Latest / "Isn't That Interesting?!": Breast Cancer Survivor Stories + Remarkable Research / Season 2 / Episode 2: Dr. Hani Goodarzi {San Francisco, CA}
Transcript
- 0:01Cancer cells hack our systems, usually without a trace.
- 0:06But paired with artificial intelligence, researchers have
- 0:09exposed some of these invaders. And they're working on an
- 0:13earlier detection with something as simple as routine blood work.
- 0:17Here's Martha Kaley, founder and executive directorofearlier.org,
- 0:21with this week's edition of Isn't That Interesting?
- 0:27Hi folks, This is Doctor Godardzi and I wanted you to
- 0:30meet him. He's really done some fabulous
- 0:34research and I'm so excited. And he has also recorded a Ted
- 0:39talk that is so informative and you can listen to it over and
- 0:45over again because you learned something new every time.
- 0:48So please enjoy listening to him talking about his his work.
- 0:53And then we'll just regroup at the end.
- 0:56So I'm just going to turn it over to you.
- 0:58Thank you, Doctor Garcia. Thank you so much for having me.
- 1:03I'm happy to start with a short introduction of myself so you
- 1:07and others know a bit more about me, and then I'm happy to tell
- 1:11you a little bit about our work. So first of all, I'm Honey.
- 1:15I'm an associate professor at UCSF, an ACCORD investigate at
- 1:19ARC Institute. My background is in a
- 1:22combination of machine learning and cancer biology.
- 1:24I did my PhD in quantitative and computational biology at
- 1:27Princeton University a long time ago, and I did my postdoc at ROK
- 1:32for university with, I was really studying cancer
- 1:35progression and metastasis. I really try to understand
- 1:38basically how what makes cancer cells tick and make them more
- 1:43aggressive and what are the ways that we could actually
- 1:46counteract that in practice. I started my lab back in 2016 at
- 1:51UCSF and we have a fairly hybrid lab.
- 1:54About half of the lab is focused on computational biology,
- 1:58machine learning and AI, and the other half focused on the
- 2:02combination of cancer biology and Arnab biology.
- 2:05So that's kind of the the scope of our research.
- 2:08One of the first projects that I started at UCSF when I when I,
- 2:14when I basically started my lab was this idea of trying to
- 2:18figure out if we take a step back and look at cancer cells,
- 2:23not as an extension of a healthy cell gone awry, but as an
- 2:30evolutionary entity that might have come up with the strategies
- 2:34and ways of doing things that doesn't even really exist in a
- 2:37healthy cell to begin with. Can we actually find something?
- 2:41Can we, can we find this cancer specific pathways of disease
- 2:45progression in cancer that we don't really see in healthy
- 2:51cells engaged in healthy cells? So again, this idea that for the
- 2:55most part throughout my careers, me and many others have been
- 2:59focused on this idea of there are, you know, certain knobs in
- 3:02the cell that you can turn up and down, you know, like
- 3:06oncogenes or torn suppressors and cancer cells basically turn
- 3:09these up and down. But our idea was that cancer is
- 3:13an evolutionary disease, sometimes takes decades for
- 3:18basically a single transformed cell to become a full blown
- 3:21anesthetic disease. And there's a lot of
- 3:24evolutionary force applied from the body and the environment
- 3:27onto onto cancer that shapes this evolution.
- 3:32And like evolution solves many interesting, many problems,
- 3:36interesting, interesting ways in, in nature.
- 3:39The idea was whether cancer cells can similarly come up with
- 3:44ingenious ways to to solve these problems, of course, to our
- 3:48detriment as patients. So that was kind of the the
- 3:52thesis of the work when when we started our lab in kind of one
- 3:56of the main areas that we wanted to invest time and effort in.
