Elucidating the Intricacies of Cytokine Release Syndrome (CRS) in Hematological Malignancies and the Associated Risk Factors: A National Study

Introduction: Hematologic malignancies have been associated with an elevated risk of Cytokine Release Syndrome (CRS) due to cytotoxic chemotherapies and the emergence of recent Chimeric Antigen Receptor T (CAR-T) cellular therapies. Nevertheless, there are several additional risk factors that warrant further investigation. The aim of our study was to explore the relationship between various risk factors and CRS in patients with hematological malignancies.

Materials and Methods: We employed the National Inpatient Sample (NIS) data from 2019 and 2020 to identify individuals with primary discharge diagnoses of hematological malignancies, including leukemias, lymphomas, and multiple myeloma, and a secondary diagnosis of Cytokine Release Syndrome (CRS). The outcomes included calculating mortality, length of stay, and total cost of treatment in CRS patients with hematological cancers. Subsequently, we conducted a multivariate regression analysis to evaluate the likelihood of CRS in patients with different associated risk factors.

Results: A total of 200,590 patients were hospitalized with hematological malignancies, of which 340 developed CRS. No statistically significant differences were observed in baseline demographic characteristics such as age, sex, insurance and income status, race, hospital teaching, rural, and size status. However, the odds of mortality were increased in CRS patients (OR 3.32, 95% CI 2.93-3.76, P<0.001). Total charges were significantly increased in CRS patients (+USD 654,286, 95% CI 375,835-932,636, P<0.001), but no difference was noted in length of stay between the two groups (+3.13, 95% CI 0.38-5.88, P=0.025). Fluid and electrolyte disorders (OR 2.71, 95% CI 2.47-2.97, P<0.001), obesity (OR 1.15, 95% CI 1.01-1.32, P=0.027), and heart failure (OR 1.39, 95% CI 1.2-1.6, P<0.001) demonstrated a higher risk of association with CRS. CRS patients were also more likely to have palliative care involvement (OR 1.71, 95% CI 1.52-1.92, P<0.001). Conversely, hypertension (OR 0.84, 95% CI 0.76-0.93, P=0.001) and major depressive disorder (OR 0.74, 95% CI 0.64-0.86, P<0.001) were associated with decreased risk of CRS in hematological cancer patients.

Conclusion: CRS in hematological cancer patients is a significant concern due to its association with increased mortality and overall hospitalization costs, without any apparent difference in length of stay. Additionally, obesity, heart failure, and fluid and electrolyte disorders have been identified as key risk factors for CRS in these patients. In order to achieve substantial improvements in patient outcomes within hospitals and reduce the likelihood of adverse events, it is of paramount importance to place a strong emphasis on the holistic management of these conditions, while simultaneously adhering to evidence-based practices that are in accordance with current research and clinical guidelines.

Introduction

Cytokine Release Syndrome (CRS) is a complex and potentially life-threatening immune system reaction that has garnered significant attention in the context of treating patients with hematological malignancies [1]. This phenomenon is particularly associated with immunotherapies, including chimeric antigen receptor (CAR) T-cell therapy, which has shown remarkable efficacy in the treatment of certain hematological malignancies [2]. Understanding the intricacies of CRS, its clinical manifestations, and the management strategies in patients with hematological malignancies is crucial for optimizing therapeutic outcomes. Hematological malignancies, such as leukemia, lymphoma, and myeloma, are characterized by the uncontrolled growth and proliferation of abnormal blood cells. These disorders contribute to 6.5% of global cancer cases, constituting around 9.0% of cancer cases in both the U.S. and Europe [3]. Additionally, they are typically associated with the elderly population, with the median age for most of these diseases being approximately 65-70 years [4]. Between 2014 and 2018, the U.S. Cancer Statistics estimated approximately 553,000 individuals were diagnosed with a hematological malignancy, and this number is projected to increase in the future [5]. Traditional treatment modalities, such as chemotherapy and radiation therapy, have been the mainstay in managing these malignancies [6]. However, recent advancements in immunotherapy, specifically CAR T-cell therapy, have demonstrated unprecedented success in inducing durable remissions. Despite these promising results, the emergence of CRS as a potential adverse event has become a focal point of concern [7].

