6-thioguanine nucleotide levels and anti-tumour necrosis factor alpha trough levels in inflammatory bowel disease patients on combination therapy: a retrospective multicentre study.

Abstract

Introduction: Thiopurine co-therapy with anti-tumour necrosis factor alpha (anti-TNFalpha) agents is associated with higher anti-TNFalpha drug levels and reduced immunogenicity in patients with Inflammatory Bowel Disease (IBD). However, the 6-thioguanine nucleotide (6-TGN) threshold level required to confer this benefit remains unclear. Aims & Methods: This study aimed to evaluate the relationship between serum 6-TGN levels and trough levels of infliximab (IFX) and adalimumab (ADA) in IBD patients on combination therapy. We performed a retrospective multicentre study of IBD patients receiving combination anti-TNFalpha and thiopurine maintenance therapy for at least 3 months duration with stable doses for at least 2 months. All 6-TGN levels measured within 3 months prior to or on the same date as IFX or ADA trough levels, between January 2015 and August 2021, were included. Undetectable 6-TGN levels (<5pmol/8x108 RBC) were excluded. Primary outcomes were IFX and ADA levels. Secondary outcomes were antibodies to IFX (ATI) or ADA (ATA). Spearman's rho was used to correlate 6-TGN and anti-TNFalpha levels. Receiver operator characteristic curves were used to define the optimal 6-TGN cut-off levels associated with therapeutic anti-TNFa levels (>=5mcg/ml for IFX, >=7.5mcg/ml for ADA), using the Liu index. The Mann-Whitney U test was used to compare 6-TGN levels between patients with and without detectable ATI. Result(s): A total of 743 paired 6-TGN and anti-TNFalpha levels (640 IFX, 103 ADA) from 354 patients (263 IFX, 91 ADA) were included. 363/640 (57%) IFX levels and 65/103 (63%) ADA levels were measured in patients on standard anti- TNFalpha dosing (5mg/kg every 8 weeks for IFX, 40mg every 2 weeks for ADA). The median 6-TGN level was 252pmol/8x108 RBC, median IFX level was 6.45mcg/ml and median ADA level was 7.70mcg/ml. 6-TGN and IFX levels were significantly positively correlated (rho 0.132, p=0.0008), however the correlation between 6-TGN and ADA levels was not significant (rho 0.145, p=0.143). The optimal 6-TGN cut-off that predicted therapeutic IFX levels was 261pmol/8x108 RBC (area under the curve (AUC) 0.57, sensitivity 0.58, specificity 0.56) for patients on standard IFX dosing and 227.5pmol/8x108 RBC (AUC 0.58, sensitivity 0.57, specificity 0.58) for patients on escalated IFX dosing. The optimal 6-TGN cut-off that predicted therapeutic ADA levels was 245pmol/8x108 RBC (AUC 0.57, sensitivity 0.62, specificity 0.53) for patients on standard ADA dosing and 201pmol/8x108 RBC (AUC 0.63, sensitivity 0.80, specificity 0.46) for patients on escalated ADA dosing. Seven patients had detectable ATI. Patients with ATI had significantly lower 6-TGN levels compared to those without ATI (median 6-TGN 133 vs 253pmol/8x108 RBC, p=0.010). No patients had detectable ATA. Conclusion(s): 6-TGN levels are associated with IFX trough levels, however the association between 6-TGN and ADA levels remains unclear. Target 6-TGN levels in the lower end of the therapeutic 6-TGN range may be adequate to maintain therapeutic anti-TNFalpha levels, particularly in patients on escalated anti-TNFalpha dosing. However, the optimal 6-TGN cut-offs associated with therapeutic IFX and ADA levels that were calculated from our multicentre data showed poor predictive ability, and may differ depending on individual laboratory assays.