Radiothérapie Moléculaire et Innovation Thérapeutique

The gut microbiota influences the clinical responses of cancer patients to immunecheckpoint inhibitors (ICIs). However, there is no consensus definition of detrimental dysbiosis. Based on metagenomics (MG) sequencing of 245 non-small cell lung cancer (NSCLC) patient feces, we constructed species-level co-abundance networks that were clustered into species-interacting groups (SIGs) correlating with overall survival. Thirty-seven and forty-five MG species (MGSs) were associated with resistance (SIG1) and response (SIG2) to ICIs, respectively. When combined with the quantification of Akkermansia species, this procedure allowed a person-based calculation of a topological score (TOPOSCORE) that was validated in an additional 254 NSCLC patients and in 216 genitourinary cancer patients. Finally, this TOPOSCORE was translated into a 21-bacterial probe set-based qPCR scoring that was validated in a prospective cohort of NSCLC patients as well as in colorectal and melanoma patients. This approach could represent a dynamic diagnosis tool for intestinal dysbiosis to guide personalized microbiota-centered interventions.

BACKGROUND: Combining radiotherapy (RT) with immuno-oncology (IO) therapy (IORT) may enhance IO-induced antitumor response. Quantitative imaging biomarkers can be used to provide prognosis, predict tumor response in a non-invasive fashion and improve patient selection for IORT. A biologically inspired CD8 T-cells-associated radiomics signature has been developed on previous cohorts. We evaluated here whether this CD8 radiomic signature is associated with lesion response, whether it may help to assess disease spatial heterogeneity for predicting outcomes of patients treated with IORT. We also evaluated differences between irradiated and non-irradiated lesions. METHODS: Clinical data from patients with advanced solid tumors in six independent clinical studies of IORT were investigated. Immunotherapy consisted of 4 different drugs (antiprogrammed death-ligand 1 or anticytotoxic T-lymphocyte-associated protein 4 in monotherapy). Most patients received stereotactic RT to one lesion. Irradiated and non-irradiated lesions were delineated from baseline and the first evaluation CT scans. Radiomic features were extracted from contrast-enhanced CT images and the CD8 radiomics signature was applied. A responding lesion was defined by a decrease in lesion size of at least 30%. Dispersion metrices of the radiomics signature were estimated to evaluate the impact of tumor heterogeneity in patient's response. RESULTS: A total of 94 patients involving multiple lesions (100 irradiated and 189 non-irradiated lesions) were considered for a statistical interpretation. Lesions with high CD8 radiomics score at baseline were associated with significantly higher tumor response (area under the receiving operating characteristic curve (AUC)=0.63, p=0.0020). Entropy of the radiomics scores distribution on all lesions was shown to be associated with progression-free survival (HR=1.67, p=0.040), out-of-field abscopal response (AUC=0.70, p=0.014) and overall survival (HR=2.08, p=0.023), which remained significant in a multivariate analysis including clinical and biological variables. CONCLUSIONS: These results enhance the predictive value of the biologically inspired CD8 radiomics score and suggests that tumor heterogeneity should be systematically considered in patients treated with IORT. This CD8 radiomics signature may help select patients who are most likely to benefit from IORT.

Given that a significant fraction of buildings and architectural heritage in Europe's historical centers are masonry structures, the selection of proper diagnosis, technological surveys, non-destructive testing, and interpretations of crack and decay patterns is paramount for a risk assessment of possible damage. Identifying the possible crack patterns, discontinuities, and associated brittle failure mechanisms within unreinforced masonry under seismic and gravity actions allows for reliable retrofitting interventions. Traditional and modern materials and strengthening techniques create a wide range of compatible, removable, and sustainable conservation strategies. Steel/timber tie-rods are mainly used to support the horizontal thrust of arches, vaults, and roofs and are particularly suitable for better connecting structural elements, e.g., masonry walls and floors. Composite reinforcing systems using carbon, glass fibers, and thin mortar layers can improve tensile resistance, ultimate strength, and displacement capacity to avoid brittle shear failures. This study overviews masonry structural diagnostics and compares traditional and advanced strengthening techniques of masonry walls, arches, vaults, and columns. Several research results in automatic surface crack detection for unreinforced masonry (URM) walls are presented considering crack detection based on machine learning and deep learning algorithms. In addition, the kinematic and static principles of Limit Analysis within the rigid no-tension model framework are presented. The manuscript sets a practical perspective, providing an inclusive list of papers describing the essential latest research in this field; thus, this paper is useful for researchers and practitioners in masonry structures.

