Non-invasive assessment of NAFLD as systemic disease-A machine learning perspective

Abstract

Background & aims: Current non-invasive scores for the assessment of severity of non-alcoholic fatty liver disease (NAFLD) and identification of patients with non-alcoholic steatohepatitis (NASH) have insufficient performance to be included in clinical routine. In the current study, we developed a novel machine learning approach to overcome the caveats of existing approaches.

Methods: Non-invasive parameters were selected by an ensemble feature selection (EFS) from a retrospectively collected training cohort of 164 obese individuals (age: 43.5±10.3y; BMI: 54.1±10.1kg/m2) to develop a model able to predict the histological assessed NAFLD activity score (NAS). The model was evaluated in an independent validation cohort (122 patients, age: 45.2±11.75y, BMI: 50.8±8.61kg/m2).

Results: EFS identified age, γGT, HbA1c, adiponectin, and M30 as being highly associated with NAFLD. The model reached a Spearman correlation coefficient with the NAS of 0.46 in the training cohort and was able to differentiate between NAFL (NAS≤4) and NASH (NAS>4) with an AUC of 0.73. In the independent validation cohort, an AUC of 0.7 was achieved for this separation. We further analyzed the potential of the new model for disease monitoring in an obese cohort of 38 patients under lifestyle intervention for one year. While all patients lost weight under intervention, increasing scores were observed in 15 patients. Increasing scores were associated with significantly lower absolute weight loss, lower reduction of waist circumference and basal metabolic rate.

Conclusions: A newly developed model (http://CHek.heiderlab.de) can predict presence or absence of NASH with reasonable performance. The new score could be used to detect NASH and monitor disease progression or therapy response to weight loss interventions.

Fig 1. Performance of the new score within the training cohort.

Based on available non-invasive parameters in an obese cohort of 164 patients a new score was generated. This score correlated strongly with the NAS (A), assessed by two independent pathologists, in this cohort. For separation between NAFL (NAS ≤ 4) and NASH (NAS > 4) an AUC of 0.73 was achieved (B).

Fig 2. The new score achieves reasonable performance in a validation cohort.

The newly generated score was evaluated in an independent cohort from the University Hospital Würzburg (A, B) with 122 patients. For the validation cohort the new score achieved a Spearman correlation coefficient with the NAS of 0.28 (A) and was able to differentiate between NAFL (NAS ≤ 4) and NASH (NAS > 4) with an AUC of 0.7 (B).

Fig 3. Differences in response to weight loss therapy indicated by the new score.

In a subgroup of 38 patients receiving weight loss therapy for 1 year (Tables 2 and 3) only 23 patients (60%) exhibited a reduction of the newly generated score, while weight and most other metabolic health indicators were improved. Individuals with an increase of the score (15 patients, 40%) during study duration had significantly lower absolute weight loss (A), lower reduction of waist circumference (B), nominally lower reduction in body fat content (C), and a significantly less pronounced impairment of basal metabolic rate (D).

Fig 4. The new score indicates liver and adipose tissue function as possible determinants for weight loss efficiency.

38 patients receiving weight loss therapy for 1 year (Tables 2 and 3) were grouped by either reduction (23 patients) or increase (15 patients) of the new score. In the group with decreased score values at end of the study duration we found significantly higher serum albumin (A) and ghrelin (B) concentrations at start and end of the study. Conversely, concentration of the unsaturated fatty acid palmitic acid (C16:0) was significantly lower at both time points in patients with decrease of the score during weight loss therapy.