CAR T Therapy Beyond Cancer: The Evolution of A Living Drug: Perspective
CAR T Therapy Beyond Cancer: The Evolution of A Living Drug: Perspective
CAR T Therapy Beyond Cancer: The Evolution of A Living Drug: Perspective
In the past decade, CAR T therapy has transformed the field of oncol- Fine-tuning these receptors has led to new generations of CARs that
ogy, treating previously incurable haematologic cancers1–3. Although eliminate inhibitory domains, introduce dominant-negative receptors
this therapy rose to prominence owing to its success in cancer, this or mutate co-stimulatory domains to improve efficacy13.
therapeutic rationale stretches back to its initial development to redi- The success of CAR T cells in contexts in which other therapies
rect T cells to treat human immunodeficiency virus (HIV)4–6. These failed suggests that cell and gene therapy is a new pillar of treatment
initial trials were unsuccessful in treating HIV but demonstrated the for cancer14. CAR T cells have many distinctive features in comparison
long-term persistence of engineered T cells in immunocompromised to traditional therapies. The intrinsic specificity of a CAR is unrivalled
patients. The subsequent application of this approach to refractory by current small-molecule approaches. CAR T cells rely on the cytotox-
B cell and plasma cell malignancies propelled the field of adoptive cell icity of a T cell, taking advantage of this highly efficient endogenous
therapy forward7–10. Since, there has been substantial effort to apply pathway. The additional benefit of a living drug is that a single CAR
this approach to other cancers11. Early results indicate that the next T cell can exponentially expand and kill hundreds, if not thousands,
advances may be in fields beyond cancer, in which CAR T cells could of target cells. Conversely, the pool of CAR T cells can contract when
find widespread application (Fig. 1). antigen is no longer present and remain on patrol for years15. Last, CAR
T cells have demonstrated clinical feasibility. With more than 15,000
patients dosed, we have yet to see any autologous products transform
A new pillar of therapy into malignant cells. This safety profile in a condition for which the dis-
The principle at the foundation of CAR T therapy is to couple the ease burden is high bodes well for application to other diseases with less
potency of a T cell with the specificity of an antibody to precisely kill target burden. Thus, the possibility to leverage the specificity, potency
diseased cells. The single-chain variable fragment confers specificity and clinical safety of CAR T cells to treat other diseases is compelling.
while the intracellular signalling domains activate T cell-mediated cyto-
toxicity (Fig. 2). Synthetic engineering allows for refinement of CARs
in each specific context. For example, first-generation CAR constructs Challenges facing CAR T therapy
were composed of a CD4 extracellular domain in combination with the Despite the clinical efficacy of CAR T therapy in blood cancers, serious
CD3ζ signalling domain4. These constructs lacked potency and spurred clinical complications can occur. High levels of CAR T cell expansion
second-generation CARs that include a co-stimulatory domain, such and target cell killing over a short time can result in cytokine release
as CD28 or 4-1BB. This improved the effector function and persistence, syndrome (CRS)16. In severe cases of CRS, patients may experience
and these CAR T cells are approved by the US Food and Drug Adminis- high-grade fevers, hypotension and even multi-organ failure. Another
tration (FDA) for the treatment of various cancers at present12 (Fig. 2). major impediment is ‘on-target, off tumour toxicity’17. This occurs when
1
Center for Cellular Immunotherapies, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA, USA. 2Department of Pathology and Laboratory Medicine, Perelman
School of Medicine at the University of Pennsylvania, Philadelphia, PA, USA. 3Parker Institute for Cancer Immunotherapy at University of Pennsylvania, Philadelphia, PA, USA. 4Cardiovascular
Institute, Department of Medicine, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA, USA. 5Department of Medicine, Perelman School of Medicine at the
University of Pennsylvania, Philadelphia, PA, USA. 6Department of Physiology and Institute for Diabetes, Obesity, and Metabolism, Perelman School of Medicine at the University of
Pennsylvania, Philadelphia, PA, USA. 7Institute for Regenerative Medicine, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA, USA. 8Department of Cell and
Developmental Biology, Perelman School of Medicine at the University of Pennsylvania, Philadelphia, PA, USA. ✉e-mail: [email protected]; [email protected]
the target antigen is present in vital tissues and can result in severe tox-
icity and even death18. Additionally, to facilitate the engraftment and
expansion of the engineered T cells, lymphodepleting chemotherapy Fig. 2 | CAR T cells are engineered to be precise and powerful killers. The
is administered before CAR T cell infusion. These chemotherapies are components of a CAR, including the variable light (VL) and variable heavy (VH)
genotoxic, leading to increased risk for cancer and other diseases19. chains that comprise the single-chain variable fragment (scFv), which is fused
Autologous CAR T cell manufacturing is an emerging technology, and to a co-stimulatory domain and CD3ζ. This synthetic receptor confers
the costs to engineer and generate T cells and treat patients are high15. antigen-specific cytotoxicity.
