- Email: [email protected]
695 Insufficient Infliximab Exposure Predisposes to Immunogenicity and Enhanced Clearance of Infliximab in IBD
AGA Abstracts
significant reduction in TRI (4.2 vs 2.1 µg/mL; p=0.02). Four patients in Cohort A (14.2%) had an unfavourable evolution of IFX pharmacokinetic defined by a decrease of TRI < 1 µg/mL or undetectable TRI with positive ATI at one year, five in Cohort B (18.5%), and 14 in Cohort C (53.8%) (p=0.01 between and C and between B and C). By ROC analysis (AUROC: O.93), a threshold of 6-TGN <105 pmoles was associated with this unfavourable pharmacokinetic (sensitivity: 67%; specificity: 92%; Likelihood ratio: 7.67). Conclusions: AZA dose reduction in IBD patients on combination therapy is as effective as the maintenance of AZA at the same dose and may improve AZA safety profile. A threshold of 6-TGN< 105pmoles was associated with an unfavourable evolution of IFX pharmacokinetics.
dose, immunomodulators, albumin and serum CRP were collected. PK and ATI were analysed simultaneously by nonlinear mixed-effects modelling. Results: 997 IFX concentrations and 756 ATI measurements from 332 IBD patients (253 Crohn's disease, 79 ulcerative colitis) were included. Eighty per cent of patients received IFX as the first anti-TNF agent; 43% received concomitant immunomodulation. Mean (SD) IFX dose was 5.47±1.33 mg/kg. ATIs were detected in 75/332 (23%) patients; ATI titers > 30 AU/mL were consistently associated with undetectable IFX concentrations. IFX clearance was affected by: body weight (40-149 kg) ranging from 0.27 to 0.53 L/Day, serum albumin (2-5.4 g/dL) from 0.93 to 0.24 L/Day and titres of ATIs (0-53000 AU/ml) from 0.36 L/Day to 15.93 L/Day (P<0.001). Insufficient exposure below an IFX trough level of 3 µg/ml resulted in a 4 fold increased risk of ATI development. Our pharmacokinetic model, based on the factors that were identified in this study, was able to accurately predict serum IFX concentrations and ATI development. Conclusions: ATI, high body weight and low albumin increase IFX clearance. IFX exposure below 3 µg/ml increases risk of ATI significantly. Identification of influential PK and ATI factors improves prediction of infliximab levels, potentially allowing individualised dosing and cost reduction
694 A Variable Number of Tandem Repeat Polymorphism in the Promotor Region of the Neonatal FC Receptor Affects Anti-TNF Serum Levels in IBD Thomas Billiet, Isabelle Cleynen, Marc Ferrante, Gert Van Assche, Ann Gils, Severine Vermeire Introduction: Ample evidence exists that anti-TNF trough levels (TL) during induction determine short and long-term outcome in IBD. Sex, inflammatory burden (CRP), albumin and BMI are known to influence these early concentrations. The neonatal Fc (FcRn) or Brambell receptor is responsible for extending serum half-life of albumin and IgGs and harbours a variable number of tandem repeats (VNTR) polymorphism in the promotor region of the FcRn gene.1 We hypothesized that anti-TNF TL are affected by this polymorphism. Methods: We analysed infliximab (IFX) TL in 368 IBD patients (215 CD, 153 UC), all IFX naïve, who received IFX 5mg/kg at week 0, 2 & 6. A second independent cohort of 139 IBD patients (100 CD, 39 UC), adalimumab (ADM) naïve and treated with 160/80/ 40mg at week 0, 2 & 4, was analysed for ADM TL. This represented a combined total of 1014 TL measurements. All patients were genotyped for the VNTR polymorphism with PCR as previously described.1 TL were measured