PerspectiveIMMUNOTHERAPY

Treg cells—the next frontier of cell therapy

See allHide authors and affiliations

Science  12 Oct 2018:
Vol. 362, Issue 6411, pp. 154-155
DOI: 10.1126/science.aau2688

In the past decade, effector T cells, engineered to express highly specific chimeric antigen receptors (CARs) or antigen-specific T cell receptors (TCRs) that recognize tumor antigens, have been shown to be highly effective adoptive cell therapies (ACTs) that have revolutionized certain cancer treatments. In 2017, the first two CAR–T cell therapies were approved by the U.S. Food and Drug Administration for the treatment of various CD19+ B cell lymphomas, and extensive clinical trials are now under way to expand the therapies to multiple solid tumor settings (1, 2). Here, we speculate on the next generation of immune cell therapies for non-cancer diseases. Specifically, we highlight the progress toward developing a new class of ACT using regulatory T cells (Tregs) to treat autoimmune diseases—including type 1 diabetes (T1D), rheumatoid arthritis, inflammatory bowel disease (IBD), graft-versus-host disease (GvHD) that can occur after bone marrow transplantation, and organ transplant rejection—and potentially to treat nonimmune diseases such as Alzheimer's disease, Parkinson's disease, heart disease, and type 2 diabetes (3).

Although Tregs constitute only 1 to 2% of peripheral blood lymphocytes, they are the master controllers of self-tolerance (whereby immune cells can recognize foreign substances and ignore self tissues, thus avoiding autoimmunity), tissue inflammation, and long-term immune homeostasis (4). In the most severe Treg deficiency, IPEX (immunodysregulation, polyendocrinopathy, enteropathy, X-linked) syndrome, mutations in the key Treg lineage–specific transcription factor FOXP3 (forkhead box P3) lead to Treg defects resulting in lethal multi-organ inflammation and autoimmunity (5). Importantly, reduced numbers and/or function of Tregs have been implicated in the pathology of many common autoimmune diseases. Indeed, in some disease settings, defects in Treg functions or reduced Treg growth and survival factors such as interleukin-2 (IL-2), coupled with a local tissue environment of high concentrations of inflammatory cytokines such as IL-6, IL-12, and IL-1, results in Treg instability, uncontrolled inflammation, and excessive tissue damage (6). Thus, a number of ongoing studies are repurposing approved drugs, such as Treg growth factors (for example, IL-2) and Treg-stabilizing factors (for example, rapamycin), to enhance Treg function as an approach to controlling a variety of autoimmune diseases, GvHD, and organ transplant rejection (7). These results underscore the opportunity to directly alter pathological immune responses by boosting Treg activity—which is also the goal of Treg-based ACT.

The first preclinical proof-of-concept studies to treat autoimmune diseases were performed using polyclonal Tregs, then termed suppressor cells, isolated on the basis of the expression of multiple cell surface markers including CD4, CD25, and CD62L (4). The success in these efforts and in subsequent studies using purified FOXP3+ Tregs has led to several clinical trials in a variety of disease settings including organ transplant rejection, GvHD, T1D, and autoimmune syndromes (8, 9). The cells are purified from the peripheral blood of each patient, grown ex vivo in the presence of antibodies to CD3 and CD28 accompanied by high-dose IL-2 to expand the highly enriched Treg population, and, after adequate characterization, adoptively transferred into patients. This polyclonal Treg ACT has been shown to be safe in phase 1 studies, and some studies have suggested clinical activity (8, 9). The first results from a phase 2 study in T1D are expected in 2019 (NCT02691247); other efforts to determine efficacy in GvHD are under way (for example, NCT01795573 and NCT01937468).

Moreover, it has become increasingly clear that nonimmune diseases such as cardiovascular diseases, obesity, type 2 diabetes, and degenerative diseases of muscle and brain are exacerbated by inflammation. Accumulating evidence in preclinical studies suggests that Tregs can quell inflammation and reduce morbidity in these diseases by contributing to tissue homeostasis and repair by producing the epidermal growth factor receptor (EGFR) ligand amphiregulin in damaged tissues including influenza-infected lungs, muscular dystrophy–affected muscle, nerve demyelination associated with multiple sclerosis (MS), and high-fat diet–induced obesity (10, 11). A recent clinical trial in amyotrophic lateral sclerosis reported that Treg ACT may reduce disease progression (12). In addition, Tregs have been shown to increase bone marrow engraftment, reduce immune responses to gene therapy, and facilitate wound healing (12).

So where will polyclonal Tregs be most effective? Preclinical data show that polyclonal Tregs are efficacious in controlling some diseases, such as lupus nephropathy and IBD, but far less so in others such as T1D and MS. For example, Tregs specific for a pancreatic islet antigen were 50 to 100 times more efficient in blocking T1D progression than polyclonal Tregs when transferred into mice with autoimmune diabetes (which is similar to T1D). Because Tregs have a TCR repertoire that is skewed toward self-antigens, we speculate that large tissues such as skin and gut with abundant and more diverse antigens may be recognized by a larger fraction of TCRs in the Treg repertoire than smaller tissues such as pancreatic islets. Thus, larger tissues can activate sufficient Tregs in a polyclonal population to achieve therapeutic effect, whereas smaller tissues will require antigen-specific Tregs for ACT.

