26 April 2020 – Sunday – #42

A few milestones. The world is approaching three million confirmed cases of Covid-19 and has passed 200,000 deaths.

In better news, Nicole got to take her four year old daughter outside for the first time since the Spanish lockdown started in mid-March. She sent me a photo to prove it!

As many parts of the world move toward a relaxation of Covid-19 lockdowns, I want to do a whirlwind tour today of what’s happening in Covid-19 science and technology.

First, it’s important to note that, while doctors and scientists understand a lot about how the Covid-19 virus enters the body, they are still figuring out what happens after. Nearly six months into the pandemic and the news this week is that even in a mild or unnoticed case of Covid-19, some patients develop blood clots that result in strokes. Covid-19 devastation is so varied that we’re bound to find out new problems over the next few months.

The most severe cases appear to result form Covid-19 triggered cytokine storms that damage tissue and organs all over the body. Here’s list of documented problem areas so far:

• Lungs
• Heart and blood vessels
• Kidney
• Liver
• Intestines
• Brain
• Eyes

So, once a Covid-19 infection starts, its open season on vital parts of the human body. Medical response to severe cases often takes a multidisciplinary team because of the range of organs attacked.

There are two basic ways forward on the science and technology front. One is using testing and tracking to attenuate transmission, the other is a therapy to resolve or prevent infection. Most of today’s entry is about testing and tracking because that’s the phase we’re heading towards in the next month or so.

On the testing front, many countries are far, far behind on their Covid-19 testing programs. I don’t know how many tests are needed, but the range is somewhere between 5 per 1,000 people per day (the rate that turned around the Covid-19 outbreak in South Korea) and 70 per 1,000 people per day (the rate Paul Romer estimates will contain Covid-19 without any restrictions). To make the world safe, the entire planet needs to test at these rates. The US is currently performing 300k tests per day, or not quite 1 per 1,000 per day.

At a basic level, we still don’t have great data on infection rates in the US and other countries. The recent ballyhooed studies in Santa Clara County and Southern California that reported higher than expected infection rates have not done well going through peer review. In San Francisco, USCF is teaming up with local neighborhood organizations to test the entire Mission district. Until there are higher levels of testing, these studies are the only way to understand infection rates, and they’re hard to do correctly.

Infections rates help with forecasting, but they also indicate whether a population has achieved herd immunity. With most infections, when 60% of the population achieves immunity either by infection or preferably by vaccination, infectious outbreaks end. With Covid-19, the science isn’t clear yet how long immunity lasts after exposure and whether 60% immunity will protect the herd. Better infection rate data will help understand that.

I reported before about efforts to increase Covid-19 RNA testing by eliminating the need for reagents to extract viral RNA. Paul Buchheit, the guy responsible for your Gmail, has proposed a ubiquitous daily testing scheme using kiosks equipped with Surface Plasmon Resonance (SPR) scanners for Covid-19 testing. I kind of understand SPR, but let’s just say that we know SPR has been used for at least 10 years to detect viruses.

In theory, a Covid-19 SPR scanner takes a small saliva sample and spits out a Covid-19 test result on the spot in ten minutes. Buchheit’s goal is to deploy millions of SPR testing stations around the US by the end of 2020.

“We’re planning to start operating the first scanner within a month. It’s a fully automated system, similar to a kiosk or turnstile. If all goes well, there will be millions of scanners deployed by this fall, ensuring that every school and essential business can reopen while remaining safe and virus-free.”

Paul Buchheit

I love this idea and I hope Buchheit pulls it off. As with other Covid-19 advances like treatments and vaccines, though, it’s one thing to find a solution, it’s quite another to deploy it at scale, especially in the absence of a US federal government or world body that can coordinate the effort.

As tests generate heaps of data, the data themselves become an asset in attenuating Covid-19. Big data provides opportunities not available five years ago. As I wrote about earlier, CMU and UMass Amherst are training machine models on search queries and other data to forecast Covid-19.