using an in-house-developed ELISA. The median TL during anti-TNF induction (from week 2 & 6 for IFX, from week 2 & 4 for ADM) was calculated for every patient. Genotype, patient- and disease-related factors were analysed in multiple linear regression with median TL as dependent variable. Results: Distribution of VNTR genotypes was similar as previously described.1 Only genotypes with a frequency > 5% (VNTR2/3 (n=83 or 16,4%) and VNTR3/3 (n=414 or 81.6%)) were further considered for analysis. Multiple regression in the IFX group withheld several factors highly associated with IFX TL but not the VNTR genotype (see table 1). When excluding patients who later on developed antibodies to IFX (ATI) (n=53), predictors remained unchanged but now also included the VNTR2/3 genotype (p=0.03), resulting in a predicted 16% lower median TL as opposed to the VNTR3/3 genotype. Prior anti-TNF use other than IFX predicted a 30% lower median TL than anti-TNF naïve patients (p=0.001). In the ADM group, the VNTR2/ 3 genotype predicted a 23% lower median ADM TL (p=0.007). The combined presence of VNTR2/3, male sex and prior IFX predicted a 40% lower median ADM TL as opposed to those without these risk factors (p=0.045). Conclusion: The VNTR2/3 genotype in the FcRn gene is associated with lower IFX but also lower ADM induction TL. Developing anti-drug antibodies for both IFX or ADM later on and the prior use of another anti-TNF agent were also associated with lower induction TL. We could furthermore confirm previously identified risk factors for lower induction TL. We therefore propose that patients, in whom these risk factors are present, might benefit from a higher anti-TNF induction dose to ensure optimal disease outcome. 1 Sachs et al. A variable number of tandem repeats polymorphism influences the transcriptional activity of the neonatal Fc receptor a-chain promoter. Immunology 2006; 119: 83-89. Table 1
Relationship between IFX concentrations falling below target concentration and probability of ATI for different albumin concentrations
696 Association of Ustekinumab Trough Concentrations With Clinical, Biochemical and Endoscopic Outcomes Robert Battat, Uri Kopylov, Talat Bessissow, Alain Bitton, Albert Cohen, Myriam Martel, Ernest G. Seidman, Waqqas Afif Background: Ustekinumab (UST), an interleukin-12/23p40 inhibitor, is effective in Crohn's disease (CD). For TNF-alpha inhibitors, trough concentrations and the presence of antidrug antibodies are significantly associated with clinical and endoscopic outcomes. Such an association has not been previously established for UST. We investigated the association between UST trough concentrations and clinical, biologic and endoscopic response in CD patients. Methods: Anti-TNF exposed CD patients treated with UST (2013-2015) at a tertiary care center were recruited. All patients were induced with UST 90 mg at week 0,1,2 and followed by 90 mg every 8 weeks. Outcomes assessed were: clinical response (decreased Harvey Bradshaw Index (HBI) by 3 points), clinical remission (HBI £4) and endoscopic response (50% reduction in the simplified endoscopic activity score for CD (SES-CD) or a SES-CD £2). Physician's global assessment was used if HBI or SES-CD was not applicable (ileostomy or ileal pouch). Outcomes were assessed for longitudinal patients at 2 months and 6 months and at any time point >6 months for cross-sectional patients. UST and UST antibody concentrations were assessed using a liquid phase assay (Prometheus