Objective: To report an overview and update on Artificial Intelligence (AI) and COVID-19 using chest Computed Tomography (CT) scan and chest X-ray images (CXR). Machine Learning and Deep Learning Approaches for Diagnosis and Treatment were identified. Methods: Several electronic datasets were analyzed. The search covered the years from January 2019 to June 2021. The inclusion criteria were studied evaluating the use of AI methods in COVID-19 disease reporting performance results in terms of accuracy or precision or area under Receiver Operating Characteristic (ROC) curve (AUC). Results: Twenty-two studies met the inclusion criteria: 13 papers were based on AI in CXR and 10 based on AI in CT. The summarized mean value of the accuracy and precision of CXR in COVID-19 disease were 93.7% ± 10.0% of standard deviation (range 68.4–99.9%) and 95.7% ± 7.1% of standard deviation (range 83.0–100.0%), respectively. The summarized mean value of the accuracy and specificity of CT in COVID-19 disease were 89.1% ± 7.3% of standard deviation (range 78.0–99.9%) and 94.5 ± 6.4% of standard deviation (range 86.0–100.0%), respectively. No statistically significant difference in summarized accuracy mean value between CXR and CT was observed using the Chi square test (p value > 0.05). Conclusions: Summarized accuracy of the selected papers is high but there was an important variability; however, less in CT studies compared to CXR studies. Nonetheless, AI approaches could be used in the identification of disease clusters, monitoring of cases, prediction of the future outbreaks, mortality risk, COVID-19 diagnosis, and disease management.

Locally advanced rectal cancer (LARC) is a heterogeneous group of tumors where a risk-adapted therapeutic strategy is needed. Short-course radiotherapy (SCRT) is a more convenient option for LARC patients than preoperative long-course RT plus capecitabine. Histone-deacetylase inhibitors (HDACi) have shown activity in combination with RT and chemotherapy in the treatment of solid tumors. Valproic acid (VPA) is an anti-epileptic drug with HDACi and anticancer activity. In preclinical studies, our group showed that the addition of HDACi, including VPA, to capecitabine produces synergistic antitumour effects by up-regulating thymidine phosphorylase (TP), the key enzyme converting capecitabine to 5-FU, and by downregulating thymidylate synthase (TS), the 5-FU target.

To evaluate MRI for neoadjuvant therapy response assessment in locally advanced rectal cancer (LARC) using dynamic contrast enhanced-MRI (DCE-MRI) and diffusion weighted imaging (DWI), we have compared magnetic resonance volumetry based on DCE-MRI (V(DCE)) and on DWI (V(DWI)) scans with conventional T2-weighted volumetry (V(C)) in LARC patients after neoadjuvant therapy. Twenty-nine patients with LARC underwent MR examination before and after neoadjuvant therapy. A manual segmentation was performed on DCE-MR postcontrast images, on DWI (b-value 800 s/mm²), and on conventional T2-weighted images by two radiologists. DCE-MRI, DWI, and T2-weigthed volumetric changes before and after treatment were evaluated. Nonparametric sample tests, interobserver agreement, and receiver operating characteristic curve (ROC) were performed. Diagnostic performance linked to DCE-MRI volumetric change was superior to T2-w and DW-MRI volumetric changes performance (specificity 86%, sensitivity 93%, and accuracy 93%). Area Under ROC (AUC) ofV(DCE) was greater than AUCs ofV(C) andV(DWI) resulting in an increase of 15.6% and 11.1%, respectively. Interobserver agreement between two radiologists was 0.977, 0.864, and 0.756 forV(C),V(DCE), andV(DWI), respectively.V(DCE) seems to be a promising tool for therapy response assessment in LARC. Further studies on large series of patients are needed to refine technique and evaluate its potential value.

Background: Our aim was to investigate preoperative chemoradiation therapy (pCRT) response in locally advanced rectal cancer (LARC) comparing standardized index of shape (SIS) obtained from dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) with intravoxel-incoherent-motion-modelling-derived parameters by diffusion-weighted imaging (DWI). Materials and methods: Eighty-eight patients with LARC were subjected to MRI before and after pCRT. Apparent diffusion coefficient (ADC), tissue diffusion (D _t ), pseudodiffusion (D _p ) and perfusion fraction (f) were calculated and percentage changes ∆ADC, ∆D _t , ∆D _p , ∆f were computed. SIS was derived comparing DCE-MRI pre- and post-pCRT. Nonparametric tests and receiver operating characteristic (ROC) curves were performed. Results: A total of 52 patients were classified as responders (tumour regression grade; TRG ⩽ 2) and 36 as not-responders (TRG > 3). Mann–Whitney U test showed statistically significant differences in SIS, ∆ADC and ∆D _t between responders and not-responders and between complete responders (19 patients with TRG = 1) versus incomplete responders. The best parameters to discriminate responders by nonresponders were SIS and ∆ADC, with an accuracy of 91% and 82% (cutoffs of −5.2% and 18.7%, respectively); the best parameters to detect pathological complete responders were SIS, ∆f and ∆D _p with an accuracy of 78% (cutoffs of 38.5%, 60.0% and 83.0%, respectively). No increase of performance was observed by combining linearly each possible couple of parameters or combining all parameters. Conclusion: SIS allows assessment of preoperative treatment response with high accuracy guiding the surgeon versus more or less conservative treatment. DWI-derived parameters reached less accuracy compared with SIS and combining linearly DCE- and DWI-derived parameters; no increase of accuracy was obtained.

Despite the improvements in diagnosis and treatment, colorectal cancer (CRC) is the second cause of cancer deaths in both sexes. Therefore, research in this field remains of great interest. The approval of bevacizumab, a humanized anti-vascular endothelial growth factor (VEGF) monoclonal antibody, in combination with a fluoropyrimidine-based chemotherapy in the treatment of metastatic CRC has changed the oncology practice in this disease. However, the efficacy of bevacizumab-based treatment, has thus far been rather modest. Efforts are ongoing to understand the better way to combine bevacizumab and chemotherapy, and to identify valid predictive biomarkers of benefit to avoid unnecessary and costly therapy to nonresponder patients. The BRANCH study in high-risk locally advanced rectal cancer patients showed that varying bevacizumab schedule may impact on the feasibility and efficacy of chemo-radiotherapy. OBELICS is a multicentre, open-label, randomised phase 3 trial comparing in mCRC patients two treatment arms (1:1): standard concomitant administration of bevacizumab with chemotherapy (mFOLFOX/OXXEL regimen) vs experimental sequential bevacizumab given 4 days before chemotherapy, as first or second treatment line. Primary end point is the objective response rate (ORR) measured according to RECIST criteria. A sample size of 230 patients was calculated allowing reliable assessment in all plausible first-second line case-mix conditions, with a 80 % statistical power and 2-sided alpha error of 0.05. Secondary endpoints are progression free-survival (PFS), overall survival (OS), toxicity and quality of life. The evaluation of the potential predictive role of several circulating biomarkers (circulating endothelial cells and progenitors, VEGF and VEGF-R SNPs, cytokines, microRNAs, free circulating DNA) as well as the value of the early [18F]-Fluorodeoxyglucose positron emission tomography (FDG-PET) response, are the objectives of the traslational project. Overall this study could optimize bevacizumab scheduling in combination with chemotherapy in mCRC patients. Moreover, correlative studies could improve the knowledge of the mechanisms by which bevacizumab enhance chemotherapy effect and could identify early predictors of response. EudraCT Number: 2011-004997-27 ClinicalTrials.gove number, NCT01718873