Molecular profiling of gastric cancer in a population with high HIV prevalence reveals a shift to MLH1 loss but not the EBV subtype

The human immunodeficiency virus (HIV) pandemic heavily affects sub‐Saharan Africa. It is associated with persistently active Epstein‐Barr virus (EBV) infection. To determine if this translates into increased frequency of EBV‐associated gastric cancer (EBVaGC), we evaluated molecular profiles of gastric cancer (GC) in Zambia. Patients with GC or premalignant gastric lesions were enrolled in Lusaka, Zambia. We used patients without any of these lesions as a control group. Chromogenic in situ hybridization (CISH) on tumor tissue was used to identify EBVaGC. To identify the microsatellite unstable subtype, immunofluorescence staining for MutL homolog 1 (MLH1) was used. Exposure to EBV and HIV was assessed serologically. We enrolled 369 patients (median age 52 years [IQR 41‐65]; 198 (54%) female). Of these, 72 (20%) had GC and 35 (9%) had gastric premalignant lesions (PL). CISH identified EBVaGC in 5/44 (11%) of those with adequate tissue, while MLH1 loss was identified in 29/45 (64%). Both GC and PL were associated with the highest titers of antibodies to Early antigen‐diffuse (OR 2.5, 95% CI 1.0‐6.1, P = .048 and OR 3.9, 95% CI 1.1‐12.9, P = .03, respectively) at high concentrations. Human immunodeficiency virus infection was associated with a range of antibodies to EBV, but not with any cancer subtype. Despite the association of HIV with persistent EBV, the proportion of EBVaGC in Zambia is similar to populations with a low prevalence of HIV infection. The proportion of microsatellite unstable tumors may be higher than other populations. Human immunodeficiency virus infection is associated active Epstein‐Barr virus (EBV) infection. We found that this does not necessarily translate into an increased proportion of EBV‐associated gastric cancer. We also found a much higher proportion of microsatellite unstable tumors than that reported elsewhere. These findings provide good preliminary data that could be helpful in understanding not only gastric carcinogenesis but probably avenues for new therapies as well.