Pia Abola (00:05):
Welcome to Countering Covid a podcast by LGC Biosearch Technologies, where we speak with clinical lab scientists and industry experts to understand their perspectives on the curtain feature state of SARS-CoV-2 testing. My name is Pia Abola and I'll be your host. On this episode, we're speaking with three scientists from Combinati, a company that is using their digital PCR platform to conduct wastewater testing for SARS-CoV-2 for organizations such as universities and local governments. Here with us today, we have Robert Lynn.
Robert Lynn (00:50):
Pia Abola (00:51):
Christina Bouwens (00:53):
Pia Abola (00:54):
And Adam Langston.
Adam Langston (00:57):
Pia Abola (00:59):
Tell me about Combinati and what it is that you do.
Adam Langston (01:05):
Combinati has been around for four to five years and we are at our core a life science research tool provider. Our focus for the last several years has been on digital PCR and we leverage the power of microfluidics for what we consider to be the most accurate, easiest to use digital PCR platform on the market.
Pia Abola (01:25):
How did Combinati get started doing wastewater surveillance?
Adam Langston (01:32):
Yeah, I think Combinati three to four years ago was really focused on cancer recurrence monitoring as the predominant application for digital PCR and still I think that is a major focus for the technology. Given the emergence of SARS-CoV-2 or Covid-19, it became pretty evident within the wastewater community that digital PCR was the superior tool for what I call needle in a haystack application. We're looking for that genetic needle in a very complex sample matrix. Our focus largely shifted in the middle of last year to support those scientists that were looking to enhance their wastewater surveillance capabilities.
Pia Abola (02:20):
You weren't monitoring cancer recurrence in wastewater, were you?
Adam Langston (02:25):
No, this was in essentially liquid biopsy and human samples.
Pia Abola (02:30):
Tell me a little bit about wastewater surveillance and how it fits into pandemic management.
Robert Lynn (02:38):
Absolutely, wastewater surveillance is essentially looking at the disease causing agent in the waste stream, as opposed to in individual humans or in patients. It's quite useful in that it does not depend on interaction with the population but what you're looking at your sample essentially is a collected pool of sample downstream from the entire population or a given community.
Pia Abola (03:01):
How has wastewater surveillance been used in other situations?
Adam Langston (03:06):
I think wastewater surveillance, this isn't the first time it's been used as a matrix to monitor certain things but this has certainly shined a spotlight on how it could be used for other diseases and areas outside of SARS-CoV-2.
Robert Lynn (03:26):
We are by no means experts in this area and a lot of the things we learned from looking at the literature, talking to folks that's been doing modern wastewater for a while. Wastewater monitoring definitely didn't start with the SARS-CoV-2. Actually, it has been used to monitor polio for a long time. Obviously polio is not something that the America has a primary concern with but overseas, it actually has been used for a while to good amount of success. We've actually have worked with somebody who's been collecting wastewater for polio for a while. When SARS-CoV-2 happened, they were able to go back and look at the samples they already collected to look for SARS-CoV-2 and lo and behold they were able to find it in samples that they have already collected for other reasons to start with.
Pia Abola (04:11):
Interesting. Do you feel that you guys have advanced knowledge about wastewater monitoring for a pandemic response or infectious disease response?
Robert Lynn (04:29):
I would say digital PCR is a great tool in this area as well as others. This has been a opportunity in the sense that we were able to leverage this as an opportunity to share how digital PCR can be used. I think America certainly hasn't seen pandemic in a level like this. I think this really... It's helping bring to light that this can be a tool and this can be a widely deployed tool and it serves communities large and small.
Pia Abola (05:02):
Okay. Transitioning to digital PCR, so you mentioned that you guys do digital PCR and it's great for wastewater surveillance. How is it different from typical SARS-CoV-2 assays that rely on qPCR?
Christina Bouwens (05:23):
Yeah, I can take this question. Visual PCR really differs from qPCR because it's a way of quantifying nucleic acids that doesn't require a standard curve like qPCR does. One of the major advantages it has for wastewater monitoring for Covid specifically is that you can have more consistency between labs because you're not relying on preparing the exact same standard curve across all these different facilities, with all these different limitations, different preparation strategies. By doing digital PCR, what we're doing is we're taking what would be a really standard [oak 00:06:04] qPCR reaction and we're splitting it up into thousands of reactions, so that there are very few molecules present in all these different thousands of reactions.
Christina Bouwens (06:14):
After you've done your PCR, you can individually look at all those individual reactions and just count for the presence or absence of an original molecule. Then that way all what we're doing is absolute quantification rather than comparing to some curve or a positive control that would just fluoresce at a different cycle threshold, which is how things are standardly done in qPCR.
