A Rare Complication of CAR T-Cell Therapy: What Two Patient Cases Reveal About T-Cell Lymphoma After Cilta-cel for Multiple Myeloma

Table of Contents

• Key Points
• Background: A Powerful Therapy With a New Safety Question
• Study Methods: How This Investigation Was Conducted
• Patient 1: A 51-Year-Old Man's Journey
• Patient 2: A 54-Year-Old Woman's Journey
• Genomic Findings: Unraveling the Causes
• Clinical Implications: What This Means for Patients
• Study Limitations: What We Still Don't Know
• Recommendations for Patients and Doctors
• Frequently Asked Questions
• Source Information

Key Points

• CAR T-cell therapy for multiple myeloma rarely leads to T-cell lymphoma: 22 cases among over 27,000 doses in the U.S.
• Both patients had TET2-mutated T-cell clones detectable years before CAR T-cell therapy, suggesting a pre-existing risk factor.
• The CAR gene was found inside lymphoma cells, but direct causation by insertional mutagenesis was not proven.
• Viral infections, including Covid-19 and parvovirus B19, coincided with expansion of abnormal CAR+ T cells in both cases.
• Patients should monitor for unusual skin lesions or swollen lymph nodes after CAR T-cell therapy and report them promptly.

Background: A Powerful Therapy With a New Safety Question

Chimeric antigen receptor (CAR) T-cell therapy is one of the most exciting advances in cancer treatment. The approach works by taking a patient's own T cells — the immune system's "soldier" cells — and genetically engineering them in a laboratory to recognize and attack cancer cells. These modified cells are then infused back into the patient, where they hunt down and destroy the malignancy. The word "chimeric" refers to the fact that the engineered receptor combines parts from different sources to create a new, cancer-seeking protein on the surface of the T cells.

Ciltacabtagene autoleucel (cilta-cel) is one such therapy, approved by the U.S. Food and Drug Administration (FDA) for the treatment of multiple myeloma, a cancer of plasma cells in the bone marrow. Cilta-cel is designed to target a protein called B-cell maturation antigen (BCMA), which is found on the surface of myeloma cells. The therapy is manufactured using a harmless virus (a lentivirus) to deliver the CAR gene into the patient's T cells — a process called transduction.

While CAR T-cell therapies have produced remarkable results in patients with blood cancers, a new safety concern has emerged. As of December 2023, the FDA was aware of 22 cases in which T-cell malignant neoplasms (cancers of T cells) developed after the administration of approved CAR T-cell therapies. This is against a backdrop of more than 27,000 doses of CAR T-cell therapy given in the United States. In three of those patients for whom data were available, the CAR transgene (the engineered gene) was detected inside the malignant cells. However, detailed clinical and genomic information had not been published — until now.

This report, published in the New England Journal of Medicine, provides the first detailed clinical and genetic characterization of two patients from the phase 3 CARTITUDE-4 clinical trial who developed a type of peripheral T-cell lymphoma — not otherwise specified (PTCL-NOS) after receiving cilta-cel. The researchers call this condition CAR transgenic T-cell lymphoproliferative neoplasm (CTTLN).

Study Methods: How This Investigation Was Conducted

The two patients described in this report were participants in the CARTITUDE-4 trial, a phase 3, randomized study that compared cilta-cel against standard therapies in patients with lenalidomide-refractory multiple myeloma (myeloma that no longer responds to the drug lenalidomide) who had received one to three previous lines of treatment. The trial showed that cilta-cel significantly improved progression-free survival compared with standard care.

The trial sponsors, Johnson & Johnson and Legend Biotech USA, were involved in collecting, analyzing, and interpreting the data. All patients provided written informed consent. The researchers used an extensive array of sophisticated laboratory techniques to investigate the T-cell lymphomas that developed in these two patients, including:

• Biopsy analysis with microscopy (histology) and immunophenotyping to identify which proteins the cancer cells carried on their surface
• Flow cytometry, a technique that counts and sorts cells based on their surface markers
• Quantitative polymerase chain reaction (PCR) to measure the amount of lentiviral DNA in tissue samples
• In situ hybridization and immunohistochemistry to detect the CAR protein within the cancer cells
• Whole-genome sequencing and targeted DNA sequencing to identify genetic mutations
• CAR integration-site analysis to determine exactly where in the genome the CAR gene had inserted itself
• T-cell receptor (TCR) sequencing to determine whether the T cells were monoclonal (all derived from a single abnormal cell)
• Tests for replication-competent lentivirus to ensure the vector virus itself was not causing infection