IntroductionWe report long-term efficacy and overall survival (OS) results from a randomised, double-blind, phase 2 study (NCT02022098) investigating xevinapant plus standard-of-care chemoradiotherapy (CRT) vs. placebo plus CRT in 96 patients with unresected locally advanced squamous cell carcinoma of the head and neck (LA SCCHN).MethodsPatients were randomised 1:1 to xevinapant 200 mg/day (days 1–14 of a 21-day cycle for 3 cycles), or matched placebo, plus CRT (cisplatin 100 mg/m2 every 3 weeks for 3 cycles plus conventional fractionated high-dose intensity-modulated radiotherapy [70 Gy/35 F, 2 Gy/F, 5 days/week for 7 weeks]). Locoregional control, progression-free survival, and duration of response after 3 years, long-term safety, and 5-year OS were assessed.ResultsThe risk of locoregional failure was reduced by 54% for xevinapant plus CRT vs. placebo plus CRT but did not reach statistical significance (adjusted hazard ratio [HR] 0.46; 95% CI, 0.19–1.13; P = .0893). The risk of death or disease progression was reduced by 67% for xevinapant plus CRT (adjusted HR 0.33; 95% CI, 0.17–0.67; P = .0019). The risk of death was approximately halved in the xevinapant arm compared with placebo (adjusted HR 0.47; 95% CI, 0.27–0.84; P = .0101). OS was prolonged with xevinapant plus CRT vs. placebo plus CRT; median OS not reached (95% CI, 40.3-not evaluable) vs. 36.1 months (95% CI, 21.8–46.7). Incidence of late-onset grade ≥3 toxicities was similar across arms.ConclusionsIn this randomised phase 2 study of 96 patients, xevinapant plus CRT demonstrated superior efficacy benefits, including markedly improved 5-year survival in patients with unresected LA SCCHN.

PURPOSE: While there is still a significant need to identify potential biomarkers that can predict which patients are most likely to respond to immunotherapy treatments, radiomic approaches have shown promising results. The objectives of this study were to evaluate whether a previously validated radiomics signature of CD8 T-cells could predict progressions at a lesion level and whether the spatial heterogeneity of this radiomics score could be used at a patient level to assess the clinical response and survival of melanoma patients. METHODS: Clinical data from patients with advanced melanoma treated in our center with immunotherapy were retrieved. Radiomic features were extracted and the CD8 radiomics signature was applied. A progressive lesion was defined by an increase in lesion size of 20% or more. Dispersion metrics of the radiomics signature were estimated to evaluate the impact of interlesion heterogeneity on patient's response. Fine-tuned cut-offs for predicting overall survival were evaluated after splitting data into training and test sets. RESULTS: A total of 136 patients were included in this study, with 1120 segmented lesions at baseline, and 1052 lesions at first evaluation. A low CD8 radiomics score at baseline was associated with a significantly higher risk of lesion progression (AUC=0.55, p=0.0091), especially for lesions larger than >1 mL (AUC=0.59 overall, p=0.0035, with AUC=0.75, p=0.002 for subcutaneous lesions, AUC=0.68, p=0.01, for liver lesions and AUC=0.62, p=0.03 for nodes). The least infiltrated lesion according to the radiomics score of CD8 T-cells was positively associated with overall survival (training set HR=0.31, p=0.00062, test set HR=0.28, p=0.016), which remained significant in a multivariate analysis including clinical and biological variables. CONCLUSIONS: These results confirm the predictive value at a lesion level of the biologically inspired CD8 radiomics score in melanoma patients treated with anti-PD1-based immunotherapy and may be interesting to assess the disease spatial heterogeneity to evaluate the patient prognosis with potential clinical implication such as tumor selection for focal ablative therapies.

Globally, cervical cancers continue to be one of the leading causes of cancer-related deaths. The primary treatment of patients with early-stage disease includes surgery or radiation therapy with or without chemotherapy. The main challenge in treating these patients is to maintain a curative approach and limit treatment-related morbidity. Traditionally, inoperable patients are treated with radiation therapy solely and operable patients undergo upfront surgery followed by adjuvant (chemo) radiotherapy in cases with poor histopathological prognostic features. Patients with locally advanced cervical cancers are treated with concurrent chemoradiotherapy followed by an image-guided brachytherapy boost. In these patients, the main pattern of failure is distant relapse, encouraging intensification of systemic treatments to improve disease control. Ongoing trials are evaluating immunotherapy in locally advanced tumours following its encouraging efficacy reported in the recurrent and metastatic settings. In this article, clinical evidence of neoadjuvant and adjuvant treatments in cervical cancer patients is reviewed, with a focus on potential strategies to improve patients' outcome and minimize treatment-related morbidity.

