ABSTRACT
Central nervous system (CNS) involvement in B‐cell acute lymphoblastic leukemia (B‐ALL) is associated with relapse, treatment refractoriness, and poor prognosis. Although chimeric antigen receptor (CAR) T cell therapy has demonstrated remarkable efficacy in relapsed/refractory (R/R) B‐ALL, its efficacy and safety in CNS leukemia (CNSL) remain unclear. We retrospectively analyzed 113 R/R B‐ALL patients who received CAR‐T cell therapy, stratifying them into CNS‐positive (n = 22) and CNS‐negative (n = 91) groups based on the presence of CNSL prior to infusion. At Day 28 after CAR‐T infusion, the overall complete remission (CR) rate was 81.8%, with no significant difference between groups. The incidence of cytokine release syndrome (CRS) and neurotoxicity was comparable. The 3‐year cumulative incidence of relapse (CIR), event‐free survival (EFS), and overall survival (OS) were similar between groups. However, CNSL patients exhibit a higher cumulative relapse rate following CAR‐T‐induced remission, with an increased risk of CNS relapse compared to patients without CNS involvement. Multivariate analysis identified allo‐HSCT as consolidation post‐CAR‐T as an independent protective factor for improved EFS and OS in CNSL patients. In conclusion, CAR‐T therapy offers similar efficacy and safety in R/R B‐ALL patients regardless of CNS involvement. Furthermore, CAR‐T cell therapy was not sufficient to maintain sustained remission, and consolidative allo‐HSCT may improve long‐term survival in this high‐risk group.
Keywords: central nervous system leukemia, chimeric antigen receptor, hematopoietic stem cell transplantation, relapsed/refractory B‐cell acute lymphoblastic leukemia
1. Introduction
Although most patients with B‐cell acute lymphoblastic leukemia (B‐ALL) achieve complete remission (CR) following initial induction therapy, approximately 50% eventually relapse [1], and about 20% exhibit primary refractory disease [2]. The treatment of relapsed or refractory B‐ALL (R/R B‐ALL) remains a major clinical challenge. These patients generally respond poorly to conventional chemotherapy, achieving limited remission and facing a high risk of subsequent relapse. Previous studies have shown that the overall response rate (ORR) to conventional chemotherapy in R/R B‐ALL patients ranges from 14% to 48%, with a median overall survival (OS) of only 3.0–6.5 months [3, 4, 5, 6]. As the number of prior lines of salvage therapy increases, the efficacy declines further—the ORR drops to approximately 11% and the median OS to merely 2.9 months in patients who have received three or more lines of therapy [7]. Chimeric antigen receptor T‐cell (CAR‐T) immunotherapy has emerged in recent years as a novel therapeutic strategy, offering new treatment prospects for patients with R/R B‐ALL. Multiple clinical studies [8] have demonstrated that CAR‐T therapy can achieve an ORR of 75%–90% in R/R B‐ALL, with CR rates ranging from 69% to 87%.
The central nervous system (CNS) represents the most common site of extramedullary infiltration in adult acute lymphoblastic leukemia (ALL). Although the incidence of CNS leukemia (CNSL) at initial diagnosis in adult ALL is relatively low (approximately 6%–7%) [9, 10], its presence generally indicates a poor prognosis. Previous studies have reported that adult ALL patients experiencing CNS relapse have a median OS of less than 1 year [11]. Nevertheless, the suitability of CAR‐T therapy for R/R B‐ALL patients with CNSL remains controversial. Due to concerns regarding the risk of neurotoxicity, most CAR‐T clinical trials have excluded patients with active CNSL, leaving these patients with very limited effective treatment options. Existing evidence suggests that in R/R B‐ALL patients with active CNS involvement receiving CD19‐targeted CAR‐T therapy, high CNS tumor burden is an independent risk factor for severe neurotoxicity [12, 13] and is significantly associated with shorter post‐infusion event‐free survival (EFS) [13]. Conversely, Leahy et al. [14] reported that CD19 CAR‐T therapy demonstrated the potential to eradicate CNS disease in pediatric and adolescent/young adult (AYA) with R/R B‐ALL and concomitant CNSL, without a significant increase in the risk of neurotoxicity.
However, most studies on CAR‐T therapy for R/R B‐ALL with CNSL have focused on pediatric and young patients, and the conclusions remain inconsistent. Therefore, this study aimed to systematically evaluate, through retrospective analysis, the efficacy and safety of CAR‐T therapy in pediatric and adult patients with R/R B‐ALL complicated by CNSL.
