Dosimetric evaluation of synthetic CT for head and neck radiotherapy generated by a patch-based three-dimensional convolutional neural network.

PURPOSE: To develop and evaluate a patch-based convolutional neural network (CNN) to generate synthetic computed tomography (sCT) images for magnetic resonance (MR)-only workflow for radiotherapy of head and neck tumors. A patch-based deep learning method was chosen to improve robustness to abnormal anatomies caused by large tumors, surgical excisions, or dental artifacts. In this study, we evaluate whether the generated sCT images enable accurate MR-based dose calculations in the head and neck region. METHODS: We conducted a retrospective study on 34 patients with head and neck cancer who underwent both CT and MR imaging for radiotherapy treatment planning. To generate the sCTs, a large field-of-view T2-weighted Turbo Spin Echo MR sequence was used from the clinical protocol for multiple types of head and neck tumors. To align images as well as possible on a voxel-wise level, CT scans were nonrigidly registered to the MR (CTreg ). The CNN was based on a U-net architecture and consisted of 14 layers with 3 × 3 × 3 filters. Patches of 48 × 48 × 48 were randomly extracted and fed into the training. sCTs were created for all patients using threefold cross validation. For each patient, the clinical CT-based treatment plan was recalculated on sCT using Monaco TPS (Elekta). We evaluated mean absolute error (MAE) and mean error (ME) within the body contours and dice scores in air and bone mask. Also, dose differences and gamma pass rates between CT- and sCT-based plans inside the body contours were calculated. RESULTS: sCT generation took 4 min per patient. The MAE over the patient population of the sCT within the intersection of body contours was 75 ± 9 Hounsfield Units (HU) (±1 SD), and the ME was 9 ± 11 HU. Dice scores of the air and bone masks (CTreg vs sCT) were 0.79 ± 0.08 and 0.70 ± 0.07, respectively. Dosimetric analysis showed mean deviations of -0.03% ± 0.05% for dose within the body contours and -0.07% ± 0.22% inside the >90% dose volume. Dental artifacts obscuring the CT could be circumvented in the sCT by the CNN-based approach in combination with Turbo Spin Echo (TSE) magnetic resonance imaging (MRI) sequence that typically is less prone to susceptibility artifacts. CONCLUSIONS: The presented CNN generated sCTs from conventional MR images without adding scan time to the acquisition. Dosimetric evaluation suggests that dose calculations performed on the sCTs are accurate, and can therefore be used for MR-only radiotherapy treatment planning of the head and neck.

Unilateral versus bilateral irradiation in squamous cell head and neck cancer in relation to patient-rated xerostomia and sticky saliva.

BACKGROUND AND PURPOSE: To investigate the association between radiation technique with patient-rated moderate and severe xerostomia and sticky saliva. MATERIALS AND METHODS: One hundred and fifty patients treated with bilateral or unilateral irradiation for head and neck cancer were included. The salivary glands and the oral cavity were delineated on plannings-CT scans. Xerostomia and sticky saliva were assessed using the EORTC QLQ-H&N35 questionnaire at baseline and 6 and 12 months. RESULTS: At 6 months a significant association between radiation technique and the mean parotid dose (MD(parb)) and xerostomia was observed (Odds ratio (OR)-technique: 2.55; p=0.04 and OR-MD(parb): 1.04; p=0.009). Considering the individual salivary glands, only the mean dose in the contralateral parotid gland (MD(parcl)) is associated with xerostomia (OR: 1.04; p<0.0001). Moreover, the threshold dose for a 50% probability (P50) on xerostomia increased from 21 Gy with bilateral to 44 Gy for unilateral irradiation. CONCLUSIONS: Both technique and MD(parb) influence the risk of xerostomia in irradiated patients. Of all individual salivary glands, only MD(parcl) is of utmost importance for xerostomia. The shift in the P50 observed for xerostomia suggests that sparing of the contralateral parotid gland is compensated by hyperfunction of the contralateral parotid gland.

Quality-of-life after radiotherapy for advanced laryngeal cancer: Results of a phase III trial of the Dutch Head and Neck Society.

