Machine learning to predict overall short-term mortality in cutaneous melanoma.

BACKGROUND: Cutaneous malignant melanoma (CMM) ranks among the ten most frequent malignancies, clinicopathological staging being of key importance to predict prognosis. Artificial intelligence (AI) has been recently applied to develop prognostically reliable staging systems for CMM. This study aims to provide a useful machine learning based tool to predict the overall CMM short-term survival. METHODS: CMM records as collected at the Veneto Cancer Registry (RTV) and at the Veneto regional health service were considered. A univariate Cox regression validated the strength and direction of each independent variable with overall mortality. A range of machine learning models (Logistic Regression classifier, Support-Vector Machine, Random Forest, Gradient Boosting, and k-Nearest Neighbors) and a Deep Neural Network were then trained to predict the 3-years mortality probability. Five-fold cross-validation and Grid Search were performed to test the best data preprocessing procedures, features selection, and to optimize models hyperparameters. A final evaluation was carried out on a separate test set in terms of balanced accuracy, precision, recall and F1 score. The best model was deployed as online tool. RESULTS: The univariate analysis confirmed the significant prognostic value of TNM staging. Adjunctive clinicopathological variables not included in the AJCC 8th melanoma staging system, i.e., sex, tumor site, histotype, growth phase, and age, were significantly linked to overall survival. Among the models, the Neural Network and the Random Forest models featured the best prognostic performance, achieving a balanced accuracy of 91% and 88%, respectively. According to the Gini importance score, age, T and M stages, mitotic count, and ulceration appeared to be the variables with the greatest impact on survival prediction. CONCLUSIONS: Using data from patients with CMM, we developed an AI algorithm with high staging reliability, on top of which a web tool was implemented ( unipd.link/melanomaprediction ). Being essentially based on routinely recorded clinicopathological variables, it can already be implemented with minimal effort and further tested in the current clinical practice, an essential phase for validating the model's accuracy beyond the original research context.

Androgen-deprivation therapies for prostate cancer and risk of infection by SARS-CoV-2: a population-based study (N = 4532).

BACKGROUND: Cell entry of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) depends on binding of the viral spike (S) proteins to angiotensin-converting enzyme 2 and on S protein priming by TMPRSS2. Inhibition of TMPRSS2 may work to block or decrease the severity of SARS-CoV-2 infections. Intriguingly, TMPRSS2 is an androgen-regulated gene that is up-regulated in prostate cancer where it supports tumor progression and is involved in a frequent genetic translocation with the ERG gene. First- or second-generation androgen-deprivation therapies (ADTs) decrease the levels of TMPRSS2. Here we put forward the hypothesis that ADTs may protect patients affected by prostate cancer from SARS-CoV-2 infections. MATERIALS AND METHODS: We extracted data regarding 9280 patients (4532 males) with laboratory-confirmed SARS-CoV-2 infection from 68 hospitals in Veneto, one of the Italian regions that was most affected by the coronavirus disease 2019 (COVID-19) pandemic. The parameters used for each COVID-19-positive patient were sex, hospitalization, admission to intensive care unit, death, tumor diagnosis, prostate cancer diagnosis, and ADT. RESULTS: There were evaluable 9280 SARS-CoV-2-positive patients in Veneto on 1 April 2020. Overall, males developed more severe complications, were more frequently hospitalized, and had a worse clinical outcome than females. Considering only the Veneto male population (2.4 million men), 0.2% and 0.3% of non-cancer and cancer patients, respectively, tested positive for SARS-CoV-2. Comparing the total number of SARS-CoV-2-positive cases, prostate cancer patients receiving ADT had a significantly lower risk of SARS-CoV-2 infection compared with patients who did not receive ADT (OR 4.05; 95% CI 1.55-10.59). A greater difference was found comparing prostate cancer patients receiving ADT with patients with any other type of cancer (OR 4.86; 95% CI 1.88-12.56). CONCLUSION: Our data suggest that cancer patients have an increased risk of SARS-CoV-2 infections compared with non-cancer patients. However, prostate cancer patients receiving ADT appear to be partially protected from SARS-CoV-2 infections.

