Coagulation Biomarker Signature As a Novel Differential Diagnostic Tool for CRS/Icans and Infection Post-CAR-T Therapy

Background:

Differentiating between cytokine release syndrome (CRS)/immune effector cell-associated neurotoxicity syndrome (ICANS) and infections post-CAR-T therapy remains challenging. Previous biomarkers (such as IL-6, CRP, etc.) and clinical presentations of CRS/ICANS have been inadequate for accurate differential diagnosis.

Methods:

Patients with relapsed/refractory (R/R) acute myeloid leukemia (AML) or acute lymphoblastic leukemia (ALL) enrolled in phase I-II CAR-T trials (Clinical Trials: NCT06209671,NCT05170568, NCT05548088, NCT 03327285) and those with R/R ALL who received KTE-19 (brexucabtagene autoleucel, ChiCTR2300073872) at Peking University People's Hospital were included in the study. Patients with clear signs of infection or coagulation disorders during lymphodepletion chemotherapy were excluded. A control group was established, comprising propensity score-matched AML/ALL patients who had a fever of ≥38°C and confirmed bloodstream infections (BSI) with carbapenem-resistant Enterobacteriaceae (CRE) or Pseudomonas aeruginosa. Adjusted fibrinogen levels were calculated by deducting transfused blood products (3.0g fibrinogen or 1000 ml frozen plasma were considered equivalent to 100mg/dl of measured plasma fibrinogen). Diagnosis of disseminated intravascular coagulation (DIC) was based on the CDSS system.

Results:

Forty-nine patients who underwent CAR-T therapy were enrolled, including 15 patients with commercial KTE-19 and 34 patients in phase I-II trials, comprising 9 AML patients and 40 ALL patients. The median age was 31 years (range: 17-67 years). Twenty-six patients (53%) had previously undergone allogeneic hematopoietic stem cell transplantation (HCT), with 3 patients experiencing extramedullary relapse and minimal residual disease positivity (MRD+) before CAR-T therapy.

Forty-one patients (83%) experienced CRS, with 17 patients (34%) having grade 1 and 24 patients (49%) having grades 2 to 3 CRS (grade 2: 35%, grade 3: 14%). All patients with ICANS also had CRS of grade ≥2. Post-CAR-T cell infusion, 19 patients had fibrinogen levels <1.5g/L (range: 0.29-7.21g/L). Equivalent fibrinogen was calculated through blood product infusion, ranging from -4.91 to 7.21g/L. Twenty-two (44%) patients exhibited increased D-dimer levels (range: 230-3.5*10E5 ng/ml), and 38% had prolonged prothrombin time (PT) (range: 8.5-33.1s). The severity of CRS correlated with bone marrow (BM) blast percentage (r=0.431, P < 0.01) and HCT history (r=-0.406, P < 0.01). PT prolongation started at a median of 8.5 days (IQR: 6.3-10 days) and lasted for a median of 10 days (IQR: 8.3-11 days). D-dimer levels ≥5mg/L began to rise at a median of 8.5 days (IQR: 7.3-9.8 days) and recovered after a median of 12 days (IQR: 8.5-15 days). Recovery times for fibrinogen and D-dimer were longer in patients with grade ≥2 CRS. Among the people who had coagulation disorder, 4 patients (8%) were diagnosed with DIC according to the CDSS.

Distinct coagulation disorder characteristics were observed between the CRS ≥2 and infection groups. The CRS ≥2 group had lower adjusted fibrinogen levels (187 vs. 276 mg/dl, 95% CI: 141-283 vs. 209-383; p<0.001) and higher D-dimer levels (641 vs. 413 ng/ml, 95% CI: 284-2080 vs. 230-1021; p<0.001). Hypofibrinogenemia occurred at a median of 4 days (range: 1-11 days; IQR: 4-6 days) vs. 15 days (range: 2-18 days; IQR: 14-15 days) for grade ≥2 CRS and BSI, respectively. Elevated D-dimer levels began at a median of 2 days (range: 0-8 days; IQR: 2-3 days) vs. 15 days (range: 4-18 days; IQR: 10-17 days) for CRS ≥2 and BSI, respectively.

Conclusion:

Lower adjusted fibrinogen and higher D-dimer levels may serve as novel differential diagnostic tools for distinguishing CRS/ICANS from infections following CAR-T therapy.

No relevant conflicts of interest to declare.