Breakthroughs in Technology for Cancer Screening and Cancer Therapies - A Discussion Francis deSouza

December 23, 2021

Charles Ryan and Francis deSouza, discuss the future of cancer care, Illumina, and its recent acquisition of Grail. 


Francis Aurelio deSouza is an American entrepreneur and business executive. He is the president and chief executive officer (CEO) of Illumina.

Charles J. Ryan, MD, is the President and CEO of the Prostate Cancer Foundation. Dr. Ryan is an internationally recognized genitourinary (GU) oncologist with expertise in the biology and treatment of advanced disease as well as the supportive care of all men with prostate cancer. He was most recently the Director of the Hematology, Oncology and Transplantation Division in the Department of Medicine at the University of Minnesota, Minneapolis. He also served as the Associate Director for Clinical Research in the Masonic Cancer Center and held the B.J. Kennedy Chair in Clinical Medical Oncology.

Read the Full Video Transcript

Charles Ryan: Hello from the Milken Global Conference 2021.  We are in Los Angeles today and I am delighted to be joined by Francis deSouza, who is the CEO of Illumina. And we were at a panel today talking about the future of healthcare, the future of cancer care, and Illumina and its recent acquisition of Grail are making a lot of interest around the cancer community about the technology that brings the potential for cancer screening and cancer therapies. So thank you so much for joining us, Francis. It's a pleasure to talk to you again.  Tell us a little bit about your company and what the work with Grail is all about.

Francis deSouza: Sure. First, thank you for having me. We at Illumina sell products that do genomic sequencing. So we sell machines typically that allow our customers to put in biological samples. So that could be blood or saliva or plant material. And we read out what DNA, what genomic material, are in those samples. There's a whole set of applications that our machines are used for. They are used in the research community for people to try and understand more about how your genome translates into health and disease.

So lots of fantastic research is happening to try and connect genomes to health states and disease states. But we also use it in a number of applications. In the clinic, for example, our machines are used for non-invasive prenatal tests for pregnant parents to assess the health of their baby. We also use it in children's hospitals, as doctors use genomic data to diagnose children with genetic diseases. So about 5% of kids are born with genetic diseases and they use our sequencers to identify the disease. We also use it in cancer hospitals to match cancer patients with the right therapies and a whole set of other applications, including in the IT world, people are looking to store IT data in DNA.

Charles Ryan: Really?

Francis deSouza: So a whole set of applications. Now, Grail, and the screening test you talked about, is one application we are particularly excited about. And what it is, is a blood test that looks for 50 types of cancer in the human body. And so it's a test you do, and it can detect if a person has any one of those 50 cancers across stages, stage one to stage four. Now we know that 10 million people a year die of cancer, 600,000 a year in the US alone. And we also know that catching cancer early gives a person the best chance of survival, that even in some of the deadliest cancers if you catch cancer in stage one or two, you can have a greater than 90% chance of five-year survival. But if you catch it late, that can drop to less than 10%. So we know early detection is better, and that's what this test is about.

Charles Ryan: So the test is called the Grail Galleri test. And just to be clear, it is a cell-free DNA test that patients can have drawn, a tube of blood. It's just one tube of blood, that is all that's needed, correct?

Francis DeSouza: That's right.

Charles Ryan: And you said that it can help to identify the stage of cancer, but is the Galleri test identifying sort of a, yes, no, there is a presence of cancerous cells here, and the stage has to be followed up with subsequent imaging and other testing? Or are there correlations between the cell-free DNA that you can make directly to the stage of cancer?

Francis deSouza: Yeah. The Grail Galleri test itself not only tells a person if they have cancer or not, but it identifies the tissue of origin. And so the cell-free fragments, the tumor fragments that we see in the blood are very specific to organ type. And so this is a doctor-ordered test, say a doctor orders it for a person, and then the doctor gives you back the results. And the doctor typically will tell you either it's all clear or that there is a follow-up needed around this organ, a pancreas, for example. And that's when you do sort of the follow-on to, it could be imaging, for example, to actually identify the cancer.

Charles Ryan: Mm-hmm (affirmative). So give us an example of where the Galleri test is going to do something that we just can't do because we do not have other tests to do them to identify cancers. Probably some of the more rare cancers, correct?

