Awad, A. B., M. Mabrouk, T. Awad, N. Zayed, S. Mousa, and M. Saeed, "Performance evaluation of decision tree classifiers for the prediction of response to treatment of hepatitis C patients", Proceeding PervasiveHealth '14 Proceedings of the 8th International Conference on Pervasive Computing Technologies for Healthcare, pp. 186-190, 2014.
Shousha, H. I., A. H. Awad, D. A. H. Omran, M. M. Elnegouly, and M. Mabrouk, "Data Mining and Machine Learning Algorithms Using IL28B Genotype and Biochemical Markers Best Predicted Advanced Liver Fibrosis in Chronic Hepatitis C.", Japanese journal of infectious diseases, vol. 71, issue 1, pp. 51-57, 2018. Abstract

IL28B single nucleotide polymorphism (rs12979860) is an etiology-independent predictor of hepatitis C virus (HCV)-related hepatic fibrosis. Data mining is a method of predictive analysis which can explore tremendous volumes of information from health records to discover hidden patterns and relationships. The current study aims to evaluate and compare the prediction accuracy of scoring system like aspartate aminotransferase-to-platelet ratio index (APRI) and fibrosis-4 (FIB-4) index versus data mining for the prediction of HCV-related advanced fibrosis. This retrospective study included 427 patients with chronic hepatitis C. We used data mining analysis to construct a decision tree by reduced error (REP) technique, followed by Auto-WEKA tool to select the best classifier out of 39 algorithms to predict advanced fibrosis. APRI and FIB-4 had sensitivity-specificity parameters of 0.523-0.831 and 0.415-0.917, respectively. REPTree algorithm was able to predict advanced fibrosis with sensitivity of 0.749, specificity of 0.729, and receiver operating characteristic (ROC) area of 0.796. Out of the 16 attributes, IL28B genotype was selected by the REPTree as the best predictor for advanced fibrosis. Using Auto-WEKA, the multilayer perceptron (MLP) neural model was selected as the best predictive algorithm with sensitivity of 0.825, specificity of 0.811, and ROC area of 0.880. Thus, MLP is better than APRI, FIB-4, and REPTree for predicting advanced fibrosis for patients with chronic hepatitis C.

Omran, D. A. H., A. H. Awad, M. A. E. R. Mabrouk, A. F. Soliman, and A. O. Abdelaziz, "Application of data mining techniques to explore predictors of HCC in Egyptian patients with HCV-related chronic liver disease.", Asian Pacific journal of cancer prevention : APJCP, vol. 16, issue 1, pp. 381-5, 2015. Abstract

BACKGROUND: Hepatocellular carcinoma (HCC) is the second most common malignancy in Egypt. Data mining is a method of predictive analysis which can explore tremendous volumes of information to discover hidden patterns and relationships. Our aim here was to develop a non-invasive algorithm for prediction of HCC. Such an algorithm should be economical, reliable, easy to apply and acceptable by domain experts.

METHODS: This cross-sectional study enrolled 315 patients with hepatitis C virus (HCV) related chronic liver disease (CLD); 135 HCC, 116 cirrhotic patients without HCC and 64 patients with chronic hepatitis C. Using data mining analysis, we constructed a decision tree learning algorithm to predict HCC.

RESULTS: The decision tree algorithm was able to predict HCC with recall (sensitivity) of 83.5% and precession (specificity) of 83.3% using only routine data. The correctly classified instances were 259 (82.2%), and the incorrectly classified instances were 56 (17.8%). Out of 29 attributes, serum alpha fetoprotein (AFP), with an optimal cutoff value of ≥50.3 ng/ml was selected as the best predictor of HCC. To a lesser extent, male sex, presence of cirrhosis, AST>64U/L, and ascites were variables associated with HCC.

CONCLUSION: Data mining analysis allows discovery of hidden patterns and enables the development of models to predict HCC, utilizing routine data as an alternative to CT and liver biopsy. This study has highlighted a new cutoff for AFP (≥50.3 ng/ml). Presence of a score of >2 risk variables (out of 5) can successfully predict HCC with a sensitivity of 96% and specificity of 82%.

