Machine learning improves the prediction of significant fibrosis in Asian patients with metabolic dysfunction-associated steatotic liver disease - The Gut and Obesity in Asia (GO-ASIA) Study

BACKGROUND

The precise estimation of cases with significant fibrosis (SF) is an unmet goal in non-alcoholic fatty liver disease (NAFLD/MASLD).

AIMS

We evaluated the performance of machine learning (ML) and non-patented scores for ruling out SF among NAFLD/MASLD patients.

METHODS

Twenty-one ML models were trained (N = 1153), tested (N = 283), and validated (N = 220) on clinical and biochemical parameters of histologically-proven NAFLD/MASLD patients (N = 1656) collected across 14 centres in 8 Asian countries. Their performance for detecting histological-SF (≥F2fibrosis) were evaluated with APRI, FIB4, NFS, BARD, and SAFE (NPV/F1-score as model-selection criteria).

RESULTS

Patients aged 47 years (median), 54.6% males, 73.7% with metabolic syndrome, and 32.9% with histological-SF were included in the study. Patients with SFvs.no-SF had higher age, aminotransferases, fasting plasma glucose, metabolic syndrome, uncontrolled diabetes, and NAFLD activity score (p < 0.001, each). ML models showed 7%-12% better discrimination than FIB-4 to detect SF. Optimised random forest (RF) yielded best NPV/F1 in overall set (0.947/0.754), test set (0.798/0.588) and validation set (0.852/0.559), as compared to FIB4 in overall set (0.744/0.499), test set (0.722/0.456), and validation set (0.806/0.507). Compared to FIB-4, RF could pick 10 times more patients with SF, reduce unnecessary referrals by 28%, and prevent missed referrals by 78%. Age, AST, ALT fasting plasma glucose, and platelet count were top features determining the SF. Sequential use of SAFE < 140 and FIB4 < 1.2 (when SAFE > 140) was next best in ruling out SF (NPV of 0.757, 0.724 and 0.827 in overall, test and validation set).

CONCLUSIONS

ML with clinical, anthropometric data and simple blood investigations perform better than FIB-4 for ruling out SF in biopsy-proven Asian NAFLD/MASLD patients.

The Global Bioequivalence Harmonisation Initiative (GBHI): Report of the fifth international EUFEPS/AAPS conference

The series of conferences of the Global Bioequivalence Harmonisation Initiative (GBHI) was started in 2015 by the European Federation for Pharmaceutical Sciences (EUFEPS). All GBHI meetings so far were co-organised together with the American Association of Pharmaceutical Scientists (AAPS). Beginning with the 3rd workshop US-FDA joined as co-sponsor - to support global harmonisation of regulatory recommendations for bioequivalence (BE) assessment. At the 5th GBHI conference, the following BE topics were intensively discussed, and the following main conclusions were drawn: (1) Statistical considerations for BE assessment in specific situations covering scaling approaches for highly variable drug (HVD) products, two-stage adaptive design and opportunities of modelling and simulation to support BE: even though special BE study concepts like adaptive designs are not often used in practise so far, a majority of the workshop participants were in favour of a more frequent application of such approaches. The regulatory conditions relevant in this context need further concretisation and harmonisation between the regions. Moreover, modelling and simulation were considered as a promising and evolving approach, also for BE development programmes. (2) Fed versus fasting conditions in BE trials: Findings that BE between generic products could be confirmed only after fasted administration but failed under fed conditions seem more an exception than the rule. Obviously, BCS class IV compounds are most problematic in this context. Differences in critical excipients such as surfactants or pH-modifiers may be relevant reasons for different sensitivity for interactions in fasted versus fed conditions. Consequently, such deviations in composition of generic preparations should be avoided. Moreover, confirmation of BE may be generally difficult comparing different dosage forms, such like capsules versus tablets, especially in fed state. (3) BE assessment of locally acting drug products applied topically to the skin: Appropriateness and potential benefit of in-vitro tests as alternatives to clinical efficacy studies have been comprehensively discussed. In addition to the already well-established in-vitro release and permeation tests, other techniques were suggested, e.g., Raman spectroscopy or dermal open flow microperfusion. Validation of those methods is challenging and, despite significant progress already achieved during previous years, more research is needed before they may be fully accepted for regulatory purposes. (4) BE evaluation of narrow therapeutic index (NTI) drugs: The discrepancies amongst regulatory agencies in necessity of tighter BE acceptance ranges, the recommendations for inclusion of peak and total drug exposure into BE assessment with more restrictive criteria and the importance of comparison of the product-related within-subject variability for NTI drugs were debated. Arguments in favour and against the different approaches were presented and discussed but need further consideration before harmonisation can be achieved. The highly interactive meeting and extensive exchange between regulators and scientists from industry and academia resulted in useful progress in open BE issues and supported the goal of science-driven harmonisation.

