Abstract
Background
Chimeric antigen receptor T-cell (CAR-T) therapy has emerged as a promising immunotherapy for various malignancies. However, its use is associated with challenges, including cytokine release syndrome (CRS), a potentially severe complication. This retrospective study aims to analyze the risks, outcomes, and healthcare burden of CRS in patients undergoing CAR-T therapy.
Method
Data from the 2020 National Inpatient Sample (NIS) were utilized, comprising 415 CAR-T-related hospitalizations. They were categorized into those with CRS (n = 68) and those without CRS (n = 347). Baseline characteristics, including age, gender, race, income, insurance status, and comorbidities, were compared. Outcomes of interest included in-hospital mortality, length of stay (LOS), total hospital charges, and access to complications, associations, and interventions. Statistical analyses, including multivariable models, were employed to assess associations.
Results
Hospitalizations with CRS did not exhibit significant differences in age, gender, race, income, or insurance status compared to those without CRS. The multivariable analysis showed no statistically significant difference in mortality (adjusted odds ratio (aOR) = 2.48, 95% confidence interval (CI): 0.71 to 8.69, p = 0.151), LOS (coefficient = -2.1 days, 95% CI: -5.43 to 1.21, p = 0.207), or total hospital charges (coefficient = $207,456, 95% CI: $6119 to $421,031, p = 0.057) between the two groups. The CRS group had a higher incidence of fever (aOR = 1.91, 95% CI: 1.15 to 3.17, p = 0.014), acute respiratory failure (aOR = 2.10, 95% CI: 1.01 to 4.40, p= 0.049), and the need for intubation/mechanical ventilation (aOR = 2.59, 95% CI: 1.14 to 5.88, p = 0.024). Hemophagocytic lymphohistiocytosis (HLH) was significantly associated with CRS (aOR = 6.72, 95% CI: 2.03 to 22.18, p = 0.002).
Conclusion
While the development of CRS in CAR-T-treated patients did not significantly increase mortality, LOS, or total hospital charges, it was associated with specific risks and outcomes, including fever, respiratory failure, and HLH. This study emphasizes the importance of vigilance in recognizing and managing CRS in CAR-T therapy to optimize patient outcomes. The findings contribute valuable insights to guide clinical decision-making in the context of CAR-T therapy.
Keywords: healthcare burden, outcomes, crs, cytokine release syndrome, car t, chimeric antigen receptor t-cell
Introduction
Chimeric antigen receptor T-cell (CAR-T) therapy is a form of immunotherapy in which a patient's T cells are collected and genetically modified to improve their targeting ability and facilitate the elimination of cancer cells [1]. CAR-T cells have been effective in treating various types of B-cell lymphomas, leukemias, and multiple myeloma, resulting in long-lasting remissions and often eliminating the cancer cells completely [2-5].
Even though CAR-T therapies are promising, they are not without their challenges. Most patients relapse and experience significant treatment-related toxicity, including cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS), which have substantial health and economic implications [6].
CRS is a systemic inflammatory response that can be triggered by a variety of factors such as infections and certain drugs [7]. The primary manifestation of CRS is a fever, alongside a range of symptoms, including fatigue, muscle aches, gastrointestinal problems (nausea, vomiting, diarrhea), tachycardia, and skin rashes [8]. CRS can also result in life-threatening complications, including cardiac dysfunction, adult respiratory distress syndrome, neurotoxicity, renal failure, disseminated intravascular coagulation, multiorgan failure, and circulatory collapse [9]. Elevated levels of various cytokines, especially interleukin-6 (IL-6) and interleukin-2 (IL-2), are found in the serum of patients experiencing toxicities due to CRS, such as after CAR-T-cell infusions. The immune system is continuously interacting with cytokines released by CAR-T cells or macrophages, which creates a complex interplay [10]. CRS usually develops after the first week of CAR-T cell infusion, and with proper management, symptoms often subside within one to two weeks [11]. It is possible to predict CRS severity based on clinical factors such as disease burden, marrow involvement, lymphodepletion, and high CAR-T doses, as well as patient-specific factors such as pre-existing inflammation (baseline serum ferritin) and baseline endothelial activation (thrombocytopenia) [12]. Due to the potential occurrence of serious adverse events associated with CAR-T therapy, the FDA has implemented a Risk Evaluation and Mitigation Strategy (REMS). This mandates only skilled personnel to be able to administer this therapy and necessitates continuous availability of drugs like tocilizumab in case of adverse events. Tocilizumab, an IL-6 receptor antagonist, manages severe CRS from CAR-T cells without harming T cells, while blinatumomab, a bispecific antibody that redirects effector T cells to B cells with its anti-CD3 and anti-CD19 arms, aims to prevent CRS through strategies such as disease reduction, corticosteroids, and dose adjustment [13].