Patient 1: A 51-Year-Old Man's Journey

Patient 1 was a 51-year-old man with multiple myeloma, diagnosed in December 2019. His cancer had a high-risk cytogenetic profile, meaning his myeloma cells carried specific chromosome abnormalities (deletion of 17p and gain of 1q) that typically indicate a more aggressive disease. He was enrolled in the CARTITUDE-4 trial in November 2021 after his disease progressed despite an initial treatment regimen that included:

• Bortezomib, cyclophosphamide, and dexamethasone for 6 months
• High-dose melphalan with autologous (self-donated) stem-cell transplantation
• Lenalidomide as maintenance therapy for 1 year

He was randomly assigned to receive cilta-cel. After apheresis (the process of collecting his T cells for engineering), he received two cycles of bridging therapy with daratumumab, pomalidomide, and dexamethasone to control his myeloma while the CAR T cells were being manufactured. He then underwent lymphodepletion (a chemotherapy regimen using fludarabine and cyclophosphamide to make room for the new cells) and received a single cilta-cel infusion in February 2022.

A Promising Initial Response

The treatment initially worked extremely well. The CAR+ T-cell count in his blood peaked on day 14 after infusion at 77 cells per cubic millimeter, before declining to 3 cells per cubic millimeter by day 92. He achieved what is known as a stringent complete response — defined by the International Myeloma Working Group as negative immunofixation in serum and urine, disappearance of soft-tissue plasmacytomas, less than 5% plasma cells in the bone marrow, a normal free light-chain ratio, and no clonal cells detectable in the bone marrow by immunohistochemistry or immunofluorescence. He also became negative for minimal residual disease (MRD), meaning no cancer cells could be found using ultra-sensitive testing (sensitivity threshold of 1.0 × 10⁻⁶, or one cancer cell in a million). This response lasted for more than 2 years after his cilta-cel infusion.

The patient did experience elevated levels of serum cytokines (immune signaling molecules) after infusion, but he did not develop cytokine release syndrome (CRS) or immune effector cell–associated neurotoxicity syndrome (ICANS) — two known side effects of CAR T-cell therapy. However, on day 31 after infusion, he developed coronavirus disease 2019 (Covid-19). The symptoms resolved 12 days later after treatment with the antiviral drug sotrovimab. During and after the Covid-19 illness, his serum cytokine levels (interferon-γ, interleukin-6, and interleukin-10) were again elevated. The level of interleukin-7 peaked once more on day 88.

The Lymphoma Appears

Five months after the infusion, a rapidly growing red plaque (erythematous nasofacial plaque) appeared on his face. A biopsy showed an infiltrate of atypical (abnormal) T cells. These cells were positive for the markers CD2 and CD3 (normal T-cell markers) but negative for CD4, CD8, CD7, CD56, ALK, Epstein–Barr virus–encoded RNA (EBER), terminal deoxynucleotidyl transferase (TdT), CD30, and cytotoxic T-cell markers including granzyme B. An FDG-PET scan revealed enlarged lymph nodes in the neck (bilateral cervical lymphadenopathy) that showed increased metabolic activity. A biopsy of these lymph nodes showed a similar T-cell infiltrate. The condition was initially classified as PTCL-NOS in accordance with current diagnostic criteria.

Analysis of the lymph node biopsy revealed a striking finding: it contained 0.8 copies of lentiviral DNA per cell, and staining techniques showed that 90 to 100% of the cells were CAR+. On day 162 after infusion (after the lymphoma had been diagnosed but before any lymphoma treatment), the CAR+ T cells in his blood had re-expanded to 378 cells per cubic millimeter and were negative for both CD4 and CD8 — a "double-negative" pattern. This expansion happened despite the fact that soluble BCMA was undetectable in his blood from day 89 onward (meaning his myeloma remained in deep remission).

Fortunately, tests showed that replication-competent lentivirus (a functional virus that could reproduce and spread) was undetectable in his blood at baseline and at 3 and 6 months after infusion — ruling out one possible safety concern about the manufacturing vector.

