Detect Early, Treat Early: Innovations Shaping the Fight Against Cancer

Sonia Sennik: and I’m Sonia Sennik, CEO at Creative Destruction Lab.

John Stackhouse: This is Disruptors x CDL: The Innovation Era.

Sonia, it’s April, and that means spring is maybe not here, but around the corner. And of course, we’re seeing daffodils both in some parks, but also on people’s lapels, which is a good reminder that April is also Cancer Awareness Month and Cancer Awareness Month. And all those daffodils, if you weren’t aware, were introduced here in Canada way back in the 1950s as a symbol of Spring of Hope, and of that great slogan that cancer can be beaten.

Sonia Sennik: John, cancer remains one of the leading causes of death worldwide with one in five people expected to develop cancer in their lifetime. But in Canada, the number is even higher.

At about two in five Canadians are projected to face a cancer diagnosis. Seeing the treatment and disease management process up close and personal, I’m [00:01:00] sure is something that many of our listeners have experienced in some way, shape, or form.

John Stackhouse: Yeah, I suspect we’ve all been through excruciating aspects of cancer, both in our own families as well as social circles, and there’s no other word.

It is excruciating on the patient, first and foremost, but also on the families and support networks of everyone who endures and suffers cancer. It’s also extraordinary and inspiring to me always to see the quality of care in this country. It is getting better just as the diagnosis of cancer is getting better, not nearly fast enough, but one of the things I love about that slogan, cancer can be beaten is its expression of hope.

The Canadian Cancer Society doesn’t say it will be beaten, but it can be beaten. If we all do something about it, and we’ll hear on this episode some of the amazing things that all of us can take advantage of with technology that don’t get us there to defeating cancer, but certainly give us all a better [00:02:00] shot.

Sonia Sennik: The Canadian Cancer Society works with us, John here at Creative Destruction Lab, with a vision of putting together an early stage program focused solely on cancer prevention treatment and survivorship technologies. Today on our podcast, we have one of our mentors, Jesse Salk, as well as one of our CDL alumni graduates.

Peter from OxCan Analytics. These folks are dedicating their lives at the forefront of innovation and technology, of improving the experience of oncology patients around the world.

John Stackhouse: And I love it that CDL has embraced cancer as a pursuit for innovation for anyone who’s listening, who is thinking about a career at innovation, and you can start that at any age.

It doesn’t need to be just about the next delivery app. Whether you’re a technologist or a marketer or a digital program manager, you too can join the battle to defeat cancer and do it through innovation.

Sonia Sennik: Absolutely, John. Coding to cure cancer. We’re gonna hear from our guests today about how artificial intelligence is playing a role in [00:03:00] transforming oncology prevention treatments and survivorship.

John Stackhouse: We’ve got a great episode, so let’s get at it.

Sonia Sennik: First, we’ll hear from Andrea Seale, the Chief Executive Officer of the Canadian Cancer Society, which is the country’s largest national charitable funder of research into over a hundred types of cancer. The Canadian Cancer Society is committed to uniting and inspiring Canadians to take control of cancer funding high performance research that improves cancer outcomes and addresses the greatest opportunities for progress.

Canadian Cancer Society is also the founding partner of our CDL Cancer program, supporting innovators at the cutting edge of science.

Andrea Seale: Progress in science and technology is helping us to improve cancer survival rates decade after decade. And when we consider emerging technologies, we could talk about genomics, ai, radio, theranostics immunotherapy, and the Canadian Cancer Society is funding really important development in all of those areas.

But some of the most exciting [00:04:00] potential I see on the horizon is about finding cancer earlier. So let’s detect it when it is stage one or even sooner. And today we do this through some of our healthcare screening programs. So mammograms, scans, fit tests, but science is allowing us. To see evidence of cancer in smaller and smaller increments of material like in the molecules in your breath or your urine, or even fragments of tumor DNA that are circulating in your bloodstream and seeing it early gives us better treatment options, and this is really what we all want.

I was in a lab that’s using breathalyzer technology to try to identify a molecular signature for lung cancer and. This is a great example of a simple, portable, inexpensive technology. If it’s like that, it could let us screen more people, screen at younger ages, screen outside our big cancer centers.

