Cost-effectiveness of weekly adaptive radiotherapy versus standard IMRT in head and neck cancer alongside the ARTIX trial.

BACKGROUND AND PURPOSE: We performed a cost-effectiveness analysis (CEA) comparing an adaptive radiotherapy (ART) strategy, based on weekly replanning, aiming to correct the parotid gland overdose during treatment and expecting therefore to decrease xerostomia, when compared to a standard IMRT. MATERIALS AND METHODS: We conducted the ARTIX trial, a randomized, parallel-group, multicentric study comparing a systematic weekly replanning ART to a standard IMRT. The primary endpoint was the frequency of xerostomia at 12 months, measured by stimulating salivary flow with paraffin. The CEA was designed alongside the ARTIX trial which was linked to the French national health data system (SNDS). For each patient, healthcare consumptions and costs were provided by the SNDS. The reference case analysis was based on the primary endpoint of the trial. Sensitivity and scenario analyses were performed. RESULTS: Of the 129 patients randomly assigned between 2013 and 2018, only 2 records were not linked to the SNDS, which provides a linkage proportion of 98.4%. All of the other 127 records were linked with good to very good robustness. On the intent-to-treat population at 12 months, mean total costs per patient were €41,564 (SD 23,624) and €33,063 (SD 16,886) for ART and standard IMRT arms, respectively (p = 0.033). Incremental cost effectiveness ratio (ICER) was €162,444 per xerostomia avoided. At 24 months, ICER was €194,521 per xerostomia avoided. For both progression-free and overall survival, ART was dominated by standard IMRT. CONCLUSION: The ART strategy was deemed to be not cost-effective compared with standard IMRT for patients with locally advanced oropharyngeal cancer.

The dosimetric parameters impact on local recurrence in stereotactic radiotherapy for brain metastases.

OBJECTIVES: Stereotactic radiotherapy (SRT) for brain metastases (BM) allows very good local control (LC). However, approximately 20%-30% of these lesions will recur. The objective of this retrospective study was to evaluate the impact of dosimetric parameters on LC in cerebral SRT. METHODS: Patients treated with SRT for 1-3 BM between January 2015 and December 2018 were retrospectively included. A total of 349 patients with 538 lesions were included. The median gross tumour volume (GTV) was 2 cm3 (IQR, 0-7). The median biological effective dose with α/ß = 10 (BED10) was 60 Gy (IQR, 32-82). The median prescription isodose was 71% (IQR, 70-80). Correlations with LC were examined using the Cox regression model. RESULTS: The median follow-up period was 55 months (min-max, 7-85). Median overall survival was 17.8 months (IQR, 15.2-21.9). There were 95 recurrences and LC at 1 and 2 years was 87.1% (95% CI, 84-90) and 78.1% (95% CI, 73.9-82.4), respectively. Univariate analysis showed that systemic treatment, dose to 2% and 50% of the planning target volume (PTV), BED10 > 50 Gy, and low PTV and GTV volume were significantly correlated with better LC. In the multivariate analysis, GTV volume, isodose, and BED10 were significantly associated with LC. CONCLUSION: These results show the importance of a BED10 > 50 Gy associated with a prescription isodose <80% to optimize LC during SRT for BM. ADVANCES IN KNOWLEDGE: Isodose, BED, and GTV volume were significantly associated with LC. A low isodose improves LC without increasing the risk of radionecrosis.

Computed tomography synthesis from magnetic resonance imaging using cycle Generative Adversarial Networks with multicenter learning.

Background and Purpose: Addressing the need for accurate dose calculation in MRI-only radiotherapy, the generation of synthetic Computed Tomography (sCT) from MRI has emerged. Deep learning (DL) techniques, have shown promising results in achieving high sCT accuracies. However, existing sCT synthesis methods are often center-specific, posing a challenge to their generalizability. To overcome this limitation, recent studies have proposed approaches, such as multicenter training . Material and methods: The purpose of this work was to propose a multicenter sCT synthesis by DL, using a 2D cycle-GAN on 128 prostate cancer patients, from four different centers. Four cases were compared: monocenter cases, monocenter training and test on another center, multicenter trainings and a test on a center not included in the training and multicenter trainings with an included center in the test. Trainings were performed using 20 patients. sCT accuracy evaluation was performed using Mean Absolute Error, Mean Error and Peak-Signal-to-Noise-Ratio. Dose accuracy was assessed with gamma index and Dose Volume Histogram comparison. Results: Qualitative, quantitative and dose results show that the accuracy of sCTs for monocenter trainings and multicenter trainings using a seen center in the test did not differ significantly. However, when the test involved an unseen center, the sCT quality was inferior. Conclusions: The aim of this work was to propose generalizable multicenter training for MR-to-CT synthesis. It was shown that only a few data from one center included in the training cohort allows sCT accuracy equivalent to a monocenter study.

