A Role for Thiopurine Metabolites in the Synergism Between Thiopurines and Infliximab in Inflammatory Bowel Disease

A Role for Thiopurine Metabolites in the Synergism Between Thiopurines and Infliximab in... Abstract Background Interactions between principal cytotoxic thiopurine metabolites, that is 6-thioguanine nucleotides [6-TGN], and infliximab [IFX] and anti-IFX antibodies [Abs] may contribute to higher effectiveness of IFX-thiopurine combination therapy than monotherapies in inflammatory bowel disease. Methods To examine if thiopurine metabolites influenced trough IFX and anti-IFX Abs, 89 patients previously assessed for anti-IFX Abs were included. To assess if IFX influenced thiopurine metabolites, eight patients who had responded to 12 weeks of intensified IFX at a constant thiopurine dosing were included. Results In the first cohort, IFX-thiopurine combination therapy reduced anti-IFX Ab detection [8/40; 20%] as compared with IFX monotherapy [22/49; 45%], odds ratio [OR] 0.31 [0.12–0.80], p < 0.05. 6-TGN was significantly lower in anti-IFX Ab-positive patients (50 pmol/8 × 108 red blood cells [RBC] vs 105, p < 0.01). All anti-IFX Ab-positive patients had 6-TGN < 117 pmol/8 × 108 RBC (sensitivity 100% [63–100], specificity 47% [29–65], area under the curveROC = 0.82, p < 0.01). Trough IFX was similar between anti-IFX Ab-negative patients in IFX monotherapy and IFX-thiopurine combination therapy [5.1 μg/mL vs 4.9, p = 0.76]. 6-TGN and IFX did not correlate [rP = 0.04, p = 0.83; rS = 0.02, p = 0.89, respectively]. In the second cohort, trough IFX increased during IFX intensification [ΔIFX median 6.5 μg/mL, p = 0.02], but 6-TGN was stable [6-TGN at Weeks 0, 4, 8, 12: 90 pmol/8 × 108 RBC, 93, 101, 90; p > 0.05]. Methylated mercaptopurine metabolite associations were consistently negative. Conclusions Superior effect of IFX-thiopurine combination therapy over monotherapies partly relates to decrease in anti-IFX Abs, which associates with 6-TGN levels and has a lower therapeutic threshold than during thiopurine monotherapy. Additional benefit likely ascribes to synergy between different anti-inflammatory modes of action rather than direct drug interactions. Inflammatory bowel disease, infliximab, antibodies against infliximab, therapeutic drug monitoring, thiopurines, 6-TGN 1. Introduction Patients with inflammatory bowel disease [IBD] usually need anti-inflammatory treatment, and tumour necrosis factor [TNF] inhibitors such as infliximab [IFX] are a cornerstone in the therapeutic armentarium.1 Unfortunately, up to half of patients lose the effect of IFX maintenance therapy.2 This can be caused by formation of antibodies against IFX [anti-IFX Abs] leading to decreased or absent circulating levels of active drug.3 Continued IFX therapy in the presence of anti-IFX Abs can also result in severe hypersensitivity reactions.4 Patients who develop anti-IFX Abs are prone to later develop anti-adalimumab Abs, leading to repeat loss of response.5 Thus, it is pertinent to find ways to minimise the risk of anti-IFX Ab formation. Previous studies have convincingly shown that combination therapy with IFX and thiopurines is more effective than monotherapy with either agent in immunosuppressor-naïve IBD patients.6–8 Whereas TNF inhibitors are large-molecular weight monoclonal antibodies that target the pro-inflammatory cytokines, TNF-α, thiopurines (i.e. 6-mercaptopurine [6-MP] and its pro-drug, azathioprine [AZA]), are small-molecular drugs that undergo complex enzymatic metabolism with formation of both therapeutic active and inactive metabolites.9 The primary immunosuppressive metabolites are 6-thioguanine nucleotides [6-TGN], which are incorporated into DNA and RNA and thereby induce apoptosis in, for example, lymphocytes.10 The mechanism for improved effectiveness of IFX-thiopurine combination treatment over monotherapies has been hypothesised to originate from thiopurines’ decreasing immunogenicity of IFX or, alternatively, that thiopurines directly influence circulating IFX levels.11 It has relatively consistently been observed that fewer patients receiving combination therapy have detectable anti-IFX Abs6,8,11,12─and it was recently reported that 6-TGN associated with anti-IFX Abs.11 The primary aim of this study was to investigate the interplay between principal cytotoxic thiopurine metabolites (6-TGN and methylated mercaptopurine metabolites [6-MeMP]) and IFX and anti-IFX Abs. Furthermore, we aimed to to establish thresholds for thiopurine metabolites that associate with improved pharmacokinetics of IFX or less likelihood of anti-IFX Abs. 2. Methods 2.1. Study design 2.1.1. Influence of thiopurine metabolites on infliximab and anti-infliximab antibodies The first part was a retrospective cohort study that examined if thiopurine metabolites associate with trough IFX levels and anti-IFX Abs. The study population comprised all patients treated with IFX at our tertiary IBD centre since 2009, in whom IFX and anti-IFX Abs had been assessed [at the discretion of the treating physician until 2012 and hereafter at predefined time points according to local guidelines: after 1 year of treatment and then at every 6 months─or at any time of suspected treatment failure]. The two principal cytotoxic cytosolic thiopurine metabolites, 6-TGN and 6-MeMP, were determined in red blood cells [RBC] in stored samples at time points corresponding exactly to those when IFX and anti-IFX Abs had been measured. We included patients who had received IFX induction treatment and at least one maintenance infusion. Patients who received concomitant thiopurines [AZA or 6-MP] were included only if they had received a stable dose for at least 2 months before sampling, in order to allow for thiopurine metabolites to achieve steady-state erythrocyte concentrations.13 Patients were excluded if they received systemic steroid treatment > 5 mg/day, allopurinol, methotrexate, or ciclosporin. This study was approved by the Danish Health and Medical Authority [H-15018092] and the Danish Data Protection Agency [HGH-2016–019:04434]. 2.1.2. Influence of infliximab on thiopurine metabolites The second part of the study was a post-hoc analysis of data from a randomised controlled trial, and examined if IFX influences the concentration of thiopurine metabolites. The study population comprised eight Crohn’s disease patients with active luminal disease despite standard IFX maintenance therapy. These patients had received an intensified IFX regimen comprising 5 mg/kg every 4 weeks for a total of 12 weeks, during which all eight patients had a clinical response (Crohn’s Disease Activity Index [CDAI] decrease of ≥70).14,15 From Week 12 until end of follow-up at Week 20, two patients had received IFX every 4 weeks, two had IFX every 6 weeks, two had IFX every 8 weeks, one had discontinued IFX, and one was lost to follow-up. All patients received steady concomitant thiopurine treatment with either AZA [n = 6] or 6-MP [n = 2] throughout all 20 study weeks. Trough IFX had been measured at Weeks 0, 4, 8, 12, and 20, and were supplemented with measurements of 6-TGN and 6-MeMP in stored samples obtained at the same time points. This study was approved by the Danish Medicines Agency [EudraCT 2009-009926-94], the regional ethics committees [HA-2009-009], and the Danish Data Protection Agency [2007-58-0015; 750.89-2]. 2.2. Analyses 2.2.1. Infliximab and anti-infliximab antibodies Circulating trough IFX concentrations were measured in serum by radioimmunoassay [RIA] (limit of detection [LOD] 0.07 μg/mL)] until end of 2012, after which it was routinely replaced by reporter gene assay [RGA; iLiteTM] [LOD 0.13 μg/mL].16,17 Samples were classified as anti-IFX Ab-positive or -negative according to the LOD of the assay, and measured by RIA until mid-2014 and then by enzyme immunoassay [EIA] or RGA [iLiteTM].16–18 All analyses were done by a commercial provider under blinded conditions [Euro Diagnostica, SE; formerly Biomonitor A/S, DK]. In the event that a patient had both a positive and a negative anti-IFX Ab test at different time points, the positive sample and the corresponding IFX trough level were used. The most recent sample was used if more than one negative anti-IFX Ab test was available. 2.2.2. Thiopurine metabolites RBC concentrations of 6-TGN and 6-MeMP metabolites were determined in biobanked buffy-coat samples [stored at -80°C] which had all been collected exactly at the day of IFX and anti-IFX Ab measurements. Quantification was carried out by hot acid hydrolysis followed by ultra-performance liquid chromatography, as previously described.19–21 Control blood spiked with known amounts of thioguanine and methylmercaptopurine monophosphate nucleotides [Jena Bioscience, DE] was used for calibrators. Thiopurine metabolite levels were normalised to buffy-coat haemoglobin [Hgb] and measured in nmol/mmol Hgb followed by conversion to pmol/8 x 108 RBC [Abacus 380 blood analyser, HU]. 2.2.3. Thiopurine methyltransferase phenotype Thiopurine methyltransferase [TPMT] enzyme activity was assessed in hypotonic buffy-coat lysates by chromatography as described above, except that a BEH C18 column and a water/formic acid-methanol buffer system were used for separation. Calibrators were 6-MeMP spiked into blank matrix. Haemoglobin levels were used for calculating TPMT activity units [1U = 1 nmol 6-MeMP x h-1 x mL packed erythrocytes-1] [Abacus 380 blood analyser]. 2.3. Statistics Descriptive data were presented as percentages for discrete variables, and as median with interquartile range [IQR] for continuous variables. Correlations were investigated using linear correlation analysis expressed as Pearson’s correlation coefficient [rp], and by non- parametric correlation analysis expressed as Spearman’s correlation coefficient [rs]. Univariate analyses of discrete variables were done by Fisher’s exact test or chi square test, as appropriate. Continuous variables were compared by Mann-Whitney test [unpaired] or Wilcoxon signed-rank test [paired]. Therapeutic thresholds for thiopurine metabolites were determined by receiving operator curve [ROC] analysis. Multivariate logistic [anti-IFX Abs] or linear [IFX] regression was done by entering the following variables into the models: gender, disease type, body mass index [BMI], frequency of IFX infusions, 6-TGN, 6-MeMP. Values below LOD were set to null. Missing data were excluded. Analyses were done in SPSS version 24 [IBM, NY, USA] or in GraphPad Prism version 5 [GraphPad Software, CA, USA]. Two-sided p-values < 0.05 were significant. 