Canada’s such a big country. Imagine if eventually we could have at [00:05:00] home cancer diagnostics. Less pressure on our medical system, we could avoid more invasive time consuming surgeries for patients. It’s really promising, and it’s so important that we work together on this

Sonia Sennik: Now we welcome Dr. Jesse Salk, a pioneer in cancer innovation, whose groundbreaking work is reshaping how we detect and treat cancer. Jesse is the co-founder of Twin Strand Biosciences, a company dedicated to enhancing the accuracy of DNA sequencing through revolutionary duplex sequencing technology.

Jesse, welcome to the podcast.

Jesse Salk: Thank you very much,

John Stackhouse: Jesse. Before we get going, we should point out to our listeners that in addition to that amazing resume, you’re also the grandson of Jonah Salk, who developed one of the first successful polio vaccines. I wonder if you can give us some insight into. Your grandfather and how that inspired what you’re doing today.

Jesse Salk: Well, I knew him growing up as a child and to me he was just a grandfather like [00:06:00] any other who brought presents and played and that sort of thing. I think in sort of retrospect now being a scientist and a physician, one of the things that I say I’m most proud about being related to him is the mentality that I remember.

He tried to pass on to me that you’re only as good as your legacy, and so it’s about what one does for others and the memories you leave and the things you do. Beyond yourself that are really the most memorable and lasting.

John Stackhouse: What a beautiful legacy for him to have left to you in addition to the extraordinary legacy he left to humanity.

Tell us a bit before we get into your company and your work, the sort of legacy that you’re trying to build.

Jesse Salk: I am a physician scientist. I came out of academics and I spent many years developing basic science tools and research and publishing papers. One of the things I learned early on was that there’s a huge amount of power in academics, but there’s also a lot of limitations in terms of the scope and breadth that one can deploy these new [00:07:00] technologies.

And so I developed a technology and started a company called Twin Strand, which was really based around that intersection of tools for scientists and, uh, things that can benefit patients and customers globally.

Sonia Sennik: Jesse, for our listeners who may be unfamiliar with the term duplex sequencing, how does that differ from traditional DNA sequencing methods?

Jesse Salk: Duplex sequencing is a technique I developed with colleagues from the University of Washington more than a decade ago that uses special biochemistry and special informatics to significantly improve the accuracy of DNA sequencing. So normal DNA sequencers work pretty well and have an error rate of around 1%, and duplex sequencing drops this to below one in 10 million to allow detection of extremely rare variants for applications like detecting the presence of residual cancer after a curative intent treatment or detecting the mutagenic [00:08:00] signature of a chemical in the environment. That’s a potential carcinogen. And things along those lines, really extreme use cases.

Sonia Sennik: So the more accurate the DNA sequencing, the better the outcomes. How does your technology impact the lives of patients?

Jesse Salk: I originally developed the technique with colleagues when we were studying the formation of cancer and early cancer processes, meaning things that humans are exposed to, either related to their normal endogenous aging process, or chemicals in the environment that can mutate, DNA, can actually change it.

And although we’re familiar with being. Born with a certain set of DNA and thinking about that being immutable during life. That’s not actually quite true. Every cell in our body undergoes a very small number of genetic changes with time, and some of these are the things that ultimately lead to cancer formation.

So this technique allows better understanding of early cancer formation processes and the things that drive it, potentially allows better [00:09:00] early detection of cancer when it’s more treatable. Allows detection of relapse of cancer early when something can be done about it. Cancer’s tricky. It’s not a disease that’s from an inherited single gene or a single virus or a single bacteria that causes it.

It’s a heterogeneous disorder that’s sort of interplays and is intertwined with aging. Our bodies naturally develop mutations and we have ways of preventing those from growing, but eventually some of them can let cells grow and expand into cancers, and we. Know that we don’t go from a state of normal to a state of cancer overnight.

This is a gradual process and better ways of resolving how those mutations occur, what caused them, and what allows the cells to carry them to grow and begin that early cancer process formation will hopefully lead to better insights for better cancer prevention and early detection.

Sonia Sennik: Jesse, you spent quite a bit of time with us this year in our CDL Cancer program.

I’m [00:10:00] curious to know what innovations you’re seeing that are exciting you the most.

