A small study from the Mayo Clinic of patients on stable doses of thiopurines showed a trend toward increased 6TGN levels and leucopenia after the addition of sulphasalazine and mesalazine, but not balsalazide[42]; however, elevations in 6TGN do not seem to be dose-dependent. A Dutch group subjected 17 patients on stable doses of thiopurines to 2 g of mesalazine for 4 weeks, dose escalation of the 5-ASA to 4 g for another 4 weeks, followed by cessation for Paclitaxel 4 weeks. Median 6TGN levels increased from 370 at baseline to 553 with 2 g mesalazine (P ≤ 0.05). Escalation to 4 g of mesalazine

did not lead to significant increases in 6TGN levels (median 553 to 572). After mesalazine washout, 6TGN levels decreased to 449 (P ≤ 0.05). Interestingly, 6MMP levels decreased significantly from a median of 1676 to 880 (P ≤ 0.05), but this required 4 g of mesalazine. There was also a favorable improvement (i.e., fall) in the 6MMP : 6TGN ratio.[43] Two Australian studies have highlighted the ability of thiopurine metabolite testing to improve outcomes with thiopurine therapy.[27, 28] Both clustered patient results into five

groups (see Table 1): underdosed/rapid metabolisers, non-adherent, refractory, ‘thiopurine shunters’ and overdosed patients. Haines et al. studied 63 consecutive IBD patients who, despite a stable dose of thiopurines for the last 3 months, had persistent clinically active disease.

Kennedy et al. Navitoclax reviewed the outcomes of 151 consecutive patients undergoing metabolite testing. 6TGN levels were subtherapeutic in 29% and 43% of patients with active disease and non-adherence was 9% and 1%, respectively in each study. The metabolite results led to the addition of allopurinol in 9% and 10% of patients, respectively, in each study. Metabolite testing revealed that 40% and 58% of patients Atazanavir were truly refractory to thiopurine therapy, and 13% and 21% of patients were overdosed. In the study by Haines et al., 87% of patients improved with thiopurine optimization and 15 patients avoided a change of therapy, compared to only 18% of patients who were not optimised (P = 0.0001). Three patients with therapeutic levels whose doctors ignored the algorithm and dose-escalated showed no improvement. In the study by Kennedy et al., 74% of patients with subtherapeutic 6TGN levels improved with dose escalation, and across the entire cohort, optimization of thiopurines improved outcomes in 38% of patients. These two manuscripts make a compelling case for the application of thiopurine metabolite testing in order to optimise the dosage and use of thiopurines and achieve better outcomes. In some centres, metabolite testing is accepted as standard of care for IBD patients on thiopurines.