Therapy-related acute myeloid leukemia with 24-month latency after CD19 CAR-T cell therapy in relapsed/refractory...

Therapy-related myeloid neoplasms (t-MNs) have emerged as a serious late complication of CD19-directed chimeric antigen receptor (CAR) T-cell therapy. While large-scale studies have characterized the epidemiology of t-MNs, the mechanisms underlying their development remain incompletely understood, and cases with extended latency are rarely documented. We report a case of therapy-related acute myeloid leukemia (t-AML) occurring 24 months after initial CD19 CAR-T cell therapy, with longitudinal genomic monitoring demonstrating evolution from wild-type TP53 to biallelic TP53 loss.

A 65-year-old male was diagnosed with germinal center B-cell (GCB)-type diffuse large B-cell lymphoma (DLBCL) in February 2019. After first-line R-CHOP chemotherapy and subsequent relapse, he received two courses of CD19 CAR-T cell therapy (June 2020 and February 2021) with sintilimab maintenance. Twenty-four months after the first CAR-T infusion, he presented with pancytopenia. Bone marrow examination revealed increased primitive and immature mononuclear cells, with flow cytometry confirming 9.49% abnormal myeloid blasts (CD34+, CD117+, MPO+) consistent with AML-M5b. Cytogenetic analysis demonstrated a complex karyotype: 44-45, XY, del(5)(q14q33),-16, der(17;18)(q10;q10),-19, add(21q). Next-generation sequencing identified biallelic TP53 loss (copy number loss and p.P278A point mutation in exon 8). Notably, NGS at initial DLBCL diagnosis had shown wild-type TP53. The patient received azacitidine plus venetoclax, achieving partial remission, followed by child-to-parent haploidentical allogeneic hematopoietic stem cell transplantation. DLBCL remained in remission throughout AML-directed therapy. The patient ultimately died from COVID-19 pneumonia in December 2022.

This case illustrates a rare but serious late complication of CD19 CAR-T cell therapy, characterized by an exceptionally long 24-month latency and documented evolution from wild-type TP53 to biallelic loss. While causality cannot be established from a single case, these findings are consistent with the hypothesis that prolonged genotoxic stress and inflammatory cytokine exposure may promote clonal selection of TP53-mutated hematopoietic cells. This report underscores the necessity of extended hematologic surveillance beyond the first 12 months post-CAR-T infusion, baseline CHIP screening prior to therapy, and systematic longitudinal genomic profiling in future cases to validate these observations.