3. Results 3.1. Influence of thiopurine metabolites on anti-IFX Abs and IFX 3.1.1. Study population The study population comprised 89 patients, of whom 40 patients [45%] received IFX-thiopurine combination therapy [AZA 88%], and 49 patients [55%] had IFX monotherapy [Table 1]. The majority of patients had Crohn’s disease [71%] with a moderate disease duration [median 7.5 years], and had received long-term IFX therapy at time of study [median 15 months]. Table 1. Characteristics of cohort 1   All patients n = 89  IFX monotherapy n = 49  IFX-thiopurine combination therapy n = 40  p-Value  Male sex, n [%]  46 [52]  28 [57]  18 [45]  0.29  Age, years median [IQR]  40 [29-50]  41 [31-50]  38 [28-48]  0.69  Crohn’s disease, n [%]  63 [71]  35 [71]  28 [70]  1.00  Disease duration, years median [IQR]  8 [3-15]  8 [3-14]  8 [4-15]  0.95  Age at diagnosis, years median [IQR]  26 [20-35]  26 [20-39]  26 [20-33.25]  0.76  Active smoking, n [%]  16 [18]  9 [18]  7 [18]  1.00  Body mass index [BMI], median [IQR]  25 [22-28]  25 [23-27]  24 [22-28]  0.44  Montreal classification [CD]  Age at onset        < 0.01   A1 [< 17 years old], n [%]  6 [10]  6 [17]  0 [0]     A2 [1740 years old], n [%]  46 [73]  20 [57]  26 [93]     A3 [> 40 years old], n [%]  11 [17]  9 [26]  2 [7]    Location        0.80   L1 [Ileal], n [%]  5 [8]  3 [9]  2 [7]     L2 [Colonic], n [%]  37 [59]  22 [63]  15 [54]     L3 [Ileocolonic], n [%]  21 [33]  10 [29]  11 [39]     L4 [Isolated upper disease], n [%]  0 [0]  0 [0]  0 [0]    Behaviour        0.06   B1 [Non-stricturing, non-penetrating], n [%]  30 [48]  21 [60]  9 [32]     B2 [Stricturing], n [%]  16 [25]  8 [23]  8 [29]     B3 [Penetrating], n [%]  17 [27]  6 [17]  11 [39]    Perianal, n [%]  20 [32]  7 [20]  13 [46]  < 0.05  Montreal Classification [UC]  Extent        0.88   E1 [Ulcerative proctitis], n [%]  2 [8]  1 [7]  1 [8]     E2 (Left sided [distal]), n [%]  14 [54]  7 [50]  7 [58]     E3 (Extensive [pancolitis]), n [%]  10 [38]  6 [43]  4 [33]    Extraintestinal manifestations, n [%]  19 [21]  12 [24]  7 [18]  0.42  Previous intestinal resection, n [%]  27 [30]  13 [27]  14 [35]  0.39  Concomitant thiopurine   Azathioprine, n [%]  -  -  35 [88]  -   Azathioprine dose, mg/kg median [IQR]  -  -  1.7 [1.5-2.0]  -   Mercaptopurine dose, mg/kg median [IQR]  -  -  0.8 [0.7-1.0]  -   Time on thiopurines, months median [IQR]  -  -  21 [12-40]  -  TPMT activity   Normal [homozygote wild type], n [%]  -  -  36 [90]  -   Intermediate [heterozygote mutant], n [%]  -  -  4 [10]  -   Enzyme deficiency [homozygote mutant], n [%]  -  -  0 [0]  -   Concomitant aminosalicylates [in colitis], n [%]  5 [19]  0 [0]  5 [42]  < 0.05  IFX regimen        0.84   5 mg/kg, q < 8 weeks, n [%]  35 [39]  18 [37]  17 [43]     5 mg/kg, q = 8 weeks, n [%]  43 [48]  25 [51]  18 [45]     5 mg/kg, q > 8 weeks, n [%]  11 [12]  6 [12]  5 [13]     Time on IFX, months median [IQR]  15 [11-27]  13 [8-38]  17 [12-25]  0.97   Previous episodic IFX, n [%]  36 [40]  19 [39]  17 [43]  0.83  Albumin mg/dL, median [IQR]  43 [40-45]  43 [40-45]  43 [40-45]  0.50  CRP mg/dL, median [IQR]  3 [3-3]  3 [3-5]  3 [3-3]  0.10  Haemoglobin mM, median [IQR]  8.6 [8.0-9.2]  8.8 [8.3-9.2]  8.4 [7.7-9.1]  < 0.05  White blood cells × 09/L, median [IQR]  6.6 [5.1-8.0]  6.8 [5.8-8.0]  6.1 [4.7-8.1]  0.16    All patients n = 89  IFX monotherapy n = 49  IFX-thiopurine combination therapy n = 40  p-Value  Male sex, n [%]  46 [52]  28 [57]  18 [45]  0.29  Age, years median [IQR]  40 [29-50]  41 [31-50]  38 [28-48]  0.69  Crohn’s disease, n [%]  63 [71]  35 [71]  28 [70]  1.00  Disease duration, years median [IQR]  8 [3-15]  8 [3-14]  8 [4-15]  0.95  Age at diagnosis, years median [IQR]  26 [20-35]  26 [20-39]  26 [20-33.25]  0.76  Active smoking, n [%]  16 [18]  9 [18]  7 [18]  1.00  Body mass index [BMI], median [IQR]  25 [22-28]  25 [23-27]  24 [22-28]  0.44  Montreal classification [CD]  Age at onset        < 0.01   A1 [< 17 years old], n [%]  6 [10]  6 [17]  0 [0]     A2 [1740 years old], n [%]  46 [73]  20 [57]  26 [93]     A3 [> 40 years old], n [%]  11 [17]  9 [26]  2 [7]    Location        0.80   L1 [Ileal], n [%]  5 [8]  3 [9]  2 [7]     L2 [Colonic], n [%]  37 [59]  22 [63]  15 [54]     L3 [Ileocolonic], n [%]  21 [33]  10 [29]  11 [39]     L4 [Isolated upper disease], n [%]  0 [0]  0 [0]  0 [0]    Behaviour        0.06   B1 [Non-stricturing, non-penetrating], n [%]  30 [48]  21 [60]  9 [32]     B2 [Stricturing], n [%]  16 [25]  8 [23]  8 [29]     B3 [Penetrating], n [%]  17 [27]  6 [17]  11 [39]    Perianal, n [%]  20 [32]  7 [20]  13 [46]  < 0.05  Montreal Classification [UC]  Extent        0.88   E1 [Ulcerative proctitis], n [%]  2 [8]  1 [7]  1 [8]     E2 (Left sided [distal]), n [%]  14 [54]  7 [50]  7 [58]     E3 (Extensive [pancolitis]), n [%]  10 [38]  6 [43]  4 [33]    Extraintestinal manifestations, n [%]  19 [21]  12 [24]  7 [18]  0.42  Previous intestinal resection, n [%]  27 [30]  13 [27]  14 [35]  0.39  Concomitant thiopurine   Azathioprine, n [%]  -  -  35 [88]  -   Azathioprine dose, mg/kg median [IQR]  -  -  1.7 [1.5-2.0]  -   Mercaptopurine dose, mg/kg median [IQR]  -  -  0.8 [0.7-1.0]  -   Time on thiopurines, months median [IQR]  -  -  21 [12-40]  -  TPMT activity   Normal [homozygote wild type], n [%]  -  -  36 [90]  -   Intermediate [heterozygote mutant], n [%]  -  -  4 [10]  -   Enzyme deficiency [homozygote mutant], n [%]  -  -  0 [0]  -   Concomitant aminosalicylates [in colitis], n [%]  5 [19]  0 [0]  5 [42]  < 0.05  IFX regimen        0.84   5 mg/kg, q < 8 weeks, n [%]  35 [39]  18 [37]  17 [43]     5 mg/kg, q = 8 weeks, n [%]  43 [48]  25 [51]  18 [45]     5 mg/kg, q > 8 weeks, n [%]  11 [12]  6 [12]  5 [13]     Time on IFX, months median [IQR]  15 [11-27]  13 [8-38]  17 [12-25]  0.97   Previous episodic IFX, n [%]  36 [40]  19 [39]  17 [43]  0.83  Albumin mg/dL, median [IQR]  43 [40-45]  43 [40-45]  43 [40-45]  0.50  CRP mg/dL, median [IQR]  3 [3-3]  3 [3-5]  3 [3-3]  0.10  Haemoglobin mM, median [IQR]  8.6 [8.0-9.2]  8.8 [8.3-9.2]  8.4 [7.7-9.1]  < 0.05  White blood cells × 09/L, median [IQR]  6.6 [5.1-8.0]  6.8 [5.8-8.0]  6.1 [4.7-8.1]  0.16  IQR, interquartile range; CD, Crohn’s disease; UC, ulcerative colitis; IFX, infliximab; q, frequency of dosing; CRP, C-reactive protein; TPMT, thiopurine methyltransferase. View Large 3.1.2. Antibodies against infliximab IFX-thiopurine combination therapy significantly reduced detection of anti-IFX Abs (8/40 [20%]) compared with IFX monotherapy (22/49 [45%]), odds ratio [OR] 0.31 [0.12-0.80], p < 0.05. Anti-IFX Abs appeared functionally active, since their presence resulted in undetectable trough IFX [median < LOD, IQR < LOD] as opposed to IFX levels of median 4.9 μg/mL, IQR 2.6-7.6 in anti-IFX Ab-negative patients [Figure 1]. Figure 1. View largeDownload slide Antibodies against infliximab. Infliximab [IFX] trough levels in patients’ positive [+] or negative [-] for anti-IFX antibodies [Abs]. Figure 1. View largeDownload slide Antibodies against infliximab. Infliximab [IFX] trough levels in patients’ positive [+] or negative [-] for anti-IFX antibodies [Abs]. Levels of 6-TGN were significantly lower in anti-IFX Abs-positive patients receiving IFX-thiopurine combination therapy [median 50 pmol/8 × 108 RBC, IQR 11-66] compared with those without detectable anti-IFX Abs [median 105 pmol/8 × 108 RBC, IQR 70-158, p < 0.01] [Figure 2A]. This association persisted in both univariate and multivariate logistic regression models [Table 2]. ROC analysis also revealed a significant difference in 6-TGN between anti-IFX Ab-positive and -negative patients (area under the curve [AUC] = 0.82 [0.68-0.97], p < 0.01). Of note, all patients with detectable anti-IFX Abs had 6-TGN < 117 pmol/8 × 108 RBC (sensitivity 100 [63-100], specificity 47 [29-65]). There was no difference in 6-MeMP between anti-IFX Ab-positive and -negative patients [median 150 pmol/8 × 108 RBC, IQR 35-1000 vs median 214, IQR 86-569, p > 0.05] [Figure 2B]. Table 2. Logistic regression of association between antibodies against infliximab and multiple study variables Variable  Univariate logistic regression coefficient, [SE]  p-Value  Multivariate regression coefficient, [SE]  p-Value  Male  -0.303  [0.450]  0.50  -1.569  [1.445]  0.28  Crohn’s disease  -0.295  [0.486]  0.54  -1.468  [1.815]  0.42  Body mass index [BMI]  0.002  [0.059]  0.97  -0.026  [0.113]  0.82  IFX frequencya  -  -  0.77  -  -  0.84   = q8  -0.257  [0.482]  0.59  -0.893  [1.526]  0.56   > q8  0.201  [0.769]  0.79  20.775  [14443]  1.00  6-TGN, median [IQR]  0.021  [0.009]  0.02  0.066  [0.030]  0.03  6-MeMP, median [IQR]  0.000  [0.001]  0.75  -0.003  [0.002]  0.12  Variable  Univariate logistic regression coefficient, [SE]  p-Value  Multivariate regression coefficient, [SE]  p-Value  Male  -0.303  [0.450]  0.50  -1.569  [1.445]  0.28  Crohn’s disease  -0.295  [0.486]  0.54  -1.468  [1.815]  0.42  Body mass index [BMI]  0.002  [0.059]  0.97  -0.026  [0.113]  0.82  IFX frequencya  -  -  0.77  -  -  0.84   = q8  -0.257  [0.482]  0.59  -0.893  [1.526]  0.56   > q8  0.201  [0.769]  0.79  20.775  [14443]  1.00  6-TGN, median [IQR]  0.021  [0.009]  0.02  0.066  [0.030]  0.03  6-MeMP, median [IQR]  0.000  [0.001]  0.75  -0.003  [0.002]  0.12  6-MeMP, methylated mercaptopurine metabolites; 6-TGN, 6-thioguanine nucleotides; IFX, infliximab; q, frequency of dosing; IQR, interquartile range; SE, standard error. a < q8 reference. View Large Figure 2. View largeDownload slide Thiopurine metabolites and antibodies against infliximab. 6-thioguanine nucleotides [6-TGN]: [A] and methylated mercaptopurine metabolites [6-MeMP]; [B] in patients positive [+] or negative [-] for anti-IFX antibodies [Abs]. Figure 2. View largeDownload slide Thiopurine metabolites and antibodies against infliximab. 6-thioguanine nucleotides [6-TGN]: [A] and methylated mercaptopurine metabolites [6-MeMP]; [B] in patients positive [+] or negative [-] for anti-IFX antibodies [Abs]. 3.1.3. Infliximab IFX trough levels were similar in anti-IFX Ab-negative patients on IFX monotherapy [median 5.1 μg/mL, IQR 2.4-8.6] compared with those on IFX-thiopurine combination therapy [median 4.9 μg/mL, IQR 3.1-6.5, p > 0.05]. There was no correlation between 6-TGN and IFX [rP = 0.04, p = 0.83; rS = 0.02, p = 0.89; respectively] [Figure 3A], or 6-MeMP and IFX [rS = -0.15, p = 0.43; rP = 0.05, p = 0.81; respectively] [Figure 3B]. Lack of associations between IFX and multiple study variables, including thipurine metabolites, were also found in both univariate and multiple regression models [Table 3]. The frequency of IFX infusions was, however, expectedly associated with IFX trough levels. Table 3. Multiple regression of association between infliximab and multiple study variables Variable  Univariate regression coefficient, [SE]  p-Value  Multivariate regression coefficient, [SE]  p-Value  Male  -0.357  [0.969]  0.71  -0.846  [1.660]  0.61  Crohn’s disease  0.534  [1.057]  0.62  1.672  [1.766]  0.35  Body mass index [BMI]  0.050  [0.120]  0.68  0.062  [0.178]  0.73  IFX frequencya  -2.402  [0.667]  0.001  -1.401  [1.123]  0.22  6-TGN, median [IQR]  0.012  [0.009]  0.20  0.018  [0.10]  0.09  6-MeMP, median [IQR]  -0.001  [0.001]  0.41  -0.002  [0.002]  0.20  Variable  Univariate regression coefficient, [SE]  p-Value  Multivariate regression coefficient, [SE]  p-Value  Male  -0.357  [0.969]  0.71  -0.846  [1.660]  0.61  Crohn’s disease  0.534  [1.057]  0.62  1.672  [1.766]  0.35  Body mass index [BMI]  0.050  [0.120]  0.68  0.062  [0.178]  0.73  IFX frequencya  -2.402  [0.667]  0.001  -1.401  [1.123]  0.22  6-TGN, median [IQR]  0.012  [0.009]  0.20  0.018  [0.10]  0.09  6-MeMP, median [IQR]  -0.001  [0.001]  0.41  -0.002  [0.002]  0.20  6-MeMP, methylated mercaptopurine metabolites; 6-TGN, 6-thioguanine nucleotides; IFX, infliximab; IQR, interquartile range; SE, standard error. a < q8 or q = 8 or q > 8. View Large Figure 3. View largeDownload slide Thiopurine metabolites and infliximab. Corresponding levels of infliximab [IFX] and 6-thioguanine nucleotides [6-TGN]: [A] and methylated mercaptopurine metabolites [6-MeMP]; [B] in patients negative for anti-IFX antibodies [Abs]. Figure 3. View largeDownload slide Thiopurine metabolites and infliximab. Corresponding levels of infliximab [IFX] and 6-thioguanine nucleotides [6-TGN]: [A] and methylated mercaptopurine metabolites [6-MeMP]; [B] in patients negative for anti-IFX antibodies [Abs]. 