Combination therapy of Tocilizumab and steroid for management of COVID-19 associated cytokine release syndrome

Abstract Cytokine release syndrome (CRS) or cytokine storm is thought to be the cause of inflammatory lung damage, worsening pneumonia and death in patients with COVID-19. Steroids (Methylprednislone or Dexamethasone) and Tocilizumab (TCZ), an interleukin-6 receptor antagonist, are approved for treatment of CRS in India. The aim of this study was to evaluate the efficacy and safety of combination therapy of TCZ and steroid in COVID-19 associated CRS. This retrospective cohort study was conducted at Noble hospital and Research Centre (NHRC), Pune, India between April 2 and November 2, 2020. All patients administered TCZ and steroids during this period were included. The primary endpoint was incidence of all cause mortality. Secondary outcomes studied were need for mechanical ventilation and incidence of systemic and infectious complications. Baseline and time dependent risk factors significantly associated with death were identified by Relative risk estimation. Out of 2831 admitted patients, 515 (24.3% females) were administered TCZ and steroids. There were 135 deaths (26.2%), while 380 patients (73.8%) had clinical improvement. Mechanical ventilation was required in 242 (47%) patients. Of these, 44.2% (107/242) recovered and were weaned off the ventilator. Thirty seven percent patients were managed in wards and did not need intensive care unit (ICU) admission. Infectious complications like hospital acquired pneumonia, blood stream bacterial and fungal infections were observed in 2.13%, 2.13% and 0.06% patients respectively. Age ≥ 60 years (P = .014), presence of co-morbidities like hypertension (P = .011), IL-6 ≥ 100 pg/ml (P = .002), D-dimer ≥ 1000 ng/ml (P < .0001), CT severity index ≥ 18 (P < .0001) and systemic complications like lung fibrosis (P = .019), cardiac arrhythmia (P < .0001), hypotension (P < .0001) and encephalopathy (P < .0001) were associated with increased risk of death. Combination therapy of TCZ and steroids is likely to be safe and effective in management of COVID-19 associated cytokine release syndrome. Efficacy of this anti-inflammatory combination therapy needs to be validated in randomized controlled trials.


