Pepsinogen and Helicobacter pylori: Serum biomarkers for gastric cancer risk in a diverse United States population

Abstract

BackgroundSerum Pepsinogen (PG) and Helicobacter pylori (Hp) have potential as biomarkers to identify persons at high risk for gastric cancer (GC) in low incidence populations. We explored the use of these biomarkers in a diverse US population. MethodsSingle institution study using serum samples from 32 newly diagnosed GC (before treatment) and 60 non-cancer participants. ELISA-based PG-I, PG-II, Hp IgG, and Hp virulence seromarker (CagA, VacA) tests were conducted to examine differences between GC and non-GC participants. ResultsMedian age was 58 (IQR 48–68); 56 % were females; 30.1 % were white, 14.8 % black, 28.7 % Hispanic, 6.1 % Asian and 18.2 % other/unknown race/ethnicity. Median values of Hp (47.9 vs 12.6 U/mL, p = 0.003) and pepsinogen ratio (PGR=PGI/PGII) (4.0 vs 7.5, p = 0.003) differed between GC and non-GC. Performance of pepsinogen tests using “standard” cut-offs (PGI ≤ 70 ng/mL and PGR ≤ 3) were highly specific (91.7 %), but not sensitive (34.4 %), while Hp IgG test using “standard” cut-off (≥ 30 U/mL) was sensitive (78.8 %), but less specific (41.2 %). Optimized cut-off values identified in our population using Youden’s Index were PGR ≤ 5.2 and Hp ≥ 17.5 U/mL. Using a combination of these values resulted in a significant increase in test sensitivity (87.9 %) with lower specificity (50.8 %). ConclusionThe combination of pepsinogens and Hp show promise as biomarkers of GC risk in a racially and ethnically diverse US population. Optimal biomarker cut-off points for US populations may differ from those established in East Asia. By adjusting cut-offs there is potential to design GC risk stratification tools tailored specifically for the diverse population within the US.