Lean Indian patients with non-alcoholic fatty liver disease (NAFLD) have less metabolic risk factors but similar liver disease severity as non-lean patients with NAFLD

INTRODUCTION

Although most patients with NAFLD are obese or overweight, some are lean with normal BMI. Our aim was to assess differences in clinicopathological profile and liver disease severity among lean and non-lean NAFLD.

METHODS

Data of 1040 NAFLD patients over last 10 years was analysed. BMI < 23 kg/m2 categorised lean patients. Non-invasive assessment of steatosis was done by ultrasound and controlled attenuation parameter (CAP) while fibrosis was assessed with FIB-4 and liver stiffness measurement (LSM). FibroScan-AST (FAST) score was used for non-invasive prediction of NASH with significant fibrosis. Histology was reported using NASH-CRN system.

RESULTS

149 (14.3%) patients were lean while 891 (85.7%) patients were non-lean. Diabetes mellitus [25 (16.7%) vs 152 (17.05%), p > 0.99], elevated triglycerides [81 (54.3%) vs 525 (58.9%), p = 0.33] and low HDL [71(47.6%) vs 479(53.7%), p = 0.18] were observed in a similar proportion. Lean patients were less likely to have central obesity [72 (48.3%) vs 788 (88.4%), p < 0.001], hypertension [16 (10.7%) vs 239(26.8%), p < 0.001] and metabolic syndrome [21 (14.09%) vs 290 (32.5%), p < 0.001]. No difference in steatosis assessment was noted using ultrasound (p = 0.55) or CAP (0.11). FAST [0.38 (0.18-0.66) vs 0.39 (0.27-0.73), p = 0.53], FIB-4 [1.08 (0.65-1.91) vs 1.09 (0.66-1.94), p = 0.94] and LSM [6.1 (4.8-7.9) vs 6.2 (4.7-8.6), p = 0.19) were similar. Liver biopsy was available in 149 patients [lean: 19 (12.7%), non-lean: 130 (87.3%)]. There was no difference in the number of patients with NASH [4 (21.05%) vs 20 (15.3%), p = 0.51], significant fibrosis [2 (10.5%) vs 32 (24.6%), p = 0.25] or advanced fibrosis [1 (5.26%) vs 18 (13.84%), p = 0.47].

CONCLUSION

Although metabolic co-morbidities are less common, there is no difference in liver disease severity among both groups.

Infrared spectroscopic laser scanning confocal microscopy for whole-slide chemical imaging

Chemical imaging, especially mid-infrared spectroscopic microscopy, enables label-free biomedical analyses while achieving expansive molecular sensitivity. However, its slow speed and poor image quality impede widespread adoption. We present a microscope that provides high-throughput recording, low noise, and high spatial resolution where the bottom-up design of its optical train facilitates dual-axis galvo laser scanning of a diffraction-limited focal point over large areas using custom, compound, infinity-corrected refractive objectives. We demonstrate whole-slide, speckle-free imaging in ~3 min per discrete wavelength at 10× magnification (2 μm/pixel) and high-resolution capability with its 20× counterpart (1 μm/pixel), both offering spatial quality at theoretical limits while maintaining high signal-to-noise ratios (>100:1). The data quality enables applications of modern machine learning and capabilities not previously feasible - 3D reconstructions using serial sections, comprehensive assessments of whole model organisms, and histological assessments of disease in time comparable to clinical workflows. Distinct from conventional approaches that focus on morphological investigations or immunostaining techniques, this development makes label-free imaging of minimally processed tissue practical.

