Introduction

Since its introduction as a tuberculosis vaccine in 1921, bacillus Calmette-Guérin (BCG) continues to thrive as new uses emerge in disparate fields of immunology. BCG contains the avirulent tuberculosis strain Mycobacterium bovis, and its main appeal for the potential purpose of treating autoimmunity is its induction of the cytokine tumor necrosis factor (TNF). This book is the work of participants in a conference, “BCG and TNF Signaling in Autoimmunity,” which was held in London, England on October 19, 2013. The conference was held with three purposes in mind: to bring together for the first time researchers working on BCG and TNF for the treatment and prevention of human autoimmune diseases; to advance research by sharing animal and human findings and best practices; and to discuss how ongoing collaboration can facilitate clinical trials. Many of the investigators in attendance had never met before.

Dr. Bhagirath Singh of the University of Western Ontario began the meeting, speaking about the prevention of type 1 diabetes in rodent models by immunotherapy with mycobacterial adjuvants. Dr. Singh has performed abundant work on the prevention and treatment of type 1 diabetes in two animal models—the non-obese diabetic (NOD) mouse and the biobreeding (BB) rat. His research has shown that immunotherapy using BCG or complete Freund’s adjuvant (CFA) reverses insulitis and increases insulin production if given to pre-diabetic mice or even diabetic mice with new-onset autoimmunity. Some of the possible mechanisms include the induction of innate immune cells and regulatory T cells, as well as downregulation or death of pathogenic T cells. The data in support of BCG’s synergy with the host’s innate immune response of TNF induction were presented from many angles. A review of Dr. Singh’s groundbreaking studies over the past 15–20 years sheds historical light on the utility of BCG for type 1 diabetes in rodents.

Dr. Graham Rook of University College London is one of the pioneers of the concept that the infectious environment protects people from autoimmunity. The developed countries have witnessed huge increases in the prevalence of a wide range of chronic inflammatory disorders, including allergies and autoimmune diseases. The increases in these diseases occur against a backdrop of economic development and urbanization in which direct exposure to infectious agents has receded. Exposure to infectious diseases (e.g., tuberculosis) has immunoregulatory roles through the induction of TNF and the upregulation of regulatory T cells. Dr. Rook views “old friends” as naturally occurring organisms that have co-evolved with humans and led to long-term health benefits. These organisms have almost been eliminated from the urban environment. Obviously, the ongoing global trials with BCG are one such attempt to reintroduce infectious agents that might allow better immune system development.

Dr. Eugen Feist of Charité University of Medicine in Berlin discovered that human Sjögren’s syndrome is associated with a genetic defect in the immunoproteasomes. The defect traces to a deficiency in one of the proteins (LMP2) that forms a catalytic subunit of the proteasome. As a consequence, defective proteasomes are thwarted in two key roles: processing of the transcription factor NF-κB and processing of proteins for T cell education. Intact proteasomes are needed to liberate NF-κB from its chaperone molecule in the cytoplasm. Once liberated by TNF stimulation, NF-κB normally migrates to the nucleus to transcribe pro-survival genes. The defective proteasome leaves autoreactive T cells, but not healthy T cells, selectively vulnerable to TNF-induced death, as seen in mouse models of Sjögren’s syndrome. This provides the rationale for future testing of BCG in human clinical trials of Sjögren’s syndrome. Following the discovery that diabetes- and Sjögren’s-affected mice have proteasome defects, Sjögren’s syndrome was the first human autoimmune disease in which the protein defect (leading to defective proteasome function for T cell education and altered T cells with NF-κB defects) was identified. Over the last 15 years, these data have buttressed similar autoimmune animal data, and now broader human autoimmune diseases show proteasome defects.

Dr. Mehmet Karaci of Turkey’s Bülent Ecevit University performed a case-control epidemiological study on children with type 1 diabetes. The study found that children with at least two BCG vaccinations, especially with the first vaccination in early infancy, was protective against the development of type 1 diabetes. In contrast, children with 0–1 vaccinations developed the disease at the same rate as children in the general pediatric population. There was no relationship between vaccination date and age at diagnosis. This research builds on animal studies showing that CFA treatment of NOD mice prevents onset of type 1 diabetes, especially multi-dosing. Dr. Karaci’s data shows that multiple doses of BCG, with the first dose in the neonatal period, are most protective. The findings align with animal data on the critical time period for infectious disease exposures that trigger TNF (and even with recent data showing that latent maternal tuberculosis infection may influence fetal immune development) and suggest the potential of BCG for type 1 diabetes prevention.