2. Patients and Methods
2.1. Population
This retrospective study included patients with R/R B‐ALL who received CAR‐T cell therapy at the First Affiliated Hospital of University of Science and Technology of China between January 2014 and December 2024. All patients received CAR‐T products containing the CD28 costimulatory domain; the majority were CD19‐targeted, while a small subset received dual‐targeting CAR‐T (CD19/CD20 or CD19/CD22). Due to the retrospective design and use of legacy trial data, detailed product characteristics beyond CD28 (e.g., transduction efficiency, vector copy number) and serial peripheral blood CAR‐T monitoring data (flow cytometry or qPCR) were not archived and are unavailable. A total of 22 patients with concomitant CNSL were included in the CNS‐positive group. For comparison, 91 contemporaneous R/R B‐ALL patients without a history of CNSL were included as the CNS‐negative control group. Clinical data, treatment responses, and long‐term survival outcomes were collected and analyzed to evaluate the therapeutic efficacy and prognosis between the two groups. This study was approved by the medical ethics committee of The First Affiliated Hospital of University of Science and Technology of China (Approval No. 2026‐RE‐158). Due to the retrospective nature of the study and the use of anonymized clinical data, the requirement for informed consent was waived by the ethics committee.
2.2. Diagnosis and Management of CNSL
All CNSL episodes in this study occurred prior to study screening (i.e., before enrollment). No patient developed new or progressive CNSL during leukapheresis, bridging chemotherapy, lymphodepletion, or between lymphodepletion and CAR‐T infusion. CNSL was diagnosed according to previously established criteria when at least one of the following conditions was fulfilled: (1) definitive cytological evidence of leukemic blasts in cerebrospinal fluid (CSF) identified by morphological examination, regardless of cell count; (2) neurological manifestations consistent with CNS involvement, such as cranial nerve palsy or other focal neurological deficits, after exclusion of other potential causes; or (3) radiological findings suggestive of leukemic infiltration in the brain parenchyma, meninges, or spinal cord, as demonstrated by contrast‐enhanced magnetic resonance imaging (MRI).
All patients diagnosed with CNSL received triple intrathecal chemotherapy prior to CAR‐T cell infusion, consisting of cytarabine (Ara‐C), methotrexate (MTX), and dexamethasone (DXM). For patients with intracranial mass lesions or persistent leukemic blasts in the CSF, cranial or craniospinal irradiation was administered in combination with intrathecal chemotherapy. CAR‐T cell infusion was performed only after the resolution of active CNSL.
2.3. Definitions
High disease burden referred to more than 5% blasts in bone marrow or with extramedullary disease, and low disease burden was identified as less than 5% blasts in bone marrow but with MRD positive.
OS was defined as the time from CAR‐T cell infusion to death from any cause or last follow‐up. EFS was defined as the time from the date of CAR‐T cell infusion to the earliest occurrence of any of the following: lack of remission at Day 28 post‐infusion, relapse, death from any cause, or last follow‐up. Non‐relapse mortality (NRM) was defined as death without evidence of relapse or disease progression after CAR‐T cell infusion, with relapse treated as a competing event. Cumulative incidence of relapse (CIR) was defined as the time from CAR‐T cell infusion to the first documented relapse or disease progression, with death without relapse considered a competing event. Cytokine release syndrome (CRS) and immune effector cell–associated neurotoxicity syndrome (ICANS) were graded according to the 2019 American Society for Transplantation and Cellular Therapy (ASTCT) consensus criteria [15]. CRS grading was based on fever, hypotension, and hypoxia, while ICANS was assessed using the Immune Effector Cell‐Associated Encephalopathy (ICE) score, level of consciousness, seizure activity, motor findings, and evidence of cerebral edema.
2.4. Statistical Analysis
Categorical variables were compared using the chi‐square test or Fisher's exact test, and continuous variables were compared using the Mann–Whitney U test or Student's t‐test, as appropriate. OS and EFS were estimated using the Kaplan–Meier method, and differences between groups were assessed with the log‐rank test. The CIR and NRM were analyzed using competing risk models. In the estimation of CIR, death without relapse was considered a competing event, whereas relapse was treated as a competing event for NRM. Cumulative incidence curves were constructed using the Fine‐Gray subdistribution hazard model to account for competing risks, and group comparisons were performed using Gray's test.
Multivariate analyses were performed using the Cox proportional hazards model, and results were reported as hazard ratios (HR) with 95% confidence intervals (CI). All statistical analyses were conducted using SPSS software (version 26.0) and R software (version 4.2.2). Two‐sided p values < 0.05 were considered statistically significant.
3. Results
3.1. Patient Characteristics
A total of 113 patients with R/R B‐ALL who received CAR‐T cell therapy were included in this study. Patients were stratified into two groups according to the presence or absence of CNSL prior to CAR‐T infusion: 22 patients in the CNS‐positive group and 91 in the CNS‐negative group. The median age at infusion was 24 years (range, 6–55 years) in the CNS‐positive group and 28 years (range, 2–66 years) in the CNS‐negative group. Of the 22 patients with CNSL, 21 (95.5%) had isolated leptomeningeal involvement (positive CSF cytology and/or leptomeningeal enhancement on MRI), and only one patient (4.5%) had parenchymal involvement (intracranial mass lesion).
