The novel coronavirus disease (COVID-19) is an infectious disease caused by severe acute respiratory syndrome coronavirus 2, an enveloped nonsegmented positive-sense RNA β-coronavirus. The virus can trigger a cascade of biological events that can lead to critical illness. It is urgent to gain an in-depth understanding of the critical activators that foreshadow patients at highest risk for mortality.1 Patients who have received tocilizumab for COVID-19–related cytokine storm have had improvements in C-reactive protein (CRP), lymphocytosis, and chest computed tomography findings.2 There exists a lack of understanding on which patients would benefit from tocilizumab administration. We aimed to characterize the group of patients who received tocilizumab and failed to improve, needed mechanical ventilation, or suffered mortality. We studied clinical and laboratory variables in patients receiving tocilizumab, comparing those who did not require mechanical ventilation with those needing mechanical ventilation including those who suffered mortality.
Adult patients with polymerase chain reaction–confirmed COVID-19 infection treated with tocilizumab at Spectrum Health System Hospitals from March 20, 2020, to May 18, 2020, were studied. Laboratory indices analyzed were metabolic panel, blood count, CRP, ferritin, fibrinogen, D-dimer, lactate dehydrogenase, and triglycerides. Our institutional protocol dictated that patients received tocilizumab when they required at least 4 L of oxygen or more respiratory support and had less than 14 days of symptoms. Tocilizumab dosing was as follows: 400 mg for 50 to 60 kg, 600 mg for 60 to 85 kg, and 800 mg for greater than 85 kg. Doses could be given every 8 hours and for a maximum of 3 doses. Day 0 was defined as the day tocilizumab was started. Cytokine panels were performed on day 0. Patients were followed for a maximum of 30 days until they met the end point of hospital discharge or mortality.
Quantitative data were expressed as mean plus or minus SD, whereas nominal data were expressed as percentages. The groups studied (intubated vs nonintubated or survivors vs deceased) were analyzed using the unpaired 2-tailed t test, whereas difference for nominal data was determined using the χ2 test or the Fisher's exact test. In addition, mixed-effects general linear modeling was performed. The independent variables were days following tocilizumab administration (days −1, 0, 1, and 3; reference value: day 0) and either intubation status or mortality.
For all analyses, significance was assessed at P < .05. All analyses were performed using Stata version 16.1 (StataCorp LLC, College Station, Texas).
Baseline characteristics revealed more type 2 diabetes in intubated patients and more smokers in the deceased group (Table 1 ). Neutrophil-to-lymphocyte ratio (NLR) increased from day negative 1 to 0 in intubated patients (P = .057) and decreased in the nonintubated group from 6.9 to 3 compared with 6.8 to 6.7 from day 0 to 3 in the intubated group. Similarly, in the survivors, there was a marked decrease in NLR from day 0 to day 3 from 7.1 to 3.3 (51.2% decrease), whereas in the nonsurvivors, there was an increase from 5 to 6.8 (33.6% increase). In all groups, mean absolute neutrophil count was highest on day 0. In nonsurvivors, day 0 absolute lymphocyte count was 34.1% lower (P = .046). No significant interactions were found in NLR from day negative 1 to day 0 in all groups.
Table 1.