The role of immune microenvironment in small-cell lung cancer: Distribution of PD-L1 expression and prognostic role of FOXP3-positive tumour infiltrating lymphocytes.

INTRODUCTION: The prognosis of small-cell lung cancer (SCLC) is dismal and new effective therapies are needed. Immunotherapy looks promising, but no molecular predictive markers are currently available, and data on immune microenvironment are very limited. METHODS: We retrospectively analysed 104 SCLC cases. Immunohistochemistry evaluation of PD-L1 was performed both on tumour cells (TCs) and on tumour-infiltrating immune cells (TIICs) by using anti-PD-L1 22C3 antibody (DAKO) and categorised by using 1% as cut-off point. Tumour-infiltrating lymphocytes (TILs) were characterised by using anti-CD8 and anti-FOXP3 antibodies. Semi-quantitative score was used and categorised as positive versus negative/low. The relation of molecular markers with prognosis and with clinical variables was evaluated. RESULTS: The analysis included 66 stage I-III patients (48 surgically resected, 18 treated with radical-intent chemoradiotherapy) and 38 metastatic cases. In the overall study population, PD-L1 was expressed on TCs and TIICs in 25% and 40% of cases, respectively. The proportion of PD-L1-positive cases was significantly higher in stage I-III versus metastatic patients (32% versus 13%, p: 0.034 for TCs; 51.5% versus 21% for TIICs, p: 0.002). CD8- and FOXP3-positive TILs were present in 59% and 72% of samples, respectively. The presence of FOXP3-TILs was associated with improved prognosis among non-metastatic patients, with a hazard ratio for survival of 0.32 (95% confidence interval [CI]: 0.16-0.7, p: 0.006) for univariate analysis, and 0.37 (95% CI: 0.17-0.81, p: 0.013) for multivariate analysis. CONCLUSIONS: Immune contexture of SCLC may differ according to stage. The presence of FOXP3-positive TILs is a potential prognostic marker for stage I-III SCLCs and warrants further investigation.

In-hospital mortality, 30-day readmission, and length of hospital stay after surgery for primary colorectal cancer: A national population-based study.

INTRODUCTION: The simultaneous assessment of multiple indicators for quality of care is essential for comparisons of performance between hospitals and health care systems. The aim of this study was to assess the rates of in-hospital mortality and 30-day readmission and length of hospital stay (LOS) in patients who underwent surgical procedures for colorectal cancer between 2005 and 2014 in Italy. METHODS: All patients in the National Italian Hospital Discharge Dataset who underwent a surgical procedure for colorectal cancer during the study period were included. The adjusted odd ratios for risk factors for in-hospital mortality, 30-day readmission, and LOS were calculated using multilevel multivariable logistic regression. RESULTS: Among the 353 941 patients, rates of in-hospital mortality and 30-day readmission were 2.5% and 6%, respectively, and the median LOS was 13 days. High comorbidity, emergent/urgent admission, male gender, creation of a stoma, and an open approach increased the risks of all the outcomes at multivariable analysis. Age, hospital volume, hospital geographic location, and discharge to home/non-home produced different effects depending on the outcome considered. The most frequent causes of readmission were infection (19%) and bowel obstruction (14.6%). CONCLUSIONS: We assessed national averages for mortality, LOS and readmission and related trends over a 10-year time. Laparoscopic surgery was the only one that could be modified by improving surgical education. Higher hospital volume was associated with a LOS reduction, but our findings only partially support a policy of centralization for colorectal cancer procedures. Surgical site infection was identified as the most preventable cause of readmission.

A 3-year interval is too short for re-screening women testing negative for human papillomavirus: a population-based cohort study.

