Search

Deep learning models based on medical images play an increasingly important role for cancer outcome prediction. The standard approach involves usage of convolutional neural networks (CNNs) to automatically extract relevant features from the patient's image and perform a binary classification of the occurrence of a given clinical endpoint. In this work, a 2D-CNN and a 3D-CNN for the binary classification of distant metastasis (DM) occurrence in head and neck cancer patients were extended to perform time-to-event analysis. The newly built CNNs incorporate censoring information and output DM-free probability curves as a function of time for every patient. In total, 1037 patients were used to build and assess the performance of the time-to-event model. Training and validation was based on 294 patients also used in a previous benchmark classification study while for testing 743 patients from three independent cohorts were used. The best network could reproduce the good results from 3-fold cross validation [Harrell's concordance indices (HCIs) of 0.78, 0.74 and 0.80] in two out of three testing cohorts (HCIs of 0.88, 0.67 and 0.77). Additionally, the capability of the models for patient stratification into high and low-risk groups was investigated, the CNNs being able to significantly stratify all three testing cohorts. Results suggest that image-based deep learning models show good reliability for DM time-to-event analysis and could be used for treatment personalisation.

Objectives: The objective of the paper was to assess real-life experience in the management of head and neck cancer (HNC) patients during the COVID-19 pandemic in radiotherapy departments and to evaluate the variability in terms of adherence to American Society of Radiation Oncology (ASTRO) and European Society for Radiotherapy and Oncology (ESTRO) recommendations., Materials and Methods: In May 2020, an anonymous 30-question online survey, comparing acute phase of outbreak and pre-COVID-19 period, was conducted. Two sections exploited changes in general management of HNC patients and different HNC primary tumors, addressing specific statements from ASTRO ESTRO consensus statement as well., Results: Eighty-eight questionnaires were included in the demographic/clinical workflow analysis, and 64 were analyzed for treatment management. Forty-eight percent of radiotherapy departments became part of oncologic hubs. First consultations reduced, and patients were addressed to other centers in 33.8 and 18.3% of cases, respectively. Telematic consultations were used in 50% of follow-up visits and 73.9% of multidisciplinary tumor board discussions. There were no practical changes in the management of patients affected by different primitive HNCs. Hypofractionation was not favored over conventional schedules., Conclusions: Compared to pre-COVID era, the clinical workflow was highly re-organized, whereas there were no consistent changes in RT indications and schedules.

Management of patients with head and neck cancers (HNCs) is challenging for the Radiation Oncologist, especially in the COVID-19 era. The Italian Society of Radiotherapy and Clinical Oncology (AIRO) identified the need of practice recommendations on logistic issues, treatment delivery and healthcare personnel's protection in a time of limited resources. A panel of 15 national experts on HNCs completed a modified Delphi process. A five-point Likert scale was used; the chosen cut-offs for strong agreement and agreement were 75% and 66%, respectively. Items were organized into two sections: (1) general recommendations (10 items) and (2) special recommendations (45 items), detailing a set of procedures to be applied to all specific phases of the Radiation Oncology workflow. The distribution of facilities across the country was as follows: 47% Northern, 33% Central and 20% Southern regions. There was agreement or strong agreement across the majority (93%) of proposed items including treatment strategies, use of personal protection devices, set-up modifications and follow-up re-scheduling. Guaranteeing treatment delivery for HNC patients is well-recognized in Radiation Oncology. Our recommendations provide a flexible tool for management both in the pandemic and post-pandemic phase of the COVID-19 outbreak.

Improvements in prognosis of head-and-neck squamous cell carcinoma (HNSCC) have paralleled with an increase in health-care costs, so that an economic evaluation is of growing importance. Presently, most of the evidence is from insurance-based studies in the USA. Between 2007 and 2010, 879 HNSCC patients were identified through the population-based cancer registry of the Friuli Venezia Giulia region, including 266 oral, 187 oropharyngeal, 136 hypopharyngeal, and 290 laryngeal cancers. Health-care costs from diagnosis to treatment initiation and in the following 2 years were retrieved through a record linkage with the regional health data warehouse. This database collected comprehensive health information on all resident citizens. Generalized linear models with a gamma distribution and log-link function were applied to model costs. The average health-care cost from diagnosis up to 2 years after treatment initiation was €20,184 (95% confidence interval: €19,634 - 20,733). Heterogeneity emerged according to cancer site, elective treatment, and retreatment for cancer persistence/recurrence (no: €13,896; yes: €24,599; p < 0.001). An advanced stage was associated with increased costs stage (I: €12,969; II: €18,276; III: €26,229; IV: €25,574; p < 0.001) as the result of treatment complexity and elevated frequency of patients retreatment due to recurrence. These findings further support strategies to diagnose patients at an earlier cancer stage and the accurate definition of diagnostic and treatment pathways, to start treating patients when radical unimodal approach is still feasible. Besides the advantage in prognosis due to timely curative treatments, this would reduce the economic burden of cancer treatment.

We analyzed the distribution of DRB1, DQA1, DQB1, and DPB1 allelic variants in 48 rheumatoid arthritis (RA) patients, compared with 109 Italian random controls, using PCR amplification and hybridization with specific oligonucleotides. We confirm the previously reported increase of DR4 specificity, in comparison with healthy Italian individuals. In particular, we find a statistically significant positive association of DRB1*0401 and DRB1*0404 alleles with RA. However, when we compare the DR4+ groups, none of the DRB1*04 alleles is increased in the RA group. By sequence analysis, performed on 10 patients, we demonstrate that the DRB1*04 genes of RA show no difference from the DRB1*04 sequences previously published. From the molecular analysis of the other DRB1 polymorphic variants, we find a trend of positive association of DRB1*0101 in DR4-negative patients versus DR4-negative healthy controls and, in the group of DR4-negative and/or DR1-negative patients, a similar increase of DRB1*06. Also, we observe in RA patients a statistically significant increase of DQA1*0301 and DQB1*0302 accompanied by a significant decrease of DQA1*0201, DQA1*0501 and DQB1*0201. Finally, from the analysis of DPB1 gene, it can be assessed that the distribution of DPB1 alleles does not differ significantly between RA patients and healthy controls.

Over a period of seven years (from 1980 to 1987) 58 acute viral hepatitis A cases were admitted to the Departments for Infectious Diseases. This number represents 10.76% of all viral hepatitis patients. Most of the patients had travelled to foreign countries or to areas in Italy with high hepatitis A virus (HAV) circulation. An assessment of specific immunization against HAV in normal healthy people aged 1-60, performed by ELISA, showed a very low number of sero-positive subjects. The disappearance of HAV in Italy, which is similar to other western European countries, is probably due to the improved socio-economical conditions; the "imported cases" seem to have little impact on public health.

10363 sera from healthy adults, volunteer blood donors (18-65 year aged) have been tested for detection of HBsAg. The frequency of Australia antigen was found to be 0,88% in all sera; 1,01% in male and 0,24% in female donors. The frequency in volunteer blood donors of both sexes has been in the first half of the life (18-40) of 1,07% and in the second half (41-65) of 0,58%. Possible epidemiological significances were discussed.