OBJECTIVES: human experts, on a testing cohort. METHODS: We enrolled 143 patients (71 GBM and 72 BM) in a retrospective bicentric study from January 2010 to May 2019 to train the classifier. Post-contrast 3DT1 MR images were performed on a 3-Tesla MR unit and 100 radiomic features were extracted. Selection and optimization of the Machine Learning (ML) classifier was performed using a nested cross-validation. Sensitivity, specificity, balanced accuracy, and area under the receiver operating characteristic curve (AUC) were calculated as performance metrics. The model final performance was cross-validated, then evaluated on a test set of 37 patients, and compared to human blind reading using a McNemar's test. RESULTS: The ML classifier had a mean [95% confidence interval] sensitivity of 85% [77; 94], a specificity of 87% [78; 97], a balanced accuracy of 86% [80; 92], and an AUC of 92% [87; 97] with cross-validation. Sensitivity, specificity, balanced accuracy and AUC were equal to 75, 86, 80 and 85% on the test set. Sphericity 3D radiomic index highlighted the highest coefficient in the logistic regression model. There were no statistical significant differences observed between the performance of the classifier and the experts' blinded examination. CONCLUSIONS: The proposed diagnostic support system based on radiomic features extracted from post-contrast 3DT1 MR images helps in differentiating solitary BM from GBM with high diagnosis performance and generalizability.

The use of multicentric data is becoming essential for developing generalizable radiomic signatures. In particular, Magnetic Resonance Imaging (MRI) data used in brain oncology are often heterogeneous in terms of scanners and acquisitions, which significantly impact quantitative radiomic features. Various methods have been proposed to decrease dependency, including methods acting directly on MR images, i.e., based on the application of several preprocessing steps before feature extraction or the ComBat method, which harmonizes radiomic features themselves. The ComBat method used for radiomics may be misleading and presents some limitations, such as the need to know the labels associated with the "batch effect". In addition, a statistically representative sample is required and the applicability of a signature whose batch label is not present in the train set is not possible. This work aimed to compare a priori and a posteriori radiomic harmonization methods and propose a code adaptation to be machine learning compatible. Furthermore, we have developed AutoComBat, which aims to automatically determine the batch labels, using either MRI metadata or quality metrics as inputs of the proposed constrained clustering. A heterogeneous dataset consisting of high and low-grade gliomas coming from eight different centers was considered. The different methods were compared based on their ability to decrease relative standard deviation of radiomic features extracted from white matter and on their performance on a classification task using different machine learning models. ComBat and AutoComBat using image-derived quality metrics as inputs for batch assignment and preprocessing methods presented promising results on white matter harmonization, but with no clear consensus for all MR images. Preprocessing showed the best results on the T1w-gd images for the grading task. For T2w-flair, AutoComBat, using either metadata plus quality metrics or metadata alone as inputs, performs better than the conventional ComBat, highlighting its potential for data harmonization. Our results are MRI weighting, feature class and task dependent and require further investigations on other datasets.

Background Tumors rewire their metabolism to achieve robust anabolism and resistance against therapeutic interventions like cisplatin treatment. For example, a prolonged exposure to cisplatin causes downregulation of pyridoxal kinase (PDXK), the enzyme that generates the active vitamin B6, and upregulation of poly ADP-ribose (PAR) polymerase-1 (PARP1) activity that requires a supply of nicotinamide (vitamin B3) adenine dinucleotide. We investigated the impact of the levels of PDXK and PAR on the local immunosurveillance (ie, density of the antigen presenting cells and adaptive immune response by CD8 T lymphocytes) in two different tumor types. Methods Tumors from patients with locally advanced cervical carcinoma (LACC) and non-small cell lung cancer (NSCLC) were stained for PAR, PDXK, dendritic cell lysosomal associated membrane glycoprotein (DC-LAMP) and CD8 T cell infiltration. Their correlations and prognostic impact were assessed. Cisplatin-resistant NSCLC cell clones isolated from Lewis-lung cancer (LLC) cells were evaluated for PAR levels by immunoblot. Parental (PAR low ) and cisplatin-resistant (PAR high ) clones were subcutaneously injected into the flank of C57BL/6 mice. Tumors were harvested to evaluate their immune infiltration by flow cytometry. Results The infiltration of tumors by CD8 T and DC-LAMP + cells was associated with a favorable overall survival in patients with LACC (p=0.006 and p=0.008, respectively) and NSCLC (p<0.001 for both CD8 T and DC-LAMP cells). We observed a positive correlation between PDXK expression and the infiltration by DC-LAMP (R=0.44, p=0.02 in LACC, R=0.14, p=0.057 in NSCLC), and a negative correlation between PAR levels and CD8 T lymphocytes (R=−0.39, p=0.034 in LACC, R=−0.18, p=0.017 in NSCLC). PARP1 is constitutively hyperactivated in cisplatin-resistant LLC cells manifesting elevated intracellular levels of poly(ADP-ribosyl)ated proteins (PAR high ). Tumors formed by such cancer cells injected into immunocompetent mice were scarcely infiltrated by CD8 T (p=0.028) and antigen presenting cells (p=0.086). Conclusions Oncometabolic features can impact local immunosurveillance, providing new functional links between cisplatin resistance and therapeutic failure.

