Stratified and subgroup analyses of five single-nucleotide polymorphisms (SNPs) within the major histocompatibility complex region, which were independently associated with classical Hodgkin lymphoma (cHL). Results are shown for A) rs2248462 (class I region, MICB), B) rs2395185 (class II region, HLA-DRA), C) rs2734986 (class I region, HLA-A), D) rs6904029 (class I region, HCG9), and E) rs6903608 (class II region, HLA-DRA). Odds ratios (ORs, represented by boxes with the area of each box inversely proportional to the variance of the estimate) and 95% confidence intervals (CIs, error bars) were derived using multiple logistic regression assuming a log-additive genetic model and adjusting for sex (male or female), country (up to eight indicator variables after excluding one country as the reference, depending on the analysis: France, Germany, Spain, Czech Republic, Ireland, United Kingdom, Denmark, Sweden, and the Netherlands), and eight principal components analysis eigenvectors for the genome-wide association study (GWAS) analyses only. The dashed vertical line represents the OR of the SNP in the analysis of total cHL among all subjects, and the width of the diamond is the corresponding 95% CI. In the analysis stratified by study, the GWAS included the EPILYMPH study (EPILYMH-GWAS), Scandinavian Lymphoma Etiology Study (SCALE-GWAS), the Scotland and Newcastle Lymphoma Group, and Young Adult Hodgkin Case–Control Study analyzed together (referred to as the UK Studies-GWAS), and the Northern Dutch Hodgkin Lymphoma Study (Netherlands-GWAS). The independent replication was done with data from the EPILYMPH study (EPILYMPH - Replication), and the Scotland and Newcastle Lymphoma Group, Young Adult Hodgkin Case–Control Study, and Epidemiology and Genetics Lymphoma Case–Control Study analyzed together (referred to as the UK Studies-Replication). Phomogeneity was on the basis of the Cochran Q test statistic and was used to evaluate between-study heterogeneity in results. Associations between the SNPs and cHL subgroups (including histological subtype, EBV status, and age) were performed, and P values (Phomogeneity) for the χ2 test of homogeneity are presented to indicate differences in the OR between subgroup analyses. For rs2248462 (A) and rs 2395185 (B), results were derived from a combined analysis of the GWAS and independent replication phase results using an inverse variance weighting meta-analysis. All statistical tests were two-sided. Ca = case patients; Chr = chromosome; Co = control subjects; MCHL = mixed cellularity Hodgkin lymphoma; NSHL = nodular sclerosis Hodgkin lymphoma.