The use of polyclonal Tregs can potentially suppress protective immunity against tumors and infectious diseases by transferring a large number of Tregs of broad undefined specificity. Thus, developing antigen-specific Treg therapy will likely provide a more effective and safer alternative. Importantly, in multiple preclinical models, Tregs that are selected ex vivo to be specific for limited tissue antigens or even single antigens can mediate localized dominant bystander suppression of effector T cells of diverse specificities. This concept of enriching target tissue-specific Tregs has already been translated into early-phase clinical trials. Tregs have been isolated from organ transplant recipients and stimulated with donor organ–derived antigen-presenting cells (APCs) to selectively expand donor-specific Tregs; these are more effective at suppressing organ rejection than polyclonal Tregs in humanized mouse transplant models (13). These alloantigen-reactive Tregs are being tested to prevent rejection in several clinical trials with the goal of reducing or even fully withdrawing immunosuppressive drugs (for example, NCT 02474199).

Applying selective expansion approaches to tissue antigens other than transplant antigens is far more challenging. This is because frequencies of alloantigen-reactive Tregs are as high as 10%, whereas frequencies of tissue antigen-specific Tregs are orders of magnitude lower. In addition, repeated stimulation of Tregs ex vivo can destabilize Tregs and lead to outgrowth of non-Tregs. Expression of engineered antigen-specific receptors (CARs and TCRs), a successful strategy in cancer immunotherapy, offers a solution to achieving antigen specificity of therapeutic Tregs. Alloantigen-specific CARs, reactive to human leukocyte antigen-A2 on donor tissue, can prevent allograft rejection in humanized mouse models (14). Similarly, Tregs engineered to express transgenic TCRs specific for autoantigens such as Factor IX in hemophilia, myelin oligodendrocyte glycoprotein in MS, and pancreatic islet–specific antigens in T1D have shown enhanced efficacy in preclinical models. Moreover, new approaches have been developed that allow Tregs to be “differentiated” to selectively control diseases mediated by subsets of T cells, including T helper 1 (TH1)–like Tregs to treat T1D and organ transplant rejection, TH2-like Tregs for asthma and allergy, and TH17-like Tregs to treat MS (15). A combination of antigen receptor engineering and differentiation approaches to direct Tregs to specific sites of disease activity will lead to more robust and effective treatments.

Sites of therapeutic Treg action

In the lymph node, where primary immunity occurs, Tregs can shut down proliferation and differentiation of pathogenic T cells. In inflamed tissues, Tregs can alter effector T cell expansion and activation, cytokine production, innate immune cell activation, chemoattraction of inflammatory cells, and epitope spreading.

GRAPHIC: VERONICA FALCONIERI HAYS/SCIENCE FROM J. A. BLUESTONE

In this regard, two features of Tregs are worth emphasizing. First, unlike cancer-killing T cells that must bind to target cells directly, in many cases there is no requirement for therapeutic Tregs to directly contact effector T cells to inhibit function. Treg-mediated immune suppression and tolerance typically occur through alteration of the local tissue microenvironment both in the draining lymph node and in the affected tissue (see the figure). Immune regulation by Tregs depends on antigen engagement normally provided by APCs. Once induced, these suppressive activities, carried out by soluble mediators and cell surface receptors, affect all cells and tissue in the vicinity in a paracrine fashion. Thus, it is possible to direct Tregs using engineered CARs and TCRs to any cells or even natural or synthetic multivalent ligands in the target tissue to achieve local immune suppression. Second, Tregs can mediate “infectious tolerance.” Tregs not only dampen inflammation but also create a tissue microenvironment conducive to the emergence of additional immune-suppressive populations, including Tregs with additional specificities, myeloid-derived suppressor cells, and other subsets of suppressive T cells, such as type 1 regulatory T (Tr1) cells. Through this infectious spread of tolerance, the effects of therapeutic Tregs are amplified and prolonged, even if the adoptively transferred Treg cells do not survive indefinitely.

With all the advantages of a living drug that can traffic to target tissue, automatically carry out therapeutic actions tailored to the tissue microenvironment, expand or contract in number depending on the therapeutic demands, and persist long-term for durable recovery, an inherent challenge of Treg cell therapy is to ensure their lineage stability and prevent antigen-specific Tregs from turning into effector T cells that are capable of inflicting tissue injury (6). This risk may be mitigated by engineering Treg cells to include suicide genes, to secrete autocrine IL-2, deleting receptors for proinflammatory cytokines, and stabilizing FOXP3 expression. It is also important to recognize that Treg ACT alone may not be sufficient to induce tolerance because in many settings the preexisting effector T cells may be refractory to Treg control or may outnumber the therapeutic Tregs. It is likely that durable remission will require a combination of drugs that reduce effectors and inflammation before administering antigen-specific Tregs, or perhaps Tregs may be genetically engineered to reduce inflammation, compromise effector T cells, and promote tolerance.

One of the challenges of ACT involves the cost and technical difficulty associated with personalized treatments. This challenge is being addressed for ACT in cancer treatments by developing universal T cells for off-the-shelf use. Using gene editing approaches, T cells from unrelated individuals are modified by deleting polymorphic human leukocyte antigens to protect the cells from being destroyed by the host immune system and by eliminating their endogenous TCRs to prevent nonspecific activities. The next few years will be critical in determining the efficacy and practicality of Treg ACT. However, if the explosion of cell therapy approaches in cancer is any indication, we believe that Tregs will be the next frontier in the development of antiinflammatory and tolerogenic therapies.

References and Notes

Acknowledgments: J.A.B. and Q.T. acknowledge support from the JDRF, Caladrius Biosciences, Becton-Dickinson, Juno Therapeutics, and Pfizer. J.A.B. and Q.T. have patents and pending patents on polyclonal and antigen-specific Tregs and consult for Third Rock Ventures and Juno in this area. J.A.B. serves on the board of Pfizer.

Navigate This Article