Focal radiation therapy (RT) has attracted considerable attention as a combinatorial partner for immunotherapy (IT), largely reflecting a well-defined, predictable safety profile and at least some potential for immunostimulation. However, only a few RT-IT combinations have been tested successfully in patients with cancer, highlighting the urgent need for an improved understanding of the interaction between RT and IT in both preclinical and clinical scenarios. Every year since 2016, ImmunoRad gathers experts working at the interface between RT and IT to provide a forum for education and discussion, with the ultimate goal of fostering progress in the field at both preclinical and clinical levels. Here, we summarize the key concepts and findings presented at the Sixth Annual ImmunoRad conference.

Background and PurposeAutomatic segmentation methods have greatly changed the radiotherapy workflow, but still need to be extended to target volumes. In this paper, Deep Learning (DL) models were compared for Gross Tumor Volume (GTV) segmentation in locally advanced cervical cancer, and a novel investigation into failure detection was introduced by utilizing radiomic features.Methods and MaterialsWe trained eight DL models (UNet, VNet, SegResNet, SegResNetVAE) for 2D and 3D segmentation. Ensembling individually trained models during cross-validation generated the final segmentation. To detect failures, binary classifiers were trained using radiomic features extracted from segmented GTVs as inputs, aiming to classify contours based on whether their Dice Similarity Coefficient (DSC)<T and DSC⩾T. Two distinct cohorts of T2W pre-RT MR images captured in 2D sequences were used: one retrospective cohort consisting of 115 LACC patients from 30 scanners, and the other prospective cohort, comprising 51 patients from 7 scanners, used for testing.ResultsSegmentation by 2D-SegResNet achieved the best DSC, Surface Dice Similarity Coefficient (SDSC3mm), and 95th Hausdorff Distance (95HD): DSC=0.72±0.16, SDSC3mm=0.66±0.17, and 95HD=14.6±9.0 mm without missing segmentation (M=0) on the test cohort. Failure detection generated precision (P=0.88), recall (R=0.75), F1-score (F=0.81), and accuracy (A=0.86) using Logistic Regression (LR) classifier on the test cohort with a threshold T=0.67 on DSC values.ConclusionsOur study revealed that segmentation accuracy varies slightly among different DL methods, with 2D networks outperforming 3D networks in 2D MRI sequences. Doctors found the time-saving aspect advantageous. The proposed failure detection could guide doctors in sensitive cases.

Valvular Heart Disease (VHD) is a known late complication of radiotherapy for childhood cancer (CC), and identifying high-risk survivors correctly remains a challenge. This paper focuses on the distribution of the radiation dose absorbed by heart tissues. We propose that a dosiomics signature could provide insight into the spatial characteristics of the heart dose associated with a VHD, beyond the already-established risk induced by high doses. We analyzed data from the 7670 survivors of the French Childhood Cancer Survivors' Study (FCCSS), 3902 of whom were treated with radiotherapy. In all, 63 (1.6%) survivors that had been treated with radiotherapy experienced a VHD, and 57 of them had heterogeneous heart doses. From the heart-dose distribution of each survivor, we extracted 93 first-order and spatial dosiomics features. We trained random forest algorithms adapted for imbalanced classification and evaluated their predictive performance compared to the performance of standard mean heart dose (MHD)-based models. Sensitivity analyses were also conducted for sub-populations of survivors with spatially heterogeneous heart doses. Our results suggest that MHD and dosiomics-based models performed equally well globally in our cohort and that, when considering the sub-population having received a spatially heterogeneous dose distribution, the predictive capability of the models is significantly improved by the use of the dosiomics features. If these findings are further validated, the dosiomics signature may be incorporated into machine learning algorithms for radiation-induced VHD risk assessment and, in turn, into the personalized refinement of follow-up guidelines.