Weekly Adaptive Radiotherapy vs Standard Intensity-Modulated Radiotherapy for Improving Salivary Function in Patients With Head and Neck Cancer: A Phase 3 Randomized Clinical Trial.

Importance: Xerostomia is a major toxic effect associated with intensity-modulated radiotherapy (IMRT) for oropharyngeal cancers. Objective: To assess whether adaptive radiotherapy (ART) improves salivary function compared with IMRT in patients with head and neck cancer. Design, Setting, and Participants: This phase 3 randomized clinical trial was conducted in 11 French centers. Patients aged 18 to 75 years with stage III-IVB squamous cell oropharyngeal cancer treated with chemoradiotherapy were enrolled between July 5, 2013, and October 1, 2018. Data were analyzed from November 2021 to May 2022. Interventions: The patients were randomly assigned (1:1) to receive standard IMRT (without replanning) or ART (systematic weekly replanning). Main Outcomes and Measures: The primary end point was the frequency of xerostomia, measured by stimulating salivary flow with paraffin. Secondary end points included salivary gland excretory function measured using technetium-99m pertechnetate scintigraphy, patient-reported outcomes (Eisbruch xerostomia-specific questionnaire and the MD Anderson Symptom Inventory for Head and Neck Cancer questionnaire), early and late toxic effects, disease control, and overall and cancer-specific survival. Results: A total of 132 patients were randomized, and after 1 exclusion in the ART arm, 131 were analyzed: 66 in the ART arm (mean [SD] age at inclusion, 60 [8] years; 57 [86.4%] male) and 65 in the standard IMRT arm (mean [SD] age at inclusion, 60 [8] years; 57 [87.7%] male). The median follow-up was 26.4 months (IQR, 1.2-31.3 months). The mean (SD) salivary flow (paraffin) at 12 months was 630 (450) mg/min in the ART arm and 584 (464) mg/min in the standard arm (P = .64). The mean (SD) excretory function of the parotid gland at 12 months, measured by scintigraphy, improved in the ART arm (48% [17%]) compared with the standard arm (41% [17%]) (P = .02). The 2-year-overall survival was 76.9% (95% CI, 64.7%-85.4%) in both arms. Conclusions and Relevance: This randomized clinical trial did not demonstrate a benefit of ART in decreasing xerostomia compared with standard IMRT. No significant differences were found in secondary end points except for parotid gland excretory function, as assessed by scintigraphy, or in survival rates. Trial Registration: ClinicalTrials.gov Identifier: NCT01874587.

Achievable Dosimetric Constraints in Stereotactic Reirradiation for Recurrent Prostate Cancer.

PURPOSE: Stereotactic body radiation therapy has been proposed as a salvage treatment for recurrent prostate cancer after irradiation. One crucial issue is choosing appropriate dose-volume constraints (DVCs) during planning. The objectives of this study were to (1) quantify the proportion of patients respecting the DVCs according to the Urogenital Tumor Study Group GETUG-31 trial, testing 36 Gy in six fractions, (2) explain geometrically why the DVCs could not be respected, and (3) propose the most suitable DVCs. METHODS AND MATERIALS: This retrospective dosimetric analysis included 141 patients treated for recurrent prostate cancer with Cyberknife (Accuray), according to GETUG-31 DVCs: V95% ≥ 95% for the planning target volume (PTV), V12Gy < 20% and V27Gy < 2 cc for the rectum, and V12Gy < 15% and V27Gy < 5 cc for the bladder. The percentage of patients not respecting the DVCs was quantified. Correlations between the DVCs and anatomic structures were examined. New DVCs were proposed. RESULTS: Only 19% of patients respected all DVCs, with a mean PTV of 18.5 cc (range, 3-48 cc), although the mean PTV was 40.5 cc (range, 3-174 cc) in the whole series. A total of 98% of the patients with a clinical target volume (CTV)/prostate ratio >0.5 could not respect the DVCs in the organs at risk. The target coverage and organ-at-risk sparing decreased significantly with increase in the values of PTV, CTV, CTV/prostate ratio, the overlapping volume between the PTV and bladder wall and between the PTV and rectal wall. Threshold values of PTV, >20 cc and 40 cc, allowed for the PTV and bladder DVCs, respectively. To improve DVC respect in case of large target volume, we proposed the following new DVCs: V12Gy < 25% and 25% and V27Gy < 2 cc and 5 cc for the rectum and bladder, respectively. CONCLUSIONS: GETUG-31 DVCs are achievable only for small target volumes (CTV more than half of the prostate). For a larger target volume, new DVCs have been proposed.

