In this paper we present a method for the statistical assessment of cancer predictors which make use of gene expression profiles. The methodology is applied to a new data set of microarray gene expression data collected in Casa Sollievo della Sofferenza Hospital, Foggia--Italy. The data set is made up of normal (22) and tumor (25) specimens extracted from 25 patients affected by colon cancer. We propose to give answers to some questions which are relevant for the automatic diagnosis of cancer such as: Is the size of the available data set sufficient to build accurate classifiers? What is the statistical significance of the associated error rates? In what ways can accuracy be considered dependant on the adopted classification scheme? How many genes are correlated with the pathology and how many are sufficient for an accurate colon cancer classification? The method we propose answers these questions whilst avoiding the potential pitfalls hidden in the analysis and interpretation of microarray data. We estimate the generalization error, evaluated through the Leave-K-Out Cross Validation error, for three different classification schemes by varying the number of training examples and the number of the genes used. The statistical significance of the error rate is measured by using a permutation test. We provide a statistical analysis in terms of the frequencies of the genes involved in the classification. Using the whole set of genes, we found that the Weighted Voting Algorithm (WVA) classifier learns the distinction between normal and tumor specimens with 25 training examples, providing e = 21% (p = 0.045) as an error rate. This remains constant even when the number of examples increases. Moreover, Regularized Least Squares (RLS) and Support Vector Machines (SVM) classifiers can learn with only 15 training examples, with an error rate of e = 19% (p = 0.035) and e = 18% (p = 0.037) respectively. Moreover, the error rate decreases as the training set size increases, reaching its best performances with 35 training examples. In this case, RLS and SVM have error rates of e = 14% (p = 0.027) and e = 11% (p = 0.019). Concerning the number of genes, we found about 6000 genes (p < 0.05) correlated with the pathology, resulting from the signal-to-noise statistic. Moreover the performances of RLS and SVM classifiers do not change when 74% of genes is used. They progressively reduce up to e = 16% (p < 0.05) when only 2 genes are employed. The biological relevance of a set of genes determined by our statistical analysis and the major roles they play in colorectal tumorigenesis is discussed. The method proposed provides statistically significant answers to precise questions relevant for the diagnosis and prognosis of cancer. We found that, with as few as 15 examples, it is possible to train statistically significant classifiers for colon cancer diagnosis. As for the definition of the number of genes sufficient for a reliable classification of colon cancer, our results suggest that it depends on the accuracy required.

Aberrant DNA methylation of CpG islands of cancer-related genes is among the earliest and most frequent alterations in cancerogenesis and might be of value for either diagnosing cancer or evaluating recurrent disease. This mechanism usually leads to inactivation of tumour-suppressor genes. We have designed the current study to validate our previous microarray data and to identify novel hypermethylated gene promoters. The validation assay was performed in a different set of 8 patients with colorectal cancer (CRC) by means quantitative reverse-transcriptase polymerase chain reaction analysis. The differential RNA expression profiles of three CRC cell lines before and after 5-aza-2'-deoxycytidine treatment were compared to identify the hypermethylated genes. The DNA methylation status of these genes was evaluated by means of bisulphite genomic sequencing and methylation-specific polymerase chain reaction (MSP) in the 3 cell lines and in tumour tissues from 30 patients with CRC. Data from our previous genome search have received confirmation in the new set of 8 patients with CRC. In this validation set six genes showed a high induction after drug treatment in at least two of three CRC cell lines. Among them, the N-myc downstream-regulated gene 2 (NDRG2) promoter was found methylated in all CRC cell lines. NDRG2 hypermethylation was also detected in 8 out of 30 (27%) primary CRC tissues and was significantly associated with advanced AJCC stage IV. Normal colon tissues were not methylated. The findings highlight the usefulness of combining gene expression patterns and epigenetic data to identify tumour biomarkers, and suggest that NDRG2 silencing might bear influence on tumour invasiveness, being associated with a more advanced stage.

Introduction: Hepatitis D virus (HDV) infection remains a significant global health challenge due to its severity and high risk of progression to cirrhosis and hepatocellular carcinoma (HCC). Bulevirtide, a novel HDV entry inhibitor, has shown promise in managing chronic hepatitis D by blocking viral entry into hepatocytes. This study evaluated the efficacy and safety of bulevirtide in reducing HDV RNA levels and improving liver function in a real-life cohort of Italian patients with HDV infection. Methods: This multicenter prospective trial enrolled 108 consecutive patients with chronic HDV infection, from June 2023 to June 2024, who received 2 mg/day of bulevirtide in combination with a nucleoside/nucleotide analogue for hepatitis B virus (HBV) infection. Patients with any stage of liver fibrosis or compensated cirrhosis were included. Data collected included demographic and clinical characteristics, liver function tests, HDV RNA levels, and adverse events at baseline and 6 months. Results: The virological response was achieved in 54.6% of patients (n = 59), with 36 demonstrating undetectable HDV RNA levels. Among responders, ALT levels decreased significantly from 67.0 U/mL [IQR 44.0–116.3] to 31.5 U/mL [IQR 24.0–36.5, p = 0.001], and AST levels from 66.0 U/mL [IQR 46.5–91.0] to 32.5 U/mL [IQR 28.0–38.0, p = 0.021]. Median HDV RNA dropped from 29,800 IU/mL [IQR 3100–375,000] to 0 IU/mL [IQR 0–291, p < 0.001]. No significant predictors of response emerged. Mild adverse events, including pruritus (5.6%) and injection-site reactions (1.9%) and flu-like syndrome (0.9) were reported, with no treatment discontinuation. Conclusions: Bulevirtide effectively reduces HDV RNA levels and improves liver function with a favorable safety profile, offering a promising therapeutic option for chronic hepatitis D. Further large-scale studies are needed to confirm these findings and explore long-term outcomes.

Spontaneous HBsAg seroclearance has been mainly studied in populations from Asia, Australia, the Pacific Islands, and Polynesia. For the first time, we evaluated the spontaneous HBsAg seroclearance and its possible associated factors and the risk of disease progression in HBeAg-negative patients with inactive infection all coming from the same region in South Italy. In this multicenter retrospective study, 146 patients were selected after 18 months of observation and followed for a median of 82 months (IQR 60–107). For our analyses, they were divided into three groups based on their HBsAg levels: 1000 IU/mL. Crude and adjusted hazard ratios (HRs) for HBsAg seroclearance were determined. During the follow-up period, three patients (2.0%) showed a disease progression with an increased liver stiffness, whereas 17 (11.6%) cleared the HBsAg. Patients with HBsAg levels <100 IU/mL had the highest probability of HBsAg seroclearance compared to the other two groups (p = 0.009). In the multivariate analysis, the HBsAg level <100 IU/mL was the only parameter independently associated with HBsAg seroclearance (adjusted HR = 3.53; CI 1.29–9.69; p = 0.01). In patients with chronic HBV inactive infection, HBsAg levels <100 IU/mL predicted the highest probability of HBsAg seroclearance.