Introduction
In December 2019, Wuhan city, the capital of Hubei province in China, became the centre of an outbreak of viral pneumonia caused by a novel beta Coronavirus. It was named severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) due to its sequence homology with SARS-COV-1. [1] The disease caused by SARS-COV-2 was later designated Coronavirus disease 2019  in February 2020, by World health organization (WHO). [2] COVID-19 spread rapidly worldwide and India was no exception. By November 2, 2020 there have been more than 9 million infections and 0.14 million deaths due to COVID-19 in India. [3] Severe and critical disease develops in approximately 15% and 5% of COVID-19 patients. In these patients, pneumonia leads to acute respiratory distress syndrome (ARDS) that could need invasive mechanical ventilation. [1,[4][5][6][7] Critically ill COVID-19 patients have a mortality rate ranging from 35 to 62%. [8][9][10] 1.1. Role of host inflammatory response in severe COVID-19 COVID-19 is characterized by 2 phases; viral replication phase and the host inflammatory response phase. [11] Host inflammatory response phase is usually seen 7 days after symptom onset. SARS-CoV-2 replicates within the pulmonary tissue, activates innate immune response, leading to production of cytokines (Interleukin-1 beta (IL-1B), Interleukin-6 (IL-6) and Tumor necrosis factor (TNF)) by alveolar macrophages and recruitment of adaptive immunity cells. The transition between innate and adaptive immune responses is critical for the clinical trajectory of SARS-CoV-2 infection. [11][12][13][14][15] Adaptive immune response controlled by immune regulatory cells can be a protective immune response or a dysregulated and exacerbated inflammatory response. [12] Amongst severe and critically ill COVID-19 patients, there is a dysregulated pulmonary and systemic immune response. This dysregulated immune response also known as cytokine release syndrome (CRS) or cytokine storm is characterized by elevation in systemic inflammatory markers (C-reactive protein (CRP), ferritin, lactate dehydrogenase (LDH) and Ddimer) and aberrant pro-inflammatory cytokine secretion (IL-6, soluble IL-2 receptor [IL-2R], IL-10, TNF-a) by pulmonary alveolar macrophages. It is also accompanied by depleted adaptive immune response in the form of lymphopenia (decline in CD4+ T cell, CD8+ T cell, Natural killer cell but not in B cell subset) and decreased Interferon gamma (IFN-g) expression in CD4+ T cells (CD4+ T cell dysfunction). [12][13][14][15] Cytokine storm leads to destruction of alveolar epithelial cells, increased pulmonary vascular permeability, worsening pneumonia, increased risk of thrombosis and progression to acute respiratory distress syndrome (ARDS). [15] Rising levels of interleukin-6 (IL- 6) in severe COVID-19 have been associated with increased likelihood of ARDS, mechanical ventilation and mortality. [16][17][18][19] 1.2. Role of immunomodulatory drugs in tackling cytokine release syndrome Steroids, namely Dexamethasone and Methylprednisolone have been extensively used to resolve hyperinflammation and inflammatory lung damage in COVID-19. [20][21][22][23] After the publication of RECOVERY trial, Dexamethasone was approved by World health organization (WHO) as an immunomodulatory drug for use in hospitalized COVID-19 patients who require oxygen. The benefit of Dexamethasone was greatest for patients who were receiving invasive mechanical ventilation. [21] In patients with severe COVID-19, Methylprednisolone therapy was associated with reduced rate of ICU transfer, reduced rate of progression to mechanical ventilation and higher probability of extubation in patients already on mechanical ventilation. [22,23] However, steroids alone might not be able to tackle cytokine storm in all patients. Other Immunomodulatory drugs such as selective cytokine inhibitors could be of incremental benefit to suppress the hyperinflammation if used in combination with steroids.
Tocilizumab (TCZ) is a recombinant humanized monoclonal interleukin-6 receptor (IL-6R) antibody of the IgG1 subtype. TCZ specifically binds and inhibits soluble and membrane-bound IL-6 receptors (sIL-6R and mIL-6R) and terminates downstream intracellular signal transduction. [24,25] It has been used for the treatment of rheumatoid arthritis, [26] systemic juvenile idiopathic arthritis, [27] Castleman disease, [28] Crohn's disease [29] and cytokine release syndrome (CRS) caused by chimeric antigen receptor T-cell (CAR-T) immunotherapy. [30] Indian council of Medical research (ICMR) guidelines published by Government of India for guidance of physicians, have included TCZ as investigational therapy for off label use in patients with moderate disease with progressively increasing oxygen requirements and in mechanically ventilated patients not improving despite use of steroids. [31] This recommendation stems from multiple retrospective and small prospective studies,  that have suggested strong clinical benefits due to use of TCZ in CRS in form of reduced risk of invasive mechanical ventilation or death in patients. However data from randomized controlled trials (RCT) has shown disappointing results with evidence, if at all, of modest efficacy. [59][60][61][62][63] RCT's have shown that TCZ reduces need for ICU admission and noninvasive and invasive ventilation in patients with severe COVID-19. [61,62] However, none of them have reported mortality benefit. Out of the 5 RCT's, only the EMPACTA trial [62] used concomitant TCZ and steroids in management of CRS. As a result, more evidence regarding positioning of TCZ as an immunomodulatory therapy in Dravid et al. Medicine (2021)  COVID-19 needs to be published. Data from resource limited settings like India regarding use of TCZ and steroids in management of CRS has also been scarce. [64,65]

Aims and objectives
The aim of this single center retrospective cohort study conducted in Pune, India was to estimate the efficacy and safety of combination therapy of TCZ and steroid in management of COVID-19 associated CRS.

