What do we already know about long-term risks from CAR-T in blood cancers?
The most concrete long-term risk data comes from patients treated for blood cancers, not solid tumors. A large real-world study of 1,480 DLBCL patients who received CAR-T therapy found a significantly higher incidence of acute myelogenous leukemia (AML) (3.23% vs. 1.18%) and myelodysplastic syndrome (MDS) (2.93% vs. 1.11%) compared to matched patients who did not receive CAR-T [1]. This means that for every 100 patients treated, roughly 2 more developed these serious bone marrow cancers than would be expected without CAR-T. Another study of 355 lymphoma patients across two centers found a 14% overall cumulative incidence of any subsequent malignancy at 36 months, including solid tumors (6.1%), hematologic cancers (4.5%), and skin cancers (4.2%) [2]. These figures establish a baseline for risk, but they come from patients with blood cancers, not solid tumors.
Why might the long-term risks be different—and possibly underestimated—for solid tumors?
Solid tumors present unique challenges that could create new or worse long-term risks. First, CAR-T cells struggle to infiltrate solid tumors and often fail to persist long enough to be effective, which has led to extremely low response rates [5]. To overcome this, researchers are developing new delivery methods, like a 3D scaffold that boosted CAR-T cell expansion 50-fold in the lab and maintained expansion for up to 30 days in a cervical tumor model [3]. While this improves efficacy, it also means CAR-T cells may remain active in the body for much longer than intended, raising the possibility of chronic inflammation or autoimmune damage that hasn't been studied long-term. Second, finding target antigens that are only on tumor cells and not on healthy tissue is much harder for solid tumors—a 2022 study on Ewing sarcoma found that truly tumor-specific antigens are rare, and many potential targets are also expressed on the placenta or fetal tissues, which could lead to off-tumor attacks if those cells are present [4]. These factors mean the risk profile for solid tumors may include not just secondary cancers, but also autoimmune complications and organ damage that are not yet well documented.
Is the evidence strong enough to say risks are being underestimated?
The evidence points toward underestimation, but with important caveats. The two largest studies on long-term risks [1][2] both focus on blood cancers and have relatively short follow-up (mean 494 days in one [1], 36 months in the other [2]), so they may miss late-emerging risks. For solid tumors, the research is still preclinical or early-stage—the scaffold study [3] and antigen discovery work [4] are not designed to measure long-term safety in humans. A 2024 review explicitly states that side effects from CAR-T in solid tumors can be 'potentially severe' and that the tumor microenvironment poses major limitations [5]. However, no study here directly measures long-term risks in solid tumor patients, so the claim of underestimation is based on extrapolation from blood cancer data and the known biological hurdles. The honest answer is that we don't yet have enough human data to quantify the risk, but the mechanisms suggest it could be significant.
About These Sources
This answer is built on 5 peer-reviewed studies — published from 2022 to 2025, 4 from 2024 or later, 5 in Q1 journals, collectively cited 83 times — selected as the most relevant from 5 studies that passed quality screening, drawn from 56 papers retrieved from a database of over 500 million.
Sources used in this answer
Risk of second primary malignancy following chimeric antigen receptor T-cell therapy: A real-world cohort study
In a real-world cohort of 1,480 DLBCL patients per group, CAR-T therapy was associated with significantly higher rates of AML (3.23% vs. 1.18%) and MDS (2.93% vs. 1.11%) compared to non-CAR-T patients, but no significant difference in solid tumor incidence over a mean follow-up of 494 days.
Subsequent Malignancies After CD19-Targeted Chimeric Antigen Receptor T Cells in Patients With Lymphoma
In a retrospective study of 355 lymphoma patients across two centers, the cumulative incidence of any subsequent malignancy at 36 months was 14%, with solid tumors at 6.1%, hematologic at 4.5%, and dermatologic at 4.2%; no T-cell malignancies were observed.
Lymph node-biomimetic scaffold boosts CAR-T therapy against solid tumor
A preclinical study using a 3D biomimetic scaffold for CAR-T delivery in a cervical tumor model achieved 50-fold in vitro and 15-fold in vivo CAR-T cell expansion, with sustained activity for up to 30 days, suggesting potential for prolonged immune activation.
Abstract IA011: Identifying tumor-restricted target antigens for adoptive cellular immunotherapy to treat Ewing Sarcoma using multi-omic discovery platforms
In Ewing sarcoma, transcriptomic and immunopeptidome analysis identified 24 overexpressed genes, but only 11 were cell-surface targets suitable for CAR-T; most potential antigens were oncofetal or placental, raising off-tumor toxicity concerns.
The dilemmas and possible solutions for CAR-T cell therapy application in solid tumors
A 2024 review states that CAR-T response rates in solid tumors remain extremely low and side effects potentially severe, primarily due to the tumor microenvironment limiting infiltration and persistence.