OBJECTIVE: To compare the results from an initial negative human papillomavirus (HPV) test with re-screening after 3 years in women attending two HPV-based screening programmes. DESIGN: Population-based cohort study. SETTING: Two cervical service screening programmes in Italy. POPULATION: Women aged 25-64 years invited to screening from April 2009 to October 2015. METHODS: Eligible women were invited to undergo an HPV test. Those with a negative HPV test went on to the next screening round 3 years later. Cytology triage was performed for HPV+ (HPV by Hybrid Capture 2) samples, with immediate colposcopy (if abnormal) and HPV re-testing 1 year later (if negative). MAIN OUTCOME MEASURES: Participation rate, positivity at HPV and at triage, referral rate to colposcopy, positive predictive value for cervical intraepithelial neoplasia grade 2+ (CIN2+) at colposcopy, and detection rate for CIN2+. RESULTS: We present the results from 48 751 women at the first screening and 22 000 women at re-screening 3 years later. The response rate was slightly higher at the second screening (74.5 versus 72.1% at the first screening; referral rate, RR 1.11; 95% confidence interval, 95% CI, 1.07-1.14). Compared with the first screening, we observed a significant reduction at the second screening in terms of HPV positivity (RR 0.55, 95% CI 0.51-0.60), referral rate to colposcopy (RR 0.47, 95% CI 0.41-0.53), CIN2+ detection rate (RR 0.24, 95% CI 0.13-0.39), and positive predictive value (PPV) for CIN2+ at colposcopy (RR 0.51, 95% CI 0.29-0.87). CONCLUSIONS: The very low frequency of disease and inadequate PPV at colposcopy indicate that a 3-year interval after a negative HPV test is too short. TWEETABLE ABSTRACT: Three years after a negative HPV the frequency of cervical disease is so low that re-screening is inefficient.

Management of Helicobacter pylori infection-the Maastricht V/Florence Consensus Report.

Important progress has been made in the management of Helicobacter pylori infection and in this fifth edition of the Maastricht Consensus Report, key aspects related to the clinical role of H. pylori were re-evaluated in 2015. In the Maastricht V/Florence Consensus Conference, 43 experts from 24 countries examined new data related to H. pylori in five subdivided workshops: (1) Indications/Associations, (2) Diagnosis, (3) Treatment, (4) Prevention/Public Health, (5) H. pylori and the Gastric Microbiota. The results of the individual workshops were presented to a final consensus voting that included all participants. Recommendations are provided on the basis of the best available evidence and relevance to the management of H. pylori infection in the various clinical scenarios.

Genetic risk of subsequent esophageal cancer in lymphoma and breast cancer long-term survival patients: a pilot study.

The occurrence of a second primary esophageal carcinoma (EC) in long-term cancer survivors may represent a late effect of previous radio-chemotherapeutic treatment. To identify the genetic factors that could increase this risk, we analyzed nine variants within ERCC1, XPD, XRCC1 and XRCC3 DNA repair pathway genes, and GSTP1, TP53 and MDM2 genes in 61 patients who received radio-chemotherapy for a prior lymphoma or breast cancer; 29 of them had a second primary EC. This cohort consists of 22 esophageal squamous cell carcinoma (ESCC) and 7 esophageal adenocarcinoma (EADC) patients. A validation cohort of 154 patients with sporadic EC was also included. The XPD Asp312Asn (rs1799793) was found to be associated with the risk of developing second primary ESCC (P=0.015). The resultant variant was also involved in the onset of sporadic ESCC (P=0.0018). To know in advance who among long-term cancer survivors have an increased risk of EC could lead to a more appropriate follow-up strategy.

The stomach in health and disease.

The stomach is traditionally regarded as a hollow muscular sac that initiates the second phase of digestion. Yet this simple view ignores the fact that it is the most sophisticated endocrine organ with unique physiology, biochemistry, immunology and microbiology. All ingested materials, including our nutrition, have to negotiate this organ first, and as such, the stomach is arguably the most important segment within the GI tract. The unique biological function of gastric acid secretion not only initiates the digestive process but also acts as a first line of defence against food-borne microbes. Normal gastric physiology and morphology may be disrupted by Helicobacter pylori infection, the most common chronic bacterial infection in the world and the aetiological agent for most peptic ulcers and gastric cancer. In this state-of-the-art review, the most relevant new aspects of the stomach in health and disease are addressed. Topics include gastric physiology and the role of gastric dysmotility in dyspepsia and gastroparesis; the stomach in appetite control and obesity; there is an update on the immunology of the stomach and the emerging field of the gastric microbiome. H. pylori-induced gastritis and its associated diseases including peptic ulcers and gastric cancer are addressed together with advances in diagnosis. The conclusions provide a future approach to gastric diseases underpinned by the concept that a healthy stomach is the gateway to a healthy and balanced host. This philosophy should reinforce any public health efforts designed to eradicate major gastric diseases, including stomach cancer.