Genetic Clues in Patient 1

Sequencing of the T-cell receptor (TCR) β-chain in the lymph node biopsy showed that a single monoclonal (identical) sequence accounted for 91% (122,362 of 134,205) of all T cells — confirming that the lymphoma arose from a single abnormal T cell. Remarkably, this same sequence was detectable in the drug product (the manufactured CAR T-cell infusion bag) at a very low frequency (approximately 0.0002%), meaning the abnormal clone existed before infusion, though at an almost invisible level.

CAR integration-site analysis revealed that the viral vector had inserted itself predominantly into a specific location: the 3′ untranslated region of a gene called PBX2 (91% of reads; 19,200 of 21,073). PBX2 is a homeobox gene involved in development. Reassuringly, the level of PBX2 RNA expression in the patient's lymph node biopsy was similar to the median level seen in 136 samples of T-cell lymphoma from patients who had not received CAR T-cell therapy.

The researchers also analyzed the drug product itself to understand whether the lentivirus had a preferential insertion pattern. They assessed 63,564 cells and detected 5,429 integration sites. The most abundant integration site accounted for only 0.02% of integration events in all cells and 0.07% of total reads — indicating that the drug product did not have any preferential integration site that would suggest the manufacturing process itself was defective.

Targeted DNA sequencing of the initial skin biopsy detected a missense variant (a type of mutation that changes one amino acid) in the TET2 gene: TET2 H1416R, with a variant allele frequency (VAF) of 39%. This same variant was found in lymph node biopsy samples by whole-exome sequencing (VAF 43%) with no abnormal gene copy number. Given an estimated tumor purity of 90%, this finding is consistent with the TET2 H1416R variant being heterozygous and clonally dominant in the tumor — meaning it was present in the majority of cancer cells. Critically, this variant was not detected in bone marrow aspirate DNA from the time of trial entry (sensitivity threshold 0.5%). However, using an ultra-sensitive technique called TwinStrand duplex sequencing, the variant was found in leukapheresis samples collected for autologous stem-cell transplantation 2 years before the patient was ever exposed to CAR T-cell therapy (VAF 0.01%). The same variant was also detected in non-transduced T cells (VAF 0.01%) from the patient before cilta-cel manufacturing. This crucial finding proves that the TET2-mutated T-cell clone existed years before the CAR T-cell therapy. The variant showed an increase in VAF from day 112 to day 162, which corresponded with the expansion of the malignant clone during the same period.

Whole-genome sequencing of the lymph node biopsy revealed a predominantly diploid tumor genome (meaning the cells had a roughly normal number of chromosomes) with an estimated tumor mutational burden of 1.26 mutations per million bases and no dominant mutational signature. Additional findings included a germline (inherited) heterozygous JAK3 variant (V722I), a somatic (acquired) nonsense PTPRB variant (W2059*), a focal duplication involving the region from the 5′ untranslated region through intron 16 of NFKB2, and a second noncoding TET2 splice variant (c.−193+1G→C; VAF 47%).

Treatment Journey and Outcome for Patient 1

The patient received a chemotherapy regimen called CHOEP (cyclophosphamide, doxorubicin, vincristine, etoposide, and prednisone) on a 21-day schedule. He achieved a complete metabolic response, meaning his FDG-PET scans normalized to a Deauville score below 3 (on a scale of 1 to 5, where higher scores indicate greater cancer activity). His circulating CAR+ T cells also dropped significantly to 42 cells per cubic millimeter.

However, he relapsed soon after treatment ended — and this relapse coincided with a second bout of Covid-19. The researchers checked whether his lymphoma's TCR sequence showed specificity for Covid-19 or any other viral antigen, but it did not. Subsequent treatments included:

1. Gemcitabine, dexamethasone, cisplatin, and alemtuzumab
2. Consolidation with fludarabine–melphalan and allogeneic (donor) stem-cell transplantation — however, engraftment failed, and an autologous (self) stem-cell infusion was performed; relapse occurred within 3 months after the allogeneic transplantation
3. The JAK3 inhibitor tofacitinib — but localized disease progression was observed within 1 month of starting this treatment
4. Local radiotherapy (50 Gy in 25 fractions from August through September 2023) followed by weekly pegylated interferon alfa-2a (August 2023 through March 2024)
5. Weekly extracorporeal photopheresis (a procedure in which blood is treated with light-activated drugs, from November 2023 through April 2024)

Encouragingly, since June 2023 his abnormal CAR+ T-cell counts have been below the lower limit of quantification in the blood, and since January 2024 in the bone marrow (last assessed in August 2024). He has no evidence of clinical or radiologic disease and was successfully weaned off photopheresis in April 2024.