BACKGROUND/PURPOSE: To report on health-related quality-of-life (HRQoL) of patients with laryngeal cancer, treated in a randomized trial comparing accelerated radiotherapy with carbogen and nicotinamide (ARCON) against accelerated radiotherapy alone (AR). MATERIAL/METHODS: HRQoL was assessed using the HRQoL Questionnaire-C30 (QLQ-C30) and the Head & Neck cancer module (QLQ-H&N35) at baseline, at completion of radiotherapy and at 6, 12, and 24months post-baseline. RESULTS: From 269/345 patients (AR: 136/174; ARCON: 133/171) data on HRQoL were available for analysis. Moderate to severe clinical impact of the treatment was observed for nearly all items of the QLQ-C30 and QLQ-H&N35 between baseline and end-of-treatment. At 6months, scores returned to baseline level with exception of dry mouth, sticky saliva, and taste/smell. No difference between AR and ARCON was observed. At 2years from baseline, the percentage of patients reporting moderate to severe complaints of dry mouth, sticky saliva, or changes in taste/smell was 30%, 22% and 18%, respectively, while the majority of patients had no or few complaints of swallowing (79%) or speech (64%). CONCLUSIONS: With accelerated radiotherapy, high local tumor control was obtained while maintaining good speech and swallowing function. Long-term dry mouth, sticky saliva and changes in taste/smell are limited to one quarter of patients. (ClinicalTrials.gov number, NCT00147732).

Does radiation dose to the salivary glands and oral cavity predict patient-rated xerostomia and sticky saliva in head and neck cancer patients treated with curative radiotherapy?

BACKGROUND AND PURPOSE: To investigate the association between the mean salivary gland and oral cavity dose, with patient-rated moderate and severe xerostomia and sticky saliva. PATIENTS AND METHODS: One hundred and fifty-seven patients treated with bilateral irradiation for head and neck cancer were included. The parotid and submandibular glands and the oral cavity were delineated on plannings-CT scans. At baseline and 6 and 12 months self-reported xerostomia and sticky saliva were assessed using the EORTC QLQ-H&N35 questionnaire. RESULTS: At 6 months a significant association between the mean parotid (MD(par)) and mean submandibular dose (MD(subm)) and xerostomia was observed (OR - MD(par): 1.17; P=0.002 and OR - MD(subm): 1.08; P = 0.02). Between MD(par) and MD(subm), a significant interaction term was present. No significant association was found with the oral cavity dose. Xerostomia was reversible depending on MD(par) and MD(subm). Considering Sticky saliva, a significant association was found at 6 and 12 months with MD(subm) (OR: 1.03; P < 0.001). The P50 for sticky saliva increased with elapsing time. CONCLUSIONS: Both MD(par) and MD(subm) influence the risk of xerostomia in irradiated patients at 6 months. This probability as a function of the mean parotid dose significantly depended on the mean dose in the submandibular glands. Sticky saliva mainly depends on MD(subm).

Prospective evaluation of health-related quality of life in long-term oral and oropharyngeal cancer survivors and the perceived need for supportive care.

PURPOSE: To evaluate long-term changes in health related quality of life (HRQOL) in oral/oropharyngeal cancer survivors and their need for and use of supportive care. METHODS: Between 1999 and 2001, 80 advanced oral or oropharyngeal cancer patients treated with free-flap reconstruction and postoperative radiotherapy were included in a prospective study of whom 27 patients were long-term survivors (mean 9.2 years, range 8-11 years). The HRQOL of 26 patients (response rate 96%) was assessed with the EORTC QLQ-C30 and QLQ-H&N35 questionnaires at four points in time: pretreatment (baseline), and at 6 months, 12 months (short term) and 8-11 years (long-term) follow up. A study specific questionnaire was developed to evaluate the need for and use of supportive care (allied health services, peer contact, psychosocial care, and complementary care) and was completed at the period of treatment and at long-term follow up. RESULTS: A number of HRQOL domains worsened significantly (p < 0.01) in the long-term: emotional functioning, social functioning, swallowing, speech, taste/smell, dry mouth, sticky saliva and coughing assessed by the mixed effects statistical model. At time of treatment, the need for supportive care was the highest for a dental hygienist (77%), a physical therapist (73%), a speech therapist (42%), a dietician (38%), and a special diet (62%). At long-term follow up, the need for supportive care was limited to a dental hygienist (46%) and a physical therapist (23%). Only small differences were observed between the perceived need for and actual use of supportive care. CONCLUSION: A range of HRQOL domains in head and neck cancer survivors were deteriorated in the long-term compared to baseline and to the first year after treatment. At time of treatment and less frequently at long-term follow up, patients reported needing and using a variety of supportive care services.

External validation of three dimensional conformal radiotherapy based NTCP models for patient-rated xerostomia and sticky saliva among patients treated with intensity modulated radiotherapy.

