk with radiation dose, after accounting for dose heterogeneity (P=.046). This trend became nonsignificant after adjustment for doses of epipodophyllotoxins, alkylating agents, and platinum compounds and the first cancer on multivariate analysis (P=.388). The role of the radiation dose appeared to be dwarfed, mostly by the alkylating agents (odds ratio 6.9, 95% confidence interval 1.9-25.0). Among the patients who have received >16 Gy to the ABM, the radiogenic risk of secondary leukemia was about 4 times greater in the subgroup with no alkylating agents than in the subgroup receiving ≥10 g/m(2)., Conclusions: Notwithstanding the limitations resulting from the size of our study population and the quite systematic co-treatment with chemotherapy, the use of detailed information on the radiation dose distribution to ABM enabled consideration of the role of radiation therapy in secondary leukemia induction after childhood cancer., (Copyright © 2015 Elsevier Inc. All rights reserved.)

Purpose: To establish dose-volume effect correlations for late bladder and rectum side effects in patients treated for locally advanced cervical cancer with concomitant chemoradiation followed by pulsed-dose rate image-guided adaptive brachytherapy., Material and Methods: The dosimetric data, converted in 2 Gy equivalent, from 217 patients were confronted to late morbidity defined as any event lasting or occurring 90 days after treatment initiation. Toxicity was assessed using the CTC-AE 3.0. Probit analyses and Log rank tests were performed to assess relationships., Results: One hundred and sixty-one urinary and 58 rectal events were reported, affecting 98 (45.1%) and 51 (23.5%) patients, respectively. Cumulative incidences for grade 2-4 bladder and rectal morbidity were 24.3% and 9.6% at 3 years, respectively. Significant relationships were observed between grade 2-4 and 3-4 events and D0.1cm(3) and D2 cm(3) for the bladder and between grade 1-4 and 2-4 event probability and rectal D2 cm(3). The effective doses for 10% grade 2-4 morbidity were 65.3 Gy (59.8-81.3), and 55.4 Gy (15.7-63.6), respectively, for the rectum and bladder. Without considering urinary and rectal incontinence, for which the pertinence of correlating them with D2 cm(3) is questionable, ED10 were 68.5 Gy (62.9-110.6) and 65.5 Gy (51.4-71.6 Gy). When sorting patients according to D2 cm(3) levels, patients with high D2 cm(3) had significantly lower morbidity free survival rates for grade 1-4 and 2-4 urinary and rectal morbidity., Conclusion: Significant dose-volume effect relationships were demonstrated between the modern dosimetric parameters and the occurrence of late rectal and urinary morbidity in patients treated with pulsed-dose-rate brachytherapy. Further studies are required to refine these relationships according to clinical cofactors, such as comorbidities., (Copyright © 2015 Elsevier Ireland Ltd. All rights reserved.)

Purpose: To report the application of the global risk analysis (GRA) in the pulsed-dose rate (PDR) brachytherapy workflow., Material and Methods: Analyses were led by a multidisciplinary working group established within the unit with the guidance of a quality engineer. First, a mapping of hazardous situations was developed as a result of interactions between the patient workflow for a treatment using PDR brachytherapy split into 51 sub-phases with a comprehensive list of the hazards that he/she faces (44). Interactions, when relevant, were sorted by level of priority: to be treated immediately, secondarily (the group is not entitled to treat the situation), or later (safe situation). Secondly, for each high priority dangerous situation, scenarios were developed to anticipate their potential consequences. Criticality was assessed, using likelihood and severity scales and a matrix, which allocated risks into categories: acceptable (C1), tolerable under control (C2) and unacceptable (C3). Then, corrective actions were proposed and planned when relevant, after assessment of their feasibility with a scale of effort. Finally, the criticality of the scenarios was reevaluated, taking into account the implementation of these actions, leading to a residual risk mapping, which could trigger additional proposals of actions., Results: Two thousand one hundred and eighty-four potential interactions between the list of hazards and the workflow were analyzed. Mapping of dangerous situations identified 213 relevant interactions, from which 61 were considered with high priority. One hundred and twenty-six scenarios were generated: 68 with a low criticality (74.3%), 58 with an intermediate score (25.7%). No scenario with the highest criticality was individualized. Twenty-one corrective actions were planned. Mapping of residual risk resulted in the disappearance of most C2 risks, leaving 5 C2 scenarios (4%), for which four monitoring indicators were implemented in addition to the corrected actions decided on., Conclusion: The implementation of the GRA appeared feasible, and led to implement 21 corrective actions, based on scenarios and not on incidents., (Copyright © 2015 Société française de radiothérapie oncologique (SFRO). Published by Elsevier SAS. All rights reserved.)

The purpose of this study was to investigate the suitability of a graphics library based model for the assessment of linear accelerator radiation leakage. Transmission through the shielding elements was evaluated using the build-up factor corrected exponential attenuation law and the contribution from the electron guide was estimated using the approximation of a linear isotropic radioactive source. Model parameters were estimated by a fitting series of thermoluminescent dosimeter leakage measurements, achieved up to 100 cm from the beam central axis along three directions. The distribution of leakage data at the patient plane reflected the architecture of the shielding elements. Thus, the maximum leakage dose was found under the collimator when only one jaw shielded the primary beam and was about 0.08% of the dose at isocentre. Overall, we observe that the main contributor to leakage dose according to our model was the electron beam guide. Concerning the discrepancies between the measurements used to calibrate the model and the calculations from the model, the average difference was about 7%. Finally, graphics library modelling is a readily and suitable way to estimate leakage dose distribution on a personal computer. Such data could be useful for dosimetric evaluations in late effect studies.

