Outcomes in patients treated with CAR-T therapy who were admitted to intensive care

June 2021 Clinical practice Jolien Blokken
3D Rendering of a Natural Killer Cell (NK Cell) destroying a cancer cell

Side effects such as cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS) often occur after the administration of chimeric antigen receptor (CAR)-T cell therapy and frequently require intensive care unit admission. The international CARRTAS study describes the management and outcomes of critically ill CAR-T recipients in intensive care.

Expert opinion of dr. Jan Van Droogenbroeck, Haematologist, AZ Sint-Jan

The international CARRTAS study was performed in 21 intensive care units (ICU) in Europe, Canada and the United States. Of the 942 patients who were treated with CAR-T’s, 27% had to be referred to an ICU and 26% were included in this overview. “As the therapeutic area for CAR-T cells will most likely expand substantially in the near future, these researchers tried to map the challenges associated with cytokine release syndrome (CRS) and the increased risk of severe bacterial infections. I expect that active immunotherapy is here to stay, so we must be armed against its side effects. This article is a call to optimise intensive care for patients treated with CAR-T cells.”


Chimeric antigen receptor (CAR)-T cell therapy has emerged as a promising treatment for refractory haematological malignancies.1–3 However, the success of CAR-T cells comes at the cost of severe side-effects such as cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS). These complications often require intensive care unit (ICU) admission and can lead to multi-organ failure and dead, even before patients can be assessed for a clinical response.2-4 Data are needed on the outcomes of CRS and ICANS in patients admitted to ICUs and the relation of these adverse events to CAR-T cell therapies.

The CARTTAS study

CAR-T cell Toxicity and Sepsis (CARTTAS) is an international, multicentre, observational cohort study established by the Nine-I study group.6 In total, 21 ICUs from eight countries (France, Spain, the US, the UK, Russia, Canada, Germany and Austria) enrolled patients who received CAR-T treatment. Both patients who were enrolled in clinical trials and those receiving commercially approved products were included. Eligible patients were aged 18 years or older, had received CAR-T cell therapy in the past 30 days and had been admitted to intensive care for any reason. Investigators retrospectively included patients admitted between February 1, 2018, and February 1, 2019, and prospectively included patients admitted between March 1, 2019, and February 1, 2020. Primary endpoint of the study was mortality from ICU admission to day 90. Between February 1, 2018, and February 1, 2020, 942 patients received CAR-T cell therapy in the 21 participating centres, including 803 (85%) for diffuse B-cell or follicular lymphoma, 106 (11%) for B-cell acute lymphocytic leukaemia, and 33 (4%) for multiple myeloma. Among them, 258 (27%) required admission to intensive care and 241 (26%) were included in the CARTTAS registry.7

ICU admission and the need for life-saving treatment

Within this cohort, the 90-day mortality across all participants was 22.4% with an ICU mortality of 5.8% and a hospital mortality of 14.9%. The median time from CAR-T cell infusion to death was 25 days. Admission to intensive care was needed within a median of 4.5 days from the CAR-T cell infusion. Within 24 hours after admission to the ICU, 41.5% of patients had isolated CRS, 41.5% had CRS and ICANS and 2.9% had isolated ICANS. Grade 3–4 CRS within one day of admission to intensive care was found in 50 (25%) of 200 patients while grade 3–4 ICANS was observed in 38 (35%) of 108 patients. A bacterial infection developed in 30 (12%) patients, including ten patients with CRS. Life-saving treatments within 24 hours after admission to intensive care were necessary in 31% of patients, primarily consisting of vasoactive drugs (27%), invasive mechanical ventilation (7%) and high-flow nasal oxygen (5%). The number of patients requiring immediate lifesaving therapies did not differ significantly between centres. Three factors were independently associated with 90-day mortality: a Clinical Frailty Score greater than 3 (HR[95%CI]:  2.51[1.37–4.57], p< 0.010), microbiologically documented infection (HR[95%CI]: 2.12[1.11-4.08], p= 0.020) and need for lifesaving therapy within 24 hours after admission to intensive care (HR[95%CI]: 1.80[1.05–3.10], p= 0.030).7


Intensive care management is an integral part of CAR T-cell therapy. CRS, ICANS and infections occur in a substantial number of patients and are associated with mortality. Studies aimed at standardising the management of CAR T-cell recipients admitted to ICU are therefore warranted.7


  1. Maude SL, et al. N Engl J Med. 2018;378:439-48.
  2. Schuster SJ, et al. N Eng J Med. 2017;377:2545-54.
  3. Neelapu SS, et al. N Eng J Med. 2017;377:2531-44.
  4. Chou CK, et al. Bone Marrow Transplant. 2019;54(Suppl 2):780-84.
  5. Azoulay E, et al. Intensive Care Med. 2017;43:1366-82.
  6. Azoulay E, et al. Intensive Care Med. 2017;43:1808-19.
  7. Azoulay E, et al. Lancet Haematol. 2021;8:e355-64.