2. Material and Methods
2.1. Patients
Patients diagnosed with B-cell malignancies who received either ASCT or CAR-T-cell therapy at the Department of Hematology and Hemotherapy, Hospital Universitario Central de Asturias (Spain), were retrospectively evaluated. All procedures were conducted in accordance with the ethical standards of the institutional research committee (Comité de Ética de Investigación con medicamentos del Principado de Asturias, code CEImPA 2024.300) and with the Declaration of Helsinki. Informed consent was obtained from all individuals at the time of treatment for the use of anonymized clinical data.
A total of 275 patients were included in the analysis: 259 patients received ASCT and 16 patients received CAR-T-cell therapy. All patients were treated at a single center and were monitored as part of routine clinical care. Patients treated with CAR-T-cell therapy received lymphodepletion followed by infusion of axicabtageneciloleucel, n = 13, tisagenlecleucel, n = 2 and brexutabtageneciloleucel, n = 1. The malignancies treated with ASCT included Diffuse Large B-Cell Lymphoma (DLBCL), Follicular Lymphoma, and Hodgkin Lymphoma, while the CAR-T group consisted of DLBCL patients, n = 15 and mantle cell lymphoma, n = 1.
2.2. Clinical Assessments and Follow-Up
Clinical data were collected retrospectively from medical records, including treatment history, laboratory results, imaging, and clinical outcomes. Particular attention was paid to hematologic parameters, the occurrence of persistent cytopenias, and signs of myelodysplastic syndrome (MN). Persistent cytopenias were defined as the requirement for transfusional support or continuous use of growth factors (e.g., G-CSF, EPO) for at least 3 months post-CAR-T infusion. Bone marrow assessments (BMA), including morphology, cytogenetics (karyotype, FISH), and molecular studies, were performed in patients who presented with cytopenias after therapy. In patients who developed t-MN, the most recent bone marrow biopsy (BMB) prior to diagnosis was retrospectively evaluated (referred to as the “baseline” or “pre-MN” sample).
Adverse events related to CAR-T-cell therapy, including cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS), were documented and graded using the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI CTCAE), version 5.0. All patients in the CAR-T cohort received the standard lymphodepletion regimen of fludarabine and cyclophosphamide. Specific treatment for ICANS, typically high-dose corticosteroids or anakinra, was initiated based on clinical presentation and grading. Patients were followed longitudinally after cellular therapy, with the time to MN diagnosis recorded in those who developed the disease. In one case, retrospective analysis of a baseline bone marrow sample was performed to assess for pre-existing clonal hematopoiesis. Specifically, next-generation sequencing (NGS) was performed to detect somatic variants in common MN associated genes (e.g., TET2, ASXL1, and DNMT3A) and to detect copy number variations (CNVs) including chromosome 7 abnormalities.
2.3. Statistical Analysis
The incidence of MN was compared between the ASCT and CAR-T-cell groups using Fisher’s exact test. Due to the small number of events, time from treatment to MN diagnosis, as well as duration of follow-up, were analyzed using the Mann–Whitney U test. Statistical significance was defined as p < 0.05. Analyses were performed using standard statistical software.
3. Results
The characteristics of the patients who were treated with ASCT or CAR-T-cell therapy are shown in
Table 1. Notably, significant differences were observed in the proportion of patients developing MN following both therapeutic modalities. Three of the 259 patients with B-cell lymphoma (1.15%) developed t-MN following ASCT, while 2 of the 16 patients developed myeloid neoplasms after CAR-T-cell therapy (12.5%) (
p = 0.03) (
Table 1). Notably, the median time to MN development was numerically shorter after CAR-T-cell therapy than after ASCT (15.5 vs. 69 months) (
p = 0.096).
It is noteworthy that all patients who developed MN were of advanced age and had received extensive prior treatment before undergoing cellular therapy, but patients with MN after CAR-T-cell therapy presented specific clinical and cytogenetic features, including persistent cytopenias and CRS (
Table 2). Moreover, both CAR-T patients had persistent CAR-T cells detected at the time of myeloid neoplasm diagnosis. For context, of the 14 CAR-T patients who did not develop MN, 4 (28.6%) had detectable CAR-T cells at their last follow-up visit.
No evidence of chromosome 7 alterations was observed before MN diagnosis, but 7q deletion and monosomy 7 were detected 14 and 17 months after CAR-T-cell infusion.