Overview of the HECKTOR Challenge at MICCAI 2022: Automatic Head and Neck Tumor Segmentation and Outcome Prediction in PET/CT.

This paper presents an overview of the third edition of the HEad and neCK TumOR segmentation and outcome prediction (HECKTOR) challenge, organized as a satellite event of the 25th International Conference on Medical Image Computing and Computer Assisted Intervention (MICCAI) 2022. The challenge comprises two tasks related to the automatic analysis of FDG-PET/CT images for patients with Head and Neck cancer (H&N), focusing on the oropharynx region. Task 1 is the fully automatic segmentation of H&N primary Gross Tumor Volume (GTVp) and metastatic lymph nodes (GTVn) from FDG-PET/CT images. Task 2 is the fully automatic prediction of Recurrence-Free Survival (RFS) from the same FDG-PET/CT and clinical data. The data were collected from nine centers for a total of 883 cases consisting of FDG-PET/CT images and clinical information, split into 524 training and 359 test cases. The best methods obtained an aggregated Dice Similarity Coefficient (DSCagg) of 0.788 in Task 1, and a Concordance index (C-index) of 0.682 in Task 2.

Practical and technical key challenges in head and neck adaptive radiotherapy: The GORTEC point of view.

Anatomical variations occur during head and neck (H&N) radiotherapy (RT) treatment. These variations may result in underdosage to the target volume or overdosage to the organ at risk. Replanning during the treatment course can be triggered to overcome this issue. Due to technological, methodological and clinical evolutions, tools for adaptive RT (ART) are becoming increasingly sophisticated. The aim of this paper is to give an overview of the key steps of an H&N ART workflow and tools from the point of view of a group of French-speaking medical physicists and physicians (from GORTEC). Focuses are made on image registration, segmentation, estimation of the delivered dose of the day, workflow and quality assurance for an implementation of H&N offline and online ART. Practical recommendations are given to assist physicians and medical physicists in a clinical workflow.

Meta-analysis and Critical Review: Association Between Radio-induced Lymphopenia and Overall Survival in Solid Cancers.

Purpose: Immune system modulation, with the use of immune checkpoint inhibitors, has drastically changed the field of oncology. Strong preclinical data indicate that radiation therapy (RT) may enhance the response rate to such drugs via in situ vaccination, although these data do not consider immune radiotoxicity. This meta-analysis investigates whether radio-induced lymphopenia (RIL) is associated with overall survival (OS). Methods and Materials: A systematic literature search and quantitative analysis were planned, conducted, and reported per the Preferred Reporting Items for Systematic Reviews and Meta-analyses and Quality of Reporting of Meta-analyses checklists. The literature from January 1990 to March 2021 was searched to identify clinical studies with OS data in patients treated with RT and presenting with lymphopenia. A random-effect model was employed for the meta-analysis. Heterogeneity was assessed using the I2 statistic. Publication bias was estimated using a P-curve analysis. Results: A total of 56 studies with 13 223 patients and 11 types of cancers were selected. The mean follow-up time was 35.9 months. Over a third of patients had RIL (37.25%). After removing outlying studies (n = 14), the between-study heterogeneity variance was estimated at t2 = 0.018 (P = .01) with an I2 value of 36.0% (95% confidence interval, 6%-56%). The results showed that RIL was significantly associated with worse OS (hazard ratio: 1.70; 95% confidence interval, 1.55-1.86; P < .01; 95% prediction interval, 1.27-2.26). A subgroup analysis was performed based on the type of primary tumor, and a difference between the subgroups was found (P < .01). Based on the P-curve analysis, a significant evidential value was found, and no significant publication bias was identified among the studies. Conclusions: RIL is a significant prognostic factor for mortality in virtually all solid cancers. Pooled-effect estimates indicate a significantly reduced risk of death in patients without RIL. Tailoring RT regimens to spare the immune system and updating dosimetric constraints for new organs at risk, such as major blood vessels, organs with rich blood supplies, bones, and all lymph node areas, may improve prognoses.

