Granulocyte colony-stimulating factor in neutropenia management after CAR-T cell therapy: A safety and efficacy evaluation in refractory/relapsed B-cell acute lymphoblastic leukemia

To the Editor: After chimeric antigen receptor T (CAR-T) cell therapy, over 90% of patients experience either mild or severe cytopenia, with the prevalence of Grades 3–4 neutropenia ranging from 53% to 94.3%.[1,2] Granulocyte colony-stimulating factor (G-CSF), known for its ability to enhance the proliferation and differentiation of neutrophils, has been extensively employed for the prevention and treatment of neutropenia in various contexts. However, its application in patients with refractory/relapsing B-cell acute lymphoblastic leukemia (R/R B-ALL) after CAR-T cell therapy remains a topic of debate.[3] Several studies have suggested that G-CSF may exert anti-inflammatory effects by modulating the release of inflammatory factors, such as interleukin-12 (IL-12), TNF-α, and IFN-γ, or by influencing the differentiation of T-cell subsets.[4] However, other researchers have posited that G-CSF could exacerbate cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS) through the augmentation of antigen-presenting cell function. Moreover, the severity of ICANS has been found to be strongly associated with elevated levels of serum G-CSF in B-ALL.[5] To provide a basis of knowledge for this topic, further research is warranted, which should include additional clinical and experimental studies of G-CSF and evaluation of the impact of G-CSF on CRS and ICANS in the content of B-ALL. This comprehensive information will enable clinicians to make more informed decisions when formulating treatment strategies for patients. We, therefore, conducted a retrospective analysis of 168 R/R B-ALL patients who developed neutropenia within one month after receiving CD19 CAR-T cell therapy at Tianjin First Central Hospital from June 2017 to December 2022. Informed consent for participation in this clinical trial (registered at www.chictr.org.cn as ChiCTR-ONN-16009862 and ChiCTR1800015164) was obtained from all patients and their families at Tianjin First Central Hospital. This study was approved by the Ethics Committee of Tianjin First Central Hospital (No. 2015002X), and the relevant clinical trials fully adhered to the principles outlined in the Declaration of Helsinki. Among these patients, 89 patients received G-CSF at a dose of 2.5–5.0 mg·kg−1·day−1 during neutropenia (G-CSF group) and 79 patients did not receive G-CSF (control group). Overall survival (OS) was defined as the duration from CAR-T cell infusion to death from any cause. Progression-free survival (PFS) was defined as the period from CAR-T cell infusion to disease progression or death. Statistical analyses were conducted using SPSS Statistics 26.0 (IBM Corp, Armonk, NY, USA) and GraphPad Prism 9.0 (GraphPad Software, CA, USA). First, we evaluated the efficacy of G-CSF application by analyzing the differences in the incidence and duration of severe neutropenia, as well as the occurrence and type of infections, between the G-CSF and control groups one month after CAR-T cell therapy. Neutropenia and infections were defined and graded according to the Common Terminology Criteria for Adverse Events 5.0 (CTCAE 5.0) and the CSCO-CAR-T Cell Therapy for Malignant Hematological Diseases Guidelines (2024). For patients in the G-CSF group, subcutaneous injection of filgrastim (2.5–5.0 μg·kg−1·day−1) was promptly administered if the absolute neutrophil count (ANC) fell below 2.0 × 109/L at any time within one month after CAR-T cell infusion. The administration was discontinued when the ANC value exceeded and remained over 0.5 × 109/L for two consecutive days. Subsequently, we compared the toxicity and antitumor activity following CAR-T cell therapy between the two cohorts. CRS and ICANS were graded based on the consensus of the American Society of Transplantation and Cell Therapy. The efficacy of CAR-T cells was assessed through bone marrow cell morphology and flow cytometry analysis conducted 28 days after infusion. Additionally, we employed flow cytometry to analyze patients’ peripheral blood samples to monitor the CAR-T expansion and persistence. Specimens were collected at predetermined time points after infusion to determine the relative Cmax (maximum concentration) of CAR-T/CD3+ T