Most studies that describe patients who were treated with CAR-T cell therapy have analyzed the clinical experience from the starting point of lymphodepletion until the very end of treatment but with limited information given regarding the patients who were admitted to the ICU with no mention of the resources used or required life-saving interventions that could have improved their different outcomes. Therefore, the information concerning this segment of practice is scarce [
8]. Current ICU practices in CAR-T cell recipients admitted to the ICU are deduced from the general ICU population and are not individualized accordingly.
As intensive care physicians, it is important to acknowledge the currently available information and experiences of the other medical specialties regarding immunotherapies and contribute by creating a standard of care for future ICU patients treated with CAR-T cell therapy. It is mandatory to have know-how regarding the CAR products available, toxicities observed, and available treatment of toxicities, and to update ourselves with the specifics of practices and interventions applicable within the ICU when caring for critically ill CAR patients.
2. Outcomes of CAR-T Cell Therapy Recipients Admitted to the ICU
Given the matter that around one-third of the total recipients of CAR-T cells required ICU admission, it represents an important fact to remember that highlights the need to have an available bed or dedicated high-dependency unit with trained medical personnel. Even though the incidence of ICU admission showed a substantial level of heterogeneity among the studies (I2 of 80%), this rate of admission was similar to the one mentioned in the literature [6,18,19]. Moreover, the readmission rate to the ICU within 30 days was around 18%, which suggests that physicians should expect readmission after an initial successful recovery in a significant proportion of patients. Reasons for readmission are not mentioned in the assessed studies, but they should be clearly stated for clinical and economic purposes.
Major reasons for ICU admission for patients treated with CAR-T cell are the development of CRS and ICANS, associated sepsis, single or multiple organ failure, hemophagocytic lymphohistiocytosis, stroke, and difficult airway management (bulky cervical or mediastinal lymphadenopathy, and tumor infiltration) [20,21]. CRS and ICANS were present in a third of these patients at the moment of admission to the ICU. Hypotension, altered mental status, AKI, and acute respiratory failure were the main reasons for ICU admission in these studies. This underlines that hematologists should carefully assess these patients daily on the wards for neurological, respiratory, cardiovascular, and kidney organ failure. Usually, the patients deteriorate clinically over time, and there is a failure to identify the process [22,23]. A delay in admission to the ICU and more than one organ failure are associated with increased mortality [24,25]. Validated scores such as MEWS/NEWS should be used daily by the medical personnel to assess the patients on the ward, and with the help of these, early admission to the ICU could be suggested [26,27,28].
CRS is a complication of CAR-T cell treatment. Hypotension, AKI, arrhythmias, and fever are the most frequent signs and symptoms of CRS presentation. The presence of these signs and symptoms should prompt physicians to do a thorough screening for CRS, which could lead to organ failure if not taken care of. Noteworthy, the same signs and symptoms could be present in sepsis, so it is worth remembering that sepsis is a differential diagnosis when dealing with CRS. A comprehensive sepsis screening should rapidly follow with the institution of sepsis treatment if suspected [14,29]. The lymphodepletion regimen given prior to CAR-T cell therapy leads to immunosuppression, which can lead to bacterial infections. Patients with immunosuppression and CAR–T cell-associated toxicities have the highest risk for infection, which is reported to be around 23–43% during the first month after CAR-T therapy infusion [30,31]. Cytokine identification could be used to differentiate between CRS and sepsis.
The most required interventions initiated in ICU in CAR-T cell recipients were high-flow nasal oxygen or noninvasive ventilation, invasive mechanical ventilation, infusion of vasoactive drugs, and a small proportion even required renal replacement therapy (RRT) during ICU stay. Notable, while AKI was one of the main reasons for ICU admission and one of the main signs of CRS presentation, RRT was only required in a low percentage of cases. The lack of diagnostic criteria for AKI within the assessed manuscripts makes it impossible to make assumptions. Future studies should focus on what is the real percentage of patients presenting with AKI and which one of the patients could be helped by initiating RRT early. RRT with Cytosorb® or other hemadsorbers can be used as a bridge therapy for refractory CRS, as well as for managing sepsis, due to the elimination/adsorption of cytokines through the filter [33,34].
The mean length of hospital stay was around 22 days, ICU stay around 5 days, and admission to the intensive care after CAR-T infusion was around 5 days, so physicians should expect a relatively long duration of hospital stay, which increases costs.
A mean SOFA value of 4.5 calculated at the admission moment corresponds to a mortality of around 15–20%, according to the literature [
35,
36]. Hematologic patients are considered to be a frail patient population [
25,
37]. Putting these two facts together, there is a suggestion that there is an increased risk of mortality among hematologic patients treated with CAR-T cells, even though the reported mortality of CAR-T cell recipients admitted to the ICU is around 6%, which could be considered low. A deeper look into this finding is required due to the fact that the mortality rate is disease, treatment, or sepsis-related, and the SOFA calculated in the mentioned studies is not solely performed on oncohematological patients. By having a dedicated team, environment, guidelines, and standard of care, the rate of detection of complications can increase, and earlier treatment started, which could improve the overall survival rate.
The focus of ICU management of these patients should reside on several problems. For an optimal management strategy in cases of respiratory failure, a lung protective strategy and a review of chest and neck imaging in the hematologic population should become standard practice to avoid airway emergencies and adverse outcomes. Regarding fluid management, goal-directed fluid therapy should be followed and helped by the use of performing point-of-care cardiac and lung ultrasound to guide it [
38]. The choice of the most appropriate intravenous fluid, like crystalloids or colloids, and the use of vasoactive agents remain to be studied in the setting of pulmonary capillary leak, CRS-related cardiomyopathy, or oliguric renal failure [
39]. Unfortunately, when sepsis is present in these patients, there are arguments that it is correlated with the risk of particularly poor outcomes [
9]. This suggests that a comprehensive search for the site of infections is mandatory.
Studies that focus on death from the progression of the disease in contrast to death from toxicity of treatment will have to be conducted to establish the incriminating factors. While immunotherapies are in continuous development, with newer CAR-T cell therapies entering the market, the indications for this treatment could become broader [
1,
40]. Toxicities and complications magnitude will have to be considered and anticipated because, at the moment, they are difficult to predict.
Earlier admission to the ICU or high dependency unit (HDU) of patients at high risk of CRS might also improve outcomes, as seen in previous studies on hematological malignancies, and might paradoxically reduce the duration of stay in the ICU [
41]. Having a hematological pathology does not seal access to an ICU, and it does not mean there are no benefits if hospitalized in the ICU, as many might think. The current risk stratification models in the context of immunotherapies are limited. There is not a single perfect tool for early prediction for ICU admission, but by adopting different scores like MEWS/NEWS and having a second opinion from an ICU physician, the threshold for admission could be set lower. Risk prediction models in the context of critically ill patients following immunotherapies could be developed by using clinical prediction models or machine learning algorithms.
There is a dire need to create a standard of care and research to change and improve the current practice, with approaches focusing on reducing the incidence of CRS and ICANS, the need for ICU admission, early sepsis screening, antimicrobial prophylaxis, and research on management strategies to enhance the care of critically ill patients following CAR-T cell therapy, characterization of the clinical course for future prevention planning, all of which work to improve the overall outcomes and increase the quality of life of these patients.
The endless opportunities that immunotherapy has to offer pose a challenge to both hematologists and intensive care physicians. The future is pictured as a worldwide standard of care as ICU management becomes integrated with overall treatment opportunities. More clinical trials are required to address the current state of things, with trials that should also address the different CAR-T therapies available.
Encyclopedia
