Subscribe and listen to the episode on iTunes.
Alice Lloyd George: Welcome to the latest episode of Flux. This is the first episode recorded remotely during pandemic times, so it’s an experiment. Myself, my guest, and producer Dan are all distributed.
Alex Bisignano: It’s great to be here.
ALG: Alex studied molecular biology at Princeton, where we met. After graduating he founded a clinical genomics company called Recombine, which sold in 2016. He’s now building Phosphorus, a full stack genomics as a service company. So Phosphorus has been doing genetic testing for a while. But what’s really interesting is how the business has rapidly evolved to also fight the Coronavirus pandemic. In fact last week, Phosphorus’s at home saliva sample kit was approved by the FDA, which is huge. It is the second at-home saliva testing kit to be approved. Congratulations on the news, Alex. Perhaps you can get into what your company was doing before the pandemic broke out, and what you have now been focusing on for the last few months.
AB: Absolutely. Phosphorus is a genomics company. Typically we are focused on inherited germline diseases and helping to diagnose and prevent those. Things like inherited cancers, heart disease, vision disorders, neurodegenerative disease. We had developed a testing platform using state of the art genomic technologies to detect these and license those methodologies to other laboratories. Now that COVID-19 has become the number one issue plaguing the world we have decided to focus our efforts on developing and launching tests for SARS-CoV-2 virus.
ALG: To clarify — the virus that causes the disease is SARS-CoV-2 and the disease itself is COVID-19 right?
AB: That’s right. So we’ve developed two tests. One is a PCR based test to look for viral RNA in a patient’s saliva. The other is an antibody test to detect two types of antibodies that our immune systems develop in fighting the infection. Our PCR based test uses a technique called qPCR. It’s a technique that we used regularly in our genomics business as a quality control measure before running next generation sequencing of DNA samples. We took our experience and even the instrumentation we had there and applied it towards detecting the Coronavirus. In order to do this, there are a couple of different approaches to developing tests. The first thing we did is acquire synthetic versions of the virus and create synthetic samples. This helped us validate analytically that we were able to detect the virus. That’s all that’s required under the emergency use authorization (EUA) of the FDA in order to go live. Subsequently we launched a clinical trial to recruit patient samples in real time, at urgent and primary care centers in New York City. We tested those patient samples with our methodology and compared them to other laboratory results. We thought this was an important step because as we start doing more testing, understanding clinical performance is key. It also allows us to develop a biobank of samples that we can use for ongoing internal proficiency testing.
ALG: How did you put out the call to get live patients and what was the process like collecting those samples?
AB: We had to first develop a study protocol. Number one when you’re doing this type of research is to ensure that ethically you’re not doing anything that could harm patients. We had that protocol reviewed by an independent review board and approved back in late March. What that allowed us to do is go to sites where we have incredible physician partners who said, yes we’ll help with this. We have research associates — EMT and medical students — really incredible humans who’ve gone to the frontlines and helped to educate patients on the study. Then we collected saliva samples for testing the qPCR assay. We also collected blood samples from these patients. Some patients were called in for testing a week or so after their infections. We were then able to use the blood samples as validation in the development of our antibody tests.
ALG: It’s amazing to see people like you on the frontlines mobilizing so quickly. What does it look like to scale this up fast? Big corporations like Abbott are trying to spin up their ID Now testing. But even for the big companies it’s still early days. They are still dealing with problems like false negatives and how to ramp production. How does all that work for a startup of your size?
AB: The major instrument that’s done most of the PCR testing to date is made by Roche — the Cobas system. These systems are sold out until the fall. Not just scalability, but getting new testing up and running is something that we need in this country. We had instruments already for our qPCR assay and we made a decision to build our tests off of a system that’s manufactured by a company called Bio-Rad. While these systems are still in demand, they’re not as sold out as some of the larger manufacturers. That’s allowed us to have a bit more security in our supply chain as we think about scaling up.
ALG: That’s interesting. Not a bad idea to diversify your supply chain and source equipment from alternative vendors that are less backlogged. So when do you think you’ll start to produce these tests and how will you distribute them to patients?
