2 Feb Episode 107 part 2 on prognostic markers

Tue, 02/02/2021 - 20:34

Dear colleagues,

The “preliminary results”of Sputnik V were just published in the Lancet.  At first view, this Ad26/Ad5 vector vaccine seems at least as good (or superior) to the “Western” vector vaccines (Astra-Zeneca en Janssen).  There is a good number of 60 + in the trial, but their mean age is only 65 (attachment 1).

A colleague made also a critical remark on the older participants in the Janssen (Ad26) trial: in fact there are only few and they are relatively young (see attachment 2)  

Another colleague shared a critical NYT article about the lack of transparency on the Chinese vaccines. Unfortunately, it seems that Chinese officials also cast doubts about the Western vaccines. COVID vaccines as just another ping-pong joustin and “apple of discord” between the West and China?  See https://www.nytimes.com/2021/01/25/business/china-covid-19-vaccine-backlash.html


As promised the second part on prognostic parameters.


  1. Very interesting paper on the relation between serum neutralization and disease severity (Ep 107-9).  It was repeatedly shown that severe disease was paradoxically associated with higher neutralizing titers than mild disease.  This is confirmed here (Fig 3 H slide 2).  Also the overall IgG anti-RBD (receptor-binding domain) is associated with disease severity (Fig 3 E).  However, if you calculate an index  of NT50 (50% neut titers)/anti-RBD IgG, which gives you a neutralization potency per IgG, you see that a lower potency is associated with intubation and mortality (Fig 3 K and M).

This paper also confirms the known associations between severity on one hand and age, sex higher IL-6, CRP,D-dimer, LDH as well as lymphopenia (slide 3). Moreover also associated        -      With severe disease: IL-8, IFN-gamma, TNF-alpha, CCL2, CXCL10 and soluble PD-L1

  • With mild disease: IFN-alpha, IL-2, IL-4, IL7, IL-15 (gamma chain cytokines), CCL-3, CCL-5 (beta chemokines) and Granzyme B   

Moreover higher antibody potency was associated with high GM-CSF and IL-33, whereas low antibody potency was associated with high IL-6

Remarkably steroid treatment resulted in lower anti-RBD IgG and NT50, but unchanged potency (NT50/anti-RBD IgG), whereas anti-IL6 (tociluzimab) resulted in increased potency (but at the expence of IgG levels) (slide4).  Unclear whether this has any therapeutic impact….

 A nice summary is shown in slide 5.

  1. A Canadian study by Pitre et al investigates relatively simple hematological and inflammatory markers as predictors of mortality (Ep 107-10). From Table 2 (slide 6), it is evident that potentially interesting markers are: higher total white blood cells  (WBC), lower lymphocytes, as well as the ratios neutrophils/Lymphocytes (NLR); platelets over WBC and lymphocytes over WBC are interesting hematological predictors.  Also higher C-reactive protein (CRP), lower albumin and the ratio between both, known as mGPS (modified Glasgow prognostic score) are interesting. Finally they propose that  CRP ≥ 78.4 mg/L, NLR ≥ 6.1, L/WBC < 0.127, and a mGPS 2 vs. 1 or 0 were associated with the highest risk of 28 day mortality.


  1. Mandel et al in Israel compared the value of various the inflammatory cytokines IL-1beta, IL-6, IL-8 and TNF-alpha in a relatively small study and finds that IL-6 is the best predictor, also as compared to age. (Ep 107-11 and slide 7).


  1. The most comprehensive study until now was recently published in JCI Insight (Ep 107-12): it investigates 66 biomarkers longitudinally by whole blood transcriptomics in 175 Italian patients with various outcome.  As shown in the graphic summary (slide 8). As can be seen, 12 biomarkers at admission were associated with mortality:
  • the cytokines: IL-2, IL-6, IL-10 and IL-15
  • the chemokine CCL-2
  • soluble receptors : s-TNFR-SF1A, s-ST2 (for IL-33), s-VEGFR1 (vascular endothelial growth factor receptor
  • neutrophil-activation: MMP-9 (metalloprotein); s100A (Ca-binding protein), NGAL (neutrophil gelatinase-associated lipocalin)
  • Iron metabolism: Ferritin

Clearly, these molecules refer to several molecular and cellular mediators of the “hyper-inflammation”.  As you can see, out of these 12, IL-10, IL-15 and the receptors S-TNFSF1A and s-ST2 were most consistently associated with mortality over time.


  1. With regard to the association of iron metabolism and various adverse outcomes of COVID-19, a Chinese study from Hubei(Ep 107-13) shows clear-cut results. As can be seen in Table 1 (slide 9), baseline values of low serum iron, serum transferrin and iron binding capacity as well as high levels of serum ferritin are associated with various adverse outcomes.  As could be expected ferritin is directly correlated with the pro-inflammatory IL-6, while the former 3 parameters are inversely correlated with IL-6 (slide 10).  The authors propose a mechanism by which IL-6 stimulates leads to hepcidin activation, which reduces release of iron from enterocytes into the circulation and uptake from the serum to macrophages.  This uptake would further promote SARS-CoV-2 replication and also stimulate further cytokine production….


  1. Finally, there is also a clear-cut association between hyper-potassemia  and 30 days mortality (Ep 107-14 and slide 11), which is to be expected.  




Besides very prominent epidemiological (age-sex-comorbidities) and clinical markers of respiratory and cardiac function, there is a number of well-established biological prognostic markers now:

  • Hematology: lymphopenia; NLR and P/WBC
  • Inflammatory markers; CRP, ferritin (and other markers of iron-metabolism),
  • Coagulation: D-dimer
  • Tissue damage: LDH
  • Cardiac:: troponin, BNP, CK-MB
  • Liver and kidney markers
  • Endothelial markers: soluble E- selectin, P-selectin; Angiopoetin-2 and soluble ICAM-1
  • The neutralization potency (NT50/anti-RBD IgG)

(This list is incomplete, I know)


The exact role of various cytokines, chemokines, proteases etc  remains to be determine: see the differences between Ep 107-9 and 107-12, but IL-6 increase seems consistent.


Large studies, using standardized techniques, in combination with bio-informatics should provide us with a tool, such as the one, developed in the UK (discussed in Ep 107-1).  It could  be very useful to perform a rapid risk assessment in newly admitted COVID-19 patients and potentially direct the treatment (to be discussed in the next Episodes).


Best wishes,