3.2. Influence of IFX on thiopurine metabolites 3.2.1. Study population The study population comprised eight patients with Crohn’s disease who all had manifest treatment failure of IFX maintenance therapy at time of inclusion, with response to 12 weeks of intensified IFX therapy at a constant thiopurine dosing throughout 20 weeks [Table 4]. Table 4. Characteristics of cohort 2.   Patients n = 8  Male sex, n [%]  3 [38]  Age, years median [IQR]  39 [30-45]  Disease duration, years median [IQR]  10 [5-24]  Age at diagnosis, years median [IQR]  26 [20-32]  Active smoking, n [%]  3 [38]  Body mass index [BMI], median [IQR]  2419 [25]  Montreal Classification    Age at onset     A1 [< 17years old], n [%]  0 [0]   A2 [17–40 years old], n [%]  5 [62]   A3 [> 40 years old], n [%]  3 [38]   Location     L1 [Ileal], n [%]  0 [0]   L2 [Colonic], n [%]  4 [50]   L3 [Ileocolonic], n [%]  3 [37]   L4 [Isolated upper disease], n [%]  1 [13]   Behaviour     B1 [Non-stricturing, non-penetrating], n [%]  6 [75]   B2 [Stricturing], n [%]  0 [0]   B3 [Penetrating], n [%]  2 [25]  Extraintestinal manifestations, n [%]  4 [50]  Previous intestinal resection, n [%]  2 [25]  TPMT activity     Normal [homozygote wild type], n [%]  8 [100]   Intermediate [heterozygote mutant], n [%]  0 [0]   Enzyme deficiency [homozygote mutant], n [%]  0 [0]  IFX regimen     5 mg/kg every 6 weeks, n [%]  4 [50]   5 mg/kg every 8 weeks, n [%]  4 [50]   Time on IFX, months median [IQR]  8 [6-32]   Previous episodic IFX, n [%]  1 [13]  Albumin mg/dL, median [IQR]  43 [41-44]  CRP mg/dL, median [IQR]  2 [0-18]  Haemoglobin mM, median [IQR]  8.5 [8.2-8.9]  White blood cells × 109/L, median [IQR]  8.2 [6.5-13.1]    Patients n = 8  Male sex, n [%]  3 [38]  Age, years median [IQR]  39 [30-45]  Disease duration, years median [IQR]  10 [5-24]  Age at diagnosis, years median [IQR]  26 [20-32]  Active smoking, n [%]  3 [38]  Body mass index [BMI], median [IQR]  2419 [25]  Montreal Classification    Age at onset     A1 [< 17years old], n [%]  0 [0]   A2 [17–40 years old], n [%]  5 [62]   A3 [> 40 years old], n [%]  3 [38]   Location     L1 [Ileal], n [%]  0 [0]   L2 [Colonic], n [%]  4 [50]   L3 [Ileocolonic], n [%]  3 [37]   L4 [Isolated upper disease], n [%]  1 [13]   Behaviour     B1 [Non-stricturing, non-penetrating], n [%]  6 [75]   B2 [Stricturing], n [%]  0 [0]   B3 [Penetrating], n [%]  2 [25]  Extraintestinal manifestations, n [%]  4 [50]  Previous intestinal resection, n [%]  2 [25]  TPMT activity     Normal [homozygote wild type], n [%]  8 [100]   Intermediate [heterozygote mutant], n [%]  0 [0]   Enzyme deficiency [homozygote mutant], n [%]  0 [0]  IFX regimen     5 mg/kg every 6 weeks, n [%]  4 [50]   5 mg/kg every 8 weeks, n [%]  4 [50]   Time on IFX, months median [IQR]  8 [6-32]   Previous episodic IFX, n [%]  1 [13]  Albumin mg/dL, median [IQR]  43 [41-44]  CRP mg/dL, median [IQR]  2 [0-18]  Haemoglobin mM, median [IQR]  8.5 [8.2-8.9]  White blood cells × 109/L, median [IQR]  8.2 [6.5-13.1]  IQR, interquartile range; IFX, infliximab; CRP, C-reactive protein; TPMT, thiopurine methyltransferase. View Large 3.2.2. Influence of infliximab on thiopurine metabolites As expected, circulating trough IFX levels increased substantially during the 12-week period of IFX intensification in the parent trial [ΔIFX median 6.5 μg/mL, IQR 1.9-9.8; p < 0.05].14,15 In contrast, thiopurine metabolite levels remained unchanged [6-TGN medians at Weeks 0, 4, 8, 12, and 20: 90 pmol/8 × 108 RBC, 93, 101, 90, and 80, respectively; 6-MeMP: 403, 559, 294, 860, and 291, respectively; p > 0.05] [Figure 4]. Figure 4. View largeDownload slide Thiopurine metabolites during IFX intensification. 6-thioguanine nucleotides [6-TGN]: [A] and methylated mercaptopurine metabolites [6-MeMP]; [B] across time in eight Crohn’s disease patients, all with infliximab [IFX] treatment failure at Week 0, and treated with an intensified IFX regimen until Week 12 resulting in resolution of symptoms. The patients were followed for 20 weeks, during which all received stable thiopurine treatment. Figure 4. View largeDownload slide Thiopurine metabolites during IFX intensification. 6-thioguanine nucleotides [6-TGN]: [A] and methylated mercaptopurine metabolites [6-MeMP]; [B] across time in eight Crohn’s disease patients, all with infliximab [IFX] treatment failure at Week 0, and treated with an intensified IFX regimen until Week 12 resulting in resolution of symptoms. The patients were followed for 20 weeks, during which all received stable thiopurine treatment. 4. Discussion In the first part of this study, we initially observed that significantly fewer patients receiving IFX-thiopurine combination treatment had detectable anti-IFX Abs as compared with those receiving monotherapy. This is consistent with previous pivotal randomised controlled trials.6–8 To further explore this finding, we measured thiopurine metabolite levels and found that 6-TGN levels were substantially lower in IFX Ab-positive than in IFX Ab-negative patients. This observation is congruent with another recently published observation.11 The association between 6-TGN and anti-IFX Abs in two independent cohort studies calls for studies to unravel the precise underlying mechanisms for this observation. We also examined whether circulating trough IFX levels themselves were directly associated with 6-TGN levels independently of anti-IFX Ab formation. In our study, IFX and thiopurine metabolite levels were measured at matched time points, but no such association was found. This observation contrasts with a previous reporting of a moderate positive correlation between IFX and 6-TGN levels.11 However, in that study the correlation analysis was carried out in unselected patients receiving thiopurines rather than in the more relevant anti-IFX Abs-negative subset, as we did. Hence, we believe that it is essential to exclude the potentially confounding effect of anti-IFX Abs to ensure that the correlation is in fact related to 6-TGN. Taken together, our findings are in keeping with the known superior clinical efficacy of combination therapy with IFX and thiopurines over monotherapies, and support therapeutic drug monitoring to optimise 6-TGN levels in order to minimise risk of anti-IFX Abs.6–8 As mentioned above, several studies now support the view that combination treatment with thiopurines and IFX results in lower risk of anti-IFX Abs than IFX monotherapy. However, only approximately 20% of patients receiving IFX actually develop anti-IFX Abs, indicating that other mechanisms may contribute to the overall additive effect of combination treatment.22 These mechanisms are not well understood, but our data support that 6-TGN does not directly associate with circulating IFX levels. Moreover, in the second part of the study we observed that the opposite mechanism did not appear operative, as 6-TGN levels remained stable in a setting of highly increased IFX exposure in an otherwise homogeneous population receiving steady thiopurine dosing. A previous study of 32 patients reported a temporary increase of 6-TGN levels 1-3 weeks after an IFX infusion [non-trough] in patients receiving steady azathioprine therapy, but with 6-TGN having dropped to pre-infusion levels at time of next scheduled infusion [trough].23 We believe that our study provides useful data on these interactions, as IFX levels and 6-TGN levels were examined in two different scenarios with either a steady [part 1] or an intensified [part 2] dose of IFX and showed no association between IFX levels and 6-TGN levels or vice versa. Taken together, it seems likely that the improved efficacy of the combination treatment, beyond reducing risk of anti-IFX Abs, reflects that thiopurines and IFX have different anti-inflammatory mechanisms of action and therefore act synergistically to reduce the inflammatory load more effectively when given in combination─rather than through direct drug interactions.24,25 The findings of the present study have clinical perspectives. Thiopurines are associated with a well-known panel of side effects such as nausea, vomiting, and flu-like symptoms as well as hepatotoxicity and bone marrow depression. These side effects may be reduced or eliminated by shifting to low-dose thiopurine-allopurinol combination treatment in patients who predominantly shunt the drug to 6-MeMP rather than 6-TGN, but this is not always feasible, and some patients tolerate only lower doses. For those also receiving IFX, it is essential to reduce immunogenicity, and the ROC analysis of our data shows that no patients with a 6-TGN level of ≥117 pmol/8 × 108 RBC had detectable anti-IFX Abs. Interestingly, this figure is very close to that reported by another group where patients with 6-TGN < 125 pmol/8 × 108 RBC had higher risk of developing anti-IFX Abs.11 It has been proposed that a 6-TGN threshold of 230-260 pmol/8 × 108 RBCs in patients on thiopurine monotherapy is best correlated to clinical remission.22,26 However, taking into account the present findings, an even lower threshold of ~120 pmol/8 × 108 RBC seems to be sufficient to reduce risk of anti-IFX Abs in patients receiving IFX-thiopurine combination therapy. This favours use of lower thiopurine doses during combination therpay─especially in patients intolerant to standard weight-based thiopurine dosing. Our study along with others shows that 6-MeMP does not influence anti-IFX Abs or circulating IFX, thus supporting use of allopurinol to shunt thiopurine metabolism in the direction of 6-TGN.9–11 The major limitations in cohort 1 are the retrospective observational design including inherent restrictions regarding causality, and that testing was neither completely standardised nor predefined, introducing risk of selection bias. As the objective was to explore potential drug interactions between thiopurine metabolites and IFX, treatment outcomes were not assessed and data from patients with Crohn’s disease and ulcerative colitis were pooled to yield power [cohort 1]. However, the majority of patients had surprisingly low 6-TGN