Patient 2: A 54-Year-Old Woman's Journey

Patient 2 was a 54-year-old woman with IgG kappa multiple myeloma, diagnosed in November 2019. Her myeloma also carried high-risk cytogenetic features (a translocation t[4;14] and gain of 1q). She received frontline therapy with bortezomib, lenalidomide, and dexamethasone, followed by high-dose melphalan with autologous stem-cell transplantation, and then lenalidomide maintenance for 1 year.

She was randomly assigned to the cilta-cel group in October 2021. After apheresis, she received two cycles of bridging therapy (daratumumab, pomalidomide, and dexamethasone), lymphodepletion with fludarabine and cyclophosphamide, and then a single cilta-cel infusion in January 2022. She did not experience cytokine release syndrome or ICANS.

Her CAR+ T-cell counts were 29 cells per cubic millimeter on day 21 and 8 cells per cubic millimeter on day 42. On day 364 after infusion, she had achieved a stringent complete response with MRD-negative status (sensitivity threshold 1.0 × 10⁻⁵), sustained for more than 2 years after the infusion (last follow-up, June 2024).

Viral Infection and T-Cell Expansion

On day 84 after infusion, the patient developed a parvovirus B19 infection with a high viral load (1,341,375 copies per milliliter). This virus can cause a condition called fifth disease and can be serious in immunocompromised patients. She was treated with high-dose intravenous immune globulin (IVIG). The viral load gradually decreased and cleared by 8 months after the cilta-cel infusion. Importantly, her CAR+ T-cell counts increased during the prolonged infection — from 38 cells per cubic millimeter at the onset of infection to 457 cells per cubic millimeter on day 196 — and did not return to pre-infection levels. The expanded T cells were predominantly CD4–CD8 double-negative and negative for EBER. Replication-competent lentivirus was undetectable in the blood at 6 and 12 months after treatment.

The Lymphoma Appears

Sixteen months after cilta-cel infusion, several skin masses developed on her face, torso, and breasts. The lesions showed spontaneous regression and recurrence in different locations over the following months without treatment. FDG-PET imaging showed progressive disease in the skin, lymph nodes, breasts, lungs, and bones (both above and below the diaphragm). Immunohistochemistry and in situ hybridization confirmed that the cells were predominantly CAR+.

Her blood CAR T-cell counts increased rapidly beginning on day 702, peaking at 15,883 cells per cubic millimeter on day 728 — a dramatic expansion. These circulating cells had a monotypic CD4–CD8 double-negative immunophenotype and appeared pleiomorphic (variable in size and shape) with small-to-medium, basophilic cytoplasm. TCRβ sequencing detected a monoclonal sequence accounting for 67% of total sequences identified — confirming that a single malignant T-cell clone had taken over. This dominant sequence was not detectable in her first-line autograft apheresis product, meaning the clone was not present years earlier (unlike Patient 1).

Genetic Clues in Patient 2

The CAR was confirmed on the surface of circulating T cells, and CAR integration-site analysis of both a lymph node biopsy and circulating CAR T cells revealed a predominant integration into intron 1 of one copy of the tumor suppressor gene ARID1A (a gene that helps regulate how DNA is packaged and accessed in cells). ARID1A was highly expressed in the tumor biopsy sample — its expression level was higher than in 91.6% of all 19,551 genes assessed. However, this rank was actually lower than that seen in a comparison cohort of 104 patients with T-cell lymphoma who had not received CAR T-cell therapy (median rank, 97.4%; range, 93.5 to 98.9). In other words, ARID1A expression in this patient's tumor was within the normal range seen in T-cell lymphoma generally, though at the lower end.

The insertion site was also checked against databases of regulatory elements (TFBS Conserved, VISTA Enhancers, and JASPAR). Only one predicted transcription factor binding site — for a factor called RFX4 — was identified, with a moderate confidence score of 418 out of 1000. A literature search found no studies reporting that RFX4 affects the transcription of ARID1A or any other gene. So, at present, it remains unclear whether the ARID1A insertion contributed meaningfully to the cancer.