Francis deSouza: Yeah. And some of the more common ones. What's startling even today is that 71% of all cancer deaths are to cancers where there is no screening.

Charles Ryan: Hmm.

Francis deSouza: So even though we know that, if you catch cancer early, you give a person the best chance of survival, the reality is we have screens for very few cancers. In fact, in the Grail Galleri test, of the 50 cancers it looks for, only five of them have other screens. And so you are talking about cancers like, for example, pancreatic cancer, which can be very deadly but has no screen. In fact, I was watching the testimonials coming in from people that have ordered the test over the last few months since it has gone public, and one that stuck with me was from a clinical oncologist. And she was saying how one of her patients had done the Grail Galleri test. He's a man in his fifties, presents as healthy, had just done a colonoscopy, had no symptoms but did it because a friend of his told him to do the test. And so he did the test, and she said they found a stage two pancreatic cancer. Now, in her career, she said she has never seen a stage two pancreatic cancer in a patient because you don't catch pancreatic-

Charles Ryan: Right, right.

Francis deSouza: ... cancer typically in stage two, because there are no symptoms.

Charles Ryan: Right.

Francis deSouza: Now, fortunately for this man, he was able to get treated, he is now healthy, and his prognosis is excellent. Now that's something she never thought-

Charles Ryan: Right.

Francis deSouza: ... she'd able to say in her career, somebody who had been diagnosed with pancreatic cancer is now healthy again.

Charles Ryan: But the algorithm went, he did the test, the test said there is evidence of pancreatic cancer here, he then goes to an oncologist or another physician who orders a CAT scan or other things, and they look and they scan the pancreas and they see it, and they stage it at stage two. You can't stop at the blood test and say, "You have stage two pancreatic cancer." All you can do is say, "There's probably evidence of pancreatic cancer. We need to do more testing." Right?

Francis deSouza: Exactly. That's exactly it.

Charles Ryan: All right.

Francis deSouza: Exactly. And then you go down the workflow-

Charles Ryan: Right, right, right.

Francis deSouza: ... for that organ type of cancer. Yeah.

Charles Ryan: So the potential criticism, I'm sure you're getting a lot of it, is, well, this is great. We do have cancers that for which we cannot screen, but what about the identification of cancers that are false positives? And what about the worry and the cost and all of those factors that this could bring in? So I guess my question is, what are you perceiving that patients, maybe they're not even patients yet, let's call them treatment test subjects. How are they perceiving things? And what about payers and healthcare systems as well?

Francis deSouza: Yes, it's a great question. In fact, when we designed the test, and we first discovered this phenomenon back in 2013. So we did a lot of work both at Illumina, and then when Grail spun out, they did these very large clinical studies. Over 100,000 enrolled in those studies. And there were a few things we knew were important. One, that it had to have sensitivity and we had to understand what type of cancers it could detect. And we had to be very clear on what stages it could detect. So we had to do a lot of work on that. Next, we knew that you needed to be able to identify the tissue of origin, that you couldn't put out a blood test like this if you couldn't tell the tissue of origin, because that would create a lot of anxiety in a population.

So it was critical, we said, for any test to be successful, it has to find the tissue of origin, which this test does. We also said, then, that you have to have a very low false-positive rate because you do not want to put out a population-wide screening test that generates a lot of false positive tests. And so we set this really high bar to say, "It has to have a false-positive rate that is less than 1%." Now that gives you a very strong positive predictive value. And after those studies, it turns out that the Grail Galleri test has a false positive rate of between 0.5% and 1%, so a very low false-positive rate. And all of those things were essential, we felt, for a successful screening test. And that's what Grail came with.

Now, what we are hearing, and I'll tell you, I did the test myself. And I did the test, I got the results back in six days. And there is some level of anxiety when you do the test for the first time, and fortunately, my results came back clear. And I'm going to do that test, personally, now every year as part of my physical. I'm over 50. And we believe if you're over 50 or if you have some kind of family history of cancer, it's probably worth looking at a test like this and talking to your doctor about it. This could be a really big breakthrough, obviously, for cancer mortality in the long term.

