Answers are grounded in this paper and cite their sources.
Reviews prognostic value of PNPLA3 testing for liver-related events in steatotic liver disease, indicating clinical utility depends on follow-up time, ethnicity, and baseline risk.
Non-alcoholic fatty liver disease (NAFLD), newly renamed metabolic dysfunction-associated steatotic liver disease (MASLD),1 has a worldwide prevalence of at least 30% among adults and is becoming the leading cause of liver cirrhosis and hepatocellular carcinoma (HCC).2 Since only a minority of subjects with MASLD progress towards liver-related events (LRE), risk stratification is crucial to optimize the allocation of resources for the care of MASLD individuals. Different algorithms aimed to accomplish this goal are being generated by using different strategies, including the evaluation of the inborn genetic risk. Indeed, MASLD heritability is elevated, with estimates ranging between 20% and 70%.3 The rs738409 C > G, a common sequence variant in PNPLA3 encoding for the p.I148M aminoacidic substitution, is the variant accounting for the largest fraction of variability of SLD development4 and progression from simple steatosis to the occurrence of LRE.5, 6 However, PNPLA3 single nucleotide polymorphism (SNP) genotyping in the clinical management of MASLD is not widespread. This is likely because this variant and composite genetic risk scores have not yet unequivocally provided additional prognostic information compared to the strongest clinical predictor of LRE, that is, stage of liver fibrosis. In this context, the results presented by Seko et al.7 deserve particular attention. Indeed, by studying a cohort of 1550 individuals with biopsy-proven MASLD from six Japanese medical centres, the authors found that PNPLA3 CG/GG robustly predicted LRE with a 16-fold increased risk. Importantly, the risk was independent of all the other variables, including fibrosis as estimated by the FIB-4 index. By employing a risk stratification model based on the combination of FIB-4 (>2.67 vs. ≤2.67) and PNPLA3 genotype (CG/GG vs. CC), they identified three subsets of individuals with very different cumulative risks of LRE over a follow-up of 10 years (Figure 1). Strikingly, among the individuals with the protective PNPLA3 CC genotype, none was diagnosed with hepatocellular carcinoma (HCC) and only one experienced a single LRE during the follow-up. Given the impressive findings, Seko et al. suggested that risk stratification in Japan should best rely on the combination of FIB-4 class and PNPLA3 rs738409 genotype. Moreover, they speculated that therapeutic strategies capable of lowering the hepatic levels of the variant PNPLA3 protein, if confirmed effective in the ongoing studies in MASLD individuals, should be eagerly awaited in this Country. Indeed, while the latest experimental studies in the homozygous PNPLA3 148 M/M (rs738409 GG) knock-in-mutant mice are further extending our knowledge of the mechanisms by which the variant PNPLA3 protein promotes the severity of the MASLD phenotype,8 implicating even changes in the gut microbiota, PNPLA3 silencing with antisense oligonucleotides was already proven effective in reducing inflammation and fibrosis in this same animal model.9 In the last 15 years, a growing number of common variants have been associated with MASLD development and progression.10 The majority primarily affects hepatic fat content (HFC) and as a result, hepatic inflammation and fibrosis, consistent with the evidence that elevated HFC cause MASLD progression.11 The implementation of polygenic risk scores (PRS) by leveraging these variants has been a main effort aimed to maximize the information concerning individual genetic risk profile.11, 12 From a prognostic point of view, different PRSs have been associated with MASLD-related fibrosis, cirrhosis, hepatic decompensation and HCC.11, 13, 14 Notably, in a study including 260 687 participants of European descent from the UK Biobank over a median follow-up of 9 years, a PRS for HFC improved the diagnostic accuracies and positive predictive values for LRE of different clinical fibrosis scores.14 This is consistent with the notion that, while liver fibrosis is undoubtedly the strongest predictor of MASLD-related hepatic and extrahepatic outcomes,15 a prediction based only on fibrosis stage is incomplete and somehow near-sighted. Indeed, taking a snapshot of the fibrosis stage does not give information on the propensity and the speed of disease progression.16 In fact, the additional prognostic information gathered by these two components can be captured only by detailing the individual metabolic and