Endoscopic submucosal dissection: European Society of Gastrointestinal Endoscopy (ESGE) Guideline

This Guideline is an official statement of the European Society of Gastrointestinal Endoscopy (ESGE). The Grading of Recommendations Assessment, Development, and Evaluation (GRADE) system 1 2 was adopted to define the strength of recommendations and the quality of evidence. Main Recommendations: 1 ESGE recommends endoscopic en bloc resection for superficial esophageal squamous cell cancers (SCCs), excluding those with obvious submucosal involvement (strong recommendation, moderate quality evidence). Endoscopic mucosal resection (EMR) may be considered in such lesions when they are smaller than 10 mm if en bloc resection can be assured. However, ESGE recommends endoscopic submucosal dissection (ESD) as the first option, mainly to provide an en bloc resection with accurate pathology staging and to avoid missing important histological features (strong recommendation, moderate quality evidence). 2 ESGE recommends endoscopic resection with a curative intent for visible lesions in Barrett's esophagus (strong recommendation, moderate quality evidence). ESD has not been shown to be superior to EMR for excision of mucosal cancer, and for that reason EMR should be preferred. ESD may be considered in selected cases, such as lesions larger than 15 mm, poorly lifting tumors, and lesions at risk for submucosal invasion (strong recommendation, moderate quality evidence). 3 ESGE recommends endoscopic resection for the treatment of gastric superficial neoplastic lesions that possess a very low risk of lymph node metastasis (strong recommendation, high quality evidence). EMR is an acceptable option for lesions smaller than 10 - 15 mm with a very low probability of advanced histology (Paris 0-IIa). However, ESGE recommends ESD as treatment of choice for most gastric superficial neoplastic lesions (strong recommendation, moderate quality evidence). 4 ESGE states that the majority of colonic and rectal superficial lesions can be effectively removed in a curative way by standard polypectomy and/or by EMR (strong recommendation, moderate quality evidence). ESD can be considered for removal of colonic and rectal lesions with high suspicion of limited submucosal invasion that is based on two main criteria of depressed morphology and irregular or nongranular surface pattern, particularly if the lesions are larger than 20 mm; or ESD can be considered for colorectal lesions that otherwise cannot be optimally and radically removed by snare-based techniques (strong recommendation, moderate quality evidence).(AU)

BRAF and RAS mutations as prognostic factors in metastatic colorectal cancer patients undergoing liver resection.

BACKGROUND: Despite major advances in the management of metastatic colorectal cancer (mCRC) with liver-only involvement, relapse rates are high and reliable prognostic markers are needed. METHODS: To assess the prognostic impact of BRAF and RAS mutations in a large series of liver-resected patients, medical records of 3024 mCRC patients were reviewed. Eligible cases undergoing potentially curative liver resection were selected. BRAF and RAS mutational status was tested on primary and/or metastases by means of pyrosequencing and mass spectrometry genotyping assay. Primary endpoint was relapse-free survival (RFS). RESULTS: In the final study population (N=309) BRAF mutant, RAS mutant and all wild-type (wt) patients were 12(4%), 160(52%) and 137(44%), respectively. Median RFS was 5.7, 11.0 and 14.4 months respectively and differed significantly (Log-rank, P=0.043). At multivariate analyses, BRAF mutant had a higher risk of relapse in comparison to all wt (multivariate hazard ratio (HR)=2.31; 95% CI, 1.09-4.87; P=0.029) and to RAS mutant (multivariate HR=2.06; 95% CI, 1.02-4.14; P=0.044). Similar results were obtained in terms of overall survival. Compared with all wt patients, RAS mutant showed a higher risk of death (HR=1.47; 95% CI, 1.05-2.07; P=0.025), but such effect was lost at multivariate analyses. CONCLUSIONS: BRAF mutation is associated with an extremely poor median RFS after liver resection and with higher probability of relapse and death. Knowledge of BRAF mutational status may optimise clinical decision making in mCRC patients potentially candidate to hepatic surgery. RAS status as useful marker in this setting might require further studies.