Cytokine Release Syndrome (CRS) is well documented in the literature as one of the most prevalent toxicities related to CAR T-cell therapy. According to a review of published clinical trials, CRS has an incidence of 42-100%, with up to 46% of patients developing severe CRS following the infusion of CAR T-cells [8]. CRS is initiated by the activation of immune cells, particularly T-cells, leading to the rapid release of pro-inflammatory cytokines [9]. In the context of hematological malignancies, the infusion of CAR T-cells contributes to the activation cascade, triggering a surge in cytokine production [10]. This cytokine storm, characterized by elevated levels of interleukin-6 (IL-6), interferon-gamma (IFN-γ), and tumor necrosis factor-alpha (TNF-α), among others, results in systemic inflammation and a range of clinical manifestations [10,11]. Clinical manifestations of CRS in patients with hematological malignancies are diverse, encompassing flu-like symptoms, such as fever, chills, and myalgias, to more severe manifestations, including hypotension, respiratory distress, and multi-organ dysfunction [12]. The severity of CRS is often graded according to established criteria, such as the consensus grading criteria determined by experts in the field for the American Society for Transplantation and Cellular Therapy (ASTCT), allowing clinicians to assess the degree of immune activation and tailor therapeutic interventions accordingly [13]. Understanding the underlying mechanisms of CRS is crucial for developing targeted interventions to mitigate its impact on patients with hematological malignancies. 

While some studies have explored the adverse outcomes associated with cytokine release syndrome, there is still a significant knowledge gap and lack of information concerning the risk factors linked to CRS, particularly with the emergence of new targeted therapy modalities. In light of these considerations, we conducted a retrospective study aimed at comprehensively analyzing adverse outcomes in patients admitted with hematological malignancies who developed CRS. Additionally, our investigation sought to identify specific risk factors that were more likely to be associated with CRS development and the impact on resource utilization.

Undertaking extensive research within the expansive National Inpatient Sample database (NIS), a vast repository comprising data from roughly eight million hospital stays per annum, we gained access to the most comprehensive all-payer inpatient care database in the United States. This meticulously constructed database is derived from billing data sourced from State Inpatient Databases and is designed to reflect the operations of approximately 20% of community hospitals across the nation. Our examination of this extensive dataset involved a thorough analysis that allowed us to uncover crucial insights that informed our research objectives. Specifically, we utilized data from the National Inpatient Sample for the years 2019 and 2020 to identify and analyze patients aged 18 years and above who were hospitalized with a primary discharge, including any of these hematological malignancies (Leukemia, lymphoma, and multiple myeloma). The cohort was subsequently divided into two subgroups based on the concurrent presence and absence of CRS. Our study utilized the International Classification of Diseases, Tenth Revision, Clinical Modification/Procedure Coding System (ICD-10-CM/PCS) as the basis for our coding methodology. This observational approach allowed for a comprehensive analysis of mortality rates, length of stay, total cost of hospitalization, and identification of risk factors associated with increased odds of CRS.

Categorical data were represented as percentages to provide deeper insight into their distribution across different groups, while continuous variables were characterized by mean values and standard deviations to capture their diverse nature within the cohorts. To compare outcomes among different categorical variables, we employed Pearson's chi-square test or Fisher's exact test, ensuring robustness in analyzing continuous variables using Student's t-test. Undertakings of univariate and multivariate analyses were executed with utmost care to scrutinize the interactions of diverse risk factors with CRS in hematological cancer patients. The primary objective of this study was to pinpoint the precise risk factors contributing to the emergence of CRS in this class of hematological cancer patients. However, it is crucial to recognize that the investigation did not venture beyond mortality rates and resource utilization in examining adverse inpatient events. Acquiring such information is essential for optimizing resource distribution and fine-tuning practices to prevent future occurrences. Moreover, these observations could offer priceless guidance for the ongoing care of patients by enabling healthcare providers to make well-informed decisions grounded in probabilistic calculations. The statistical examinations were conducted with utmost precision using Stata 17 software, thereby significantly contributing to the exploration of research questions and hypotheses and enhancing the overall validity and rigor of the study.