Pia Abola (06:38):
Cool. What makes your platform different from droplet digital PCR?
Christina Bouwens (06:48):
I started my journey in digital PCR actually using emulsion based digital PCR. I found it really challenging and it is great I could see the advantages of digital PCR over just regular PCR, qPCR but it was challenging to do. It actually. It was something that was really tough for anybody else in my original lab at the time to do, I was the only person who could actually do it because it required a lot of skill to effectively get through that method. Then I found out about Combinati and we use a very different approach to partitioning. It's not emulsion based, it's actually done on a physical plate. There's no work involved from the view of the user, which really helps to improve not only just user workflow but also your overall accuracy. If you're not doing any of the physical partitioning yourself or transitioning of all of these reagents.
Pia Abola (07:48):
What limited detection do you have?
Christina Bouwens (07:52):
Yeah, the limit of detection for digital PCR is on the average of just several target molecules per sample volume that you look at. You'll frequently hear down to one or two copies per reaction. That is truly because we're partitioning these reagents so that you're just looking at the end of the day on that individual molecules in final partitions. It can be very, very low down to single copies.
Pia Abola (08:18):
I imagine, Oh, go ahead.
Robert Lynn (08:22):
I was going to jump in and say, it's not just that it's very sensitive, it's also the confidence you have in how consistent that measurement is. If you're just looking at a CT or a CQ, small differences really... It's hard for a user to interpret a very small difference between CT. With digital PCR, because you have the absolute quantification, even small differences, you can have confidence you're actually detecting that it. In general, any monitoring applications, being able to confidently interpret small changes, it's very, very important. Digital PCR is especially suited in wastewater monitoring for this reason as well.
Pia Abola (09:00):
Is what you're saying that the possibility of having a false negative is not as high, something like that.
Christina Bouwens (09:12):
Yeah, the possibility of having a false negative is much lower with digital PCR. If you can imagine, if you're looking at a negative sample in a qPCR reaction. At some point it may fluoresce at some cycle unknown. With digital PCR, what we're actually looking at is just the presence or absence of a quan. Your negatives don't look like background signal, your negatives truly look negative. Because we're actually counting the molecules that are there, we have that confidence, like Rob said for those single molecules that could be present, which gives you more confidence in either a true negative call or a true positive. You don't have that level of confidence with qPCR.
Adam Langston (09:56):
Yeah, I think to the same point and to Rob's earlier point, for monitoring purposes, eliminating that requirement for a standard curve is a huge advantage. If we're looking... if we take dorms as an example, because I know a lot of college campuses are using this method to monitor whether or not they have students staying in those dorms testing positive. If they want to actually correlate that data, they in many cases do want to correlate that data to the number of potential positive cases.
Adam Langston (10:28):
If you've got several hundred students living in one facility and you're able to monitor the wastewater coming from that individual building, if you see a very minor change in the concentration of that genetic material, if you're using qPCR, that could actually be a result of minor changes in the standard curve that you're generating day over day with a different technician or a different pipette. By using digital PCR. We can really standardize the testing method so that if we do see a small increase in the concentration of that particular target, it actually can be better correlated to the frequency of positive cases.
Pia Abola (11:09):
Then why isn't digital PCR used more in conventional testing?
Adam Langston (11:15):
That's a really good question but I'm happy to speak to you. Conceptually digital PCR has been around for awhile, certainly not as long as qPCR but for a long, long time digital PCR has been perceived as being very cumbersome, very challenging, very hands-on and very expensive. Our technology really is focused on trying to lower that barrier to entry for those who are looking to make the conversion from Q to DPCR. We try to reduce the overall cost but we also really are focused on making sure that it's a very assessable method and that really any scientist can walk into a lab and not feel intimidated by a multi instrument, complex hours, long workflow. That's why we're starting to really, not only is it technically more advantageous for this particular application but it's less intimidating than it was 10 years ago. We're seeing more and more labs start to adopt this approach, particularly for wastewater surveillance.
Christina Bouwens (12:20):
Also say, thinking back on why digital PCR is such a good technology for wastewater monitoring and more of a why the competency platform is so useful for wastewater monitoring is because we're not reliant on this tube of emulsions to do our digital PCR reaction. We're actually looking at all of our partitions on a flat fixed plate. We can actually reject anything that doesn't look like a true positive partition.
