Donor derived T-cell large granular lymphocyte leukemia after cord blood transplant for Pediatric T-Cell lymphoblastic leukemia

Large granular lymphocyte (LGL) leukemia is characterized by a clonal expansion of either CD3(+) cytotoxic T or CD3(−) NK cells. LGL leukemia most commonly affects the elderly with a median age of 60, although cases in younger adults and very rarely pediatric patients have been reported (1–4). In most cases T-LGL leukemia is associated with an indolent clinical course characterized by cytopenias and autoimmune phenomena, with infections due to neutropenia being among the most common presenting symptoms (4–7). LGL leukemia mortality is mainly due to severe infections,although this occurs in <10% of patients and the overall survival at 10 years is approximately 70% (8,9). Donor-derived T-LGL leukemia has been reported after hematopoietic stem cell transplant (HSCT) with both bone marrow and peripheral blood donor sources (4,6). T-LGL leukemia is a rare post-HSCT neoplasm that is rarely diagnosed in the pediatric population. Here we present the case of a female who underwent an unrelated umbilical cord blood transplant for T-cell acute lymphoblastic leukemia (T-ALL) and was subsequently diagnosed with donor-derived T-LGL at 11 years of age.

At 6 years of age, the patient was diagnosed with T-ALL, and had refractory disease with 25% T lymphoblasts in the bone marrow following induction chemotherapy. She then underwent salvage chemotherapy and achieved complete remission without detection of minimal residual disease by flow cytometry. She subsequently received an unrelated, 4/6 matched cord blood transplant (CBT) with single mismatches at HLA-B and DRB1. The preparative transplant conditioning regimen contained fludarabine, cyclophosphamide and total body irradiation (13.2 Gy) with cranial boost. Graft versus host disease (GVHD) prophylaxis included cyclosporine and mycophenolate mofetil. Early post-transplant, the patient developed acute skin grade IIB GVHD which was treated with high-dose prednisone. Following completion of an initial prednisone taper, she developed delayed acute gastrointestinal GVHD grade IIA that was treated with lower dose prednisone, beclomethasone, and budesonide. The patient developed a mild, intermittent skin rash which was treated with topical steroid ointment as needed. She also developed transaminitis nine months after transplant in the setting of her cyclosporine taper that was presumed liver GVHD. She was transitioned to sirolimus and low-dose prednisone was restarted with good response. She ultimately completed systemic immune suppression approximately 14 months after transplant and aside from mild skin flares treated with topical immune suppression only, had been without signs or symptoms of chronic GVHD since that time.

Two years following CBT, the patient presented with sepsis, pancytopenia and adrenal insufficiency. Bone marrow evaluation revealed no leukemia and appropriate cellularity and no other abnormalities. The patient received 1 week of empiric antibiotioic therapy and recovered blood counts within 1–2 weeks. Subsequently at 3 years following CBT she again presented with pancytopenia. She recovered to normal blood counts in < 1 week but 2 months later developed neutropenia. A bone marrow exam at the time demonstrated no evidence of leukemia, approximately 50% marrow cellularity, and her chimerisms demonstrated 100% donor for CD3, CD33, CD56 and CD19 cells by fractionated chimerism analysis of peripheral blood leukocytes using amplified fragment length polymorphism on the sorted populations. The patient was seen by Infectious Disease and Immunology specialists for further workup for etiologies of her cytopenias. Testing for CMV and HHV-6 from peripheral blood were negative, EBV was detected at low level of 126 International Units (IU)/mL, and parvovirus testing was positive at 7,500 IU/mL. Serology testing was positive for parvovirus B19 IgG and IgM negative. The patient was treated for chronic parvovirus with high-dose IVIG (1g/kg x 5 days) monthly. Despite monthly IVIG infusions, the patient had persistent parvoviremia with a PCR level at 17,000 IU/ml after 12 months. The patient progressed from persistent neutropenia to pancytopenia. Bone marrow evaluation was repeated at approximately 4 years following CBT and revealed a T-LGL clonal population that was CD8 positive with decreased CD5 expression. Irregular T-LGLs were detected in the peripheral blood (Figure 1A) were subsequnently determined by immunophenotyping to be present in excess in the peripheral blood and bone marrow, comprising 21% (absolute LGL count of 924) and 39% of white cells respectively with abnormally decreased expression of CD2, CD5 and CD8, positive for CD3 and CD7, and negative for CD4, CD56 and CD34 (Figure 1B). No T-ALL population was identified by flow cytometry, and spleen size was normal. Further molecular analysis demonstrated the T-cell receptor gamma gene was clonally rearranged (Figure 2). Flow cytometry cell sorting of the T-LGL population from peripheral blood for chimerism analysis confirmed 99% donor-status of the T-LGL population with a 46,XY normal male karyotype. Rheumatologic evaluation revealed a positive c-ANCA titer of 1:640 and positive anti-MPO antibodies (obtained after IVIG administration), but she had no laboratory values or symptoms suggestive of vasculitis or rheumatologic disorders. The patient began treatment with oral methotrexate 10 mg/m² weekly. Mutational analysis with amplicon-based next-generation sequencing was performed on unsorted peripheral blood and found to be positive for a STAT3 p.Y640F mutation (c.1919A>T,hg19:Chr17:40474482A>T), variant allele fraction 11%, corresponding to a likely heterozygous mutation in the 21% peripheral LGLs. After 3 months of oral methotrexate, the patient had no response in blood counts, and with identification of the STAT3 mutation, treatment was switched to oral cyclophosphamide 50 mg weekly. The patient had no change in blood counts after 4 months on cyclophosphamide, however she had multiple infectious complications including several episodes of neutropenic colitis, typhlitis, and cholecystitis requiring IV antibiotics. Due to the lack of response to cyclophosphamide, she then was switched to tocafitinib, a JAK3 inhibitor based on recent reports of success as salvage therapy in LGL, while undergoing evaluation for second transplant (10). The patient remained on tocafitinib for approximately 4 months prior to transplant conditioning and experienced a rise in her platelet count with stabilization of her hematocrit, however had persistent severe neutropenia with an undetectable neutrophil count.