Notably, although the median follow-up was significantly longer with ASCT, as first CAR-T cell infusion was performed in April 2023, (57 vs. 3 months, p < 0.001), none of the patients developed chromosome 7 alterations after ASCT (Descriptive).
Given the novelty of these findings, we present a case study of patients with relapsed Diffuse Large B-Cell Lymphoma (DLBCL) who developed high-risk MN and Clonal Cytopenias of Uncertain Significance (CCUS) associated with chromosome 7 alterations following CAR-T-cell therapy.
3.1. Case 1 (Patient 1 in Table 2)
A 74-year-old woman was diagnosed with DLBCL, germinal center-like, stage IV, International Prognostic Index (IPI) score 3, and central nervous system (CNS) IPI score 3 (intermediate risk) in May 2021 on the basis of immunohistochemistry (IHC) analysis of cervical lymph node biopsy data, which revealed CD20+, CD10-, CD79a+, BCL2+, BCL6+, PAX5+, MUM1+, CMYC+, and Ki-67 (+50%). Bone marrow assessment (BMA) revealed no lymphoproliferative infiltration. She was treated with six cycles of rituximab, cyclophosphamide, adriamycin, vincristine, and prednisone (R-CHOP) and achieved a complete response (CR) on computed tomography (CT).
In January 2022, she relapsed and received second-line treatment with rituximab, gemcitabine, dexamethasone, and cisplatin (R-GDP). An interim positron emission tomography-computed tomography (PET-CT) scan revealed stable disease after two cycles of treatment, but progression of adenopathy was observed the following month. She received third-line treatment with polatuzumab–rituximab–bendamustine and achieved a second CR in September 2022. The patient relapsed in April 2023 and was considered eligible for CAR-T-cell therapy. She received bridging therapy for her progressive disease with rituximab, gemcitabine, and oxaliplatin (R-GEMOX) prior to CAR-T-cell therapy.
On admission for CAR-T-cell infusion, her blood count revealed grade 2 anemia with normal platelet and leukocyte counts. After lymphodepletion, the patient received an axicabtageneciloleucel infusion in July 2023. On the first day after infusion, the patient experienced grade 2 cytokine release syndrome (CRS), which was treated with tocilizumab and dexamethasone and a second dose of tocilizumab on day +5. She also developed hypogammaglobulinemia, grade 3 anemia, grade 4 neutropenia, and grade 3 thrombocytopenia according to the National Cancer Institute Common Terminology Criteria for Adverse Events (NCI CTCAE) version 5.0, which were initially attributed to the hematopoietic toxicity of lymphodepletion therapy.
Owing to persistent pancytopenia, BMA was performed on day +28 after CAR-T-cell therapy, which resulted in hypoplasia and no dysplasia, but no evidence of lymphoproliferative infiltration or increased CD34 cellularity was observed by flow cytometry. Karyotype analysis performed on day +28 was normal, and no FISH abnormalities were detected for 5q and 7q deletion at that time. She was subsequently diagnosed with early grade III immune effector cell-associated hematotoxicity and received supportive care with packed red blood cell and platelet transfusions, periodic filgrastim, and recombinant human erythropoietin (EPO). Cytopenias persisted, and eltrombopag was added on day +31 (off-label use for PCTT), with a partial response observed. On day 60 after axicabtageneciloleucel therapy, the patient experienced radiologic and metabolic progression, which was confirmed by biopsy in September 2023. Owing to the loss of CD20 expression and myelotoxicity, she was treated with polatuzumab. After 3 cycles, her blood count improved, with normalization of her leukocyte and platelet counts, but anemia persisted. She received 7 cycles of polatuzumab and achieved her third CR 9 months after CAR-T-cell infusion. Despite the end of treatment, a decrease in blood counts with macrocytosis was noted.
Seventeen months after CAR-T-cell therapy, the cytopenias rapidly worsened, and following a bone marrow biopsy, the patient was diagnosed with mylodysplastic neoplasm (former myelodyspastic syndrome with ring sideroblasts). No metaphases were obtained in the karyotype analysis; however, FISH analysis revealed deletion of chromosomes 5 q, 7q, and 13q. MN was classificated in the very-high-risk group according to the Revised International Prognostic Scoring System (R-IPSS) (score 7). At the time of MN diagnosis, CAR-T cells were present in the blood. She started treatment with azacitidine, but consolidation with allogeneic hematopoietic SCT declined due to comorbidities.