Study setting
This study was conducted at Noble hospital and Research Centre (NHRC), Pune, Western India. NHRC is a tertiary level private hospital designated for clinical management of COVID-19 patients in Pune since March 23, 2020. Pune is located in the state of Maharashtra, Western India and was one of the epicenters of COVID-19 epidemic in India. As of November 2, 2020, Maharashtra state has reported more than 1.71 million cases of COVID-19 and more than 45,000 deaths. [3] Till November 2, 2020, NHRC has admitted 2831 COVID-19 patients. NHRC provides clinical care, diagnostic and treatment services to COVID-19 patients at a subsidized cost. Patients are referred from primary care physicians, private practitioners of alternative medical systems, primary level COVID care centers (both government run and private owned) and other tertiary level hospitals. Data of all hospitalized COVID-19 patients is entered into an electronic database (Lifeline electronic database, Manorama infosystems, Kolhapur, India). Data was obtained from electronic health record of each individual by manual abstraction. It included hospitalization dates, demographics, comorbidities, clinical examination data, laboratory data (including inflammatory markers), microbiology reports and imaging reports (X-ray chest and high-resolution computerized tomography scan (HRCT chest)).

Study population
Patients were eligible for inclusion in this analysis if they were admitted to NHRC between April 2, 2020 and November 2, 2020 and were administered TCZ. Criteria for prescribing TCZ were developed by the Department of Infectious Diseases and Department of Critical care medicine. Patients were administered TCZ if they satisfied following inclusion criteria for hyperinflammation or CRS: (1) Reverse-transcriptase polymerase chain reaction (RT-PCR) test positive for SARS-CoV-2 RNA or positive COVID antibody test. (2) Lung Imaging: Moderate or severe pneumonia on High resolution Computerized tomography scan (HRCT) of chest (CT severity index > = 8 [66] ) or X-ray chest showing evidence of pneumonia. (3) Day 7 to 14 since onset of symptoms. (4) Rapidly worsening respiratory status despite use of steroids and antiviral drugs: Hypoxia (room air oxygen saturation (SPO2) < 90 mm Hg) and tachypnea (respiratory rate > 30 per minute) at rest or after minimal exertion which requires supplemental oxygen. OR (5) Rapidly worsening respiratory status despite use of steroids and antiviral drugs: Requirement of noninvasive or invasive ventilation to mitigate hypoxia and tachypnea. OR (6) PaO2/FiO2 ratio (ratio of arterial oxygen partial pressure (PaO2 in mmHg) to fractional inspired oxygen (FiO2) of less than 300 on room air. (7) Elevated inflammatory markers: IL-6 (> 100 pg/ml or 5-fold increase from prior level) or one out of D-dimer (> 1000 ng/ ml), Ferritin (> 1000 ng/ml) and CRP (> 10 mg/ml) being elevated.
Exclusion criteria: Following patients were not prescribed TCZ during hospital admission or were excluded from the analysis.
(1) Pregnant females. Tocilizumab was given intravenously at 8 mg/kg bodyweight (up to a maximum of 800 mg in two infusions, 24 h apart). Additional doses were given at the discretion of infectious disease and critical care physician if it was thought that additional dosing would be needed for morbidly obese patients (body weight > 100 kg) or to reduce persistent hyperinflammation. Patients or their immediate family members signed an informed consent form prior to TCZ administration. The language in the consent form was non-prescriptive, saying that TCZ can be used off-label in patients with COVID-19 induced hyperinflammation as per ICMR guidelines. The consent form cautions that the evidence for benefit is modest; a risk of infectious complications exists but in view of limited treatment options in patients with severe pneumonia, therapy can be considered. The cost of TCZ (approximately 500 dollars for 1 vial of 400 mg) was borne by the patient as an out of pocket expense or by third party reimbursement via medical health insurance.

Data collection
For all patients administered TCZ, we scrutinized inpatient case files until hospital discharge, death, or December 2, 2020-the date on which the database was locked-whichever happened first. Number of TCZ doses given, type of steroid and dose of steroid given and other concomitant medication prescribed to patient were noted. Time to TCZ administration since onset of symptoms and time to TCZ therapy since hospital admission was calculated. Oxygenation and ventilation outcomes in patients were noted. Systemic complications (including infectious complications) developing in patients during hospital admission were also noted. All patients who died during hospital admission were identified and a death audit to look for complications and cause of death was undertaken. Ordinal scale for COVID-19 severity was noted for all patients at hospital admission, during hospital stay and at discharge or death. The 8 level ordinal scale is as follows: 1 = ambulatory and no restriction of activities; 2 = ambulatory and restriction of activities due to use of home oxygen therapy or complications; 3 = Hospitalized but no oxygen therapy; 4 = Hospitalized with oxygen therapy by nasal prongs; 5 = Hospitalized with oxygen therapy by Non re-breathing mask; 6 = Hospitalized with severe disease and on high flow nasal oxygen (HFNO) or Noninvasive ventilation; 7 = Hospitalized with severe disease and on invasive mechanical ventilation; 8 = Death. All patients who recovered, got discharged and had outpatient follow-up visit at 15 and 30 days after discharge were identified. Their outpatient follow-up visits were traced from electronic database to look for delayed complications. There was no control group in our study as all patients with suspicion of hyperinflammation or CRS ended up getting TCZ and steroid.

