Kevan Herold, Yale Immunology Department, multi-center disease onset studies, T cell depletion
by Adam Burrack, PhD
Readers familiar with this blog series will be aware that I have focused on T cells of the immune system, which appear to be the “final effectors” of the autoimmune response leading to development of type 1 diabetes (T1D). Matthias von Herrath studies CD8 T cells and their activation in a mouse model of virus-induced beta cell destruction, Pere Santamaria studies a specific population of auto antigen-specific CD8 T cells that play a role in diabetes onset. Katie Haskins studies beta cell secretory granule proteins that are targeted by different autoreactive CD4 T cells in NOD mice, several of which are targets of CD4 T cells in the human disease as well. Jeff Bluestone studies the biology of T cell activation in general, and is currently investigating the potential utility of down-regulatory CD4 T cells to delay or stop new-onset diabetes in an on-going clinical trial.
Today I will describe the wide-ranging research of one of the leaders in immune-directed therapy – and specifically in T cell depletion-based clinical trials – for the treatment of new-onset T1D. Kevan Herold has been a long-time member of the immunology department at Yale University and has worked extensively with others in the field in the area of T cell inhibition and depletion to stop new-onset T1D.
Translating research outcomes from single centers, with 5-15 NOD mice per experimental group, to large multi-center clinical trials, with dozens if not hundreds of humans per group, is not always straight-forward. In fact, there are more than 140 published methods to prevent or cure new-onset T1D in NOD mice, none of which have translated to success in clinical trials in human subjects. Dr Herold is one of several Principle Investigators (PIs) in a working group derived from the Immune Tolerance Network which has decided to test protocols from high-impact publications across several different NOD mouse colonies simultaneously. Other PIs in this group include Mark Atkinson at the University of Florida, Matthias von Herrath in San Diego, and Ron Gill at the University of Colorado – all top-tier, well-regarded researchers who have access to separate NOD mouse colonies in different regions of the country.
This group tested CD20-based depletion of B cells combined with oral insulin treatment, two published methods to reverse new-onset diabetes in NOD mice. In this multi-center study across hundreds of NOD mice, neither therapy alone, nor the combination, reversed new-onset diabetes. This lack-of-benefit result may be surprising to some readers, but highlights what many consider a fundamental flaw in basic research: publication bias. Exciting results are often taken through to clinical trials without comprehensive proof of efficacy from multi-center animal studies. In addition, some exciting studies cannot be replicated in the same mouse model, which the same experimental protocol, at different research institutions. The NOD mouse is a great model of type 1 diabetes – the best animal model we have – but to base a clinical trial off of one manuscript using less than 50 mice may not be the best approach. Small-scale NOD mouse studies can identify potentially exciting candidate therapies, but what the immune tolerance network advocates for is an intermediate step of validation in a multi-center study (in the same model, with the same protocol) prior to taking the time and effort to conduct a clinical trial. Unfortunately, academic science is built around the publication of novel results in new model systems throughout an individual researcher’s career, which does not lend itself to the process of (a) discovery, then (b) validation, followed by (c) scaling, followed by (d) application in a clinical trial.
The observed lack of long-term benefit from B cell depletion (measured by maintenance of C-peptide levels, or in this case the lack of maintenance of C-peptide levels over time following treatment) shown in this multi-center NOD mouse study suggests that the clinical trial in which this reagent was applied may have been based on imperfect data. This new consortium aims to avoid the effort wasted in clinical trials with imperfect methods by conducting more thorough animal studies up front so that clinical trial reagents have a better chance of success.
Toward the end of developing better therapies to use in new-onset cessation clinical trials, Dr Herold has been at the forefront of T cell depletion studies targeting the CD3 molecule. In a recent publication describing results from a 2-year follow-up study, Dr Herold and others described an increase of C-peptide levels in a sub-group of patients treated with CD3-depleting antibodies (which target T cells in general and inhibit T cell activation). Subjects who responded better to this therapy – maintained a higher level of C-peptide production – were characterized by better blood sugar level control at beginning of the study (lower HbA1c – below 7.0%) and less insulin use per body weight (lower insulin dose per kilogram of body weight per day). The deeper question of why some subjects respond to therapy with better preservation of C-peptide production than other subjects is an open question that many researchers in the field are keen to understand.
Toward the end of developing an even better mouse model of T1D to study the response of human patient-derived T cells to beta cells-derived peptides, Dr Herold is part of a group which has derived a ‘humanized’ mouse model of T1D. Using this type of mouse model, researchers can implant human-derived islets (from cadavers or iPS-derived sources). In the absence of a mouse immune system these insulin-producing cells will not be destroyed. Researchers can then transplant T cells derived from human T1D patients to study the response of autoreactive, islet-specific T cells to insulin-producing beta cells. This combination of human islets and human autoreactive T cells affords researchers the opportunity to study human autoimmune disease recurrence – outside the context of a human patient – and to develop therapies to stop this T cell response.
Finally, the Herold lab has developed a sensitive test to measure beta cell death in peripheral blood in the circulation. Dr Herold’s group has developed a test to measure changes in insulin gene DNA associated with beta cell damage, indicative of on-going autoimmunity. This test provides an alternative to measuring auto antibody levels or trying to detect rare autoreactive T cells in the peripheral blood of subjects at-risk of developing T1D.
Is sum, the Herold lab is working on the cutting edge of several areas related to T1D, and is collaborating with groups throughout the country on developing better pre-clinical models of T1D as well as testing T cell-directed therapies more comprehensively before conducting clinical trials in human subjects. We at YOUglycemia – and many others – applaud the efforts of Dr Herold and colleagues.