cells (%), and patient survival status, including PFS and OS, was evaluated up to December 31, 2023. The frequency distributions and percentages were used to compare qualitative variables. For continuous variables with a normal distribution, a t-test was used; for non-normally distributed continuous variables, a rank-sum test was used; and for categorical variables, chi-squared test was used. Kaplan–Meier survival curves were compared using the log-rank test. P <0.05 was considered statistically significant. The basic characteristics were similar between the two groups [Table 1], with a median white blood cell count of 4.59 × 109/L and a median neutrophil count of 2.31 × 109/L before the lymphocyte clearance regimen, and there was no statistically significant difference between the two groups [Supplementary Table 1, https://links.lww.com/CM9/C202]. There was also no statistically significant difference between the groups in the source of the CAR-T cells, isolated from either donors or the patients themselves (P = 0.151). Additionally, there was no marked variation in the median infused dose of CAR-T cells (2.40 × 106 cells/kg vs. 2.90 × 106 cells/kg, P = 0.073). However, after CAR-T cell infusion, the lowest recorded ANC was significantly greater in the G-CSF group than in the control group (0.35 × 109/L vs. 0.14 × 109/L, P = 0.015). Although the incidence of Grades 3–4 neutropenia was similar between the two groups (93.3% vs. 93.7%, P = 0.473), the duration of Grades 3–4 neutropenia was shorter in the G-CSF group (5 [range: 2–7] days vs. 8 [range: 3–12] days, P = 0.001) [Supplementary Figure 1A, https://links.lww.com/CM9/C202]. A total of 115 patients (68.5%) developed infections, which were mainly bacterial infections (97 patients), with a small number of fungal (13 patients) and viral (5 patients) infections. There were no significant differences in the overall risk (69.6% vs. 67.1%, P = 0.721) or severity of infection between the two groups [Supplementary Table 1, https://links.lww.com/CM9/C202]. Table 1 - Baseline characteristics of R/R B-ALL patients before CAR-T cell therapy. Baseline characteristics G-CSF* (N = 89) Control† (N = 79) All patients (N = 168) Statistic values P value Age (years), median (range) 37 (22–51) 28 (19–43) 31 (20–47) 2.238 0.052 Sex, n (%) 1.026 0.305 Male 51 (57.3) 39 (49.4) 90 (53.6) Female 38 (42.7) 40 (50.6) 78 (46.4) Gene mutations or fusions, n (%) 1.753 0.658 Ph-positive 16 (18) 20 (25.3) 36 (21.4) TP53 mutation 12 (13.5) 9 (11.4) 21 (12.5) MLL rearrangement 8 (9) 9 (11.4) 17 (10.1) Karyotype, n (%) 0.079 0.937 Complex karyotype 17 (19.1) 14 (17.7) 31 (18.5) Abnormal karyotype 25 (28.1) 23 (29.1) 48 (28.6) Normal karyotype 37 (41.6) 33 (41.8) 70 (41.7) Others 10 (11.2) 9 (11.4) 19 (11.3) Extramedullary disease, n (%) 1.262 0.207 Central nervous system leukemia 16 (18.0) 11 (13.9) 27 (16.1) Other extramedullary manifestations 8 (9.0) 4 (5.1) 12 (7.1) Blasts in BM before lymphodepletion treatment, n (%) 1.675 0.094 <5 30 (33.7) 38 (48.1) 67 (40.5) ≥5 and <20 11 (12.4) 7 (8.9) 16 (8.9) ≥20 and <50 19 (21.3) 14 (17.7) 32 (18.9) ≥50 29 (32.6) 20 (25.3) 47 (28.1) Number of prior therapy lines, n (%) 0.788 0.431 <5 38 (42.7) 37 (46.8) 75 (44.6) ≥5 and <10 34 (38.2) 28 (35.4) 62 (36.9) ≥10 17 (19.1) 14 (17.7) 31 (18.5) Prior-HSCT, n (%) 18 (20.2) 19 (24.1) 37 (22.0) 0.130 0.897 *Patients who received G-CSF after CAR-T cell infusion. †Patients who did not receive G-CSF after CAR-T cell infusion. ANC: Absolute neutrophil count; BM: Bone marrow; CAR-T: Chimeric antigen receptor T; G-CSF: Granulocyte colony-stimulating factor; HSCT: Hematopoietic stem cell transplantation; R/R B-ALL: Refractory/relapsing B-cell acute lymphoblastic leukemia; WBC: White blood cell. After CAR-T cell therapy, 70.8% (119/168) of patients developed CRS, with the majority having Grades 1–2. Patients who developed CRS or ICANS received treatments tailored to the severity of their condition, including tocilizumab, corticosteroids, and plasmapheresis. The incidence of CRS was slightly greater in the G-CSF group (77.5% vs. 63.3%, P = 0.043) [Supplementary Table 2, https://links.lww.com/CM9/C202]. However, there were no statistically significant differences in the severity (P = 0.365) [Supplementary Figure 1B, https://links.lww.com/CM9/C202] or duration of CRS (5 [range: 1–8] days vs. 4 [range: 