[DocSavage87]

Here's the paper they reference:

https://journals.sagepub.com/doi/full/10...1620974755

I don't want to dismiss it out of hand, but my best guess is that their "proprietary SARS-CoV-2 binding" test had poor specificity and that these were cross reactions.  However, if this is replicated with validated test with known specificity then it would change our understanding of where and when the virus emerged.  Certainly the Chinese have tried to deflect blame by saying it came from Italy.

And yes, if the virus emerged in Italy in September 2019, then I would expect that it was in the US by October 2019 at the latest.

BC

Agree that it seems likely due to poor specificity given the time course of the outbreak in Italy.  They were using anti-SARS-CoV-2 RBD Ab detection.

I believe the "proprietary SARS-CoV-2 binding test" they referred to was described in their earlier paper:
https://www.biorxiv.org/content/10.1101/...1.full.pdf

I noticed in a recent Nature paper (https://www.nature.com/articles/s41421-020-00224-3) this blurb: 

We also tested whether antibodies from COVID-19 patients cross-reacted with antigens from other CoV. We performed ELISA using plasma from the same patients with SARS-CoV RBD and S1 subunits of MERS-CoV, HCoV-HKU1, HCoV-229E, and HCoV-NL63. The latter two are circulating CoV from the Alphacoronavirus family.  Our results show that SARS-CoV RBD which has 73.8–74.9% amino acid identity with SARS-CoV-2 RBD cross-binds with antibodies from COVID-19 patients, as there is a positive correlation (Pearson correlation = 0.4316) between AUC levels measured by SARS-CoV-2 RBD and SARS-CoV RBD (Fig. 1d). This is expected considering the similarity of the RBD sequence between these two strains and that antibodies elicited by SARS-CoV-infected patients can cross-neutralize SARS-CoV-2. Antigens from the circulating human CoVs, including HCoV-HKU1, HCoV-229E, and HCoV-NL63 seemed to bind antibodies from COVID-19 patients, as well as healthy controls in varying degree, suggesting prior widespread infection of these CoVs in the general population (Fig. 1e and Supplementary Fig. S2a, b).

There look to be better approaches per that paper using S1 antigen.  Also it's not clear to me how their actually used cutoff translates to the sensitivity/specificity of their test.  This isn't my field so I'll leave it to someone else to examine. I'd be interested in learning how to interpret that ROC given their explanation.

[DocSavage87]

Here's the paper they reference:

https://journals.sagepub.com/doi/full/10...1620974755

I don't want to dismiss it out of hand, but my best guess is that their "proprietary SARS-CoV-2 binding" test had poor specificity and that these were cross reactions.  However, if this is replicated with validated test with known specificity then it would change our understanding of where and when the virus emerged.  Certainly the Chinese have tried to deflect blame by saying it came from Italy.

And yes, if the virus emerged in Italy in September 2019, then I would expect that it was in the US by October 2019 at the latest.

BC

Agree that it seems likely due to poor specificity given the time course of the outbreak in Italy.  They were using anti-SARS-CoV-2 RBD Ab detection.

I believe the "proprietary SARS-CoV-2 binding test" they referred to was described in their earlier paper:
https://www.biorxiv.org/content/10.1101/...1.full.pdf

I noticed in a recent Nature paper (https://www.nature.com/articles/s41421-020-00224-3) this blurb: 

We also tested whether antibodies from COVID-19 patients cross-reacted with antigens from other CoV. We performed ELISA using plasma from the same patients with SARS-CoV RBD and S1 subunits of MERS-CoV, HCoV-HKU1, HCoV-229E, and HCoV-NL63. The latter two are circulating CoV from the Alphacoronavirus family.  Our results show that SARS-CoV RBD which has 73.8–74.9% amino acid identity with SARS-CoV-2 RBD cross-binds with antibodies from COVID-19 patients, as there is a positive correlation (Pearson correlation = 0.4316) between AUC levels measured by SARS-CoV-2 RBD and SARS-CoV RBD (Fig. 1d). This is expected considering the similarity of the RBD sequence between these two strains and that antibodies elicited by SARS-CoV-infected patients can cross-neutralize SARS-CoV-2. Antigens from the circulating human CoVs, including HCoV-HKU1, HCoV-229E, and HCoV-NL63 seemed to bind antibodies from COVID-19 patients, as well as healthy controls in varying degree, suggesting prior widespread infection of these CoVs in the general population (Fig. 1e and Supplementary Fig. S2a, b).

There look to be better approaches per that paper using S1 antigen.  Also it's not clear to me how their actually used cutoff translates to the sensitivity/specificity of their test.  This isn't my field so I'll leave it to someone else to examine. I'd be interested in learning how to interpret that ROC given their explanation.