levels, thus warranting caution when extrapolating the results. Even though we cannot rule out that lack of association between 6-TGN and IFX in cohort 1 was due confounders, the multiple regression analysis did not indicate that this was the case; and it was not apparent that patients receiving IFX-thiopurine combination therapy had a markedly more severe disease course than those receiving IFX monotherapy [Table 1]. The use in cohort 1 of different analytical techniques across time, due to changes in technology by the supplier company, is unlikely to have had major implications in IFX quantifications as the IFX assays have been shown to correlate well.16,17 Bias may, however, have been introduced regarding anti-IFX Ab detection, as assays are less congruent and have different drug sensitivity [RIA has moderate drug sensitivity;18,27 EIA is not prone to drug interference;16,18 RGA is a a cell-based assay which measures the functional activity of drug and anti-drug Abs at the cellular TNF receptor level, thus rendering it sensitive to IFX [> ~0.65 µg/mL] ─only samples without detectable IFX by RGA were assessed for anti-IFX Abs by RGA, whereas samples with detectable IFX were generally tested for anti-IFX Abs by EIA to accommodate for potential drug interference.17 As thiopurines [AZA and 6-MP] were handed out in our clinic to the patients, adherence is likely to be high although non-adherence in patients with low thiopurine metabolite levels cannot completely be ruled out. Although cohort 2 was very small, our observations add weight to lack of influence of IFX on thiopurine metabolite levels, because the cohort consisted of a highly homogeneous population receiving a steady thiopurine dosing with repeated standardised measurements of thiopurine metabolites and IFX and anti-IFX Abs across time, and with the only variable changed being the increased exposure of IFX. In conclusion, superior effectiveness of IFX-thiopurine combination therapy in part relates to decreased anti-IFX Abs. Importantly, this effect seems to be influenced by 6-TGN and, with a relatively low 6-TGN threshold of ≥ ~120 pmol/8 × 108 RBC, to substantially reduce or completely eliminate risk of anti-IFX Abs. Additional benefit from IFX-thiopurine combination therapy over monotherapies appears related to synergy between different anti-inflammatory modes of action rather than to direct interactions between thiopurine metabolites and IFX or vice versa. The results support the view that therapeutic drug monitoring can be implemented during thiopurine-IFX combination treatment to optimise 6-TGN levels and lower 6-MeMP levels, in order to improve efficacy and reduce risk of side effects. For those patients who only tolerate lower doses, a 6-TGN level of about 120 pmol/8 × 108 RBC as of now seems to be a sound target threshold during combination therapy. Funding This study was kindly supported by the Danish Colitis and Crohn’s Disease Society, Jacob Madsen and Olga Madsens Fund, and Herlev-Gentofte Hospital Research Council. Conflict of Interest Within the past year, CS has served as speaker for Takeda and as advisory board member for Janssen. JB has served as advisory board member for Abbvie, Janssen, and Takeda; and as primary investigator for Abbvie and Janssen. Author Contributions Study design: DVM, JB, CS. Collection of data: DVM, CS. Analysis and interpretation of data: DVM, JB, CS. Drafting the manuscript: DVM, JB, CS. Revising the manuscript and approval of final manuscript: all authors. References 1. Lichtenstein GR, Hanauer SB, Sandborn WJ; Practice Parameters Committee of American College of Gastroenterology. Management of Crohn’s disease in adults. Am J Gastroenterol  2009; 104: 465– 83; quiz 464, 484. Google Scholar CrossRef Search ADS PubMed  2. Gisbert JP, Panés J. Loss of response and requirement of infliximab dose intensification in Crohn’s disease: a review. Am J Gastroenterol  2009; 104: 760– 7. Google Scholar CrossRef Search ADS PubMed  3. Steenholdt C, Bendtzen K, Brynskov J, Ainsworth MA. Optimizing treatment with TNF inhibitors in inflammatory bowel disease by monitoring drug levels and antidrug antibodies. Inflamm Bowel Dis  2016; 22: 1999– 2015. Google Scholar CrossRef Search ADS PubMed  4. Steenholdt C, Svenson M, Bendtzen K, Thomsen OØ, Brynskov J, Ainsworth MA. Severe infusion reactions to infliximab: aetiology, immunogenicity and risk factors in patients with inflammatory bowel disease. Aliment Pharmacol Ther  2011; 34: 51– 8. Google Scholar CrossRef Search ADS PubMed  5. Frederiksen MT, Ainsworth MA, Brynskov J, Thomsen OO, Bendtzen K, Steenholdt C. Antibodies against infliximab are associated with de novo development of antibodies to adalimumab and therapeutic failure in infliximab-to-adalimumab switchers with IBD. Inflamm Bowel Dis  2014; 20: 1714– 21. Google Scholar CrossRef Search ADS PubMed  6. Colombel JF, Sandborn WJ, Reinisch Wet al.  ; SONIC Study Group. Infliximab, azathioprine, or combination therapy for Crohn’s disease. N Engl J Med  2010; 362: 1383– 95. Google Scholar CrossRef Search ADS PubMed  7. Colombel JF, Reinisch W, Mantzaris GJet al.   Randomised clinical trial: deep remission in biologic and immunomodulator naïve patients with Crohn’s disease - a SONIC post hoc analysis. Aliment Pharmacol Ther  2015; 41: 734– 46. Google Scholar CrossRef Search ADS PubMed  8. Panaccione R, Ghosh S, Middleton Set al.  . Combination therapy with infliximab and azathioprine is superior to monotherapy with either agent in ulcerative colitis. Gastroenterology  2014; 146: 392– 400.e3. Google Scholar CrossRef Search ADS PubMed  9. Coskun M, Steenholdt C, de Boer NK, Nielsen OH. Pharmacology and optimization of thiopurines and methotrexate in inflammatory bowel disease. Clin Pharmacokinet  2016; 55: 257– 74. Google Scholar CrossRef Search ADS PubMed  10. Moon W, Loftus EVJr. Review article: recent advances in pharmacogenetics and pharmacokinetics for safe and effective thiopurine therapy in inflammatory bowel disease. Aliment Pharmacol Ther  2016; 43: 863– 83. Google Scholar CrossRef Search ADS PubMed  11. Yarur AJ, Kubiliun MJ, Czul Fet al.  . Concentrations of 6-thioguanine nucleotide correlate with trough levels of infliximab in patients with inflammatory bowel disease on combination therapy. Clin Gastroenterol Hepatol  2015; 13: 1118– 24.e3. Google Scholar CrossRef Search ADS PubMed  12. Bar-Yoseph H, Waterman M, Almog Ret al.  . Prevention of antidrug antibody formation to infliximab in Crohn’s patients with prior failure of thiopurines. Clin Gastroenterol Hepatol  2017; 15: 69– 75. Google Scholar CrossRef Search ADS PubMed  13. Pozler O, Chládek J, Malý Jet al.  . Steady-state of azathioprine during initiation treatment of pediatric inflammatory bowel disease. J Crohns Colitis  2010; 4: 623– 8. Google Scholar CrossRef Search ADS PubMed  14. Steenholdt C, Brynskov J, Thomsen OØet al.  . Individualised therapy is more cost-effective than dose intensification in patients with Crohn’s disease who lose response to anti-TNF treatment: a randomised, controlled trial. Gut  2014; 63: 919– 27. Google Scholar CrossRef Search ADS PubMed  15. Steenholdt C, Brynskov J, Thomsen OØet al.  . individualized therapy is a long-term cost-effective method compared with dose intensification in Crohn’s Disease patients failing infliximab. Dig Dis Sci  2015; 60: 2762– 70. Google Scholar CrossRef Search ADS PubMed  16. Steenholdt C, Ainsworth MA, Tovey Met al.  . Comparison of techniques for monitoring infliximab and antibodies against infliximab in Crohn’s disease. Ther Drug Monit  2013; 35: 530– 8. Google Scholar CrossRef Search ADS PubMed  17. Steenholdt C, Bendtzen K, Brynskov J, Thomsen OØ, Ainsworth MA. Clinical implications of measuring drug and anti-drug antibodies by different assays when optimizing infliximab treatment failure in Crohn’s disease: post hoc analysis of a randomized controlled trial. Am J Gastroenterol  2014; 109: 1055– 64. Google Scholar CrossRef Search ADS PubMed  18. Steenholdt C, Frederiksen MT, Bendtzen K, Ainsworth MA, Thomsen OØ, Brynskov J. Time course and clinical implications of development of antibodies against adalimumab in patients with inflammatory bowel disease. J Clin Gastroenterol  2016; 50: 483– 9. Google Scholar CrossRef Search ADS PubMed  19. Dervieux T, Boulieu R. Simultaneous determination of 6-thioguanine and methyl 6-mercaptopurine nucleotides of azathioprine in red blood cells by HPLC. Clin Chem  1998; 44: 551– 5. Google Scholar PubMed  20. Dervieux T, Boulieu R. Identification of 6-methylmercaptopurine derivative formed during acid hydrolysis of thiopurine nucleotides in erythrocytes, using liquid chromatography-mass spectrometry, infrared spectroscopy, and nuclear magnetic resonance assay. Clin Chem  1998; 44: 2511– 5. Google Scholar PubMed  21. Lennard L. Commentary on: Differences in nucleotide hydrolysis contribute to the differences between erythrocyte 6-thioguanine nucleotide concentrations determined by two widely used methods. Clin Chem  2003; 49: 1551; author reply 1551–2. Google Scholar CrossRef Search ADS PubMed  22. Vande Casteele N, Herfarth H, Katz J, Falck-Ytter Y, Singh S. American Gastroenterological Association Institute Technical Review on the Role of Therapeutic Drug Monitoring in the Management of Inflammatory Bowel Diseases. Gastroenterology  2017; 153: 835– 57.e6. Google Scholar CrossRef Search ADS PubMed  23. Roblin X, Serre-Debeauvais F, Phelip JM, Bessard G, Bonaz B. Drug interaction between infliximab and azathioprine in patients with Crohn’s disease. Aliment Pharmacol Ther  2003; 18: 917– 25. Google Scholar CrossRef Search ADS PubMed  24. Olesen CM, Coskun M, Peyrin-Biroulet L, Nielsen OH. Mechanisms behind efficacy of tumor necrosis factor inhibitors in inflammatory bowel diseases. Pharmacol Ther  2016; 159: 110– 9. Google Scholar CrossRef Search ADS PubMed  25. Tiede I, Fritz G, Strand Set al.  . CD28-dependent Rac1 activation is the molecular target of azathioprine in primary human CD4+ T lymphocytes. J Clin Invest  2003; 111: 1133– 45. Google Scholar CrossRef Search ADS PubMed  26. Moreau AC, Paul S, Del Tedesco Eet al.  . Association between 6-thioguanine nucleotides levels and clinical remission in inflammatory disease: a meta-analysis. Inflamm Bowel Dis  2014; 20: 464– 71. Google Scholar CrossRef Search ADS PubMed  27. Svenson M, Geborek P, Saxne T, Bendtzen K. Monitoring patients treated with anti-TNF-alpha biopharmaceuticals: assessing serum infliximab and anti-infliximab antibodies. Rheumatology [Oxford]  2007; 46: 1828– 34. Google Scholar CrossRef Search ADS PubMed  Copyright © 2017 European Crohn’s and Colitis Organisation (ECCO). Published by Oxford University Press. All rights reserved. For permissions, please email: journals.permissions@oup.com http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Crohn's and Colitis Oxford University Press

A Role for Thiopurine Metabolites in the Synergism Between Thiopurines and Infliximab in Inflammatory Bowel Disease

Loading next page...
