OBJECTIVES

To assess effect of GLP1a and/or SGLT2i on development of advanced liver disease in T2DM patients with NASH/NAFLD or at risk of NASH.

METHODS

Data collected from a proprietary US EMR database were limited to adult T2DM (ICD10 E11. 9) patients diagnosed with NASH/NAFLD (ICD10 K76, 75.81) or with a risk profile of [Age >50 + ALT >30 U/L + BMI >30] and without viral hepatitis or evidence of alcohol abuse. Index date was the first calculable FIB4 between Jul 2015 to Jun 2017 with >1y history and >2y follow-up or to death. Advanced liver disease was defined as cirrhosis (compensated or decompensated), hepatocellular carcinoma, or liver transplant. All-cause death was not considered. Cox proportional models were used to determine hazard ratios (HR) between propensity-score (PS) matched groups. All multiple regression models used complete-case analysis and included treatment, MACE, FIB4, CVD, gender, age, race/ethnicity, and all covariates with p <0.200 by individual models.

RESULTS

Of 30MM adult patients, 31323 met all study criteria and received SGLT2i and/or GLP1a at index (treated, 10%, 3208) or never received these therapies prior to or during follow up from index (not-treated, 90%, 28115). Mean follow up was 36.5 months. Advanced liver disease at index was significantly lower for the treated group (16%, 516/3208 v 29%, 8169/28115 not-treated, p<0.001). PS-matching of patients without advanced liver disease at index yielded 1720 pairs. Cox hazard ratios for developing advanced liver disease were favorable for treatment (0.74 [0.63-0.87] p<0.001) and unfavorable for FIB4 >2.67 (2.17 [1.59-2.96] p<0.001), MACE (2.32 [1.91-2.82], p<0.001), Age ≥75 (1.84 [1.91-2.82] p=0.005), and CVD (1.27 [1.07-1.52] p=0.007) at index.

CONCLUSIONS

In T2DM patients with NASH/NAFLD or at risk of NASH, the risk of developing advanced liver disease was lower in patients treated with SGLT2i and/or GLP1a.