Concomitant medications given for COVID-19 therapy
NHRC follows the ICMR guidelines published by Government of India for COVID-19 management. [31] These guidelines get updated from time to time. All hypoxic COVID-19 patients who received TCZ were already receiving antiviral agents, intravenous steroids (Dexamethasone 6 mg per day or Methylprednisolone 40 mg BD) and systemic anticoagulation (Low molecular weight heparin (Enoxaparin) or Unfractionated Heparin) as a standard of care. Intravenous steroids were continued for a maximum of 10 days followed by shift to oral Prednisolone in tapering doses over next 10 days. Hydroxychloroquine and Lopinavir/ritonavir were initially recommended as standard antiviral therapy at NHRC. Once clinical trial data about lack of efficacy and toxicity was published, their use as antivirals was discontinued. [67][68][69] Remdesvir was used as antiviral of choice at our institute since July 26, 2020 and was administered to all patients presenting with moderate or severe pneumonia. [70] Convalescent plasma therapy (CPT) was also used in a subset of patients as an antiviral agent. Adjunctive antibiotic and antifungal therapy in patients to prevent bacterial and fungal super-infections was decided by the infectious disease physician. After administration of TCZ, anti-fibrotic agents like Pirfenidone (daily dosage ranging from 800 to 2000 mg per day) and Nintedanib (daily dosage of 150 mg twice a day) were added to the treatment protocol for patients suspected of developing lung fibrosis.
3.5. Outcomes 3.5.1. Primary endpoint. Deaths in the cohort after TCZ and steroid administration.
3.5.2. Secondary outcomes. (A) Number of patients who received noninvasive or invasive ventilation: Patients who required mechanical ventilation (noninvasive or invasive) in our cohort were identified. NHRC adopted the delayed intubation and delayed invasive mechanical ventilation (IMV) policy for patients with COVID-19 associated ARDS. Patients not maintaining arterial oxygen saturation (SPO2) > 90% on supplemental oxygen (Non re-breathing mask at 15 liters/minute) were initially started on noninvasive ventilation (NIV: High flow nasal oxygen (HFNO) or Bi-level positive airway pressure (BIPAP)). The indications for invasive mechanical ventilation (IMV) in our hospital was respiratory failure on NIV, defined by any of the following criteria: 1. respiratory rate of 40 or more breaths per minute; 2. respiratory distress with activation of accessory respiratory muscles; 3. the need for fractional inspired oxygen (FiO2) of 100% to maintain an oxygen saturation (SPO2) level of 90%, or a PaO2/FiO2 ratio of less than 100; 4. Neurological deterioration (altered consciousness with a Glasgow Coma Scale score of 10 or lower) on treatment.
(B) Number of patients who could be weaned off NIV or IMV and discharged from hospital.
(C) Improvement in 8 point ordinal scale reflecting declining disease severity.
(D) Incidence of infectious complications (bacterial or fungal super-infections): Patients developing fever (Temperature > 38.3 degrees Celsius), productive cough, leucocytosis (WBC count > 15,000 cells/mm 3 with predominance of neutrophils and shift to left) or increased serum procalcitonin despite adequate antibiotic and antifungal therapy were investigated for super-infections. All positive blood and respiratory cultures (bacterial and fungal culture) were assessed by an Infectious Diseases physician to decide infection versus colonization. Infections were included if they occurred > 48 h after TCZ administration.
The use of database for clinical research was approved by the institutional review board (IRB) of Noble hospital and Research Centre, Pune, India.

