Sublingual Microcirculatory Imaging As a Novel Tool to Monitor for Cytokine Release Syndrome after Chimeric Antigen Receptor T-Cell Therapy

Abstract

Background:Chimeric antigen receptor T-cell (CAR-T) therapy is an FDA-approved therapy for relapsed or refractory diffuse large B-cell lymphoma (DLBCL) and acute lymphoblastic leukemia (ALL). A common side effect of CAR-T therapy is cytokine release syndrome (CRS), and its severity ranges from mild to severe, and occasionally resulting in death. Patients at particularly high risk for severe CRS may benefit from earlier supportive care and rescue therapies, such as tocilizumab. Although the median onset of CRS has been reported as two days, no existing prognostic tools adequately assist the bedside clinician with triaging which patients will decompensate and warrant escalation of care. For example, biomarkers such as CRP and ferritin are ineffective in predicting CRS severity. Evaluation of the sublingual microcirculation of patients receiving CAR-T therapy may serve as a valuable surveillance tool.The sublingual microcirculation (defined as blood vessels <20um) has been investigated clinically in various inflammatory settings (e.g., sepsis, hemorrhage, surgery). Persistent alterations in microcirculatory flow correlate with worse clinical outcomes. We have used a handheld dark-field microscope, the Cytocam (Bradeus, Netherlands) to image and analyze the sublingual microcirculation of patients receiving CAR-T therapy.Methods:Video images are collected in real time. Prior to interpretation, all video clips were assessed for quality based on the current microcirculation consensus document. Each acceptable clip was then analyzed by using two point-of-care (POC) scoring systems: 1) the microvascular flow index (MFI) and 2) the point-of-care microcirculation (POEM) score. The MFI is determined by dividing the video monitor into four equal quadrants and grading the overall flow of each quadrant with a score from 0-3 (0 = no flow; 1 = intermittent flow; 2 = sluggish flow; 3 = normal flow). The POEM score utilizes an ordinal 1-5 scale (1 = critically impaired; 2 = impaired; 3 = normal with marked heterogeneity; 4 = normal with mild heterogeneity; 5 = normal) that is a composite of four measurements assessing flow impairment and heterogeneity. For each enrolled patient, a baseline measurement was made immediately prior to CAR-T cell infusion with follow up measurements occurring every six hours beginning at hour 18 after cell infusion until hour 72. After hour 72, measurements were made daily until they were deemed no longer at risk for CRS.Results:At this time there is mature data on 7 patients. All patients received Axicabtagene Ciloleucel- a CD19 directed CAR-T cell therapy (Yescarta, Kite Pharma). All patients had normal baseline microcirculation (MFI > 2.6, POEM=5) and normal or near-normal microcirculation at the end of the study period. No patients developed severe CRS (grade 3 and above). Three patients developed grade 2 CRS that required tocilizumab. Patients #1 and #2 both had significant microcirculatory impairments ≥12 hours prior to developing symptoms severe enough to warrant tocilizumab. Patient #3 had normal microcirculation through the first four days of therapy and developed hypotension on day six. We captured a subtle change from a normal MFI and POEM score to mild impairment with both scoring algorithms on day five, one day prior to clinical manifestations of decompensation. For logistical reasons, subsequent data were unable to be obtained. MFI and POEM scores for all patients are listed below in Table 1. The remaining four patients developed grade 1 CRS with associated mild microcirculatory changes.Conclusions:In this pilot study, POC microcirculatory assessments were successfully used to monitor patients undergoing CAR-T therapy. Patients with more severe CRS manifested lower MFI and POEM scores and maintained their nadir longer than those with milder CRS. Our data suggest that CAR-T patients developing CRS manifest early signs of sublingual microcirculatory dysfunction. Moreover, these microcirculatory defects present prior to the development of standard clinical abnormalities, such as macrocirculatory derangements. While further investigation is ongoing, this tool could be used for earlier identification of patients at risk for CRS in order to deliver earlier appropriate therapies, and ultimately to improve patient outcomes. DisclosuresNo relevant conflicts of interest to declare.