Plexiform fibromyxoma of the gallbladder.

We report the unusual case of a plexiform fibromyxoma, occasionally assessed in a lithiasic gallbladder. The full thickness assessment of the gallbladder wall revealed an intra-mural, well demarked multi-nodular tumor (1 cm), consisting of a plexiform growth of spindle cells, included within a fibromyxoid stroma with a rich micro-vascular network. The tumor cells featured no nuclear atypia, nor mitotic activity. At the immunohistochemical profiling, the spindle shaped cells unequivocally featured vimentin, SMA, HHF35, collagen IV, and CD34; no cells expressed CD117, PDGFRA, CD10, desmin, GFAP, EMA, and S-100. Faint STAT6 nuclear expression was observed in isolated tumor cells. The molecular profiling did not revealed any CKIT and PDGFRA genes mutations. The uncommon site of the tumor presentation and its aberrant CD34 expression both confer to the reported case a unique place among the myxoid tumors of the gastrointestinal tract.

Value of real-time gastric juice analysis in upper gastrointestinal endoscopy.

AIM: Esophagogastroduodenoscopy (EGDS) cannot identify microscopic lesions. We determined the contribution of real-time gastric juice analysis in detecting lesions non-detectable with the simple endoscopic inspection. METHODS: Endoscopy, histology and gastric juice analysis were performed in 216 patients. We assessed six diagnostic strategies: EGDS (strategy-1), EGDS with antral biopsies (hematoxylin-eosin staining) in hypochlorhydrics (strategy-2) or all patients (strategy-3), EGDS with antral and fundic biopsies (hematoxylin-eosin staining) in hypochlorhydrics (strategy-4) or all patients (strategy-5), EGDS with antral and fundic biopsies (hematoxylin-eosin + immunohistochemical staining) in hypochlorhydrics (strategy-6). Then, we determined how many of the pathological conditions identified by the complete histological evaluation would have been detected by each strategy. RESULTS: In total, 220 pathological conditions were identified. Hypochlorhydria was correlated (r=0.67; P<0.01) with histological lesions (85% lesions were detected in hypochlorhydrics) and high ammonium levels, with H.pylori infection (r=0.69; P<0.01). Strategy-1 identified only 5% conditions, while strategies 3 and 5 detected 68.6% and 83.2% conditions, respectively. Strategies 2, 4 and 6 (based on gastric juice analysis) yielded detection rates (61.4%, 75.5%, 90.9%) similar to or better than those of strategies 3 and 5. CONCLUSION: Real-time gastric juice analysis provided information about the presence of gastric lesions in an otherwise "normal" stomach at EGDS. It improved the diagnostic yield and optimized resource utilization without any additional effort by the endoscopist.

In vivo molecular imaging of HER2 expression in a rat model of Barrett's esophagus adenocarcinoma.

Human epidermal growth factor receptor 2 (HER2) is involved in the malignant progression of several human cancers, including esophageal adenocarcinoma (EAC). The purpose of this study was to evaluate HER2 overexpression and to explore the feasibility of confocal laser endomicroscopy for in vivo molecular imaging of HER2 status in an animal model of Barrett's-related EAC. Rats underwent esophagojejunostomy with gastric preservation. At 30 weeks post-surgery, the esophagus of 46 rats was studied; endoscopic and histological findings were correlated with HER2 immunofluorescence on excised biopsies and gross specimens. At this age, 23/46 rats developed Barrett's esophagus (BE), and 6/46 had cancer (four EAC and two squamous cell carcinomas). A significant overexpression of HER2 was observed in esophageal adenocarcinoma compared with normal squamous esophagus (9.4-fold) and BE (6.0-fold). AKT and its phosphorylated form were also overexpressed in cancer areas. Molecular imaging was performed at 80 weeks post-surgery in four rats after tail injection of fluorescent-labeled anti-HER2 antibody. At this age, 3/4 rats developed advance adenocarcinoma and showed in vivo overexpression of HER2 by molecular confocal laser endomicroscopy with heterogeneous distribution within cancer; no HER2 signal was observed in normal or Barrett's tissues. Therefore, HER2 overexpression is a typical feature of the surgical induced model of EAC that can be easily quantified in vivo using an innovative mini-invasive approach including confocal endomicroscopy; this approach may avoid limits of histological evaluation of HER2 status on 'blinded' biopsies.