AB: We just received FDA approval for our viral PCR test using saliva samples. This is exciting because it can be used at home unsupervised. We are pending FDA approval for our blood based antibody test but hope we will receive it soon, and are currently in the market for both tests today. Right now we’re a small lab. We’ll be starting at just over 1000 total tests a day but hoping to scale that capacity up over the next couple of months. As far as who we’re trying to work with and how we’re working on distributing the tests, we believe that it’s important to get them to healthcare providers, hospitals, private practices — any groups that need to understand what to do to bring people back to work safely and build safe, Covid-free spaces and protocols. We’ve heard that medical providers are having a hard time with their P&Ls as they’ve been shut down to a large degree. To ensure that we can keep delivering care for Covid patients, we need these institutions to open up more broadly.
ALG: So you’ve collected the live human swab samples as well as the serology tests for the antibodies. Is it not that common to use real patient data? What’s the ratio of companies using synthetic versus real patient data?
AB: So far most of the tests on the market are reporting outcomes from synthetic data that’s posted on the FDA’s website. That’s been the FDA’s primary concern — that the analytical work in the laboratory be good and that’s what they’ve been looking at in many of the companies that they’ve reviewed. There have been very few, to my knowledge, large scale clinical studies and clinical data sets published. Partly it’s because getting that data early on in assay development is challenging. Most laboratories only have access and publish that data after they’ve been on the market. We wanted to lead with data because there are so many unknowns with respect to performance of these assays. As we’re seeing in the news, different antibody tests are all over the map. Having developed one in the lab I can tell you that there is a range of challenges — from which protein do you look for in the virus? All the way through, how do you find representative blood samples from patients where you know exactly what day of infection they are at with respect to which antibodies they might be producing? There’s a lot of variables that are important to control for.
ALG: It seems hugely important to use real patient data. But I understand it’s difficult to obtain. What was it like for you to be in those rooms with Covid patients and witness the testing in person?
AB: As we started the study I was on site helping with the patient data collection and training our new staff. As you’d expect, everyone is fearful. Testing is one of the ways that we can help people understand the risk and allay these fears. I have nothing but incredible gratitude for not just the physicians and our staff, but the patients who understood that by participating it might not help them so much as future patients that we are able to get this test to. It’s been an interesting ride for us to make sure that we’re doing this ethically and responsibly and with the right access to data ultimately.
ALG: Most of the companies we’ve mentioned so far are actually medical device companies, whereas you are first and foremost a genetics company. I’d love to get into that. We’re all racing to understand COVID-19. Some people show mild or no symptoms and others develop the disease severely. Obviously age and underlying health conditions like diabetes or heart disease are factors. But there are also young people who are healthy that are dying. I haven’t seen much about genetics in the public discussion. But clearly genes play a role in malaria, HIV and other diseases. What are your thoughts on how genetics ties in?
AB: Genetics is a dimension that needs to be studied as we progress in our understanding of this virus. We’ve already been in discussions with academic partners on taking samples that we’re doing testing on — whether it’s blood samples for antibody testing or saliva samples for PCR testing — and banking them. Such that for consenting patients, we de-identifying the samples to use for genetic studies in the future. I like to say that the outliers inform the mean. By that I mean when you look at more extreme cases, whether it’s someone who’s incredibly young that’s affected by something that typically only affects older populations, you might find something that is unique with respect to their genetics or other factors that ends up being a good predictor of severity. We see that in areas like hypercholesterolemia or other heart diseases. These methods of thinking about disease need to be applied here. We’re hopeful that studies and ongoing work that we do on this virus will involve sequencing different types of genes to examine different theses for what else might help us predict not just infection rates, but the severity of the infection.
ALG: I’ve heard talk of the ACE-2 gene. Is that one of the things people are investigating?
AB: Definitely. The viral spike protein is able to gain access to our cells by binding that gene as its primary target. The thought is that with people who express that gene more, the rate of infection for them is higher. Not necessarily the severity of the disease. There had been studies early on coming out of China showing that men might be more susceptible because of this pathway. Though it could be linked to smoking, it could be linked to hypertension or could be linked to a number of other kind of co-morbid diseases that cause expression of that gene. But it’s a great example of the way that genetics might impact our ability to predict who’s at risk.