A total of 200,590 patients with hematological malignancies were hospitalized, among whom 340 developed CRS. Table 1 presents a comparison of the baseline demographic characteristics, insurance and income status, hospital characteristics (such as size, location, and region in the USA), and discharge disposition between the two groups of patients. No statistically significant differences were observed in the baseline characteristics of both groups of patients, except for the Charlson comorbidity index, which showed that a greater percentage of patients with CRS had a Charlson score of 2 (54.55%) compared to those without CRS (34.16%). Conversely, patients without CRS had a higher proportion of Charlson scores of 3 or more (56.38%) compared to those with CRS (27.94%). This difference was found to be statistically significant (P less than 0.001). It was observed that patients with CRS had a lower prevalence of patients with HLD (16.18%) compared to those without CRS (26.85%) (P=0.042). Additionally, a significantly greater percentage of patients with CRS had undergone CAR-T cellular therapy (48.53%) compared to those without CRS (0.87%) (P less than 0.001). (Table 1).

Table 1: Comparison of baseline characteristics of hematological cancer patients with and without CRS

Regarding mortality rates, patients without CRS had a rate of 6.5%, while patients with CRS had a rate of 5.9%. After adjusting for confounding variables, it was found that the odds of mortality were significantly increased in patients with CRS (OR 3.32, 95% CI 2.93-3.76, P less than 0.001). Moreover, the total cost of hospitalization was found to be higher in hematological cancer patients with CRS (+$654,286, 95% CI $375,935-$932,636, P less than 0.001). However, no significant difference was observed in the length of stay between the two groups (OR 3.13, 95% CI 0.38-5.88, P=0.025) (Table 2 & 3).

Table 2: Rate and Odds Ratio of Mortality in Hematological cancer patients with and without CRS

Table 3: Comparison of Length of stay and total charges in Hematological cancer patients with and without CRS

Odds ratios for the association of several risk factors with the development of CRS were also calculated. Female patients were found to have a lower likelihood of developing CRS as compared to male patients (OR 0.78, 95% CI 0.71-0.85, P less than 0.001). Patients older than 46 had decreased odds of association with CRS as compared to those aged 18-35 (46-64 (OR 0.71,95% CI 0.6-0.84, P less than 0.001), greater than 65 (OR 0.76, 95% CI 0.62-0.92, P=0.006), while patients aged 36-45 did not show any statistically significant association with the development of CRS (OR 0.86, 95% CI 0.69-1.07, P=0.186). As compared to white patients, Black and Hispanic patients had decreased odds of association with CRS (OR 0.81, 95% CI 0.71-0.94, P=0.005 & OR 0.85, 95% CI 0.73-0.99, P=0.047 respectively), while patients of other races did not show any statistically significant difference in the association with CRS (OR 1.01, 95% CI 0.83-1.31, P=0.687). Fluid and electrolyte disorders, heart failure, and obesity were found to have higher odds of developing CRS as compared to those without these conditions (OR 2.71, 95% CI 2.47-2.97, P less than 0.001), (OR 1.39, 95% CI 1.2-1.6, P less than 0.001), and (OR 1.15, 95% CI 1.01-1.32, P=0.027) respectively. Additionally, patients with CRS had an increased likelihood of palliative care involvement (OR 1.71, 95% CI 1.52-1.92, P less than 0.001). Hypertension, major depressive disorder, and patients who underwent hematopoietic stem cell transplant had decreased odds of association with CRS (OR 0.84, 95% CI 0.76-0.93, P=0.001), (OR 0.74, 95% CI 0.64-0.86, P less than 0.001), and (OR 0.11, 95% CI 0.08-0.15, P less than 0.001) respectively. However, no significant difference was observed in the association with CRS in patients who underwent chemotherapy and CAR-T cellular therapy (OR 1.41, 95% CI 0.96-2.08, p=0.076 & OR 1.62, 95% CI 0.88-2.98, P=0.121 respectively) (Table 4).