Prognostic significance of AMPK activation in advanced stage colorectal cancer treated with chemotherapy plus bevacizumab.

BACKGROUND: AMP-activated protein kinase (AMPK) has a central role in cellular energy sensing and is activated in preclinical tumour models following anti-vascular endothelial growth factor (VEGF) therapy. The possible predictive or prognostic role of AMPK status in cancer patients treated with anti-VEGF drugs has not been investigated so far. METHODS: Expression of components of the AMPK pathway including phosphorylated AMPK (pAMPK), phosphorylated acetyl-Coa carboxylase (pACC) and liver kinase B1 (LKB1) was investigated by immunohistochemistry in 48 colorectal cancers treated with FOLFIRI plus bevacizumab. Correlation between pAMPK and pACC and associations between the AMPK pathway scores and clinico-pathological characteristics were assessed. Overall survival (OS) was estimated through Kaplan-Meier method, whereas hazard ratios were computed to identify prognostic factors. RESULTS: Fourteen patients (29.2%) were included in the pAMPK-negative group (score ≤5), whereas 34 patients (70.8%) were included in the pAMPK-positive group (score >5). The Spearman's coefficient for the correlation between pAMPK and pACC scores in primary tumour samples was 0.514 (P=0.0002). Low pAMPK levels were associated with worse OS (P-value 0.0002) but not with PFS, whereas low pACC levels were associated both with worse OS and PFS (P-value 0.0007 and 0.01, respectively). CONCLUSIONS: Our findings suggest that high tissue AMPK activation is a prognostic biomarker in this cohort of metastatic colorectal cancer patients.

Use of grafts from anti-HBc-positive donors in liver transplantation: a 5-year, single-center experience.

INTRODUCTION: Liver transplantation (OLT) is the treatment of choice for advanced hepatic disease. The growing gap between waiting list patients and the number of donations has led to acceptance of less than optimal donors. The aim of this study was to evaluate the 5-year experience with anti hepatitis B core antigen (HBc)-positive liver donors. PATIENTS AND METHODS: All recipients of anti-HBc-positive grafts from January 2005 to December 2010 were evaluated annually after OLT for liver disease etiology, Model for End-Stage Liver Disease (MELD) score, and the presence of hepatocellular carcinoma (HCC) liver biopsy histology and serology for hepatitis B virus (HBsAg, anti-HBs, HBV-DNA), hepatitis C virus, and hepatitis D virus as well as antiviral prophylaxis to prevent de novo HBV. RESULTS: Among the 249 OLT performed from January 2005 to December 2010, (9.3%) cases used grafts from anti-HBc-positive donors. Etiologics of liver disease among the recipients were HBV (n = 13; 32.5%), HCV (n = 13; 32.5%) or other causes (n = 14; 35%). In 20 of the 40 patients (50%), HCC was found in the explanted organ. Of 40 recipients of anti-HBc-positive grafts 11 died, and 7 (17.5%) required retransplantation. Various regimens were employed as post-transplantation antiviral prophylaxis: (l) Immune globulin (25.8%); (2) Oral antiviral drugs (9.7%); and (3) combined prophylaxis (51.6%) or no treatment (12.9%). No difference was observed in patient or graft survival in relation to the etiology of liver disease, the MELD score, or the presence of HCC at the time of OLT, except graft survival was significantly reduced among recipient who underwent transplantation for non-HBV or non-HCV liver diseases compared with those engrafted due to viral hepatitis (P = .0062). No difference was observed in histologic features (grading and staging) compared with the antiviral prophylactic therapy; the 2 patients (5%) who developed de novo HBV had not received prophylaxis after OLT. CONCLUSIONS: Matching anti-HBc-positive grafts to recipients without HBV infection before OLT, may be especially safe.