Prometheus Biosciences (United States)

Lay Summary Vedolizumab clearance (CL) outperforms serum trough concentrations in predicting therapeutic outcomes in Crohn’s disease. Higher CL was significantly associated with reduced remission rates, supporting CL as a pharmacokinetic marker for optimizing vedolizumab therapy.

Abstract Background Individual blood markers have been associated with future risk of Crohn's disease (CD). However, there is a need to understand which combination of biomarkers will be most predictive to facilitate CD prevention trial recruitment. We aimed to develop and validate a blood-based integrative risk score for CD development. Methods We analyzed data from 200 patients with CD and 100 age, sex, and race-matched healthy controls (HC) from PREDICTS, a nested case-control study of US active-component service members (ACSM). Longitudinal serum samples were collected at four time points up to 6+ years prior to diagnosis. Anti-microbial antibodies (Prometheus), proteomic markers (Olink inflammation panel), and anti-granulocyte macrophage-colony stimulating factor autoantibodies (anti-GM-CSF) were assayed in each sample. Participants were randomly divided into equal sized training and testing sets. Time-varying trajectories of marker abundance were estimated for each biomarker. Logistic regression modeled disease status as a function of each marker for different time points and multivariate modeling was performed via logistic lasso regression. A risk score to predict CD onset within 2 years was developed using penalized mixed-effects logistic regression. Prediction models were fit on the testing set and predictive performance evaluated via receiver operating characteristic (ROC) curves and area under the curve (AUC). Results The AUCs of individual markers were dynamic over time, with some having stable predictive capacity and others increasing or decreasing closer to time of diagnosis (Figure 1A). The AUCs for a model combining antibodies (Prometheus and anti-GM-CSF) and proteins were consistently above 0.8 starting 4 years prior to diagnosis, primarily driven by proteomic markers (Figure 1B). An integrative model predicting CD onset within 2 years incorporated 10 biomarkers and significantly associated with CD onset (Figure 1C). The model resulted in an AUC of 0.87 with a specificity of 99% and positive predictive value of 84% at a classification threshold of 0.7 (Figure 1D). Stratifying the integrative model scores into quartiles revealed a clear gradient in CD incidence within 2 years, increasing from 2% in the first quartile to 57.7% in the fourth quartile (Figure 1E). The relative risk (RR) for developing CD among ACSM in the top quartile, compared to the lower quartiles, was 10.4 (95% CI 6.9, 15.6). Conclusion An integrative blood-based model combining anti-microbial antibodies and proteins predicts onset of CD. Serologic and proteomic markers have dynamic changes years before diagnosis, highlighting the varying preclinical phases of CD and how leveraging biomarker dynamics can predict time to diagnosis. Figure 1

Granulomatosis with polyangiitis (GPA) is a chronic relapsing systemic autoimmune vasculitis. Current treatment of GPA is unsatisfactory, as it relies on strong immunosuppressive regimens, with either CYC or rituximab, which reduce the immunogenicity of several vaccines and are risk factors for a severe form of COVID-19. This emphasizes the need to identify new drug targets and to develop treatment strategies with less harmful side effects. Since CD4+ effector memory T cells (TEM) play a key role in the pathogenesis of GPA, we aimed in this study to modulate CD4+TEM cell activity via Kv1.3 blockade using the specific peptide inhibiter, ShK-186. Peripheral blood samples from 27 patients with GPA in remission and 16 age- and sex-matched healthy controls (HCs) were pre-incubated in vitro in the presence or absence of ShK-186, followed by stimulation with phorbol myristate acetate, calcium ionophore and brefeldin-A. The effect of ShK-186 on the cytokine production (IFNγ, TNFα, IL-4, IL-17, IL-21) within total and subsets of CD4+ T helper (CD4+TH) cells were assessed using flow cytometry. ShK-186 reduced the expression level of IFNγ, TNFα, IL-4, IL-17 and IL-21 in CD4+TH cells from patients with GPA in vitro. Further analysis performed on sorted CD4+T cell subsets, revealed that ShK-186 predominantly inhibited the cytokine production of CD4+TEM cells. ShK-186 treatment reduced the production of the pro-inflammatory cytokines to the level seen in CD4+ TH cells from HCs. Modulation of cellular effector function by ShK-186 may constitute a novel treatment strategy for GPA with high specificity and less harmful side effects.