0–8] days, P = 0.319) between the two groups. Interestingly, the median IL-6 max value in the G-CSF group (49.80 pg/mL) was lower than that in the control group (69.45 pg/mL), but no statistically significant difference was observed (P = 0.306). Only one patient in the G-CSF group developed Grade 2 ICANS, and no deaths occurred due to Grades 3–4 ICANS. The expansion of CAR-T cells was monitored using flow cytometry after CAR-T cell infusion, and statistical analysis was performed based on the relative Cmax. Although the median relative Cmax in the G-CSF group (2.6%, range: 0.8–10.8%) was comparatively lower than that in the control group (3.3%, range: 0.9–16.4%), no statistically significant difference was observed (P = 0.544). Furthermore, there were no significant differences in the detectable duration of CAR-T cells in the peripheral blood samples. In terms of antitumor activity, 68.5% (61/89) of patients in the G-CSF group achieved minimal residual disease (MRD)–CR, 4.5% (4/89) MRD+CR, 6.7% (6/89) PR, and 20.2% (21/89) NR one month after infusion. There was no significant difference in the overall response rate between the G-CSF and control groups (79.7% vs. 80.5%, P = 0.632) [Supplementary Table 3, https://links.lww.com/CM9/C202]. Further analysis revealed that the median PFS for patients in the G-CSF group was 541 days, and for patients in the control group was 587 days (P = 0.503, Supplementary Figure 2A, https://links.lww.com/CM9/C202). In addition, the median OS was 881 days in the G-CSF group and 757 days in the control group (P = 0.825, Supplementary Figure 2B, https://links.lww.com/CM9/C202), but the difference was not statistically significant. These findings indicated that the administration of G-CSF does not affect the prognosis of patients treated with CAR-T cell therapy. This is a relatively large real-world observation on the application of G-CSF in R/R B-ALL patients receiving CAR-T cell therapy. We observed that while G-CSF did not significantly reduce the incidence of Grades 3–4 neutropenia or the occurrence of infection, it was effective in shortening the duration of Grades 3–4 neutropenia following CAR-T cell therapy. In terms of safety, although G-CSF was associated with an increased incidence of CRS, there was no marked difference in CRS severity between the two groups. In addition, only one patient in the G-CSF group developed lower-grade ICANS. The efficacy, PFS, and OS after CAR-T cell therapy were not significantly different between the two groups. Certain limitations of this study should be considered to accurately contextualize the findings. Primarily, our study is limited by its single-center, retrospective design. This framework potentially limits the generalizability of the results. Furthermore, the administration and maintenance time of G-CSF varied among different clinicians, and this heterogeneity may introduce confounding variables. It should also be noted that the management protocols for CRS and ICANS have changed since 2018. These factors collectively necessitate a cautious interpretation of our conclusions regarding the influence of G-CSF on adverse reactions following CAR-T cell therapy. In summary, our findings suggest that the use of G-CSF for managing neutropenia after CAR-T cell therapy may be a clinically viable option. However, given the limitations of our study, future studies need to be multicenter, prospective, and designed to obtain more comprehensive results. Funding This work was supported by grants from the General Project of the National Natural Science Foundation of China (No. 81970180), the Science and Technology Project of Tianjin Municipal Health Committee (No. TJWJ2022QN030), the Key projects of Tianjin Applied Basic Research and Multi-Investment Fund (No. 21JCZDJC01240), the Science and Technology Project of Tianjin Municipal Health Committee (No. TJWJ2022XK018), Tianjin Key Medical Discipline (Specialty) Construction Project (No. TJYXZDXK-056B), Tianjin Municipal Natural Science Foundation (No. 22JCQNJC00820), Tianjin Health Research Project (No. TJWJ2023QN027), Tianjin Health Bureau Project (No. ZC20074), Tianjin Key Medical Discipline(Specialty) Construction Project (No. TJWJ2023XK010), and Henan Provincial Youth and Middle-aged Health Science and Technology Talents Excellent Youth Program (No. 20230273). Conflicts of interest None.