Figure 1.

Figure 1A: Peripheral blood cytology showing large granular lymphocytes with excess cytoplasms, irregular nuclear contours, and coarse chromatin.

Figure 1B: Flow cytometry analysis of bone marrow and peripheral blood showing LGL population.

Figure 2.

T cell receptor gamma gene rearrangement

Donor derived T-LGL leukemia following HSCT mostly affects individuals with a median age of 60 years and is rarely described in children (11). We found a single report in pediatrics of a 16 year-old male who developed T-LGL leukemia following allogeneic related donor HSCT for T-cell lymphoma (3). The patient developed EBV-positive post-transplant lymphoproliferative disorder (PTLD), but had persistent neutropenia and splenomegaly despite treatment for PTLD with rituximab 9 months after HSCT. A splenectomy was performed with flow cytometric analysis that showed an aberrant T-cell population consistent with T-LGL. In this patient, testing for a STAT3 exon 21 mutation was negative in both pre-HSCT and post-HSCT specimens. Reports from the adult literature include one by Gill et al who described 7 patients that developed T-LGL leukemia following HSCT, with 3 patients having disease derived from donor T-cells (4). Interestingly, none of the patients showed cytopenias, autoimmune phenomenon or organ infiltration, which are features typical of de novo T-LGL leukemia. Rather, all patients in the case series showed lymphocytosis, and 5 had documented CMV viremia.

We describe here a pediatric patient with donor-derived LGL leukemia post-CBT for T-ALL. LGL leukemia is a clonal lymphoproliferative disease associated with chronic inflammation and autoimmunity caused by stimulation resulting in oligoclonal LGL expansion (5). Chronic activation by a virus with structural relationship to the human T-cell leukemia/lymphoma virus (HTLV)-family has been proposed as a potential etiologic stimuli that contributes to the initial expansion of a population of T cells leading to STAT3 activation and emergence of a dominant clone (5,12). Kondo et al. have previously reported a 35-year-old female who developed parvovirus B19-associated pure red cell aplasia with T-LGL leukemia (13). Previous case series have described a link with CMV viremia suggesting chronic aberrant immune triggers from inadequate clearance of viruses in the post-HSCT period. We hypothesize the patient presented here had inadequate clearance of viruses, as evidenced by the chronic low level parvoviremia, which resulted in chronic aberrant immune stimulation and may have led to T-cell stimulation and LGL leukemia. LGL leukemia is considered an indolent disease with no curative chemotherapy options, posing a unique challenge for individuals diagnosed in childhood. Although the 10 year overall survival is approximately 70%, this is certainly a less meaningful measure in children as this group of patients is at significant risk for infectious morbidity and mortality from prolonged neutropenia.(8) The only reported curative treatment for patients with LGL is HSCT, although experience is very limited in this population and mortality due to toxicity and relapse are common (14).

This case report highlights the therapeutic challenges of diagnosing and treating LGL in children. The majority of LGL in older patients is indolent and the therapeutic approach of immunosuppressive agents is limited in its ability to produce long-lasting remissions, however this strategy may be insufficient in young children. In cases with suboptimal response to multiple therapeutic modalities, such as the case presented here, a second HSCT is the only curative option. The profound cytopenias and the infectious comorbidities resulting from chronic LGL, including chronic viremia, both of which the patient described above expderienced, make this a very high-risk treatment choice. This case demonstrates consideration of LGL leukemia is warranted among post-HSCT patients, including children, with unexplained cytopenias or evidence of chronic infections, including viruses.