 
/lp/ou_press/a-role-for-thiopurine-metabolites-in-the-synergism-between-thiopurines-11XHJouGDf
Publisher
Oxford University Press
Copyright
Copyright © 2017 European Crohn’s and Colitis Organisation (ECCO). Published by Oxford University Press. All rights reserved. For permissions, please email: journals.permissions@oup.com
ISSN
1873-9946
eISSN
1876-4479
D.O.I.
10.1093/ecco-jcc/jjx149
Publisher site
See Article on Publisher Site

Abstract

Abstract Background Interactions between principal cytotoxic thiopurine metabolites, that is 6-thioguanine nucleotides [6-TGN], and infliximab [IFX] and anti-IFX antibodies [Abs] may contribute to higher effectiveness of IFX-thiopurine combination therapy than monotherapies in inflammatory bowel disease. Methods To examine if thiopurine metabolites influenced trough IFX and anti-IFX Abs, 89 patients previously assessed for anti-IFX Abs were included. To assess if IFX influenced thiopurine metabolites, eight patients who had responded to 12 weeks of intensified IFX at a constant thiopurine dosing were included. Results In the first cohort, IFX-thiopurine combination therapy reduced anti-IFX Ab detection [8/40; 20%] as compared with IFX monotherapy [22/49; 45%], odds ratio [OR] 0.31 [0.12–0.80], p < 0.05. 6-TGN was significantly lower in anti-IFX Ab-positive patients (50 pmol/8 × 108 red blood cells [RBC] vs 105, p < 0.01). All anti-IFX Ab-positive patients had 6-TGN < 117 pmol/8 × 108 RBC (sensitivity 100% [63–100], specificity 47% [29–65], area under the curveROC = 0.82, p < 0.01). Trough IFX was similar between anti-IFX Ab-negative patients in IFX monotherapy and IFX-thiopurine combination therapy [5.1 μg/mL vs 4.9, p = 0.76]. 6-TGN and IFX did not correlate [rP = 0.04, p = 0.83; rS = 0.02, p = 0.89, respectively]. In the second cohort, trough IFX increased during IFX intensification [ΔIFX median 6.5 μg/mL, p = 0.02], but 6-TGN was stable [6-TGN at Weeks 0, 4, 8, 12: 90 pmol/8 × 108 RBC, 93, 101, 90; p > 0.05]. Methylated mercaptopurine metabolite associations were consistently negative. Conclusions Superior effect of IFX-thiopurine combination therapy over monotherapies partly relates to decrease in anti-IFX Abs, which associates with 6-TGN levels and has a lower therapeutic threshold than during thiopurine monotherapy. Additional benefit likely ascribes to synergy between different anti-inflammatory modes of action rather than direct drug interactions. Inflammatory bowel disease, infliximab, antibodies against infliximab, therapeutic drug monitoring, thiopurines, 6-TGN 1. Introduction Patients with inflammatory bowel disease [IBD] usually need anti-inflammatory treatment, and tumour necrosis factor [TNF] inhibitors such as infliximab [IFX] are a cornerstone in the therapeutic armentarium.1 Unfortunately, up to half of patients lose the effect of IFX maintenance therapy.2 This can be caused by formation of antibodies against IFX [anti-IFX Abs] leading to decreased or absent circulating levels of active drug.3 Continued IFX therapy in the presence of anti-IFX Abs can also result in severe hypersensitivity reactions.4 Patients who develop anti-IFX Abs are prone to later develop anti-adalimumab Abs, leading to repeat loss of response.5 Thus, it is pertinent to find ways to minimise the risk of anti-IFX Ab formation. Previous studies have convincingly shown that combination therapy with IFX and thiopurines is more effective than monotherapy with either agent in immunosuppressor-naïve IBD patients.6–8 Whereas TNF inhibitors are large-molecular weight monoclonal antibodies that target the pro-inflammatory cytokines, TNF-α, thiopurines (i.e. 6-mercaptopurine [6-MP] and its pro-drug, azathioprine [AZA]), are small-molecular drugs that undergo complex enzymatic metabolism with formation of both therapeutic active and inactive metabolites.9 The primary immunosuppressive metabolites are 6-thioguanine nucleotides [6-TGN], which are incorporated into DNA and RNA and thereby induce apoptosis in, for example, lymphocytes.10 The mechanism for improved effectiveness of IFX-thiopurine combination treatment over monotherapies has been hypothesised to originate from thiopurines’ decreasing immunogenicity of IFX or, alternatively, that thiopurines directly influence circulating IFX levels.11 It has relatively consistently been observed that fewer patients receiving combination therapy have detectable anti-IFX Abs6,8,11,12─and it was recently reported that 6-TGN associated with anti-IFX Abs.11 The primary aim of this study was to investigate the interplay between principal cytotoxic thiopurine metabolites (6-TGN and methylated mercaptopurine metabolites [6-MeMP]) and IFX and anti-IFX Abs. Furthermore, we aimed to to establish thresholds for thiopurine metabolites that associate with improved pharmacokinetics of IFX or less likelihood of anti-IFX Abs. 2. Methods 2.1. Study design 2.1.1. Influence of thiopurine metabolites on infliximab and anti-infliximab antibodies The first part was a retrospective cohort study that examined if thiopurine metabolites associate with trough IFX levels and anti-IFX Abs. The study population comprised all patients treated with IFX at our tertiary IBD centre since 2009, in whom IFX and anti-IFX Abs had been assessed [at the discretion of the treating physician until 2012 and hereafter at predefined time points according to local guidelines: after 1 year of treatment and then at every 6 months─or at any time of suspected treatment failure]. The two principal cytotoxic cytosolic thiopurine metabolites, 6-TGN and 6-MeMP, were determined in red blood cells [RBC] in stored samples at time points corresponding exactly to those when IFX and anti-IFX Abs had been measured. We included patients who had received IFX induction treatment and at least one maintenance infusion. Patients who received concomitant thiopurines [AZA or 6-MP] were included only if they had received a stable dose for at least 2 months before sampling, in order to allow for thiopurine metabolites to achieve steady-state erythrocyte concentrations.13 Patients were excluded if they received systemic steroid treatment > 5 mg/day, allopurinol, methotrexate, or ciclosporin. This study was approved by the Danish Health and Medical Authority [H-15018092] and the Danish Data Protection Agency [HGH-2016–019:04434]. 2.1.2. Influence of infliximab on thiopurine metabolites The second part of the study was a post-hoc analysis of data from a randomised controlled trial, and examined if IFX influences the concentration of thiopurine metabolites. The study population comprised eight Crohn’s disease patients with active luminal disease despite standard IFX maintenance therapy. These patients had received an intensified IFX regimen comprising 5 mg/kg every 4 weeks for a total of 12 weeks, during which all eight patients had a clinical response (Crohn’s Disease Activity Index [CDAI] decrease of ≥70).14,15 From Week 12 until end of follow-up at Week 20, two patients had received IFX every 4 weeks, two had IFX every 6 weeks, two had IFX every 8 weeks, one had discontinued IFX, and one was lost to follow-up. All patients received steady concomitant thiopurine treatment with either AZA [n = 6] or 6-MP [n = 2] throughout all 20 study weeks. Trough IFX had been measured at Weeks 0, 4, 8, 12, and 20, and were supplemented with measurements of 6-TGN and 6-MeMP in stored samples obtained at the same time points. This study was approved by the Danish Medicines Agency [EudraCT 2009-009926-94], the regional ethics committees [HA-2009-009], and the Danish Data Protection Agency [2007-58-0015; 750.89-2]. 2.2. Analyses 2.2.1. Infliximab and anti-infliximab antibodies Circulating trough IFX concentrations were measured in serum by radioimmunoassay [RIA] (limit of detection [LOD] 0.07 μg/mL)] until end of 2012, after which it was routinely replaced by reporter gene assay [RGA; iLiteTM] [LOD 0.13 μg/mL].16,17 Samples were classified as anti-IFX Ab-positive or -negative according to the LOD of the assay, and measured by RIA until mid-2014 and then by enzyme immunoassay [EIA] or RGA [iLiteTM].16–18 All analyses were done by a commercial provider under blinded conditions [Euro Diagnostica, SE; formerly Biomonitor A/S, DK]. In the event that a patient had both a positive and a negative anti-IFX Ab test at different time points, the positive sample and the corresponding IFX trough level were used. The most recent sample was used if more than one negative anti-IFX Ab test was available. 2.2.2. Thiopurine metabolites RBC concentrations of 6-TGN and 6-MeMP metabolites were determined in biobanked buffy-coat samples [stored at -80°C] which had all been collected exactly at the day of IFX and anti-IFX Ab measurements. Quantification was carried out by hot acid hydrolysis followed by ultra-performance liquid chromatography, as previously described.19–21 Control blood spiked with known amounts of thioguanine and methylmercaptopurine monophosphate nucleotides [Jena Bioscience, DE] was used for calibrators. Thiopurine metabolite levels were normalised to buffy-coat haemoglobin [Hgb] and measured in nmol/mmol Hgb followed by conversion to pmol/8 x 108 RBC [Abacus 380 blood analyser, HU]. 2.2.3. Thiopurine methyltransferase phenotype Thiopurine methyltransferase [TPMT] enzyme activity was assessed in hypotonic buffy-coat lysates by chromatography as described above, except that a BEH C18 column and a water/formic acid-methanol buffer system were used for separation. Calibrators were 6-MeMP spiked into blank matrix. Haemoglobin levels were used for calculating TPMT activity units [1U = 1 nmol 6-MeMP x h-1 x mL packed erythrocytes-1] [Abacus 380 blood analyser]. 2.3. Statistics Descriptive data were presented as percentages for discrete variables, and as median with interquartile range [IQR] for continuous variables. Correlations were investigated using linear correlation analysis expressed as Pearson’s correlation coefficient [rp], and by non- parametric correlation analysis expressed as Spearman’s correlation coefficient [rs]. Univariate analyses of discrete variables were done by Fisher’s exact test or chi square test, as appropriate. Continuous variables were compared by Mann-Whitney test [unpaired] or Wilcoxon signed-rank test [paired]. Therapeutic thresholds for thiopurine metabolites were determined by receiving operator curve [ROC] analysis. Multivariate logistic [anti-IFX Abs] or linear [IFX] regression was done by entering the following variables into the models: gender, disease type, body mass index [BMI], frequency of IFX infusions, 6-TGN, 6-MeMP. Values below LOD were set to null. Missing data were excluded. Analyses were done in SPSS version 24 [IBM, NY, USA] or in GraphPad Prism version 5 [GraphPad Software, CA, USA]. Two-sided p-values < 0.05 were significant. 