Statistical methods
Continuous variables were summarized using median and interquartile range (IQR), while categorical variables were summarized using frequency and percentages. Continuous variables were compared using a Mann-Whitney U test. Categorical variables were compared using Chi-square test, Proportion test and Fishers' exact test. Baseline and time dependent risk factors significantly associated with death were identified by Relative risk estimation. Baseline risk factors included were age (< 60 years or ≥ 60 years), gender, co-morbidities like diabetes, hypertension, ischemic heart disease and chronic kidney disease and baseline investigations like IL-6 (≥ 100 versus < 100 pg/ml), absolute lymphocyte count (< 1000 versus ≥ 1000 cells/mm 3 ), D-dimer (≥ 1000 versus < 1000 ng/ml) and CT severity index (≥ 18 versus < 18). Time dependent risk factors included systemic complications like lung fibrosis, arrhythmia, hypotension, new onset or worsening thrombocytopenia, hepatitis, acute kidney injury and encephalopathy. The P value .05 was considered as statistically significant. All data was analyzed by SPSS version 12.0.

Baseline demographic data
Of the total 2831 COVID-19 patients admitted in NHRC, 522 were administered TCZ. Seven patients were given TCZ therapy prior to transfer to NHRC and hence were excluded from the analysis (Number of patients included in final analysis -515). Baseline demographic data, preexisting co-morbidities, presenting symptoms in patients at admission and investigations performed prior to TCZ therapy are enumerated in Tables (Table 3).

Systemic and Infectious complications
Systemic complications (fatal and nonfatal) observed in patients in our cohort are enumerated in Table 4. Septic shock (persistent hypotension requiring vasopressors to maintain systolic blood pressure > 100 mm Hg and a serum lactate level greater than 2 mmol/L) was noted in 58 ( 18.837)) were also associated with increased risk of death ( Figure 1, Table 5).

Delayed complications
Out of the 380 patients who recovered and were discharged, 72 (18.9%) needed short term home oxygen therapy after hospital discharge. Of these, two patients reported worsening of respiratory symptoms leading to respiratory failure and death, while 70 patients could be weaned off short term oxygen therapy. Four patients developed late infectious complications (multidermatomal herpes zoster, herpes zoster opthalmicus, acute bacterial cholecystitis and Escherichia coli (E. coli) bacteremia and septic shock) within 1 month of discharge from hospital. Three patients recovered after treatment, but the patient having E. coli bacteremia progressed to septic shock, ARDS and died. One patient developed midbrain encephalitis, 27 days after discharge from hospital and died (Total number of deaths: 139).