Pia Abola (12:48):
I wanted to pull it back to wastewater surveillance and ask a little bit about the technical challenges. I would imagine that with qPCR, I would be really worried about false positives in something like wastewater surveillance. What are the challenges that digital PCR has, or maybe what are the advantages it has for wastewater surveillance?
Christina Bouwens (13:17):
One of the major advantages that digital PCR has for wastewater surveillance is that it is fairly tolerant to a lot of common PCR inhibitors. Wastewater is a really complex matrix. It's from sample to sample and from site to site, it's going to vary significantly. Digital PCR overall is a technology and because we're reducing those reactions sizes down, it just fundamentally it's more tolerant to inhibitors. That just makes it really nice for this application.
Pia Abola (13:48):
What does it look like when you do the monitoring, the sampling of wastewater? Is your company going out there and actually going into manholes and pulling up samples or does somebody ship it to you?
Adam Langston (14:09):
Fundamentally we are the technology provider. We supply the technology required to actually get those numbers from those samples. Our users, there is some variation in how the sampling is done. I think the wastewater community is working to develop best practices, composite versus grab sampling and how it's filtered and concentrated. We're starting to see some standardizing of that approach but we are not directly involved in that aspect. We do work in many cases very closely with the scientists that are using our technology and helping them to optimize that process and optimize sample prep before those materials go on the absolute queue platform. We are not directly, we are not... I'm not diving down into manholes, collecting wastewater myself but we do work pretty closely with the scientists that are involved in that.
Pia Abola (15:10):
Does your technology, your platform have challenges with all the different sample prep or when it comes to you, when the sample comes to you, is it all pretty much uniform?
Robert Lynn (15:26):
That's a great question. I would say in general, sample processing, from raw sewage to the sample that's on the platform, it comes out reasonably clean. Really the idea is to get as much assets as possible. By the time it gets onto the platform, we're looking at you, like that's a sample, not relatively pure. Going back to your question and what Adam said, people are doing it very differently. Different sides, it really depends on what access they have to... You may have access to man holes, you may get water from waste treatment plants.
Robert Lynn (15:59):
Really a lot of variation comes from really upstream. That's something interesting from our experience. Talking to different users and different people out there, really that is very different. Your sample type is different, your processing method is different, your concentration method might be different. A lot of it is often limited by availability. Because now a lot more people are doing this than before so people have to find ways that they can do this.
Pia Abola (16:28):
Does that affect at all your ability to provide an answer? It sounds if you can get down to really low levels of detection, you can always return pretty much a number of detected... What do you return to them, number of detected molecules?
Christina Bouwens (16:47):
Yeah, that's really interesting. The space has really developed. Since the beginning of the pandemic to now we return a report of copies per reaction, which you can then extrapolate depending on your upstream processing methods. Typically I think people report it in genome copies per ml. It's like per milliliter of wastewater, we saw this many targets. In the very beginning, we were actually using the CDC recommended targets for clinical sampling. We're looking at the N1 gene and the N2 gene. N1 and N2 regions of the N gene of SARS-CoV-2 and the human RNSP. We quickly figured out that that was not the right target for wastewater. We've been working with a few different targets and since this is another advantage of digital PCR and having many colors to look at is in one reaction, we're probing for the N1, N2 and recently we've included controls, like the human fecal control PMMOV.
Christina Bouwens (17:48):
Sometimes you can include a process controls to give you an idea of how your upstream processing varies from site to site. We're using [B Kovi 00:17:58] Using all four of these targets in the same reaction can really give you insight into not only the concentration of these target viral molecules but also I can help you relate these samples back to other populations. Using some of those controls, you can assess for inhibition rates because even though digital PCR is very tolerant to inhibitors, it's not completely immune. Having this information can give you a more clear picture of what you're looking at. Even if you're not detecting viral molecules, you still have some information that you can get from most every sample
Pia Abola (18:30):
I wanted to ask you, could you expand on that? I'm not quite sure. What is that telling you? You've got... it's tellIng you that you're amplifying with the human sample. You should always expect to see that, is that right?
Christina Bouwens (18:45):
Yeah, PMMOV there's a lot of different targets that you can use them. These are just viral targets that are really commonly spread or commonly shed in unexpected in our wastewater. What you can use these controls for is first to just make sure everything's working right. If you have extracted nucleic acid from wastewater and you're not seeing any PMMOB, maybe your process didn't work at all. The other thing that it can tell you is how many people contribute to that size. We've heard a lot about how great wastewater can be for almost like pseudo contract tracing. Where you can really identify different locations, either very broad, on the city scale, all the way down to individual streets.