Targeted sequencing of circulating T cells identified a TET2 variant (Y1902H) with a VAF of 39%, along with a deletion involving the TET2 locus (copy ratio 0.719; z score −27.977). This pattern is consistent with biallelic alteration — meaning both copies of the TET2 gene were affected — and with clonality (tumor purity estimated at 55%). Just as with Patient 1, the TET2 Y1902H variant was found in leukapheresis samples collected 22 months before CAR T-cell exposure, at a VAF of 0.4%.

A skin biopsy showed pleiomorphic lymphocytes with blastoid (immature, rapidly dividing) features in the dermis and hypodermis. The cells were CD4–CD8 double-negative T lymphocytes (CD3+, CD2+, CD5+, CD7+/−, and CD99+). They were positive for T-cell–restricted intracellular antigen (TIA1) but negative for EBER, perforin, CD56, TdT, CD34, and CD1a, with a high Ki-67 proliferation index of 90% — indicating that nearly all cells were actively dividing. A similar infiltrate with the identical immunophenotype was found in inguinal lymph nodes, and the condition was classified as PTCL-NOS.

The patient received chemotherapy (CHOEP) and then underwent a haploidentical allogeneic (half-matched donor) stem-cell transplantation in May 2024.

Genomic Findings: Unraveling the Causes

When we put both cases together, a coherent — although still incomplete — picture emerges. The researchers propose that several factors likely worked together to cause these lymphomas.

First, both patients had pre-existing TET2-mutated T cells. TET2 is a gene that normally helps regulate how blood stem cells develop and mature. Mutations in TET2 are well known in the field of clonal hematopoiesis — a condition in which blood cells carrying certain mutations expand abnormally but don't yet cause cancer. In Patient 1, the TET2 variant was detectable 2 years before CAR T-cell therapy at a frequency of 0.01%. In Patient 2, it was detectable 22 months before at a frequency of 0.4%. This means the "seed" of the future lymphoma existed long before the CAR T cells were ever manufactured.

Second, the CAR gene inserted into the genome near or within these abnormal T cells. In Patient 1, the lentiviral vector integrated predominantly into the 3′ untranslated region of PBX2. In Patient 2, it integrated into intron 1 of ARID1A. Both genes are involved in important cellular processes — PBX2 is a homeobox gene (involved in development), and ARID1A is a chromatin-regulating tumor suppressor gene frequently mutated in cancers. However, in both patients, the expression levels of these genes were not significantly abnormal compared with T-cell lymphoma controls. This means the evidence that the gene insertion itself caused the cancer ("insertional mutagenesis") is currently inconclusive.

Third, the malignant T cells likely acquired additional genetic changes over time. Patient 1's tumor had a TET2 H1416R variant that increased in frequency over time, plus additional somatic variants and a germline JAK3 variant. Patient 2's tumor had biallelic TET2 alterations (a missense variant on one copy and a deletion on the other).

Fourth, external factors may have contributed. Both patients experienced significant viral infections after their CAR T-cell therapy — Patient 1 had two episodes of Covid-19, and Patient 2 had a prolonged parvovirus B19 infection. Viral infections are known to stimulate T-cell proliferation, and the timing of the patients' T-cell expansions appeared to coincide with these infections. Interestingly, in Patient 2, the CAR+ T-cell counts increased dramatically during the parvovirus infection and never returned to baseline. Additional contributing factors may include the patients' prior myeloma treatments (both had received lenalidomide maintenance, melphalan, and cyclophosphamide — drugs associated with an increased risk of second primary malignant neoplasms) and the fact that multiple myeloma itself carries an elevated risk for lymphoma as a second primary malignancy.

The researchers emphasize that the contribution of insertional mutagenesis — the direct disruption of a gene by the viral insertion itself — to the development of the T-cell lymphoma is currently unclear. Neither of the two integration site genes showed clear expression disruption, and there was no evidence that the drug products themselves had any preferential integration site that would suggest a manufacturing problem.

Clinical Implications: What This Means for Patients

First and foremost, patients should understand that these complications are extremely rare. The FDA was aware of only 22 reported cases of T-cell malignant neoplasms among more than 27,000 doses of approved CAR T-cell therapies administered in the U.S. — a rate of less than 0.1%. Moreover, not all of those cases had confirmed CAR transgene expression in the malignant clones, and detailed data were only available for a small number.

