Interleukin‐1 blockade with anakinra in acute leukaemia patients with severe COVID‐19 pneumonia appears safe and may result in clinical improvement

Abstract

As of 17 May 2020 the number of patients infected by coronavirus disease 2019 (COVID-19) worldwide has exceeded 4·5 million.1 A subgroup of patients with COVID-19 pneumonia develop a hyperinflammatory syndrome which has a similar cytokine release profile to secondary haemophagocytic lymphohistiocytosis (HLH).2 Immunomodulatory drugs are hypothesised to abrogate the dysfunctional immune response in hyperinflammatory COVID-19 and are currently being investigated in clinical trials. Interleukin-1 (IL-1) blockage with anakinra has been shown to be safe and is associated with clinical improvement in patients with hyper-inflammatory COVID-19.3 Preliminary reports suggest that patients with an underlying malignancy have inferior outcomes from COVID-19.4, 5 Many haematology patients with COVID-19 will not be able to access novel immunomodulatory agents through clinical trials due to threshold laboratory values or recent use of other biologic agents. Therefore, off-label use of accessible therapeutic agents that have demonstrated benefit should be considered in haematology patients with concomitant COVID-19. In this report we demonstrate that anakinra is safe in haematology patients and resulted in a clinical improvement in three patients with acute leukaemia and confirmed or suspected COVID-19 pneumonia with a life-threatening hyper-inflammatory syndrome. A 40-year-old male patient with newly diagnosed acute myeloid leukaemia (AML) was commenced on induction chemoimmunotherapy (daunorubicin 50 mg/m2, cytarabine 100 mg/m2, gemtuzumab ozogamicin 3 mg/m2) (Table I). On day 12, after starting treatment he became pyrexial and empirical antibiotics and antifungal agents were started. High-resolution computed tomography (HRCT) scan of the chest demonstrated ground glass opacities in the right upper lobe (Fig 1A). A combined nose and throat swab for SARS-CoV-2 was negative but a diagnosis of presumed COVID-19 was made based on the typical appearances on imaging and the absence of another identifiable cause. Due to increasing oxygen requirements, the patient was transferred to the intensive care unit (ICU). A diagnosis of haemophagocytic lymphohistiocytosis (HLH) was considered after ferritin levels of 55 043 µg/l were noted along with persistent pyrexia, refractory thrombocytopenia (5 × 109/l), raised triglycerides (5·7 mmol/l) and a coagulopathy [International Normalized Ratio (INR) 2.4]. The H Score, which has a specificity of 86% for scores above 163, was calculated as 195. He was started on subcutaneous anakinra at a dose of 100 mg three times a day (TDS), dexamethasone and intravenous immunoglobulin (IVIg). The following day his oxygen requirements reduced and he defervesced. The anakinra and corticosteroids were weaned, the serum ferritin level fell, and the patient was discharged 35 days after commencing chemotherapy. A 31-year-old male, newly diagnosed with AML, was admitted for induction therapy (daunorubicin 50 mg/m2, cytarabine 100 mg/m2, gemtuzumab ozogamicin 3 mg/m2). On day 11 after starting chemotherapy he became pyrexial (38·4°C) and was therefore commenced on broad spectrum antibiotics. A repeat chest X-ray at this time showed features of bilateral airspace opacification. Combined nose and throat swabs were negative for a respiratory virus panel by polymerase chain reaction (PCR); SARS-CoV-2 was not detected when tested retrospectively. On day 22, rigours, tachycardia and desaturation prompted transfer to ICU for high flow oxygen at a FiO2 of 35%. On admission to ICU the serum ferritin was >100 000 µg/l, along with raised triglycerides (2·9 mmol/l), a coagulopathy (INR 1·66) and a pancytopenia. The H Score was calculated as 195 and he was started on subcutaneous anakinra 100 mg TDS, dexamethasone and IVIg. Along with a reduction in temperature, the ferritin reduced to 35 760 µg/l four days after starting anakinra and the oxygen requirements began decreasing after five days. After seven days in ICU he was discharged back to the ward, where anakinra and steroids were progressively reduced. HRCT prior to discharge showed bilateral ground glass changes with patchy distribution and small areas of peribronchial consolidation, consistent with COVID-19 (Fig 1B). A 36-year-old man with acute lymphoblastic leukaemia (ALL) presented with collapse and fever five days post completion of a second cycle of blinatumomab to eliminate minimal residual disease prior to allogeneic stem cell transplantation. Laboratory tests showed a lymphopenia (0·62 × 109/l) and a mild thrombocytopenia (121 × 109/l). Broad spectrum antibiotics were commenced. The patient desaturated two days later and combined nose and throat swabs sent for SARS-CoV-2 reverse transcription (RT)-PCR were found to be positive. A HRCT on day 11 showed widespread ground glass changes (Fig 1C). The ferritin rose to 8 961 µg/l along with a significantly rising C-reactive protein, increasing oxygen requirements and persistent pyrexia. He became progressively pancytopenic, with high triglycerides (3 mmol/l), a coagulopathy (INR 1·44) and an H Score calculated as 204. Anakinra was started at 200 mg intravenously twice a day. Initially, the ferritin continued to rise to a peak of 25 382 µg/l and the platelets continued to fall. Ten days after starting anakinra the patient defervesced and oxygen requirements were sustainably reduced. Anakinra was weaned and the clinical picture continued to improve on the ward before discharge 31 days after admission. We highlight that severe COVID-19 pneumonia can result in a life-threatening hyper-inflammatory syndrome in haematology patients post chemoimmunotherapy. In these patients, IL-1 blockade with anakinra was safe and resulted in clinical improvement. All three cases support the importance of screening for hyper-inflammatory states in patients with COVID-19 and acute leukaemia and support the use of immunomodulatory agents for patients with this phenotype. At this stage of the pandemic, results of large randomized trials are not available and evidence-based treatment protocols have yet to be established. Emerging evidence suggests that immunomodulatory agents such as anakinra, may improve outcomes in hyper-inflammatory COVID-19. We provide further evidence of the utility of this agent in the clinical context described and are the first to report its safe administration in patients with acute leukaemia affected by COVID-19. JWD wrote the paper along with TAF and BC. RH provided the radiological images. BC and PDK were the consultants in charge of the patients and PDK proofread the paper. None of the authors had relevant conflicts of interest.

Topics & Concepts