Raziky, M. E., W. F. Fathalah, Z. Zakaria, H. G. E. deen, A. Abul-Fotouh, A. Salama, A. B. Awad, G. Esmat, and M. Mabrouk, "Predictors of Virological Response in 3,235 Chronic HCV Egyptian Patients Treated with Peginterferon Alpha-2a Compared with Peginterferon Alpha-2b Using Statistical Methods and Data Mining Techniques.", Journal of interferon & cytokine research : the official journal of the International Society for Interferon and Cytokine Research, vol. 36, issue 5, pp. 338-46, 2016. Abstract

Despite the appearance of new oral antiviral drugs, pegylated interferon (PEG-IFN)/RBV may remain the standard of care therapy for some time, and several viral and host factors are reported to be correlated with therapeutic effects. This study aimed to reveal the independent variables associated with failure of sustained virological response (SVR) to PEG-IFN alpha-2a versus PEG-IFN alpha-2b in treatment of naive chronic hepatitis C virus (HCV) Egyptian patients using both statistical methods and data mining techniques. This retrospective cohort study included 3,235 chronic hepatitis C patients enrolled in a large Egyptian medical center: 1,728 patients had been treated with PEG-IFN alpha-2a plus ribavirin (RBV) and 1,507 patients with PEG-IFN alpha-2b plus RBV between 2007 and 2011. Both multivariate analysis and Reduced Error Pruning Tree (REPTree)-based model were used to reveal the independent variables associated with treatment response. In both treatment types, alpha-fetoprotein (AFP) >10 ng/mL and HCV viremia >600 × 10(3) IU/mL were the independent baseline variables associated with failure of SVR, while male gender, decreased hemoglobin, and thyroid-stimulating hormone were the independent variables associated with good response (P < 0.05). Using REPTree-based model showed that low AFP was the factor of initial split (best predictor) of response for either PEG-IFN alpha-2a or PEG-IFN alpha-2b (cutoff value 8.53, 4.89 ng/mL, AUROC = 0.68 and 0.61, P = 0.05). Serum AFP >10 ng/mL and viral load >600 × 10(3) IU/mL are variables associated with failure of response in both treatment types. REPTree-based model could be used to assess predictors of response.

Saad, Y., A. Awad, W. Al Akel, W. Doss, T. Awad, and M. Mabrouk, "Data mining of routine laboratory tests can predict liver disease progression in Egyptian diabetic patients with hepatitis C virus (G4) infection: a cohort study of 71 806 patients.", European journal of gastroenterology & hepatology, vol. 30, issue 2, pp. 201-206, 2018. Abstract

OBJECTIVES: Hepatitis C virus (HCV) and diabetes mellitus (DM) are prevalent diseases worldwide, associated with significant morbidity, mortality, and mutual association. The aims of this study were as follows: (i) find the prevalence of DM among 71 806 Egyptian patients with chronic HCV infection and its effect on liver disease progression and (ii) using data mining of routine tests to predict hepatic fibrosis in diabetic patients with HCV infection.

PATIENTS AND METHODS: A retrospective multicentered study included laboratory and histopathological data of 71 806 patients with HCV infection collected by Egyptian National Committee for control of viral hepatitis. Using data mining analysis, we constructed decision tree algorithm to assess predictors of fibrosis progression in diabetic patients with HCV.

RESULTS: Overall, 12 018 (16.8%) patients were diagnosed as having diabetes [6428: fasting blood glucose ≥126 mg/dl (9%) and 5590: fasting blood glucose ≥110-126 mg/dl (7.8%)]. DM was significantly associated with advanced age, high BMI and α-fetoprotein (AFP), and low platelets and serum albumin (P≤0.001). Advanced liver fibrosis (F3-F4) was significantly correlated with DM (P≤0.001) irrespective of age. Of 16 attributes, decision tree model for fibrosis showed AFP was most decisive with cutoff of 5.25 ng/ml as starting point of fibrosis. AFP level greater than cutoff in patients was the first important splitting attribute; age and platelet count were second important splitting attributes.

CONCLUSION: (i) Chronic HCV is significantly associated with DM (16.8%). (ii) Advanced age, high BMI and AFP, low platelets count and albumin show significant association with DM in HCV. (iii) AFP cutoff of 5.25 is a starting point of fibrosis development and integrated into mathematical model to predict development of liver fibrosis in diabetics with HCV (G4) infection.

