Anti-inflammatory Therapy With Anakinra in Newly Diagnosed Type 1 Diabetes

Type 1 diabetes mellitus (T1D) is caused by autoimmune and autoinflammatory destruction of the insulin-producing beta cells in the pancreatic islets of Langerhans. Historically, treatment for this condition has consisted of insulin replacement therapy and dietary modification. Recent studies have demonstrated the potential benefits of immunomodulatory therapy in patients with newly diagnosed T1D to prevent further immune-mediated damage. Thus far, there has been a paucity of completed research using agents that target pro-inflammatory cytokines dysregulated in T1D.

IL1B, the gene encoding the proinflammatory cytokine interleukin-1β (IL-1β) is significantly overexpressed in peripheral blood mononuclear cells (PBMC) in patients with newly diagnosed T1D compared to healthy controls. There is ample precedent to support the involvement of IL-1β in the pathogenesis of diabetes. In particular, incubation of human or animal islets or insulinoma cell lines with IL-1β inhibits insulin secretion and leads to apoptosis of beta cells.

Additionally, the IL-1 receptor antagonist protein, anakinra, improves glycemic control and insulin secretory capacity in patients with type 2 diabetes.However, no data have been published regarding efficacy of agents targeting IL-1β in modifying the course of the disease in patients with T1D. Anti-Interleukin-1 in Diabetes Action (AIDA) is a randomized, placebo controlled clinical trial of anakinra in 160 adults with newly diagnosed type 1 diabetes. This trial is currently recruiting subjects. Review of clinicaltrials.gov demonstrates two other planned trials of IL-1 agents in newly diagnosed T1D, one using canakinumab, a monoclonal antibody directed at IL-1β, and another using rilonacept, a cytokine trap targeting IL-1β. To assess the effectiveness of these drugs in future studies and in clinical practice, it will be valuable to identify biomarkers that allow us to monitor their therapeutic effects. However, to the best of our knowledge, the pattern of changes in gene expression induced by IL-1β in normal PBMC has not been systematically studied, making it difficult to identify candidate biomarkers for validation studies.

In this exploratory study, we aimed to determine which of the characteristic changes in gene expression from patients with newly diagnosed T1D are IL-1β-mediated, using both in vitro and in vivo approaches. We also determined the effect size of a short course of anakinra therapy on glycemic control, insulin dosing, and C-peptide area under the curve during mixed-meal tolerance testing (MMTT). Finally, we evaluated the tolerability of anakinra in children and adolescents with newly diagnosed T1D.

Methods In vitro studies To determine the effects of IL-1β on gene expression in peripheral blood mononuclear cells (PBMC), blood samples were collected from 7 healthy adult volunteers under an IRB-approved protocol.

Isolating serum. Blood was collected in EDTA tubes (BD, Franklin Lakes, NJ) and centrifuged at 2500 rpm for 15 minutes. The plasma layer was treated with topical thrombin (5000 U/ml, King Pharmaceuticals, Bristol, TN) equaling 5% total volume and incubated at 38° for 20 minutes. The resulting clot was removed from the serum and discarded.

Cell culture. PBMC were isolated from the cellular fraction by centrifugation using Lymphocyte Separation Medium (Mediatech, Manassas, VA) and washed with PBS (Mediatech, Manassas, VA). Cells were plated at 3 x 106 per well in 6 well plates in RPMI 1640 (Mediatech) supplemented with 10% fetal bovine serum (Atlanta Biologicals, Lawrenceville, GA), 10% autologous human serum, 5.5 mM glucose, 100 U penicillin, 100 µg/ml streptomycin, 50 µmol/l 2-mercaptoethanol and 5% HEPES. IL-1β (15 ng/ml final concentration, Abcam, Cambridge, MA), or 9.3 µg/ml S100b (Sigma, St. Louis, MO) were added to some wells, All experimental conditions were plated in triplicate. Cells were collected after 24 hours of incubation at 37°C in a 5% CO2 atmosphere.