Genetic mutations or chromosomal abnormalities were detected in 12 of 21 evaluable patients (57.1%) in the CNS‐positive group (one patient with bone marrow dry tap was excluded) and in 56 of 91 patients (61.5%) in the CNS‐negative group. Philadelphia chromosome positivity (Ph+) was found in 9 of 21 evaluable patients (42.9%) in the CNS‐positive group and in 24 of 91 patients (26.4%) in the CNS‐negative group. The T315I mutation, which confers resistance to first‐ and second‐generation tyrosine kinase inhibitors (TKIs), was detected in 15 of these 33 Ph + patients (45.5%): 5 cases in the CNS‐positive group (5/9, 55.6% of Ph + in this group) and 10 cases in the CNS‐negative group (10/24, 41.7% of Ph + in this group). The most frequently involved genes included BCR/ABL, T315I, E2A/PBX1, WT1, MLL/AF4, TLS‐ERG, HOX11, CREBBP, FLT3, KRAS, Tyr253His, and TEL/AML1.
Before lymphodepleting conditioning, five patients (22.7%) in the CNS‐positive group received bridging chemotherapy due to high tumor burden, compared with 13 patients (14.3%) in the CNS‐negative group. Among the latter, two patients developed severe myelosuppression after bridging chemotherapy and could not complete the planned lymphodepleting regimen before CAR‐T infusion.
Baseline clinical characteristics of the two groups are summarized in Table 1. No significant differences were observed between the CNS‐positive and CNS‐negative groups with respect to age, sex, genetic mutations or cytogenetic abnormalities, pre‐CAR‐T infusion disease status, number of relapses, bone marrow blast percentage, or lymphodepletion chemotherapy regimen (all p > 0.05).
TABLE 1.
Clinical characteristics of patients with R/R B‐ALL.
| Clinical characteristics | CNS‐positive (n = 22) | CNS‐negative (n = 91) | Statistic | p |
|---|---|---|---|---|
| Age | 0.092 | 0.762 | ||
| < 18 year | 8 (36.4) | 30 (33.0) | ||
| ≥ 18 year | 14 (63.6) | 61 (67.0) | ||
| Sex | 3.791 | 0.052 | ||
| Female | 7 (31.8) | 50 (54.9) | ||
| Male | 15 (68.2) | 41 (45.1) | ||
| Gene mutation a /chromosome abnormality | 0.362 | 0.548 | ||
| None | 9 (42.9) | 35 (38.5) | ||
| Present | 12 (57.1) | 56 (61.5) | ||
| Philadelphia chromosome | 1.810 | 0.178 | ||
| Negative | 12 (57.1) | 67 (73.6) | ||
| Positive | 9 (42.9) | 24 (26.4) | ||
| Previous transplantation | 1.363 | 0.239 | ||
| No | 16 (72.7) | 76 (83.5) | ||
| Yes | 6 (27.3) | 15 (16.5) | ||
| Pre‐CAR‐T infusion disease status | 1.483 | 0.390 | ||
| MRD‐positive | 4 (8.2) | 13 (14.3) | ||
| No Remission | 17 (77.3) | 77 (84.6) | ||
| Extramedullary involvement | 1 (4.5) | 1 (1.1) | ||
| Number of relapses before infusion | 2.133 | 0.144 | ||
| < 2 times | 9 (40.9) | 68 (74.7) | ||
| ≥ 2 times | 13 (59.1) | 23 (25.3) | ||
| Bone marrow blasts (%) before infusion | 1.456 | 0.491 | ||
| < 5% | 6 (27.3) | 15 (16.5) | ||
| 5%–25% | 4 (18.2) | 22 (24.2) | ||
| > 25% | 12 (54.5) | 54 (59.3) | ||
| Lymphodepletion chemotherapy regimen | 2.629 | 0.274 | ||
| None | 0 (0.0) | 2 (2.2) | ||
| FC | 17 (77.3) | 54 (59.3) | ||
| Decitabine + FC | 5 (22.7) | 35 (38.5) | ||
| CAR‐T target | 0.083 | 1.000 | ||
| CD19 | 20 (90.9) | 84 (92.3) | ||
| CD19/CD20 | 1 (4.5) | 3 (3.3) | ||
| CD19/CD22 | 1 (4.5) | 4 (4.4) |
Note: One patient in the CNS‐positive group was excluded from genetic analysis due to bone marrow dry tap. Percentages for genetic and chromosome abnormalities in this group are based on 21 evaluable patients.
Abbreviations: CAR‐T, chimeric antigen receptor T cells; CNS, central nervous system; FC, Fludarabine + Cyclophosphamide; MRD, minimal residual disease.
Included: BCR/ABL, T315I, E2A/PBX1, WT1, MLL/AF4, TLS‐ERG, HOX11, CREBBP, FLT3, KRAS, Tyr253His, TEL/AML1.
3.2. CAR‐T‐Related Toxicities
Adverse events following CAR‐T cell infusion are summarized in Table 2. In the CNS‐positive group, the overall incidence of CRS was 86.4% (19/22), with 18.2% (4/22) experiencing grade ≥ 3 CRS. ICANS occurred in 13.6% (3/22) of patients, with 4.5% (1/22) experiencing grade ≥ 3 ICANS. In the CNS‐negative group, the overall incidence of CRS was 84.6% (77/91), with grade ≥ 3 CRS in 24.2% (22/91). ICANS was observed in 9.9% (9/91) of patients, with 5.5% (5/91) experiencing grade ≥ 3 ICANS. No significant differences were observed between the two groups in overall incidence of CRS, grade ≥ 3 CRS, or the incidence of ICANS and grade ≥ 3 ICANS (p > 0.05).