ALG: If you look at history, most pathogens that have hit at large scale have affected the young. COVID-19 is such a peculiar disease in that it takes out the old and has mostly left younger populations untouched. There’s been wild conspiracy theories relating to the aging population and social security burdens, but I’m wondering what biological explanations there are. Is there a genetic argument for why this is happening?
AB: As we age different biological processes slow down or are performed less efficiently. For example, conversion of cholesterol into hormones is something that tends to become less efficient as we get older. That results not just in increased cholesterol, but a lower presence of certain hormones in our body. Another example is blood clotting. As we age risk from thrombosis and blood clotting tends to become more pronounced. Many of these different pathways that we know change with age could be candidates for severity of this virus, because there’s a correlation with how they change over time. We certainly know that the virus is hitting our older population more harshly than the younger population.
ALG: It’s strange since newborns don’t have a fully developed immune system.
AB: That is one of the other areas of interest — heterogeneity in the population of T cells and B cells, our immune cells. Younger children and especially people under the age of 20 and 30 more actively create diversity in the immune cells they produce. As we get older our thymus gland shuts off and we stop evaluating new immune cells. There’s a lot of unknown unknowns that we’ll be able to study once we get through the horrible acute phase of the virus getting into the population.
ALG: Fast forwarding into a hypothetical future. Say we sequence a lot of data and build out these biobanks. What might it look to know which people are at risk and which ones aren’t? How might that play out with ending the lockdown?
AB: Yeah it’s an important point. I don’t want to overpromise on our ability to identify who are truly at high risk. But the more finely we can narrow that width and understand that segment of the population, the better we can develop return to work policies. If we were able to with high precision, identify people who were the high risk people for the cytokine storm and the intubation, those who would benefit from taking extra precautions, we could structure a return to normal that entails those individuals maintaining higher levels of personal safety and then actually target them first with the vaccine or whatever interventions we’re able to get out. There’s huge value in identifying strong predictors of severity of this disease.
ALG: Not to stray too far down a science fiction path, but people have been talking about bio passports and bio identity. If we are able to identify who’s immune and who’s at risk, there are implications for the labor market. We might look at redistributing those who are immune such that they become essential workers in jobs and professions that are more likely to be in contact with the virus. And when events of 50+ people return at some point, people who test negative might get a bio pass to attend. In China they have already have so called green light red light systems. What do you think of these social structures and policies?
AB: They’re fascinating because these are in some ways personal privacy and employment issues as much as they are personal safety issues. Those two things are definitely coming to a head the discussions I’m having around this. We think it’s important to get testing out quickly to our health care providers and the institutions providing care for not just Covid patients, but patients in other areas. Otherwise those systems will not be able to get up and functioning. But the concept of applying labor based on people’s Covid status or their susceptibility to the disease raises ethical questions around mandatory testing, mandatory disclosure. Good legal minds are thinking about whether this is even possible given the current state of laws and whether laws will change around this quickly in order to accommodate a society where the virus and its impact are acute? There’s an ongoing discussion on many levels that I’ve either been a part of or been witness to in my early work here. The bio passport is interesting — are you getting a wristband, a passport or something that allows you to share your status with your employer or a restaurant or wherever we’re going in the future. This is an important question around public health versus personal privacy that Covid is bringing to light in a way that we would not have expected six months ago.
ALG: It’s really forced conversations that were previously on the sidelines to the forefront. What Apple and Google are doing with contact tracing is interesting. Clearly Silicon Valley has mobilized fast and the rest of the country was initially quick to scoff at that response. But as we look to build out pandemic solutions, companies like yours that have that mentality of building quickly can come up against bureaucratic institutions and government bodies who act slower. Or segments of society who think differently. Whenever you have those discrepancies in approach it exacerbates the tensions and strains.
AB: Yeah. This also becomes more pronounced as we understand the data on the disease. If 5 percent of people are extremely high risk versus 0.1 percent — not suggesting either of those are the real numbers — that will affect the degree to which society is going to respond from all these perspectives.
ALG: It’s particularly interesting to see the divergence in how countries have responded. Look at Sweden or Taiwan. Have you seen compelling best practices in other countries?