- 4:00And former postdoc of mine, Doctor Lisa Fish was actually
- 4:04the brave soul who who took up this kind of answering this
- 4:07question. And for for a variety of
- 4:10reasons, we focused on small non coding RNAs specifically as a
- 4:16way of going at finding these kind of cancer specific pathways
- 4:21of disease. I also wanted to take a quick
- 4:23segue. I mentioned that why my lab is
- 4:26really interested in metastasis. A lot of our work is actually
- 4:28focused on breast cancer and to some extent, to some extent
- 4:33colon cancer, where a lot of our research is focused on breast
- 4:36cancer metastasis specifically. So a lot of our work over the
- 4:40years have been focused on this disease, kind of progression of
- 4:44of this disease. So Lisa, I started basically by
- 4:48asking the question of are there small non coding RNAs that are
- 4:53cancer specific, breast cancer specific basically we don't even
- 4:56see them in healthy normal tissues, but they appear in
- 4:59cancer cells. So what is the idea there?
- 5:02The idea is that small non coding RNAs, there are these
- 5:06like tiny fragments of RNA in the cell that have regulatory
- 5:11potential, like functional potential built into them
- 5:15because they're very versatile. They can interact with our
- 5:18antibody proteins, for example, They can target, you know,
- 5:21interact with other RNAs, they can interact with DNA.
- 5:24So if you want to kind of mess with function of molecules in in
- 5:29the cell in important ways, this RNA is kind of an approach, a
- 5:35very good approach to go about it.
- 5:37And you know, my work as during my postdoc had kind of pointed
- 5:42the way to, you know, a small RNA is actually being this
- 5:45really versatile and rich repertoire of kind of RNA space
- 5:50that was fairly underexplored and and under tapped and to be
- 5:53honest continues to be like that.
- 5:55So the, the first step of our process was like, if we can find
- 6:00cancer specific RNA small RNAs, then it means that we can like
- 6:05limit our search into those and be like, does any of these RNAs
- 6:09actually end up doing anything important in, in breast cancer
- 6:13progression? So the but the first answer, the
- 6:16first question to answer for us was, well, we do know to some
- 6:20extent that there are cancer specific small RNAs based on our
- 6:25earlier work. But really how many?
- 6:28It turns out we were actually quite surprised by the answer.
- 6:31And the answer is like a lot. We identified thousands upon
- 6:34thousands of these small RNAs that are very cancer specific.
- 6:38So we started with profiling, you know, breast cancer cell
- 6:41line models that are kind of our go to models for studying breast
- 6:45cancer in the lab. And you know, we, we saw a lot
- 6:48of these cancer emergent RNAs in that in those lines and we were
- 6:52like, maybe, you know, these are cell lines.
- 6:54So, you know, maybe we shouldn't read too much into it.
- 6:56So we also looked at patient derived xenograft models, for
- 6:59example, as like more faithful models of breast cancer.
- 7:02And that was thanks to collaboration with another group
- 7:06at at UCSF under doctor, under Google's lab.
- 7:09And finally, we also looked at clinical samples in a largest
- 7:13extent thanks to the Cancer Genome Atlas where they had done
- 7:16a small RNA sequencing and basically thousands of samples
- 7:20including large number of breast cancer tumors.
- 7:24So when we put all of these together, we actually again saw
- 7:28large set of these kind of breast cancer emergent RNAs that
- 7:31are not present in healthy cells but but appear in cancer cells.
- 7:35And we, the way that we approach them is very similar to how you
- 7:39would approach basically any kind of oncogenic driving
- 7:43pathway. Kind of do higher expression of
- 7:47these RNAs lead to, you know, kind of adverse short term or
- 7:51long term clinical outcomes? Do I see, for example, patients
- 7:55relapse earlier have higher expression of any of these RNAs?
- 7:59And I also wanted to take a step back and mention that we, we, we
- 8:03annotated these RNA, so we got to name them.
- 8:06And so we've called them orphan non coding RNAs or on Carnas.