PURPOSE: The purpose of this study was to investigate the ability of predictive models for patient-rated xerostomia (XER(6M)) and sticky saliva (STIC(6M)) at 6 months after completion of primary (chemo)radiation developed in head and neck cancer patients treated with 3D-conformal radiotherapy (3D-CRT) to predict outcome in patients treated with intensity modulated radiotherapy (IMRT). METHODS AND MATERIALS: Recently, we published the results of a prospective study on predictive models for patient-rated xerostomia and sticky saliva in head and neck cancer patients treated with 3D-CRT (3D-CRT based NTCP models). The 3D-CRT based model for XER(6M) consisted of three factors, including the mean parotid dose, age, and baseline xerostomia (none versus a bit). The 3D-CRT based model for STIC(6M) consisted of the mean submandibular dose, age, the mean sublingual dose, and baseline sticky saliva (none versus a bit). In the current study, a population consisting of 162 patients treated with IMRT was used to test the external validity of these 3D-CRT based models. External validity was described by the explained variation (R(2) Nagelkerke) and the Brier score. The discriminative abilities of the models were calculated using the area under the receiver operating curve (AUC) and calibration (i.e. the agreement between predicted and observed outcome) was assessed with the Hosmer-Lemeshow "goodness-of-fit" test. RESULTS: Overall model performance of the 3D-CRT based predictive models for XER(6M) and STIC(6M) was significantly worse in terms of the Brier score and R(2) Nagelkerke among patients treated with IMRT. Moreover the AUC for both 3D-CRT based models in the IMRT treated patients were markedly lower. The Hosmer-Lemeshow test showed a significant disagreement for both models between predicted risk and observed outcome. CONCLUSION: 3D-CRT based models for patient-rated xerostomia and sticky saliva among head and neck cancer patients treated with primary radiotherapy or chemoradiation turned out to be less valid for patients treated with IMRT. The main message from these findings is that models developed in a population treated with a specific technique cannot be generalised and extrapolated to a population treated with another technique without external validation.

Development of NTCP models for head and neck cancer patients treated with three-dimensional conformal radiotherapy for xerostomia and sticky saliva: the role of dosimetric and clinical factors.

PURPOSE: The purpose of this multicentre prospective study was to investigate the significance of the radiation dose in the major and minor salivary glands, and other pre-treatment and treatment factors, with regard to the development of patient-rated xerostomia and sticky saliva among head and neck cancer (HNC) patients treated with primary (chemo-) radiotherapy ((CH)RT). METHODS AND MATERIALS: The study population was composed of 167 consecutive HNC patients treated with three-dimensional conformal (3D-CRT) (CH) RT. The primary endpoint was moderate to severe xerostomia (XER6m) as assessed by the EORTC QLQ-H&N35 at 6 months after completing (CH)RT. The secondary endpoint was moderate to severe sticky saliva at 6 months (STIC6m). All organs at risk (OARs) potentially involved in salivary function were delineated on planning-CT, including the parotid, submandibular and sublingual glands and the minor glands in the soft palate, cheeks and lips. Patients with moderate to severe xerostomia or sticky saliva at baseline were excluded. The optimum number of variables for a multivariate logistic regression model was determined using a bootstrapping method. RESULTS: The multivariate analysis showed the mean parotid dose, age and baseline xerostomia (none versus a bit) to be the most important predictors for XER6m. The risk of developing xerostomia increased with age and was higher when minor baseline xerostomia was present in comparison with patients without any xerostomia complaints at baseline. Model performance was good with an area under the curve (AUC) of 0.82. For STIC6m, the mean submandibular dose, age, the mean sublingual dose and baseline sticky saliva (none versus a bit) were most predictive for sticky saliva. The risk of developing STIC6m increased with age and was higher when minor baseline sticky saliva was present in comparison with patients without any sticky saliva complaints at baseline. Model performance was good with an AUC of 0.84. CONCLUSIONS: Dose distributions in the minor salivary glands in patients receiving 3D-CRT have limited significance with regard to patient-rated symptoms related to salivary dysfunction. Besides the parotid and submandibular glands, only the sublingual glands were significantly associated with sticky saliva. In addition, reliable risk estimation also requires information from other factors such as age and baseline subjective scores. When these selected factors are included in predictive models, instead of only dose volume histogram parameters, model performance can be improved significantly.

NTCP models for patient-rated xerostomia and sticky saliva after treatment with intensity modulated radiotherapy for head and neck cancer: the role of dosimetric and clinical factors.