Christina Bouwens (19:30):
To tell the difference between, I saw 10 copies of SARS-CoV-2 in sample A and 10 in a sample B. It's really important to know that the size of the community that you're looking at. If it's 10 and just a tiny street, that's a lot more concerning than a 10 across the whole county. That's what that human normalization control can be for. It can give you a little bit more context about your sample. Then like Rob was mentioning, there's a high amount of variance in how you process the samples. Not only how you collect them but do you concentrate them? What methods are using for extraction? It varies significantly because so many people are doing it and people are using their... all their tools and everybody's tools vary.
Christina Bouwens (20:16):
I'm using process controls, I like something that shouldn't exist in normal wastewater. B Kovi can be really helpful. You can put it in at the beginning and then use that amount at the very end and get a readout between sample sites or different sample processing steps and how much you vary. It doesn't necessarily have to inform... it can just be informative in the sense that you know the level of variation between things. It can help you also to just provide an inline orthogonal measure of efficiency of those types of methods.
Robert Lynn (20:50):
For sure, like Christina mentioned we started using the CDC targets and that really is for human samples. That's actually how we started but as we engage with more researchers, as we learn more about what people are doing and what the challenges are, we realize as a small company, we can be agile and really look to improve and try to serve the need of the community. Christina spoke to the two of the most very important things. One is the human fecal load normalization. That's in our assay, the B Kovi [bovi 00:18:30] Corona virus as a process cultural and that's the only in our assay. The assay as it currently stands, looks for SARS-CoV-2 targets along with one human fecal control and the process control. Really that's leveraging the full capability of the instrument to be able to provide as much information as possible in that one single sample.
Pia Abola (21:47):
How many different targets can digital PCR monitor?
Christina Bouwens (21:52):
There's so many, I think I would say the advantage of digital PCR is that it's very adaptable. While we were looking into the targets to include in this assay, as I said, there's a lot of those human normalization and even the process control. Lots of people have lots of different... they have lots of different preferences and suggestions. Even for the engine for qPCR, there's many, many targets on SARS-CoV-2 that you can look for. The nice thing is at the heart of it, it's a really simple qPCR reaction with a lot more information. That's one of the advantages is we can be really flexible to include targets because at the end of the day it's just PCR. That's a really good advantage for digital. Right now we look at four targets. I think once you start going about four the assay design can get complex for making sure that things are specific and sensitive enough to actually give you the information that you need. Four is a really good balance.
Robert Lynn (22:51):
I would say, fundamentally the multi-pattern capability really depends on the architecture of the technology. It's not just your digital PCR, you're talking about qPCR, how many can you multiplex well. There are things that can inherently increase the multiplexing capability of a given reaction. A lot of those can be applied to digital PCR as well.
Pia Abola (23:14):
I'm so curious about variants It's one of the biggest things to worry about now, the prevalence of all these variants. I would imagine with the ability to monitor multiple targets, you can track different variants. Is that something you're doing or you're looking at?
Adam Langston (23:34):
It's certainly something that we're looking at. I can't speak too extensively at this moment because some things are in their early, early efforts. Theoretically speaking, if we're looking at snips or single nucleotide polymorphisms, digital PCR is very capable of differentiating and quantifying those different targets. There we are having conversations with researchers around the country and internally about expanding our capability with regard to quantitation of SARS-CoV-2 variants.
Pia Abola (24:10):
I know that you can't talk about who your customers are, the researchers you work with. Do you have any ideas of how they're using this information? Is it really just monitoring dorms or how does this information get used?
Adam Langston (24:28):
Sure. Yeah, I think we have several different users. We have a number of users in academia, certain municipalities are using this information as well. There is some variation, some state government agencies in addition to smaller local municipalities. I think it's one customer that I can speak about is the University of California, Santa Cruz. They have been using this technology for this very particular application for a number of months. They're also helping with the city of Watsonville in central California.
Adam Langston (25:04):
Not only are they using this approach to monitor the prevalence of Covid-19 on campus but also a larger scale in a community of around 50,000 people to monitor, to try to get an idea of what their hospital situation might look like in the next two to three weeks. There is a handful of different use cases but those probably capture two of the most prevalent.
Pia Abola (25:30):
I was hoping you could expand on those a little bit. You mentioned that you can use it to anticipate hospital occupancy?
Adam Langston (25:43):
Yeah, I think being able to monitor the prevalence of SARS-CoV-2 in wastewater gives you insight into what is happening within the community without relying on positive clinical tests. That can give you a preview as to what is happening within the community before you start seeing hospitals fill up. It's a precursor dataset, if you will.