Validation by RT-PCR. Total RNA was isolated using RNA-Stat 60 (Tel-Test, Friendswood, TX). The High Capacity cDNA Reverse Transcription Kit (Applied Biosystems, Carlsbad, CA) was used with 750 ng RNA to create cDNA. Real time PCR was performed on the Roche LightCycler 480 system using 5 ul of the cDNA sample, the Roche LightCycler Probes Master kit and human IL1B primers (Applied Biosystems). RNA was quantitated by delta Ct values using GADPH as the internal control (Applied Biosystems).

Microarray analysis. Triplicate samples for each experimental condition were pooled for further analysis. From 2-5 µg of total RNA, double-stranded cDNA containing the T7-dT(24) promoter sequence were generated using GeneChip® One-Cycle cDNA Synthesis Kits (Invitrogen, Santa Clara, CA). Synthesis of cRNA used 200ng of cDNA for in vitro transcription, amplification and labeling steps according to the manufacturer's instructions using the Illumina RNA amplification kit (Ambion Inc, Austin, TX). 1.5 µg of amplified biotin-labeled cRNA was subsequently hybridized to Illumina Human HT-12 v3.0 Beadchip microarrays according to standard protocols. at the Baylor Institute for Immunology Research, Dallas, TX.

Slides were scanned on an Illumina Beadstation 500 and data processed with Beadstudio software. For each chip, raw intensity data were normalized to the mean intensity of all measurements on that chip. Data were imported into Genespring GX11 (Agilent) for further analysis. Probe sets were selected if "Present" or "Marginal" in at least 50% of samples in any group. Principal component analysis was performed to detect outliers. Class comparisons were performed using t-tests after log transformation with the Type 1 error rate controlled using Benjamini-Hochberg false discovery rates (FDR).

In vivo studies Subjects. The study was approved by the Institutional Review Board of the University of Texas Southwestern Medical Center. Written informed consent was obtained from parents or legal guardians, and assent for participation was obtained from subjects aged 10 years and older. Patients at Children's Medical Center Dallas between ages 6 and 18 years with Type 1 diabetes within one week of diagnosis were eligible. Exclusion criteria included treatment with systemic or inhaled corticosteroids or any other immunomodulatory drug, active infection, history of mycobacterial disease, pregnancy or lactation, use of a live vaccine within 90 days of study enrollment, and severe comorbidities (such as chronic kidney disease, heart failure, or uncontrolled hypertension). Patients were excluded if it was unclear whether they had Type 1 or Type 2 diabetes.

Data collected from the medical charts included age, gender, race/ethnicity, weight, height, hemoglobin A1c (HbA1c) and beta-hydroxybutyrate at diagnosis. A blood sample was obtained at study entry; a portion was analyzed for screening labs including alanine aminotransferase (ALT), aspartate aminotransferase (AST), blood urea nitrogen (BUN), creatinine, complete blood count (CBC) with differential, and serum pregnancy test for all menstruating females and any female over age 10 years. Additionally, 20 mL was processed for microarray analysis.

Diabetes care At diagnosis, all subjects were placed on a basal-bolus insulin regimen with glargine and either lispro or aspart. For the duration of the study, insulin doses were adjusted per standard clinic protocol with target glucose of 80-140 mg/dL fasting and 80-180 mg/dL before meals. At each study visit, we recorded the subject's current insulin doses and weight to allow calculation of the total daily dose (units/kg/day).

Anakinra After study enrollment, all subjects started anakinra (Kineret; Amgen, Thousand Oaks, CA) as a subcutaneous daily injection. Subjects weighing more than 25 kg at the time of enrollment received 100 mg daily whereas those weighing 25 kg or less received 50 mg daily. Anakinra was continued for a total of 28 days with no dose adjustment.

Diabetes autoantibodies Per standard protocol, all patients were tested for antibodies to insulin, protein tyrosine phosphatase receptor type N (insulinoma-associated antigen (IA-2)), and glutamic acid decarboxylase (ARUP Laboratories, Salt Lake City, UT). The results of these studies were typically not available at the time of study enrollment and therefore were not used as criteria for study entry. However, we excluded patients with negative results for all three antibodies from further analysis.

