Autosomal recessive human CTLA-4 deficiency with autoimmune infiltration

In this issue of the JHI, Catak and colleagues (1) report a case of a child who had severe immunodysregulation with onset during infancy (Table 1). This patient had chronic diarrhea with poor growth and lymphocytic infiltration in the gastrointestinal tract, which was initially diagnosed as celiac disease, and he developed recurrent immune thrombocytopenic purpura with hepatosplenomegaly. He also had mild recurrent upper respiratory tract infections with hypogammaglobulinemia and poor vaccine titers before immunoglobulin replacement therapy was started. Whole-genome sequencing identified a novel homozygous CTLA4 hypomorphic missense variant. This substitution of a proline for serine at residue 172 within the conserved transmembrane domain resulted in reduced surface expression and increased lysosomal degradation of CTLA-4 following its internalization from the cell surface. The patient had been previously treated with methylprednisolone and azathioprine, and only later with abatacept was disease adequately controlled. Consistent with the loss of CTLA-4 driving the immunodysregulation in vivo, abatacept normalized the increased activation and proliferation of the patient’s CD4⁺ T cells when they were cocultured in vitro with antigen-presenting cells (APC). Detailed phenotyping before and after abatacept initiation showed normalization of various immune abnormalities, including the increases in circulating activated follicular helper T cells (T_FH), effector memory and exhausted T cells, IFN-γ and IL-10 expression in T cells, as well as activated and atypical memory B cell subsets. The patient is currently 9 years old and remains clinically well after starting abatacept, with resolution of his hepatosplenomegaly and no recurrence of thrombocytopenia, neutropenia, or diarrhea.

Cytotoxic T lymphocyte-associated protein 4 (CTLA-4, also known as CD152) plays a crucial role in dampening T cell functions to maintain lymphocyte homeostasis (reviewed in 2, 10). When T cells become activated, CTLA-4 expression translocates to the cell surface, where it can bind CD80/CD86 on APC. Binding of CTLA-4 with CD80/CD86 leads to the latter’s internalization by trans-endocytosis. The resulting depletion of CD80/CD86 molecules on the APC impairs the ability of the APC to deliver costimulatory signals via CD80/CD86 binding of CD28 on the T cells. Because CTLA-4 has a higher affinity than CD28 for CD80/CD86, it outcompetes CD28, thereby inhibiting T cell activation signals to limit T cell proliferation and T cell IL-2 production.

Much of what we know about how CTLA-4 functions came initially from studies of Ctla4-deficient (Ctla4^−/−) mice, which, similarly to the abovementioned CTLA-4-deficient patient, develop an early-onset multiorgan lymphocytic infiltration (Table 1) (6, 7). Of note, mice rendered genetically deficient for Ctla-4, specifically within regulatory T cells (T_reg), develop a less severe form of disease, suggesting that loss of constitutive expression of CTLA-4 on T_reg cells, along with loss of inducible CTLA-4 expression on activated conventional T cells, both contribute to pathogenesis (8). Overall, mechanistic studies in mice indicate that CTLA-4 functions dually in T_reg and conventional effector T cells, which cooperate to maintain immune homeostasis in vivo.

Understanding how CTLA-4 functions in mice provided a rationale for the development of blocking antibodies against CTLA-4, such as ipilimumab, to enhance T cell activity against cancers. In a pivotal trial, ipilimumab improved overall survival in patients with advanced melanoma, which led to the development of other forms of immune checkpoint inhibitors. However, treatment with CTLA-4 inhibitors is associated with severe dose-dependent inflammation, often involving the gastrointestinal tract, liver, skin, or endocrine organs, and autoimmunity when present can be life-threatening (Table 1) (2). These presentations are reminiscent of the Ctla4^−/− mice, although the prevalence, organs predominantly affected, and associated autoimmunity differ somewhat, which might reflect the doses of CTLA-4 inhibitors used and their administration in adulthood as opposed to complete gene deficiency since birth (Table 1).

