The expanding spectrum of CAR-T related T-cell lymphomas

In this issue of Haematologica, Terakawa and colleagues¹ report a clinically important case illustrating the complex interplay between immune effector cell (IEC) therapies, immune dysregulation, and clonal T-cell evolution. The authors describe a patient with relapsed diffuse large B-cell lymphoma (DLBCL) who developed severe hemophagocytic lymphohistiocytosis (HLH) concurrently with a secondary peripheral T-cell lymphoma during treatment with the CD3×CD20 bispecific antibody epcoritamab, following prior chimeric antigen receptor T-cell (CAR-T) therapy with tisagenlecleucel. Through molecular analyses, they demonstrate that the malignant T-cell population originated from a pre-existing T-cell clone detectable in the leukapheresis material used for CAR-T manufacturing, which harbored alterations affecting the TET2 locus and subsequently acquired a pathogenic TP53 mutation during the clinical course. Importantly, rare CAR-T-associated T-cell lymphomas have emerged as severe and increasingly recognized complications as CAR-T cell therapies gain broader regulatory approval and become more widely accessible.² Initially, most reported cases involved T-cell lymphomas derived from CAR-transduced T cells with detectable CAR transgene integration.² However, an expanding body of literature now describes secondary T-cell lymphomas arising after CAR-T therapy in which the malignant cells lack detectable CAR expression. To date, a limited but growing number of such cases have been reported (Table 1), highlighting that CAR expression is not a prerequisite for malignant T-cell evolution in this setting.

Table 1.Clinical spectrum of CAR-negative T-cell lymphoma.

Clinically, these CAR-negative secondary T-cell lymphomas share several features with CAR-positive counterparts, including frequent predisposition for extranodal manifestations. Virtually all CAR-negative T-cell lymphomas presented a mature, often cytotoxic immunophenotype, frequently characterized by loss of the pan-T-cell antigen CD5,^1,3-5 a feature well recognized in cytotoxic T-cell neoplasms, including T-large granular lymphocytic leukemia (T-LGLL).⁶ Moreover, many of these cases are classified as peripheral T-cell lymphoma, not otherwise specified (PTCL-NOS),^1,3,5,7 underscoring that their biology and clinical presentation do not consistently align with established pathogenic categories of specified mature T-cell lymphomas.

Figure 1.Proposed pathogenetic concept of immune effector cell-associated clonal T-cell evolution. (Left) A polyclonal T-cell compartment containing a pre-existing, burdened T-cell population harboring clonal hematopoiesis-associated alterations, most prominently disruption of the epigenetic regulator TET2 (highlighted in red). These altered clones may persist at low frequency without overt clinical manifestation. (Middle) Following immune effector cell therapy, including chimeric antigen receptor (CAR) T-cell manufacturing and integration, antigenic and inflammatory selective pressure promotes oligoclonal T-cell expansion. At this stage, both CAR-positive and CAR-negative T-cell populations with shared TET2 disruption may co-exist, reflecting therapy-driven clonal selection rather than CAR transgene-dependent transformation. (Right) Under sustained selective pressure, analogous to Darwinian evolution, as exemplified by adaptive radiation in Darwin’s finches, acquisition of a secondary oncogenic hit enables the outgrowth of a dominant monoclonal T-cell population. In the case reported by Terakawa et al.,¹ this step is exemplified by the acquisition of a pathogenic TP53 mutation, resulting in overt peripheral T-cell lymphoma. Notably, both oligoclonal and monoclonal T-cell expansions may present with HLH, indicating that immune effector cell (IEC) therapy-associated hyperinflammation and overt T-cell malignancy represent different points along a shared biological spectrum shaped by immune-mediated selection.

Surveying the genomic landscape of reported CAR-T-associated T-cell lymphomas, loss-of-function alterations affecting TET2 emerge as a recurrent theme.^1,7,8 This pattern, previously recognized in CAR-positive T-cell lymphomas,² suggests shared pathogenic mechanisms centered on therapy-driven clonal selection. The case reported by Terakawa et al.¹ is particularly instructive in this regard, as it demonstrates that such selective pressure may not be exerted by CAR-T therapy alone but can be compounded by subsequent T-cell-engaging immunotherapies, such as bispecific antibodies. TET2 appears to play a central role in this process. Experimental data have identified TET2 as a critical regulator of CAR-T cell proliferation, acting as a safeguard against uncontrolled expansion.⁹ In murine models, biallelic loss of TET2 enables antigen-independent CAR-T cell proliferation, driven by sustained expression of the AP-1 transcription factor BATF3 and activation of an MYC-dependent proliferative program. Importantly, TET2 is also among the most frequently mutated genes in clonal hematopoiesis of indeterminate potential (CHIP), a condition characterized by age-related expansion of hematopoietic clones harboring mutations in epigenetic regulators.¹⁰ This observation provides a conceptual framework in which pre-existing TET2-altered T-cell clones may persist silently for years, only to undergo malignant transformation upon exposure to strong immune and inflammatory selective pressures (Figure 1). Secondary genetic events, distinct across individual patients, may then drive overt lymphomagenesis. In the case presented by Terakawa et al., this evolutionary step is compellingly illustrated by the acquisition of a TP53 nonsense mutation, effectively abrogating a critical tumor suppressor pathway.

An additional layer of complexity is introduced by the frequent overlap between HLH and the detection of aberrant clonal T-cell populations in this setting. Such overlap is evident not only in the present case but also in prior reports, including the CAR-positive T-cell lymphoma described by Kobbe and colleagues.¹¹ HLH is a well-recognized complication of T-cell lymphomas,¹² yet its co-existence with clonal CAR-T or CAR-negative T-cell expansions underscores that the boundary between IEC-associated hyperinflammation and overt T-cell malignancy may be fluid rather than discrete. Indeed, spontaneous regression of clonal CAR-T populations has been reported, as in the case described by Maurer et al.,⁸ and transient clonal expansions resolving after short-course corticosteroid therapy have been observed in other patients, including the case reported by Mehr et al.⁴ These observations highlight the critical importance of integrating clinical phenotype, temporal dynamics, and molecular findings in the diagnostic evaluation of suspected CAR-T-associated T-cell lymphomas. Current diagnostic criteria for T-cell lymphomas, largely developed in the context of immunocompetent or conventionally treated patients, may be insufficient in the era of immune effector cell therapies. In selected cases with ambiguous findings, close clinical and molecular follow-up may guide the diagnosis.

Collectively, the case presented by Terakawa and colleagues¹ expands the spectrum of CAR-T-associated T-cell lymphomas and underscores the role of immune-mediated selective pressure in driving clonal T-cell evolution. As immune-based therapies continue to reshape hematologic oncology, refined diagnostic frameworks and heightened clinical vigilance will be essential to distinguish transient immune dysregulation from emerging T-cell malignancy.

Footnotes

• Received February 19, 2026
• Accepted February 26, 2026

Correspondence

T. Braun till.braun@uk-koeln.de

Disclosures

No conflicts of interest to disclose.

Contributions

TS and TB prepared the figures and wrote the editorial; TB also provided resources.

Funding

TS is supported by a Köln Fortune MD scholarship; TB is supported by an MSSO scholarship and received research grants from the German José-Carreras Leukemia Foundation (DJCLS 10R_2025), the Walter-Schulz-Stiftung, and the Sander Stiftung (N. 2023.084.1), and is part of the ImmuneT-ME consortium (EPPERMED2024-522).