3. Results 3.1. Influence of thiopurine metabolites on anti-IFX Abs and IFX 3.1.1. Study population The study population comprised 89 patients, of whom 40 patients [45%] received IFX-thiopurine combination therapy [AZA 88%], and 49 patients [55%] had IFX monotherapy [Table 1]. The majority of patients had Crohn’s disease [71%] with a moderate disease duration [median 7.5 years], and had received long-term IFX therapy at time of study [median 15 months]. Table 1. Characteristics of cohort 1   All patients n = 89  IFX monotherapy n = 49  IFX-thiopurine combination therapy n = 40  p-Value  Male sex, n [%]  46 [52]  28 [57]  18 [45]  0.29  Age, years median [IQR]  40 [29-50]  41 [31-50]  38 [28-48]  0.69  Crohn’s disease, n [%]  63 [71]  35 [71]  28 [70]  1.00  Disease duration, years median [IQR]  8 [3-15]  8 [3-14]  8 [4-15]  0.95  Age at diagnosis, years median [IQR]  26 [20-35]  26 [20-39]  26 [20-33.25]  0.76  Active smoking, n [%]  16 [18]  9 [18]  7 [18]  1.00  Body mass index [BMI], median [IQR]  25 [22-28]  25 [23-27]  24 [22-28]  0.44  Montreal classification [CD]  Age at onset        < 0.01   A1 [< 17 years old], n [%]  6 [10]  6 [17]  0 [0]     A2 [1740 years old], n [%]  46 [73]  20 [57]  26 [93]     A3 [> 40 years old], n [%]  11 [17]  9 [26]  2 [7]    Location        0.80   L1 [Ileal], n [%]  5 [8]  3 [9]  2 [7]     L2 [Colonic], n [%]  37 [59]  22 [63]  15 [54]     L3 [Ileocolonic], n [%]  21 [33]  10 [29]  11 [39]     L4 [Isolated upper disease], n [%]  0 [0]  0 [0]  0 [0]    Behaviour        0.06   B1 [Non-stricturing, non-penetrating], n [%]  30 [48]  21 [60]  9 [32]     B2 [Stricturing], n [%]  16 [25]  8 [23]  8 [29]     B3 [Penetrating], n [%]  17 [27]  6 [17]  11 [39]    Perianal, n [%]  20 [32]  7 [20]  13 [46]  < 0.05  Montreal Classification [UC]  Extent        0.88   E1 [Ulcerative proctitis], n [%]  2 [8]  1 [7]  1 [8]     E2 (Left sided [distal]), n [%]  14 [54]  7 [50]  7 [58]     E3 (Extensive [pancolitis]), n [%]  10 [38]  6 [43]  4 [33]    Extraintestinal manifestations, n [%]  19 [21]  12 [24]  7 [18]  0.42  Previous intestinal resection, n [%]  27 [30]  13 [27]  14 [35]  0.39  Concomitant thiopurine   Azathioprine, n [%]  -  -  35 [88]  -   Azathioprine dose, mg/kg median [IQR]  -  -  1.7 [1.5-2.0]  -   Mercaptopurine dose, mg/kg median [IQR]  -  -  0.8 [0.7-1.0]  -   Time on thiopurines, months median [IQR]  -  -  21 [12-40]  -  TPMT activity   Normal [homozygote wild type], n [%]  -  -  36 [90]  -   Intermediate [heterozygote mutant], n [%]  -  -  4 [10]  -   Enzyme deficiency [homozygote mutant], n [%]  -  -  0 [0]  -   Concomitant aminosalicylates [in colitis], n [%]  5 [19]  0 [0]  5 [42]  < 0.05  IFX regimen        0.84   5 mg/kg, q < 8 weeks, n [%]  35 [39]  18 [37]  17 [43]     5 mg/kg, q = 8 weeks, n [%]  43 [48]  25 [51]  18 [45]     5 mg/kg, q > 8 weeks, n [%]  11 [12]  6 [12]  5 [13]     Time on IFX, months median [IQR]  15 [11-27]  13 [8-38]  17 [12-25]  0.97   Previous episodic IFX, n [%]  36 [40]  19 [39]  17 [43]  0.83  Albumin mg/dL, median [IQR]  43 [40-45]  43 [40-45]  43 [40-45]  0.50  CRP mg/dL, median [IQR]  3 [3-3]  3 [3-5]  3 [3-3]  0.10  Haemoglobin mM, median [IQR]  8.6 [8.0-9.2]  8.8 [8.3-9.2]  8.4 [7.7-9.1]  < 0.05  White blood cells × 09/L, median [IQR]  6.6 [5.1-8.0]  6.8 [5.8-8.0]  6.1 [4.7-8.1]  0.16    All patients n = 89  IFX monotherapy n = 49  IFX-thiopurine combination therapy n = 40  p-Value  Male sex, n [%]  46 [52]  28 [57]  18 [45]  0.29  Age, years median [IQR]  40 [29-50]  41 [31-50]  38 [28-48]  0.69  Crohn’s disease, n [%]  63 [71]  35 [71]  28 [70]  1.00  Disease duration, years median [IQR]  8 [3-15]  8 [3-14]  8 [4-15]  0.95  Age at diagnosis, years median [IQR]  26 [20-35]  26 [20-39]  26 [20-33.25]  0.76  Active smoking, n [%]  16 [18]  9 [18]  7 [18]  1.00  Body mass index [BMI], median [IQR]  25 [22-28]  25 [23-27]  24 [22-28]  0.44  Montreal classification [CD]  Age at onset        < 0.01   A1 [< 17 years old], n [%]  6 [10]  6 [17]  0 [0]     A2 [1740 years old], n [%]  46 [73]  20 [57]  26 [93]     A3 [> 40 years old], n [%]  11 [17]  9 [26]  2 [7]    Location        0.80   L1 [Ileal], n [%]  5 [8]  3 [9]  2 [7]     L2 [Colonic], n [%]  37 [59]  22 [63]  15 [54]     L3 [Ileocolonic], n [%]  21 [33]  10 [29]  11 [39]     L4 [Isolated upper disease], n [%]  0 [0]  0 [0]  0 [0]    Behaviour        0.06   B1 [Non-stricturing, non-penetrating], n [%]  30 [48]  21 [60]  9 [32]     B2 [Stricturing], n [%]  16 [25]  8 [23]  8 [29]     B3 [Penetrating], n [%]  17 [27]  6 [17]  11 [39]    Perianal, n [%]  20 [32]  7 [20]  13 [46]  < 0.05  Montreal Classification [UC]  Extent        0.88   E1 [Ulcerative proctitis], n [%]  2 [8]  1 [7]  1 [8]     E2 (Left sided [distal]), n [%]  14 [54]  7 [50]  7 [58]     E3 (Extensive [pancolitis]), n [%]  10 [38]  6 [43]  4 [33]    Extraintestinal manifestations, n [%]  19 [21]  12 [24]  7 [18]  0.42  Previous intestinal resection, n [%]  27 [30]  13 [27]  14 [35]  0.39  Concomitant thiopurine   Azathioprine, n [%]  -  -  35 [88]  -   Azathioprine dose, mg/kg median [IQR]  -  -  1.7 [1.5-2.0]  -   Mercaptopurine dose, mg/kg median [IQR]  -  -  0.8 [0.7-1.0]  -   Time on thiopurines, months median [IQR]  -  -  21 [12-40]  -  TPMT activity   Normal [homozygote wild type], n [%]  -  -  36 [90]  -   Intermediate [heterozygote mutant], n [%]  -  -  4 [10]  -   Enzyme deficiency [homozygote mutant], n [%]  -  -  0 [0]  -   Concomitant aminosalicylates [in colitis], n [%]  5 [19]  0 [0]  5 [42]  < 0.05  IFX regimen        0.84   5 mg/kg, q < 8 weeks, n [%]  35 [39]  18 [37]  17 [43]     5 mg/kg, q = 8 weeks, n [%]  43 [48]  25 [51]  18 [45]     5 mg/kg, q > 8 weeks, n [%]  11 [12]  6 [12]  5 [13]     Time on IFX, months median [IQR]  15 [11-27]  13 [8-38]  17 [12-25]  0.97   Previous episodic IFX, n [%]  36 [40]  19 [39]  17 [43]  0.83  Albumin mg/dL, median [IQR]  43 [40-45]  43 [40-45]  43 [40-45]  0.50  CRP mg/dL, median [IQR]  3 [3-3]  3 [3-5]  3 [3-3]  0.10  Haemoglobin mM, median [IQR]  8.6 [8.0-9.2]  8.8 [8.3-9.2]  8.4 [7.7-9.1]  < 0.05  White blood cells × 09/L, median [IQR]  6.6 [5.1-8.0]  6.8 [5.8-8.0]  6.1 [4.7-8.1]  0.16  IQR, interquartile range; CD, Crohn’s disease; UC, ulcerative colitis; IFX, infliximab; q, frequency of dosing; CRP, C-reactive protein; TPMT, thiopurine methyltransferase. View Large 3.1.2. Antibodies against infliximab IFX-thiopurine combination therapy significantly reduced detection of anti-IFX Abs (8/40 [20%]) compared with IFX monotherapy (22/49 [45%]), odds ratio [OR] 0.31 [0.12-0.80], p < 0.05. Anti-IFX Abs appeared functionally active, since their presence resulted in undetectable trough IFX [median < LOD, IQR < LOD] as opposed to IFX levels of median 4.9 μg/mL, IQR 2.6-7.6 in anti-IFX Ab-negative patients [Figure 1]. Figure 1. View largeDownload slide Antibodies against infliximab. Infliximab [IFX] trough levels in patients’ positive [+] or negative [-] for anti-IFX antibodies [Abs]. Figure 1. View largeDownload slide Antibodies against infliximab. Infliximab [IFX] trough levels in patients’ positive [+] or negative [-] for anti-IFX antibodies [Abs]. Levels of 6-TGN were significantly lower in anti-IFX Abs-positive patients receiving IFX-thiopurine combination therapy [median 50 pmol/8 × 108 RBC, IQR 11-66] compared with those without detectable anti-IFX Abs [median 105 pmol/8 × 108 RBC, IQR 70-158, p < 0.01] [Figure 2A]. This association persisted in both univariate and multivariate logistic regression models [Table 2]. ROC analysis also revealed a significant difference in 6-TGN between anti-IFX Ab-positive and -negative patients (area under the curve [AUC] = 0.82 [0.68-0.97], p < 0.01). Of note, all patients with detectable anti-IFX Abs had 6-TGN < 117 pmol/8 × 108 RBC (sensitivity 100 [63-100], specificity 47 [29-65]). There was no difference in 6-MeMP between anti-IFX Ab-positive and -negative patients [median 150 pmol/8 × 108 RBC, IQR 35-1000 vs median 214, IQR 86-569, p > 0.05] [Figure 2B]. Table 2. Logistic regression of association between antibodies against infliximab and multiple study variables Variable  Univariate logistic regression coefficient, [SE]  p-Value  Multivariate regression coefficient, [SE]  p-Value  Male  -0.303  [0.450]  0.50  -1.569  [1.445]  0.28  Crohn’s disease  -0.295  [0.486]  0.54  -1.468  [1.815]  0.42  Body mass index [BMI]  0.002  [0.059]  0.97  -0.026  [0.113]  0.82  IFX frequencya  -  -  0.77  -  -  0.84   = q8  -0.257  [0.482]  0.59  -0.893  [1.526]  0.56   > q8  0.201  [0.769]  0.79  20.775  [14443]  1.00  6-TGN, median [IQR]  0.021  [0.009]  0.02  0.066  [0.030]  0.03  6-MeMP, median [IQR]  0.000  [0.001]  0.75  -0.003  [0.002]  0.12  Variable  Univariate logistic regression coefficient, [SE]  p-Value  Multivariate regression coefficient, [SE]  p-Value  Male  -0.303  [0.450]  0.50  -1.569  [1.445]  0.28  Crohn’s disease  -0.295  [0.486]  0.54  -1.468  [1.815]  0.42  Body mass index [BMI]  0.002  [0.059]  0.97  -0.026  [0.113]  0.82  IFX frequencya  -  -  0.77  -  -  0.84   = q8  -0.257  [0.482]  0.59  -0.893  [1.526]  0.56   > q8  0.201  [0.769]  0.79  20.775  [14443]  1.00  6-TGN, median [IQR]  0.021  [0.009]  0.02  0.066  [0.030]  0.03  6-MeMP, median [IQR]  0.000  [0.001]  0.75  -0.003  [0.002]  0.12  6-MeMP, methylated mercaptopurine metabolites; 6-TGN, 6-thioguanine nucleotides; IFX, infliximab; q, frequency of dosing; IQR, interquartile range; SE, standard error. a < q8 reference. View Large Figure 2. View largeDownload slide Thiopurine metabolites and antibodies against infliximab. 6-thioguanine nucleotides [6-TGN]: [A] and methylated mercaptopurine metabolites [6-MeMP]; [B] in patients positive [+] or negative [-] for anti-IFX antibodies [Abs]. Figure 2. View largeDownload slide Thiopurine metabolites and antibodies against infliximab. 6-thioguanine nucleotides [6-TGN]: [A] and methylated mercaptopurine metabolites [6-MeMP]; [B] in patients positive [+] or negative [-] for anti-IFX antibodies [Abs]. 3.1.3. Infliximab IFX trough levels were similar in anti-IFX Ab-negative patients on IFX monotherapy [median 5.1 μg/mL, IQR 2.4-8.6] compared with those on IFX-thiopurine combination therapy [median 4.9 μg/mL, IQR 3.1-6.5, p > 0.05]. There was no correlation between 6-TGN and IFX [rP = 0.04, p = 0.83; rS = 0.02, p = 0.89; respectively] [Figure 3A], or 6-MeMP and IFX [rS = -0.15, p = 0.43; rP = 0.05, p = 0.81; respectively] [Figure 3B]. Lack of associations between IFX and multiple study variables, including thipurine metabolites, were also found in both univariate and multiple regression models [Table 3]. The frequency of IFX infusions was, however, expectedly associated with IFX trough levels. Table 3. Multiple regression of association between infliximab and multiple study variables Variable  Univariate regression coefficient, [SE]  p-Value  Multivariate regression coefficient, [SE]  p-Value  Male  -0.357  [0.969]  0.71  -0.846  [1.660]  0.61  Crohn’s disease  0.534  [1.057]  0.62  1.672  [1.766]  0.35  Body mass index [BMI]  0.050  [0.120]  0.68  0.062  [0.178]  0.73  IFX frequencya  -2.402  [0.667]  0.001  -1.401  [1.123]  0.22  6-TGN, median [IQR]  0.012  [0.009]  0.20  0.018  [0.10]  0.09  6-MeMP, median [IQR]  -0.001  [0.001]  0.41  -0.002  [0.002]  0.20  Variable  Univariate regression coefficient, [SE]  p-Value  Multivariate regression coefficient, [SE]  p-Value  Male  -0.357  [0.969]  0.71  -0.846  [1.660]  0.61  Crohn’s disease  0.534  [1.057]  0.62  1.672  [1.766]  0.35  Body mass index [BMI]  0.050  [0.120]  0.68  0.062  [0.178]  0.73  IFX frequencya  -2.402  [0.667]  0.001  -1.401  [1.123]  0.22  6-TGN, median [IQR]  0.012  [0.009]  0.20  0.018  [0.10]  0.09  6-MeMP, median [IQR]  -0.001  [0.001]  0.41  -0.002  [0.002]  0.20  6-MeMP, methylated mercaptopurine metabolites; 6-TGN, 6-thioguanine nucleotides; IFX, infliximab; IQR, interquartile range; SE, standard error. a < q8 or q = 8 or q > 8. View Large Figure 3. View largeDownload slide Thiopurine metabolites and infliximab. Corresponding levels of infliximab [IFX] and 6-thioguanine nucleotides [6-TGN]: [A] and methylated mercaptopurine metabolites [6-MeMP]; [B] in patients negative for anti-IFX antibodies [Abs]. Figure 3. View largeDownload slide Thiopurine metabolites and infliximab. Corresponding levels of infliximab [IFX] and 6-thioguanine nucleotides [6-TGN]: [A] and methylated mercaptopurine metabolites [6-MeMP]; [B] in patients negative for anti-IFX antibodies [Abs]. 