Cytokine profile in plasma of severe COVID-19 does not differ from ARDS and sepsis

BACKGROUND Elevated levels of inflammatory cytokines have been associated with poor outcomes among COVID-19 patients. It is unknown, however, how these levels compare with those observed in critically ill patients with acute respiratory distress syndrome (ARDS) or sepsis due to other causes. METHODS We used a Luminex assay to determine expression of 76 cytokines from plasma of hospitalized COVID-19 patients and banked plasma samples from ARDS and sepsis patients. Our analysis focused on detecting statistical differences in levels of 6 cytokines associated with cytokine storm (IL-1β, IL-1RA, IL-6, IL-8, IL-18, and TNF-α) between patients with moderate COVID-19, severe COVID-19, and ARDS or sepsis. RESULTS Fifteen hospitalized COVID-19 patients, 9 of whom were critically ill, were compared with critically ill patients with ARDS (n = 12) or sepsis (n = 16). There were no statistically significant differences in baseline levels of IL-1β, IL-1RA, IL-6, IL-8, IL-18, and TNF-α between patients with COVID-19 and critically ill controls with ARDS or sepsis. CONCLUSION Levels of inflammatory cytokines were not higher in severe COVID-19 patients than in moderate COVID-19 or critically ill patients with ARDS or sepsis in this small cohort. Broad use of immunosuppressive therapies in ARDS has failed in numerous Phase 3 studies; use of these therapies in unselected patients with COVID-19 may be unwarranted. FUNDING Funding was received from NHLBI K23 HL125663 (AJR); The Bill and Melinda Gates Foundation OPP1113682 (AJR and CAB); Burroughs Wellcome Fund Investigators in the Pathogenesis of Infectious Diseases #1016687 NIH/NIAID U19AI057229-16; Stanford Maternal Child Health Research Institute; and Chan Zuckerberg Biohub (CAB).