For context, the CARTITUDE-4 trial showed that cilta-cel significantly improved progression-free survival compared with standard care in patients with relapsed or refractory myeloma. The two patients described here both achieved deep, durable remissions of their myeloma — a stringent complete response with undetectable minimal residual disease, lasting more than 2 years. CAR T-cell therapy remains a highly effective, potentially life-saving treatment for many patients with blood cancers, and its benefits generally far outweigh the very small risk of this rare complication.

However, this report provides important insights for clinical practice. Both patients had detectable TET2-mutated T-cell clones before receiving CAR T-cell therapy — clones that were present in leukapheresis material used to manufacture the CAR T cells. While these variants were present at very low frequencies (0.01% and 0.4%, respectively), their presence raises the question of whether pre-screening patients for such mutations could help identify those at higher risk for this rare complication. The researchers note that TET2 mutations are common in the aging population (they are a hallmark of clonal hematopoiesis of indeterminate potential), so routine screening of all patients may not be practical.

The cases also highlight the possible role of viral infections in stimulating CAR+ T-cell proliferation. Both patients had significant viral infections that preceded or accompanied the expansion of their malignant T-cell clones. While this observation is intriguing, the researchers caution that their TCR sequences did not show specificity for the viruses tested — meaning the T cells were not directly responding to the viruses.

Study Limitations: What We Still Don't Know

This report describes only two cases, which limits its generalizability. Several important questions remain unanswered:

• Why do some patients develop this complication while the vast majority do not? The presence of pre-existing TET2 mutations in both patients is a striking observation, but TET2 mutations are relatively common in older adults, and CTTLN remains extremely rare.
• Did the CAR gene insertion actually cause the cancer? In these two cases, the integration sites (PBX2 and ARID1A) did not show significant expression dysregulation, so the contribution of insertional mutagenesis remains unclear. It's possible the CAR gene insertion played no direct causal role — the malignant cells may have simply been "bystanders" that carried the CAR gene because they were transduced T cells. The authors could not prove causation.
• What is the true incidence of this complication? With only 22 reported cases out of 27,000 doses, the risk appears very low, but under-reporting is possible, and only a fraction of cases had samples available for testing.
• Could other CAR products or manufacturing processes carry different risks? This report is specific to cilta-cel, which is manufactured with lentiviral transduction. Other CAR T-cell products use different vectors and manufacturing methods, and their risk profiles may differ.
• What is the long-term outcome for patients who develop CTTLN? One patient achieved durable remission after multiple treatments, while the other was still in the course of treatment at the time of reporting. More follow-up is needed.

Recommendations for Patients and Doctors

Based on this report, patients considering CAR T-cell therapy should keep the following points in mind:

1. Talk to your doctor about the risks and benefits. CAR T-cell therapy offers a chance for deep, durable remission in patients with relapsed or refractory myeloma. The risk of developing a T-cell lymphoma is extremely small, but it is now a known risk that should be disclosed as part of informed consent.
2. Be alert to new or unusual symptoms after treatment. Patients who have received CAR T-cell therapy should be monitored for unusual skin lesions, swollen lymph nodes, or other concerning symptoms. In both cases, the lymphoma appeared as skin lesions or enlarged lymph nodes. Early detection can allow for prompt treatment.
3. Understand that unusual T-cell expansions can occur after viral infections. In these two patients, Covid-19 and parvovirus B19 infections were associated with T-cell expansion. If you develop a significant infection after CAR T-cell therapy, discuss monitoring of your T-cell counts with your healthcare team.
4. Participate in long-term follow-up. The FDA requires long-term safety monitoring for patients receiving CAR T-cell therapies. Staying engaged with these registries helps doctors learn more about the true frequency of rare complications and how to prevent or treat them.
5. For doctors and researchers: The authors suggest that this complication may be driven by transduction of pre-existing mutated T-cell clones followed by acquisition of further oncogenic changes. Future research should explore whether pre-screening for TET2 mutations or other clonal hematopoiesis markers could identify higher-risk patients, and whether alternative manufacturing approaches could reduce the risk of transducing abnormal clones.

Bottom line: This report represents an important step forward in understanding a rare complication of a highly effective therapy. For the overwhelming majority of patients, the benefits of CAR T-cell therapy in treating refractory blood cancers far outweigh the tiny risk of this complication. But as with all medical treatments, ongoing surveillance and openness about risks remain essential.

Frequently Asked Questions

What is the risk of developing a T-cell lymphoma after CAR T-cell therapy?