Drug Permeability - Best Practices for Biopharmaceutics Classification System (BCS)-Based Biowaivers: A workshop Summary Report

The workshop "Drug Permeability - Best Practices for Biopharmaceutics Classification System (BCS) Based Biowaivers" was held virtually on December 6, 2021, organized by the University of Maryland Center of Excellence in Regulatory Science and Innovation (M-CERSI), and the Food and Drug Administration (FDA). The workshop focused on the industrial, academic, and regulatory experiences in generating and evaluating permeability data, with the aim to further facilitate implementation of the BCS and efficient development of high-quality drug products globally. As the first international permeability workshop since the BCS based biowaivers was finalized as the ICH M9 guideline, the workshop included lectures, panel discussions, and breakout sessions. Lecture and panel discussion topics covered case studies at IND, NDA, and ANDA stages, typical deficiencies relating to permeability assessment supporting BCS biowaiver, types of evidence that are available to demonstrate high permeability, method suitability of a permeability assay, impact of excipients, importance of global acceptance of permeability methods, opportunities to expand the use of biowaivers (e.g. non-Caco-2 cell lines, totality-of-evidence approach to demonstrate high permeability) and future of permeability testing. Breakout sessions focused on 1) in vitro and in silico intestinal permeability methods; 2) potential excipient effects on permeability and; 3) use of label and literature data to designate permeability class.

Testing Herzberg’s two factor theory on millennials

This study tests Herzberg’s two-factor theory of motivation on working millennials. The purpose of the study is to evaluate what the assumptions and the findings of the original study hold for millennials. With the increasing trend of amalgamation of work life and personal life, job satisfaction and general job happiness have never been more on an individual’s personal life. This impact is far-reaching, with strong indications of affecting the employees’ physical, mental, and social health. In such a scenario, organizations need to keep their employees motivated and satisfied. For studying Herzberg’s two-factor theory of motivation on working millennials, we have selected respondents with work experience of at least one year. We have not restricted the study to any particular industry; hence our sample is across all industries. The type of sampling used was convenient sampling. From the analysis, we have found that for the concerned sample, hygiene factors tower above the motivating factors in terms of job satisfaction for millennials. From the data collection and analysis, we proved that both motivators and hygiene factors affect job satisfaction, unlike what has been said by Herzberg. This study will contribute to the exciting literature on motivation.

Study of trust of government handling of Covid-19 in India and USA and disinformation tactics used by the government

This research aims to find out the status of trust the people of the world’s two prominent democracies – The United States of America, known worldwide as the most powerful democracy, and the Republic of India, known as the World’s Largest Democracy in the handling of the COVID-19 epidemic, that has gripped the entire world by storm. Also, the second objective of this study is to find out if the population of the two nations believe that their governments have actively used disinformation tactics – once thought to be used only by a despot or autocratic governments, on its populace to control the COVID-19 panic and hysteria surrounding it. This study also aims to understand the relationship between trust and disinformation, if any. The study aims to fulfill its objective via individual responses. The survey was conducted via Google Forms and was floated via social media apps like Face book, Linked In, Whatsapp, Instagram, and popular chat sites such as Omegle and Reddit.

Clinicopathological Profile and Outcome of a Large Cohort of Patients with Nonalcoholic Fatty Liver Disease from South Asia: Interim Results of the Indian Consortium on Nonalcoholic Fatty Liver Disease