Despite the available treatment strategies for CRS, CRS associated with CAR-T therapy still carries the risk of morbidity and mortality. Therefore, we conducted a retrospective study to analyze the risks, outcomes, and healthcare burden of CRS in patients who underwent CAR-T therapy.
Materials and methods
This retrospective study utilized data extracted from the National Inpatient Sample (NIS), a database supported by the Agency for Healthcare Research and Quality (AHRQ) and part of the Health Care Cost and Utilization Project (HCUP) [14]. The NIS database represents a roughly 20% stratified sample of discharges from almost 1000 US hospitals across all 50 states. It is important to note that the NIS database is the largest publicly available database in the United States for inpatient care, covering all types of payers. For this analysis, we worked with the 2020 NIS database, which includes hospitalizations occurring from January 1, 2020, to December 31, 2020. Additionally, because the NIS data is de-identified, this study was exempt from the need for Institutional Review Board (IRB) approval.
Study population
We employed the International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) codes to identify hospitalizations related to CAR-T therapy. Subsequently, we divided these hospitalizations into two distinct groups: one with CRS and the other without CRS. Specifically, for CAR-T hospitalizations, we used the ICD-10 codes XW033C3 and XW043C3, while for CRS cases, we utilized the ICD-10 codes D89.83, D89.831, D89.832, D89.833, D89.834, D89.835, and D89.839. We excluded categories such as "No charge," "Other," and "Missing value" for the primary payer (insurance status). Our inclusion criteria were restricted to adults aged 18 years and older.
Outcomes of interest
Our primary areas of interest included in-hospital mortality, length of stay, and the overall costs during the hospital stay. Our secondary goals involved exploring complications and associations related to CRS. To assess mortality, we used the NIS variable "DIED," and the length of hospital stays was determined through the NIS variable "LOS." To calculate the total charges associated with hospitalization, we used the variable "TOTCHG." Additionally, to access complications, associations, and interventions, we used the pertinent ICD-10 codes as provided in the Supplemental Table.
Statistical analysis
All analyses were conducted in accordance with the Healthcare Cost and Utilization Project regulations, which involved appropriate stratification, clustering, and weighting of samples [15]. We calculated odds ratios for binary variables and coefficients for continuous variables. Initially, we performed univariate analyses to determine unadjusted odds ratios. Subsequently, in the multivariable analysis, we included only those variables that showed a significant association with the outcome of interest in the univariate analysis, with a significance level of P < 0.05. To construct a multivariate analysis model, we incorporated potential confounding variables, such as age, gender, race, income quartile based on zip code, hospital division, hospital bed size, insurance status, and the Charlson Comorbidity Index score. The Charlson Comorbidity Index score encompasses conditions like myocardial infarction, congestive heart failure, peripheral arterial disease, cerebrovascular disease, dementia, chronic obstructive pulmonary disease, connective tissue disease, peptic ulcer disease, liver disease, diabetes, hemiplegia or paraplegia, chronic kidney disease, diabetes with end-organ damage, solid tumors, leukemia, lymphoma, and AIDS/HIV, all of which are conditions associated with high mortality rates [16].
Categorical variables were compared using the Fisher exact test, and continuous variables were compared using the Student's t-test. All P-values were two-sided, and the threshold for statistical significance was set at P < 0.05. The statistical analysis was carried out using STATA version 17, developed by StataCorp LLC in College Station, TX.
Results
We identified 415 inpatient encounters linked to CAR-T therapy. Among them, 68 had CRS, while 347 did not. Table 1 presents the baseline characteristics of CAR-T hospitalizations, categorized into those with CRS and those without CRS, and provides details on characteristics such as median age, sex, race, median household income, insurance status, Charlson Comorbidity Index score, admission type, census division, and hospital bed-size, along with corresponding p-values.
Table 1. Baseline characteristics of CAR-T therapy hospitalizations.
CRS: cytokine release syndrome; CAR-T: chimeric antigen receptor T-cell.