Jesse Salk: There’s quite a lot, uh, in the cancer stream around artificial intelligence, whether that’s interpreting pathology or radiology slides or developing new drugs. So that’s an exciting area. That’s, um, something that I’m definitely anticipating is gonna really be impacting, uh, science and medicine going forward.

There’s also a lot of focus I’m seeing on health economics and ways of taking technologies that might be cool but are also expensive, and not just throwing more expensive solutions at our problems, but taking some of the problems we have and looking at the most cost effective and, uh, realistic ways to get the most out of the tools we do have.

I think the breadth of people in the room, both companies and mentors is enormous and it’s always a pleasure to participate and see something. I learn something new each session.

Sonia Sennik: Geoffrey Hinton, years ago, I think in 2017, famously said, [00:11:00] in five years there’ll be no more radiologists. I think each of us probably know a radiologist and radiologists are very much still in business.

What does the pathway look like for AI to start making really tangible changes when it comes to cancer prevention or treatment?

Jesse Salk: I think AI is obviously a tricky term because it’s not quite the same as human intelligence. I think there are things that many of the tools we have now can do better than humans.

A lot of tasks that I tend to forget or my colleagues would rather not spend their time doing so we can spend our time thinking and focusing and more sophisticated. Management decisions and strategy. Like I said before, cancer is an incredibly complex disease and we’ve over the last decade, developed more and more new tools for increasing the depth and breadth of data streams for measuring different molecular happenings in cancer and other diseases.

I think taking very complex disparate signals and integrating together using AI to create [00:12:00] models and learning that we can use to, from that information to predict outcomes or identify specific choke points where we can therapeutically intervene is probably one of the most relevant things. So it’s using.

Pattern matching that we as humans are not necessarily good for, to get additional insights into some of the data we have from new technologies that already exist and those that are coming on down the road.

John Stackhouse: Presumably, Jesse, you could do a lot of this in your lab, and even as an academic, you’ve chosen to create your own company, Twin Strand Biosciences. Tell us a bit about the origins of the company and what, as a scientist you felt you wanted to do also as an entrepreneur.

Jesse Salk: So I spent many, many years in academics, so I really, uh, appreciate that and understand the importance of academics.

A little kid that asked me the other day, what grade I was in after he told me he was in second or third grade, and I [00:13:00] thought it was sort of funny and I thought about it and you know, I realized that I had actually graduated in the 29th grade if you had a medical school and college and residency and fellowship.

And so I’ve been at that for a really long time. So, although on one hand I really am motivated by advancing science and teaching and, and taking care of patients, I also found that there’s a lot of challenges being able to deploy new technologies at the scale and scope and to the number of people that I, you know, would, would always want to be able to do.

And so starting a company was scary but exciting opportunity and I learned a whole lot in the process. You know, I think the future of advancing science and medicine is always gonna come at the interface of academics. And commercial, whether one is fully on the academic side or one is on the commercial side, there’s always a crossover point, and that’s where that intersection of, uh, innovation comes from the most.

Sonia Sennik: And Jesse, how do you balance being an entrepreneur and a practicing oncologist, [00:14:00] and how does each role inform the other?

Jesse Salk: It’s challenging, but just like, uh, anybody does multiple things. It’s challenging. As an oncologist, I still see patients half a day a week. I take care of oncology patients over at the local VA here in Seattle, but I also teach, I supervise residents and fellows and medical students and, and teach them how to be better doctors.

And as an entrepreneur, starting a company, I teach customers. I teach scientists at the company and I teach investors about the technologies and the opportunities we have. I think there’s more overlaps than differences, but it is challenging. It is challenging context, switching from one thing to the other, but just like context switching, running a lab and being a parent, or being a doctor and taking your kids to a soccer game, these are just part of life and one has to make it work.

John Stackhouse: Jesse, switching back to cancer research, what recent developments in cancer therapies maybe excite you most and [00:15:00] where do you think that will take us?

Jesse Salk: I think one of the biggest thematic changes in cancer therapy over the last, uh, 10 years that I’ve seen is the development of many therapies focused around harnessing the immune system.

The immune system is this amazing, adaptable, evolvable system that’s, you know, taken billions of years to, to get to the state where it is, and it’s really powerful for adapting to new changes and new threats to our body. And one of the ways cancer cells avoid being cleared out by the immune system is doing certain tricky molecular changes that hide.

