S25 © 2019 by the Crohn’s & Colitis Foundation and the AGA Institute. Crohn’s & Colitis Congress® Background: Monitoring ustekinumab (UST) concentrations and antibodies-to ustekinumab (ATU) during IBD treatment may allow more informed decisions in assessing exposure/response and appropriate dosing. To aid in interpreting results in the context of Janssen’s published UST results, the reliability of diferent assays measuring UST and ATU were compared with those from Janssen. This abstract reports the comparison of the UST and ATU assays from Laboratory Corporation of America (LabCorp, Calabasas, CA, USA) and Janssen (JRD, Spring House, PA, USA). Results from the other companies will be reported at a later time. Methods: Blinded test samples, prepared by JRD, were sent to the LabCorp and JRD labs for UST and ATU assessments. Results were reported to JRD for integrated analyses. All assays were tested for specificity, selectivity, accuracy and precision. ATU assays were evaluated for sensitivity, drug interference, and potential interfer- ence of IL-12. UST and ATU were tested at LabCorp using Meso Scale Discovery (MSD®, Rockville, MD, USA) electrochemiluminescent immunoassays (ECLIA) and at JRD using different MSD ECLIAs. The lower limit of quantification was 0.4 mg/mL for the LabCorp UST concentration assay and 0.1688 mg/mL for the JRD UST con- centration assay. Results: Strong agreement was observed between the JRD and LabCorp UST assays. Specifcity was demonstrated when both UST assays accurately detected 1.0 or 10.0 mg/mL of UST but did not detect other human monoclonal antibodies (mAb). The presence of ATU titers up to 200 or IL-12 concentrations up to 100 pg/ mL did not interfere with the UST assessment in either assay. Accuracy was con- frmed by 3 independent measurements of UST-spiked (0.06-32 mg/mL) human psoriasis (PSO) sera. However, accuracy was not confrmed when testing incurred samples from PSO subjects previously exposed to ustekinumab at concentrations below approximately 1.3 mg/mL. Both UST assays were precise, as determined by inter-occasion reproducibility. Strong agreement was observed between the JRD and LabCorp ATU assays. Both ATU assays specifically detected anti-UST antibodies; results were not affected by high titer antibodies against other human mAb. Both ATU assays demonstrated drug tolerance to 8.0 mg/mL of UST, with the LabCorp assay demonstrating drug tolerance to 32.0 mg/mL. Concentrations of free or bound IL-12 (≤100 pg/ mL) did not interfere with ATU detection in either assay. Both ATU assays were reproducible. Conclusion: Our study results indicate that the LabCorp UST concentration and ATU assays strongly correlate with those from JRD. The substantial agreement between the LabCorp and JRD assays may provide support to clinicians in their use of these assays, and for understanding their patients’ UST and ATU results relative to published data from clinical studies of UST. P050 METHOTREXATE POLYGLUTAMATES AS BIOMARKERS OF TREATMENT RESPONSE IN PEDIATRIC INFLAMMATORY BOWEL DISEASE Valentina Shakhnovich, Taina Jausurawong, Ryan Morrow, Mara Becker, Ryan Funk Background: Therapeutic drug monitoring methods are lacking for methotrexate (MTX), an immunomodulator used in Infammatory Bowel Disease (IBD) treatment, alone or in conjunction with infiximab (IFX). Although studies demonstrate a lack of relationship between MTX plasma concentrations and efcacy/toxicity, recent studies in rheumatoid and juvenile idiopathic arthritis suggest a positive correla- tion between therapeutic response and accumulation of intracellular MTX metab- olites, MTX polyglutamates (MTX-Glu). The aim of this cross-sectional investigation was to test the hypothesis that intracellular MTX-Glu are associated with therapeu- tic response in pediatric IBD. Methods: 21 children (5-21 years; 90% Crohn’s disease), receiving combination therapy with IFX and MTX for maintenance treatment of IBD, had MTX-Glu 1-6 mea- sured in erythrocytes, using HPLC-MS/MS methodologies, and serum IFX troughs and anti-IFX antibodies (Anti-IFX) measured, using a NF-kB luciferase gene-re- porter assay (ARUP Laboratories). Medical records were reviewed and disease activity determined (active vs quiescent) via consensus by 2 pediatric gastroen- terologists on Physician Global Assessment. Only children on stable dosing were included (i.e., no changes to dose or interval of either drug for at least 2 IFX cycles). Nonparametric tests (e.g., Mann-Whitney U) were used to compare MTX-Glu, IFX, anti-IFX, laboratory and demographic data in children with active vs quiescent disease, and