| With CRS | Without CRS | Total CAR-T therapy hospitalizations | P-value | |
| Number of hospitalizations | 68 | 347 | 415 | |
| Mean age | 59.1 years | 60.3 years | 60.1 years | P = 0.517 |
| Sex | P = 0.506 | |||
| Male | 43 (63.24%) | 208 (59.65%) | 251 (60.24%) | |
| Female | 25 (36.76%) | 139 (40.35%) | 164 (39.76%) | |
| Race | P = 0.795 | |||
| White | 53 (78.46%) | 268 (77.22%) | 321 (77.42%) | |
| Black | 6 (9.23%) | 26 (7.4%) | 32 (7.69%) | |
| Hispanic | 5 (6.15%) | 33 (9.47%) | 38 (8.93%) | |
| Asian or Pacific Islander | 3 (4.62%) | 10 (2.96%) | 13 (3.23%) | |
| Native American | 0 (0%) | 1 (0.3%) | 1 (0.25%) | |
| Other | 1 (1.54%) | 9 (2.66%) | 10 (2.48%) | |
| Median household income | P = 0.766 | |||
| 0-25th percentile | 10 (14.93%) | 58 (16.82%) | 68 (16.5%) | |
| 26th to 50th percentile | 15 (22.39%) | 68 (19.52%) | 83 (20.0%) | |
| 51st to 75th percentile | 15 (22.39% | 94 (27.03%) | 109 (26.25% | |
| 76th to 100th percentile | 28 (40.30%) | 127 (36.64%) | 155 (37.25%) | |
| Insurance status | P = 0.533 | |||
| Medicare | 24 (34.85%) | 153 (44.01%) | 177 (42.50%) | |
| Medicaid | 5 (7.58%) | 25 (7.19%) | 30 (7.25%) | |
| Private insurance | 38 (56.06%) | 164 (47.31%) | 202 (48.75%) | |
| No insurance | 1 (1.52%) | 5 (1.50%) | 6 (1.50%) | |
| Charlson Comorbidity Index score | P = 0.038 | |||
| 0 | 1 (1.47%) | 0 (0%) | 1 (0.24%) | |
| 1 | 0 (0%) | 1 (0.29%) | 1 (0.24%) | |
| 2 | 44 (64.71%) | 191 (55.04%) | 235 (56.63%) | |
| 3 or more | 23 (33.82%) | 155 (44.67%) | 178 (42.89%) | |
| Admission type | P = 0.445 | |||
| Non-elective | 21 (30.88%) | 92 (26.51%) | 113 (27.23%) | |
| Elective | 47 (69.12%) | 255 (73.49%) | 302 (72.77%) | |
| Census division | P = 0.215 | |||
| New England | 6 (8.82%) | 35 (10.09%) | 41 (9.88%) | |
| Middle Atlantic | 8 (11.76%) | 64 (18.44%) | 72 (17.35%) | |
| East North Central | 12 (17.65%) | 57 (16.43%) | 69 (16.63%) | |
| West North Central | 7 (10.29%) | 29 (8.36%) | 36 (8.67%) | |
| South Atlantic | 11 (16.18) | 54 (15.56%) | 65 (15.66%) | |
| East South Central | 2 (2.94%) | 6 (1.73%) | 8 (1.93%) | |
| West South Central | 2 (2.94%) | 32 (9.22%) | 34 (8.19%) | |
| Mountain | 2 (2.94%) | 9 (2.59%) | 11 (2.65%) | |
| Pacific | 18 (26.47%) | 61 (17.58%) | 79 (19.04%) | |
| Hospital bed-size | P = 0.864 | |||
| Small | 10 (13.24%) | 49 (14.12%) | 59 (13.98%) | |
| Medium | 10 (14.71%) | 44 (12.68%) | 54 (13.01%) | |
| Large | 48 (72.06%) | 254 (73.20%) | 302 (73.01%) |
Table 2 compares mortality, length of stay, and total hospital charges in CAR-T hospitalizations with and without CRS. The CRS group had a mortality rate of 5.89% (4/68) compared to 2.9% (10/347) in the non-CRS group. The adjusted odds ratio (aOR) for mortality was 2.48 (95% CI: 0.71 to 8.69, p = 0.151). The mean length of stay for CRS was 16.9 days, and for non-CRS, it was 18.9 days, with a coefficient of -2.1 days (95% CI: -5.43 to 1.21, p = 0.207). Mean total hospital charges were $1,148,539 for CRS and $967,146 for non-CRS, with a coefficient of $207,456 (95% CI: $6119 to $421,031, p = 0.057).
Table 2. Mortality, length of stay, and total hospital charges.
CRS: cytokine release syndrome.
| With CRS | Without CRS | Adjusted odds ratio (aOR)/coefficient (multivariable analysis) | P-value | |
| Mortality | 5.89% (4/68) | 2.9% (10/347) | aOR= 2.48 (95 CI: 0.71 to 8.69) | P = 0.151 |
| Mean length of stay | 16.9 days | 18.9 days | Coefficient = -2.1 days (95 CI: -5.43 days to 1.21 days) | P = 0.207 |
| Mean hospital charges | $1,148,539 | $967,146 | Coefficient = $207,456 (95 CI: $6119 to $421,031) | P = 0.057 |
Table 3 displays the risks and outcomes associated with CRS, comparing occurrences with and without CRS. Fever was observed in 66.2% of the group with CRS and 51.9% of the group without CRS with an aOR of 1.91 (95% CI: 1.15 to 3.17, p = 0.014). Acute respiratory failure was observed in 10% of the group with CRS and 5% of the group without CRS with an aOR of 2.10 (95% CI: 1.01 to 4.40, p = 0.049). Intubation/mechanical ventilation was observed in 11.7% of the group with CRS and 5% of the group without CRS with an aOR of 2.59 (95% CI: 1.14 to 5.88, p = 0.024).
Table 3. Risks and outcomes associated with CRS.
Statistically significant p-values are in bold.
CRS: cytokine release syndrome; AKI: acute kidney injury; CVA: cerebrovascular accident; MI: myocardial infarction; CHF: congestive heart failure.
| Risks and outcomes | With CRS | Without CRS | Adjusted odds ratio (aOR) (multivariable analysis) | P-value |
| Fever | 66.2% (45/68) | 51.9% (180/347) | aOR = 1.91 (95 CI:1.15 to 3.17) | P = 0.014 |
| Sepsis | 8.8% (6/68) | 6.9% (24/347) | aOR = 1.68 (95 CI: 0.56 to 5.04) | P = 0.348 |
| Vasopressor support | 3% (2/68) | 1.4% (5/347) | aOR = 2.39 (95% CI: 0.40 to 14.35) | P = 0.334 |
| Encephalopathy | 20.6% (14/68) | 22.2% (77/347) | aOR = 1.03 (95% CI: 0.50 to 2.09) | P = 0.940 |
| Seizures | 4.4% (3/68) | 1.1% (4/347) | aOR = 3.96 (95% CI: 0.84 to 18.72) | P = 0.082 |
| Acute respiratory failure | 10% (7/68) | 5% (17/347) | aOR = 2.10 (95% CI: 1.01 to 4.40) | P = 0.049 |
| Intubation/mechanical ventilation | 11.7% (8/68) | 5% (17/347) | aOR = 2.59(95% CI: 1.14 to 5.88) | P = 0.024 |
| Clostridioides difficile infection | 4.4% (3/68) | 5.2% (18/347) | aOR = 0.94 (95% CI:0.26 to 3.40) | P = 0.921 |
| AKI | 20.6% (14/68) | 14.4% (50/347) | aOR =1.69 (95% CI: 0.84 to 3.40) | P = 0.141 |
| Acute CVA | 0% (0/68) | 1.44% (5/347) | aOR = N/A | N/A |
| Acute liver injury | 1.5% (1/68) | 0.9% (3/347) | aOR = 1.71(95% CI: 0.16 to 18.24) | P = 0.651 |
| Hepato/splenomegaly | 4.4% (3/68) | 2.6% (9/347) | aOR = 1.73(95% CI: 0.57 to 5.27) | P = 0.326 |
| Acute MI | 0% (0/68) | 1.2% (4/347) | aOR = N/A | N/A |
| CHF | 3% (2/68) | 3.5% (12/347) | aOR = 0.85 (95% CI: 0.17 to 4.13) | P = 0.833 |
Other outcomes, including sepsis, vasopressor support, encephalopathy, seizures, Clostridioides difficile infection, acute kidney injury (AKI), acute cerebrovascular accident (CVA), acute liver injury, hepato/splenomegaly, acute myocardial infarction (MI), and congestive heart failure (CHF), are presented with corresponding aORs and p-values.
Table 4 outlines hematological outcomes and interventions associated with CRS. In the CRS group, hemophagocytic lymphohistiocytosis (HLH) was observed in 7.4% compared to 1.2% in the non-CRS group, with an aOR of 6.72 (95% CI: 2.03 to 22.18, p = 0.002). The risk of anemia was lower in the group with CRS compared to the group without CRS, with an aOR of 0.57 (95% CI: 0.34 to 0.94, p = 0.029).
Table 4. Hematological outcomes and interventions associated with CRS.
Statistically significant p-values are in bold.
CRS: cytokine release syndrome.
| Outcomes & interventions | With CRS | Without CRS | Adjusted odds ratio (aOR) (multivariable analysis) | P-value |
| Anemia | 73.5% (50/68) | 82% (284/347) | aOR = 0.57 (95% CI: 0.34 to 0.94) | P = 0.029 |
| Thrombocytopenia | 19.1% (13/68) | 15% (52/347) | aOR = 1.42 (95% CI: 0.67 to 2.99) | P = 0.350 |
| Pancytopenia | 66.2% (45/68) | 62.2% (216/347) | aOR = 1.40 (95% CI: 0.73 to 2.66) | P = 0.307 |
| Hemophagocytic lymphohistiocytosis (HLH) | 7.4% (5/68) | 1.2% (4/347) | aOR = 6.72 (95% CI: 2.03 to 22.18) | P = 0.002 |
| Pulmonary embolism | 0% (0/68) | 0.8% (3/347) | aOR = N/A | N/A |
| Major bleeding | 0% (0/68) | 2.3% (8/347) | aOR = N/A | N/A |
| Intracranial hemorrhage | 0% (0/68) | 1.4% (5/347) | aOR = N/A | N/A |
| Gastrointestinal bleeding | 1.5% (1/68) | 2.3% (8/347) | aOR = 0.67 (95% CI: 0.09 to 5.34) | P = 0.704 |
| Disseminated intravascular coagulation | 4.4% (3/68) | 2.6% (9/347) | aOR = 1.73 (95% CI: 0.41 to 7.34) | P = 0.448 |
| Red blood cell transfusion | 5.8% (4/68) | 9.2% (32/347) | aOR = 0.62 (95% CI: 0.23 to 1.62) | P = 0.318 |
| Platelets transfusion | 4.4% (3/68) | 5.4% (19/347) | aOR = 0.89 (95% CI: 0.22 to 3.62) | P = 0.873 |
| Fresh frozen plasma transfusion | 1.5% (1/68) | 0.3% (1/347) | aOR = 5.1 (95% CI: 0.30 to 90.06) | P = 0.255 |
| Cryoprecipitate transfusion | 2.9% (2/68) | 2% (7/347) | aOR = 1.64 (95% CI: 0.26 to 10.34) | P = 0.590 |
Other outcomes and interventions, including thrombocytopenia, pancytopenia, pulmonary embolism, major bleeding, intracranial hemorrhage, gastrointestinal bleeding, disseminated intravascular coagulation (DIC), red blood cell transfusion, platelets transfusion, fresh frozen plasma (FFP) transfusion, and cryoprecipitate transfusion, are presented with corresponding aORs and p-values.
Discussion
CAR-T therapy remains limited by significant toxicities, including CRS and ICANS. CRS occurs in around 70% of patients after CD19 CAR-T cell therapy, with published incidence rates ranging from 35% to 93%, depending on the product infused and disease treated [17]. CRS initially manifests with fever and can progress to life-threatening capillary leak with hypoxia and hypotension, with subsequent multiple organ toxicities and hematological complications. In addition to fever, cytopenias, and hypofibrinogenemia, a profound rise in the serum ferritin, soluble CD25, and cytokines, such as interferon-gamma (IFN-γ), IL2, and IL6, in severe CRS suggests many similarities to macrophage activation syndrome/hemophagocytic lymphohistiocytosis (MAS/HLH) [18,19]. The Lee scale and ASCTC (American Society for Transplantation and Cellular Therapy) grading systems categorize CRS severity into four grades: Grade 1, characterized by mild symptoms and no need for intervention; Grade 2, requiring symptomatic treatment or infusion interruption but responding promptly; Grade 3, exhibiting prolonged symptoms, recurrence after initial improvement, or hospitalization due to clinical sequelae; and Grade 4, the most severe, necessitating pressor or ventilatory support due to life-threatening consequences [20].
The present study was conducted to analyze real-world data to find risks, predictors, and outcomes of CRS in hospitalized patients treated with CAR-T therapy using the NIS. Several literature reviews and incidence studies have been conducted in the past to better understand the pathophysiology and treatment options of CRS secondary to CAR-T therapy [21,22]. However, hardly any studies have been able to conclude how the development of CRS in hospitalized patients treated with CAR-T therapy affects mortality, length of stay (LOS), demographic associations, and outcomes like organ toxicities and hematological complications [23].
Our study focused on 415 hospitalizations involving CAR-T therapy. Among these, 68 (16.38%) experienced CRS. We examined the baseline characteristics of these hospitalizations and determined that the median age was 60 years, with no significant age difference between the two groups (p-value = 0.517). The study comprised 290 males (60.24%) and 193 females (39.76%) encounters; once again, we observed no significant difference in the gender distribution for CRS incidence (p-value = 0.506). This finding aligns with other studies [24]. The study encompassed diverse racial and ethnic groups (white, black, Hispanic, Asian, Native American), and CRS incidence exhibited no significant variation, indicating no racial disparity. No significant associations were identified between CRS and the patient’s median household income percentile, as well as insurance status, including Medicare, Medicaid, private insurance, or lack of insurance. However, our study did reveal a significant association in the CRS group, with a Charlson Comorbidity Index score of 2 or more demonstrating a higher incidence of CRS. There was no significant association between CRS and different geographical locations or hospital bed sizes.
Although the incidence of mortality in the CRS group was higher in the present study sample (5.89% vs. 2.89%), there was no statistically significant difference in mortality between both study groups (aOR = 2.48; p = 0.151). This could potentially be secondary to treatment and interventions, including therapy with steroids, tocilizumab, or anakinra, along with supportive measures [25]. The difference in LOS and mean total hospital charges between both groups was found to be statistically insignificant (coefficient = -2.1 days and p = 0.207; coefficient = $207,456 and p = 0.057).
Based on our analysis, we found that the group with CRS had a significant association with fever (aOR = 1.91; p = 0.015), acute respiratory failure (aOR = 2.10; p = 0.049), and requiring intubation/mechanical ventilation (aOR = 2.59; p = 0.024). Other risks and outcomes that were studied and which showed no statistical significance were sepsis (aOR = 1.68; p = 0.348), vasopressor support requirement (aOR = 2.39; p = 0.334), encephalopathy (aOR = 1.03; p = 0.940), Clostridioides difficile infection (aOR = 0.94; p = 0.921), seizures (aOR = 3.96; p = 0.082), AKI (aOR = 1.69; p = 0.141), acute cerebrovascular event (p = NA), acute liver injury (aOR = 1.71; p = 0.651), hepatomegaly & splenomegaly (aOR = 1.73; p = 0.326), acute MI (p = NA), and CHF (aOR = 0.85; p = 0.833).
This study was extended to include hematological complications associated with CRS and CAR-T therapy. It revealed that the group with CRS had a statistically significant association with the development of HLH (aOR = 6.72; p = 0.002). This finding is related to the underlying pathophysiology of hyperinflammation and increased cytokine activity in CRS [18-20]. There was a statistically lower incidence of anemia in CRS-related CAR-T therapy when odds ratios were compared between the study groups (aOR = 0.5; p = 0.029), while there was a higher incidence of thrombocytopenia (19.1% vs. 15%) and pancytopenia (66.2% vs. 62.2%) in the CRS group. However, there was no statistical difference when comparing odds ratios (aOR = 1.42 and p = 0.350 for thrombocytopenia; aOR = 1.40 and p = 0.307 for pancytopenia). There were not enough reports on catastrophic bleeding, pulmonary embolism, and intracranial hemorrhage. CAR-T therapy with CRS was associated with DIC (4.4% vs. 2.6%) compared to those without CRS; however, this association was found to be insignificant (aOR = 1.73; p = 0.445). The need for red blood cell transfusion, platelet transfusion, FFP transfusion, and cryoprecipitate transfusion showed no statistically significant difference between hospitalizations of CAR-T patients with CRS and those without CRS (all p > 0.05).
Limitations
The NIS database lacks data regarding pre-admission and post-discharge information, limiting our ability to conduct long-term follow-up. Additionally, our analysis captures hospitalizations rather than individual patients, potentially resulting in duplicated data for readmissions. Furthermore, our dataset lacks significant details such as imaging, laboratory values, coagulation panel results, treatment strategies, and cause of death analyses. It is also important to note that our findings establish associations rather than causal relationships with the events we studied. Given that our data are cross-sectional in nature, it is important to interpret our findings with caution. Our analyses were carried out using retrospective registry data, which introduces the possibility of selection bias due to potential selective reporting and the use of ICD codes to define the patient cohort.
Despite these limitations, our study offers valuable insights into the occurrence of CRS in patients receiving CAR-T therapy. Our ultimate aim is to enhance patient outcomes by providing guidance for clinical decision-making. While acknowledging the existence of coding errors and variations, it is worth noting that our study is based on a substantial sample drawn from this database and represents a diverse population across the United States, and incorporates data from numerous medical centers.
Conclusions
Based on the above analysis, it appears that the association of CRS in patients treated with CAR-T therapy did not increase the risk of mortality as well as the LOS in US hospitalized patients. The study does highlight a statistically significant association of CRS with the incidence of fever, acute respiratory failure, requirement of intubation, and HLH in patients treated with CAR-T requiring hospitalization. Although the study concludes limited risk of mortality with the development of CRS in CAR-T-treated hospitalized patients, clinicians should be vigilant with prompt recognition of CRS in such patients and initiate appropriate management.
Appendices
Table 5. Supplemental table.
CAR-T: chimeric antigen receptor T-cell; CRS: cytokine release syndrome; AKI: acute kidney injury; CVA: cerebrovascular accident; CHF: congestive heart failure; HLH: hemophagocytic lymphohistiocytosis; DIC: disseminated intravascular coagulation; FFP: fresh frozen plasma.