Surveillance laboratory tests Upon completion of anakinra therapy (4-5 weeks after diagnosis), a blood sample was collected and sent for ALT, AST, BUN, creatinine, and CBC with differential. Per standard protocol, all subjects had point-of-care hemoglobin A1c (DCA Vantage Analyzer, Siemens Healthcare Diagnostics, Deerfield, IL) and capillary glucose tested at clinic visits 4-5 weeks after diagnosis, 4 months after diagnosis, and 7 months after diagnosis.

Adverse event monitoring During the 28 days of anakinra therapy, study personnel called subjects weekly to document frequency of hypoglycemia and presence of rash, injection site reactions (swelling, erythema, and pain at the site of anakinra administration), headaches, fevers, and any other adverse events. After the completion of anakinra therapy, patients were assessed for adverse events at each subsequent study visit.

Mixed meal tolerance tests (MMTTs) MMTTs were conducted at the University of Texas Southwestern Medical Center Clinical Translational Research Center (CTRC). Subjects underwent MMTTs essentially as described (8) at 3-4 weeks after diagnosis and again at 7 months after diagnosis. C-peptide analyses were performed in the laboratory of Dr. Philip Raskin (UT Southwestern Medical Center, Dallas, TX).

Microarray blood sample processing Microarray blood samples were collected in EDTA tubes (BD Vacutainer) at study enrollment, 3-4 weeks after diagnosis, and upon completion of anakinra therapy. PBMC were isolated and stored at -80°C within 4 hours of the blood draw. Cells were lysed in RLT lysis buffer containing β-mercaptoethanol (Qiagen, Valencia, CA). Total RNA was extracted using the RNeasy® Mini Kit according to the manufacturer-recommended protocol (Qiagen, Valencia, CA) and analyzed as described above.

Control groups Because subjects in the anakinra study were not randomized, we used two different pre-existing control groups. Subjects in control group A were previously enrolled at our institution under a separate, IRB-approved protocol in a randomized, controlled trial examining the effect of the initial choice of longer-acting insulin on the rate of loss of beta cell function. Subjects were enrolled during their hospitalization for diabetes diagnosis and randomized to receive either NPH or basal insulin (glargine or detemir). Diabetes care and routine clinic visits, schedule and procedures for mixed meal tolerance testing and microarray analysis were the same as for our anakinra-treated subjects.. Control group A includes all patients in that study randomized to receive basal insulin, except those with negative autoantibody results.

Subjects in control group B were identified in a chart review of all new diabetes diagnoses from April 2008 through November 2009 in patients aged 9-18 years (to match date range of recruitment and age of anakinra-treated subjects). Subjects were included in this control group if they had at least one positive autoantibody and were not enrolled in the anakinra study. We collected age, gender, race/ethnicity, weight, height, and beta-hydroxybutyrate at diagnosis. We also collected HbA1c and total daily insulin dose (expressed as units/kg/day) at diagnosis and for clinic visits at 1 month, 4 months, and 7 months after diagnosis.

A subset of 10 of these subjects consented to enroll in an IRB-approved study in which blood was drawn for microarray analysis at diagnosis and again 1 month after diagnosis. Microarray procedures were the same as described above for the anakinra-treated subjects.

Statistical analysis Descriptive statistics were compiled and all comparative analyses performed using SAS version 9.2 (Cary, NC). C-peptide area under the curve (AUC) was calculated for each MMTT using the trapezoidal method.. Insulin-dose adjusted A1c (IDAA1c) was calculated as outlined by Mortensen, et. al. (9), IDAA1c = hemoglobin A1C (percent) + [4 × insulin dose (units per kilogram per 24 h)]. Hemoglobin A1c and IDAA1c between groups were compared using Wilcoxon rank-sum tests. Since total daily insulin dose was normally distributed, between groups comparisons were performed by standard t-test.