It is very early as a test. And so today it's a self-pay test and it costs $950. But we have started the conversations with payers, the Grail team has. And what we know is that not only are the outcomes better if you treat cancer early, but it's much cheaper in a lot of cases to treat cancer in stage one or-

Charles Ryan: Yeah.

Francis deSouza: ... stage two than it is in stage three or stage four.

Charles Ryan: Right.

Francis deSouza: And so the modeling, the healthcare cost modeling we've done, indicates that not only can you increase the odds of a better outcome for the patient, but you will take a lot of cost out of the healthcare system-

Charles Ryan: Right, right.

Francis deSouza: ... if you could do a big stage shift in when you identify cancer.

Charles Ryan: Right, right. So, I mean, one of the challenges in this issue of the stage that I keep coming back to is when is that moment when that cancer goes from stage one to stage two based on the cell-free DNA? And so you may have the opportunity to serially test patients who have negative tests after negative tests who then get a positive test, right? And I guess the question is, from the standpoint of quantification of the DNA, everybody has cell-free DNA, cancer, or no cancer. As you said, in the setting of cancer, some of your cell-free DNA is taken up by the cancer cells. So do you think you'll be able to give the world data on that moment of time when-

Francis deSouza: Yeah. We will learn a lot as the-

Charles Ryan: Yeah, yeah.

Francis deSouza: ... stuff gets rolling. The one thing I would say is, look, depending on the type of cancer, depending on the patient, the doctor may decide that even though there is a positive-

Charles Ryan: Yeah.

Francis deSouza: ... that the right thing for that patient may be just waitful watching. To say, "Look, let's watch. And let's do another test in a year or six months and see if this cancer is progressing and what the rate is." And so I think we now have that option too, that says, "Look, the action could be to just check in from time to time, maybe do more enhanced screening at a shorter time interval to see if it is moving." The other thing I'll tell you from a cost perspective is we rolled it out as a benefit to our employees. And so now if you are over 50 and you are an Illumina employee, you're eligible to do this test. And this is one of the few times, I think, as a CEO where you get the chance to roll out a benefit to your employees that not only is a loved benefit, but actually lowers the costs you incur.

Charles Ryan: Oh, uh-huh (affirmative). You're right.

Francis deSouza: And so this is one of those rare situations where patients really appreciate the benefit.

Charles Ryan: Sure.

Francis deSouza: And so what we are seeing now is even in advance of payers reimbursing, we're seeing employers now start to sign up, especially the larger employers that are self-insured. They are signing up to offer it as a benefit, knowing also that not only does it alleviate some of the human sufferings, but it also lowers the cost associated with insuring their employees.

Charles Ryan: Outstanding. And it sounds like a lot more to come. We'll look forward to seeing whether the uptake of the test and what we can find out from the serial testing of patients over time. I'm struck by this issue on a couple of different levels. I treat kidney cancer, and one of the things we will see from time to time is somebody who has a CAT scan for some completely unrelated reason, and they detect a kidney tumor. And they might even have metastatic kidney cancer. And of course, you don't know how long that cancer has been sitting there. And so what we do now in this setting is we kind of watch it at first. And there are patients who will go years perhaps before they ever need therapy for their kidney cancer. And if you contrast that with somebody who has a symptom, gets this kidney removed, and then three months later develops lung metastases, you're looking at two very, almost totally different diseases in terms of their natural history.

So I'm thinking as we think about this, this is a little bit like that kidney cancer situation where you detect cancer, but you don't know the trajectory, and so you have to kind of follow it for a while. So there might be active surveillance, watchful waiting-type scenarios with many cancers. I treat prostate cancer, of course, as well. And prostate cancer is one where a large number of patients go on active surveillance protocols.

So this does bring in a whole new set of issues to discuss in how we move forward in terms of following cancer patients. And it's going to create a lot of opportunities for knowledge and expansion for oncologists. Let's put it that way. In other words, they have to learn how to adjust to these types of tests and figure out what to do. So this will create a lot of thought, I think, and I congratulate you and your company on this amazing step forward and look forward to future conversations where we might be able to hear more about the progress.

Francis deSouza: Thank you. I look forward to updating you as we make progress. And thank you for having me.

Charles Ryan: My pleasure.
email news signup