- 8:10So you know, on sounds like oncology, but but in reality
- 8:14that there's nothing to do with that is more about this idea
- 8:16that these are orphan RNAs that the kind of in the sense that
- 8:19they appear out of nowhere in the context of cancer.
- 8:23So then annotated is oncoronase and we kind of prioritize them
- 8:28in terms of their potential role in disease, cancer progression
- 8:32in breast cancer, the same way that you would kind of nominate
- 8:36important genes in terms of association of studies and based
- 8:41on short term and long term clinical outcomes.
- 8:44So we nominated a couple of these and you know, we took the
- 8:48top one on our list and and we did a much further deep dive
- 8:53into kind of understanding how is this RNA cancer specific?
- 8:57It ends up being kind of a fragment that is generated from
- 9:02the RNA component of telomerase, a gene called Turk.
- 9:06And I mean Turk itself tends to be fairly silenced in healthy
- 9:10cells and actually re expression of telomerase activity and and
- 9:14Turk is one of the hallmarks of cancer.
- 9:17So that precursor longer RNA itself is is cancer specific.
- 9:22But then on top of that, the we found like a couple of our new
- 9:25body proteins that bind this longer RNA, the longer Turk RNA
- 9:30and cut it to a shorter fragment that we identified as an orphan
- 9:33non coding RNA and again renamed it T3P, really referring to
- 9:38being end of Turk RNA. And we also dug a lot deeper
- 9:45into kind of the biology of this small RNA and, and how it could
- 9:48actually play a functional role in breast cancer progression.
- 9:53Now, as we were doing all of this, we're really not thinking
- 9:55about anything about detection or, you know, cancer detection
- 9:59or anything like that. We're really solving this
- 10:01problem that I mentioned, right? How can we look at kind of these
- 10:05cancer specific pathways of kind of functions that appear only in
- 10:11cancer cells? And our idea was that if you can
- 10:13find these kinds of pathways, we can target them very effectively
- 10:17because there is no kind of on target toxicity, no, no healthy
- 10:21cell has them to begin with. So if you target them, it will
- 10:24really just target the cancer cell.
- 10:26But as we were working on this way on this project, quite
- 10:30serendipitously, we actually had another project in the lab that
- 10:33was focused on understanding how small RNAs are secreted from
- 10:39cells, all cells, including cancer cells and healthy cells.
- 10:43They package RNAs and secrete them.
- 10:45This is actually one of the areas that we know very little
- 10:47about. You know, we had you think these
- 10:49are important for cell cell communications in the body, but
- 10:52really it's an area that's again massively underexplored.
- 10:56So we were kind of we have an area of research in our lab to
- 10:58this state really focused on kind of secretion of RNAs.
- 11:03And as we were working on that, we kind of quite serendipitously
- 11:07one day noticed that we do see now that we have annotated on
- 11:11coronation breast cancer there. Now we actually do see them
- 11:14outside of cancer cells secreted by cancer cells into the media.
- 11:19And that, that was kind of the light bulb moment for us that
- 11:23well, if on coronation or cancer is specific and are can are, you
- 11:29know, secreted by some of them are secreted by cancer cells,
- 11:33then if you see enough of them in the right patterns and you
- 11:35know, the right numbers, then you have like conviction that
- 11:39there's an underlying cancer cell is actually secreting them.
- 11:42So that was kind of the, the really like light bulb moment
- 11:46that we could actually do something with this when it
- 11:48comes to kind of liquid biopsy applications.