And so many of the new treatments that we’ve developed, uh, checkpoint inhibitors and CAR T cells and other immunomodulators have really, uh, begun. To address those vulnerabilities and weaknesses in cancer, let the aspects of the immune system work really well overcome these relatively straightforward changes.

And although we’ve [00:16:00] made major progress for coming up with treatments that can be sometimes curative in a stage four setting, which was never possible before. For many cancers or being, uh, vastly more tolerable than past chemotherapies. I think there’s an extraordinary number of opportunities ahead of us, and when we look back in 10 years, I’m sure we’re gonna say that even where we’re at right now that I just said, we should be so proud of that technology of the present is actually probably gonna be medieval compared to where we are in a decade.

So I’m really excited to be in oncology because of that.

Sonia Sennik: Jesse, for any of the entrepreneurs, scientists or epidemiologists listening, what advice would you have if they’re looking to make a difference in the field of biotechnology and healthcare?

Jesse Salk: I think the most important thing is to do what you wanna do, not what you think you should wanna do.

Take the things that you’re passionate about and find ways to use those to not just create papers or a reputation for yourself, but get them out in the [00:17:00] world. And that means working with colleagues in academia, it means starting companies or working with other companies. It means being kind and creative and generous to the broader community that we’re all part of.

As many folks have said, we all stand on the shoulders of giants and are only able to move forward because of the past work behind us. And so be sure to respect the past scientists that you’ve come from and pass it on to the next generation as you move forward and make your legacy in the world.

John Stackhouse: Jesse, that’s truly inspiring.

Thank you for being on the podcast.

Jesse Salk: Thank you for having me.

Sonia Sennik: We’re now joined by Peter Liu, co-founder and CEO at Oxford Cancer Analytics, a company at the forefront of early cancer detection using advanced machine learning and blood-based diagnostics. Peter is also a Creative Destruction Lab alumni founder, graduating from our CDL Cancer program in 2024. Peter, welcome to the podcast.

Peter Liu: Thanks very much, Sonia.

Sonia Sennik: So Peter, let’s start with your [00:18:00] founding story. What inspired you and your team of researchers to launch Oxford Cancer Analytics?

Peter Liu: So I came from a clinical medicine and cancer research background, being trained as a medical doctor in Toronto. And then, uh, having completed my PhD in Oxford, I’ve been focusing on cancer innovation for the past 13 years.

The first half of my career, I focus on novel, innovative treatments for cancers and also the mechanism to which cancers are initiated. But during my, um, clinical experiences, I started realizing actually. When detected early, a lot of cancers can be subject to, uh, treatment with curative intent. But unfortunately, if you look at a lot of the, uh, the deadliest cancers, a cancer that unfortunately kill the most normal people, those are often detected too late.

For example, lung cancer is the leading cost of cancer deaths worldwide, and over 70% of patients are often detected at the late stage when this can be a death sentence for many. Whereas you’re able to [00:19:00] detect this cancer early, there’s a significant increase in those who are able to survive beyond five to 10 years or more.

And that’s why I started shifting my focus on the, uh, early cancer detection front and the idea that a simple blood test is able to pick up materials in the blood predictive of cancer is very exciting. The fact that we’re able to detect many cancers in a minimally invasive manner using the same kind of blood tests that any of us could have done at the family doctor or GPS office for early detection.

The ability to fundamentally transform the way that we get approached cancer prevention from one that’s reactive to proactive. I think when it comes to, um, innovation nowadays, especially in the health tech space. There needs to be intersection amongst three different perspectives. One is the clinical utility.

Second is the scientific feasibility, and thirdly is a commercial impact. You need to get all three of these right, intersected at the right [00:20:00] point. Having seen at least two outta three of these, um, from the clinical medicine partner with the research part, I’ve realized that what may be clinically feasible and promising may not be scientifically feasible and vice versa. A very innovative scientific idea without being fully integrated with existing clinical pathways may often have trouble being able to reach the hands of patients. The ability to bring innovation that can benefit patients, that can benefit clinicians with a feasible scientific idea will not reach the hands of patients we need the most without that commercial support.