To mitigate the risk of radioactive isotope dissemination, the development of preventative and curative measures is of particular interest. For mass treatment, the developed solution must be easily administered, preferably orally, with effective, nontoxic decorporating properties against a wide range of radioactive isotopes. Currently, most orally administered chelation therapy products are quickly absorbed into the blood circulation, where chelation of the radioactive isotope is a race against time due to the short circulation half-life of the therapeutic. This report presents an alternative therapeutic approach by using a functionalized chitosan (chitosan@DOTAGA) with chelating properties that remains within the gastrointestinal tract and is eliminated in feces, that can protect against ingested radioactive isotopes. The polymer shows important in vitro chelation properties towards different metallic cations of importance, including (Cs(I), Ir(III), Th(IV), Tl(I), Sr(II), U(VI) and Co(II)), at different pH (from 1 to 7) representing the different environments in the gastrointestinal tract. An in vivo proof of concept is presented on a rodent model of uranium contamination following an oral administration of Chitosan@DOTAGA. The polymer partially prevents the accumulation of uranium within the kidneys (providing a protective effect) and completely prevents its uptake by the spleen.

Emerging countries are currently facing an increasing burden of cancer while they do not have adequate prevention, monitoring, and research capabilities to tackle the disease. Cancer outcomes are influenced by several factors, including different cancer patterns, national cancer screening guidelines, current stage of disease, and access to quality care and treatments. Discrepancies in cancer care between emerging and developed countries require actions to achieve global health equity. The process of pioneering a sister relationship in the oncology field can thwart the global burden of cancer. The objective of such cooperation programs should include research and training programs, evidence-based oncology practice, and quality cancer. Building global connections will therefore be the novel approach to addressing the global burden of cancer.

Background: Cardiac disease (CD) is a primary long-term diagnosed pathology among childhood cancer survivors. Dosiomics (radiomics extracted from the dose distribution) have received attention in the past few years to assess better the induced risk of radiotherapy (RT) than standard dosimetric features such as dose-volume indicators. Hence, using the spatial information contained in the dosiomics features with machine learning methods may improve the prediction of CD. Methods: We considered the 7670 5-year survivors of the French Childhood Cancer Survivors Study (FCCSS). Dose-volume and dosiomics features are extracted from the radiation dose distribution of 3943 patients treated with RT. Survival analysis is performed considering several groups of features and several models [Cox Proportional Hazard with Lasso penalty, Cox with Bootstrap Lasso selection, Random Survival Forests (RSF)]. We establish the performance of dosiomics compared to baseline models by estimating C-index and Integrated Brier Score (IBS) metrics with 5-fold stratified cross-validation and compare their time-dependent error curves. Results: An RSF model adjusted on the first-order dosiomics predictors extracted from the whole heart performed best regarding the C-index (0.792 ± 0.049), and an RSF model adjusted on the first-order dosiomics predictors extracted from the heart's subparts performed best regarding the IBS (0.069 ± 0.05). However, the difference is not statistically significant with the standard models (C-index of Cox PH adjusted on dose-volume indicators: 0.791 ± 0.044; IBS of Cox PH adjusted on the mean dose to the heart: 0.074 ± 0.056). Conclusion: In this study, dosiomics models have slightly better performance metrics but they do not outperform the standard models significantly. Quantiles of the dose distribution may contain enough information to estimate the risk of late radio-induced high-grade CD in childhood cancer survivors.