Table 4: Risk factors of CRS in Hematological cancer patients

In our analysis of 200,590 patients with hematological malignancies that were hospitalized, 340 of them developed CRS.  Our study found that CRS was linked to elevated rates and odds of mortality.  Cai et al. conducted a meta-analysis examining the adverse events and time to death associated with various toxic effects of CAR-T cellular therapy. The authors found that CRS was a common adverse event and had a significantly shorter median time to death compared to other adverse events. This underscores the critical clinical implications of CRS as an adverse event that requires meticulous attention, particularly in light of the latest treatment modalities such as CAR-T cellular therapies, which are thought to be associated with a heightened risk of CRS [14]. Our retrospective study also uncovered the association between CRS and increased hospitalization costs. There are a number of factors that can lead to higher hospitalization costs as a result of CRS. For one, it is important to consider the symptom severity of CRS. CRS can lead to a large array of symptoms ranging from flu-like symptoms to organ dysfunction. Hospitalization costs incurred may increase substantially as a result of more severe symptoms, or ICU admission. Additionally, the administration of anti-cytokine medications or other medical interventions may contribute to increased costs. Other factors include the costs of laboratory and diagnostic tests, the collaboration of various medical specialists such as oncologists, immunologists and critical care physicians, and unplanned admissions as a result of unforeseen CRS [15].

From our investigation it is evident that there are a number of risk factors associated with CRS in hematological cancer patients. Our study identified fluid and electrolyte disorders, obesity, and heart failure as risk factors for CRS. Conversely, hypertension and major depressive disorder were found to be associated with a decreased risk of CRS in hematological patients. The results of the study demonstrate that hematological cancer patients with fluid and electrolyte disorders are at increased risk of CRS. Current evidence suggests that IL-6 is involved in regulating the secretion of vasopressin, which, when elevated along with increased water intake, likely contributes to hyponatremia [16,17]. While the exact mechanism behind these electrolyte disorders remains unclear, these findings highlight the importance of closely monitoring electrolytes and volume status to mitigate risk of CRS in patients being treated for hematological cancer. Obesity is recognized as a major factor contributing to the production of inflammatory cytokines. Excessive adipose tissue, characteristic of obesity, can function as an endocrine organ releasing various proinflammatory mediators. These stimulate immune cells to release inflammatory cytokines such as TNF-a and IL-6, leading to chronic low levels of inflammation throughout the body [18]. The inflammatory and immunological changes seen in obesity not only impact end-organ pathology, but also bear significance for the host’s response to cancer immunotherapy [19]. Consistent with the findings in this study, Canter et al. found that individuals with preexisting inflammation, such as those who are obese, may be at increased risk of immunotherapy toxicities like CRS. Furthermore, these observations have been substantiated by mouse studies and patient observations suggesting that chronic levels of low-grade inflammation are likely exacerbated once CRS occurs [20,21]. Ultimately, given the significant impact of excessive adipose tissue on immunologic processes, more research is needed to elucidate its effect on responses to immunotherapy.

Although the role of the immune system in cardiac pathology remains in the early stages of comprehension, it is evident that the heart relies heavily on immune stability to maintain functionality [22]. Conditions like chronic heart failure, myocardial infarction, and cardiomyopathies have been associated with elevated levels of proinflammatory cytokines, although it is unclear whether these markers are an indicator of disease severity or actively contribute to cardiac dysfunction [23]. Given this, a growing body of evidence underscores the close relationship between heart failure and CRS [22,23]. While existing research suggests that CRS can lead to heart failure, the findings presented here further explore this relationship, revealing that heart failure also independently poses a risk factor for CRS in hematologic cancer patients. While the etiology of hypertension, like obesity, is associated with increased levels of proinflammatory cytokines and other immunological changes, the findings presented here indicate a decreased risk of CRS in patients with hematological cancer. Although prior research has identified hypertension as an independent risk factor for COVID-19 induced CRS, this study marks the first to demonstrate hypertension’s association with a decreased risk of CRS in hematological cancer patients [24]. Given the scarcity of studies exploring the connection of hypertension to CRS, this relationship warrants further attention.