Hostile to CAR Large target cell burden Tumour Target cell accessible Disease cell;
T cell infiltration and high mutational burden microenvironment, to CAR T cell small target cell burden
hypoxic and and low mutational burden
nutrient depleted
Healthy cell
CAR T cell
Treg cell
Necrotic cell
Cancer cell
Fibroblast
Fig. 3 | Advantages of CAR T therapy in other diseases in comparison to a much smaller target burden, partial clearance of diseased cells can be
cancer. a, Cancer presents several substantial barriers to CAR T cell efficacy therapeutic, a lower mutational load, and are generally not localized in such an
including a large target burden, nearly total clearance of malignant cells, a high inhospitable environment.
mutational load and a hostile TME. b, By contrast, other diseases generally have
absent in most chronic diseases, and often the diseased tissue is tar- after CAR T cell infusion, the patient had increased myalgia and elevated
geted by the immune system physiologically. Thus, a synthetic immune serum creatinine kinase; however, this was followed by a sharp improve-
approach to aid in the clearance of the pathologic cells is both feasible, ment in physical function and normalization of creatinine kinase and
and quite possibly, synergistic with endogenous immune function32. other clinical and laboratory measurements over the 180 days post CAR
Hence, the notion of using CAR T cells to target diseases other than T cell infusion. Further imaging at 3 months post CAR T cell infusion
cancer is compelling. revealed resolution of myositis in quadriceps and hamstrings and an
amelioration of alveolitis. The patient did have mild CRS, which was
Autoimmune and inflammatory diseases. CAR T therapy in autoim- managed and resolved in 3 days. It is possible that the acute increases
mune disease is already showing signs of clinical efficacy. A case report in in myalgia and serum creatinine kinase are a result of CRS; however,
2021 repurposed approved CD19 CAR T cells to target B cells in a patient this needs to be explored in more detail. In this patient, once again
with systemic lupus erythematosus (SLE), a life-threatening autoim- B cells reappeared after a few months with no report of symptom recur-
mune disease33. In 2022, results from five patients were published34. rence. This is a single case and thus caution must be emphasized.
Remarkably, all patients showed expansion of the CAR T cells in vivo, This clinical work was built on earlier preclinical reports using CD19
rapid depletion of B cells, and a resolution of SLE symptoms and of CAR T cell products to treat autoimmune diseases in which B cells are
markers of end-organ damage. All five patients discontinued their the source of pathology. CD19 CAR T cells in mouse models of SLE
immunosuppressive drug regimen and were declared to be in drug-free could both prevent and treat the disease37,38. Other engineered immune
remission. Surprisingly, short-term follow-up revealed that naive B cells approaches include the use of a chimaeric autoantibody receptor. This
re-emerge a few months following CAR T cell infusion without return strategy fuses an epitope, rather than a single-chain variable fragment,
of disease symptoms. The administration of CAR T cells was well tol- to intracellular signalling domains, to selectively eliminate pathologic
erated, with only mild cases of CRS, consistent with the notion that a B cells that recognize the given epitope. This has shown promise in
lower target cell burden may reduce the severity of CRS. Large trials and mouse models of pemphigus vulgaris and myasthenia gravis39,40. A
long-term follow-up will be necessary to elucidate whether the reso- similar approach specifically targeted dysfunctional B cells in a mouse
lution of immunopathology is temporary or durable. These patients model of haemophilia41. These approaches allow for the selective deple-
received lymphodepletion and it is difficult to disentangle whether tion of only pathologic B cells rather than inducing B cell aplasia. Trials
this independently or coordinately with CAR T cells may have improved are underway to test the safety and efficacy of these approaches in
symptoms. With so few patients treated, it is still too early to tell whether humans (NCT04422912).
chronic B cell aplasia may occur in some patients with SLE. The paradox A recent report also implicated the potential of CAR T cells in the
raised by the rapid B cell recovery in SLE in contrast to the durable B cell context of severe asthma42. In two separate animal models, CAR T cells
aplasia in some patients with leukaemia after therapy with the same CAR targeted eosinophils and protected from asthma attack. This protection
T cells may be explained by understanding the kinetics and niches of was durable, providing a potential advantage over antibody-based thera-
CD19 CAR T cells in SLE as compared to those of patients with cancer35. pies for chronic allergic disorders with pathogenic eosinophils. Emerg-
CD19 CAR T cells have similarly been deployed in a patient with an ing evidence also suggests potential efficacy in type 1 diabetes, using
inflammatory myopathy due to antisynthetase syndrome36. The patient CAR T cells that target the antigen-presenting cells that activate auto-
presented with severe weakness and a computerized tomography immune T cells43. However, the resolution of diabetes was not durable.
