Modest variation in Lp(a) levels but consistent ASCVD risk across South Asian subgroups (MASALA and UK Biobank) (JACC Adv 2026)

Original title: Lipoprotein(a) in Disaggregated South Asian Groups: An Analysis of MASALA and the UK Biobank

JACC Adv · 13 October 2026 · 7

Bal KS, Reyes MP, Kanaya AM, Taub P, Tsai MY, Kandula NR, Islam N, Wilkinson MJ, Ramsis M, Mahmud E, Shah NS, Kulkarni A et al.

This cohort analysis of MASALA and UK Biobank data evaluated Lp(a) concentrations and ASCVD risk across disaggregated South Asian populations. Among 2,074 MASALA participants, median Lp(a) was 20.7 [10.5-43.0] mg/dL, with modest variation by country of origin (Bangladesh 23, India 19, Pakistan 22, Sri Lanka 17 mg/dL). In 6,845 UK Biobank participants (median Lp(a): 31 [13-65] nmol/L), the association between Lp(a) and incident ASCVD did not differ by country of origin or South Asian ethnicity (P interaction > 0.05). While nongenetic factors showed only modest associations with Lp(a) levels, the consistent risk gradient across subgroups confirms that Lp(a)-driven ASCVD risk is uniform regardless of South Asian ancestry.

Original abstract

Background: South Asian individuals have a high prevalence of elevated lipoprotein(a) [Lp(a)], but little is known about the variation between disaggregated South Asian subgroups.

Objectives: The authors aimed to evaluate Lp(a) levels, nongenetic influences, and the association with incident atherosclerotic cardiovascular disease (ASCVD) in South Asian populations using data from the Meditators of Atherosclerosis in South Asians Living in America (MASALA) and the UK Biobank (UKB) studies.

Methods: Lp(a) levels were assessed by South Asian country of origin and by state among Indian-origin participants. The association between nongenetic factors and Lp(a) levels was evaluated in multivariable linear regression models. Cox proportional hazards models evaluated the association between Lp(a) and ASCVD events in UKB participants.

Results: Among 2,074 MASALA participants born in the South Asian subcontinent (mean age: 56.7 ± 9.6 years), median [Q1-Q3] Lp(a) was 20.7 [10.5-43.0] mg/dL with significant variation by country of origin (P = 0.035). Median Lp(a) was 23 [11-48], 19 [10-37], 22 [11-47], and 17 [10-34] mg/dL in participants from Bangladesh, India, Pakistan, and Sri Lanka, respectively. Vegetarian diet, statin use, postmenopausal status, and fewer years in the United States were associated with higher Lp(a) levels while alcohol consumption was associated with lower Lp(a) levels. However, these associations were all very modest. Among 6,845 UKB participants (mean age: 55.1 ± 8.7 years; median Lp(a): 31 [13-65] nmol/L), the associations between Lp(a) and ASCVD did not vary by country of origin or South Asian ethnicity (P interaction > 0.05).

Conclusions: There is modest variation in Lp(a) levels, but no difference in the association between Lp(a) and ASCVD outcomes, across South Asian subgroups.

Summary written by lp-a.org from the published abstract; figures as published. Page updated 16 October 2026. Methods.