AB: The only thing I’ve seen that is compelling is data on wearing masks. At least the U.S. is now fully on board that we should be wearing masks. Countries that had experience with the SARS epidemic have performed better at the immediate phase of infections. Part of it is experience. Part of it is probably cultural. Availability of testing is another area. I wish we would have had that earlier on. But hopefully we get brought up to speed as labs like mine and others are able to pivot into making our resources available.
ALG: You’ve been in the biology domain for some time now. How does it feel to look at your career and come up against this global crisis, does it feel like your work has culminated in this moment?
AB: It’s been an interesting personal journey. There was that moment that we had as an organization where we asked, do we have the skill set and the resources to do something? Early in February we thought, is it necessary for us to step in? Lack of focus is often the reason that startups don’t execute well. And our focus was not virology or infectious disease. Then it became apparent in March that this was going to have sweeping implications. It was almost magical when the team said, we have the ability to do something. Let’s do it. That wasn’t an easy decision. We had to reorient absolutely everything and figure out how to maintain our current customers. I had to convince my staff that we would put in the right safety procedures. Now instead of socially isolating from the virus, we were bringing the virus to us. Right. So there were a lot of interesting thoughts going through our heads as we made the decision to get into this. But I consider it a good thing that our response is, what can we do to help? We’re lucky that we’re in a position to help. From a business perspective, we’re lucky that Covid is something that can provide us with an opportunity to apply our infrastructure.
ALG: It’s a brave response. Both on the personal and the business side. You have VC investors, so I presume your board supports this direction. On that, I’m interested in your thoughts on the capital markets. Over the last 20 years there’s been a divide among venture capitalists — they either look at software or they look at biotech, but not both. Boston firms aside, gone are the days of generalist funds like Kleiner Perkins backing say Genentech. You’ve raised in Silicon Valley. Do you think there will now be a shift in terms of capital going towards companies focused on hard bio?
AB: That’s a great question. There has been in the past decade, from my perspective fundraising for two different companies, major shifts in capital allocation towards biological companies that exhibit informatics or computer science data structures. Genomics has been one of the forefronts there. Raising money for my first company, there was resistance to the fact that even though we were a genomics information technology company, that we had to own a clinical lab. The clinical lab was almost a drag to a lot of venture capitalists, especially on the West Coast. As we’ve progressed people have seen that this is a regulatory operation too. You need businesses that tackle not just the informatics and the artificial intelligence, but the regulations, the payers, the healthcare models. I’ve seen more venture capitalists understand that great artificial intelligence and data science can’t be applied to this industry in a vacuum. I’ve now seen Covid responses with funds already looking to fund this area. But there’s going to be more of an appreciation that good business models in healthcare need to bring together the infrastructure for healthcare with new applications of technology and all the the wonderful advances in data science and A.I.
ALG: Well I hope that in the coming months and years we will see more capital and talent flow towards hard sciences and biotech. It’s unfortunate that this pandemic is going on, but there will be a silver lining if it focuses more energy into the hard sciences. That will be a good outcome for decades to come. Thank you so much, you and your team at Phosphorus, for everything you’re doing on the frontlines. We are lucky to have such minds focused on this problem.
AB: We’re fortunate to be able to operate and do something in this environment.
Could genetics explain why some COVID-19 patients fare worse than others?
Certain genetic differences might separate people who fall severely ill with COVID-19 from those who contract the…
Human leukocyte antigen susceptibility map for SARS-CoV-2
Genetic variability across the three major histocompatibility complex (MHC) class I genes (human leukocyte antigen…
Five Variables that Can Affect COVID-19 Symptoms, Prognosis, and Outcomes
May 7, 2020 What explains the range of COVID-19 infection outcomes? From asymptomatic to a bout of coughing and fatigue…
Opinion | Covid-19 quickly kills some while others don't show symptoms. Can genetics explain this?
Andrea Ganna is a group leader at the Institute for Molecular Medicine Finland at the University of Helsinki. Benjamin…
FDA Authorizes COVID-19 Saliva Test by Phosphorus Diagnostics
(Reuters) - The U.S. Food and Drug Administration on Monday allowed emergency use of a COVID-19 saliva test made by…