3.2. Case 2 (Patient 2 in Table 2)
A 61-year-old man was diagnosed with DLBCL (GCB, stage IV, National Comprehensive Cancer Network-International Prognostic (Index NCCN-IPI score 4)) in February 2022 on the basis of IHC analysis of axillary lymph node biopsy, which revealed CD19+, CD20+, CD3-, CD5-, CD10+, CD79a+, CD138+, BCL2+, BCL6+, PAX5+, and Ki-67 positivity (+70–80%). BMA revealed lymphoproliferative infiltration with 8% BCL2+ and BCL6 rearrangement and IGH translocation by fluorescence in situ hybridization (FISH). The results of the MYC and chromosome 7 analyses were normal.
After five cycles of R-CHOP with intrathecal therapy, a discordant response was observed, with a reduction in lymphadenopathies in some areas and progression in others. He started second-line treatment with rituximab, dexamethasone, high-dose cytarabine, and cisplatin (R-DHAP). The PET-CT scan after three cycles revealed no response, and the patient was considered eligible for CAR-T-cell therapy as a third-line treatment. He received bridging therapy for progressive disease with R-GEMOX before receiving CAR-T-cell therapy. His blood counts were normal on admission.
After lymphodepletion, he received axicabtageneciloleucel infusion in December 2022. Within the first 2 weeks, the patient experienced persistent grade 1 CRS, which was improved by two doses of tocilizumab and a single dose of dexamethasone. He also developed grade 1 immune effector cell-associated neurotoxicity syndrome (ICANS), which resolved spontaneously and without specific treatment. He developed grade 3 anemia, grade 4 neutropenia, and thrombocytopenia according to NCI CTCAE version 5.0, which were initially attributed to the hematopoietic toxicity of lymphodepletion therapy. He received supportive care with packed red blood cell and platelet transfusions, periodic filgrastim, and EPO. However, persistent pancytopenia persisted, and BMA was performed on day +33 after CAR-T-cell therapy.
Morphological assessment revealed decreased cellularity with iron overload, no dysplasia, and no evidence of lymphoproliferative infiltration or increased CD34 cellularity. Cytopenias persisted, and iron chelator and eltrombopag were added on day +100. At that time, the PET-CT scan revealed a complete metabolic response, which was maintained in subsequent studies. He achieved independence from transfusion products, but neutropenia persisted, and BMA was performed 10 months after CAR-T-cell therapy. The morphological assessment revealed normal cellularity, no dysplasia, and no signs of lymphoproliferative infiltration or increased CD34 cellularity. Four months later, the cytopenias remained unchanged, and BMA showed normal cellularity without infiltration or clear signs of dysplasia. Karyotype analysis revealed 47, XY, +Y[12]/46, XY[8]. MN-associated gene analysis revealed two truncating variants affecting the genes TET2 (13.6%) and ASXL1 (14.5%) and chromosome 7 loss at a low frequency (<10% variant allele frequency). These findings were confirmed by FISH. BMA was repeated two months later, and no significant changes were detected via morphological or flow cytometry studies. However, karyotype analysis revealed 47, XY, +Y[4]/45, XY, −7[4]/46, XY[12], and FISH confirmed chromosome 7 monosomy in 30% of nucleated cells.
Once clonality was confirmed, we retrospectively investigated the presence of clonality in the basal bone marrow sample collected before CAR-T-cell infusion at the time of lymphoma diagnosis in 2022. The presence of somatic variants in TET2 and ASXL1 was excluded, with a sensitivity of 3%. Monosomy 7 was also not detected in the bone marrow at this time; however, the sensitivity of the next-generation sequencing panel designed to detect copy number variations (CNVs) is only 10%. In February 2024, 14 months after CAR-T-cell infusion, a diagnosis of Clonal Cytopenia of Uncertain Significance (CCUS) was made. At that time, the patient had persistent CAR-T cells in his blood.
4. Discussion
The present study analyzed the incidence of myeloid neoplasms in patients with B-cell malignancies following ASCT and CAR-T-cell therapy. Despite the limited number of patients who develop MN and the shorter follow-up after CAR-T-cell therapy, this analysis suggests a potentially greater incidence of MN and a shorter duration of MN development following CAR-T-cell therapy than after ASCT. Patients with MN post-CAR-T-cell therapy exhibit lymphodepletion prior to CAR-T-cell therapy and CRS and persistent cytopenias post-CAR-T-cell therapy (PCTT), which are recognized risk factors for MN [
3,
8].
