3.1 Patient characteristics

We retrospectively analyzed 20 r/r B-ALL patients who received CAR-T treatment between June 2017 and September 2021 at the First Affiliated Hospital of Soochow University. T Ten patients received Rituximab treatment before CAR-T cell infusion (Rituximab group), while the remaining ten patients did not receive rituximab treatment (Control group). The demographic and clinical characteristics of the r/r B-ALL patients are shown in Table 1, and there were no significant differences in baseline characteristics between the two groups. Both groups received treatment during the same period, ensuring no significant differences in supportive care (Table 1). Before treatment, the expression levels of CD19, CD20, and CD22 on the surface of leukemia cells were measured by flow cytometry in both groups (Rituximab group: CD19: 65.65%, CD20: 49.47%, CD22: 74.12%; Control group: CD19: 83.1%, CD20: 6.84%, CD22: 67.08%) (Fig. 1A).

Table 1 Patient characteristics3.2 Clinical response

28 days after CAR-T cells infusion, 8 out of 10 patients (80%) in the Rituximab group achieved minimal residual disease negative complete remission (MRD-negative CR), maintaining CR for a median time of 11.15 months (0.46–46.93) post-CAR-T therapy. In comparison, 6 out of 10 patients (60%) in the Control group achieved MRD-negative CR, maintaining CR for a median time of 5.03 months (0.13–28) post-CAR-T therapy (p = 0.158) (Table 1, Fig. 1B). At the time of analysis, 8 out of 10 patients (80%) remained alive in the Rituximab group, while 3 out of 10 patients (30%) remained alive in the Control group (Fig. 1C). Among the patients in the Rituximab group, 6 (60%) proceeded to allo-HSCT, with 5 out of 6 (83.3%) patients achieving CR and survival. In contrast, 8 patients (80%) in the Control group underwent allo-HSCT, but only 3 out of 8 (37.5%) patients were alive with CR (Fig. 1D). Immune-modulating cytokines, including TNF-α, IFN-γ, IL-2, IL-6, IL-4, and IL-10, were induced in patients following 19–22 CAR-T and rituximab infusion, showing more significant elevation in the Rituximab Group compared to the Control Group (Fig. 1E).

3.3 Long-term survival

The median follow-up periods for the Rituximab and Control groups were 29.27 and 9.83 months, respectively. Our findings suggest that adding Rituximab may lead to a more favorable prognosis compared to the Control group. The 2-year overall survival (OS) and leukemia-free survival (LFS) rates were both higher in the Rituximab group compared to the Control group (90% vs 26.7%, p = 0.0342; 41.7% vs 25%, p = 0.308) (Fig. 1G). Subgroup analysis revealed that Rituximab significantly improved OS and LFS for relapsed patients (88.9% vs 25%, p = 0.0203; 87.5% vs 15%, p = 0.0112) (Fig. 2A, B). Furthermore, the combination of CAR-T and Rituximab improved long-term prognosis in patients who had failed multiple lines of therapy (>5) (OS 100% vs 28.5%, p = 0.0205; LFS 50% vs 0%, p = 0.0119) (Fig. 2C, D). Following CAR-T treatment, subgroups of patients who underwent bridging allo-HSCT or not were analyzed. The long-term prognosis of the Rituximab group was found to be superior to that of the Control group. For patients who received allo-HSCT after CAR-T therapy, OS in the Rituximab group was significantly longer than in the Control group (83.3% vs 33.3%, p = 0.0469) (Fig. 2E, F). For patients who did not undergo allo-HSCT after CAR-T therapy, LFS in the Rituximab group was significantly better than in the Control group (33.3% vs 0%, p = 0.0389) (Fig. 2G, H).

3.4 Safety

There were no significant differences between the two groups in terms of hematological and non-hematological toxicities. All patients experienced grade 1–3 cytokine release syndrome (CRS) symptoms, such as fever and hypotension, which were effectively managed through symptomatic treatment. Two patients in the Rituximab group developed headaches, while patients in the Control group did not exhibit any neurotoxicity syndromes. Moreover, all patients in our study experienced hematological toxicities. The median time to platelet and neutrophil count recovery was similar in both groups. Additionally, 50% (5/10) of patients in the Rituximab group experienced infections, compared to 30% (3/10) in the control group. All adverse events were reversible and manageable (Table 1).

3.5 Expansion and killing of CD19/CD22 CAR-T cells was improved by pre-treatment of tumor cells with Rituximab

Our clinical data indicated that Rituximab combined with CD19/CD22 CAR-T treatment effectively improved the long-term prognosis of r/r B-ALL patients (Fig. 3A). Subsequently, we evaluated the killing ability of Rituximab combined with CAR-T on tumor cells and the proliferation of CAR-T in vitro.

Firstly, we evaluated the tumor killing effect by selecting CD19, CD20, and CD22 simultaneously antigen-expressing B-ALL cells (NALM-6) as well as primary cells (Fig. 3B). Subsequently, we identified the concentration of Rituximab in vitro that was effective for killing tumor cells (NALM-6, Primary cells) without impacting CAR-T activity (Fig. 3C, Supplementary 1A). We used a concentration of 350 nM Rituximab to treat tumor cells, followed by flow cytometry to detect the expression of CD19 and CD22. The results showed that the expression of CD19 and CD22 had not been obviously altered (Supplementary 1C). We also selected an efficacy target ratio based on the killing of NALM-6 by Rituximab combined with CAR-T cells, and eventually, we chose a 1:1 efficacy target ratio for the subsequent experiments (Fig. 3D).