Abstract BACKGROUND High-grade gliomas are the most common type of primary brain tumor in adult patients. Despite a multimodal treatment, the prognosis remains poor. Accurate stratification of survival is crucial for choosing the best course of therapy. The objective of this study was to use radiomics analysis based on multiparametric MRI combined with clinical data of high-grade gliomas patients with incomplete resection to identify a predictive signature for the stratification of patients according to their overall survival. MATERIAL AND METHODS In this multicentric retrospective study, we included a sub-cohort of 237 patients treated for high-grade gliomas, from 2009 to 2021, at three French medical centers. This data collection was carried out as part of a national funding to promote the integration of AI tools into clinical routine in oncology (AI.DReAM project). Clinical and pathological data, together with the MR images before treatment and at the time of relapse were collected.Patients with gross tumor resection were excluded to homogenize the automatic contouring. Three tumor subregions were generated by a deep learning framework based in the BraTS challenge on the MR images at the time of radiotherapy: label 1, corresponding to the necrosis part of the tumor; label 2, the hypersignal in the T2/FLAIR sequence; and label 3, corresponding to the enhancing part after the injection of gadolinium. The cohort was split into a training data set of 189 patients with cross-validation to ensure robustness and evaluated using a test set of 48 patients. The survival stratification of the two classes was based on the median in month of overall survival (OS) in our cohort with short survival defined as OS inferior to 17 months. The open-source software Pyradiomics and the ensemble classifiers XGBoost, in conjunction with grid search were used for calculating radiomic parameters. RESULTS In total, 976 patients were collected. For this analysis, 237 patients with high-grade gliomas were included, 82 % being IDH-wildtype glioblastoma. One-hundred and nineteen patients underwent subtotal tumor resection. A total of 936 radiomics were extracted from the three tumor subregions and 14 clinical data features were added to the algorithm. The areas under the ROC curve (ROC-AUC) for survival prediction based on radiomic and clinical parameters alone were 0.61 and 0.60 respectively on the test set. The combination of radiomic features with clinical features showed superior performance with ROC-AUC of 0.67 in the 5-fold cross-validation and 0.64 in the test set. CONCLUSION In this preliminary study, we achieved a reproducible estimation of patient survival using radiological and clinical features. In future analyses, we will seek to better account for the biological heterogeneity of high-grade glioma by increasing the number of labels and having them corrected by experts in the field.

Najjar YG, Navrazhina K, Ding F, et al . Ipilimumab plus nivolumab for patients with metastatic uveal melanoma: a multicenter, retrospective study. J Immunother Cancer 2020;8:e000331. doi: 10.1136/jitc-2019-000331 This article has been corrected since it was published online. The author, Igor