And that’s where we decided to take a commercial approach. So with a group of highly skilled scientists, physicians, medical researchers, regulatory experts, and commercial experts we’re able to bring these innovations to hands of patients who need it most.

Sonia Sennik: And maybe just in the simplest terms, Peter, how does OXcan’s liquid biopsy work and how is it different from other diagnostic tools?

From a patient perspective? [00:21:00]

Peter Liu: Yeah, for sure. Uh, Sonia and John, when was the last time, uh, you were at a family doctor’s office for a blood test?

Sonia Sennik: Within the last few months. I’m a great citizen. Peter. John, what about you?

John Stackhouse: It would’ve been last summer.

Peter Liu: Well, you know what? For me it was just a couple months ago.

It’s the same kind of blood test you had done there that can tell you whether you are at a higher risk of cancer and whether we would recommend you to proceed with further confirmatory testing. The reason why we have decided to go with a blood-based approach is because it can be very cost effective, is fully integrated with existing pathways.

All the infrastructures are already in place. The challenge is really knowing what to test for in the blood and whether to, um, identify whether someone is at an increased risk of having cancer. And that’s why a lot of our R & D is focused on analyzing within the blood to identify the risk of cancer.

So all in all, it sounds very simple, simple blood test that many people would’ve experience, but the science behind it, it’s a lot more complex, but requires a lot of hard [00:22:00] work.

John Stackhouse: Well, simple is good, and as a male of a certain age, I get that mailing here in Ontario from the health ministry for a prostate test, which I’m glad to do.

But point being, we should all be doing more, and the more that the system does to make it easier for us, probably the more uptake there’s gonna be.

Peter Liu: Yeah, precisely and very similar to, um, the colorectal screening guideline. There is an existing screening guideline for lung cancer, which is a leading cause of cancer that’s worldwide.

But fortunately, many countries, including the US has reached some hurdles in terms of people adopting it. There is, uh, an exposure to radiation. Plus there’s also cost and there’s only so many people you get screened through these gigantic donors. So that’s where we come in as through a simple blood test.

And that’s why actually, uh, we’ve been doing quite well in terms of working with some of the, um, the largest, uh, screening programs for lung cancer in order to better triage and to allow more patients to be screened, uh, [00:23:00] faster at a lower cost. So that, and we can detect, um, more lung cancers. Earlier.

Sonia Sennik: Peter, cancer’s such a personal and emotional disease for many. How has working so closely on this issue affected your outlook, either professionally or personally?

Peter Liu: Yeah, Sonia, I grew up in, uh, Calgary, Alberta, and I remember when I first started my undergrad, I was volunteering at the Tom Baker Cancer Center working with quite a lot of families and patients who were struggling with cancer. What I saw were each individual human being, each with their own identity, their dreams, and I saw cancer as a disease that deprived people of their fundamental identity, often altered people’s personality and to behave in a way that were not them, but also deprive them of their, of some of the most fundamental dreams and hopes.

So these encounters ground [00:24:00] the purpose of what I did, and that’s how I started Paths to Cancer Innovation. Fast forward over a decade later, I continue to be motivated by these day one stories, uh, from patients. I have also lost some very close family members, uh, of my own, uh, two cancer, including lung cancer when I was developing this company.

So not only this personal, it’s also meant for, uh, other families and people who are. Currently fighting cancer, but also people who are focused and will continue to fight cancer. In a way, it’s unfortunate that my own family member would not be able to benefit from this technology, but I hope other families and other people will be struggling with cancer will have the opportunity to benefit from this, the whole ambitions to drastically transform the way that we detect cancer and treat cancer to enable more treatment, security attempt to save lives.

John Stackhouse: Peter, as we move towards close, I wonder if you can share a bit about what’s next for OXcan and, and [00:25:00] maybe also give us a sense of where you see AI taking your company, your work, and the broader field.

Peter Liu: Yeah, so AI at its core is a tool to help us analyze data and get an output from it. So what’s important is what goes inside of it and what comes out of it.

So AI is only as good as the data you put in there. So that’s why at Oxford Cancer Analytics, we have formulated a new generation of proteomic technology so that we can unravel previously unseen data in the blood for the first time so that we can feed this data into the machine learning algorithms so that it can tell us what new biomarkers can be, uh, used for early cancer detection.

