Background. Rheumatoid arthritis (RA) is an autoimmune disease mainly affecting the joints but has also extra articular manifestations and an increased cardiovascular (CV) co-morbidity. Rheumatoid factor (RF) and antibodies against citrullinated proteins/peptides (ACPA) are diagnostically important and are related to a more severe disease. The aetiology is unknown but RA is considered a complex disease caused by both genetic and environmental factors. The heritability is estimated to be 60% with the main contribution from the HLA region. The relative homogeneity of the population in northern Sweden due to low immigration and founder effects has shown to be suitable for genetic studies. Objectives. The aim of this thesis has been to identify genes contributing to the susceptibility of RA and the CV co-morbidity in particular. To achieve this, multi-case families from the four northern most counties of Sweden were collected for linkage studies to identify susceptibility genes. Association studies with genetic polymorphisms in genes, involved in inflammation or being of importance for atherothrombotic manifestations (ATM) in the general population, were performed in RA-patients concerning ATM e.g. myocardial infarction, angina pectoris with intervention, stroke/TIA, deep vein thrombosis/pulmonary embolism (DVT/PE) at follow-up. Methods & Results. 47 families with 134 affected and 216 unaffected relatives were included in a genome-wide linkage study (GWL) performed with microsatellite markers at an average of 10cM resolution analysed using ABI PRISM 3730 DNA sequencer and non-parametric multipoint linkage in the Merlin program. Eight linked loci were identified with HLA as the most significant and a novel region on chromosome 14. In a follow-up analysis on a custom Illumina chip, with 13 additional families, yielding a total of 198 affected and 197 unaffected relatives. The majority of the 1536 single nucleotide polymorphisms (SNPs) used in the Illumina follow-up analyses was focused on chromosome 14. Statistical analyses with linkage and transmission disequilibrium test narrowed the region to 4 cM, a region containing multiple plausible RA candidate genes (Paper I). In Paper II  serum samples from 163 affected and 157 first degree relatives were analysed with EliA ACPA assay on ImmunoCAP250 for ACPA (IgA, IgG, IgM) and RF (IgA, IgM) isotypes. Both concentrations and frequencies were increased among the relatives compared with controls but lower compared with RA-patients and with a different relative distribution of the isotypes. The genetic contribution to ATM was studied in Paper III and IV using selected SNPs analysed using ABI PRISM 7900HT sequence detector system. In Paper III, RA-patients (n=467) were compared with age and sex matched controls (n=696) with respect to SNPs in tumor necrosis factor receptor II (TNFRII)(M196R), ß-fibrinogen -455 (G-455A), plasminogen activator inhibitor type-1 (PAI-1) (4G/5G) and Factor XIIIA (Val34Leu). Hypertension was predicted by TNFRII R allele and to a higher extent in combination with the A-allele in ß-fibrinogen. The 4G allele in PAI-1 was more frequent in patients with ischemic heart disease (IHD) and the FXIIIA Leu34 variant in patients with DVT/PE. In Paper IV, the minor allele of the polymorphism in growth differentiation factor 15 (GDF15) was found to be associated with RA (n=696) per se but also to ATM, a SNP in the 9p21.3 locus was also associated with ATM. A significant association to stroke was found in female patients homozygote for the minor allele of GDF15. Stoke among male patients was significantly associated with carrying the major allele of two SNPs in the CD40 gene. DVT/PE was associated with the minor allele of GDF15. Conclusion. A novel locus on chromosome 14 of importance for RA susceptibility in northern Sweden was found. The minor allele of TNFRII separately and together with the minor allele of ß-fibrinogen -455 was associated with hypertension and the 4G allele in PAI-1 was associated with IHD and  the Leu34 variant was associated with DVT/PE in RA patients. The GDF15 minor allele was associated with RA per se, ATM and DVT/PE in RA patients and a genotype in the SNP on 9p21.3 was associated with ATM. Stroke among females was associated with GDF15 and stroke among males with two SNPs in CD40.

Background Rheumatoid arthritis (RA) is a systemic progressive destructive joint disease with an increased risk for co-morbidity and premature death if untreated. Cardiovascular disease (CVD) is the main cause of death but also other co-morbid conditions contribute to the patient’s shorter life expectancy. Inflammation is important for the development of CVD, but knowledge of its relationship with other co-morbidities is sparse. Early disease modifying anti rheumatic drugs (DMARDs) can suppress disease activity and improve the long-term outcome. The aim of this thesis was to evaluate prospectively aspects of disease activity and severity in a large cohort of patients with early RA. Predictive and prognostic markers, e.g., antibodies against citrullinated proteins/peptides (ACPAs), occurring in early disease and with implications for disease outcome and co-morbidity were evaluated. Methods Patients with early RA (i.e., symptomatic for ≤12 months) have, since December 1995, been consecutively included in a large survey of prospective and observational studies on the progression of RA and the development of co-omorbidity. Autoantibodies, inflammatory, genetic markers and radiographs have been analyzed. In paper I, 210 RA patients and 102 controls were followed regularly for two years. The predictive value of four different ACPAs in relation to disease activity and radiological progression was evaluated. In Paper II (n = 700) and in Papers III-IV (n =950), patients with early RA from the four northern-most counties of Sweden were followed regularly for 5 years. Data on risk factors and co-morbidity was collected, according to the study protocol, from clinical records and self-reported questionnaires from patients at inclusion into the study cohort and after five years. The predictive value of traditional and potential disease related risk factors for new cardiovascular events (CVE) was evaluated (II). In Paper III, the impact of age at the onset RA, stratified as being young onset RA (<58 years; YORA) and late onset RA (≥58 years; LORA) on disease activity, severity and chosen treatment, was evaluated. In Paper IV, the development of new co-morbidities after RA onset and their relation to inflammatory activity was assessed. Results The presence of anti-mutated citrullinated vimentin (MCV ) antibodies was associated with a more severe disease course, estimated by disease activity score, erythrocyte sedimentation rate (ESR) and swollen joint count after 24 months, compared with anti-CCP2, anti-CCP3, and anti CCP3.1 antibodies. In Paper II, the incidence of a new CVE during 5 years was explained by several of the traditional CV risk factors, and potentiated by a high disease activity. Treatment with DMARDs decreased the risk. In Paper III, LORA patients were associated with greater disease activity/severity at disease onset and over time compared with YORA who were more often ACPA positive. YORA patients were treated earlier with DMARDs, whilst LORA patients were more often treated with corticosteroids and less so with DMARDs early in the course of their disease. In Paper IV, 53%of patients already had one or more co-morbidities already at the onset of RA. After 5 years, 41% of the patients had developed at least one new co-morbidity. ESR at baseline and accumulated disease activity were associated with a new co-morbidity after five years. Conclusion Early RA patients sero-positive for anti- MCV antibodies appeared to have a higher disease activity over time. The occurrence of a new CVE in early RA patients was predicted by traditional risk factors for CVD which were potentiated by a high disease activity. Treatment with DMARDs decreased the risk. Patients with young onset of RA were associated with a higher frequency of ACPA. Late onset of RA was associated with higher disease activity/severity at inclusion and over time. However, LORA patients were more often treated with corticosteroids and less so with DMARDs early in the disease course. Development of a new co-morbidity during the five years following diagnosis was related to ESR.