genetic risk profiles. Lastly, behavioural factors, such as the specific composition of the diet, the level of physical activity and the type of sedentary behaviour, can impact the natural history of MASLD and even offset the harmful effect of predisposing genetic factors.17 As alluded to, some studies showed that the addition of genetics to clinical risk scores modestly improves their predictive performance.18 One might wonder whether the design of a study may affect the capability of bringing out the role of genetics in prognostic stratification. Indeed, there are at least two aspects to be considered in this context: the duration of the follow-up and the overall risk level of the study population. The longer the follow-up, the higher the probability that the effect of genetic predisposition will emerge over the baseline stage of fibrosis. Indeed, the predictive potential of the baseline clinical fibrosis scores reduces with the increase in the duration of the follow-up.19 Moreover, the higher the level of risk, that is, the relative incidence rate of LRE, the higher the probability that clinically relevant results are obtained. Indeed, in multivariable models, the effect of genetic risk is partially diluted by its contribution to the baseline stage of fibrosis, and only a high relative rate of LRE can ensure adequate statistical power. What specific characteristics of the study cohort by Seko et al. provided the PNPLA3 variant with a striking prognostic potential? There are at least two explanations: the long duration of follow-up (7.1 years, range 1.0–24.0), providing enough time for LRE to occur, and the geographical origin of the study. Indeed, individuals of Asian descent are enriched with the PNPLA3 variant as compared to other African and European ancestries,1, 20, 21 and in the present study the prevalence of the SNP is further increased due to the study setting, that is, subjects with biopsy-proven MASLD from hepatology tertiary referral centres. Notably, Asians have lower BMI, which could have mitigated the effect of the PNPLA3 variant.22 However, at any given BMI, Asians display higher levels of visceral fat, which is known to potentiate the effect of the PNPLA3 and other variants on the risk of fatal and non-fatal advanced liver disease.23 Moreover, genetics, epigenetics and lifestyle may modify the penetrance of genetic variants.24 Interestingly, in this study, liver-related mortality is the first cause of death during follow-up, far surpassing cardiovascular-related death, which is well recognized as the principal cause of death in MAFLD individuals from Western countries.15 This finding is in line with previous studies in Japanese subjects with MASLD,25 and it may be due to a more advantageous metabolic profile and cardiovascular care in this Country. However, the low cardiovascular death should not change the relationship between the PNPLA3 variant and liver-related mortality. Indeed, Mendelian randomization studies have convincingly excluded a causal association between the PNPLA3 rs730409 variant and ischemic heart disease.26 Consistently, in population studies, the PNPLA3 SNP is not associated with ischemic-heart disease-related, or extrahepatic cancer-related, mortality.27 Rather, it is tempting to speculate that the low rate of competitive deaths from CVD has favoured the high relative rate of LRE and liver-related deaths observed in this cohort, therefore helping to disclose their association with the PNPLA3 variant. Consistently, a recent study concludes that PNPLA3 genotyping helped redefining the risk of cirrhosis of subjects in the intermediate FIB-4 class only in those individuals with diabetes.5 Conversely, the improvement of risk prediction by different risk SNPs was modest at the population level,19 where a PRS for HFC improved risk prediction only in those at high metabolic risk and/or in medium-high classes of clinical fibrosis scores.14 In conclusion, the body of evidence that assessing the inborn risk may improve risk stratification in individuals with MASLD is growing. The potential is undoubted, and it seems clear that genetic profile should be assessed early in life to maximize its value and in tertiary centre, rather than at the general population, level. Further research should define the better settings for its application and dictate the rules for a convenient clinical translation. Data sharing is not applicable to this article as no data sets were generated or analysed during the current study.
No takes yet. Share an insight, caveat, or question.
Vespasiani‐Gentilucci et al. (2023) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: