Cytokine release syndrome was an independent risk factor associated with hypoalbuminemia for patients with relapsed/refractory hematological malignancies after CAR-T cell therapy

From May 2021 to December 2021, 71 patients after CAR-T cell therapy were enrolled in this clinical trial (ChiCTR2100046474) conducted at the Bone Marrow Transplantation Center in The First Affiliated Hospital of Zhejiang University School of Medicine. All admitted patients were evaluated at the time of admission, referring to NRS 2002 Nutrition Risk Screening Form. This clinical trial was designed to explore nutritional status during CAR-T treatment and evaluate the risk factors of post-CAR-T hypoproteinemia and its correlation with CRS. The inclusion criteria were as follows: 1) Age ≥ 13 and < 70 years; 2) Expected survival time is more than 3 months; 3) Patients provided informed consent. The exclusion criteria were as follows: 1) Women who are pregnant or breastfeeding; 2) Patients had other uncontrolled medical conditions that the investigator deemed unsuitable for enrollment: 3) Patients with incomplete data.

All participants provided written informed consent. The patients were divided into the CRS group (n = 60) and the non-CRS group (n = 11). The treatment protocol included a lymphodepletion regimen consisting of 2 days of cyclophosphamide 0.5 g/m² and 3 days of fludarabine 30 mg/m²/d, as well as CAR-T cell infusion. Patients with CRS received supportive treatment with tocilizumab (humanized monoclonal antibody against the IL-6 receptor) or corticosteroids. Ethical approval was obtained from the Ethics Committee of the First Affiliated Hospital of Zhejiang University School of Medicine. The study was registered at the Chinese Clinical Trial Registry (ChiCTR2100046474).

Data on blood nutrition, clinical baseline, CRS-related indicators, and severity were collected for all patients treated with CAR-T cells at our center. The primary outcomes were nutrition-related indicators and the incidence of hypoalbuminemia. In order to observe the dynamic changes in the nutritional status of patients during the CAR-T cell therapy, nutrition-related indicators at different time points were recorded, including serum albumin, total protein, triglyceride, total cholesterol, body mass index (BMI), grip strength, upper arm length, mid-calf circumference (MAC), and triceps skinfold thickness (TSF). CRS peak was defined as the day with the highest or most frequent fever. The prerequisite for CRS diagnosis is fever (temperature ≥ 38 °C which is not attributable to any other cause, within 24 h to 3 weeks after CAR-T cell administration). The diagnosis should also be combined with Hypotension, Hypoxia, and other systemic symptoms, according to the American Society for Transplantation and Cellular Therapy (ASTCT) Consensus Grading for Cytokine Release Syndrome (see supplementary material for detail) [12]. The severity of CRS could be classified into no CRS (grade 0), mild CRS (grade 1 or grade 2 CRS), and severe CRS (grade 3 or grade 4) subcategories. Serum albumin < 35 g/L is identified as hypoalbuminemia.

A total of 71 patients after CAR-T cell therapy (41males and 30 females) were enrolled, with a median age of 55 years (range, 17 to 69 years) and a median BMI of 22.08 kg/m² (range, 2.73 to 32.15). Among the enrolled patients, 35 (49.3%) were diagnosed with acute leukemia, 17(23.9%) patients with lymphoma, and 19(26.8%) with multiple myeloma. The median chemotherapy cycle of the cases was 7. Nine patients undergone auto-HSCT and 3 patients receive allo-HSCT. During the study period, 60 patients developed CRS and 11 patients did not. Detailed baseline characteristics are shown in Table 1.

The nutritional status of all patients was assessed when they were admitted to the hospital, during CAR-T cell infusion treatment, and discharged. For patients with CRS, nutrition-related indicators at CRS onset, peak, and recovery stages were recorded. For patients without CRS, these indicators were monitored at 1 week and 2 weeks after CAR-T cell infusion (Fig. 1A). Serum albumin and total protein showed a significant downward trend at the beginning of the CRS and reached their lowest value at CRS peak, which was not obvious in the non-CRS group (Fig. 1B, C). Triglyceride level, total cholesterol level, upper arm length, calf circumference, and Triceps skinfold thickness decreased obviously after CAR-T cell infusion with or without CRS, while BMI and grip strength remained stable during treatment (Fig. 1D, E, F, G, H, I, J).

To determine risk predictors for post-CAR-T hypoalbuminemia, multivariate logistic regression was applied to the three significant variables identified by univariate analysis (P < 0.1). The three significant variables are CRS (P = 0.014), baseline ALB (P = 0.022) and baseline LDH (P = 0.082). Finally, CRS (odds ratio (OR) = 13.618, 95% confidence interval (CI) 1.499–123.709; P = 0.013) and baseline ALB (OR = 0.854, 95% CI 0.754–0.967; P = 0.020) were identified as the independent factors associated with post-CAR-T hypoalbuminemia.

The incidence of hypoalbuminemia events varied based on the level of albumin measured at different stages of hospitalization (Fig. 2A). The incidence of hypoalbuminemia was lowest at admission (12.68%) and was highest at the nadir of serum albumin (45.07%). The prevalence of hypoalbuminemia was 23.94% (42/117) at discharge. The severity of CRS could be divided into non-CRS (grade 0), mild CRS (grade 1 or grade 2), and severe CRS (grade 3 or grade 4) subcategories. Patients with severe CRS tend to have a higher probability of occurrence of hypoalbuminemia. According to the nadir of serum albumin, hypoalbuminemia occurred most frequently in patients with severe CRS (78.57%), the incidence of which was slightly lower at 56.52% in the mild CRS group (Fig. 2B).

Spearman correlation analysis showed that nadir of serum albumin (r =  − 0.587, P < 0.001) and serum albumin at discharge (r =  − 0.315, P = 0.01) were negatively correlated duration of CRS (Fig. 2C, D). Patients with hypoalbuminemia had significantly longer hospital stays than patients without (P = 0.04) (Fig. 3A). And the nadir of serum album was also negatively associated with prolonged hospital stay (r =  − 0.532, P < 0.001) (Fig. 3B).