Family history, genes and gene-environment interaction in risk and prognosis of non-Hodgkin lymphoma

Sammanfattning: Non-Hodgkin lymphoma (NHL) is a heterogeneous group of malignant diseases arising from lymphocytes. There are about 40 subtypes according to the World Health Organizations classification system, differing in morphology, immunophenotype, genetic and/or clinical features. Immunodeficiency, family history, autoimmune disease and some infectious agents are established risk factors for NHL. For most incident NHL patients, however, the reason is not known. One of the most common NHL subtypes, follicular lymphoma (FL), has been associated with smoking in most but not all studies. FL risk is strongly associated with germline genetic variation in the human leukocyte antigen (HLA) DRB1 gene, and aminoacid variation in certain positions of the HLA-DRB1 molecule. In Study I, we used a casecontrol study design to test the hypothesis that smoking is a risk factor for FL among individuals with certain amino-acid combinations in HLA-DRB1, including in positions 70-74 (known as the shared epitope (SE) in rheumatoid arthritis). We found that individuals that carried two SE alleles and were former or current smokers had approximately 2 and 3,5 times increased risk of FL, respectively, as compared to individuals that carried zero SE alleles and never smoked. The interaction was significant when assessed by estimating the attributable proportions of interaction (0.15≤ APoverall≤1.0; 0.005≤ APwomen≤1.0). This finding provides further evidence for a role of smoking in follicular lymphomagenesis, and offers a new model to explore FL etiology. The prognosis of NHL not only varies by subtype, but also within subtypes for reasons that we do not fully understand. Prognostic indices are commonly used in NHL to guide choice of treatment and predict disease course. These include the host factors age and performance status along with indicators of tumor burden, and provide a rough estimation of outcome. Other host factors such as lifestyle factors and medical history as well as host genetic variation are being explored for their potential contribution to better prognostic prediction. This was the focus of Study II-IV. In Study II, we tested the hypothesis that smoking status, attained education (proxy for socioeconomic status), body mass index, ultraviolet radiation exposure, autoimmune disease or family history of hematopoietic malignancy influenced overall and lymphomarelated survival among NHL patients overall and major subtypes. We found evidence that current smoking, few years of attained education and history of autoimmune disease increased the risk of all-cause death among NHL patients overall. Smoking also increased the risk of death among patients with diffuse large B-cell lymphoma. Attained education was also associated with lymphoma-related death. Further studies are needed to understand mechanisms for these associations and how they should be accounted for in the clinical setting. In Study III, we tested the hypothesis that single nucleotide polymorphisms (SNPs) have an impact on FL survival or progression by 1) conducting a meta-analysis of two GWAS, and 2) exploring 22 SNPs previously reported to be associated with FL outcome. In the metaanalysis, no SNP was associated with FL survival at genome-wide significance, although one SNP in the ABCA10 gene on chromosome 17 was borderline associated (rs10491178: Prandom= 5.24 x10-8). In line with previous studies, two linked SNPs in IL8 and one in CD46 were negatively associated with FL progression. Our results mainly provide further evidence of an impact of SNPs involved in immune functions on follicular lymphoma outcome. In Study IV, we further investigated the hypothesis that inherited factors influence survival among patients with a lymphoid malignancy by testing concordance in survival between two first-degree relatives with lymphoid malignancies. We found that individuals with a firstdegree relative with good prognosis had a better survival than individuals with a first-degree relative with expected or poor survival among individuals with the same broad type of lymphoid malignancy, especially indolent lymphomas. Our results support a role of heritability in the outcome primarily of indolent lymphomas.