Explainable prediction model for the human papillomavirus status in patients with oropharyngeal squamous cell carcinoma using CNN on CT images.

Several studies have emphasised how positive and negative human papillomavirus (HPV+ and HPV-, respectively) oropharyngeal squamous cell carcinoma (OPSCC) has distinct molecular profiles, tumor characteristics, and disease outcomes. Different radiomics-based prediction models have been proposed, by also using innovative techniques such as Convolutional Neural Networks (CNNs). Although some of these models reached encouraging predictive performances, there evidence explaining the role of radiomic features in achieving a specific outcome is scarce. In this paper, we propose some preliminary results related to an explainable CNN-based model to predict HPV status in OPSCC patients. We extracted the Gross Tumor Volume (GTV) of pre-treatment CT images related to 499 patients (356 HPV+ and 143 HPV-) included into the OPC-Radiomics public dataset to train an end-to-end Inception-V3 CNN architecture. We also collected a multicentric dataset consisting of 92 patients (43 HPV+ , 49 HPV-), which was employed as an independent test set. Finally, we applied Gradient-weighted Class Activation Mapping (Grad-CAM) technique to highlight the most informative areas with respect to the predicted outcome. The proposed model reached an AUC value of 73.50% on the independent test. As a result of the Grad-CAM algorithm, the most informative areas related to the correctly classified HPV+ patients were located into the intratumoral area. Conversely, the most important areas referred to the tumor edges. Finally, since the proposed model provided additional information with respect to the accuracy of the classification given by the visualization of the areas of greatest interest for predictive purposes for each case examined, it could contribute to increase confidence in using computer-based predictive models in the actual clinical practice.

Corrigendum: Transfer learning approach based on computed tomography images for predicting late xerostomia after radiotherapy in patients with oropharyngeal cancer.

[This corrects the article DOI: 10.3389/fmed.2022.993395.].

Transfer learning approach based on computed tomography images for predicting late xerostomia after radiotherapy in patients with oropharyngeal cancer.

Background and purpose: Although the latest breakthroughs in radiotherapy (RT) techniques have led to a decrease in adverse event rates, these techniques are still associated with substantial toxicity, including xerostomia. Imaging biomarkers could be useful to predict the toxicity risk related to each individual patient. Our preliminary work aims to develop a radiomic-based support tool exploiting pre-treatment CT images to predict late xerostomia risk in 3 months after RT in patients with oropharyngeal cancer (OPC). Materials and methods: We performed a multicenter data collection. We enrolled 61 patients referred to three care centers in Apulia, Italy, out of which 22 patients experienced at least mild xerostomia 3 months after the end of the RT cycle. Pre-treatment CT images, clinical and dose features, and alcohol-smoking habits were collected. We proposed a transfer learning approach to extract quantitative imaging features from CT images by means of a pre-trained convolutional neural network (CNN) architecture. An optimal feature subset was then identified to train an SVM classifier. To evaluate the robustness of the proposed model with respect to different manual contouring practices on CTs, we repeated the same image analysis pipeline on "fake" parotid contours. Results: The best performances were achieved by the model exploiting the radiomic features alone. On the independent test, the model reached median AUC, accuracy, sensitivity, and specificity values of 81.17, 83.33, 71.43, and 90.91%, respectively. The model was robust with respect to diverse manual parotid contouring procedures. Conclusion: Radiomic analysis could help to develop a valid support tool for clinicians in planning radiotherapy treatment, by providing a risk score of the toxicity development for each individual patient, thus improving the quality of life of the same patient, without compromising patient care.

Long-Lasting Remission in De Novo Breast Myeloid Sarcoma Treated with Decitabine and Radiotherapy.

Myeloid sarcoma (MS) represents a rare disease with an adverse clinical outcome for patients not candidate to acute myeloid leukemia (AML)-like chemotherapies. Here we present the case of an elderly patient affected by a bilateral breast localization of MS treated with the hypomethylating agent decitabine associated to radiotherapy. The association of the two treatment modalities has allowed an optimal and long-lasting disease control.

Waiting time for radiation therapy after breast-conserving surgery in early breast cancer: a retrospective analysis of local relapse and distant metastases in 615 patients.

BACKGROUND: Postoperative radiotherapy after breast-conserving surgery (BCS) is the standard in the management of breast cancer. The optimal timing for starting postoperative radiation therapy has not yet been well defined. In this study, we aimed to evaluate if the time interval between BCS and postoperative radiotherapy is related to the incidence of local and distant relapse in women with early node-negative breast cancer not receiving chemotherapy. METHODS: We retrospectively analyzed clinical data concerning 615 women treated from 1984 to 2010, divided into three groups according to the timing of radiotherapy: ≤60, 61-120, and >120 days. To estimate the presence of imbalanced distribution of prognostic and treatment factors among the three groups, the χ2 test or the Fisher exact test were performed. Local relapse-free survival, distant metastasis-free survival (DMFS), and disease-free survival (DFS) were estimated with the Kaplan-Meier method, and multivariate Cox regression was used to test for the independent effect of timing of RT after adjusting for known confounding factors. The median follow-up time was 65.8 months. RESULTS: Differences in distribution of age, type of hormone therapy, and year of diagnosis were statistically significant. At 15-year follow-up, we failed to detect a significant correlation between time interval and the risk of local relapse (p = 0.09) both at the univariate and the multivariate analysis. The DMFS and the DFS univariate analysis showed a decreased outcome when radiotherapy was started early (p = 0.041 and p = 0.046), but this was not confirmed at the multivariate analysis (p = 0.406 and p = 0.102, respectively). CONCLUSIONS: Our results show that no correlation exists between the timing of postoperative radiotherapy and the risk of local relapse or distant metastasis development in a particular subgroup of women with node-negative early breast cancer.

Targeting mast cells tryptase in tumor microenvironment: a potential antiangiogenetic strategy.

Angiogenesis is a complex process finely regulated by the balance between angiogenesis stimulators and inhibitors. As a result of proangiogenic factors overexpression, it plays a crucial role in cancer development. Although initially mast cells (MCs) role has been defined in hypersensitivity reactions and in immunity, it has been discovered that MCs have a crucial interplay on the regulatory function between inflammatory and tumor cells through the release of classical proangiogenic factors (e.g., vascular endothelial growth factor) and nonclassical proangiogenic mediators granule-associated (mainly tryptase). In fact, in several animal and human malignancies, MCs density is highly correlated with tumor angiogenesis. In particular, tryptase, an agonist of the proteinase-activated receptor-2 (PAR-2), represents one of the most powerful angiogenic mediators released by human MCs after c-Kit receptor activation. This protease, acting on PAR-2 by its proteolytic activity, has angiogenic activity stimulating both human vascular endothelial and tumor cell proliferation in paracrine manner, helping tumor cell invasion and metastasis. Based on literature data it is shown that tryptase may represent a promising target in cancer treatment due to its proangiogenic activity. Here we focused on molecular mechanisms of three tryptase inhibitors (gabexate mesylate, nafamostat mesylate, and tranilast) in order to consider their prospective role in cancer therapy.