Purpose: To evaluate the efficiency and potential implications of the lowest dose evaluated in the maximally exposed 2cm(3) of the bladder/dose evaluated at the International Commission for Radiation Units and Measurements (ICRU) bladder point (D2cm(3)/DICRU) ratio as surrogate to locate the D2cm(3) in patients treated with MRI-guided adaptive brachytherapy for cervical cancer., Methods and Materials: The D2cm(3) area of the bladder was located in 69 patients, using the Digital Imaging and Communications in Medicine coordinates of its barycenter, with respect to the ICRU bladder point. The D2cm(3)/DICRU ratio was correlated with the longitudinal coordinate of the D2cm(3). Afterward, the ratio was used in a retrospective cohort of 216 patients to evaluate its impact in dose-effect analyses for late urinary incontinence., Results: The mean position of the D2cm(3) was 1.73±0.98 cm cranially, 0.59±0.65 cm backwardly, and 0.02±0.89 cm to the right of the ICRU point. It was located above the ICRU point in 95.7% of the patients. Its position was lower in patients with vaginal involvement at diagnosis (p=0.03). The D2cm(3)/DICRU ratio was correlated with the position of the D2cm(3) (R²=0.716, p<10(-6)). In speculating that a ratio greater than one would predict a D2cm(3) located above the ICRU point, the sensibility, specificity, positive, and negative predictive values were 95.2%, 100%, 100%, and 66.8%, respectively. Among the retrospective cohort, 85 patients had a ratio lower than 1.1, reflecting a D2cm(3) located in the lower bladder. In these patients, analyses showed significant dose relationship with Grade 2-4 incontinence (p=0.017), whereas no correlation was demonstrated in the remaining patients., Conclusions: The D2cm(3)/DICRU ratio is a relevant surrogate to estimate the localization of the D2cm(3). Significant dose-effect correlations for incontinence were established in patients with low values for this ratio., (Copyright © 2015 American Brachytherapy Society. Published by Elsevier Inc. All rights reserved.)

Purpose: To prospectively evaluate the intrafractional movements of organs at risk (OARs) and their dosimetric impact during the delivery of pulsed-dose-rate brachytherapy in cervical cancer., Patients and Methods: An MRI on Day 1 was used for treatment planning in 19 patients. CT scans were acquired at Days 1, 2, and 3 with delineation of the OARs. The MRI plan was transferred to each CT. The intersection volume between the 10 Gy isodose and the OARs were monitored, reflecting movement. Lower dose evaluated in the maximally exposed 0.1 cm(3) of an organ and lower dose evaluated in the maximally exposed 2 cm(3) of an organ (D(2cm3)) were evaluated on each CT and compared. Results were averaged considering that each CT reflected one-third of the treatment course to evaluate the delivered dose., Results: No major movements of the sigmoid and bladder were observed, whereas the rectum got significantly closer to the implant at Day 2. The consequence was an increase of 6% ± 5.3 (3.7 Gy, α/β = 3 Gy) of the delivered D(2cm3) from the planned dose, in contrast to 0.2% ± 6.1 for the bladder and 1.1% ± 6.4 for the sigmoid. The increase of the D(2cm3) of the rectum was reported in 17 patients, ranging from 0.4 to 9.4 Gy, leading to a 10.5% overcoming of the dose constraint (75 Gy). Similar tendencies were reported for lower dose evaluated in the maximally exposed 0.1 cm(3) of an organ., Conclusions: A significant systematic variation was observed for the rectum (+3.7 Gy). As significant random variations were observed, caution should be exercised when the planned D(2cm3) is close to the dose constraints., (Copyright © 2015 American Brachytherapy Society. Published by Elsevier Inc. All rights reserved.)

Purpose: To report the prognostic factors for local control in patients treated for locally advanced cervical cancer with image guided pulsed-dose rate brachytherapy., Materials/methods: Patients treated with curative intent by a combination of external beam radiotherapy and pulsed-dose rate brachytherapy were selected. Local failure was defined as any relapse in the cervix, vagina, parametria, or uterus during follow-up. Prognostic factors were selected based on log rank tests and then analyzed with a Cox model. Dose/effect correlations were performed using the probit model., Results: Two hundred and twenty-five patients treated from 2006 to 2011 were included. According to the FIGO classification, 29% were stage IB, 58% stage II, 10% stage III, and 3% stage IVA; 95% received concomitant chemotherapy. Thirty patients were considered having incomplete response or local failure. Among the selected parameters, D90 for HR-CTV, D90 for IR-CTV, the overall treatment time, the TRAK, and the HR-CTV volume appeared significantly correlated with local control in univariate analysis. In multivariate analysis, overall treatment time >55days and HR-CTV volume >30cm(3) appeared as independent. The probit analysis showed significant correlations between the D90 for both CTVs, and the probability of achieving local control (p=0.008 and 0.024). The thresholds to reach to warrant a probability of 90% of local control were 85Gy to the D90 of the HR-CTV and 75Gy to 90% of the IR-CTV (in 2Gy equivalent, α/β=10). To warrant the same local control rate, the D90 HR-CTV should be significantly increased in stage III-IV tumors, in case of HR-CTV >30cm(3), excessive treatment time, or tumor width at diagnosis >5cm (97, 92, 105, and 92Gy respectively)., Conclusions: Overall treatment time and HR-CTV volume were independent prognostic factors for local control. The D90 for HR and IR CTV were significantly correlated with local control, and D90 HR-CTV should be adapted to clinical criteria., (Copyright © 2014 Elsevier Ireland Ltd. All rights reserved.)