Perlman, Riki

Summary Among individuals of European Ancestry (EA), genome‐wide association studies (GWAS) have identified single nucleotide polymorphisms (SNPs) and polygenic risk scores (PRSs) associated with non‐Hodgkin lymphoma (NHL) risk. We evaluated subtype‐specific PRSs, based on established EA‐SNPs, in Israeli Jews (IJ) and Palestinian Arabs (PA) and performed a GWAS in the combined ethnic groups to identify new loci. We included three common pathologically confirmed subtypes: diffuse large B‐cell (DLBCL), follicular (FL) and marginal zone (MZL) lymphomas. Controls were frequency matched to cases by age and sex. Among 752 IJ (201—DLBCL; 130—FL; 54—MZL/367—controls) and 593 PA (203—DLBCL; 41—FL/349‐controls), we computed PRSs weighted by EA‐derived effect estimates and used logistic regression models adjusted for confounders. In the combined ethnic groups of IJ and PA, subtype‐specific PRSs were significantly associated with the corresponding subtype, with a 1.69‐fold, 2.21‐fold and 2.26‐fold risk for DLBCL, FL and MZL, respectively; however, these effect sizes were attenuated compared to those reported in EA and varied by ethnicity. In the GWAS of the combined ethnic groups, two novel SNPs in the 6p21.32 locus were associated with DLBCL risk. Additional GWAS studies are needed among Jewish and Arab populations to improve genetic risk prediction for NHL in these ethnic groups.

AbstractAlthough autoimmune diseases (AIDs) are known to predispose to non‐Hodgkin lymphoma (NHL), their association with NHL prognosis has rarely been investigated. We examined associations between autoimmunity and B‐cell NHL onset by comparing AID history (determined by self‐report and medication review and supplemented by chart review where possible) among 435 adult B‐NHL patients in Hadassah‐Hebrew University Medical Center, diagnosed 2009‐2014, and 414 age‐and‐sex frequency‐matched controls. We examined AIDs as a whole, B‐ and T‐cell–mediated AIDs, and autoimmune thyroid diseases. Among cases, we used Kaplan‐Meier and Cox regression models to assess the association of AID with overall survival and relapse‐free survival, adjusting for prognostically important patient and disease characteristics such as Ki67% staining, International Prognostic Index, rituximab treatment, and histological subgroup.Autoimmune diseases were associated with B‐NHL (odds ratio [OR] = 1.95; 95% confidence interval (CI), 1.31‐2.92), especially AIDs mediated by B‐cell activation (OR = 5.20; CI, 1.90‐14.3), which were particularly associated with marginal zone lymphoma (OR = 19.3; CI, 4.59‐80.9). We found that time to relapse for all B‐NHL patients with AIDs was significantly shorter (mean of 49.21 mo [±3.22]) than among patients without AID (mean of 59.74 mo [±1.62]), adjusted hazard ratio [HRadj] = 1.69 (CI, 1.03‐2.79). Specifically, in patients with diffuse large B‐cell lymphoma, of whom 91.8% had received rituximab, a history of B‐cell–mediated AIDs was associated with shorter relapse‐free survival and overall survival, HRadj = 8.34 (CI, 3.01‐23.1) and HRadj = 3.83 (CI, 1.20‐12.3), respectively.Beyond confirming the well‐known association between AIDs and B‐NHL, we found that AID is an adverse prognostic factor in B‐cell lymphoma, associated with a shortened time to relapse, suggesting that there are specific therapeutic challenges in the subgroup of patients suffering from both these diseases. Further work is required to address mechanisms of resistance to standard treatment in the setting of AID‐associated B‐NHL. In the era of immunotherapy, these findings have particular relevance.