Raziky, M. E., W. F. Fathalah, W. A. El-akel, A. Salama, G. Esmat, M. Mabrouk, R. M. Salama, and H. M. Khatab, "The Effect of Peginterferon Alpha-2a vs. Peginterferon Alpha-2b in Treatment of Naive Chronic HCV Genotype-4 Patients: A Single Centre Egyptian Study.", Hepatitis monthly, vol. 13, issue 5, pp. e10069, 2013 May. Abstract

BACKGROUND: Egypt has one of the highest (16-8%) prevalence rates of HCV infection in the world. Approximately 90% of Egyptian HCV isolates belong to a single subtype (4a), which responds less successfully to interferon therapy than other subtypes. Studies comparing the efficacy and safety of PEGIFN alfa-2a and PEGIFN alfa-2b in treatment-naive HCV-infected patients have shown conflicting results.

OBJECTIVES: Assessing the effects of Peginterferon alpha-2a versus Peginterferon alpha-2b on the sustained virological response in naive chronic HCV genotype-4 Egyptian patients.

PATIENTS AND METHODS: This retrospective study cohort consists of 3718 chronic HCV patients admitted to a large, Egyptian medical center. 1985 patients had been treated with PEG-IFN alfa-2a plus RBV and 1733 patients with PEG-IFN alfa-2b plus RBV between years 2007-2011. Efficacy outcomes were sustained virologic response (SVR) and treatment discontinuation rates due to serious adverse effects.

RESULTS: The ETR & SVR in patients treated with PEGIFN alfa-2a was 64.1% and 59.6% as compared to treatment with PEGIFN alfa-2b where these parameters were 58.2% and 53.9% respectively (P < 0.05). Treatment discontinuation rates, were similar in the two types of PEGIFN [0.66 (0.37-1.16); P = 0.15]. Significant dose reduction was evident with peginterferon alfa-2b (35.3%) than peginterferon alpha-2a (27.3 %) (P < 0.01). Patients with lower base line AFP and ALT were most likely to achieve SVR using INF alpha 2-a.

CONCLUSIONS: Peginterferon alpha-2a has a higher efficacy regarding ETR and SVR as compared to Peginterferon alfa-2b in treatment of naive chronic HCV genotype-4 patients.

Khairy, M., R. Fouad, M. Mabrouk, W. El-Akel, A. B. Awad, R. Salama, M. Elnegouly, and O. Shaker, "The impact of interleukin 28b gene polymorphism on the virological response to combined pegylated interferon and ribavirin therapy in chronic HCV genotype 4 infected egyptian patients using data mining analysis.", Hepatitis monthly, vol. 13, issue 7, pp. e10509, 2013. Abstract

BACKGROUND: Chronic HCV represents one of the common causes of chronic liver disease worldwide with Egypt having the highest prevalence, namely genotype 4. Interleukin IL-28B gene polymorphism has been shown to relate to HCV treatment response, mainly in genotype1.

OBJECTIVES: We aim to evaluate the predictive power of the rs12979860 IL28B SNP and its protein for treatment response in genotype 4 Egyptian patients by regression analysis and decision tree analysis.

PATIENTS AND METHODS: The study included 263 chronic HCV Egyptian patients receiving peg-interferon and ribavirin therapy. Patients were classified into 3 groups; non responders (83patients), relapsers (76patients) and sustained virological responders (104 patients). Serum IL 28 B was performed, DNA was extracted and analyzed by direct sequencing of the SNP rs 12979860 of IL28B gene.

RESULTS: CT, CC and TT represented 56 %, 25 % and 19% of the patients, respectively. Absence of C allele (TT genotype) was significantly correlated with the early failure of response while CC was associated with sustained virological response. The decision tree showed that baseline alpha fetoprotein (AFP ≤ 2.68 ng/ml) was the variable of initial split (the strongest predictor of response) confirmed by regression analysis. Patients with TT genotype had the highest probability of failure of response.

CONCLUSIONS: Absence of the C allele was significantly associated with failure of response. The presence of C allele was associated with a favorable outcome. AFP is a strong baseline predictor of HCV treatment response. A decision tree model is useful for predicting the probability of response to therapy.