It is extremely rare. The FDA was aware of only 22 reported cases of T-cell malignant neoplasms among more than 27,000 doses of approved CAR T-cell therapies given in the U.S., a rate of less than 0.1%. Not all cases had the CAR gene confirmed in the cancer cells.

What were the first signs of lymphoma in these two patients?

Both patients developed skin lesions: one had a rapidly growing red plaque on the face, the other had skin masses on the face, torso, and breasts. Enlarged lymph nodes also appeared. At that time, both were in deep remission from their myeloma.

Did the CAR T-cells themselves cause the cancer?

The CAR gene was found inside the cancer cells, but the exact cause remains unclear. Researchers identified pre-existing TET2 mutations in T cells years before treatment, viral infections, and additional genetic changes as possible contributing factors. Insertional mutagenesis (direct gene disruption by the viral vector) was not confirmed.

Should I be screened for TET2 mutations before CAR T-cell therapy?

Both patients had low-level TET2 mutations in T cells before therapy, but TET2 mutations are common in aging populations and this complication is extremely rare. The researchers noted that routine screening of all patients may not be practical, though it could help identify higher-risk individuals.

What should I watch for after CAR T-cell therapy?

Be alert to new or unusual symptoms such as skin lesions, swollen lymph nodes, or other concerning changes. In both cases, the lymphoma first appeared as skin lesions or enlarged lymph nodes. Significant viral infections (Covid-19, parvovirus B19) were also associated with T-cell expansion, so discuss monitoring with your healthcare team.

Is CAR T-cell therapy still worth it given this risk?

For most patients, the benefits far outweigh the tiny risk. In the CARTITUDE-4 trial, cilta-cel significantly improved progression-free survival compared with standard care. Both patients achieved a stringent complete response lasting more than 2 years. The risk of T-cell lymphoma is less than 0.1%.

What treatments did the two patients receive for the lymphoma?

Patient 1 received chemotherapy (CHOEP), an allogeneic stem-cell transplant, tofacitinib, radiotherapy, interferon, and photopheresis; he had no evidence of disease since June 2023. Patient 2 received CHOEP and a haploidentical stem-cell transplant in May 2024, and was still in treatment at the time of reporting.

Should I get a second opinion if I developed T-cell lymphoma after CAR T-cell therapy for multiple myeloma?

T-cell lymphoma after CAR T-cell therapy for multiple myeloma is extremely rare, with only 22 cases reported among more than 27,000 treatment doses in the U.S. The lymphoma may appear as skin lesions or swollen lymph nodes, sometimes months after infusion. Because this complication is uncommon and its diagnosis and treatment are complex, a second opinion can help confirm the pathology and review whether the proposed treatment—such as CHOEP chemotherapy or allogeneic transplant—is appropriate for your specific situation. Diagnostic Detectives Network provides independent expert second opinions.

Source Information

Original Article: "CAR+ T-Cell Lymphoma after Cilta-cel Therapy for Relapsed or Refractory Myeloma"

Authors: S.J. Harrison, C. Touzeau, N. Kint, K. Li, T. Nguyen, C. Mayeur-Rousse, M. Rahman, Y. Le Bris, J. Er, J. Eugene-Lamer, N.M. Haynes, J. Li, R.C. Abbott, C. Bodet-Milin, A. Moreau, E. Letouzé, N. Lendvai, J.M. Schecter, W. Deraedt, A. Banerjee, T. Lengil, M. Vogel, B. Foulk, H. Zhao, D. Smirnov, A. Slaughter, C. Lonardi, E. Lee, L. Marquez, A. Sankari, V. Plaks, J.O.C. Filho, N. Patel, D. Geng, T. Gastinne, H. Kelly, I.S. Tiong, M. Eveillard, P. Chevallier, S. Lade, P. Moreau, S. Grimmond, J. Oliaro, B. Tessoulin, and P. Blombery

Publication: New England Journal of Medicine, 2025, Volume 392, pages 677–685. DOI: 10.1056/NEJMoa2309728. Copyright © 2025 Massachusetts Medical Society.

Funding: The study was funded by Johnson & Johnson and Legend Biotech USA. Clinical trial registration: CARTITUDE-4, ClinicalTrials.gov number NCT04181827.

This patient-friendly article is based on peer-reviewed research. It is intended for educational purposes and should not replace professional medical advice. Patients with questions about CAR T-cell therapy should consult their oncology care team.