To the Editor,

22q11.2 deletion syndrome (22q11.2DS) is a rare congenital disorder resulting in a variety of immunological changes, one of which is secondary immune cytopenias that tend to be recurrent and difficult to manage^1,2. There is currently no standard regimen for treating immune cytopenias in 22q11.2DS³. We present two cases of Evans syndrome (ES) in the setting of 22q11.2DS with an absence of typical phenotypic features.

The first case is of a previously healthy 18-year-old male who presented at age 16 with mucocutaneous bleeding and thrombocytopenia with a positive Coombs test but without active hemolysis. His evaluation included CT chest, abdomen, and pelvis which showed axillary and inguinal adenopathy. A biopsy of an enlarged lymph node was performed, revealing reactive lymphoid hyperplasia. Bone marrow evaluation was negative for increased blasts with normal trilineage hematopoiesis. He was clinically diagnosed with immune thrombocytopenic purpura (ITP) but did not respond to two doses of intravenous immunoglobulin (IVIG) and his response to systemic steroids was suboptimal. Subsequently, he was initiated on romiplostim, a thrombopoietin receptor agonist (TPO-RA), with an improvement in platelet counts. Following discontinuation of romiplostim, he remained thrombocytopenic with baseline platelet counts between 80,000-100,000/µL (Figure 1). Other than occasional nosebleeds, he was otherwise asymptomatic. One year later, he presented with jaundice and was found to have warm autoimmune hemolytic anemia (AIHA). His anemia was responsive to systemic steroids. Due to persistent lymphopenia and evolving hypogammaglobinemia, a genetic panel for immunodeficiency was initiated and revealed a gross deletion inTBX1 . Further, the SNP microarray confirmed 22q11.2DS.

The second case is of an 8-year-old male with a history of partial cleft palate who was initially diagnosed with ES when he was one year old. His anemia resolved with a steroid course although he remained thrombocytopenic. He subsequently had two additional episodes of warm AIHA treated with rituximab courses due to insufficient response to steroids. He also developed steroid-induced hyperglycemia requiring insulin therapy. Due to persistent cytopenias, he underwent bone marrow evaluation to rule out the infiltrative marrow process. Incidentally, SNP microarray from his bone marrow aspirate identified 22q11.2DS. Following a diagnosis of 22q11.2DS, he had recurrent episodes of immune cytopenias either manifesting as acute warm AIHA or acute worsening of his thrombocytopenia. He typically required both steroid and rituximab during relapse of his warm AIHA. He eventually was started on romiplostim with normalization of his platelet counts. He maintained normal platelet counts following the discontinuation of romiplostim (Figure 1). In-depth immunophenotyping performed on both patients revealed mild hypogammaglobulinemia, moderate T cell lymphopenia, significantly reduced CD4+ naïve T cells, and decreased class-switched memory B cells. Table 1 below summarizes the immunologic findings in these two patients.

Multiple cases of ES have been documented in DiGeorge syndrome (DGS) individuals after the first reported case in 1990 involving an infant who developed ITP and subsequently AIHA during childhood⁴, suggesting that a thorough investigation for 22q11DS is warranted in individuals presenting with recurrent immune cytopenia, even without typical DGS features^4,5. The first case described in this paper did not have the typical features of DGS, which subsequently led to a delay in the diagnosis until 18 years upon further investigation of recurrent autoimmune cytopenia. There were many theories explaining the increased risk of autoimmunity in patients with DGS. Deshpande et al. study on 467 patients with pDGS showed patients with autoimmunity had lower levels of CD3+, CD3+CD4+, and naïve CD4+CD45RA+CD27+ T lymphocytes compared with pDGS patients without autoimmunity⁶. It was concluded that premature CD4+ T-cell aging, and lymphopenia-induced spontaneous peripheral T-cell proliferation might contribute to the pathogenesis of autoimmunity in patients with pDGS⁷. A case-control study showed patients with hemolytic anemia (HA) tend to have lower naïve CD4+ cells, higher memory CD4+ T cells, lower recent thymic emigrants, lower class-switched B cells and higher naïve B cells compared to the non-HA group¹.

Autoimmune cytopenia (AIC) in DGS seems to be resistant to treatment. In a study on treatment selection and response for patients with pDGS with AIC, 4 out of 7 pDGS patients with ES had refractory cytopenias requiring second-line immunomodulation including rituximab and sirolimus⁸. Among those four patients, there was a background of antibody deficiency and lower naïve T cell counts compared to pDGS patients with mild AIC⁸. We observed similar immunophenotypes in our two patients and recommend extensive B and T cell immunophenotyping at diagnosis and close monitoring of these biomarkers for early identification of patients at risk for severe and recurrent AIC. Although our patients demonstrated good response to TPO-RA, it may not fully address the underlying immune mechanism driving the peripheral destructions due to autoreactive T and B cells. Long term T-cell directed immunomodulation maybe the preferred therapy for certain patients with recurrent AIC.

We conclude that there is an increased risk of having AIC among patients with 22q11.2DS. This is suggested by specific immunophenotypes according to the literature. Managing cases of AIC secondary to 22q11.2DS remains controversial. There is no standard regimen to treat those patients. We suggest those patients be treated differently, as their disease course is much more complicated. Our two patients diagnosed with 22q11.2DS with ES responded well to the TPO-RA after trying multiple regimens.

Table 1. Pertinent immunophenotyping results