Deep Learning-Based Segmentation of Head and Neck Organs-at-Risk with Clinical Partially Labeled Data.

Radiotherapy is one of the main treatments for localized head and neck (HN) cancer. To design a personalized treatment with reduced radio-induced toxicity, accurate delineation of organs at risk (OAR) is a crucial step. Manual delineation is time- and labor-consuming, as well as observer-dependent. Deep learning (DL) based segmentation has proven to overcome some of these limitations, but requires large databases of homogeneously contoured image sets for robust training. However, these are not easily obtained from the standard clinical protocols as the OARs delineated may vary depending on the patient's tumor site and specific treatment plan. This results in incomplete or partially labeled data. This paper presents a solution to train a robust DL-based automated segmentation tool exploiting a clinical partially labeled dataset. We propose a two-step workflow for OAR segmentation: first, we developed longitudinal OAR-specific 3D segmentation models for pseudo-contour generation, completing the missing contours for some patients; with all OAR available, we trained a multi-class 3D convolutional neural network (nnU-Net) for final OAR segmentation. Results obtained in 44 independent datasets showed superior performance of the proposed methodology for the segmentation of fifteen OARs, with an average Dice score coefficient and surface Dice similarity coefficient of 80.59% and 88.74%. We demonstrated that the model can be straightforwardly integrated into the clinical workflow for standard and adaptive radiotherapy.

Segmentation and Classification of Head and Neck Nodal Metastases and Primary Tumors in PET/CT.

The prediction of cancer characteristics, treatment planning and patient outcome from medical images generally requires tumor delineation. In Head and Neck cancer (H&N), the automatic segmentation and differentiation of primary Gross Tumor Volumes (GTVt) and malignant lymph nodes (GTVn) is a necessary step for large-scale radiomics studies to predict patient outcome such as Progression Free Survival (PFS). Detecting malignant lymph nodes is also a crucial step for Tumor-Node-Metastases (TNM) staging and to support the decision to resect the nodes. In turn, automatic TNM staging and patient outcome prediction can greatly benefit patient care by helping clinicians to find the best personalized treatment. We propose the first model to automatically individually segment GTVt and GTVn in PET/CT images. A bi-modal 3D U-Net model is trained for multi-class and multi-components segmentation on the multi-centric HECKTOR 2020 dataset containing 254 cases. The dataset has been specifically re-annotated by experts to obtain ground truth GTVn contours. The results show promising segmentation performance for the automation of radiomics pipelines and their validation on large-scale studies for which manual annotations are not available. An average test Dice Similarity Coefficients (DSC) of 0.717 is obtained for the segmentation of GTVt. The GTVn segmentation is evaluated with an aggregated DSC to account for the cases without GTVn, which is estimated at 0.729 on the test set.

Cleaning radiotherapy contours for radiomics studies, is it worth it? A head and neck cancer study.

A vast majority of studies in the radiomics field are based on contours originating from radiotherapy planning. This kind of delineation (e.g. Gross Tumor Volume, GTV) is often larger than the true tumoral volume, sometimes including parts of other organs (e.g. trachea in Head and Neck, H&N studies) and the impact of such over-segmentation was little investigated so far. In this paper, we propose to evaluate and compare the performance between models using two contour types: those from radiotherapy planning, and those specifically delineated for radiomics studies. For the latter, we modified the radiotherapy contours to fit the true tumoral volume. The two contour types were compared when predicting Progression-Free Survival (PFS) using Cox models based on radiomics features extracted from FluoroDeoxyGlucose-Positron Emission Tomography (FDG-PET) and CT images of 239 patients with oropharyngeal H&N cancer collected from five centers, the data from the 2020 HECKTOR challenge. Using Dedicated contours demonstrated better performance for predicting PFS, where Harell's concordance indices of 0.61 and 0.69 were achieved for Radiotherapy and Dedicated contours, respectively. Using automatically Resegmented contours based on a fixed intensity range was associated with a C-index of 0.63. These results illustrate the importance of using clean dedicated contours that are close to the true tumoral volume in radiomics studies, even when tumor contours are already available from radiotherapy treatment planning.