3.2. Influence of IFX on thiopurine metabolites 3.2.1. Study population The study population comprised eight patients with Crohn’s disease who all had manifest treatment failure of IFX maintenance therapy at time of inclusion, with response to 12 weeks of intensified IFX therapy at a constant thiopurine dosing throughout 20 weeks [Table 4]. Table 4. Characteristics of cohort 2.   Patients n = 8  Male sex, n [%]  3 [38]  Age, years median [IQR]  39 [30-45]  Disease duration, years median [IQR]  10 [5-24]  Age at diagnosis, years median [IQR]  26 [20-32]  Active smoking, n [%]  3 [38]  Body mass index [BMI], median [IQR]  2419 [25]  Montreal Classification    Age at onset     A1 [< 17years old], n [%]  0 [0]   A2 [17–40 years old], n [%]  5 [62]   A3 [> 40 years old], n [%]  3 [38]   Location     L1 [Ileal], n [%]  0 [0]   L2 [Colonic], n [%]  4 [50]   L3 [Ileocolonic], n [%]  3 [37]   L4 [Isolated upper disease], n [%]  1 [13]   Behaviour     B1 [Non-stricturing, non-penetrating], n [%]  6 [75]   B2 [Stricturing], n [%]  0 [0]   B3 [Penetrating], n [%]  2 [25]  Extraintestinal manifestations, n [%]  4 [50]  Previous intestinal resection, n [%]  2 [25]  TPMT activity     Normal [homozygote wild type], n [%]  8 [100]   Intermediate [heterozygote mutant], n [%]  0 [0]   Enzyme deficiency [homozygote mutant], n [%]  0 [0]  IFX regimen     5 mg/kg every 6 weeks, n [%]  4 [50]   5 mg/kg every 8 weeks, n [%]  4 [50]   Time on IFX, months median [IQR]  8 [6-32]   Previous episodic IFX, n [%]  1 [13]  Albumin mg/dL, median [IQR]  43 [41-44]  CRP mg/dL, median [IQR]  2 [0-18]  Haemoglobin mM, median [IQR]  8.5 [8.2-8.9]  White blood cells × 109/L, median [IQR]  8.2 [6.5-13.1]    Patients n = 8  Male sex, n [%]  3 [38]  Age, years median [IQR]  39 [30-45]  Disease duration, years median [IQR]  10 [5-24]  Age at diagnosis, years median [IQR]  26 [20-32]  Active smoking, n [%]  3 [38]  Body mass index [BMI], median [IQR]  2419 [25]  Montreal Classification    Age at onset     A1 [< 17years old], n [%]  0 [0]   A2 [17–40 years old], n [%]  5 [62]   A3 [> 40 years old], n [%]  3 [38]   Location     L1 [Ileal], n [%]  0 [0]   L2 [Colonic], n [%]  4 [50]   L3 [Ileocolonic], n [%]  3 [37]   L4 [Isolated upper disease], n [%]  1 [13]   Behaviour     B1 [Non-stricturing, non-penetrating], n [%]  6 [75]   B2 [Stricturing], n [%]  0 [0]   B3 [Penetrating], n [%]  2 [25]  Extraintestinal manifestations, n [%]  4 [50]  Previous intestinal resection, n [%]  2 [25]  TPMT activity     Normal [homozygote wild type], n [%]  8 [100]   Intermediate [heterozygote mutant], n [%]  0 [0]   Enzyme deficiency [homozygote mutant], n [%]  0 [0]  IFX regimen     5 mg/kg every 6 weeks, n [%]  4 [50]   5 mg/kg every 8 weeks, n [%]  4 [50]   Time on IFX, months median [IQR]  8 [6-32]   Previous episodic IFX, n [%]  1 [13]  Albumin mg/dL, median [IQR]  43 [41-44]  CRP mg/dL, median [IQR]  2 [0-18]  Haemoglobin mM, median [IQR]  8.5 [8.2-8.9]  White blood cells × 109/L, median [IQR]  8.2 [6.5-13.1]  IQR, interquartile range; IFX, infliximab; CRP, C-reactive protein; TPMT, thiopurine methyltransferase. View Large 3.2.2. Influence of infliximab on thiopurine metabolites As expected, circulating trough IFX levels increased substantially during the 12-week period of IFX intensification in the parent trial [ΔIFX median 6.5 μg/mL, IQR 1.9-9.8; p < 0.05].14,15 In contrast, thiopurine metabolite levels remained unchanged [6-TGN medians at Weeks 0, 4, 8, 12, and 20: 90 pmol/8 × 108 RBC, 93, 101, 90, and 80, respectively; 6-MeMP: 403, 559, 294, 860, and 291, respectively; p > 0.05] [Figure 4]. Figure 4. View largeDownload slide Thiopurine metabolites during IFX intensification. 6-thioguanine nucleotides [6-TGN]: [A] and methylated mercaptopurine metabolites [6-MeMP]; [B] across time in eight Crohn’s disease patients, all with infliximab [IFX] treatment failure at Week 0, and treated with an intensified IFX regimen until Week 12 resulting in resolution of symptoms. The patients were followed for 20 weeks, during which all received stable thiopurine treatment. Figure 4. View largeDownload slide Thiopurine metabolites during IFX intensification. 6-thioguanine nucleotides [6-TGN]: [A] and methylated mercaptopurine metabolites [6-MeMP]; [B] across time in eight Crohn’s disease patients, all with infliximab [IFX] treatment failure at Week 0, and treated with an intensified IFX regimen until Week 12 resulting in resolution of symptoms. The patients were followed for 20 weeks, during which all received stable thiopurine treatment. 4. Discussion In the first part of this study, we initially observed that significantly fewer patients receiving IFX-thiopurine combination treatment had detectable anti-IFX Abs as compared with those receiving monotherapy. This is consistent with previous pivotal randomised controlled trials.6–8 To further explore this finding, we measured thiopurine metabolite levels and found that 6-TGN levels were substantially lower in IFX Ab-positive than in IFX Ab-negative patients. This observation is congruent with another recently published observation.11 The association between 6-TGN and anti-IFX Abs in two independent cohort studies calls for studies to unravel the precise underlying mechanisms for this observation. We also examined whether circulating trough IFX levels themselves were directly associated with 6-TGN levels independently of anti-IFX Ab formation. In our study, IFX and thiopurine metabolite levels were measured at matched time points, but no such association was found. This observation contrasts with a previous reporting of a moderate positive correlation between IFX and 6-TGN levels.11 However, in that study the correlation analysis was carried out in unselected patients receiving thiopurines rather than in the more relevant anti-IFX Abs-negative subset, as we did. Hence, we believe that it is essential to exclude the potentially confounding effect of anti-IFX Abs to ensure that the correlation is in fact related to 6-TGN. Taken together, our findings are in keeping with the known superior clinical efficacy of combination therapy with IFX and thiopurines over monotherapies, and support therapeutic drug monitoring to optimise 6-TGN levels in order to minimise risk of anti-IFX Abs.6–8 As mentioned above, several studies now support the view that combination treatment with thiopurines and IFX results in lower risk of anti-IFX Abs than IFX monotherapy. However, only approximately 20% of patients receiving IFX actually develop anti-IFX Abs, indicating that other mechanisms may contribute to the overall additive effect of combination treatment.22 These mechanisms are not well understood, but our data support that 6-TGN does not directly associate with circulating IFX levels. Moreover, in the second part of the study we observed that the opposite mechanism did not appear operative, as 6-TGN levels remained stable in a setting of highly increased IFX exposure in an otherwise homogeneous population receiving steady thiopurine dosing. A previous study of 32 patients reported a temporary increase of 6-TGN levels 1-3 weeks after an IFX infusion [non-trough] in patients receiving steady azathioprine therapy, but with 6-TGN having dropped to pre-infusion levels at time of next scheduled infusion [trough].23 We believe that our study provides useful data on these interactions, as IFX levels and 6-TGN levels were examined in two different scenarios with either a steady [part 1] or an intensified [part 2] dose of IFX and showed no association between IFX levels and 6-TGN levels or vice versa. Taken together, it seems likely that the improved efficacy of the combination treatment, beyond reducing risk of anti-IFX Abs, reflects that thiopurines and IFX have different anti-inflammatory mechanisms of action and therefore act synergistically to reduce the inflammatory load more effectively when given in combination─rather than through direct drug interactions.24,25 The findings of the present study have clinical perspectives. Thiopurines are associated with a well-known panel of side effects such as nausea, vomiting, and flu-like symptoms as well as hepatotoxicity and bone marrow depression. These side effects may be reduced or eliminated by shifting to low-dose thiopurine-allopurinol combination treatment in patients who predominantly shunt the drug to 6-MeMP rather than 6-TGN, but this is not always feasible, and some patients tolerate only lower doses. For those also receiving IFX, it is essential to reduce immunogenicity, and the ROC analysis of our data shows that no patients with a 6-TGN level of ≥117 pmol/8 × 108 RBC had detectable anti-IFX Abs. Interestingly, this figure is very close to that reported by another group where patients with 6-TGN < 125 pmol/8 × 108 RBC had higher risk of developing anti-IFX Abs.11 It has been proposed that a 6-TGN threshold of 230-260 pmol/8 × 108 RBCs in patients on thiopurine monotherapy is best correlated to clinical remission.22,26 However, taking into account the present findings, an even lower threshold of ~120 pmol/8 × 108 RBC seems to be sufficient to reduce risk of anti-IFX Abs in patients receiving IFX-thiopurine combination therapy. This favours use of lower thiopurine doses during combination therpay─especially in patients intolerant to standard weight-based thiopurine dosing. Our study along with others shows that 6-MeMP does not influence anti-IFX Abs or circulating IFX, thus supporting use of allopurinol to shunt thiopurine metabolism in the direction of 6-TGN.9–11 The major limitations in cohort 1 are the retrospective observational design including inherent restrictions regarding causality, and that testing was neither completely standardised nor predefined, introducing risk of selection bias. As the objective was to explore potential drug interactions between thiopurine metabolites and IFX, treatment outcomes were not assessed and data from patients with