TABLE 2.
CAR‐T cell infusion‐related adverse events in R/R B‐ALL patients.
| Adverse events | CNS‐positive (n = 22) | CNS‐negative (n = 91) | p |
|---|---|---|---|
| CRS | 0.123 | ||
| Grade 0 | 3 (13.6) | 14 (15.4) | |
| Grade 1 | 14 (63.6) | 32 (35.2) | |
| Grade 2 | 1 (4.5) | 23 (25.3) | |
| Grade 3 | 4 (18.2) | 17 (18.7) | |
| Grade 4 | 0 (0.0) | 4 (4.4) | |
| Grade 5 | 0 (0.0) | 1 (1.1) | |
| ICANS | 0.352 | ||
| Grade 0 | 19 (86.4) | 82 (90.1) | |
| Grade 1 | 0 (0.0) | 1 (1.1) | |
| Grade 2 | 2 (9.1) | 3 (3.3) | |
| Grade 3 | 0 (0.0) | 4 (4.4) | |
| Grade 4 | 1 (4.5) | 1 (1.1) |
Abbreviations: CNS, central nervous system; CRS, cytokine release syndrome; ICANS, immune effector cell‐associated neurotoxicity syndrome.
3.3. Treatment Response and Survival Outcomes
At a median follow‐up of 11.6 months (range, 0.2–84.7), the day‐28 CR rate after CAR‐T cell infusion was 81.8% (18/22) in the CNS‐positive group and 73.6% (67/91) in the CNS‐negative group, with no significant difference (p = 0.820, Table 3). In the CNS‐positive group, three patients (13.6%) were classified as no response (NR), and one patient (4.5%) died of severe pulmonary infection prior to assessment. In the CNS‐negative group, 19 patients (20.9%) were classified as NR, and five patients (5.5%) died before evaluation, including one due to severe CRS with ICANS and four due to severe pulmonary infections. Among patients achieving CR, relapse occurred in 72.2% (13/18) of CNS‐positive and 64.2% (43/67) of CNS‐negative patients, with similar 3‐year CIR (72.2% vs. 65.1%, p = 0.990, Figure 1A). CNS relapse was significantly higher in the CNS‐positive group than in the CNS‐negative group (22.7% vs. 2.2%, p = 0.003, Table 3).
TABLE 3.
Clinical outcomes of patients with R/R B‐ALL following CAR‐T cell infusion.
| Clinical outcome | CNS‐positive (n = 22) | CNS‐negative (n = 91) | p |
|---|---|---|---|
| Disease response at Day 28 post‐infusion, n (%) | 0.820 | ||
| Complete remission | 18 (81.8) | 67 (73.6) | |
| No remission | 3 (13.6) | 19 (20.9) | |
| Not evaluated | 1 (4.5) | 5 (5.5) | |
| Disease relapse, n (%) | 13/18 (72.2) | 43/67 (64.2) | 0.599 |
| CNS relapse, n (%) | 5/22 (22.7) | 2/91 (2.2) | 0.003 |
| 3‐year CIR rate (95% CI) | 72.2 (43.3–88.1) | 65.1 (51.3–75.9) | 0.990 |
| 3‐year EFS rate (95% CI) | 18.2 (7.5–44.1) | 22.2 (14.8–33.5) | 0.810 |
| 3‐year OS rate (95% CI) | 46.4 (28.0–76.7) | 42.9 (32.0–56.4) | 0.960 |
| 3‐year NRM rate (95% CI) | 15.5 (3.5–35.3) | 18.5 (10.5–28.3) | 0.780 |
Abbreviations: CI, confidence interval; CIR, cumulative incidence of relapse; CNS, central nervous system; EFS, event‐free survival; NRM, non‐relapse mortality; OS, overall survival.
FIGURE 1.
Competing risk analysis by CNS status after CAR‐T therapy in R/R B‐ALL: (A) Cumulative incidence of relapse (CIR); (B) Non‐relapse mortality (NRM).
Long‐term outcomes were comparable between the groups. The 3‐year EFS rates were 18.2% (95% CI, 7.5–44.1) versus 22.2% (95% CI, 14.8–33.5; p = 0.810, Figure 2A), the 3‐year OS rates were 46.4% (95% CI, 28.0–76.7) versus 42.9% (95% CI, 32.7–56.4; p = 0.960, Figure 2B), and the 3‐year NRM rates were 15.5% (95% CI, 3.5–35.3) versus 18.5% (95% CI, 10.5–28.3; p = 0.780, Figure 1B). None of these differences were statistically significant.
FIGURE 2.
Long‐term survival of R/R B‐ALL patients treated with CAR‐T cells according to CNS status: (A) Event‐free survival (EFS); (B) Overall survival (OS).