Comment on Mao YP, Wang SX, Gao TS, et al. Medial retropharyngeal nodal region sparing radiotherapy versus standard radiotherapy in patients with nasopharyngeal carcinoma: open label, non-inferiority, multicentre, randomised, phase 3 trial. BMJ. 2023 Feb 6;380:e072133. The retropharyngeal lymph node region is considered, together with the level II lymph nodes, to be the first echelon nodes of the nasopharyngeal lymphatic drainage, as they are involved in approximately 70% of patients with lymph node metastases.1 Covering the whole retropharyngeal in radiotherapy volumes has been considered to be the standard in the nasopharyngeal carcinoma treatment,2 although there is some evidence to suggest that the rate of lymph node involvement in the medial part of the retropharyngeal region (MRLN), between the pharyngeal constrictor muscles and the prevertebral fascia, is very low, if not negligible, compared to the lateral retropharyngeal region. In a prospective study including 3100 patients, Wang and colleagues observed only 6 patients with involved lymph nodes in the MRLN (0.2%) among the 2679 (86.4%) patients who presented involved lymph nodes.3 Tang and colleagues reported no involved medial retropharyngeal lymph nodes in a retrospective cohort of 924 consecutive patients, while 73.5% (679 of 924 patients) presented a lateral retropharyngeal lymph node metastasis at the initial presentation.4 In another study, Lin and colleagues reported only 5 patients out of 959 patients (0.3%) with lymph node metastases located in the MRLN.5 Despite these rather reassuring data, only one retrospective study evaluating the feasibility of the omission of the MRLN existed until recently.6 In this retrospective case-control study, Pan and colleagues have observed no significant difference between the 2 groups of 183 patients in terms of 5 years-local relapse-free survival LRFS (94.8% vs 94.2%), regional relapse-free survival RRFS (98.3% vs 97.1%),distant metastasis-free survival DMFS (88.5% vs 86.7%), and overall survival OS (85.5% vs 84.7%) for the study group and the control groups respectively, and a significant reduction of the incidence of grade III acute oral mucositis in the study group.6 Mao and colleagues are to be commended for their randomized non-inferiority phase III trial that provided new and strong evidence regarding the pending question of the possible omission of MRLN radiotherapy and the potential benefits in terms of reduced toxicity and improved quality of life in patients with nasopharyngeal cancers.7 This study provides us with a number of valuable information, confirming the pre-existing data. In this well-designed trial, 568 patients were recruited between November 2017, and December 2018, and randomly allocated to the MRLN sparing radiotherapy group (n = 285 patients) and the standard radiotherapy group (n = 283). Radiotherapy delineation, plan, dose and schedule were performed according to the consensus guideline (70 Gy (2.12 Gy/fraction) to the planning target volume (PTV) of the gross tumor at the primary site and the involved regional lymph nodes, 66–70 Gy to the PTV of the involved cervical lymph nodes, 60–62 Gy to the PTV of the high risk clinical target volume (CTV), and 54–56 Gy to the PTV of the low risk CTV. The characteristics of the study population were wide. Stage III and IVa represented respectively 43% and 41% of the patients in the MRLN sparing radiotherapy group,and 46% and 44% in the standard radiotherapy group. Concurrent chemoradiotherapy and Induction chemotherapy followed by concurrent chemoradiotherapy represented respectively 34% and 58% of the patients in the MRLN sparing radiotherapy group, and 32% and 59% in the standard radiotherapy group. A pre-treatment EBV DNA test was available for 88% and 89% patients in the MRLN sparing group and the standard group respectively. The primary endpoint was the three year local relapse-free survival defined as any relapse in the primary site and in the retropharyngeal lymph node region, or death from any cause. Sparing the MRLN allowed to decrease the dose to the superior and middle pharyngeal constrictors. The authors have reached their primary objective and have shown that MRLN sparing radiotherapy is non-inferior to standard radiotherapy, with a three year local relapse-free survival of 95.3% (95% confidence interval 92.8 to 97.8) in the MRLN sparing radiotherapy group, and 95.5% (93.0 to 98.0) in the standard radiotherapy group (estimated absolute difference –0.2% (lower boundary of the one-sided 97.5% confidence interval of –3.6, which is not greater than the predefined non-inferiority margin of –8%). This result was consistent in multivariate and subgroup analyses taking into account the stage of the disease, the type of chemotherapy regimen, and the quantification of EBV DNA at baseline. Only six patients (1.1%) presented a lymph node recurrence in the retropharyngeal region, but none in the MRLN. OS, DMFS, and RRFS were not significantly different between the MRLN sparing radiotherapy group and the standard radiotherapy group (95.2% vs 96.4%, 89.7% vs 92.3%, 96.9% vs 94.0% respectively). The lower DMFS observed in the sparing radiotherapy group while the RRFS seemed better (HR 0.67, P = 0.28) may probably be explained by randomness. Finally, sparing MRLN has shown to decrease the incidence of grade ≥1 acute dysphagia (25.5% v 35.1%, P = 0.01) and late dysphagia (24.0% v 34.3%, P = 0.008), and to significantly improve quality of life on multiple domains such as global health status, role functioning, social functioning, fatigue and swallowing. Overall, this study provides a high level of evidence regarding the feasibility of sparing MRLN and the potential benefit on patient quality of life by reducing treatment toxicity. These results thus make it possible to envisage a new lever of action for radiation oncologists, making it possible to improve the tolerance of the treatment, which is important in this population where the prognosis is rather good, and the survival rather long. They are part of the de-escalation approach along with the studies showing the possibility to spare the level IV and Vb of the uninvolved neck of N0-N1 nasopharyngeal cancer patients8 or the studies showing the benefit of elective irradiation to neck level Ib.9 This study has the added advantage of showing the possibility of omitting MRLN irradiation in virtually all patients with cavum cancer, regardless of their stage. Some few questions are still pending such as the generalizability of these findings to a non-endemic nasopharyngeal carcinoma cohort population, and the compatibility of the MRLN sparing with other de-escalate approaches such as dose de-escalation. However, we do believe that the results provided by Mao and colleagues’ study need to be consider in future guidelines of nasopharyngeal non-keratinising carcinoma. In the future, the implementation of even more conformal radiotherapy techniques, such as proton therapy, will certainly allow to gain more benefit of the sparing of MRLN through further dose reduction to the swallowing structures.