Patient Characteristics Between Groups
| Characteristics | Intubation |
Mortality |
||||
|---|---|---|---|---|---|---|
| No (n = 34) | Yes (n = 32) | P value | No (n = 54) | Yes (n = 12) | P value | |
| Age, y (mean) | 56.1 ± 15.2 | 57.2 ± 11.0 | .745 | 55.2 ± 13.4 | 63.3 ± 10.6 | .053 |
| Sex | .569 | .512 | ||||
| Male | 20 (58.8) | 21 (65.6) | 32 (59.3) | 9 (75.0) | ||
| Female | 14 (41.2) | 11 (34.4) | 22 (40.7) | 3 (25.0) | ||
| Race or ethnicity | .414 | .463 | ||||
| White | 11 (32.4) | 13 (40.6) | 18 (33.3) | 6 (50.0) | ||
| African American | 4 (11.8) | 3 (9.4) | 5 (9.3) | 2 (16.7) | ||
| Hispanic | 14 (41.2) | 15 (46.9) | 25 (46.3) | 4 (33.3) | ||
| Asian | 5 (14.7) | 1 (3.1) | 6 (11.1) | 0 (0) | ||
| BMI (mean) | 29.7 ± 5.0 | 31.4 ± 7.2 | .259 | 30.1 ± 5.4 | 32.2 ± 9.2 | .304 |
| GFR >60 | 28/31 (90.3) | 26/31 (83.9) | .707 | 45/50 (90.0) | 9 (75.0) | .177 |
| Diabetic | 7 (20.6) | 18 (56.3) | .003 | 19 (35.2) | 6 (50.0) | .348 |
| Smoking history | 9 (26.5) | 15 (46.9) | .085 | 13 (24.1) | 11 (91.7) | <.001 |
| Hypertension | 15 (44.1) | 21 (65.6) | .079 | 26 (48.2) | 10 (83.3) | .051 |
| Tocilizumab dose, mg | 688 ± 112 | 688 ± 113 | .979 | 693 ± 108 | 667 ± 130 | .471 |
| Number of tocilizumab doses | .147 | .659 | ||||
| 1 | 1 (2.9) | 0 (0) | 1 (1.9) | 0 (0) | ||
| 2 | 5 (14.7) | 1 (3.1) | 6 (11.1) | 0 (0) | ||
| 3 | 28 (82.4) | 31 (96.9) | 47 (87.0) | 12 (100) | ||
| Additional therapies | ||||||
| Convalescent plasma | 11 (32.4) | 14 (43.8) | .340 | 20 (37) | 5 (41.7) | .754 |
| Remdesivir | 0 | 1 (3.1) | .485 | 0 | 1 (8.3) | .182 |
| Hydroxychloroquine | 20 (58.8) | 18 (56.3) | .833 | 33 (61) | 5 (41.7) | .333 |
| Corticosteroids | 5 (14.7) | 9 (28.1) | .183 | 11 (20.4) | 3 (25) | .708 |
| Infections | ||||||
| Positive sputum cultures | 11 (20.2) | 8 (66.7) | <.001 | |||
| Candida in sputum | 2 (3.7) | 5 (41.7) | <.001 | |||
Abbreviations: BMI, body mass index; GFR, glomerular filtration rate.
NOTE: Data are presented as number (percentage) unless indicated otherwise.
Only interleukin (IL)-10 and IL-6 were higher in intubated patients than in nonintubated patients (20.4 pg/mL vs 9.5, P = .001; 26.5 pg/mL vs 10.5, P = .002); there was no difference in serum soluble IL-2 receptor, IL-10, and IL-6 in patients who survived compared with deceased. A moderate correlation was observed between day 0 NLR and IL-6 (ρ = 0.307; P = .021) and between day 0 NLR and CRP (r = 0.326; P < .001). No correlation was observed between day 0 NLR and serum soluble IL-2 receptor, IL-10, or ferritin.
Studies have found an increased NLR to be an independent risk factor for severity in patients with COVID-19 infections. This is congruent with mounting evidence on importance of neutrophil response in severe COVID-19. Although nonspecific, existing cohorts have revealed onetime NLR greater than 3.13 and an interquartile NLR range of 5.82 to 15.1 to be predictive of critical illness and persistent elevation correlated with death.3 , 4 The mean ratio dropped between days 0 and 3 across all groups except in the patients suffering mortality. Furthermore, NLR correlated with IL-6 levels and may represent a proxy to monitor response to IL-6 blockade.
Beyond neutrophilia, neutrophil activation and degranulation are the most biologically active processes in severe COVID-19.5 Increased activated neutrophil extracellular traps are associated with severity of illness in patients with severe COVID-19.6 , 7 Neutrophil extracellular trap activation contributes to neutrophilia in COVID-19 simultaneously increasing IL-1β and IL-6, giving biologic plausibility to their respective blockade being therapeutic in severe COVID-19.6 , 7
The patients who seem to have a more favorable clinical course after tocilizumab are the same patients who had a drop in neutrophil levels. This is true for other biological conditions in which patients receiving tocilizumab have clinical improvement which correlated with a drop in neutrophil count during therapy.8 , 9 The elevated IL-6 and IL-10 and persistently elevated NLR in the deceased group indicate that further studies may be needed to determine whether the elevated levels suggest an advanced disease state resistant to tocilizumab therapy. We posit that immunophenotyping such differences in prospective trials will be imperative to individualize treatment for these patients.