Head and neck tumor segmentation in PET/CT: The HECKTOR challenge.

This paper relates the post-analysis of the first edition of the HEad and neCK TumOR (HECKTOR) challenge. This challenge was held as a satellite event of the 23rd International Conference on Medical Image Computing and Computer-Assisted Intervention (MICCAI) 2020, and was the first of its kind focusing on lesion segmentation in combined FDG-PET and CT image modalities. The challenge's task is the automatic segmentation of the Gross Tumor Volume (GTV) of Head and Neck (H&N) oropharyngeal primary tumors in FDG-PET/CT images. To this end, the participants were given a training set of 201 cases from four different centers and their methods were tested on a held-out set of 53 cases from a fifth center. The methods were ranked according to the Dice Score Coefficient (DSC) averaged across all test cases. An additional inter-observer agreement study was organized to assess the difficulty of the task from a human perspective. 64 teams registered to the challenge, among which 10 provided a paper detailing their approach. The best method obtained an average DSC of 0.7591, showing a large improvement over our proposed baseline method and the inter-observer agreement, associated with DSCs of 0.6610 and 0.61, respectively. The automatic methods proved to successfully leverage the wealth of metabolic and structural properties of combined PET and CT modalities, significantly outperforming human inter-observer agreement level, semi-automatic thresholding based on PET images as well as other single modality-based methods. This promising performance is one step forward towards large-scale radiomics studies in H&N cancer, obviating the need for error-prone and time-consuming manual delineation of GTVs.

Cardiac radioablation for ventricular tachycardia: Which approach for incorporating cardiorespiratory motions into the planning target volume?

PURPOSE: To evaluate different approaches for generating a cardiorespiratory ITV for cardiac radioablation. METHODS: Four patients with ventricular tachycardia were included in this study. For each patient, cardiac-gated and respiration-correlated 4D-CT scans were acquired. The cardiorespiratory ITV was defined using registrations of the cardiac and respiratory 4D-CT images. Five different approaches, which differed in the number of incorporated cardiac phases (1, 2, 10, or 1 with a fixed 3 mm margin (FM) expansion) and respiratory phases (2 or 10), were evaluated. For each approach, a VMAT treatment plan was simulated. Target coverage (TC) and spill were evaluated geometrically and dosimetrically for each approach. RESULTS: When employing one cardiac phase, the TC did not exceed 85%. Using the two extreme phases of the cardiac and respiratory cycles resulted in a geometric TC < 88% for two patients, with a dosimetric TC of 83% for one patient. An acceptable TC for all patients (geometric TC > 89%, dosimetric TC > 92%) was only achieved when combining 10 respiratory phases with either 2 or 10 cardiac phases or a single cardiac phase with FM. The use of a single cardiac phase with FM combined with 10 respiratory phases lead to a mean geometric and dosimetric spill of 43% and 35%, respectively. CONCLUSION: For cardiac radioablation, the use of two extreme cardiac phases combined with 10 respiratory phases is a robust approach to generate a cardiorespiratory ITV. The use of a single cardiac phase with or without fixed margin expansion is not recommended based on this study.

18F-FDG PET/CT-Based Prognostic Survival Model After Surgery for Head and Neck Cancer.