Dr. Giovanni Ristori, an expert on multiple sclerosis (MS), opened his talk with the historical point that the use of BCG or immune adjuvants in animal models of MS dates back to the 1950s. His presentation then turned to his present-day human clinical trial using BCG in this disease. Dr. Ristori and his colleagues at the Sapienza University of Rome have conducted the most advanced clinical trials of BCG in patients with autoimmunity, in this case with MS. In a placebo-controlled trial, they found that BCG prevented progression of MS in patients with a single clinical episode suggestive of MS and a baseline gadolinium (Gd)-enhanced MRI scan supporting a diagnosis of MS. At the six month follow-up, 45.5% of vaccinated subjects versus 75% of placebo subjects developed one or more new Gd-enhanced lesions. No adverse events were reported. At five years, more than half of BCG recipients remained relapse-free versus 30% of placebo subjects. This Phase II BCG clinical trial demonstrates the feasibility and safety of administering BCG after the first demyelinating episode. In addition to launching a Phase III trial, Dr. Ristori and colleagues are eager to explore the mechanisms underlying the long-term beneficial effects of BCG vaccination in MS patients.

I was pleased to speak about my research group’s findings from a Phase I clinical trial that evaluated multi-dosing of BCG for the treatment of type 1 diabetes. Our early work showed that TNF-inducers like BCG and CFA can help to reverse type 1 diabetes (including the remarkable regeneration of insulin-producing cells of the pancreas) even in NOD mice with advanced disease. These data came after years of studying the biochemical mechanism by which this occurs: through TNF and NF-κB signaling errors in mouse and human diabetes and other autoimmune diseases.

Our findings in the mouse model led us to introduce BCG for treatment of advanced type 1 diabetes in humans in a randomized, placebo-controlled, proof-of-concept clinical trial. We found that multi-dose BCG in patients with advanced disease (mean duration of disease: 15 years) had the following salutary effects: a large increase in dead insulin-autoreactive T cells entering the circulation; induction of regulatory T cells, a rare subtype of T lymphocyte that helps maintain tolerance to self-antigens; and, most importantly, a transient rise in production of C-peptide (a marker of insulin production). This was the first demonstration of a resurgence of β-islet cell function in what was long considered to be a dead pancreas. It was also the first evidence that the pathogenic CD8 T cells could be killed directly.

Following these main presentations, Dr. Paul Burn of the University of South Dakota’s Sanford School of Medicine discussed some of the difficulties that have hampered success in type 1 diabetes clinical trials. Dr. Daniel Leffler of Beth Israel Deaconess Medical Center spoke to the advantages of testing novel therapies in celiac disease. Because celiac disease is often considered a more benign autoimmune disease than some of the life-threatening autoimmune diseases like type 1 diabetes or multiple sclerosis, it makes a poor candidate for trials using heavy-duty immunosuppressive drugs. Since the BCG vaccine is safe and has an impeccable safety record, it may be a plausible clinical trial drug candidate for interventional trials where safety is of utmost importance.

Dr. Tor Paaske Utheim, an ophthalmologist from Oslo University Hospital, spoke about the compelling rationale for studying BCG for the treatment of Sjögren’s syndrome, which features an autoimmune attack against exocrine glands (with hallmark symptoms of dry eye and dry mouth) and systemic manifestations including severe fatigue. Dry eye is not trivial in nature, for it can lead to infections, corneal abrasions, and, potentially, blindness. The ability to study dry eye with relative ease is one of Sjögren’s foremost advantages for future clinical trials with BCG. In addition, since both the Sjögren’s-affected NOD mouse and humans with Sjögren’s syndrome have underlying proteasome defects that predict drug efficacy with BCG/TNF induction, Sjögren’s is a prime candidate for human clinical trials.

In conclusion,...