3.4. Post‐Remission Treatment
After CAR‐T infusion, TKI therapy (dasatinib, ponatinib, or flumatinib) was administered to 4 of 9 Ph + patients (44.4%) in the CNS‐positive group and to 9 of 24 Ph + patients (37.5%) in the CNS‐negative group. Of note, among the 91 CNS‐negative patients, 22 patients (24.2%) underwent consolidative allogeneic hematopoietic stem cell transplantation (allo‐HSCT) after achieving remission. These 22 patients had significantly higher 3‐year EFS (65.0% vs. 26.2%, p = 0.003) and 3‐year OS (85.7% vs. 32.0%, p = 0.001) compared with those who did not receive allo‐HSCT. In the CNS‐positive group, 7 of 22 patients (31.8%) received consolidative allo‐HSCT.
3.5. Prognostic Factors in CNSL Patients Following CAR‐T Therapy
Among the 22 R/R B‐ALL patients with CNSL who received CAR‐T cell therapy, the median EFS was 5.7 months, and the median OS was 19.1 months. The 1‐ and 2‐year EFS rates were 31.8% and 18.2%, respectively, while the corresponding OS rates were 64.8% and 46.4% (Figure 3A,B). The cumulative relapse rate was 72.2% (13/18), with 10 relapses involving the bone marrow, 2 isolated to the CNS, and 1 extramedullary, presenting as abnormal signals in the lumbar spine, suggesting possible leukemic infiltration in the spinal or epidural region. The cumulative mortality rate was 50.0% (11/22); among these, 1 death was due to severe CRS and ICANS, 2 occurred from transplant‐related complications during consolidative allo‐HSCT (infection, acute graft‐versus‐host disease, and hemorrhagic cystitis), and 9 resulted from severe infections during salvage chemotherapy.
FIGURE 3.
Long‐term survival of R/R B‐ALL patients with CNSL after CAR‐T therapy: (A) event‐free survival (EFS); (B) overall survival (OS); (C) EFS by allo‐HSCT after CAR‐T; (D) OS by allo‐HSCT after CAR‐T.
Multivariate Cox regression analysis revealed that previous allo‐HSCT was an independent protective factor for EFS (HR = 0.150, 95% CI: 0.025–0.913, p = 0.039) in patients with CNSL. Consolidative allo‐HSCT after CAR‐T therapy was an independent protective factor for both EFS (HR = 0.128, 95% CI: 0.019–0.849, p = 0.033) and OS (HR = 0.052, 95% CI: 0.003–0.921, p = 0.044; Table 4). Among the 22 patients with CNS involvement, 6 (27.3%) had undergone prior allo‐HSCT and 7 (31.8%) received allo‐HSCT as consolidation following CAR‐T therapy, including six who underwent umbilical cord blood transplantation (UCBT). Both prior and consolidative allo‐HSCT were associated with improved survival outcomes. Among patients who underwent consolidative allo‐HSCT, the median EFS and OS were 8.2 and 16.3 months, respectively, with a 1‐year OS rate of 71.4%. At the last follow‐up, 3 of 7 patients (42.9%) remained in continuous remission, 2 (28.6%) died of transplant‐related complications, and 2 (28.6%) were lost to follow‐up. Subgroup analysis demonstrated a higher 6‐month EFS rate in the transplantation group compared with the non‐transplantation group (57.1% vs. 40.0%), while the 1‐year EFS rates were comparable (28.6% vs. 33.3%). The 1‐year OS rate also tended to be higher in the transplantation group (85.7% vs. 54.8%), however, the differences were not statistically significant (Log‐rank test, p > 0.05; Figure 3C,D).
TABLE 4.
Multivariate analysis of factors associated with EFS and OS in patients with CNSL.
| Variable | EFS | OS | ||||
|---|---|---|---|---|---|---|
| HR | 95% CI | p | HR | 95% CI | p | |
| Age (> 18 year vs. ≤ 18 year) | 0.883 | 0.088–8.854 | 0.916 | 0.467 | 0.017–12.500 | 0.650 |
| Gender (Male vs. Female) | 0.946 | 0.188–4.764 | 0.946 | 0.569 | 0.112–2.886 | 0.496 |
| Gene mutation/chromosomal abnormality (Yes vs. No) | 0.159 | 0.024–1.041 | 0.055 | 0.296 | 0.009–9.437 | 0.490 |
| Previous transplantation (Yes vs. No) | 0.150 | 0.025–0.913 | 0.039 | 0.442 | 0.023–8.360 | 0.586 |
| Radiotherapy (Yes vs. No) | 0.926 | 0.189–4.539 | 0.924 | 0.120 | 0.002–7.294 | 0.312 |
| Tumor burden before infusion (High vs. Low) | 2.092 | 0.152–28.728 | 0.581 | 21.967 | 0.132–3666.285 | 0.237 |
| Conditioning regimen (FC + Decitabine vs. FC) | 2.366 | 0.302–18.509 | 0.412 | 0.898 | 0.128–6.291 | 0.914 |
| Severe CRS (Yes vs. No) | 0.135 | 0.011–1.723 | 0.123 | 1.632 | 0.055–48.835 | 0.777 |
| Neurotoxicity (Yes vs. No) | 0.558 | 0.011–29.071 | 0.772 | 1.009 | 0.026–39.553 | 0.996 |
| Post–CAR‐T transplantation (Yes vs. No) | 0.128 | 0.019–0.849 | 0.033 | 0.052 | 0.003–0.921 | 0.044 |
| Second CAR‐T therapy (Yes vs. No) | 0.888 | 0.174–4.524 | 0.887 | 0.129 | 0.008–2.183 | 0.156 |
| Post‐CAR‐T TKI use (Yes vs. No) | 5.014 | 0.505–49.762 | 0.169 | 2.325 | 0.155–34.894 | 0.542 |
Abbreviations: CAR‐T, chimeric antigen receptor T cell; CI, Confidence interval; CRS, Cytokine release syndrome; EFS, event‐free survival; FC, Fludarabine + cyclophosphamide; HR, Hazard ratio; OS, overall survival; TKI, Tyrosine kinase inhibitor.