The aims of this multicenter study were to identify clinical and preoperative PET/CT parameters predicting overall survival (OS) and distant metastasis-free survival (DMFS) in a cohort of head and neck squamous cell carcinoma patients treated with surgery, to generate a prognostic model of OS and DMFS, and to validate this prognostic model with an independent cohort. Methods: A total of 382 consecutive patients with head and neck squamous cell carcinoma, divided into training (n = 318) and validation (n = 64) cohorts, were retrospectively included. The following PET/CT parameters were analyzed: clinical parameters, SUVmax, SUVmean, metabolic tumor volume (MTV), total lesion glycolysis, and distance parameters for the primary tumor and lymph nodes defined by 2 segmentation methods (relative SUVmax threshold and absolute SUV threshold). Cox analyses were performed for OS and DMFS in the training cohort. The concordance index (c-index) was used to identify highly prognostic parameters. These prognostic parameters were externally tested in the validation cohort. Results: In multivariable analysis, the significant parameters for OS were T stage and nodal MTV, with a c-index of 0.64 (P < 0.001). For DMFS, the significant parameters were T stage, nodal MTV, and maximal tumor-node distance, with a c-index of 0.76 (P < 0.001). These combinations of parameters were externally validated, with c-indices of 0.63 (P < 0.001) and 0.71 (P < 0.001) for OS and DMFS, respectively. Conclusion: The nodal MTV associated with the maximal tumor-node distance was significantly correlated with the risk of DMFS. Moreover, this parameter, in addition to clinical parameters, was associated with a higher risk of death. These prognostic factors may be used to tailor individualized treatment.

Impact of Neck Dissection in Head and Neck Squamous Cell Carcinomas of Unknown Primary.

PURPOSE: Management of head and neck cancers of unknown primary (HNCUP) combines neck dissection (ND) and radiotherapy, with or without chemotherapy. The prognostic value of ND has hardly been studied in HNCUP. METHODS: A retrospective multicentric study assessed the impact of ND extent (adenectomy, selective ND, radical/radical-modified ND) on nodal relapse, progression-free survival (PFS) or survival, taking into account nodal stage. RESULTS: 53 patients (16.5%) had no ND, 33 (10.2%) had lymphadenectomy, 116 (36.0%) underwent selective ND and 120 underwent radical/radical-modified ND (37.3%), 15 of which received radical ND (4.7%). With a 34-month median follow-up, the 3-year incidence of nodal relapse was 12.5% and progression-free survival (PFS) 69.1%. In multivariate analysis after adjusting for nodal stage, the risk of nodal relapse or progression was reduced with lymphadenectomy, selective or radical/modified ND, but survival rates were similar. Patients undergoing lymphadenectomy or ND had a better PFS and lowered nodal relapse incidence in the N1 + N2a group, but the improvement was not significant for the N2b or N2 + N3c patients. Severe toxicity rates exceeded 40% with radical ND. CONCLUSION: In HNCUP, ND improves PFS, regardless of nodal stage. The magnitude of the benefit of ND does not appear to depend on ND extent and decreases with a more advanced nodal stage.

Statistical harmonization can improve the development of a multicenter CT-based radiomic model predictive of nonresponse to induction chemotherapy in laryngeal cancers.

PURPOSE: To develop a radiomic model predicting nonresponse to induction chemotherapy in laryngeal cancers, from multicenter pretherapeutic contrast-enhanced computed tomography (CE-CT) and evaluate the benefit of feature harmonization in such a context. METHODS: Patients (n = 104) eligible for laryngeal preservation chemotherapy were included in five centers. Primary tumor was manually delineated on the CE-CT images. The following radiomic features were extracted with an in-house software (MIRAS v1.1, LaTIM UMR 1101): intensity, shape, and textural features derived from Gray-Level Co-occurrence Matrix (GLCM), Neighborhood Gray Tone Difference Matrix (NGTDM), Gray-Level Run Length Matrix (GLRLM), and Gray-Level Size Zone Matrix (GLSZM). Harmonization was performed using ComBat after unsupervised hierarchical clustering, used to determine labels automatically, given the high heterogeneity of imaging characteristics across and within centers. Patients with similar feature distributions were grouped with unsupervised clustering into an optimal number of clusters (2) determined with "silhouette scoring." Statistical harmonization was then carried out with ComBat on these 2 identified clusters. The cohort was split into training/validation (n = 66) and testing (n = 32) sets. Area under the receiver operating characteristics curves (AUC) were used to evaluate the ability of radiomic features (before and after harmonization) to predict nonresponse to chemotherapy, and specificity (Sp) and sensitivity (Se) were used to quantify their performance in the testing set. RESULTS: Without harmonization, none of the features identified as predictive in the training set remained significant in the testing set. After ComBat, one textural feature identified in the training set keeps a predictive trend in the testing set-Zone Percentage, derived from the GLSZM, was predictive of nonresponse in the training set (AUC = 0.62, Se = 70%, Sp = 64%, P = 0.04) and obtained a satisfactory performance in the testing set (Se = 80%, Sp = 67%, P = 0.03), although significance was limited by the size of the testing set. These results are consistent with previously published findings in head and neck cancers. CONCLUSIONS: Radiomic features from CE-CT could help in the selection of patients for induction chemotherapy in laryngeal cancers, with relatively good sensitivity and specificity in predicting lack of response. Statistical harmonization with ComBat and unsupervised clustering seems to improve the predictive value of features extracted in such a heterogeneous multicenter setting.