In proliferation experiments, we found CD19-CD22 CAR-T cells rapidly proliferated upon Nalm-6 challenge, reaching a peak level of expansion on day 5 (Fig. 3E). Rituximab presensitized tumor cells facilitated the proliferation of CAR-T cells.

Several mechanisms could be working through the therapeutic efficacies of anti-CD20 antibodies, including complement-dependent cytotoxicity (CDC), antibody-dependent cellular cytotoxicity (ADCC) and the induction of apoptosis [8]. Previous studies have indicated that direct effects mediated through binding of CD20 to the cell surface include inhibition of proliferation, induction of apoptosis, and sensitization of cancer cells to chemotherapy [9]. We divided mitomycin-treated NALM-6 and primary cells into two groups based on whether they were pretreated with Rituximab or not, and then incubated them with CD19/CD22 CAR-T cells. The other group was treated with Rituximab alone for tumor cells. FCM analysis showed that Rituximab combined with CAR-T significantly promoted tumor cell apoptosis compared to the control groups (p < 0.001). Terminal deoxynucleotidyl transferase dUTP nick end labeling (TUNEL) assays demonstrated that the number of apoptotic cells in the Rituximab combined with CAR-T group was significantly higher than that in the control groups (Fig. 3F, Supplementary 1D).

We utilized a sequential killing assay in which CAR-T cells were plated with Nalm-6 cells at a 1:1 effector-to-target ratio and replated every 3 days with fresh tumor cells to restore the initial ratio. Unlike the control groups, the combination of Rituximab with CAR-T repeatedly eliminated tumor cells for at least three rounds (Supplementary 2A). Collectively, the Rituximab combined with CAR-T demonstrated robust and sustained cytotoxicity against ALL cell lines in vitro.

To examine the effector function of rituximab combined with CAR-T, a panel of cytokines were measured during the in vitro cytotoxicity assay. Compared with the control group, a broad range of cytokines was produced by Rituximab combined with CD19-CD22 CAR-T when co-cultured with NALM-6. Increased secretion of effector cytokines and chemokines such as interferon-γ (IFN-γ), tumor necrosis factor-α (TNF-α), and interleukin-2 (IL-2) was observed (Fig. 4A).

Fig. 4

CD19/CD22 CAR-T cells undergo activation, antigen-induced differentiation. A CD19/CD22 CAR-T cells produced higher levels of cytokines. CAR-T cells and T cells were incubated with tumor cells for 12 h; the levels of cytokines in the culture supernatants were determined by ELISA assay. B–D Left panel: CD19/CD22 CAR-T were incubated with NALM-6 for 48 h, CD25, CD8+CD69, CD4+CD62L were analyzed by flow cytometry. Quantitative analysis of the frequency of CD25, CD8+CD69, CD4+CD62L; Right panel: Histogram plots of CD25, CD8+CD69, and CD4+CD62L expression on CD19/CD22 CAR T cells (n = 3 samples examined over three independent experiments)

3.6 Pre-treatment of tumor cells with Rituximab affects the surface molecule expression of CD19-CD22 CAR-T

To investigate the effects of Rituximab on the expression of surface marker molecules on CD19/CD22 CAR-T cells, we co-cultured Rituximab-treated tumor cells with CAR-T cells for 24 h. Subsequently, we assessed the expression of activation molecules and memory molecules on CAR-T cells using flow cytometry. The analysis revealed that CD25+, CD8+CD69+, and CD4+CD62L+ molecules on the surface of CD19/CD22 CAR-T cells were significantly upregulated in the Rituximab pre-treatment group compared to the group without Rituximab treatment (p < 0.001) (Fig. 4B–D).

To investigate the influence of Rituximab on the persistence of CD19/CD22 CAR-T cells, the expression of inhibitory molecules PD-1, LAG-3 and Tim-3 on CD19/CD22 CAR-T cells were monitored. Flow cytometry (FCM) analysis at 24 h showed that the expression of CD19/CD22 CAR-T cell surface exhaustion molecules (LAG-3, PD-1, TIM-3) was significantly lower in the Rituximab group than in the Control group (p < 0.001). Interestingly, compared to the control group, FCM analysis at 96 h showed that pretreatment of tumor cells with rituximab did not significantly upregulate CD19/CD22 CAR-T surface exhaustion molecules (PD-1, TIM-3) (Fig. 5A–C).

Fig. 5

Rituximab reduced the expression of exhaustion molecules (PD-1, LAG-3, Tim-3) on CD19/CD22 CAR-T. A–C Left panel: CD19/CD22 CAR-T were incubated with NALM-6 for 24–96 h, PD-1, LAG-3, and Tim-3 were analyzed by flow cytometry. Quantitative analysis of the frequency of PD-1, LAG-3, and Tim-3; Right panel: Histogram plots of PD-1, LAG-3, and Tim-3 expression on CD19/CD22 CAR T cells (n = 3 samples examined over three independent experiments)

To establish whether Rituximab combined with CAR-T was resistant to functional exhaustion, we used a sequential killing assay in which CAR-T cells were plated with Nalm-6 cells at a 1:1 effector-to-target ratio and replated every 3 days with fresh tumor cells to restore the initial ratio. After three rounds of repeated challenges, Rituximab combined with CAR-T effectively reduced CAR-T exhaustion (Supplementary 2B–D).