Hauser, G., T. Awad, J. Brok, K. Thorlund, D. Stimac, M. Mabrouk, C. Gluud, and L. L. Gluud, "Peginterferon plus ribavirin versus interferon plus ribavirin for chronic hepatitis C.", The Cochrane database of systematic reviews, vol. 2, pp. CD005441, 2014 Feb 28. Abstract

BACKGROUND: Pegylated interferon (peginterferon) plus ribavirin is the recommended treatment for patients with chronic hepatitis C, but systematic assessment of the effect of this treatment compared with interferon plus ribavirin is needed.

OBJECTIVES: To systematically evaluate the benefits and harms of peginterferon plus ribavirin versus interferon plus ribavirin for patients with chronic hepatitis C.

SEARCH METHODS: We searched the Cochrane Hepato-Biliary Group Controlled Trials Register, the Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, EMBASE, Science Citation Index-Expanded, and LILACS. We also searched conference abstracts, journals, and grey literature. The last searches were conducted in September 2013.

SELECTION CRITERIA: We included randomised clinical trials comparing peginterferon plus ribavirin versus interferon plus ribavirin with or without co-intervention(s) (e.g., other antiviral drugs) for chronic hepatitis C. Quasi-randomised and observational studies retrieved through the searches for randomised clinical trials were also considered for reports of harms. Our primary outcomes were liver-related morbidity, all-cause mortality, serious adverse events, adverse events leading to treatment discontinuation, other adverse events, and quality of life. Our secondary outcome was sustained virological response in serum, that is, undetectable hepatitis C virus RNA in serum by sensitive tests six months after the end of treatment.

DATA COLLECTION AND ANALYSIS: Two review authors independently used a standardised data collection form. We meta-analysed data with both fixed-effect and random-effects models. For each outcome, we calculated the odds ratio (OR) (for liver-related morbidity or all-cause mortality) or the risk ratio (RR) along with 95% confidence interval (CI) based on intention-to-treat analysis. We used domains of the trials to assess the risk of systematic errors (bias) and trial sequential analyses to assess the risk of random errors (play of chance).For each outcome, we calculated the RR with 95% CI based on intention-to-treat analysis. Effects of interventions on outcomes were assessed according to GRADE.

MAIN RESULTS: We included 27 randomised trials with 5938 participants. All trials had high risk of bias. We considered that the risk of bias did not impact on the quality of evidence for liver-related mortality and adverse event outcomes, but it did for virological response. All trials compared peginterferon alpha-2a or peginterferon alpha-2b plus ribavirin versus interferon plus ribavirin for participants with chronic hepatitis C. Three trials administered co-interventions (amantadine hydrochloride 200 mg daily to both intervention groups), and 24 trials were conducted without co-interventions. The effect observed between the two intervention groups regarding liver-related morbidity plus all-cause mortality (5/907 (0.55%) versus 4/882 (0.45%) was imprecise: OR 1.14 ( 95% CI 0.38 to 3.42; five trials; low quality of evidence), as was the risk of adverse events leading to treatment discontinuation (332/2692 (12.3%) versus 409/2176 (18.8%); RR 0.86, 95% CI 0.68 to 1.09; 15 trials; low quality of evidence) or regarding adverse events leading to treatment discontinuation (332/2692 (12.3%) versus 409/2176 (18.8%); RR 0.86, 95% CI 0.66 to 1.12; 17 trials; low quality of evidence). However, peginterferon plus ribavirin versus interferon plus ribavirin significantly increased the risk of neutropenia (332/2202 (15.1%) versus 117/1653 (7.1%); RR 2.15, 95% CI 1.76 to 2.61; 13 trials), thrombocytopenia (65/1113 (5.8%) versus 23/1082 (2.1%); RR 2.63, 95% CI 1.68 to 4.11; 10 trials), arthralgia (517/1740 (29.7%) versus 282/1194 (23.6%); RR 1.19, 95% CI 1.05 to 1.35; four trials), injection site reaction (627/1168 (53.7%) versus 186/649 (28.7%); RR 1.71, 95% CI 1.50 to 1.93; four trials), and nausea (606/1784 (34.0%) versus 354/1239 (28.6%); RR 1.13, 95% CI 1.01 to 1.26; four trials). The most frequent adverse event was fatigue, which occurred in 57% of participants (2024/3608). No significant difference was noted between peginterferon plus ribavirin versus interferon plus ribavirin in terms of fatigue (1177/2062 (57.1%) versus 847/1546 (54.8%); RR 1.01, 95% CI 0.96 to 1.07; 12 trials). No significant differences were reported between the two treatment groups regarding anaemia, headache, rigours, myalgia, pyrexia, weight loss, asthenia, depression, insomnia, irritability, alopecia, pruritus, skin rash, thyroid malfunction, decreased appetite, or diarrhoea. We were unable to identify any data on quality of life. Peginterferon plus ribavirin versus interferon plus ribavirin seemed to significantly increase the number of participants achieving sustained virological response (1673/3300 participants (50.7%) versus 1081/2804 patients (36.7%); RR 1.39, 95% CI 1.25 to 1.56; I(2) = 64%; 27 trials; very low quality of evidence). However, the risk of bias in the 13/27 (48.1%) trials reporting on this outcome was high and was considered only 'lower' in the remainder. Because the conventional meta-analysis did not reach its required information size (n = 14,486 participants), we used trial sequential analysis to control for risks of random errors. Again, in this analysis, the estimated effect was statistically significant in favour of peginterferon. Subgroup analyses according to risk of bias, viral genotype, baseline viral load, past treatment history, and type of intervention yielded similarly significant results favouring peginterferon over interferon on the outcome of sustained virological response.