Persistent cytopenias post-CAR-T (PCTT) are a common complication in patients treated for B-cell malignancies, with an incidence of 38% reported for axicabtageneciloleucel [
4]. The etiology of persistent cytopenias remains unclear [
9]; however, pre-existing thrombocytopenia (platelet count < 75,000/μL) and the early onset of CRS have been reported as risk factors for the development of PCTT and MN [
4]. Furthermore, baseline cytopenia and elevated baseline levels of C-reactive protein and ferritin have been demonstrated to correlate with the duration of neutropenia after CAR-T-cell therapy [
5]. These observations suggest that the inflammatory damage to the bone marrow caused by high levels of cytokines and inflammation observed after CAR-T-cell therapy, combined with a low stem cell reserve due to previous therapies, may contribute to PCTT. The prolonged presence of CAR-T cells in our patients could be another contributing factor to persistent cytopenias. This prolonged persistence, coupled with chronic inflammation, may create a microenvironment conducive to the expansion of existing or the de novo generation of myeloid clones.
A distinguishing characteristic of the myeloid neoplasm post-CAR-T cell therapy observed in this study is the presence of cytogenetic abnormalities involving chromosome 7. The loss of genes on chromosome 7 can disrupt the maturation of hematopoietic stem cells, leading to ineffective hematopoiesis in MN [
10]. Notably, none of the 259 patients whose post-ACST data were analyzed in this study presented chromosome 7 alterations, despite their advanced age and extensive pretreatment history. While monosomy 7 and 7q deletion are well-established features of therapy-related myeloid neoplasms (t-MN) following exposure to alkylating agents, this typical t-MN presentation usually manifests 5–7 years after exposure [
11]. In our CAR-T cohort, these alterations were detected relatively early, 14 and 17 months after CAR-T-cell infusion. Furthermore, Case 1 exhibited 5q del and 7q, a combination highly suggestive of prior alkylating agent exposure and often associated with
TP53 mutation (though
TP53 sequencing was not performed in this retrospective study). These findings could suggest that CAR-T-cell therapy may either accelerate this process or that there is an alternative underlying mechanism involved in the development of myeloid neoplasms between CAR-T-cell therapy and ASCT. It is plausible that CAR-T therapy, particularly the associated inflammation and bone marrow toxicity, acts as a potent selecting pressure, promoting the expansion of pre-existing, small, alkylator-induced clones that carry these high-risk alterations.
A study of bone marrow samples obtained from patients before CAR-T-cell therapy revealed that patients who developed MN already presented with molecular alterations and cytogenetic abnormalities related to this disease even in the absence of overt MN, suggesting that CAR-T-cell therapy may promote the expansion of pre-existing MN clones rather than the generation of new clones [
3,
12]. Similarly, a patient with MN post-CAR-T-cell therapy with chromosome 7 deletion has already been reported [
13]. A retrospective next-generation sequencing analysis revealed the presence of molecular alterations prior to CAR-T-cell therapy. However, the acquisition of a chromosome 7 deletion and a novel
RUNX1 mutation in this patient may prompt the development of high-risk MN [
13]. This finding is consistent with the observation that no evidence of chromosome 7 alteration was detected prior to the diagnosis of MN in our patients, despite retrospective NGS for somatic mutations in one case (Case 2), though limited by assay sensitivity, suggesting that de novo chromosome 7 alteration may be the event that triggers MN development. The precise role of CAR-T-cell therapy in this process remains unclear. The potential role of immunosuppression, inflammation, or another alternative mechanism induced by CAR-T-cell therapy in the emergence of novel molecular alterations after CAR-T-cell therapy that may contribute to the progression of preexisting MN clones warrants further investigation [
12].
MN with chromosome 7q deletion or monosomy 7 is characterized by early-onset disease, high-risk features, poor prognosis, and a risk of transformation to AML, with a median time to progression (TTP) of 11.5 months in comparison with the 80.5 months observed in patients lacking chromosome 7 alterations [
6,
7]. Consequently, long-term follow-up of these patients, particularly those with prolonged cytopenias, is needed to fully determine the risk of MN and other secondary malignancies. Furthermore, the increased likelihood of progression to AML necessitates the implementation of aggressive therapeutic strategies. Finally, the correlation of MN development after CAR-T-cell therapy with previous cytopenias and molecular alterations raises the question of whether candidates for CAR-T-cell therapy should be evaluated for the presence of clonal hematopoiesis of indeterminate potential (CHIP) before the procedure.