INTRODUCTION
Numerous reports have described an association between elevated inflammatory markers and poor outcomes in COVID-19 patients. [1][2][3] These data have sparked interest in "cytokine storm" as a major driver of illness severity in COVID-19, and multiple clinical trials are underway to test the efficacy of immunosuppressive therapies, including IL-6 antagonists. However, it is unclear if inflammatory cytokine levels are truly higher in patients with severe COVID-19 than in critically ill patients with ARDS or sepsis due to other causes.
All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Levels of IL-1b, IL-1RA, IL-6, IL-8, IL-18 and TNFα did not significantly differ between the moderate COVID-19, severe COVID-19, and ARDS/sepsis group (Figure 1, after adjustment for multiple comparisons, all p values > 0.05). There was a trend towards higher levels of IL-1RA and IL-6 in the patients with severe COVID-19 as compared to those with moderate COVID-19, consistent with prior reports. [1][2][3] There was also a trend towards higher IL-18 in the severe COVID-19 group compared to the other critical illness group (p-unadjusted = 0.01, p-adjusted = 0.10).
In the more extended exploratory analysis of 70 additional cytokines, the three patient groups did not differ strongly in principal component analysis (Supplemental Figure S1, available in the online supplement). There were small but statistically significant differences in the levels of IL-16, IL-21, IL-28A (IFNL2), and TSLP between the three groups (Supplemental Figure S2), but no cytokine levels were dramatically increased in COVID-19 compared to other causes of critical illness (Supplemental Figures S2 and   S3). Together, these data suggest that a "cytokine storm" in COVID-19 that is distinct from other critical illness (e.g. sepsis and ARDS) is unlikely.
All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Our primary goal was to directly compare levels of six inflammatory cytokines commonly associated with cytokine storm (IL-1b, IL-1RA, IL-6, IL-8, IL-18, and TNFα) between severe COVID-19 patients and patients with ARDS or sepsis due to other causes. Our findings are consistent with previous data demonstrating higher levels of inflammatory cytokines among COVID-19 patients with more severe disease. Importantly, however, our data suggest that inflammatory biomarkers in severe COVID-19 patients are not markedly elevated when directly compared to critically ill patients with ARDS or sepsis.
IL-6 levels were measured by the clinical lab (as part of clinical care) in 6 of the COVID-19 patients, including 4 with severe disease, though not at matched time points with the research blood collection. Levels ranged from <6-31 pg/mL. Given the small number of patients who had clinical measurements available, and the variation in collection times, we were unable to derive accurate concentrations based on MFI for the remaining patients. Nonetheless, these data points further support our findings that IL-6 levels in particular-while elevated above levels found in healthy subjects-are not markedly elevated in all severe COVID-19 patients compared to other critically ill patients.
This brief report calls into question the idea that "cytokine storm" is the major driver of morbidity and mortality in all severe COVID-19 patients. As Ritchie and Singanayagam have stated, it is equally possible that the higher levels of proinflammatory cytokines seen in severe COVID-19 reflect an increased burden of virus rather than "an inappropriate host response that requires correction." 9 All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. The most important limitation of this study is our small sample size. Even in our predetermined analysis of six cytokines strongly associated with cytokine storm, we lack power to detect minor differences between groups. The exploratory analyses of an additional 70 cytokines is similarly limited, but is provided as a reference for the field. In addition, we do not have measurements of cytokines over time, but only near the point of enrollment. Finally, as discussed above, we report cytokine levels by MFI per recommended Luminex analysis methods, 10 precluding direct comparison of our values to previously published data that report cytokine levels in pg/mL. (which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
Twenty-eight critically ill subjects with ARDS and sepsis who had been enrolled in the Stanford ICU biobank between 2015-2018 were selected for comparison. Briefly, the Stanford ICU Biobank recruits patients at risk for development of respiratory failure and ARDS admitted to Stanford Hospital as previously described. 11 Subjects are eligible for enrollment when decision to admit to ICU is made, either from the Emergency Department or the hospital wards, with goal enrollment in <24h of ICU transfer. All 28 patients were phenotyped for ARDS and sepsis by 2-physician consensus (AJR and JEL), based on the Berlin Criteria and Sepsis-2 criteria and using all available hospital clinical data including history, physical exam, laboratory and microbiologic data, invasive monitoring data, autopsy results, and physician summaries. The Stanford ICU is a major referral center for cancer and thus typically has high rates of immunosuppression. To assess cytokine response to infection in patients with a normal immune system (similar to the COVID population), the ARDS and sepsis patients were therefore enriched for normal baseline immune system (e.g. no metastatic cancer, bone marrow transplant, or high dose steroids) in comparison to the Biobank as a whole.
All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint this version posted May 16, 2020.

Statistics
All statistical analyses were performed using the open source statistical software R (https://www.r-project.org). Because we observed significant differences in CHEX4 MFI between the three groups tested, we corrected for CHEX4 nonspecific binding, keeping the clinical conditions (severe COVID-19 (COV-ICU), moderate COVID-19 (COV-noICU), ARDS and Sepsis), age, and sex as covariates. Each sample was normalized according to methods used by the Human Immune Monitoring Center at Stanford University. 12 Briefly, the median fluorescence intensity (MFI) of each cytokine was corrected first for plate/batch/lot artifacts by linear mixed modeling, then the average of technical replicates was log transformed. Then the log transformed average MFIs were corrected for nonspecific binding by local polynomial regression and repeated cross-validation, resulting in plate-detrended MFI (dpMFI) values. All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint this version posted May 16, 2020.

Study approval
This study was conducted according to Declaration of Helsinki principles, and was approved by the Stanford University Hospital IRB (protocol 28205). All patients or their surrogates gave written informed consent to participate in the Stanford ICU Biobank.
All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.
The copyright holder for this preprint this version posted May 16, 2020. All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.  All rights reserved. No reuse allowed without permission.
(which was not certified by peer review) is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.