- 11:51And here I actually have to acknowledge quite a bit our
- 11:56breast oncology program at UCSF, which is led by Doctor Laura,
- 12:02doctors Laura Esserman and Laura Van Fier and really the entirety
- 12:06of the ISPI team for that matter, who have been running
- 12:09one of the largest and longest running adaptive trials in
- 12:13breast cancer called ISPI twos and your adjuvant therapy study
- 12:17and and trial. And I basically took what we had
- 12:21found thus far and, you know, took it to mentor and Doctor Van
- 12:26Fier and kind of we went through, over what, you know,
- 12:28through what we had and kind of how we could leverage this to
- 12:31basically look at using blood, look at the presence or absence
- 12:35of an underlying tumor. And at the same time, I think
- 12:39the, the ISP team had it started to work on CTDNA as, as an
- 12:44approach as well. So, you know, the, the
- 12:46comparison was, was timely. So we ended up actually
- 12:50basically receiving about for about a couple of 100 patients
- 12:54pre and post treatment blood samples, you know, samples
- 12:58specifically and we processed them in the lab in a very simple
- 13:02way. We're just basically extracted
- 13:04RNAs and we we sequenced small RNAs in the blood.
- 13:10And again we knew exactly what we were looking for, which is
- 13:12basically this is a small RNAs is on Carnas that we had
- 13:16annotated. So and we gave a very simple
- 13:20score to every patient, which is basically how much of these RNA
- 13:24do you see per unit of blood and basically how does that value
- 13:29change pre and post treatment. And we actually saw that as
- 13:34expected when the patients are treated and a tumor size reduces
- 13:39in most of these patients, there is a reduction in signal in
- 13:42terms of reduction of the content of on coronaries in the
- 13:44blood of these patients. And more importantly there's
- 13:48also very clear long term clinical association that the
- 13:52more on coronaries you see post treatment in the blood the worst
- 13:57outcome for the patients. So that was kind of the first
- 14:01sizable study that that we did that really proved this point of
- 14:07there is a very clear cancer associated signal in Oncoronase.
- 14:12Now, what we did here was was very simplistic, right?
- 14:15This idea that you kind of basically sum up all Oncoronase
- 14:19in blood. In reality, on carnies have a
- 14:21lot of individuality the same way like genes have a lot of
- 14:24individuality. Sure, you could get take all on
- 14:28code genes and sum them up, but they're not all equal.
- 14:31Different ones and different cancers are more important.
- 14:34So we wanted to really be able to go after that individuality
- 14:39of oncologies and be able to understand that process.
- 14:43The the challenge is that, well that requires a lot of samples
- 14:47that that and also it requires kind of machine learning
- 14:50approaches that really didn't really exist to be able to
- 14:54leverage this presence and absence signature or what we
- 14:59call actually oncorning barcodes basically and associate that to
- 15:03presence and absence of disease. Our idea is that basically on
- 15:08Carnase are expressed in cancer cells, but a fraction of them
- 15:13are actually secreted. So not every small RNA is
- 15:16actually secreted from cancer cells.
- 15:17There is a very special grammar and code to RNA secretion.
- 15:22So a small fraction of all the on Carnase are expressed in
- 15:25cancer cells secreted and then we are taking, you know, a meal
- 15:29of blood from a patient. So you know, that's another like
- 15:33very significant dilution. We're not sampling the entirety
- 15:36of the blood, we're taking a tiny fraction of that.
- 15:39So, you know, you don't expect to see the entirety of the like
- 15:43landscape along coronaries present in the cancer.
- 15:46You will get, you know, a fraction of a fraction.
- 15:49But what you would like to do is from that fraction of a
- 15:53fraction, kind of reconstitute the original picture that what
- 15:58was in the what was in the tumor.
- 15:59And that would tell you not only that the tumor exists, but also
- 16:03something about the biology of that tumor.
- 16:05And that's kind of an area that that we have worked on quite a
- 16:08bit as well. I think the best analogy that I
- 16:10can make is that it's kind of similar, you know, to, to our
- 16:15vision, right? So once we know like what a cat
- 16:17looks like, you have a picture of a cat, you can kind of remove
- 16:22like 90% of the pixels and you only see like 10% of the image,
- 16:26but you can go, oh, that's a cat, right.
- 16:28So your, your brain, once it is learned what a cat looks like
- 16:33and what a cat is, can fill in the gaps and reconstitute and
- 16:37reconstruct that image of a cat from very partial information.
