Eric R Kallwitz1, Bamidele O Tayo2, Mark H Kuniholm3, Jianwen Cai4, Martha Daviglus5, Richard S Cooper2, Scott J Cotler6. 1. Division of Hepatology, Loyola University Medical Center, Maywood, Illinois. Electronic address: ekallwitz@lumc.edu. 2. Department of Public Health Sciences, Stritch School of Medicine, Loyola University Chicago, Maywood, Illinois. 3. School of Public Health, SUNY at Albany, Albany, New York. 4. UNC Gillings School of Global Public Health, University of North Carolina, Chapel Hill, North Carolina. 5. Institute for Minority Health Research, University of Illinois, Chicago, Illinois. 6. Division of Hepatology, Loyola University Medical Center, Maywood, Illinois.
Abstract
BACKGROUND & AIMS: Nonalcoholic fatty liver disease (NAFLD) disproportionally affects Hispanic/Latino populations. However, the magnitude varies among Hispanic/Latino ethnic groups. We investigated the mechanisms of these disparities. METHODS: We examined associations of NAFLD-associated genetic variants and continental ancestry with suspected NAFLD, levels of alanine aminotransferase (ALT), and liver fibrosis using data from the Hispanic Community Health Study/Study of Latinos-a population-based study of Hispanic/Latino adults in the United States. We evaluated data from 16,415 Hispanic/Latino adults in 4 cities from 2008 through 2011. Subjects suspected of having NAFLD or liver fibrosis were identified based on unexplained increases in levels of aminotransferases and FIB-4 score, respectively. RESULTS: Among the 9342 participants with available genetic and aminotransferase data, the PNPLA3 G allele (odds ratio [OR], 1.53; 95% CI, 1.41-1.66), TM6SF2 T allele (OR, 1.41; 95% CI, 1.20-1.67), and PPP1R3B G allele (OR, 1.16; 95% CI, 1.06-1.28) were associated with suspected NAFLD. PNPLA3 G was also associated with increased levels of ALT, except in participants with Dominican and South American backgrounds, and with liver fibrosis. The frequency of PNPLA3 G was high (41%) and TM6SF2 T (5%) was low in Hispanic/Latinos. PNPLA3 G frequency differed among Hispanic background groups with the highest proportion in Mexicans (52%) and the lowest proportion in Dominicans (23%). After adjustment for demographic, clinical, and behavioral factors, as well as PNPLA3 G, TM6SF2 T, and PPP1R3B G, American ancestry had a positive association with level of ALT (r = 6.61%; P < .001), whereas African (r = -3.84%; P < .001) and European (r = -4.31%; P < .001) ancestry were inversely associated with level of ALT. CONCLUSIONS: American ancestry and PNPLA3 G are independent predictors of ALT levels in US Hispanic/Latinos and may in part explain NAFLD disparities in US Hispanic/Latinos.
BACKGROUND & AIMS:Nonalcoholic fatty liver disease (NAFLD) disproportionally affects Hispanic/Latino populations. However, the magnitude varies among Hispanic/Latino ethnic groups. We investigated the mechanisms of these disparities. METHODS: We examined associations of NAFLD-associated genetic variants and continental ancestry with suspected NAFLD, levels of alanine aminotransferase (ALT), and liver fibrosis using data from the Hispanic Community Health Study/Study of Latinos-a population-based study of Hispanic/Latino adults in the United States. We evaluated data from 16,415 Hispanic/Latino adults in 4 cities from 2008 through 2011. Subjects suspected of having NAFLD or liver fibrosis were identified based on unexplained increases in levels of aminotransferases and FIB-4 score, respectively. RESULTS: Among the 9342 participants with available genetic and aminotransferase data, the PNPLA3 G allele (odds ratio [OR], 1.53; 95% CI, 1.41-1.66), TM6SF2 T allele (OR, 1.41; 95% CI, 1.20-1.67), and PPP1R3B G allele (OR, 1.16; 95% CI, 1.06-1.28) were associated with suspected NAFLD. PNPLA3 G was also associated with increased levels of ALT, except in participants with Dominican and South American backgrounds, and with liver fibrosis. The frequency of PNPLA3 G was high (41%) and TM6SF2 T (5%) was low in Hispanic/Latinos. PNPLA3 G frequency differed among Hispanic background groups with the highest proportion in Mexicans (52%) and the lowest proportion in Dominicans (23%). After adjustment for demographic, clinical, and behavioral factors, as well as PNPLA3 G, TM6SF2 T, and PPP1R3B G, American ancestry had a positive association with level of ALT (r = 6.61%; P < .001), whereas African (r = -3.84%; P < .001) and European (r = -4.31%; P < .001) ancestry were inversely associated with level of ALT. CONCLUSIONS: American ancestry and PNPLA3 G are independent predictors of ALT levels in US Hispanic/Latinos and may in part explain NAFLD disparities in US Hispanic/Latinos.