associations between MTX-Glu and parameters of interest explored via Pearson’s correlation (SPSS24; α=0.05). To account for variability in standard- of-care IFX and MTX dosing, data were normalized for mg/kg and interval, as appropriate. Results: MTX dose ranged 5-25mg weekly (86% oral) and, adjusted for weight (mg/kg), correlated with total MTX-Glu accumulation (ρ=0.6; α=0.01), especially long-chain MTX-Glu 3-5 (ρ=0.7; p=0.01). Children with quiescent disease (n=11) had signifcantly higher dose-adjusted total MTX-Glu than children with active dis- ease (n=10), 224 ± 94.6 vs 133 ± 85.6 (p=0.04), despite receiving comparable MTX doses (0.22 ± 0.12 vs 0.31 ± 0.18 mg/kg; p=0.39). IFX troughs (μg/ml) unadjusted (24.3 ± 16.1 vs 16.7 ± 13.0) or adjusted (adj) for IFX dose and interval (86.9 ± 59.7 vs 57.4 ± 46.8) were comparable between study groups, which were also comparable for age and disease duration (p>0.25). Neither MTX dose nor MTX-Glu correlated signifcantly with IFX trough or adjtrough (ρ -0.09 to 0.39; p>0.1). Only 1 child, with undetectable IFX trough, had anti-IFX. Conclusions: The observed positive correlation between MTX dose and MTX-Glu suggests that MTX-Glu refect systemic MTX exposure in IBD. Despite comparable IFX troughs, higher total MTX-Glu, particularly long-chain MTX-Glu, were associ- ated with quiescent vs active disease, suggesting that MTX-Glu may be important markers of disease response in IBD. P051 REAL WORLD EXPERIENCE IN THERAPEUTIC DRUG MONITORING IN INFLAMMATORY BOWEL DISEASE PATIENTS ON USTEKINUMAB Kerri Glassner, Malcolm Irani, Hoda Malaty, Bincy P. Abraham Background: Current evidence supports the use of reactive therapeutic drug monitoring(TDM) to guide treatment changes in infammatory bowel disease(IBD) patients treated with anti-TNF agents. However, there is less data available on the utility of drug monitoring with ustekinumab(UST). Aims: To determine if trough levels of UST have an impact on treatment based on clinical, laboratory, and endoscopic fndings. Methods: A total of 55 UST trough levels were collected from 46 patients with Crohn disease seen at our IBD center from May 2017 to June 2018. Levels were checked either reactively due to ongoing clinical symptoms, endoscopic, biologic markers(fecal calprotectin, ESR, CRP), or radiologic studies suggestive of active dis- ease; or drawn proactively to guide de-escalation of disease therapy. Demographic data was collected including gender, medications, age, disease duration, BMI and ethnicity. Infammatory markers(ESR, CRP), vitamin D, clinical scoring(HBI), endo- scopic and histologic data were obtained within 6 months prior to level being drawn, and at 6 month follow up. Low drug levels were defned as UST trough level less than 5ug/mL. Results: Of 45 trough levels drawn reactively, 71% were low; of 10 drawn proac- tively, 90% were low. There were 34 patients who had 41 low UST trough levels. These patients were more likely to be female,(64%, p=0.002), Caucasian(90%, p= 0.002), non obese(BMI≤30) (78%, p=0.001), and less likely to be on dual biologic therapy(35%, p=0.001), immune modulator(22%, p=0.002), or steroids(37%, p=0.01). Compared to patients with high UST levels, those with low levels were more likely to have higher inflammatory markers, ESR 24.9 vs 20.4, p=0.01, and CRP 22.1 vs 4.1, p=0.003. There was no significant difference in vitamin D, clin- ical, endoscopic, or histologic remission between patients with low and high trough levels, however only 38% had clinical scoring at the time of the drug level and 16% endoscopy/histology. Of the 41 patients with levels drawn reac- tively, 8 had follow up trough level after dose adjustement. All except one had improvement in drug level, however 88% remained subtherapeutic. There was endoscopic follow up in 19 of 41 patients with levels drawn reactively; 70% had low levels and 53% had endoscopic improvement or remained in remission/mild disease. Conclusions: The majority of patients who had drug levels checked either reac- tively or proactively had subtherapeutic levels. Low drug levels were associated with higher infammatory markers. Although there was no diference in endo- scopic remission seen between patients with low and high trough levels, over half of patients who had levels checked reactively and endoscopic follow up, improved or remained in remission/mild disease activity. This suggests that there may be utility to TDM to optimize drug levels and improve outcomes in patients on UST therapy. Downloaded from https://academic.oup.com/ibdjournal/article/25/Supplement_1/S25/5308179 by guest on 15 December 2022