Interestingly, mice heterozygous for Ctla4 null alleles do not develop disease (7). Thus, the discovery of disease in humans haploinsufficient for CTLA4 was surprising (Table 1) (11, 12), although prior studies had shown associations of increased risk for various autoimmune diseases with certain polymorphisms in CTLA4 or, alternatively, decreased expression of CTLA-4 on T cells, including T_reg cells (10). Disease in CTLA-4 haploinsufficient humans displays variable penetrance, with ∼60% of carriers affected and ∼40% of carriers remaining healthy (3). Several hundred affected patients have been reported in the literature, with lymphoproliferation, autoimmunity, and recurrent infections predominating (2, 3, 4, 5). Typical features are lymphocytic infiltrations in the gastrointestinal tract and brain, granulomatous-lymphocytic interstitial lung disease, lymphadenopathy, and splenomegaly. Some patients develop lymphomas or gastric cancers (4). Disease also features autoimmunity, typically manifesting as cytopenias, enteropathy, endocrinopathies such as type I diabetes mellitus and thyroiditis, and skin disorders such as psoriasis and vitiligo. A central role of CTLA-4 in disease pathogenesis is supported by the use of engineered fusion proteins (CTLA4-Ig) such as abatacept that treat disease not only in the CTLA-4 haploinsufficient patients but also in other patients with autoimmune diseases; however, more definitive results establishing the efficacy of such targeted therapy for CTLA-4-haploinsufficient patients await the results of the ongoing phase 2 ABACHAI clinical trial (10). By binding to CD80/CD86, CTLA4-Ig blocks CD80/CD86 engagement with CD28 to thereby interrupt delivery of costimulatory signals to the responding T cell. Finally, CTLA-4 haploinsufficiency in humans also features hypogammaglobulinemia with poor vaccine titers and recurrent infections as a part of their complex immunodysregulatory disorder (3).

Consistent with their clinical findings, Ctla-4-deficient mice exhibit increased numbers of activated T and B cells in spleen and lymph nodes as well as hypergammaglobulinemia (6, 7, 8). The increased B cell responses result from increases in T_reg, T_fh that promote B cell responses, and T follicular T_reg that are compromised in their ability to suppress B cell responses (8, 13, 14, 15). By contrast, most CTLA-4 haploinsufficient patients show normal immunological parameters in the peripheral blood, although some patients have decreased numbers of total lymphocytes, T cells, CD4⁺ T cells, CD8⁺ T cells, T_reg, B cells, and NK cells as well as hypogammaglobulinemia (3). The reason for the hypogammaglobulinemia in CTLA-4 haploinsufficient humans but not in mice is unclear but has been proposed to reflect poor survival of plasma cells due to activated T cell infiltration into the bone marrow (12). More extensive immunological analyses in the peripheral blood of CTLA-4 haploinsufficient patients also show trends of decreased numbers of T_reg having impaired suppressor function, naïve T cells, and memory B cells but increased numbers of memory T cells and circulating T_FH (5, 16). Of note, treatment with abatacept normalized T_FH numbers correlating with improved disease control, although effects on other cell subsets have not been as systematically investigated (5, 16).

Why mice heterozygous for Ctla4 null alleles remain healthy while heterozygous CTLA-4 haploinsufficient humans do not is unclear, but this disparity could reflect other factors such as HLA/genetic background and/or environment such as microbiome/microbial exposure. Because of their apparent health, mice heterozygous for Ctla4 null alleles have not been studied in-depth, although a close examination has revealed an increased germinal center area after primary immunization with superantigen Staphylococcal enterotoxin B, suggesting the possibility of other mild cellular abnormalities (9). Regardless, results from mice suggest that humans with monogenic biallelic CTLA-4 deficiency would also develop a similar severe immunodysregulatory disease. This indeed appears to be supported by Catak and colleagues’ first reported case of autosomal recessive CTLA-4 deficiency.

The identification of this case helps us better understand the overall role of CTLA-4 in humans, as well as related disorders such as LRBA or DEF6 deficiencies, which decrease CTLA-4 cell surface expression by impairing its intracellular trafficking. It will be important to identify additional patients to determine whether findings for this index patient can be generalized, including whether abatacept treatment is sufficient to avoid hematopoietic stem cell transplantation. Furthermore, future studies in either CTLA-4 haploinsufficient or additionally CTLA-4 deficient humans should help us better understand the variable disease penetrance and impaired B cell responses that are features of the humans but not their corresponding mutant mouse models.