- 16:41So that's kind of, and that's because, you know, our human
- 16:45vision is like such an kind of elaborate machine learning model
- 16:49basically that can actually fill in these gaps.
- 16:51So that's exactly the kind of problem that we needed to solve
- 16:56taking advantage of machine learning and AI as well that
- 16:59taking advantage of this like fraction of fraction of
- 17:02oncoronase that we see, how can we reconstruct the full picture.
- 17:06And the reason that we can do this is that the full picture
- 17:10comes from really our knowledge of the actual patterns in solid
- 17:15tumors. And I guess I mentioned we had
- 17:18initially done this in breast cancer, but we have since, you
- 17:22know, extended it to other cancers as well.
- 17:24So what we ended up doing is we actually with with support and
- 17:28help from, you know, folks at UCSF, the innovation ventures at
- 17:32UCSF, they helped us actually set up a spin out a company
- 17:35called XI Bio that basically took charge.
- 17:39And by given the resources that that is needed to actually
- 17:43collect the kind of data that you need and build the kind of
- 17:46machine learning team that you need to be able to realize this
- 17:50vision of how do we build catalogs of bar codes of
- 17:54oncoronase from different tissues from different tumors.
- 17:58And then use that to do pattern matching against blood collected
- 18:03from healthy versus cancer patients and be able to to
- 18:07compare and contrast. So as I mentioned, we had looked
- 18:11at 3 arms of I Spy. One of the studies that Excite
- 18:15Bio has done over the past few years is actually completing
- 18:18that study. So they've looked at I think an
- 18:21additional 7 arms of I Spy and recapitulated what we saw.
- 18:25I think there was a poster a few years ago that they presented
- 18:29that, that kind of showed the reproducibility of using
- 18:33oncarnase for, for predicting basically minimal residual
- 18:37disease and as well as kind of short term and long term
- 18:41clinical outcomes. And more importantly they have
- 18:45extended it to to other diseases as well.
- 18:48And they continue to have a lot of interest in basically breast
- 18:52cancer early detection as well. But of course you know clinical
- 18:57studies are a long and inexpensive.
- 18:59So we'll we'll see how that process moves forward.
- 19:04That is so interesting. So let me ask you a dumb
- 19:08question. So once the cancer, once the onc
- 19:13RNA is discovered and tested, are there and if it's treated,
- 19:21are there residual onc RNA is running around still in the
- 19:25blood? And if there are, how do you
- 19:29determine the when you need to do extra treatment?
- 19:36So on Carne's like absence absent the source of the signal,
- 19:42which is a tumor, right, they don't last long.
- 19:45So extracellular vesicles released by by cells in the
- 19:49body, they, you know, they, their lifespan is like hours.
- 19:52So, so you, so your. Your signal is very much tied to
- 19:58the presence and absence of. Some are very similar to like
- 20:01you know, CTDNA for example, where you look for fragments of
- 20:05DNA originating from from cancer cells.
- 20:09Great. OK.
- 20:10The there's another one that I was thinking about.
- 20:12If there is a metastasis connected with this, do you get
- 20:18different readings of the onc RNA?
- 20:22Does does it change? Does it morph into something
- 20:26different? Very, very good question.
- 20:28So the, the answer is yes and no, right?
- 20:31So there is a, there is a component of kind of different
- 20:35types of cancers that is like forms the trunk of our
- 20:38understanding or, you know, cancer, like for example, you
- 20:40know, this is a breast cancer, this is colon cancer and so on
- 20:43and so forth. But then you can go further
- 20:46refined, right? There's, you know, 3 subtypes of
- 20:49breast cancer, There are four CMS types of colon cancer and so
- 20:52on and so forth. So there's kind of this higher,
- 20:55higher resolution when it comes to kind of gene expression
- 20:59patterns in cancers and you know a part and other kind of markers
- 21:04of types of types of cancer on Carnase are not very not
- 21:09different from that. So there are key on Carnase that
- 21:12are very much cancer associated. So you know, you'll see them
- 21:14primarily breast cancer and so you can kind of associate their
- 21:19patterns of presence or absence very much as a trunk of this
- 21:22tree of like breast cancer. But then there is like
- 21:25differences between like subtypes of breast cancer for
- 21:29example, but also progression of breast cancer.