Laboratories Inc.). Results: Fifty-nine patients were included in the study. Outcomes for individual and combined cohorts, as well as subgroup analysis, are outlined in Table 1. Concurrent use of immunosuppression (methotrexate/thiopurines) was 26% and after >6 months of treatment, 78% of patients received UST every 4 weeks. In the combined cohort, the endoscopic response was 51.4% (95% CI, 34.7-67.8). Endoscopic response was greater in colonic CD, compared to other locations (80% (8/10) vs. 40.7% (11/27) (p=0.034)). At >6 months, UST antibodies were undetectable (0/49). Only one patient had undetectable UST concentrations. Using receiver operating characteristic analysis, UST > 4.5 µg/mL was associated with endoscopic response (72.2% sensitivity, 83.3% specificity, p=0.0006, AUC 0.782). UST >4.5 µg/mL, compared to lower levels, increased endoscopic response (81.3% vs. 25%, p= 0.008) and the combined outcome of steroid-free clinical remission and endoscopic response (50% vs. 15%, p=0.024). UST >5 µg/mL, compared to lower levels, was associated with normal CRP (63.6% vs. 33%, p=0.024). There was a non-significant increase in steroid-free clinical remission (55% vs 44%, p=0.46) and fecal calprotectin <200 µg/mL (19.2% vs 11.5%, p=0.44). Conclusion: UST induced and maintained steroid-free clinical remission in a large proportion of anti-TNF refractory CD patients. Patients with colonic disease had significantly improved clinical and endoscopic outcomes. Adequate UST concentrations (>4.5 µg/mL) are associated with important patient outcomes, but larger studies are required to confirm the optimal therapeutic concentration. Clinical Outcomes of Patients on Ustekinumab
Factors independently influencing median TL. All variables had a variance inflation factor of less than 2, implicating that there were no issues with multicollinearity. * Excluding patients who later developed ATI.
695 Insufficient Infliximab Exposure Predisposes to Immunogenicity and Enhanced Clearance of Infliximab in IBD Johannan F. Brandse, Anne S. Strik, Diane Mould, Yael Ashruf, Oscar S. Smeekes, Sabine Kuin, Gijs van den Brink, Geert R. D'Haens Background: Factors influencing the pharmacokinetics (PK) of infliximab (IFX) in inflammatory bowel disease (IBD) have mainly been studied in clinical trials. We studied a real world IBD patient cohort to identify patient, disease and treatment characteristics influencing IFX PK. Methods: Levels of IFX and antibodies to IFX (ATI) were measured in IBD patients at a single academic IBD center using an ELISA and antigen binding test (Sanquin laboratories, Amsterdam, The Netherlands). Gender, age, weight, disease location and behaviour, IFX
AGA Abstracts
S-144
significant reduction in TRI (4.2 vs 2.1 µg/mL; p=0.02). Four patients in Cohort A (14.2%) had an unfavourable evolution of IFX pharmacokinetic defined by a decrease of TRI < 1 µg/mL or undetectable TRI with positive ATI at one year, five in Cohort B (18.5%), and 14 in Cohort C (53.8%) (p=0.01 between and C and between B and C). By ROC analysis (AUROC: O.93), a threshold of 6-TGN <105 pmoles was associated with this unfavourable pharmacokinetic (sensitivity: 67%; specificity: 92%; Likelihood ratio: 7.67). Conclusions: AZA dose reduction in IBD patients on combination therapy is as effective as the maintenance of AZA at the same dose and may improve AZA safety profile. A threshold of 6-TGN< 105pmoles was associated with an unfavourable evolution of IFX pharmacokinetics.
