Checkpoint blockade toxicities: Insights into autoimmunity and treatment

Abstract

Checkpoint blockade has transformed not only the way cancers are treated, but also highlighted the importance of mounting a proper immune response against tumors. Despite advances in the field of immunotherapy, many patients develop a range of inflammatory toxicities that limit the efficacy of these therapies. These toxicities range from barrier site injury, such as colitis, to endocrine organ dysfunction, such as diabetes. In order to properly treat patients with cancer and avoid checkpoint blockade induced toxicities, we must gain a deeper understanding of the underlying mechanisms generating these adverse events. Cytotoxic and tissue-resident T cells likely play an important role in mediating some toxicities, though high levels of cytokines and the generation of auto-antibodies in other toxicities demonstrates these mechanisms are not all shared. Certain risk factors for specific toxicities may be able to predict who might benefit most from alternative therapies given the risk-benefit associated with checkpoint blockade. As the targets of checkpoint inhibitors have important functions in the prevention of autoimmunity, insights into risk factors and causes of toxicities will further our knowledge of fundamental immunology and enable the development of novel therapeutics.

Introduction

Monoclonal antibody blockade of the immune regulatory “checkpoint” receptors cytotoxic T lymphocyte antigen (CTLA)-4, Programmed Death (PD)-1 and its ligand PD-L1 has revolutionized the treatment of diverse cancers, leading to durable responses in a subset of patients with advanced malignancies [1,2]. The clinical success of checkpoint blockade has underscored the importance of endogenous immune responses in shaping cancer growth, and revealed a critical role for reactivation of effector memory CD8 + T cells in mediating this response [3]. However, alongside the benefits of checkpoint blockade has come a wide range of inflammatory toxicities. These toxicities can affect any organ in the body and frequently cause morbidity and delayed or discontinued immunotherapy use; in rare cases, these toxicities can even be fatal [4,5]. Inflammation from checkpoint blockade most commonly affects the barrier tissues, including the skin, gastrointestinal tract and liver, and the respiratory epithelium [5]. Endocrine organs and joints are also common targets, similar to many spontaneous autoimmune diseases [6,7]. Inflammation in the heart and nervous system is rare, affecting only a few percent of people on checkpoint blockade, but both are important cases of fatal toxicity [4]. CTLA-4 blockade is considerably more likely to cause inflammatory side effects than either PD-1 or PD-L1 blockade, and for CTLA-4 inhibitors, the frequency and severity of this inflammation is dose dependent [5,[8], [9], [10]]. When CTLA-4 and PD-1 blockade are combined to treat melanoma, nearly all patients develop some inflammatory side effects, and treatment-limiting toxicities occur in over half of patients [11].

In addition to the impact these inflammatory side effects have on clinical care, these toxicities are also an important limitation to developing novel treatments. Checkpoint inhibitor toxicities are more common with combination immunotherapies, and also occur more frequently when combined with other treatment modalities such as targeted agents and chemotherapy [[11], [12], [13], [14], [15]]. An expansive array of immunotherapies targeting other pathways are in clinical development, including antibodies targeting other co-inhibitory receptors (e.g. TIM3, LAG3 and TIGIT), agonistic antibodies against co-stimulatory receptors (e.g. OX40, CD137, CD40, GITR), as well as a variety of other immune modulating agents acting on multiple distinct pathways [16]. Many of these agents have been limited by adverse events, or dosage has been reduced to avoid them. This is particularly true for novel combination treatments, which are also more likely to generate a clinical response.