AUTHORS' CONCLUSIONS: Peginterferon plus ribavirin versus interferon plus ribavirin seems to significantly increase the proportion of patients with sustained virological response, as well as the risk of certain adverse events. However, we have insufficient evidence to recommend or reject peginterferon plus ribavirin for liver-related morbidity plus all-cause mortality compared with interferon plus ribavirin. The clinical consequences of achieved sustained virological response are unknown, as sustained virological response is still an unvalidated surrogate outcome. We found no evidence of the potential benefits on quality of life in patients with achieved sustained virological response. Further high-quality research is likely to have an important impact on our confidence in the estimate of patient-relevant outcomes and is likely to change our estimates.There is very low quality evidence that peginterferon plus ribavirin increases the proportion of patients with sustained virological response in comparison with interferon plus ribavirin. There is evidence that it also increases the risk of certain adverse events.

Hauser, G., T. Awad, K. Thorlund, D. Stimac, M. Mabrouk, and C. Gluud, "Peginterferon alpha-2a versus peginterferon alpha-2b for chronic hepatitis C.", The Cochrane database of systematic reviews, vol. 2, pp. CD005642, 2014 Feb 28. Abstract

BACKGROUND: A combination of weekly pegylated interferon (peginterferon) alpha and daily ribavirin still represents standard treatment of chronic hepatitis C infection in the majority of patients. However, it is not established which of the two licensed peginterferon products, peginterferon alpha-2a or peginterferon alpha-2b, is the most effective and has a better safety profile.

OBJECTIVES: To systematically evaluate the benefits and harms of peginterferon alpha-2a versus peginterferon alpha-2b in head-to-head randomised clinical trials in patients with chronic hepatitis C.

SEARCH METHODS: We searched the Cochrane Hepato-Biliary Group Controlled Trials Register, the Cochrane Central Register of Controlled Trials (CENTRAL) in The Cochrane Library, MEDLINE, EMBASE, Science Citation Index Expanded, and LILACS until October 2013. We also searched conference abstracts, journals, and grey literature.

SELECTION CRITERIA: We included randomised clinical trials comparing peginterferon alpha-2a versus peginterferon alpha-2b given with or without co-intervention(s) (for example, ribavirin) for chronic hepatitis C. Quasi-randomised studies and observational studies as identified by the searches were also considered for assessment of harms. Our primary outcomes were all-cause mortality, liver-related morbidity, serious adverse events, adverse events leading to treatment discontinuation, other adverse events, and quality of life. The secondary outcome was sustained virological response in the blood serum.