| Variable | ICD-10/Procedure codes used |
| CAR-T therapy | XW033C3, XW043C3 |
| CRS | D89.83, D89.831, D89.832, D89.833, D89.834, D89.835, D89.839 |
| Sepsis | A021, A227, A267, A327, A400, A401, A403, A408, A409, A41,A4101, A4102, A411, A412, A413, A414, A4150, A4151, A4152, A4153, A4159, A4181, A4189, A419, A427, A5486, B377, P360, P3610, P3619, P362, P3630, P3639, P364, P365, P368, P369, R6520, R6521, T8144XA, T8144XD, T8144XS |
| Vasopressor support | 3E030XZ, 3E033XZ, 3E040XZ, 3E043XZ, 3E050XZ, 3E053XZ, 3E060XZ, 3E063XZ |
| Acute respiratory failure | J9600, J9601, J9602, J9620, J9621, J9622, J9690, J9691, J9692 |
| Intubation/mechanical ventilation | 5A09357, 5A09457, 5A09557, 09HN7BZ, 0CHY7BZ, 0DH57BZ, 0NH17EZ, 5A1935Z, 5A1945Z, 5A1955Z, 0BH07DZ |
| Clostridium difficile infection | A047, A0471, A0472 |
| AKI | N170, N171, N172, N178, N179 |
| Acute CVA | I639, I638, I6359, I63549, I63219, I63212, I63211, I6320, I6309, I63039, I63032, I63031, I6302, I63019, I63012, I63011, I6300 |
| Encephalopathy | G92, G93.4, G93.40, G93.41, G93.49 |
| Seizures | G40 |
| Acute liver injury | K7200, K7201, K7290, K7291 |
| Hepato &/splenomegaly | R160, R161, R162 |
| Acute myocardial infarction | I2101, I2102, I2109, I2111, I2119, I2121, I2129, I213, I214 |
| CHF | I50, I97. 13 |
| Anemia | D50, D51, D52, D53, D55, D56, D57, D58, D59, D60, D61, D62, D63, D64, D46.0, D46.1, D46.2, D46.4, O99.0 |
| Thrombocytopenia | D6942, D6949, D6959, D696 |
| Pancytopenia | D61.81, D61.810, D61.811, D61.818 |
| HLH | D76.1 |
| Pulmonary embolism | I26, I260, I2602, I2609, I269, I2692, I2699 |
| Major bleeding | R58, L7622, K661, I62, I620, I6200, I6201, I6202, I6203, I621, I629, R04, R040, R041, R042, R048, R0481, R0489, R049 |
| Intracranial hemorrhage | I6000, I6001, I6002, I6010, I6011, I6012, I602, I6030, I6031, I6032, I604, I6050, I6051, I6052, I606, I607, I608, I609, I6030, I6030, I610, I611, I612, I613, I614, I615, I616, I618, I619, I6200, I6201, I6202, I6203, I621, I629 |
| GI bleeding | K920, K921, K922, K625, K928, K929, K9281, K9282, K9289, K250, K254, K260, K264, K270, K28, K621 |
| DIC | D65 |
| Red blood cell transfusion | 30233N0, 30233N1, 30243N0, 30243N1, 30273N1, 30277N1, 30233P0, 30233P1, 30243P0, 30243P1 |
| Platelets transfusion | 30233R1, 30243R0, 30243R1, 30273R1, 30277R1 |
| FFP transfusion | 30233L0, 30233L1, 30243L0, 30243L1, 30273L1, 30233K0, 30233K1, 30243K0, 30243K1, 30273K1, 30277L1, 30277K1 |
| Cryoprecipitate transfusion | 30233M0, 30233M1, 30243M0, 30243M1, 30273M1, 30277M1, 30233D1, 30243D1 |
The authors have declared that no competing interests exist.
Author Contributions
Concept and design: Rushin Patel, Mrunal Patel, Fehmida Laxmidhar, Khushboo Lakhatariya, Zalak Patel, Safia Shaikh
Acquisition, analysis, or interpretation of data: Rushin Patel, Mrunal Patel, Darshil Patel, Safia Shaikh
Drafting of the manuscript: Rushin Patel, Mrunal Patel, Fehmida Laxmidhar, Khushboo Lakhatariya, Zalak Patel, Safia Shaikh
Critical review of the manuscript for important intellectual content: Rushin Patel, Fehmida Laxmidhar, Khushboo Lakhatariya, Darshil Patel, Safia Shaikh
Human Ethics
Consent was obtained or waived by all participants in this study
Animal Ethics
Animal subjects: All authors have confirmed that this study did not involve animal subjects or tissue.