(CT) scan revealed alveolitis and interstitial lung disease. The patient’s CD19 CAR T cells used thus far target both harmful and healthy
T cells were collected, engineered and reinfused. A rapid expansion of B cells. Using chimaeric autoantibody receptor T cells to specifically
CAR T cells was seen coinciding with a rapid depletion of B cells. Soon eliminate pathologic B cells is promising but requires knowledge of the
Allogeneic route
Healthy donor T cell
Allogeneic route
Cells banked and
thawed on demand
Targeted
nanoparticle CAR
mRNA
Loss of CAR
Fig. 4 | A comparison of ex vivo and in vivo platforms being explored for such as targeted viral delivery and tLNP delivery, are much less mature, but
CAR T therapy. a, At present, autologous CAR T cell manufacturing requires have tremendous potential to disrupt this industry by potentially obviating the
extensive infrastructure, time and lymphodepletion of the patient. Allogeneic need for manufacturing facilities and lymphodepletion. tLNP delivery has the
CAR T cell manufacturing can potentially reduce the associated costs and time additional feature of only transiently expressing the CAR.
required, but still requires lymphodepletion of the patient. b, In vivo platforms,
The long-term effects of T cells that were once CAR T cells should be transduction strategies92. This may prove attractive for conditions that
profiled. Whether T cells have a memory of once being CAR T cells and require persistent CAR T cells but increases the likelihood of potential
whether there are functional consequences of this prior activity should unwanted targeting.
be determined. Ways to gain temporal control of CAR T cells have been Other approaches being explored include banking donor or stem
reviewed13. tLNPs delivering mRNA in vivo is an orthogonal way to regu- cell-derived T cells for off-the-shelf CAR T therapy2,3,93,94 (Fig. 4). This
late CAR T cells. The use of transient CAR T cells may also prove to be platform still relies on ex vivo transduction, but rather than collect-
an effective way to evaluate safety of new CAR products, allowing for ing patient cells, cells can be obtained from healthy donors or from
titration of both dose and duration of CAR T activity. Although certain induced pluripotent stem cells. Using allogeneic CAR T cells in can-
cancer contexts may necessitate a virally transduced CAR T cell prod- cer has gained substantial traction owing to its potential to lower
uct, transient CAR T cells could enable widespread adoption of this costs, simplify manufacturing and expand treatment to patients with
modality for non-malignant indications. low-quality T cells. Allogeneic approaches have already reached the
clinic95. Despite the potential benefits of allogeneic CAR T cells, the
Alternative platforms. Targeted in vivo viral transduction strategies process is more complex, requiring genetic alterations of allogeneic
are a different way to bypass the extensive and expensive CAR T cell T cells to prevent GvHD and rejection of allogeneic CAR T cells93. It is
manufacturing process (Fig. 4). Rather than transducing T cells ex vivo, likely that patients may still need to be lymphodepleted to see robust
a targeted virus delivers the CAR gene specifically to T cells in vivo. engraftment of allogeneic CAR T cells95. In urgent contexts, having
This approach produced functional human CAR T cells in humanized banked CAR T cells may be a substantial advantage. Although clinical
mice, using engineered adeno-associated virus or targeted lentivi- deployment of allogeneic CAR T cells is still in the early stages, at pre-
rus88–91. This approach may, like the tLNP approach, obviate the need sent these cells have a reduced duration of persistence in comparison
for the current lymphodepletion protocols. Viral delivery in vivo has with that of autologous CAR T cells. This is probably due to rejection by
the potential to improve accessibility and cost in comparison with the host; however, these T cells may also have cell-intrinsic defects due
ex vivo autologous CAR T therapy; however, viral vector-mediated to the required genetic modifications. This reduced persistence may
gene therapy approaches will probably remain more expensive than be attractive for conditions for which persistent targeting is undesired.
mRNA delivery. Concerns regarding this modality include the risk of The transformation of allogeneic CAR T cells has been observed and
off-target transduction, with germline transmission being the most is a major safety concern96.
serious concern. A potential immunogenic response to the viral vector
may negate the possibility of redosing patients. In contrast to mRNA
delivery, in vivo viral transduction probably will generate CAR T cells The promise of translation
that persist. Engineered viral-like particles have the potential to deliver A major reason for the interest surrounding CAR T therapy is its
gene-editing proteins and overcome many limitations of in vivo viral clinical success in leukaemia, lymphoma and myeloma. Most of the