Recommendations for postoperative radiotherapy in head & neck squamous cell carcinoma in the presence of flaps: A GORTEC internationally-reviewed HNCIG-endorsed consensus.

INTRODUCTION: Head and neck reconstructive surgery using a flap is increasingly common. Best practices and outcomes for postoperative radiotherapy (poRT) with flaps have not been specified. We aimed to provide consensus recommendations to assist clinical decision-making highlighting areas of uncertainty in the presence of flaps. MATERIAL AND METHODS: Radiation, medical, and surgical oncologists were assembled from GORTEC and internationally with the Head and Neck Cancer International Group (HNCIG). The consensus-building approach covered 59 topics across four domains: (1) identification of postoperative tissue changes on imaging for flap delineation, (2) understanding of tumor relapse risks and target volume definitions, (3) functional radiation-induced deterioration, (4) feasibility of flap avoidance. RESULTS: Across the 4 domains, international consensus (median score ≥ 7/9) was achieved only for functional deterioration (73.3%); other consensus rates were 55.6% for poRT avoidance of flap structures, 41.2% for flap definition and 11.1% for tumor spread patterns. Radiation-induced flap fibrosis or atrophy and their functional impact was well recognized while flap necrosis was not, suggesting dose-volume adaptation for the former. Flap avoidance was recommended to minimize bone flap osteoradionecrosis but not soft-tissue toxicity. The need for identification (CT planning, fiducials, accurate operative report) and targeting of the junction area at risk between native tissues and flap was well recognized. Experts variably considered flaps as prone to tumor dissemination or not. Discrepancies in rating of 11 items among international reviewing participants are shown. CONCLUSION: International GORTEC and HNCIG-endorsed recommendations were generated for the management of flaps in head and neck radiotherapy. Considerable knowledge gaps hinder further consensus, in particular with respect to tumor spread patterns.

The importance of feature aggregation in radiomics: a head and neck cancer study.

In standard radiomics studies the features extracted from clinical images are mostly quantified with simple statistics such as the average or variance per Region of Interest (ROI). Such approaches may smooth out any intra-region heterogeneity and thus hide some tumor aggressiveness that may hamper predictions. In this paper we study the importance of feature aggregation within the standard radiomics workflow, which allows to take into account intra-region variations. Feature aggregation methods transform a collection of voxel values from feature response maps (over a ROI) into one or several scalar values that are usable for statistical or machine learning algorithms. This important step has been little investigated within the radiomics workflows, so far. In this paper, we compare several aggregation methods with standard radiomics approaches in order to assess the improvements in prediction capabilities. We evaluate the performance using an aggregation function based on Bags of Visual Words (BoVW), which allows for the preservation of piece-wise homogeneous information within heterogeneous regions and compared with standard methods. The different models are compared on a cohort of 214 head and neck cancer patients coming from 4 medical centers. Radiomics features were extracted from manually delineated tumors in clinical PET-FDG and CT images were analyzed. We compared the performance of standard radiomics models, the volume of the ROI alone and the BoVW model for survival analysis. The average concordance index was estimated with a five fold cross-validation. The performance was significantly better using the BoVW model 0.627 (95% CI: 0.616-0.637) as compared to standard radiomics0.505 (95% CI: 0.499-0.511), mean-var. 0.543 (95% CI: 0.536-0.549), mean0.547 (95% CI: 0.541-0.554), var.0.530 (95% CI: 0.524-0.536) or volume 0.577 (95% CI: 0.571-0.582). We conclude that classical aggregation methods are not optimal in case of heterogeneous tumors. We also showed that the BoVW model is a better alternative to extract consistent features in the presence of lesions composed of heterogeneous tissue.