4. Discussion
Traditional intensive chemotherapy combined with allo‐HSCT has historically been the only curative option for R/R B‐ALL. While intensive chemotherapy regimens achieve complete morphological remission in approximately 60% of patients, only 48% maintain remission and survive long enough to undergo transplantation [16]. In recent years, CD19‐targeted CAR‐T therapy has significantly improved remission rates in R/R B‐ALL. Several multicenter studies have demonstrated that CD19‐targeted CAR‐T therapy achieves CR in 70%–90% of both pediatric and adult R/R B‐ALL patients [17, 18, 19, 20, 21]. Notably, a phase 1b/2 clinical trial reported a 90% overall response rate with autologous CD19‐targeted CAR‐T therapy in R/R ALL, with sustained anti‐leukemia activity even in patients with extramedullary disease, including those with CNS involvement [20]. However, most prospective studies have excluded patients with active CNSL, resulting in limited data on the early efficacy and safety of CAR‐T in this population. Additionally, existing reports predominantly focus on pediatric and AYA populations, with a lack of comprehensive evaluations across all age groups.
This retrospective study included patients of all ages with R/R B‐ALL, comparing the early efficacy and long‐term outcomes of CAR‐T therapy in those with and without CNS involvement. The results showed that by Day 28 post‐CAR‐T infusion, the CR rates were similar between the CNS‐positive and CNS‐negative groups, suggesting that CNS involvement does not impair the anti‐leukemia activity of CAR‐T therapy. In terms of safety, the incidence of CRS and ICANS was comparable between the two groups, and no increase in NRM due to additional neurotoxicity was observed in the CNS‐positive group. Long‐term follow‐up analysis revealed no statistically significant differences in 3‐year EFS, OS, NRM, or CIR between the two groups (p > 0.05). These findings suggest that, provided CNS disease is adequately controlled prior to CAR‐T infusion with conventional CNS‐directed therapies (including intrathecal chemotherapy, high‐dose systemic chemotherapy, or cranial radiation), patients with CNSL can achieve early remission rates and long‐term survival benefits comparable to those without CNS involvement. These results are consistent with previous studies in pediatric and AYA population [12, 13, 14]. Therefore, for traditionally defined “high‐risk” CNSL patients, CAR‐T therapy remains a safe, feasible, and effective anti‐leukemia strategy, which can be used either as a bridging therapy to allo‐HSCT or as a maintenance treatment transition.
Although CAR‐T therapy can rapidly induce remission in patients with R/R B‐ALL, the high relapse rate remains a significant challenge as follow‐up time extends [22]. Aamir et al. [16] reported that while CAR‐T therapy can achieve a high CR rate, the cumulative relapse rate after remission remains as high as 36%, highlighting its limited long‐term efficacy. In this study, the overall 3‐year EFS was only 22.2% in the CNS‐negative cohort. For patients with R/R B‐ALL and concurrent CNSL, the relapse rate significantly increased following CAR‐T treatment, reaching 72.2%, with a median EFS of only 5.7 months and a 2‐year EFS rate of just 18.2%. Notably, the risk of CNS relapse was particularly pronounced in this group, likely due to the physical barrier of the blood–brain barrier and the unique immune microenvironment in the CNS, which may hinder the effective clearance of residual leukemia cells. Therefore, optimizing CAR‐T therapy to achieve long‐term remission and reduce the risk of CNS relapse in R/R B‐ALL patients with CNSL remains an urgent challenge.