Abstract BACKGROUND Predicting progression-free survival (PFS) in glioblastoma (GB) is crucial for post-surgery treatment decisions. Detecting post-operative volumes via pre-radiotherapy (preRT) MRI, combined with clinical and molecular data, improves prognosis. A decision tree model using these features provides valuable insights into PFS when aligned with the new RANO resection categories. MATERIAL AND METHODS We analyzed retrospective data from 205 GB treated by STUPP protocol across three French centers (2009-2021) as part of the AIDREAM project. Data included clinical and anatomopathological information. Advanced automated segmentation of tumor subregions on preRT MRI was used, subsequently revalidated by experts, and a detailed brain atlas to refine lesion site. A survival analysis was performed using Kaplan-Meier with log-rank tests to compare survival distributions according to clinical data, surgical extension and the 4 RANO volume subcategories. A conventional univariate and multivariate Cox proportional HR analysis based on clinical/anatomopathological data and residual non-contrast-enhanced (nonCE) and contrast-enhanced (CE) volumes on pre-RT MRI was also conducted. The SurvivalTree algorithm, chosen for its capability to handle survival data and missing values, was applied in two experimental setups: one using only clinical data and another using combined clinical and volumetric data. Variables included age, sex, KPS, surgical extension, MGMT, tumor site and nonCE, CE, tumor bed and postoperative changes volumes. In our 5-fold cross-validation process, we individually assessed the importance of each feature in every fold through permutations. This approach enabled us to evaluate the consistency and dependability of feature importance across various data subsets. We then calculated the mean importance of each feature over all folds. RESULTS Median PFS was 8 months (1-62) and median OS was 18 months (2- 93). Based on the log-rank test, it was shown that extensive surgical resection and lower RANO categories improved PFS, with significant p-values. MGMT status correlated strongly with longer PFS (p &amp;lt; 0.001). Univariate analysis highlighted significant predictors including nonCE (p=0.002) and CE volume (p=0.017), while multivariate analysis confirmed only the significance of nonCE volume (HR 1.011, 95% CI 1.005-1.018, p &amp;lt; 0.001). The decision tree model’s performance increased with volumetric data, from a mean C-index of 0.546 ± 0.057 to 0.576 ± 0.066, identifying CE and nonCE volumes and MGMT status as key predictive factors. CONCLUSION Integrating surgical, volumetric, and molecular data into a decision tree model could effectively enhance the prediction of PFS in glioblastoma. Expanding the study to the entire AIDREAM cohort of 733 patients will validate the robustness of the model and its clinical utility for personalized treatment strategies.

The treatment of head and neck cancer remains a press- ing challenge in the realm of oncology. Particularly, the precise targeting in radiotherapy demands a thorough un- derstanding of the Gross Tumor Volume (GTV). However, with the persistent issue of interobserver variability and in- accuracy in GTV demarcation due to the low quality of available image acquisitions, the necessity for better tools and methodologies becomes paramount. It underscores the need for integrating diverse data sources for a comprehen- sive understanding of the tumor’s spatial extent and bio- logical characteristics.Histology and radiology, while both essential in onco- logical diagnostics, offer multi-scale information about the tumor whose synergy is often under-exploited. While radi- ology provides a macroscopic view, capturing the tumor’s overall structure, size, and location, histology delves into the microscopic, elucidating cellular and tissue-level de- tails. The granularity and precision of histological data, juxtaposed with the broader perspectives of radiological im- agery, advocate for their fusion, which can potentially rev- olutionize our understanding of tumor characteristics and their spatial distribution.Registration stands as a pivotal technique to bridge these modalities embedding multiscale information. By aligning histological slides spatially with their correspond- ing radiological scans, registration facilitates a direct pixel- wise comparison. However, this task is highly technical due to the substantial differences between these modalities and the extreme deformations that the tissue undergoes from in vivo acquisition to a tissue slide from ex vivo resected spec- imen. Our deep learning method StructuRegNet emerged as our answer to the challenges of this alignment, harness-ing rigid structures like cartilage to progressively guide the mapping. By automating this traditionally manual task, we set the foundation for a seamless integration of histological and radiological insights.With the capabilities provided by StructuRegNet, di- rect comparisons between both modalities became feasible, especially in assessing the GTV and its delineation on histo- logical data. This comparison revealed systematic overes- timations in conventional GTV definitions. Building upon this finding, we introduced a diffusion-based segmentation model tailored for histological labels on CT scans. Given that these labels are of superior quality, the model could sidestep the pitfalls encountered by previous models focused solely on GTV. This approach illuminated the path towards histopathology-enhanced GTV and introduced the concept of ambiguous delineations, hinting at the potential of non- binary volumetric dose painting in radiotherapy.Shifting from spatial to feature-level fusion, the SMuRF framework was introduced. Instead of merely relying on spatial correlations, SMuRF operates at a deeper level, fo- cusing on the inherent features and patterns within the data. Through this advanced fusion leveraging cutting-edge computer vision and deep learning methods, we achieved notable successes in predicting cancer grade and survival, outperforming traditional monomodal methods.In summary, this research underscores the transforma- tive potential of integrating histological and radiological data, augmented by artificial intelligence, in refining head and neck cancer radiotherapy. By fusing macroscopic and microscopic insights, the work paints a promising picture of individualized, precision-driven oncology treatments for the future.