by Victoria A. Fisher, M.P.H.
DCEG investigators have been evaluating the etiology of NHL for many years. As the seventh most commonly diagnosed cancer in the U.S., NHL comprises a group of related diseases of the lymph system. There are more than 30 biologically and clinically distinct subtypes of NHL, and treatment is largely determined by the particular subtype. Researchers have suspected there might be both etiologic similarities and differences among the various subtypes that could inform the search for risk factors and help understand genetic susceptibility for this complex disease. However, previous studies investigating common genetic variants have been limited in size and scope.
To shed some light on the genetic factors contributing to this heterogeneous set of malignancies, investigators in the Occupational and Environmental Epidemiology Branch—Sonja Berndt, Pharm.D., Ph.D., and Nathaniel Rothman, M.D., M.P.H., M.S.—initiated a collaborative and interdisciplinary large-scale genome-wide association study (GWAS) including 13 case-control studies from the International Lymphoma Epidemiology Consortium (InterLymph) and 9 cohorts from the NCI Cohort Consortium. DNA samples from approximately 9,000 NHL cases and 3,000 controls were scanned at NCI and made available for analysis. In addition, several extramural studies contributed data from their scans.
To date, the team has reported 27 novel independent single nucleotide polymorphisms (SNPs) for four NHL subtypes, including chronic lymphocytic leukemia/small lymphocytic lymphoma (CLL/SLL), diffuse-large B-cell lymphoma (DLBCL), follicular lymphoma (FL), and marginal zone lymphoma (MZL), with only a small amount of overlap in these findings across lymphoma subtypes.
“Findings from the NHL GWAS have made a substantial contribution to our understanding of the genetic etiology of NHL,” said Stephen J. Chanock, M.D., Director of DCEG. “This research has helped establish the basis for further investigation of the striking heterogeneity between NHL subtypes.”
Previous work has led the way for the present large-scale initiative. Early candidate gene approaches suggested some genetic variants were associated with multiple NHL subtypes or NHL as a whole, while others were likely limited to particular subtypes.
Consistent with this hypothesis, the first GWAS for CLL and FL reported distinct loci, underscoring the value of assessing specific subtypes. However, these studies were limited in sample size.
“Our initial pooled candidate gene studies and long history of successful collaboration within InterLymph laid the foundation for the NHL GWAS, which is the largest and most comprehensive attempt to date to understand the genetic basis of NHL and how it interacts with the environment,” said Dr. Rothman.
Results from the effort have been enlightening. As noted, although each NHL subtype showed a pattern of mostly distinct associations, there were some associations that were shared across subtypes, with either the same SNP or different SNPs in the same locus showing association.
“Our findings of both shared and distinct genetic risk are consistent with the results for non-genetic risk factors in the InterLymph Non-Hodgkin Lymphoma Subtypes Project,” Dr. Berndt said. (See Linkage article “Probing the Complexity of Non-Hodgkin Lymphoma Subtypes—A Consortial Effort”).
Perhaps most notable, investigators discovered genetic variants in one or more human leukocyte antigen (HLA) regions were associated with CLL, DLBCL, FL, and MZL. HLA regions include several key immune response genes, which highlights the critical role that alterations in the immune system play in lymphomagenesis.
CLL/SLL has been shown to have a strong familial component. Investigators identified 10 independent SNPs in 9 novel loci for CLL, as well as one new independent SNP at a previously established locus (Berndt, 2013). These findings almost doubled the number of risk loci for CLL.
Investigators reported five non-HLA loci associated with FL risk (Skibola, 2014). In addition, they showed striking evidence of very strong signals for FL across multiple HLA regions. These findings substantially expanded the number of loci associated with FL and provided evidence that multiple common variants outside as well as within the HLA region make a significant contribution to FL risk.
DLBCL is the most common lymphoma subtype and is clinically aggressive. With the discovery of four new loci, investigators provided new evidence for genetic susceptibility to DLBCL and point to immune pathways in the pathogenesis of DLBCL (Cerhan, 2014).
Two independent SNPS in the HLA region were associated with MZL risk, providing the first evidence for inherited susceptibility to MZL, and a role for genetic variation in the HLA region (Vijai, 2015).
Drs. Berndt, Rothman, and collaborators continue to search for novel genetic variants associated with NHL. They are extending their work to include scanning of less common subtypes, such as mantle cell lymphoma, T cell lymphoma, and together with investigators in the Genetic Epidemiology Branch (Neil E. Caporaso, M.D., Lynn R. Goldin, Ph.D., and Mary Lou McMaster, M.D.), Waldenström macroglobulinemia.
However, research goals extend beyond the discovery of new loci. Investigators are coordinating the use of the GWAS data across multiple institutions to support more than 20 intramural and extramural projects, including studies of gene-environment interactions, genetic mosaicism, the contribution to telomere length, and application of new methods of genomic analyses.
“Although we have discovered numerous variants associated with the risk of specific subtypes, the causal variants underlying these associations are unknown,” Dr. Berndt said. Plans are underway to conduct a number of secondary analyses to further elucidate the genetic architecture of NHL and pinpoint the functional variants contributing to disease.
In addition, investigators are evaluating potential genetic modifiers of established risk factors, taking advantage of previous data harmonization efforts. “Ongoing and future studies combining data on occupational and environmental exposures, lifestyle factors, medical history, and genetic risks promise to lead to further insight into this complex set of diseases,” said Dr. Rothman.
Learn more about this research and key collaborators:
• International Lymphoma Epidemiology Consortium (InterLymph)
• NCI Cohort Consortium
Berndt SI, Skibola CF, Joseph V, et al. Genome-wide association study identifies multiple risk loci for chronic lymphocytic leukemia. Nat Genet 2013 Aug;45:868-876. doi: 10.1038/ng.2652
Skibola CF, Berndt SI, Vijai J, et al. Genome-wide association study identifies five susceptibility loci for follicular lymphoma outside the HLA region. Am J Hum Genet 2014 Oct 2;95:462-471. doi: 10.1016/j.ajhg.2014.09.004.
Cerhan JR, Berndt SI, Vijai J, et al. Genome-wide association study identifies multiple susceptibility loci for diffuse large B cell lymphoma. Nat Genet 2014 Nov;46:1233-1238. doi: 10.1038/ng.3105
Vijai J, Wang Z, Berndt SI, et al. A genome-wide association study of marginal zone lymphoma shows association to the HLA region. Nat Commun 2015 Jan 8;6:5751. doi: 10.1038/ncomms6751.