Limitations of our cohort include its retrospective nature and smaller number of patients. Because all patients in the cohort received tocilizumab owing to the severe nature of their illness, we are unable to make a statement on clinical effectiveness of tocilizumab. We instead focused on biologic markers that could be indicative of disease severity and death. Our study period encompassed a time of varying hydroxychloroquine use, and dexamethasone was not widely used; however, there was no baseline difference between the groups.
This report suggests that persistently elevated NLR is associated with unfavorable outcomes, but it is unclear whether COVID-19 disease, secondary infections, or other causes are driving elevated NLR in patients with more severe disease. We suggest that an elevated NLR is one potential marker. Further prospective studies are needed to reveal the performance of NLR vs in combination with other markers (CRP, complement components, cytokines, etc). NLR should be considered as a marker for COVID-19 that is resistant to anti-inflammatory therapy, with a need for infection monitoring and aggressive life support measures.
Acknowledgment
We acknowledge Dr Amanda Holsworth for her careful editing, input, and feedback during the preparation of the article.
Footnotes
Disclosures: Dr Hartog reports serving in the advisory boards of Genentech, Inc, Orchard Therapeutics, and Pharming Healthcare, Inc; serving in the advisory board and speakers bureau of Horizon Pharmaceuticals, Inc; and serving in the speakers bureau of Takeda Pharmaceutical Company Limited, which are all unrelated to the current topic. Dr Rajasekaran reports to have unrelated National Institute of General Medical Sciences–funded parent R01 (1R01GM134307-01) on pediatric multiorgan dysfunction. The remaining authors have no conflicts of interest to report.
Funding: The authors have no funding sources to report.
References
- 1.Zhou F., Yu T., Du R. Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study. Lancet. 2020;395(10229):1054–1062. doi: 10.1016/S0140-6736(20)30566-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Mehta P., McAuley D.F., Brown M. COVID-19: consider cytokine storm syndromes and immunosuppression. Lancet. 2020;395(10229):1033–1034. doi: 10.1016/S0140-6736(20)30628-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Hwang S.Y., Shin T.G., Jo I.J. Neutrophil-to-lymphocyte ratio as a prognostic marker in critically ill septic patients. Am J Emerg Med. 2017;35(2):234–239. doi: 10.1016/j.ajem.2016.10.055. [DOI] [PubMed] [Google Scholar]
- 4.Liu J., Liu Y., Xiang P. Neutrophil-to-lymphocyte ratio predicts critical illness patients with 2019 coronavirus disease in the early stage. J Transl Med. 2020;18(1):206. doi: 10.1186/s12967-020-02374-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Morrissey S.M., Geller A.E., Hu X. Emergence of low-density inflammatory neutrophils correlates with hypercoagulable state and disease severity in COVID-19 patients. https://www.medrxiv.org/content/10.1101/2020.05.22.20106724v1
- 6.Middleton E.A., He X.Y., Denorme F. Neutrophil extracellular traps (NETs) contribute to immunothrombosis in COVID-19 acute respiratory distress syndrome. Blood. 2020;136(10):1169–1179. doi: 10.1182/blood.2020007008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Zuo Y., Yalavarthi S., Shi H. Neutrophil extracellular traps in COVID-19. JCI Insight. 2020;5(11) doi: 10.1172/jci.insight.138999. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Ghang B., Kwon O., Hong S., Lee C.K., Yoo B., Kim Y.G. Neutrophil-to-lymphocyte ratio is a reliable marker of treatment response in rheumatoid arthritis patients during tocilizumab therapy. Mod Rheumatol. 2017;27(3):405–410. doi: 10.1080/14397595.2016.1214340. [DOI] [PubMed] [Google Scholar]
- 9.Moots R.J., Sebba A., Rigby W. Effect of tocilizumab on neutrophils in adult patients with rheumatoid arthritis: pooled analysis of data from phase 3 and 4 clinical trials. Rheumatology (Oxford) 2017;56(4):541–549. doi: 10.1093/rheumatology/kew370. [DOI] [PMC free article] [PubMed] [Google Scholar]