Crohn’s disease and ulcerative colitis were pooled to yield power [cohort 1]. However, the majority of patients had surprisingly low 6-TGN levels, thus warranting caution when extrapolating the results. Even though we cannot rule out that lack of association between 6-TGN and IFX in cohort 1 was due confounders, the multiple regression analysis did not indicate that this was the case; and it was not apparent that patients receiving IFX-thiopurine combination therapy had a markedly more severe disease course than those receiving IFX monotherapy [Table 1]. The use in cohort 1 of different analytical techniques across time, due to changes in technology by the supplier company, is unlikely to have had major implications in IFX quantifications as the IFX assays have been shown to correlate well.16,17 Bias may, however, have been introduced regarding anti-IFX Ab detection, as assays are less congruent and have different drug sensitivity [RIA has moderate drug sensitivity;18,27 EIA is not prone to drug interference;16,18 RGA is a a cell-based assay which measures the functional activity of drug and anti-drug Abs at the cellular TNF receptor level, thus rendering it sensitive to IFX [> ~0.65 µg/mL] ─only samples without detectable IFX by RGA were assessed for anti-IFX Abs by RGA, whereas samples with detectable IFX were generally tested for anti-IFX Abs by EIA to accommodate for potential drug interference.17 As thiopurines [AZA and 6-MP] were handed out in our clinic to the patients, adherence is likely to be high although non-adherence in patients with low thiopurine metabolite levels cannot completely be ruled out. Although cohort 2 was very small, our observations add weight to lack of influence of IFX on thiopurine metabolite levels, because the cohort consisted of a highly homogeneous population receiving a steady thiopurine dosing with repeated standardised measurements of thiopurine metabolites and IFX and anti-IFX Abs across time, and with the only variable changed being the increased exposure of IFX. In conclusion, superior effectiveness of IFX-thiopurine combination therapy in part relates to decreased anti-IFX Abs. Importantly, this effect seems to be influenced by 6-TGN and, with a relatively low 6-TGN threshold of ≥ ~120 pmol/8 × 108 RBC, to substantially reduce or completely eliminate risk of anti-IFX Abs. Additional benefit from IFX-thiopurine combination therapy over monotherapies appears related to synergy between different anti-inflammatory modes of action rather than to direct interactions between thiopurine metabolites and IFX or vice versa. The results support the view that therapeutic drug monitoring can be implemented during thiopurine-IFX combination treatment to optimise 6-TGN levels and lower 6-MeMP levels, in order to improve efficacy and reduce risk of side effects. For those patients who only tolerate lower doses, a 6-TGN level of about 120 pmol/8 × 108 RBC as of now seems to be a sound target threshold during combination therapy. Funding This study was kindly supported by the Danish Colitis and Crohn’s Disease Society, Jacob Madsen and Olga Madsens Fund, and Herlev-Gentofte Hospital Research Council. Conflict of Interest Within the past year, CS has served as speaker for Takeda and as advisory board member for Janssen. JB has served as advisory board member for Abbvie, Janssen, and Takeda; and as primary investigator for Abbvie and Janssen. Author Contributions Study design: DVM, JB, CS. Collection of data: DVM, CS. Analysis and interpretation of data: DVM, JB, CS. Drafting the manuscript: DVM, JB, CS. Revising the manuscript and approval of final manuscript: all authors. References 1. Lichtenstein GR, Hanauer SB, Sandborn WJ; Practice Parameters Committee of American College of Gastroenterology. Management of Crohn’s disease in adults. Am J Gastroenterol  2009; 104: 465– 83; quiz 464, 484. Google Scholar CrossRef Search ADS PubMed  2. Gisbert JP, Panés J. Loss of response and requirement of infliximab dose intensification in Crohn’s disease: a review. Am J Gastroenterol  2009; 104: 760– 7. Google Scholar CrossRef Search ADS PubMed  3. Steenholdt C, Bendtzen K, Brynskov J, Ainsworth MA. Optimizing treatment with TNF inhibitors in inflammatory bowel disease by monitoring drug levels and antidrug antibodies. Inflamm Bowel Dis  2016; 22: 1999– 2015. Google Scholar CrossRef Search ADS PubMed  4. Steenholdt C, Svenson M, Bendtzen K, Thomsen OØ, Brynskov J, Ainsworth MA. Severe infusion reactions to infliximab: aetiology, immunogenicity and risk factors in patients with inflammatory bowel disease. Aliment Pharmacol Ther  2011; 34: 51– 8. Google Scholar CrossRef Search ADS PubMed  5. Frederiksen MT, Ainsworth MA, Brynskov J, Thomsen OO, Bendtzen K, Steenholdt C. Antibodies against infliximab are associated with de novo development of antibodies to adalimumab and therapeutic failure in infliximab-to-adalimumab switchers with IBD. Inflamm Bowel Dis  2014; 20: 1714– 21. Google Scholar CrossRef Search ADS PubMed  6. Colombel JF, Sandborn WJ, Reinisch Wet al.  ; SONIC Study Group. Infliximab, azathioprine, or combination therapy for Crohn’s disease. N Engl J Med  2010; 362: 1383– 95. Google Scholar CrossRef Search ADS PubMed  7. Colombel JF, Reinisch W, Mantzaris GJet al.   Randomised clinical trial: deep remission in biologic and immunomodulator naïve patients with Crohn’s disease - a SONIC post hoc analysis. Aliment Pharmacol Ther  2015; 41: 734– 46. Google Scholar CrossRef Search ADS PubMed  8. Panaccione R, Ghosh S, Middleton Set al.  . Combination therapy with infliximab and azathioprine is superior to monotherapy with either agent in ulcerative colitis. Gastroenterology  2014; 146: 392– 400.e3. Google Scholar CrossRef Search ADS PubMed  9. Coskun M, Steenholdt C, de Boer NK, Nielsen OH. Pharmacology and optimization of thiopurines and methotrexate in inflammatory bowel disease. Clin Pharmacokinet  2016; 55: 257– 74. Google Scholar CrossRef Search ADS PubMed  10. Moon W, Loftus EVJr. Review article: recent advances in pharmacogenetics and pharmacokinetics for safe and effective thiopurine therapy in inflammatory bowel disease. Aliment Pharmacol Ther  2016; 43: 863– 83. Google Scholar CrossRef Search ADS PubMed  11. Yarur AJ, Kubiliun MJ, Czul Fet al.  . Concentrations of 6-thioguanine nucleotide correlate with trough levels of infliximab in patients with inflammatory bowel disease on combination therapy. Clin Gastroenterol Hepatol  2015; 13: 1118– 24.e3. Google Scholar CrossRef Search ADS PubMed  12. Bar-Yoseph H, Waterman M, Almog Ret al.  . Prevention of antidrug antibody formation to infliximab in Crohn’s patients with prior failure of thiopurines. Clin Gastroenterol Hepatol  2017; 15: 69– 75. Google Scholar CrossRef Search ADS PubMed  13. Pozler O, Chládek J, Malý Jet al.  . Steady-state of azathioprine during initiation treatment of pediatric inflammatory bowel disease. J Crohns Colitis  2010; 4: 623– 8. Google Scholar CrossRef Search ADS PubMed  14. Steenholdt C, Brynskov J, Thomsen OØet al.  . Individualised therapy is more cost-effective than dose intensification in patients with Crohn’s disease who lose response to anti-TNF treatment: a randomised, controlled trial. Gut  2014; 63: 919– 27. Google Scholar CrossRef Search ADS PubMed  15. Steenholdt C, Brynskov J, Thomsen OØet al.  . individualized therapy is a long-term cost-effective method compared with dose intensification in Crohn’s Disease patients failing infliximab. Dig Dis Sci  2015; 60: 2762– 70. Google Scholar CrossRef Search ADS PubMed  16. Steenholdt C, Ainsworth MA, Tovey Met al.  . Comparison of techniques for monitoring infliximab and antibodies against infliximab in Crohn’s disease. Ther Drug Monit  2013; 35: 530– 8. Google Scholar CrossRef Search ADS PubMed  17. Steenholdt C, Bendtzen K, Brynskov J, Thomsen OØ, Ainsworth MA. Clinical implications of measuring drug and anti-drug antibodies by different assays when optimizing infliximab treatment failure in Crohn’s disease: post hoc analysis of a randomized controlled trial. Am J Gastroenterol  2014; 109: 1055– 64. Google Scholar CrossRef Search ADS PubMed  18. Steenholdt C, Frederiksen MT, Bendtzen K, Ainsworth MA, Thomsen OØ, Brynskov J. Time course and clinical implications of development of antibodies against adalimumab in patients with inflammatory bowel disease. J Clin Gastroenterol  2016; 50: 483– 9. Google Scholar CrossRef Search ADS PubMed  19. Dervieux T, Boulieu R. Simultaneous determination of 6-thioguanine and methyl 6-mercaptopurine nucleotides of azathioprine in red blood cells by HPLC. Clin Chem  1998; 44: 551– 5. Google Scholar PubMed  20. Dervieux T, Boulieu R. Identification of 6-methylmercaptopurine derivative formed during acid hydrolysis of thiopurine nucleotides in erythrocytes, using liquid chromatography-mass spectrometry, infrared spectroscopy, and nuclear magnetic resonance assay. Clin Chem  1998; 44: 2511– 5. Google Scholar PubMed  21. Lennard L. Commentary on: Differences in nucleotide hydrolysis contribute to the differences between erythrocyte 6-thioguanine nucleotide concentrations determined by two widely used methods. Clin Chem  2003; 49: 1551; author reply 1551–2. Google Scholar CrossRef Search ADS PubMed  22. Vande Casteele N, Herfarth H, Katz J, Falck-Ytter Y, Singh S. American Gastroenterological Association Institute Technical Review on the Role of Therapeutic Drug Monitoring in the Management of Inflammatory Bowel Diseases. Gastroenterology  2017; 153: 835– 57.e6. Google Scholar CrossRef Search ADS PubMed  23. Roblin X, Serre-Debeauvais F, Phelip JM, Bessard G, Bonaz B. Drug interaction between infliximab and azathioprine in patients with Crohn’s disease. Aliment Pharmacol Ther  2003; 18: 917– 25. Google Scholar CrossRef Search ADS PubMed  24. Olesen CM, Coskun M, Peyrin-Biroulet L, Nielsen OH. Mechanisms behind efficacy of tumor necrosis factor inhibitors in inflammatory bowel diseases. Pharmacol Ther  2016; 159: 110– 9. Google Scholar CrossRef Search ADS PubMed  25. Tiede I, Fritz G, Strand Set al.  . CD28-dependent Rac1 activation is the molecular target of azathioprine in primary human CD4+ T lymphocytes. J Clin Invest  2003; 111: 1133– 45. Google Scholar CrossRef Search ADS PubMed  26. Moreau AC, Paul S, Del Tedesco Eet al.  . Association between 6-thioguanine nucleotides levels and clinical remission in inflammatory disease: a meta-analysis. Inflamm Bowel Dis  2014; 20: 464– 71. Google Scholar CrossRef Search ADS PubMed  27. Svenson M, Geborek P, Saxne T, Bendtzen K. Monitoring patients treated with anti-TNF-alpha biopharmaceuticals: assessing serum infliximab and anti-infliximab antibodies. Rheumatology [Oxford]  2007; 46: 1828– 34. Google Scholar CrossRef Search ADS PubMed  Copyright © 2017 European Crohn’s and Colitis Organisation (ECCO). Published by Oxford University Press. All rights reserved. For permissions, please email: journals.permissions@oup.com

Journal

Journal of Crohn's and ColitisOxford University Press

Published: Mar 1, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off