[fabian2k]

It's really annoying that press releases tend to overhype research so much. As far as I understand from skimming the paper, they created a synthetic antibody against the Spike protein of SARS-CoV2 that binds very tightly.

There is no clinical data in the paper, they measured how tightly the nanobody they designed binds to the Spike protein and they did some neutralization assays against SARS-CoV2.

The advantages of the nanobodies compared to full antibodies sound interesting, for sure. But this is still a paper pretty solidly in the basic research area, and quite far from clinical use.

I found an article about this research that puts it into a bit more context:

https://www.statnews.com/2020/08/11/scientists-create-potent...

> While the lab results look promising, experts in the field advise caution because important work has not been done to test the compound in animals. “The critical thing is animal data. We’ve found things that are very potent in vitro that do nothing in vivo,” said Dimiter Stanchev Dimitrov, a professor of medicine who directs the Center for Antibody Therapeutics at the University of Pittsburgh and has created antibody-based therapeutics for numerous viruses including SARS and MERS, two other coronaviruses. He said it can take months to collect the needed data in animals. “Once these are tested in animal models, then I can get excited.”

[Duller-Finite]

Normally I agree that press releases way oversell the research, but I don't think that was the case here. Their camelid nanobody approach seems to be a fairly novel idea with some nice benefits over traditional antibodies, and most of the linked article actually does a nice job of walking you through the figures in the preprint.

Sure there's not any clinical data, but they actively admit that and I'm sure that's something they're working on. Furthermore, there have been multiple Nature papers published on SARS-CoV-2 neutralizing antibodies and antibody cocktails that use the same experiments (e.g. Vero cells) without testing in animal models or in humans. One step at a time!

[vikramkr]

Camelid nanobodies aren't that new, but this is still overselling stuff. I don't like reporting on preprint in general, and this article really doesn't explain the complexities of actually getting some early stage technology like this approved and developed. That's where you end up with the constant barrage of "oh look, another biomedical technology we'll never hear about again" comments, and the amount that this article tries to sell the tech in the headline and article really overstated the readiness of the tech and the near term impact it could have

[derefr]

If you take it as science reporting, mainly intended to be of interest to scientists or to people who follow the progress of science, then it's sensible. It's similar to the materials-science reporting they do for battery technologies.

These releases coming from University PR departments (this one's from UCSF) aren't really supposed to make it into newspapers for wide consumption. Their target audience is:

1. the people working in the same field—in other Universities, or in industry—who maybe don't have time to read journals, so you've gotta get their attention actively with a "billboard" announcement, rather than putting it in a journal they have to explicitly decide to read;

2. the people who fund the university, who want to see what sorts of neat things their money is being spent on.

Pop-science journalists sometimes glom onto these releases and make them more than they are, "retargeting" them for public consumption. You can certainly object to that. But as originally delivered, these publications are blameless for that.

[fabian2k]

I've never seen any scientists reading press releases instead of papers, they are certainly not targeted towards scientists. And you also don't have to read a full journal to notice an individual paper, it's quite common to have keyword-based alerts on Pubmed or something like that.

The level of detail in a press release makes it usually pretty useless to a scientist, it hides the important details behind language intended for non-scientists. The abstract of a paper is much more useful if you need to decide whether it is of interest for you at all.

[jrumbut]

I think you're right that the UCSF public relations team is doing neither hard hitting investigative journalism nor the sort of critical pre-Phase 1 analysis that prevents wasted effort in medical device development.

I enjoyed hearing about this new approach to infection control that is its infancy regardless. I felt a little hope for a creative solution to our current crisis and I didn't have to wade through the literature on camelid antibodies to do it.

New technology has always relied on a certain underlying optimism that you can do something that's new and better against the odds (since most fail).

Would you like this better if there was a disclaimer explaining in vitro/in vivo or just the long road from basic research to wide spread deployment?

[fabian2k]

The Statnews article I linked is a good example of reporting for this kind of very early result. It does clearly mention that this is at an early stage and that no clinical data has been collected yet. I think this is certainly interesting, it's just so tiring to see this kind of stuff turn up again and again in extremely misleading articles in mainstream media.

Though I think it would have been better to wait until peer review before making a press release. This is not a paper that has immediate clinical applications, and the peer review might still turn up some problems with the paper.

[Duller-Finite]

The article reports on the in vitro results and says they're going to be doing further validation in vivo. As I mentioned before, the same has been true for basically every other antibody being developed (monoclonals and ones isolated from patients), which were also initially published in Nature/Science/etc with only in vitro and structural data.

There are multiple aspects that are novel here: their size, the trimer (rather than a cocktail), and their stability/delivery. All of these seem worthy of reporting. They say they're about to start clinical trials to see if they could be efficacious; they don't say that they will.

[jvm_]

> While the lab results look promising, experts in the field advise caution because important work has not been done to test the compound in animals. “The critical thing is animal data. We’ve found things that are very potent in vitro that do nothing in vivo,” said Dimiter Stanchev Dimitrov, a professor of medicine who directs the Center for Antibody Therapeutics at the University of Pittsburgh and has created antibody-based therapeutics for numerous viruses including SARS and MERS, two other coronaviruses. He said it can take months to collect the needed data in animals. “Once these are tested in animal models, then I can get excited.”

https://xkcd.com/1217/

[Kwantuum]

Since you seem well informed: does the research mention anything about use as an aerosol and why that might be considered before internal use?

[fabian2k]

They wrote that regular antibodies have to be given intravenuously. These nanobodies are smaller and more stable, so they argue that it is possible to use them as aerosol. The idea seems to be to use them as prevention, which would not be feasible if you have to inject them.

I can't really judge whether that approach makes sense, my experience is in the basic research area, not with clinical stuff.

[mcintyre1994]

Is the hypothesis of an aerosol as prevention that you could spray eg. an indoor area and prevent infected people who come into that area being able to infect others in an airborne way?

[peeters]

> In an aerosol formulation they tested, dubbed “AeroNabs” by the researchers, these molecules could be self-administered with a nasal spray or inhaler.

[Kwantuum]

Right, but they do have the potential to also be used as treatment then? Considering they're easier to manufacture than synthetic antibodies.