Abstract Background Infliximab (IFX) exposure is established predictive factor of pharmacokinetic (PK) origin in inflammatory bowel disease, and expert consensus is to achieve adequate exposure during induction (1). We evaluated the performance of a Bayesian PK tool in IBD patients starting IFX. Methods Trough IFX collected immediately before the third and fourth infusion (corresponding to week 6 and 14 on a standard dosing schedule, respectively) were evaluated from 307 IBD patients starting IFX and enrolled in three cohorts of IBD. The first cohort patients was adult CD enrolled in the TAILORIX trial (n=101, mean age [SEM]: 35.0±1.4 years, 60% female) (2), a second cohort was from the PRECISION trial (n=158, mean age 15.8±0.5, 75% CD, 46% female) (3) and a third cohort consisted of pediatric patients (n=45, mean age 13.2±0.5, 71% CD, 40% female). Forecasted IFX concentration at the 4th infusion were estimated using serum IFX, antibodies to IFX and albumin determined at the third infusion using population PK calculator with Bayesian prior. The outcome variable was a CRP-based clinical remission during follow-up with CRP levels below 3 mg/L in the absence of symptoms (corresponding to remission). Sustained CRP-based clinical remission status over the maintenance period was also calculated. Statistics consisted of Kaplan Meier analysis with calculation of Hazard ratio (HR), and logistic regression. Results IFX concentration above 15 µg/mL before the third infusion had shorter time to CRP-based clinical remission as compared to IBD with IFX concentration below 15 µg/mL (mean followup: 182±11 days). Measured IFX concentrations above 10 µg/mL before the 4th infusion were also associated with a higher rate and shorter time to remission (p<0.01). Similar results were observed with forecasted concentration above 10 µg/mL at the 4th infusion (p<0.01). In the presence of IFX concentration above 15 µg/mL there was a 2.5-fold higher likelihood of sustained CRP based clinical remission status during maintenance as compared to IFX concentrations below 15 µg/mL (p<0.01). Forecasted IFX level above 10 µg/mL at 4th infusion associated with 3.9-fold higher likelihood of CRP-based clinical remission. (p<0.01). Conclusion These data further support that IFX concentrations above recommended cutoffs are best to predict enhanced disease control. Patients presenting with low forecasted exposure may benefit from dose intensification sooner than the scheduled standard infusion.

BackgroundThe aim of this study was to assess the relationship between serum vedolizumab (VDZ) concentrations and antibodies to VDZ (ATV) in a large cohort of patients with inflammatory bowel diseases. Furthermore, we evaluated the association between serum VDZ concentrations and a novel serum-based biomarker panel designated as the endoscopic healing index (EHI), developed and validated for identifying mucosal inflammation in patients with Crohn’s disease (CD).MethodsRetrospective study where results from patient samples submitted to a commercial clinical laboratory were included. Serum VDZ and ATV levels were analyzed using a drug-tolerant assay. In CD patients for whom both VDZ and EHI were available, VDZ concentrations were correlated with EHI. serum VDZ threshold analysis was performed using ROC curves, and the serum VDZ concentrations that best differentiated EHI < 20 (previously associated with endoscopic remission) were chosen.ResultsA total of 9356 patients were included in the VDZ/ATV analysis. Detectable ATV was observed in 2.9% patients with significantly lower serum VDZ concentrations when compared to those with undetectable ATV [3.9 µg/mL (0–9.0) vs. 11.3 µg/mL (5.9–20.6), p < 0.0001]. Of the patients with serum VDZ result, 287 patients had a concomitant EHI test. An inverse correlation was observed between VDZ concentration and EHI (rho = − 0.20, p < 0.001). A serum VDZ concentration ≥ 15.7 µg/ml was best correlated wi th an EHI < 20 [AUROC: 0.67 (95% CI 0.57–0.77)].ConclusionsIncidence of ATVs was low, but significantly associated with lower VDZ levels. A serum VDZ concentration threshold of ≥ 15.7 µg/ml was associated with endoscopic remission as defined by an EHI < 20.

INTRODUCTION: Therapeutic drug monitoring (TDM) of Infliximab remains an important clinical decision tool to improve outcomes in Inflammatory bowel diseases (IBD). However, one limitation of TDM in clinical gastroenterology practice is the lack of standardization across available technology platforms. Recently, the 1st international standard for infliximab (IFX) bioactivity was characterized and produced by World Health Organization (WHO). Our objective was to evaluate the performance characteristics of the Homogenous Mobility Shift assay (HMSA) against WHO calibration. METHODS: All IFX testing was conducted in a clinical laboratory accredited by the American College of Pathologists (Prometheus Biosciences, San Diego) using liquid chromatography size exclusion HMSA calibrated in house using IFX standards purchased independently from WHO. WHO Standard (NIBSC code: 16/170) was purchased as lyophilized powder, re-suspended to a final concentration of 50 µg/ml in normal healthy serum and stored at -80°C. IFX concentrations measured using in-house and WHO calibration methods were compared using de-identified specimens submitted for IFX testing. Data were analyzed using Deming’s regression and linear mixed effect models for repeated measures. IFX levels in relation to dosing intervals from patient specimens presenting with ICD-10 specific IBD codes were also evaluated. RESULTS: Between run (n = 6) precision was 9.5% (median CV; range 3.9–19.4%) and 8.6% (range: 5.2–17.8%) for 28 IFX specimens measured using in-house and WHO calibration, respectively. As presented in Figure 1, IFX levels measured using in-house calibration method (mean 13.8 ± 0.9 µg/ml) were comparable to those recovered from WHO calibration method (mean 13.9 ± 0.9 µg/ml) (Deming’s slope 1.00 CI 95%: 0.97–1.03). Linear mixed effect models also indicated comparable IFX levels between calibrations (slope = 0.986 ± 0.012; intercept = 0.11 ± 0.22; R2 = 0.97). Table 1 highlights IFX exposure levels in a total of 19,013 specimens submitted with dosing information between 2015 and 2020. A 5mg/Kg every 8 weeks dosing interval associated with 57% specimens above 5 µg/ml. Shorter intervals resulted in higher exposure. CONCLUSION: Our data indicate that IFX levels measured by HMSA in the reference laboratory yield comparable estimates to those from WHO. Calibration with WHO standards may facilitate harmonization of TDM in clinical gastroenterology practice.Figure 1Table 1