It is noteworthy that several studies have shown that consolidation with allo‐HSCT following CAR‐T therapy can effectively reduce relapse risk and improve long‐term survival outcomes [19, 23, 24]. Notably, only 24.2% of CNS‐negative patients proceeded to consolidative allo‐HSCT. Patients who received transplantation experienced markedly higher 3‐year EFS (65.0% vs. 26.2%, p = 0.003) and 3‐year OS (85.7% vs. 32.0%, p = 0.001) than those who did not, indicating a strong survival benefit associated with post‐CAR‐T transplantation. Multivariate analysis in this study further confirmed that consolidative allo‐HSCT after CAR‐T therapy is an independent protective factor for longer EFS and OS in R/R B‐ALL patients with CNSL. The 6‐month EFS rate in the transplant group was higher than in the non‐transplant group; however, due to the limited sample size, the difference did not reach statistical significance. Existing evidence suggests that consolidative allo‐HSCT after CAR‐T cell therapy may provide significant survival benefits for patients with CNSL. However, the potential value of reducing relapse risk remains to be validated by multi‐center prospective studies.
This study has several limitations. First, due to the retrospective design and the use of legacy clinical trial data, detailed CAR‐T product characteristics beyond the CD28 costimulatory domain (e.g., transduction efficiency, vector copy number) and serial peripheral blood CAR‐T monitoring data were not available, which precluded an assessment of CAR‐T persistence in vivo. Second, because no patient had active CNSL before CAR‐T infusion, lumbar puncture for CAR‐T detection in cerebrospinal fluid was not performed routinely; consequently, direct evidence of CAR‐T cells within the CNS is lacking. Third, the number of patients with parenchymal involvement was too small to allow subgroup analyses comparing treatment responses or toxicity profiles between leptomeningeal and parenchymal CNS leukemia. Fourth, the small sample size of the CNSL‐positive cohort (n = 22) limited statistical power, and some hazard ratio estimates had wide confidence intervals. Prospective studies with standardized product characterization, longitudinal sampling, and larger cohorts are needed to validate our findings.
In conclusion, CAR‐T therapy provides comparable early efficacy, safety, and long‐term survival benefits in patients with R/R B‐ALL, regardless of CNS involvement. However, CAR‐T therapy struggles to maintain sustained long‐term remission in patients with CNSL, with a notably higher risk of CNS relapse. Allo‐HSCT as consolidation following CAR‐T therapy may be a key strategy to improve long‐term survival in this patient population. Therefore, following CAR‐T‐induced remission, a comprehensive evaluation of the feasibility of consolidative transplantation should be proactively considered.
Author Contributions
Yanghong Yu: investigation, validation, software, writing – original draft, visualization, resources, writing – review and editing. Hui Xu: resources, investigation. Lei Xue: investigation, resources. Xingbing Wang: writing – review and editing, conceptualization.
Funding
This study was funded by the National Natural Science Foundation of China (Grant No. 82170221).
Conflicts of Interest
The authors declare no conflicts of interest.
Data Availability Statement
The data utilized in this study were derived from a single‐center retrospective analysis and are not publicly available due to privacy and ethical considerations.
References
- 1. Paul S., Kantarjian H., and Jabbour E. J., “Adult Acute Lymphoblastic Leukemia,” Mayo Clinic Proceedings 91 (2016): 1645–1666. [DOI] [PubMed] [Google Scholar]
- 2. Maffini E., Saraceni F., and Lanza F., “Treatment of Adult Patients With Relapsed/Refractory B‐Cell Philadelphia‐Negative Acute Lymphoblastic Leukemia,” Clinical Hematology International 1 (2019): 85–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Liedtke M., Dunn T., Dinner S., et al., “Salvage Therapy With Mitoxantrone, Etoposide and Cytarabine in Relapsed or Refractory Acute Lymphoblastic Leukemia,” Leukemia Research 38 (2014): 1441–1445. [DOI] [PubMed] [Google Scholar]
- 4. Ahn J.‐S., Yang D.‐H., Jung S.‐H., et al., “Clinical Efficacy of Mitoxantrone and Ara‐C With or Without Etoposide Salvage Chemotherapy in Adult Patients With Relapsed or Refractory Acute Lymphoblastic Leukemia: Retrospective Multicenter Study of the Korean Adult ALL Working Party,” Acta Haematologica 133 (2014): 91–97. [DOI] [PubMed] [Google Scholar]
