Skip to main content
SearchLogin or Signup

Review 2: "Sofosbuvir protects human brain organoids against SARS-CoV-2"

This study claims brain organoids are permissive to SARS-CoV-2 infection and use this model to demonstrate neuroprotective effects of Sofosbuvir. While the findings are compelling, the experiments performed and the data offered are insufficient to deem the stated claims reliable.

Published onSep 16, 2020
Review 2: "Sofosbuvir protects human brain organoids against SARS-CoV-2"

RR:C19 Evidence Scale rating by reviewer:

  • Potentially informative. The main claims made are not strongly justified by the methods and data, but may yield some insight. The results and conclusions of the study may resemble those from the hypothetical ideal study, but there is substantial room for doubt. Decision-makers should consider this evidence only with a thorough understanding of its weaknesses, alongside other evidence and theory. Decision-makers should not consider this actionable, unless the weaknesses are clearly understood and there is other theory and evidence to further support it.



In the light of the COVID19 outbreak, the current work claims that iPSC-derived brain organoids can be infected with SARS-CoV-2. The authors declare that the infected cells die. Furthermore, the authors propose that Sofobuvir can mitigate the effects caused by the virus.

There are a couple of works out there as a preprint. Compare to them (which all claim that the virus can enter the neurons), the current work differs that virus can infect organoids. Although this is somewhat surprising, the drug treatment part is exciting within the context of neurological effects.

Here are my comments:

It is a rapidly evolving field of work. The introductory paragraph sounds very primitive. It must be updated with the current trends and why studying neurological phenotypes are essential in organoids. Cite the relevant literature with recent preprint works.

By definition, infection means the virus enters the cell and produces more of it, just like ZIKV in human brain organoids. I don’t see any data showing productive infection data (viral titer, progenies, or viral mRNA quantification). Without these, the word, “infection should not be used.”

Based on the displayed items, I am unable to figure out the details. The organoids display nestin and MAP2 throughout the region, and there is no defined cytoarchitecture is shown. I am wondering why there is no VZ in which nestin can be observed as radial fibers. I am also puzzled why nestin is seen along with MAP2 in the region, which I assume to be a cortical region.

Low mag complete organoid slice along with a subset of images showing specific cell types should be given. Otherwise, it is difficult to identify the target cell types. Each cell type (nestin-positive and GFAP-positive) must be co-labled, shown with the virus.

Other concerns:

-The statistics for every figure must be rigorously performed
-Low mag and high mag images should be given
-Organoid architecture should be given
-Authors have to clarify what are the cell types are favorable for the virus. If they claim, NPCs, then the virus should be shown located at the VZ of the organoid.

Overall, I understand the current work is too elementary and purely descriptive. This is because of the current extraordinary situation and the potential limiations related to the access to the safety level labs and extraordinary times.

The drug part is attractive, and they made a clear logic why they wanted to use it.

Now it is the authors choice of how they want to proceed. Without addressing the organoid-related experiments, I don’t think the current work will help the field (it may rather reduce the quality of the work).

Considering this, I propose them to focus on the drug part (not going too much into organoid architecture, cell types, etc.), mainly focusing on viral addition and staining of organoids with anti-SARS-Cov2. Titration would be great but, I guess there is no infection. After drug treatment, the authors can show that the frequencies of positive viral cells are reduced.

At the method section, I do observe that the authors have performed viral titers. I, unfortunately, fail to see the data for organoids. It is unclear to me if they did test the productive infection. Did they?

Having these two solid data pieces in a very concise form will provide some useful information to the field.

While giving these, the authors must pay attention to tamper down their statements. At the moment, it sounds as if the observed effects are happening in the human brain, and the virus is harmful to the developing fetus.


No comments here