Introduction: An increasing number of parameters can be considered when making decisions in oncology. Tumor characteristics can also be extracted from imaging through the use of radiomics and add to this wealth of clinical data. Machine learning can encompass these parameters and thus enhance clinical decision as well as radiotherapy workflow. Methods: We performed a description of machine learning applications at each step of treatment by radiotherapy in head and neck cancers. We then performed a systematic review on radiomics and machine learning outcome prediction models in head and neck cancers. Results: Machine Learning has several promising applications in treatment planning with automatic organ at risk delineation improvements and adaptative radiotherapy workflow automation. It may also provide new approaches for Normal Tissue Complication Probability models. Radiomics may provide additional data on tumors for improved machine learning powered predictive models, not only on survival, but also on risk of distant metastasis, in field recurrence, HPV status and extra nodal spread. However, most studies provide preliminary data requiring further validation. Conclusion: Promising perspectives arise from machine learning applications and radiomics based models, yet further data are necessary for their implementation in daily care.

Modern radiation therapy techniques are characterized by high conformality to tumor volumes and steep dose gradients to spare normal organs. These techniques require accurate clinical target volume definitions and rigorous assessment of set up uncertainties using image guidance, a concept called image-guided radiation therapy. Due to alteration of patient anatomy, changes in tissue density/volumes and tumor shrinkage over the course of treatment, treatment accuracy may be challenged. This may result in excessive irradiation of organs at risk/healthy tissues and undercoverage of target volumes with a significant risk of locoregional failure. Adaptive radiation therapy (ART) is a concept allowing the clinician to reconsider the planned dose based on potential changes to accurately delivering the remaining radiation dose to the tumor while optimally minimizing irradiation of healthy tissues. There is little consensus on how to apply this concept in clinical practice. The current review investigates the current ART issues, including patient selection, clinical/dosimetric criteria and timing for re-planning, and practical technical issues. A practical algorithm is proposed for patient management in cases where ART is required.

Introduction:Stereotactic hypofractionated radiotherapy is an effective treatment for brain metastases in oligometastatic patients. Its planning is however time-consuming because of the number of organs at risk to be manually segmented. This study evaluates 2 automated segmentation commercial software.Methods:Patients were scanned in the treatment position. The computed tomography scan was registered on a magnetic resonance imaging and volumes were manually segmented by a clinician. Then 2 automated segmentations were performed (with iPlan and Smart Segmentation). RT STRUCT files were compared with Aquilab’s Artistruct segment comparison module. We selected common segmented volume ratio as the main judging criterion. Secondary criteria were Dice-Sørensen coefficients, overlap ratio, and additional segmented volume.Results:Twenty consecutive patients were included. Agreement between manual and automated contouring was poor. Common segmented volumes ranged from 7.71% to 82.54%, Dice-Sørensen coefficient ranged from 0.0745 to 0.8398, overlap ratio ranged from 0.0414 to 0.7275, and additional segmented volume ranged from 9.80% to 92.25%. Each software outperformed the other on some organs while performing worse on others.Conclusion:No software seemed clearly better than the other. Common segmented volumes were much too low for routine use in stereotactic hypofractionated brain radiotherapy. Manual editing is still needed.

Conventional cemented acetabular components are reported to have a high rate of failure when implanted into previously irradiated bone. We recommend the use of a cemented reconstruction with the addition of an acetabular reinforcement cross to improve fixation. We reviewed a cohort of 45 patients (49 hips) who had undergone irradiation of the pelvis and a cemented total hip arthroplasty (THA) with an acetabular reinforcement cross. All hips had received a minimum dose of 30 Gray (Gy) to treat a primary nearby tumour or metastasis. The median dose of radiation was 50 Gy (Q1 to Q3: 45 to 60; mean: 49.57, 32 to 72). The mean follow-up after THA was 51 months (17 to 137). The cumulative probability of revision of the acetabular component for a mechanical reason was 0% (0 to 0%) at 24 months, 2.9% (0.2 to 13.3%) at 60 months and 2.9% (0.2% to 13.3%) at 120 months, respectively. One hip was revised for mechanical failure and three for infection. Cemented acetabular components with a reinforcement cross provide good medium-term fixation after pelvic irradiation. These patients are at a higher risk of developing infection of their THA.

Helical tomotherapy reduced the incidence and severity of xerostomia. A mean dose to the parotid between 20 and 26Gy allowed preservation of salivary function without compromising treatment efficacy. However, parotid-sparing HT requiring a mean dose less than 20Gy is associated with an increased risk of recurrence.

Previous radiation is associated with an increased risk of amputation and reoperation for SSI/WC when treating a local recurrence. This information should be accounted for when deciding for the use of radiation.

Background: Sarcopenia appears to be a negative prognostic factor for poor survival outcomes and worse treatment tolerance in patients with head-and-neck squamous cell carcinoma (HNSCC). We evaluated sarcopenia's impact on overall survival (OS), disease-free survival (DFS) and chemo-radiation tolerance in patients with head-and-neck cancer (HNC) treated with chemoradiotherapy (CRT) from a monocentric observational study. Methods:We identified patients with HNC treated by CRT between 2009 and 2018 with pretreatment imaging using positron emission tomography-computed tomography scans (PET/CT). Sarcopenia was measured using the pretreatment PET/CT at the L3 vertebral body using previously published methods. Clinical variables were retrospectively retrieved. Results:Of 216 patients identified, 54 patients (25.47%) met the criteria for sarcopenia. These patients had a lower mean body mass index before treatment (21.92 vs. 25.65 cm/m 2 , p < 0.001) and were more likely to have a history of smoking (88.89% vs. 71.52%, p = 0.01), alcohol use (55.56% vs. 38.61%, p = 0.03) and positive human papilloma virus status (67.74% vs. 41.75%, p = 0.011). At 3 years of follow-up, OS and DFS were 75% and 70% versus 82% and 85% for sarcopenic and non-sarcopenic patients, respectively (p = 0.1 and p = 0.00015). On multivariate analysis, sarcopenia appeared as a pejorative factor on DFS (hazard ratio 2.174, p = 0.0001) in the overall cohort. Sarcopenic patients did not require more chemotherapy and radiation-treatment interruptions and did not suffer from more chemo-induced and radiation-induced grade 3-4 toxicities than their nonsarcopenic counterparts. Conclusion:Sarcopenia in HNSCC patients is an independent adverse prognostic factor for DFS after definitive chemoradiotherapy.

No abstract