References
- 1.Cancer T-cell therapy: building the foundation for a cure. Kamb A, Go WY. F1000Res. 2020;9:1295. doi: 10.12688/f1000research.27217.1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Chimeric antigen receptor T-cell therapies for lymphoma. Brudno JN, Kochenderfer JN. Nat Rev Clin Oncol. 2018;15:31–46. doi: 10.1038/nrclinonc.2017.128. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.A review of clinical outcomes of CAR T-cell therapies for B-acute lymphoblastic leukemia. Martino M, Alati C, Canale FA, Musuraca G, Martinelli G, Cerchione C. Int J Mol Sci. 2021;22:2150. doi: 10.3390/ijms22042150. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.CART-cell therapy: recent advances and new evidence in multiple myeloma. Martino M, Canale FA, Alati C, et al. Cancers (Basel) 2021;13:2639. doi: 10.3390/cancers13112639. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Chimeric antigen receptor T-cell therapy for T-ALL and AML. Wei W, Yang D, Chen X, Liang D, Zou L, Zhao X. Front Oncol. 2022;12:967754. doi: 10.3389/fonc.2022.967754. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Optimizing CAR T cell therapy in lymphoma. Qualls D, Salles G. Hematol Oncol. 2021;39:104–112. doi: 10.1002/hon.2844. [DOI] [PubMed] [Google Scholar]
- 7.Cytokine release syndrome. Shimabukuro-Vornhagen A, Gödel P, Subklewe M, et al. J Immunother Cancer. 2018;6:56. doi: 10.1186/s40425-018-0343-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.CAR T-cell therapy and critical care: a survival guide for medical emergency teams. Messmer AS, Que YA, Schankin C, Banz Y, Bacher U, Novak U, Pabst T. Wien Klin Wochenschr. 2021;133:1318–1325. doi: 10.1007/s00508-021-01948-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Current concepts in the diagnosis and management of cytokine release syndrome. Lee DW, Gardner R, Porter DL, et al. Blood. 2014;124:188–195. doi: 10.1182/blood-2014-05-552729. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Toxicities of chimeric antigen receptor T cells: recognition and management. Brudno JN, Kochenderfer JN. Blood. 2016;127:3321–3330. doi: 10.1182/blood-2016-04-703751. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Assessment and management of cytokine release syndrome and neurotoxicity following CD19 CAR-T cell therapy. Chou CK, Turtle CJ. Expert Opin Biol Ther. 2020;20:653–664. doi: 10.1080/14712598.2020.1729735. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Cytokine release syndrome: current perspectives. Murthy H, Iqbal M, Chavez JC, Kharfan-Dabaja MA. Immunotargets Ther. 2019;8:43–52. doi: 10.2147/ITT.S202015. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Cytokine release syndrome: who is at risk and how to treat. Frey N. Best Pract Res Clin Haematol. 2017;30:336–340. doi: 10.1016/j.beha.2017.09.002. [DOI] [PubMed] [Google Scholar]
- 14.Agency for Healthcare Research and Quality. Overview of HCUP. [ Oct; 2023 ]. 2022. https://www.ahrq.gov/data/hcup/index.html https://www.ahrq.gov/data/hcup/index.html
- 15.Adherence to methodological standards in research using the National Inpatient Sample. Khera R, Angraal S, Couch T, et al. JAMA. 2017;318:2011–2018. doi: 10.1001/jama.2017.17653. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.New ICD-10 version of the Charlson comorbidity index predicted in-hospital mortality. Sundararajan V, Henderson T, Perry C, Muggivan A, Quan H, Ghali WA. J Clin Epidemiol. 2004;57:1288–1294. doi: 10.1016/j.jclinepi.2004.03.012. [DOI] [PubMed] [Google Scholar]
- 17.CAR-T cell therapy for hematological malignancies: limitations and optimization strategies. Huang J, Huang X, Huang J. Front Immunol. 2022;13:1019115. doi: 10.3389/fimmu.2022.1019115. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Anakinra for refractory cytokine release syndrome or immune effector cell-associated neurotoxicity syndrome after chimeric antigen receptor T cell therapy. Gazeau N, Liang EC, Wu QV, et al. Transplant Cell Ther. 2023;29:430–437. doi: 10.1016/j.jtct.2023.04.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Cytokine release syndrome with chimeric antigen receptor T cell therapy. Frey N, Porter D. Biol Blood Marrow Transplant. 2019;25:0–7. doi: 10.1016/j.bbmt.2018.12.756. [DOI] [PubMed] [Google Scholar]
- 20.Grading and management of cytokine release syndrome in patients treated with tisagenlecleucel in the JULIET trial. Schuster SJ, Maziarz RT, Rusch ES, et al. https://doi.org/10.1182/bloodadvances.2019001304. Blood Adv. 2020;4:1432–1439. doi: 10.1182/bloodadvances.2019001304. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Updated safety and long term clinical outcomes in TRANSCEND NHL 001, pivotal trial of lisocabtagene maraleucel (JCAR017) in R/R aggressive NHL. Abramson JS, Gordon LI, Palomba ML, et al. J Clin Oncol. 2018;36:7505. [Google Scholar]
- 22.T cells expressing an anti-B-cell maturation antigen chimeric antigen receptor cause remissions of multiple myeloma. Ali SA, Shi V, Maric I, et al. Blood. 2016;128:1688–1700. doi: 10.1182/blood-2016-04-711903. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Pericytes regulate the blood-brain barrier. Armulik A, Genové G, Mäe M, et al. Nature. 2010;468:557–561. doi: 10.1038/nature09522. [DOI] [PubMed] [Google Scholar]
- 24.Health care resource utilization and total costs of care among patients with diffuse large B cell lymphoma treated with chimeric antigen receptor T cell therapy in the United States. Keating SJ, Gu T, Jun MP, McBride A. Transplant Cell Ther. 2022;28:404–406. doi: 10.1016/j.jtct.2022.03.021. [DOI] [PubMed] [Google Scholar]
- 25.The pathogenesis, diagnosis, prevention, and treatment of CAR-T cell therapy-related adverse reactions. Li Y, Ming Y, Fu R, et al. Front Pharmacol. 2022;13:950923. doi: 10.3389/fphar.2022.950923. [DOI] [PMC free article] [PubMed] [Google Scholar]