And, um, in terms of what comes out of it, uh, it’s also important because you need to control these parameters in a very careful way. I think a lot of times people may misuse machine learning, people may misinterpret the data, uh, coming out of it. And this is especially [00:26:00] important in the field of clinical medicine and also early cancer detection because it’s one of these fields where there’s an increasing need for humans to be extra prudent in terms of AI use.

When we started back in uh, 2020, people were only beginning to try to use machine learning in this area of data analytics. We realized that actually a lot of the machine learning used at the time were not necessarily tailored appropriately. For example, we used a lot of high dimensionality and low sample size data, whereas a lot of machine learning were meant to be applied to millions of people compared to the hundreds of thousands of, uh, sample size that were dealing with in clinical medicine and science.

So we actually had to directly design a lot of the machine learning models from the ground up in order to tailor them towards getting the most. Out of these data in terms of the future, we have further built in an explainability component to address a lot of the concerns that people have for AI, especially in the, um, clinical medicine and regulatory domain.

Some of the, uh, more modern but complex machine learning models [00:27:00] can be seen as a black box. What we have done is actually, we added an explainability component. So the AI actually talks back to us telling us exactly how it’s made that decision. So not only I think this patient has cancer, and I think these are the important biomarkers, it actually tells us, this is why I think this patient has cancer, and this is how I made that decision at.

And we have to pioneer a lot of new generation machine learning models, explainable AI models in order to best utilize this tool in the most responsible and suitable manner to maximize what we can get. From these innovative proteomics data for early lung cancer detection, I’m highly optimistic about AI.

I think it’s inevitable that AI will become an everyday part of our, uh, work and life, but also, especially certainly in our field, there’s an increasing need to approach it with responsibility and prudence. Uh, in terms of what’s next for Oxford Cancer Analytics. We’re excited to launch our product in the next year, starting from the US North America, UK, and EU, and hopefully being able [00:28:00] to benefit, uh, more patients globally.

Sonia Sennik: Thanks so much for joining us on the podcast.

Peter Liu: Thank you very much, Sonia and John.

Sonia Sennik: In the words of Coldplay’s, Chris Martin, it was all yellow. Daffodil month is upon us, and we’ve heard from some amazing visionaries and innovators in the cancer space.

John Stackhouse: There was so much to learn from in that episode, and sometimes cancer can seem overwhelming, but there’s a few very simple things all of us can do, not just in Cancer Awareness Month, but through the year.

Number one is to talk about it. This isn’t a secret that we should tuck away. It should be a common part of our conversation so that we’re all learning and sharing and supporting. We could also spread awareness through those daffodils and donating our time and money. And then lastly, perhaps most importantly, get tested.

No matter who you are, no matter what age or stage, there’s no real good excuse for not testing yourself and helping others get tested for cancer. That’s how collectively we can all live up to that motto of beating cancer. [00:29:00]

Sonia Sennik: John, one of my favorite parts of the CDL Cancer program is our patient contribution group.

So we’re piloting a new structure where we have patients in the room, in our CDL sessions with the innovators and our mentors and scientists. Engaging firsthand in technological development. Having the chance for the patients to share their stories and provide input to our technologists at these early stages is a really rare and special experience to observe.

And what I’ve learned is no two cancer journeys are the same. So as you mentioned, educating yourself, getting tested early and contributing to our ecosystem that has incredibly smart innovators that can tackle this problem, is an opportunity for all of us.

John Stackhouse: That’s so well said, Sonia. No two journeys are the same, so whatever yours is, don’t be afraid to share it.

This is Disruptors, an RBC podcast. I’m John Stackhouse.

Sonia Sennik: And I’m Sonia Sennik.

John Stackhouse: Talk to you [00:30:00] soon.

This article is intended as general information only and is not to be relied upon as constituting legal, financial or other professional advice. A professional advisor should be consulted regarding your specific situation. Information presented is believed to be factual and up-to-date but we do not guarantee its accuracy and it should not be regarded as a complete analysis of the subjects discussed. All expressions of opinion reflect the judgment of the authors as of the date of publication and are subject to change. No endorsement of any third parties or their advice, opinions, information, products or services is expressly given or implied by Royal Bank of Canada or any of its affiliates.