T1DM is caused by the incompletely understood immune mediated destruction of the only cells capable of secreting insulin, the pancreatic beta cells. It has been known for some time that most individuals with new onset T1DM have residual beta cell function at the time of their diagnosis, and that those beta cells remain functional for months, even years following diagnosis. Further, it is now recognized that preserving even marginal beta cell function has a beneficial effect on a patient's ability to safely control his/her blood sugar. We also know that instituting immunosuppressive therapy at diagnosis preserves beta cell function, at least temporarily. Nevertheless, immunosupressive therapy is typically not instituted when T1DM is first diagnosed, and for two good reasons. One, the beta cell protective effect appears to be only temporary in that treated patients eventually require insulin replacement therapy. Two, the available immunosuppressive agents are toxic: they impair normal beta cell function, and they have other undesired toxicities including renal toxicity for a sizeable minority. The latter effect has been especially worrisome because of T1DM's known intrinsic nephrotoxic effects. We have tested two alternative approaches to beta cell preservation in the Diabetes Branch: ? 1) Administration of anti-CD3 antibodies (trial completed): In 1994, it was first reported that normal blood sugar control was restored in non-obese diabetic (NOD) mice with recently diagnosed diabetes by administering an anti-mouse CD3 antibody. While the exact mechanism underlying the remarkable effect remains incompletely explored, we initially reproduced the reported studies, then also studied whether the anti-CD3 could similarly reverse disease in a rat insulin promoter-CD80 (RIP-CD80) transgenic mouse model we developed. Contrary to the experience with non-obese diabetic (NOD) mice, anti-CD3 treatment only delayed disease onset in the RIP-CD80 transgenic mouse model. Nevertheless, since the first report that anti-CD3 therapy could restore euglycemia to NOD mice with recent onset diabetes and in so doing, block ongoing immune-mediated beta cell destruction, investigators have wished to test anti-human CD3 antibody in a clinical trial of patients with recent onset T1DM. Then available anti-human CD3 antibody was associated with significant toxicity (e.g. marked lymphopenia, a cytokine release syndrome, and immune responses against the mouse anti-human antibody) however the antibody was subsequently modified to limit toxicity. In 1999, Dr. Kevan Herold at Columbia University initiated a controlled clinical trial testing whether the modified OKT3 antibody (called OKT3 gamma, ala-ala) could be safely administered to patients with new onset T1DM (diagnosed within 6 weeks), and to test whether the agent preserved beta cell function. We applied for compassionate exemption permission to enter patients into the protocol. Two patients served as controls and two patients received the agent at the NIH clinical center. Neither NIH patient treated with the agent experienced any serious toxicity from the drug (except myalgias, temporary fever, and expected transient decreases in T cell counts). Twelve month follow-up data (patients and controls) as well as 24 month follow-up data (natural history of controls) have been published. ? 2) Administration of oral interferon-alpha (study completed, analysis pending): In 1998, Brod et al reported that oral interferon-alpha administered to non-obese diabetic (NOD) mice reduced insulitis and prevented the onset of diabetes. Further, they reported that the adoptively transfering unstimulated splenocytes from interferon fed donors suppressed spontaneous diabetes in recipient animals. Consistent with this finding, spleen cells from treated animals when compared with controls produced more interferon-gamma and interleukins 4 and 10. Subsequently, ten patients with newonset type 1 diabetes were included in an open label pilot study at the University of Texas in Houston, in which more than the expected number of individuals remained in the so called 'honeymoon phase' (period of time with continued endogenous insulin production). The exact mechanisms underlying the apparent efficacy of the ingested alpha interferon remains unknown. We initiated a multi-center, controlled, double-blinded clinical trial in which patients with recent onset (diagnosed less than 6 weeks) T1DM are randomized to two different doses of oral interferon-alpha (5,000 or 30,000 Units per day) or placebo. We enrolled 128 patients between the ages of 3 and 25 years in all 5 participating centers (NIH, Bethesda; University of Texas, Houston and Dallas; Children's Hospital St. Paul, MN, Kansas City, Kansas). We monitored beta cell secretory capacity, metabolic control, insulin requirements, and frequency and severity of hypoglycemic events. Patients showed excellent compliance with the study medication and follow-up testing, returning to the NIH for the required mixed meal studies performed every 3 months for 1 year. Four serious adverse events (all judged to be likly unrelated to study drug) have been reported. One child was diagnosed with a malignant brain tumor (NIH). The second child was hospitalized with asthma (Houston) and another youngster developed staphylococcus septicemia (NIH). The fourth child was hospitalized with dehydration. Data collection has been completed and final statistical analyses are underway.