Background: Previous data from South Asia and India had shown that patients with nonalcoholic fatty liver disease (NAFLD) have mild liver disease severity. There are no data regarding long-term clinical outcomes in patients with NAFLD from South Asia. The aim of the study was to evaluate the clinicopathological profile, severity of NAFLD, and clinical outcomes in a large cohort of patients with NAFLD from South Asia. Methods: In an ongoing real-life study [Indian Consortium on nonalcoholic fatty liver disease (ICON-D)], interim data captured across 23 centers in India over 18 months was analyzed for clinicopathological profile, severity of NAFLD, and hepatic/extrahepatic events on follow-up. Results: Of 4313 patients (mean age 45 ± 12.2 years, males 52%), data on metabolic risk factors in 3553 (82.3%) patients revealed that 378 (10.6%) were lean, 575 (16.2%) overweight, 2584 (72.7%) obese; metabolic syndrome in 1518 (42.7%) and at least one metabolic risk factor in 3292 (92.6%) patients. Evidence of significant or advanced fibrosis assessed with [aspartate transaminase to platelet ratio index (APRI), n = 3196 (74%)], [fibrosis-4 (FIB-4), n = 3554 (82.4%)], [NAFLD fibrosis score (NFS), n = 1924 (44.6%)], [Fibroscan, n = 2475, (57.3%)], and histology [n = 267 (6.2%)] was present in 682 (21.3%), 676 (19%), 397 (20.6%), 715 (29%), and 41 (15.4%) patients, respectively; 246 (10%) patients on Fibroscan and 22 (8.2%) on histology had evidence of cirrhosis. On a mean follow-up 43.5 months, hepatic and extrahepatic events recorded in 1353 (31.3%) patients showed that patients with compensated cirrhosis [71 (5.2%)] had more hepatic [26 (36.7%)] and extrahepatic events [8 (11.3%)] in comparison with those without cirrhosis (P < 0.0001). Conclusion: Around one fifth of patients with NAFLD in South Asia have significant liver disease. Both hepatic and extrahepatic events on follow-up are observed more commonly in patients with nonalcoholic steatohepatitis-related compensated cirrhosis.

FibroScan-AST (FAST) Score for Nonalcoholic Steatohepatitis - Validation in an Indian Cohort

Background

The FibroScan-AST (FAST) score was recently described to detect patients with nonalcoholic steatohepatitis (NASH) having elevated nonalcoholic fatty liver disease (NAFLD) activity score (NAS ≥ 4) and significant fibrosis (≥ F2) on liver biopsy (NASH+ NAS ≥ 4 + F ≥ 2).

Aim

The aim of this study was to validate the FAST score in Indian patients with NAFLD and to derive optimal cut-offs.

Methods

Sixty patients with biopsy-proven NAFLD [men: 38 (63.3%), age 40 (32-52) years] with all parameters for assessing the FAST score within 3 months of liver histology were retrospectively analysed.

Results

Histological NASH was present in 17 patients (28.3%), while 11 (18.3%) patients had NASH + NAS ≥ 4 + F ≥ 2. The area under the curve (AUROC) of the FAST score for discriminating NASH + NAS ≥ 4 + F ≥ 2 was 0.81. Using cut-offs by Newsome et al, the rule-out cut-off (FAST: ≤ 0.35) had a negative predictive value (NPV) of 0.88 [sensitivity: 0.91, specificity: 0.14, negative likelihood ratio (LR): 0.64], while the rule-in cut-off (FAST: ≥ 0.67) had a positive predictive value (PPV) of 0.33 (sensitivity: 0.73, specificity: 0.67, positive LR: 2.22). Fifteen (25%) patients were correctly classified as per histology, while 28 (46.67%) patients fell in the grey zone. On recalculating the optimal cut-offs for our patients, the rule-out cut-off (FAST: ≤ 0.55) had an NPV of 0.95 (sensitivity: 0.90, specificity: 0.45, negative LR: 0.21), while the optimal rule-in cut-off (FAST: ≥ 0.78) had a PPV of 0.70 (sensitivity: 0.64, specificity 0.94, positive LR: 10.39). With these cut-offs, 27 (45%) patients fell in the grey zone and 29 (48.3%) were correctly classified as per histology, performing better than the cut-offs by Newsome et al (P < 0.001).

Conclusion

The FAST score demonstrates good AUROC for detecting NASH with significant fibrosis and inflammation on histology. Cut-offs should be recalibrated based on prevalence of disease.

Lay summary

India has a high burden of NAFLD with an estimated 25 million patients at potential risk for significant liver disease. Liver biopsy remains the gold standard for diagnosing NASH, although its application in routine clinical practice is limited. Noninvasive tests for the simultaneous detection of steatosis, inflammation and fibrosis are thus the need of the hour. The FAST score has been recently suggested for the noninvasive detection of NASH with significant fibrosis (≥ F2) and inflammation (NAS ≥ 4) on liver biopsy. We validated the utility of the FAST score for detecting NASH with significant fibrosis and inflammation on liver biopsy in Indian patients with NAFLD. This noninvasive, easy-to-use and nonproprietary FAST score can correctly classify disease severity in more than 50% patients. However, our results suggest that cut-offs should be recalibrated based on the anticipated prevalence of NASH + NAS ≥ 4 + F ≥ 2 in the given population.