PURPOSE: The purpose of this multicentre prospective study was to develop multivariable logistic regression models to make valid predictions about the risk of moderate-to-severe patient-rated xerostomia (XER(M6)) and sticky saliva 6 months (STIC(M6)) after primary treatment with intensity modulated radiotherapy (IMRT) with or without chemotherapy for head and neck cancer (HNC). METHODS AND MATERIALS: The study population was composed of 178 consecutive HNC patients treated with IMRT. All patients were included in a standard follow up programme in which acute and late side effects and quality of life were prospectively assessed, prior to, during and after treatment. The primary endpoints were XER(M6) and STIC(M6) as assessed by the EORTC QLQ-H&N35 after completing IMRT. Organs at risk (OARs) potentially involved in salivary function were delineated on planning-CT, including the parotid, submandibular and sublingual glands and the minor glands in the soft palate, cheeks and lips. Patients with moderate-to-severe xerostomia or sticky saliva, respectively, at baseline were excluded. The optimal number of variables for a multivariate logistic regression model was determined using a bootstrapping method. RESULTS: Eventually, 51.6% of the cases suffered from XER(M6). The multivariate analysis showed that the mean contralateral parotid gland dose and baseline xerostomia (none vs. a bit) were the most important predictors for XER(M6). For the multivariate NTCP model, the area under the receiver operating curve (AUC) was 0.68 (95% CI 0.60-0.76) and the discrimination slope was 0.10, respectively. Calibration was good with a calibration slope of 1.0. At 6 months after IMRT, 35.6% of the cases reported STIC(M6). The mean contralateral submandibular gland dose, the mean sublingual dose and the mean dose to the minor salivary glands located in the soft palate were most predictive for STIC(M6). For this model, the AUC was 0.70 (95% CI 0.61-0.78) and the discrimination slope was 0.12. Calibration was good with a calibration slope of 1.0. CONCLUSIONS: The multivariable NTCP models presented in this paper can be used to predict patient-rated xerostomia and sticky saliva. The dose volume parameters included in the models can be used to further optimise IMRT treatment.

Radiotherapy on the neck nodes predicts severe weight loss in patients with early stage laryngeal cancer.

BACKGROUND AND PURPOSE: Although patients with early stage (T1/T2) laryngeal cancer (LC) are thought to have a low incidence of malnutrition, severe weight loss is observed in a subgroup of these patients during radiotherapy (RT). The objective of this study was to evaluate weight loss and nutrition-related symptoms in patients with T1/T2 LC during RT and to select predictive factors for early identification of malnourished patients. METHODS: Of all patients with T1/T2 LC, who received primary RT between 1999 and 2007, the following characteristics were recorded: sex, age, TNM classification, tumour location, radiation schedule, performance status, quality of life, weight loss, and nutrition-related symptoms. The association between baseline characteristics and malnutrition (>5% weight loss during RT) was investigated by Cox regression analysis. RESULTS: The study population consisted of 238 patients. During RT, 44% of patients developed malnutrition. Tumour location, TNM classification, RT on the neck nodes, RT dose, nausea/vomiting, pain, swallowing, senses problems, trouble with social eating, dry mouth and the use of painkillers were all significantly associated with malnutrition. In the multivariate analysis, RTs on both the neck nodes (HR 4.16, 95% CI 2.62-6.60) and dry mouth (HR 1.72, 95% CI 1.14-2.60) remained predictive. Nevertheless, RT on the neck nodes alone resulted in the best predictive model for malnutrition scores. CONCLUSIONS: Patients with early stage laryngeal cancer are at risk of malnutrition during radiotherapy. Radiotherapy on the neck nodes is the best predictor of malnutrition during radiotherapy. Therefore, we suggest to offer nutritional counselling to all the patients who receive nodal irradiation.

Ipsilateral irradiation for oral and oropharyngeal carcinoma treated with primary surgery and postoperative radiotherapy.

PURPOSE: The purpose was to evaluate the contralateral nodal control (CLNC) in postoperative patients with oral and oropharyngeal cancer treated with ipsilateral irradiation of the neck and primary site. Late radiation-induced morbidity was also evaluated. METHODS AND MATERIALS: The study included 123 patients with well-lateralized squamous cell carcinomas treated with surgery and unilateral postoperative irradiation. Most patients had tumors of the gingiva (41%) or buccal mucosa (21%). The majority of patients underwent surgery of the ipsilateral neck (n = 102 [83%]). The N classification was N0 in 73 cases (59%), N1 or N2a in 23 (19%), and N2b in 27 cases (22%). RESULTS: Contralateral metastases developed in 7 patients (6%). The 5-year actuarial CLNC was 92%. The number of lymph node metastases was the only significant prognostic factor with regard to CLNC. The 5-year CLNC was 99% in N0 cases, 88% in N1 or N2a cases, and 73% in N2b cases (p = 0.008). Borderline significance (p = 0.06) was found for extranodal spread. Successful salvage could be performed in 71% of patients with contralateral metastases. The prevalence of Grade 2 or higher xerostomia was 2.6% at 5 years. CONCLUSIONS: Selected patients with oral or oropharyngeal carcinoma treated with primary surgery and postoperative ipsilateral radiotherapy have a very high CLNC with a high probability of successful salvage in case of contralateral metastases. However, bilateral irradiation should be applied in case of multiple lymph node metastases in the ipsilateral neck, particularly in the presence of extranodal spread. The incidence of radiation-induced morbidity is considerably lower as observed after bilateral irradiation.