High Prevalence of PNH-phenotype Cells in Patients Who Received CD19-targeted CAR T-cell Therapy

Abstract

Chimeric antigen receptor (CAR) T-cell therapy targeting CD19 has emerged as a remarkably effective treatment option for relapsed/refractory B-cell acute lymphoblastic lymphoma and diffuse large B-cell lymphoma.1–3 While cytokine-releasing syndrome (CRS) and neurotoxicity are major short-term toxicities, prolonged cytopenia is one of the most important issues among late toxicities.4,5 Cytopenia manifests as a biphasic pattern.6 Early cytopenia, which occurs several days after CAR-T cell infusion, commonly resolves within 3 weeks after CAR T-cell infusion and is attributable to lymphodepletion chemotherapy (fludarabine and cyclophosphamide in most cases) as well as CRS. In contrast, late cytopenia occurs beyond 3 weeks after infusions and typically resolves in 1 to 2 months in most cases, but could persist for more than 6 months; however, the mechanisms remain unclear.7 Glycosylphosphatidylinositol (GPI)-anchored protein-deficient (GPI [−]) cells were first observed in patients with paroxysmal nocturnal hemoglobinuria (PNH). GPI (−) cells (ie, PNH-phenotype cells) were also detected at a low frequency in patients with bone marrow failure with etiologies other than PNH, including acquired aplastic anemia (AA) or lowrisk myelodysplastic syndromes (MDS).8 Moreover, the presence of PNH-phenotype cells serves as a predictor of a good response to immunosuppressive therapy in AA or MDS, regardless of the frequency of those cells.9–11 Recent reports showed that high-resolution flow cytometry (observation of GPIanchored protein-deficient Cells in Japanese Patients with Bone Marrow Failure Syndrome and in Those Suspected of Having PNH [OPTIMA] method) can precisely detect clinically relevant minor PNH-type cell populations, defined as ≥0.003% CD11b+FLAER− granulocytes and ≥0.005% glycophorin A+CD55−CD59− erythrocytes, that were absent in healthy individuals.12,13 A large-scale study enrolling 1210 patients revealed that PNH-type cells were detected in 56.3% of AA, 18.5% of MDS, 100% of classical PNH, and 18.6% of undiagnosed bone marrow failure patients.13 Based on these observations, we assessed the presence of PNH-phenotype cells using OPTIMA method in a patient who developed severe and prolonged cytopenia on day 48 after CAR T-cell infusion (Table 1, patient no. 8). Surprisingly, PNH-type cells were detected at much higher levels than the threshold in the granulocyte (0.061%, Figure 1A) and erythrocyte (0.049%) populations. Bone marrow biopsy performed on day 48 (the same day as that of the assessment of PNH-type cells) showed severely hypocellular marrow with marked decrease in megakaryocytic and erythroid lineages and no evidence of lymphoma cells (Figure 1B). Chromosomal abnormalities or gene mutations suggestive of MDS were not detected. Severe cytopenia was resistant to low-dose corticosteroid, which was used for treatment of skin erythema in this patient, and repeat testing on day 92 showed persistence of the PNH-type cells (0.053% granulocytes and 0.029% erythrocytes). Finally, cyclosporine was started on day 155, which resulted in a remarkable improvement in the blood count. Based on this experience, we assessed prevalence of PNHphenotype cells in 17 consecutive patients (including patient no. 8 described earlier), who received CAR-T cells (Tisagenlecleucel) for relapsed/refractory acute lymphoblastic lymphoma or diffuse large B-cell lymphoma and had a partial or complete response (Table 1). The median age of the patients was 60 (range, 24-70) years. The duration from the date of CAR T-cell therapy to the assessment of PNH-type cells was a median of 84 (range, 33-252) days. Surprisingly, PNH-type cells were detected in 7 of 8 patients (87.5%) who were examined 1-2 months after infusion, when late cytopenia generally occurs. In contrast, PNHtype cells were detected in 3 of 9 patients (33.3%) who were examined beyond 2 months, by which late cytopenia generally resolves. The difference in the prevalence of PNH-type cells before and after 2 months after infusion was statistically significant (P < 0.05 with Fisher exact test). When analyzed based on the presence of neutropenia regardless of the timing of the assay, defined by an absolute neutrophil count of <1000/μL, PNH-type cells were detected in 7 of 9 patients (77.8%) with neutropenia, and 3 of 8 patients (37.5%) without neutropenia. We demonstrated that PNH-type cells are frequently detected in peripheral blood between 1 and 2 months after CAR T-cell infusion. The association between the presence of PNH-type cells and a response to immune suppressive therapy in AA has led to a suggestion that their presence is a surrogate marker for an immunological pressure on hematopoiesis.11 PNH-type cells lack GPI-anchor proteins that can be targeted by immune effector cells, and are therefore considered to have growth advantages when immunological pressure is exerted. Consequently, immune activation of hematopoiesis 1-2 months after CAR T-cell therapy may be responsible for emergence of PNH-type 1Department of Hematology, Hyogo College of Medicine Hospital, Hyogo, Japan 2Department of Surgical Pathology, Hyogo College of Medicine, Hyogo, Japan 3Department of Transfusion Medicine and Cellular Therapy, Hyogo College of Medicine Hospital, Hyogo, Japan Copyright © 2021 the Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the European Hematology Association. This is an open access article distributed under the terms of the Creative Commons AttributionNonCommercial-ShareAlike 4.0 License, which allows others to remix, tweak, and build upon the work non-commercially, as long as the author is credited and the new creations are licensed under the identical terms. HemaSphere (2021) 5:9(e628). http://dx.doi.org/10.1097/HS9.0000000000000628. Received: 20 February 2021 / Accepted: 2 July 2021