- 5. Faderl S., Balakrishnan K., Thomas D. A., et al., “Phase I and Extension Study of Clofarabine Plus Cyclophosphamide in Patients With Relapsed /Refractory Acute Lymphoblastic Leukemia,” Clinical Lymphoma Myeloma and Leukemia 14 (2014): 231–238. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. O'Brien S., Schiller G., Lister J., et al., “High‐Dose Vincristine Sulfate Liposome Injection for Advanced, Relapsed, and Refractory Adult Philadelphia Chromosome–Negative Acute Lymphoblastic Leukemia,” Journal of Clinical Oncology 31 (2013): 676–683. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Gokbuget N., Dombret H., Ribera J.‐M., et al., “International Reference Analysis of Outcomes in Adults With B‐Precursor Ph‐Negative Relapsed/Refractory Acute Lymphoblastic Leukemia,” Haematologica 101 (2016): 1524–1533. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Testa U., Sica S., Pelosi E., Castelli G., and Leone G., “Car‐T Cell Therapy in B‐Cell Acute Lymphoblastic Leukemia,” Mediterranean Journal of Hematology and Infectious Diseases 16 (2024): e2024010. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Pui C.‐H. and Thiel E., “Central Nervous System Disease in Hematologic Malignancies: Historical Perspective and Practical Applications,” Seminars in Oncology 36 (2009): S2–S16. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Reman O., Pigneux A., Huguet F., et al., “Central Nervous System Involvement in Adult Acute Lymphoblastic Leukemia at Diagnosis and/or at First Relapse: Results From the GET‐LALA Group,” Leukemia Research 32 (2008): 1741–1750. [DOI] [PubMed] [Google Scholar]
- 11. Kopmar N. E. and Cassaday R. D., “How I Prevent and Treat Central Nervous System Disease in Adults With Acute Lymphoblastic Leukemia,” Blood 141 (2023): 1379–1388. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Jacoby E., Ghorashian S., Vormoor B., et al., “CD19 CAR T‐Cells for Pediatric Relapsed Acute Lymphoblastic Leukemia With Active CNS Involvement: A Retrospective International Study,” Leukemia 36 (2022): 1525–1532. [DOI] [PubMed] [Google Scholar]
- 13. Qi Y., Zhao M., Hu Y., et al., “Efficacy and Safety of CD19‐Specific CAR T Cell–Based Therapy in B‐Cell Acute Lymphoblastic Leukemia Patients With CNSL,” Blood 139 (2022): 3376–3386. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Leahy A. B., Newman H., Li Y., et al., “CD19‐Targeted Chimeric Antigen Receptor T‐Cell Therapy for CNS Relapsed or Refractory Acute Lymphocytic Leukaemia: A Post‐Hoc Analysis of Pooled Data From Five Clinical Trials,” Lancet Haematology 8 (2021): e711–e722. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Lee D. W., Santomasso B. D., Locke F. L., et al., “ASTCT Consensus Grading for Cytokine Release Syndrome and Neurologic Toxicity Associated With Immune Effector Cells,” Biology of Blood and Marrow Transplantation 25 (2019): 625–638. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Aamir S., Anwar M. Y., Khalid F., Khan S. I., Ali M. A., and Khattak Z. E., “Systematic Review and Meta‐Analysis of CD19‐Specific CAR‐T Cell Therapy in Relapsed/Refractory Acute Lymphoblastic Leukemia in the Pediatric and Young Adult Population: Safety and Efficacy Outcomes,” Clinical Lymphoma Myeloma and Leukemia 21 (2021): e334–e347. [DOI] [PubMed] [Google Scholar]
- 17. Park J. H., Rivière I., Gonen M., et al., “Long‐Term Follow‐Up of CD19 CAR Therapy in Acute Lymphoblastic Leukemia,” New England Journal of Medicine 378 (2018): 449–459. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Maude S. L., Laetsch T. W., Buechner J., et al., “Tisagenlecleucel in Children and Young Adults With B‐Cell Lymphoblastic Leukemia,” New England Journal of Medicine 378 (2018): 439–448. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Hay K. A., Gauthier J., Hirayama A. V., et al., “Factors Associated With Durable EFS in Adult B‐Cell ALL Patients Achieving MRD‐Negative CR After CD19 CAR T‐Cell Therapy,” Blood 133 (2019): 1652–1663. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Jacoby E., Bielorai B., Avigdor A., et al., “Locally Produced CD19 CAR T Cells Leading to Clinical Remissions in Medullary and Extramedullary Relapsed Acute Lymphoblastic Leukemia,” American Journal of Hematology 93 (2018): 1485–1492. [DOI] [PubMed] [Google Scholar]
- 21. Shah B. D., Bishop M. R., Oluwole O. O., et al., “KTE‐X19 Anti‐CD19 CAR T‐Cell Therapy in Adult Relapsed/Refractory Acute Lymphoblastic Leukemia: ZUMA‐3 Phase 1 Results,” Blood 138 (2021): 11–22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Wang J., Hu Y., and Huang H., “Acute Lymphoblastic Leukemia Relapse After CD19‐Targeted Chimeric Antigen Receptor T Cell Therapy,” Journal of Leukocyte Biology 102 (2017): 1347–1356. [DOI] [PubMed] [Google Scholar]
- 23. Pan J., Yang J. F., Deng B. P., et al., “High Efficacy and Safety of Low‐Dose CD19‐Directed CAR‐T Cell Therapy in 51 Refractory or Relapsed B Acute Lymphoblastic Leukemia Patients,” Leukemia 31 (2017): 2587–2593. [DOI] [PubMed] [Google Scholar]
- 24. Jiang H., Li C., Yin P., et al., “Anti‐CD19 Chimeric Antigen Receptor‐Modified T‐Cell Therapy Bridging to Allogeneic Hematopoietic Stem Cell Transplantation for Relapsed/Refractory B‐Cell Acute Lymphoblastic Leukemia: An Open‐Label Pragmatic Clinical Trial,” American Journal of Hematology 94 (2019): 1113–1122. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data utilized in this study were derived from a single‐center retrospective analysis and are not publicly available due to privacy and ethical considerations.