Aims and objectives: The overall aim of this thesis was to assess the role of immunogenicity on long-term treatment response and immune-mediated adverse events to anti-tumour necrosis factor (TNF) agents. Objectives included: (i) assessment of the clinical value of pharmacological monitoring in adalimumab, etanercept and certolizumab-treated patients with rheumatoid arthritis (RA) (ii) performing steps towards implementation of these tests in clinical practice (iii) evaluation of the risk of lupus and vasculitis-like events (iv) establishing a prospective observational study in psoriatic arthritis (PsA) to assess future biomarkers of biologic treatment response. Methods: The thesis utilised samples and data from the Biologics in RA Genetics and Genomics Study Syndicate (BRAGGSS) to assess the clinical utility of anti-TNF pharmacological testing in RA. Anti-drug antibodies (ADAbs) were measured using a radioimmunoassay (RIA) and random timed serum drug levels measured using an ELISA over 12 months. The concordance between RIA and ELISA for the detection of ADAbs in adalimumab treated patients was then assessed. A microcosting analysis was performed to estimate the actual resource use for implementing pharmacological monitoring in clinical practice. The anti-TNF drug-specific risk of lupus and vasculitislike adverse events was investigated using data from the British Society for Rheumatology Biologics Register for RA (BSRBR-RA). Finally, a new study was established to assess predictors of treatment response in PsA patients initiated on biologics [Outcome of Treatment in Psoriatic Arthritis Study Syndicate (OUTPASS)]. Results: In BRAGGSS, 24.8% of RA patients on adalimumab had detectable ADAbs (measured by RIA) by 12 months. At 3 months, ADAb formation and low adalimumab levels were significant predictors of no response at 12 months according to the European League Against Rheumatism (EULAR) criteria (AUC 0.71 [95% CI: 0.57, 0.85). Body mass index (BMI) and poor adherence were additional predictors of low adalimumab and etanercept drug levels. However, when RIA was compared with ELISA assays to test ADAbs to adalimumab, the concordance was poor (AUC 0.65 [95% CI: 0.59-0.71]). ADAbs (measured by RIA) to certolizumab were detected in 37% patients and were significantly associated with lower drug levels over 12 months (ß=-0.037, 95% CI: -0.055, -0.018, p<0.0001). In BSRBR-RA, the combined adjusted hazard ratio of lupus and vasculitis-like events in anti-TNF treated patients compared to biologic naïve patients was 2.43 (95% CI 1.24 to 4.74). Infliximab conferred the highest overall risk followed by etanercept, however 95% CIs overlapped following adjustment. The interim analysis of OUTPASS patients revealed age, gender, baseline disease activity, BMI and use of concomitant methotrexate or sulfasalazine were associated with biologic treatment response in PsA. Conclusions: ADAbs and low drug levels are useful biomarkers of adalimumab treatment response in RA. Infliximab conferred the highest risk of lupus and vasculitislike events but the absolute risk was small. The role of ADAbs and drug levels as biomarkers to inform the management of PsA patients treated with biologic drugs requires further evaluation.