Utility of laboratory and immune biomarkers in predicting disease progression and mortality among patients with moderate to severe COVID-19 disease at a Philippine tertiary hospital

Abstract

Purpose: This study was performed to determine the clinical biomarkers and cytokines that may be associated with disease progression and in-hospital mortality in a cohort of hospitalized patients with RT-PCR confirmed moderate to severe COVID-19 infection from October 2020 to September 2021, during the first wave of COVID-19 pandemic before the advent of vaccination.

Patients and methods: Clinical profile was obtained from the medical records. Laboratory parameters (complete blood count [CBC], albumin, LDH, CRP, ferritin, D-dimer, and procalcitonin) and serum concentrations of cytokines (IL-1β, IL-2, IL-4, IL-6, IL-8, IL-10, IL-18, IFN-γ, IP-10, TNF-α) were measured on Days 0-3, 4-10, 11-14 and beyond Day 14 from the onset of illness. Regression analysis was done to determine the association of the clinical laboratory biomarkers and cytokines with the primary outcomes of disease progression and mortality. ROC curves were generated to determine the predictive performance of the cytokines.

Results: We included 400 hospitalized patients with COVID-19 infection, 69% had severe to critical COVID-19 on admission. Disease progression occurred in 139 (35%) patients, while 18% of the total cohort died (73 out of 400). High D-dimer >1 µg/mL (RR 3.5 95%CI 1.83-6.69), elevated LDH >359.5 U/L (RR 1.85 95%CI 1.05-3.25), lymphopenia (RR 1.91 95%CI 1.14-3.19), and hypoalbuminemia (RR 2.67, 95%CI 1.05-6.78) were significantly associated with disease progression. High D-dimer (RR 3.95, 95%CI 1.62-9.61) and high LDH (RR 5.43, 95%CI 2.39-12.37) were also significantly associated with increased risk of in-hospital mortality. Nonsurvivors had significantly higher IP-10 levels at 0 to 3, 4 to 10, and 11 to 14 days from illness onset (p<0.01), IL-6 levels at 0 to 3 days of illness (p=0.03) and IL-18 levels at days 11-14 of illness (p<0.001) compared to survivors. IP-10 had the best predictive performance for disease progression at days 0-3 (AUC 0.81, 95%CI: 0.68-0.95), followed by IL-6 at 11-14 days of illness (AUC 0.67, 95%CI: 0.61-0.73). IP-10 predicted mortality at 11-14 days of illness (AUC 0.77, 95%CI: 0.70-0.84), and IL-6 beyond 14 days of illness (AUC 0.75, 95%CI: 0.68-0.82).

Conclusion: Elevated D-dimer, elevated LDH, lymphopenia and hypoalbuminemia are prognostic markers of disease progression. High IP-10 and IL-6 within the 14 days of illness herald disease progression. Additionally, elevated D-dimer and LDH, high IP-10, IL-6 and IL-18 were also associated with mortality. Timely utilization of these biomarkers can guide clinical monitoring and management decisions for COVID-19 patients in the Philippines.

Keywords: COVID-19; Filipino; biomarkers; cytokines; disease progression; mortality.

Figure 1

Comparison of the clinical laboratory biomarkers throughout the course of illness between COVID-19 patients with and without disease progression. The clinical biomarkers are: (A) hs-CRP; (B) lactose dehydrogenase (LDH); (C) Ferritin; (D) albumin; (E) white blood cell (WBC); (F) neutrophil-lymphocyte ratio; (G) hemoglobin; (H) platelet count; (I) D-dimer; (J) procalcitonin; and (K) Troponin. Line estimates were created using Locally Weighted Scatterplot Smoothing (LOWESS) showing 95% CI as shaded area around the line estimate. Points at each day interval represent mean estimate computed at that time point. Broken lines show the normal ranges of the tests.

Figure 2

Comparison of the clinical laboratory biomarkers throughout the course of illness between COVID-19 survivors and nonsurvivors. The clinical biomarkers are: (A) hs-CRP; (B) lactose dehydrogenase (LDH); (C) Ferritin; (D) albumin; (E) white blood cell (WBC); (F) neutrophil-lymphocyte ratio; (G) hemoglobin; (H) platelet count; (I) D-dimer; (J) procalcitonin; and (K) Troponin. Line estimates were created using Locally Weighted Scatterplot Smoothing (LOWESS) showing 95% CI as shaded area around the line estimate. Points at each day interval represent mean estimate computed at that time point. Broken lines show the normal ranges of the tests.

Figure 3

Dynamics of serum levels of (A) inducible protein 10 (IP-10) and (B) interleukin 8 (IL-8) during the disease course in the COVID-19 patient cohort. The cytokines were measured in terms of days from illness onset. All samples collected from 400 patients were stratified into four intervals starting from illness onset. The dots represent individual measurement, and the box plots represent medians with interquartile range. The different groups were compared using the Kruskal-Wallis test with Dunn’s post hoc test. **p<0.01, ***p<0.001, ****p<0.0001.

Figure 4

Dynamics of serum cytokine levels during the disease course in COVID-19 patients based on disease progression (A, B) and mortality (C–E). The cytokines were measured in terms of days from illness onset. All samples collected from 400 patients were stratified into four intervals starting from illness onset. The dots represent individual measurements, and the box plots represent medians with interquartile range. The different groups were compared by repeated measures mixed model regression with post hoc test. *p<0.05, **p<0.01, ***p<0.001, ****p<0.0001.

Figure 5

ROC curves of serum cytokine levels to predict disease progression (A) and mortality (B) of COVID-19 patients during hospitalization.