Comparison of feature selection in radiomics for the prediction of overall survival after radiotherapy for hepatocellular carcinoma.

Hepatocellular carcinoma (HCC) is the sixth more frequent cancer worldwide. This type of cancer has a poor overall survival rate mainly due to underlying cirrhosis and risk of recurrence outside the treated lesion. Quantitative imaging within a radiomics workflow may help assessing the probability of survival and potentially may allow tailoring personalized treatments. In radiomics a large amount of features can be extracted, which may be correlated across a population and very often can be surrogates of the same physiopathology. This issues are more pronounced and difficult to tackle with imbalanced data. Feature selection strategies are therefore required to extract the most informative with the increased predictive capabilities. In this paper, we compared different unsupervised and supervised strategies for feature selection in presence of imbalanced data and optimize them within a machine learning framework. Multi-parametric Magnetic Resonance Images from 81 individuals (19 deceased) treated with stereotactic body radiation therapy (SBRT) for inoperable (HCC) were analyzed. Pre-selection of a reduced set of features based on Affinity Propagation clustering (non supervised) achieved a significant improvement in AUC compared to other approaches with and without feature pre-selection. By including the synthetic minority over-sampling technique (SMOTE) for imbalanced data and Random Forest classification this workflow emerges as an appealing feature selection strategy for survival prediction within radiomics studies.

Comparison of CBCT-based dose calculation methods in head and neck cancer radiotherapy: from Hounsfield unit to density calibration curve to deep learning.

PURPOSE: Anatomical variations occur during head and neck (H&N) radiotherapy treatment. kV cone-beam computed tomography (CBCT) images can be used for daily dose monitoring to assess dose variations owing to anatomic changes. Deep learning methods (DLMs) have recently been proposed to generate pseudo-CT (pCT) from CBCT to perform dose calculation. This study aims to evaluate the accuracy of a DLM and to compare this method with three existing methods of dose calculation from CBCT in H&N cancer radiotherapy. METHODS: Forty-four patients received VMAT for H&N cancer (70-63-56 Gy). For each patient, reference CT (Bigbore, Philips) and CBCT images (XVI, Elekta) were acquired. The DLM was based on a generative adversarial network. The three compared methods were: (a) a method using a density to Hounsfield Unit (HU) relation from phantom CBCT image (HU-D curve method), (b) a water-air-bone density assignment method (DAM), and iii) a method using deformable image registration (DIR). The imaging endpoints were the mean absolute error (MAE) and mean error (ME) of HU from pCT and reference CT (CTref ). The dosimetric endpoints were dose discrepancies and 3D gamma analyses (local, 2%/2 mm, 30% dose threshold). Dose discrepancies were defined as the mean absolute differences between DVHs calculated from the CTref and pCT of each method. RESULTS: In the entire body, the MAEs and MEs of the DLM, HU-D curve method, DAM, and DIR method were 82.4 and 17.1 HU, 266.6 and 208.9 HU, 113.2 and 14.2 HU, and 95.5 and -36.6 HU, respectively. The MAE obtained using the DLM differed significantly from those of other methods (Wilcoxon, P ≤ 0.05). The DLM dose discrepancies were 7 ± 8 cGy (maximum = 44 cGy) for the ipsilateral parotid gland Dmean and 5 ± 6 cGy (max = 26 cGy) for the contralateral parotid gland mean dose (Dmean ). For the parotid gland Dmean , no significant dose difference was observed between the DLM and other methods. The mean 3D gamma pass rate ± standard deviation was 98.1 ± 1.2%, 91.0 ± 5.3%, 97.9 ± 1.6%, and 98.8 ± 0.7% for the DLM, HU-D method, DAM, and DIR method, respectively. The gamma pass rates and mean gamma results of the HU-D curve method, DAM, and DIR method differed significantly from those of the DLM. CONCLUSIONS: For H&N radiotherapy, DIR method and DLM appears as the most appealing CBCT-based dose calculation methods among the four methods in terms of dose accuracy as well as calculation time. Using the DIR method or DLM with CBCT images enables dose monitoring in the parotid glands during the treatment course and may be used to trigger replanning.