DATA COLLECTION AND ANALYSIS: Two authors independently used a standardised data collection form. We meta-analysed data with both the fixed-effect and the random-effects models. For each outcome we calculated the relative risk (RR) with 95% confidence interval (CI) based on intention-to-treat analysis. We used domains of the trials to assess the risk of systematic errors (bias) and trial sequential analyses to assess the risks of random errors (play of chance). Intervention effects on the outcomes were assessed according to GRADE.

MAIN RESULTS: We included 17 randomised clinical trials which compared peginterferon alpha-2a plus ribavirin versus peginterferon alpha-2b plus ribavirin in 5847 patients. All trials had a high risk of bias. Very few trials reported data on very few patients for the patient-relevant outcomes all-cause mortality, liver-related morbidity, serious adverse events, and quality of life. Accordingly, we were unable to conduct meta-analyses on all-cause mortality, liver-related morbidity, and quality of life. Twelve trials reported on adverse events leading to discontinuation of treatment without clear evidence of a difference between the two peginterferons (197/2171 (9.1%) versus 311/3169 (9.9%); RR 0.84, 95% CI 0.57 to 1.22; I(2) = 44%; low quality evidence). A trial sequential analysis showed that we could exclude a relative risk reduction of 20% or more on this outcome. Peginterferon alpha-2a significantly increased the number of patients who achieved a sustained virological response in the blood serum compared with peginterferon alpha-2b (1069/2099 (51%) versus 1327/3075 (43%); RR 1.12, 95% CI 1.06 to 1.18; I(2)= 0%, 12 trials; moderate quality evidence). Trial sequential analyses supported this result. Subgroup analyses based on risk of bias, viral genotype, and treatment history yielded similar results. Trial sequential analyses supported the results in patients with genotypes 1 and 4, but not in patients with genotypes 2 and 3.

AUTHORS' CONCLUSIONS: There is lack of evidence on patient-important outcomes and paucity of evidence on adverse events. Moderate quality evidence suggests that peginterferon alpha-2a is associated with a higher sustained virological response in serum than with peginterferon alpha-2b. This finding may be affected by the high risk of bias of the included studies . The clinical consequences of peginterferon alpha-2a versus peginterferon alpha-2b are unknown, and we cannot translate an effect on sustained virological response into comparable clinical effects because sustained virological response is still an unvalidated surrogate outcome for patient-important outcomes. The lack of evidence on patient-important outcomes and the paucity of evidence on adverse events means that we are unable to draw any conclusions about the effects of one peginterferon over the other.

Awad, T., K. Thorlund, G. Hauser, D. Stimac, M. Mabrouk, and C. Gluud, "Peginterferon alpha-2a is associated with higher sustained virological response than peginterferon alfa-2b in chronic hepatitis C: systematic review of randomized trials.", Hepatology (Baltimore, Md.), vol. 51, issue 4, pp. 1176-84, 2010 Apr. Abstract

A combination of weekly pegylated interferon (peginterferon) alpha and daily ribavirin represents the standard of care for the treatment of chronic hepatitis C according to current guidelines. It is not established which of the two licensed products (peginterferon alpha-2a or peginterferon alfa-2b) is most effective. We performed a systematic review of head-to-head randomized trials to assess the benefits and harms of the two treatments. We searched the Cochrane Central Register of Controlled Trials, MEDLINE, EMBASE, and LILACS through July 2009. Using standardized forms, two reviewers independently extracted data from each eligible trial report. We statistically combined data using a random effects meta-analysis according to the intention-to-treat principle. We identified 12 randomized clinical trials, including 5,008 patients, that compared peginterferon alpha-2a plus ribavirin versus peginterferon alfa-2b plus ribavirin. Overall, peginterferon alpha-2a significantly increased the number of patients who achieved a sustained virological response (SVR) versus peginterferon alfa-2b (47% versus 41%; risk ratio 1.11, 95% confidence interval 1.04-1.19; P = 0.004 [eight trials]). Subgroup analyses of risk of bias, viral genotype, and treatment history yielded similar results. The meta-analysis of adverse events leading to treatment discontinuation included 11 trials and revealed no significant differences between the two peginterferons. Conclusion: Current evidence suggests that peginterferon alpha-2a is associated with higher SVR than peginterferon alfa-2b. However, the paucity of evidence on adverse events curbs the decision to definitively recommend one peginterferon over the other, because any potential benefit must outweigh the risk of harm.