dose, immunomodulators, albumin and serum CRP were collected. PK and ATI were analysed simultaneously by nonlinear mixed-effects modelling. Results: 997 IFX concentrations and 756 ATI measurements from 332 IBD patients (253 Crohn's disease, 79 ulcerative colitis) were included. Eighty per cent of patients received IFX as the first anti-TNF agent; 43% received concomitant immunomodulation. Mean (SD) IFX dose was 5.47±1.33 mg/kg. ATIs were detected in 75/332 (23%) patients; ATI titers > 30 AU/mL were consistently associated with undetectable IFX concentrations. IFX clearance was affected by: body weight (40-149 kg) ranging from 0.27 to 0.53 L/Day, serum albumin (2-5.4 g/dL) from 0.93 to 0.24 L/Day and titres of ATIs (0-53000 AU/ml) from 0.36 L/Day to 15.93 L/Day (P<0.001). Insufficient exposure below an IFX trough level of 3 µg/ml resulted in a 4 fold increased risk of ATI development. Our pharmacokinetic model, based on the factors that were identified in this study, was able to accurately predict serum IFX concentrations and ATI development. Conclusions: ATI, high body weight and low albumin increase IFX clearance. IFX exposure below 3 µg/ml increases risk of ATI significantly. Identification of influential PK and ATI factors improves prediction of infliximab levels, potentially allowing individualised dosing and cost reduction
694 A Variable Number of Tandem Repeat Polymorphism in the Promotor Region of the Neonatal FC Receptor Affects Anti-TNF Serum Levels in IBD Thomas Billiet, Isabelle Cleynen, Marc Ferrante, Gert Van Assche, Ann Gils, Severine Vermeire Introduction: Ample evidence exists that anti-TNF trough levels (TL) during induction determine short and long-term outcome in IBD. Sex, inflammatory burden (CRP), albumin and BMI are known to influence these early concentrations. The neonatal Fc (FcRn) or Brambell receptor is responsible for extending serum half-life of albumin and IgGs and harbours a variable number of tandem repeats (VNTR) polymorphism in the promotor region of the FcRn gene.1 We hypothesized that anti-TNF TL are affected by this polymorphism. Methods: We analysed infliximab (IFX) TL in 368 IBD patients (215 CD, 153 UC), all IFX naïve, who received IFX 5mg/kg at week 0, 2 & 6. A second independent cohort of 139 IBD patients (100 CD, 39 UC), adalimumab (ADM) naïve and treated with 160/80/ 40mg at week 0, 2 & 4, was analysed for ADM TL. This represented a combined total of 1014 TL measurements. All patients were genotyped for the VNTR polymorphism with PCR as previously described.1 TL were measured using an in-house-developed ELISA. The median TL during anti-TNF induction (from week 2 & 6 for IFX, from week 2 & 4 for ADM) was calculated for every patient. Genotype, patient- and disease-related factors were analysed in multiple linear regression with median TL as dependent variable. Results: Distribution of VNTR genotypes was similar as previously described.1 Only genotypes with a frequency > 5% (VNTR2/3 (n=83 or 16,4%) and VNTR3/3 (n=414 or 81.6%)) were further considered for analysis. Multiple regression in the IFX group withheld several factors highly associated with IFX TL but not the VNTR genotype (see table 1). When excluding patients who later on developed antibodies to IFX (ATI) (n=53), predictors remained unchanged but now also included the VNTR2/3 genotype (p=0.03), resulting in a predicted 16% lower median TL as opposed to the VNTR3/3 genotype. Prior anti-TNF use other than IFX predicted a 30% lower median TL than anti-TNF naïve patients (p=0.001). In the ADM group, the VNTR2/ 3 genotype predicted a 23% lower median ADM TL (p=0.007). The combined presence of VNTR2/3, male sex and prior IFX predicted a 40% lower median ADM TL as opposed to those without these risk factors (p=0.045). Conclusion: The VNTR2/3 genotype in the FcRn gene is associated with lower IFX but also lower ADM induction TL. Developing anti-drug antibodies for both IFX or ADM later on and the prior use of another anti-TNF agent were also associated with lower induction TL. We could furthermore confirm previously identified risk factors for lower induction TL. We therefore propose that patients, in whom these risk factors are present, might benefit from a higher anti-TNF induction dose to ensure optimal disease outcome. 1 Sachs et al. A variable number of tandem repeats polymorphism influences the transcriptional activity of the neonatal Fc receptor a-chain promoter. Immunology 2006; 119: 83-89. Table 1