Discussion
The goal of this retrospective observational cohort study conducted at a tertiary level, private hospital in Pune, India was to assess efficacy and safety of combination therapy of TCZ and steroids in tackling CRS developing in patients with severe COVID-19. Our cohort consisted of a relatively elderly     [64] strict inclusion and exclusion criteria applied while administering TCZ, [64,65] availability of data about ventilatory outcomes, [64,65] availability of data about systemic complications (including infectious complications) [64,65] and lower mortality rate. [65] In our cohort, 37% patients could be managed in isolation wards without need for intensive care. In an ideal scenario, all patients were candidates for ICU care, but overburdened healthcare system and shortage of ICU beds and ventilators meant that they were managed in wards. In addition, 53% patients required only supplemental oxygen before recovery and did not progress to mechanical ventilation. As per the CORIMUNO-19 RCT, use of TCZ among patients on supplemental oxygen reduced the need for intensive care, noninvasive or invasive mechanical ventilation by almost 40%. The effect of TCZ was numerically higher if combined with steroids. [61] According to the EMPACTA RCT [62] and the TESEO cohort, [32] the likelihood of progression to mechanical ventilation was significantly lower among patients who received TCZ plus standard care than among those who received placebo plus standard care. Reduction in need of ICU care can reduce the risk of long-term complications including death and improve health-related quality of life. Preventing progression to mechanical ventilation greatly alters patient outcomes and leads to better utilization of scarce healthcare resources.
Mortality rate in our cohort was 26%, while mortality rate amongst patients requiring ICU was 40%. Overall mortality rate was similar to that seen in multiple retrospective cohort studies. [35,36,54,57,64] Mortality rate amongst patients requiring mechanical ventilation (NIV/IMV) was 56%. This was higher as compared to other cohort studies. [40] It could be related to delayed intubation and delayed invasive mechanical ventilation policy followed at our institute. Out of 130 patients requiring invasive mechanical ventilation, only 3 could be successfully extubated. On relative risk estimation, age ≥ 60 years, presence of multiple comorbidities, increased baseline inflammatory markers (IL-6 and Ddimer), high CT severity index and development of systemic complications like lung fibrosis, arrhythmia, encephalopathy, thrombocytopenia and hypotension were significant risk factors associated with death. Females had a higher mortality rate than males in our cohort (30.4% versus 24.9%, Table 5). Early evidence indicates that males have higher overall burden, but females have a higher relative-risk of COVID-19 mortality in India. [73] Marked sex differences in access to health services, with women being less likely to be admitted to hospital than men might result in more severe cases of COVID-19 among women than men in hospital settings and higher mortality. [74] TCZ and steroid combination therapy was safe and well tolerated. Infectious complications like confirmed bacterial and fungal infections (including HAP/VAP and blood stream infections) were seen in 2% patients in the cohort. Low incidence of infectious complications could be because of adequate, prophylactic antibiotic and antifungal therapy prescribed to patients. Meropenem, Teicoplanin and Fluconazole were the commonest antibiotic and antifungal agents prescribed to patients. Transaminitis or hepatitis, which is an adverse event associated with TCZ, was seen in approximately 10% of patients. However, in view of prescription of multiple drugs like Remdesvir, Fluconazole, Doxycycline, Pirfenidone and Nintedanib along with TCZ, it was difficult to identify the exact causative agent of drug induced liver injury. Pulmonary (progression of lung fibrosis) and infectious complications were noted even after discharge from hospital. As a result, close follow up of patients for a period of 3-months after hospital discharge is essential for immediate identification of delayed complications.

Limitations
Our study has several limitations. First, it is not a randomized controlled trial, and therefore unmeasured confounding cannot be ruled out. Second, as for all retrospective studies, some individuals administered TCZ and steroids may be unreported leading to measurement bias and overestimation of safety and efficacy of the combination therapy. Third, a comparator arm was not possible in this pandemic setting. Considering the unavailability of observational or RCT data about efficacy of other regimens to tackle hyperinflammation, worsening of respiratory parameters seen in patients despite receiving potent steroids and life threatening nature of the disease characterized by sudden worsening and rapid progression to respiratory failure over few hours, a comparative arm could not be justified. Fourth, concomitant therapies like antiviral drugs, convalescent plasma, anti-fibrotic agents, supplemental oxygen and ventilation strategies can help in reducing disease severity and clinical improvement. The authors acknowledge the fact that individual contribution of each drug is difficult to estimate. However, considering the predefined criteria for introduction of each drug in patient management, strict inclusion and exclusion criteria applied while administering TCZ and steroids, large number of patients enrolled and improvement in symptoms and oxygenation after TCZ administration, the authors are reasonably confident that this study reliably captures the efficacy of this combination. Fifth, an overwhelmed health care system, inadequate workforce and lack of exhaustive reporting could be responsible for underestimation of co-morbidities, presenting symptoms and complications amongst patients in our cohort. Sixth, Body mass index (BMI) could not be calculated for patients who were bed ridden or those requiring mechanical ventilation. Seventh, CT severity index and PaO2/FiO2 ratio was not available for all patients in our cohort. Eighth, Sequential organ failure assessment (SOFA) score was not performed in patients admitted in ICU. [75] Ninth, ventilator parameters like positive end expiratory pressure (PEEP) and plateau pressure were not available for all patients started on NIV or IMV. Tenth, patients were followed up after discharge from hospital for 1 month. As a result, long term complications due to immunomodulatory therapy could not be identified. Despite these limitations, this retrospective cohort study from Western India adds to the growing body of literature on use of TCZ and steroids as an anti-inflammatory combination therapy in tackling cytokine release syndrome and resultant ARDS in COVID-19.

Conclusions
Combination therapy of TCZ and steroids is likely to be a safe and effective treatment modality in management of COVID-19 associated cytokine release syndrome. Efficacy of this antiinflammatory combination therapy needs to be validated in large randomized controlled clinical trials.