But there are also straightforward applications of testing data today. The bioinformatics company Hc1 is aggregating Covid-19 lab data to provide near real time dashboards of testing results at higher resolution than public health reporting. This gives government and private planners county level resolution of Covid-19 outbreaks.

“As the antibody tests come online we’ll be able to see who might have had the infection but was never symptomatic, for example, or who might have immunity. So we’re certainly taking the long game and building a road map to inform not only the surge and the apex of this but also how do we start looking at the recovery and what happens with respect to outcomes.”

David Dexter, CEO, Sonora Quest Laboratories on Hc1 lab test aggregation service

The Israeli company Medial EarlySign is revamping its influenza forecasting service to work with Covid-19 data. Medial will help healthcare systems organize resources to respond over the course of the pandemic.

“We think we can do a pretty good job scanning the entire population of a health system and bringing to their attention which [COVID-19] patients are most likely to die, be hospitalized, need a ventilator, etc.”

Jeremy Orr. CEO, Medial EarlySign

China, Germany, Vietnam, Taiwan, and South Korea have demonstrated that testing is key to controlling Covid-19 outbreaks. Testing is ramping up quickly elsewhere and may attenuate Covid-19 as early as this year. Aggregation of testing data will deliver more miles per test, as it were, reducing the frequency of testing needed to control Covid-19.

In the longer term, we want a treatment or a cure. Derek Lowe provides a nice summary of the three main thrusts of Covid-19 therapy development, existing drugs, monoclonal antibodies, and vaccines. Repurposing existing drugs is the quickest path to a treatment because it avoids a series of trials and because manufacturing is already understood. Repurposed drugs are unlikely to provide a cure, but they are the best bet for a treatment in the next six to twelve months.

“Don’t expect cures – the odds of something working that well when it was developed for some other use entirely are extremely small. Wandering through the shelves of the auto-parts store and throwing things out into the parking lot at random is unlikely to fix your broken lawnmower, for similar reasons. If you’re very lucky, you might find something that can be jammed on with a hammer and aligned with duct tape, and that’s pretty much what drug repurposing is like even at its best.”

Derek Lowe on drug repurposing

As Trump’s hydroxychloroquine con illustrates, it takes time to perform studies needed to know a drug or vaccine works. As much as a pitch man like Trump wants us to believe otherwise, there’s no way around that.

On the vaccine front, Johnson and Johnson has announced plans to deliver a Covid-19 vaccine in the first quarter of 2021. The quickest previous vaccine development took five years, so Johnson and Johnson needs everything to go right to hit its aggressive one year target. I hope J&J gets there, but there’s no historical data that says you should plan based on that delivery date. Most experts I’m reading think a Covid-19 vaccine is possible in three to five years.

As noted above, cytokine storms appear to trigger inflammation and damage throughout the body. The FDA is fast tracking therapies that potentially manage cytokine storms in Covid-19 patients. A non-peer-reviewed German study shows that high levels of cytokine interleukin-6, or IL -6, were associated with patients who needed ventilator support or had Acute Respiratory Distress Syndrome (ARDS). Drugs like the IL-6 receptor blocker tocilizumab, which is used to manage cytokine storms in other disorders, are under study for use in Covid-19.

Technology has helped us during the Covid-19 pandemic in another way. We’re’ using it to stay in touch and keep our sanity. Or to lose our sanity in a flurry of frustrating Zoom interactions.

One UC Berkeley student is using Minecraft to imagine his shelter-in-place graduation. Inspired by a sarcastic remark on Facebook, Bjorn Lustic started building a model of Memorial Stadium for a remote graduation. The project has blossomed into 500 students using satellite data to build a Minecraft replica of the entire UC Berkeley campus.

On a different note, congratulations to Prime Minister Boris Johnson on his return to 10 Downing. I may disagree with his politics, but it’s heartening to know he made it through a particularly difficult case of Covid-19.

To keep you inebriated entertained today, here’s your quarantini upgrade, the American 45.

---

I write Covid Diary BCN for my own sanity. If it helps yours, please share it on your social media feeds and email it to your family and friends. Thanks!