Relationship between IFX concentrations falling below target concentration and probability of ATI for different albumin concentrations
696 Association of Ustekinumab Trough Concentrations With Clinical, Biochemical and Endoscopic Outcomes Robert Battat, Uri Kopylov, Talat Bessissow, Alain Bitton, Albert Cohen, Myriam Martel, Ernest G. Seidman, Waqqas Afif Background: Ustekinumab (UST), an interleukin-12/23p40 inhibitor, is effective in Crohn's disease (CD). For TNF-alpha inhibitors, trough concentrations and the presence of antidrug antibodies are significantly associated with clinical and endoscopic outcomes. Such an association has not been previously established for UST. We investigated the association between UST trough concentrations and clinical, biologic and endoscopic response in CD patients. Methods: Anti-TNF exposed CD patients treated with UST (2013-2015) at a tertiary care center were recruited. All patients were induced with UST 90 mg at week 0,1,2 and followed by 90 mg every 8 weeks. Outcomes assessed were: clinical response (decreased Harvey Bradshaw Index (HBI) by 3 points), clinical remission (HBI £4) and endoscopic response (50% reduction in the simplified endoscopic activity score for CD (SES-CD) or a SES-CD £2). Physician's global assessment was used if HBI or SES-CD was not applicable (ileostomy or ileal pouch). Outcomes were assessed for longitudinal patients at 2 months and 6 months and at any time point >6 months for cross-sectional patients. UST and UST antibody concentrations were assessed using a liquid phase assay (Prometheus Laboratories Inc.). Results: Fifty-nine patients were included in the study. Outcomes for individual and combined cohorts, as well as subgroup analysis, are outlined in Table 1. Concurrent use of immunosuppression (methotrexate/thiopurines) was 26% and after >6 months of treatment, 78% of patients received UST every 4 weeks. In the combined cohort, the endoscopic response was 51.4% (95% CI, 34.7-67.8). Endoscopic response was greater in colonic CD, compared to other locations (80% (8/10) vs. 40.7% (11/27) (p=0.034)). At >6 months, UST antibodies were undetectable (0/49). Only one patient had undetectable UST concentrations. Using receiver operating characteristic analysis, UST > 4.5 µg/mL was associated with endoscopic response (72.2% sensitivity, 83.3% specificity, p=0.0006, AUC 0.782). UST >4.5 µg/mL, compared to lower levels, increased endoscopic response (81.3% vs. 25%, p= 0.008) and the combined outcome of steroid-free clinical remission and endoscopic response (50% vs. 15%, p=0.024). UST >5 µg/mL, compared to lower levels, was associated with normal CRP (63.6% vs. 33%, p=0.024). There was a non-significant increase in steroid-free clinical remission (55% vs 44%, p=0.46) and fecal calprotectin <200 µg/mL (19.2% vs 11.5%, p=0.44). Conclusion: UST induced and maintained steroid-free clinical remission in a large proportion of anti-TNF refractory CD patients. Patients with colonic disease had significantly improved clinical and endoscopic outcomes. Adequate UST concentrations (>4.5 µg/mL) are associated with important patient outcomes, but larger studies are required to confirm the optimal therapeutic concentration. Clinical Outcomes of Patients on Ustekinumab
Factors independently influencing median TL. All variables had a variance inflation factor of less than 2, implicating that there were no issues with multicollinearity. * Excluding patients who later developed ATI.
695 Insufficient Infliximab Exposure Predisposes to Immunogenicity and Enhanced Clearance of Infliximab in IBD Johannan F. Brandse, Anne S. Strik, Diane Mould, Yael Ashruf, Oscar S. Smeekes, Sabine Kuin, Gijs van den Brink, Geert R. D'Haens Background: Factors influencing the pharmacokinetics (PK) of infliximab (IFX) in inflammatory bowel disease (IBD) have mainly been studied in clinical trials. We studied a real world IBD patient cohort to identify patient, disease and treatment characteristics influencing IFX PK. Methods: Levels of IFX and antibodies to IFX (ATI) were measured in IBD patients at a single academic IBD center using an ELISA and antigen binding test (Sanquin laboratories, Amsterdam, The Netherlands). Gender, age, weight, disease location and behaviour, IFX
AGA Abstracts
S-144