Existing literature has previously established a bidirectional relationship between depression and inflammation with depression creating a proinflammatory state and in turn, influencing mood [25,26]. Specifically, the inflammatory cytokine IL-6 has been identified as a major contributor in the development of depression [26]. In light of current research, our study provides evidence suggesting a protective association between major depressive disorder and CRS, indicating complexities in these relationships that necessitate further investigation. One possible explanation for the observed phenomenon could be that patients who experience cytokine release syndrome (CRS) are frequently in a precarious state of health, with a substantial disease burden that encompasses both physical and psychological dimensions. These individuals are more likely to be receiving medication that has a positive impact on mood or stabilizes their condition, which may contribute to the effective management of depression or the resolution of symptoms. This could potentially account for the weaker association observed in patients with CRS. These results emphasize the pressing need to integrate mental health factors into the treatment and risk assessment of hematological cancer patients. Additional research on the emotional well-being of this patient population is merited. 

The findings of this retrospective study emphasize the imperative of comprehensive disease management, particularly in patients with hematological cancers and CRS. Comprehensive management strategies can optimize the efficacy of treatment, especially in cases where CRS is a side effect of CAR-T and other cytotoxic therapies. By managing the CRS, medical professionals can continue to treat the primary cancer. Other benefits of adopting a comprehensive treatment approach include minimizing complications associated with uncontrolled CRS, adopting an individualized approach to care, preventing the escalation of CRS symptoms, providing supportive and preventive measures such as hydration and pain management in order to provide comfort, and patient education to involve the patient and empower them in their own care. Comprehensive management strategies also enable interdisciplinary collaboration between medical personnel to adopt a holistic approach to care, and can lead to long-term monitoring and the advancement of research through the accumulation of more data and related variables that may refine treatment strategies for future patients [27].

It is important to recognize inherent limitations with the use of databases in retrospective studies. One limitation includes sampling bias, as the National Inpatient Sample (NIS) is not representative of the entire U.S. population, nor does it account for variations across regions or types of hospitals across the United States. Additionally, the NIS relies on diagnostic and procedural codes such as ICD codes, where accuracy and classification errors are often inevitable. Other limitations include the lack of longitudinal data to track follow-ups after discharge, limited clinical detail requiring additional datasets, and variability in data collection practices across hospital facilities. It is also important to note that NIS sampling methodology may alter periodically, and that certain states might have restrictions on the type of data to be included in the NIS [28].

The findings of this study provide critical insights into the relationship between hematological malignancies and their association with CRS. However, several avenues remain underexplored and merit further investigation to enhance our understanding and guide clinical practices. Firstly, given the emergent nature of CART-T cellular therapies and their impact on the development of CRS, future research is needed to optimize these therapies to mitigate risk of CRS without compromising therapeutic efficacy. Specifically, understanding the molecular pathways triggering CRS post CAR-T therapy could pave the way for targeted interventions. Secondly, while our study identified several risk factors associated with CRS, there is a need to explore these potential interactions and genetic predispositions. This may shed light on individual susceptibility to CRS, aiding in personalized treatment strategies. Lastly, as technological advancements continue to reshape healthcare, incorporating real-time monitoring and predictive analytics could aid in early detection and management of CRS, leading to improved patient outcomes. While this study offers valuable insights, the understanding of CRS in hematological malignancies remains underexplored, necessitating collaborative efforts from clinicians and researchers alike.

Our study found that CRS was associated with increased mortality and hospitalization costs in patients with hematological cancers. Key risk factors for CRS in these patients included fluid and electrolyte disorders, obesity, and heart failure, while hypertension and major depressive disorder were associated with decreased risk for CRS. These findings highlight the importance of evidence-based management for patients with hematological malignancies and CRS, which can improve patient outcomes and reduce adverse events in hospital settings. Further research is needed to understand protective mechanisms or factors contributing to these observations.