Robert Lynn (26:07):
No, I think that's absolutely fair. It's not like we're saying, if you see an uptake in your wastewater two weeks later, your hospital beds are going to be full. This really is a way to get an aggregate measurement of the level of virus that's in the community. There has to be multiple and this is not the only case that this can be seen as a leading indicator of the amount of cases that might be present in the community.
Pia Abola (26:35):
I'm curious, now that we've got the vaccine heading out into bigger distribution, are you seeing or do you anticipate any changes to your testing? The wastewater surveillance testing business?
Adam Langston (26:52):
I don't think we're going to see this stop anytime soon. I think as I had mentioned earlier, this has really shined a spotlight on the capabilities of not only digital PCR but also wastewater surveillance at large. As Rob mentioned earlier, this can be expanded to monitor other infectious diseases as well. It's a really effective tool that can be applied to other diseases. I should also say that it's expanding to other areas of science as well. I know for a long time, environmental scientists have looked for ways to study environmental DNA or eDNA in many cases for invasive species.
Adam Langston (27:33):
This has actually opened the door to using digital PCR for that particular approach as well. There are now some scientists who are actually looking for SARS-CoV-2 in the great lakes. You talk about a needle in a haystack application, in that case it's considered environmental DNA and digital PCR is incredibly well-suited for that as well. It's certainly expanding what many scientists perceive as the capabilities and benefits of digital PCR.
Pia Abola (28:04):
Thinking about the assay development and now that I'm in the right mindset that there's not all this emulsion. Do you have any different constraints or different freedoms for assay development than you would with emulsion based digital PCR?
Adam Langston (28:27):
The way our technology works, the instrument itself that actually processes your sample is entirely dry. Everything is self-contained on a consumable plate, a consumable device. In developing this technology, we really focused on being as reagent agnostic as we possibly could be. We've added some additional flexibility into digital PCR because at the end of the day, we truly are, I'll say agnostic. We are able to effectively put over existing qPCR assays and chemistries onto the absolute queue platform. That was a major focus in developing the technology and we feel we've been pretty successful in that effort.
Pia Abola (29:13):
What do you think is the most important thing for assay developers to understand about wastewater analysis using digital PCR?
Christina Bouwens (29:25):
I think in the last, I would say, four or five months, it's been really clear that having thoughts about consistency and how to use controls effectively to compare our data is going be more important than ever. There are so many variables that can be involved in with the assay that we've developed, we've addressed as many as we can in this assay at this time. There are different ways to approach it and other ways to approach it. I just think controls, controls, controls are always my number one thing as a scientist at heart.
Christina Bouwens (30:02):
Just thinking about how we use those controls upfront so that we can make the most of the datasets that we have right now because it is an incredible data set. We're never going to have as much joy as we have right now or we'll have that in the future. If we can think about controls upfront, we can learn a lot more just for response to infectious disease in general. Yeah, that's what I would say.
Robert Lynn (30:26):
I think that's an excellent point [inaudible 00:30:31], especially wastewater is a very challenging matrix. Not to say others are not complex as well, but wastewater really is different beast. Really having, like Christina said even more controls. You just understand how this matrix is affecting your reactions is really as critical. If you don't have a control, you could just see everything as negative, it could just be because you're incredibly inhibited and nothing's going to be positive anyways. Being able to know that it's going to be incredibly important to ensure you have accurate results.
Pia Abola (31:08):
What do you think clinicians should know about wastewater surveillance? Should we be doing more of it?
Robert Lynn (31:14):
I don't know if we were quite qualified to tell clinicians what to do. I will say, based on our experience, talking to different people, this really is a very useful tool. It's not going to be as precise or pinpoint as testing the individuals and deciding whether or not a person needs to be hospitalized or no, this is just a cough, you can go home it's going to be fine. Really having the ability to monitor on population level and make more broad strokes decisions. I think this is a very useful tool. On the larger scale decision level, I think wastewater monitoring can be very, very powerful and very cost effective.
Pia Abola (32:00):
Oh yeah. It's really more a question. Not the clinicians who need to know about this but public health officials. This is really a good tool.
Robert Lynn (32:10):
Absolutely. As with everything else, the information is there and how the information is used is going to have to be a case by case basis. In terms of providing valuable and valid information about a community, I think wastewater monitoring is a very powerful tool to do that.
Pia Abola (32:28):
Thank you for joining us for this episode of Countering Covid. If you're interested in additional resources on this topic, you can check the show notes or visit biosearchtech.com/covid-19, that's biosearchtech.com/covid-19. Thank you.