- 21:32So in that earlier work that I mentioned that led to
- 21:36identification of T3P from Turk. The reason that one of the main
- 21:40reasons that we actually nominated T3P was through
- 21:43comparison of poorly and highly metastatic the models of breast
- 21:47cancer that we had in our lab. And we focus on TD therapy
- 21:50because not only was it appearing in breast cancer, but
- 21:52it was further going up and highly metastatic models that we
- 21:57had. So that was kind of a key
- 21:59indicated role. This is an important on carne if
- 22:02it's expression levels is increasing.
- 22:05So that particular on carne would be, I want to say more
- 22:13central for for the change in the and to metastasis, right.
- 22:19Yeah, exactly. So there there will be basically
- 22:22like on carnage that either appear later as a disease
- 22:25progressing or you know appear earlier but then further
- 22:29increase as a disease progresses.
- 22:32So you will get differences that you can zero in on if you if you
- 22:36really that's your intention. But kind of again, similar to
- 22:39like telling apart different subtypes of breast cancer, but
- 22:43there's also kind of this general idea that you have like
- 22:48patterns, universal patterns to being a breast cancer basically.
- 22:52Right. OK.
- 22:53Now where do you see this going? Where do you envision?
- 22:58I mean, one of the purposes of doing these podcasts are for us
- 23:03to demonstrate the different disciplines that are being so
- 23:09effective of, of so many different things like blood
- 23:13biomarkers or whatever, you know, and with AI working with
- 23:19the imaging, I think that's an interesting concept.
- 23:23So where do you see this evolving to?
- 23:27Yeah, I think, I think fundamentally what we want to do
- 23:31is to be able to get a continuous source of information
- 23:36from cancer cells and understand, you know, what
- 23:38they're up to, what they're doing kind of continuously
- 23:43without the need for direct biopsies, right.
- 23:45You know, one of the things that really changed things for us
- 23:49when it comes to therapies was the ability to biopsy tumors.
- 23:53You go and bring out the tumor and take a sample, and you learn
- 23:58so much from that, right? You look at different markers,
- 24:01mutations, genes, pathways, all of those.
- 24:05And that is basically how you define your therapies, right?
- 24:09But but, but then the problem is that that's one snapshot in
- 24:13time. And we started all of this
- 24:14conversation by saying that tumors evolve, right?
- 24:17You know, in breast cancer, you know, subtype switching, for
- 24:20example, is very common. So, you know, you started with
- 24:23an HR positive breast cancer. It doesn't mean that after you
- 24:27know, a round of treatment that you think you're targeting an HR
- 24:31positive breast cancer. It continues to be HR positive
- 24:34breast cancer. Sometimes it's subtype switches
- 24:36and you don't have any idea until you go on biopsy again,
- 24:39which is which is rare. So the the point is that
- 24:44understanding what is happening in the cell in cancer cells and
- 24:49the biology of the tumor is incredibly informative about how
- 24:52we deal with it. But the reality is that we don't
- 24:56have access to that information. We get like 1 biopsy and one
- 25:00snapshot and that's it, right? Imaging is helpful about telling
- 25:04you kind of the gross phenotypes and of the tumor, but it's just
- 25:08doesn't really tell you anything about the biology.
- 25:11Now where liquid biopsies, this is where liquid biopsies can
- 25:14really change the game, right? Of course, we have been focusing
- 25:18more, you know, in early phases on just detecting the disease.
- 25:23But I'm excited about Oncoronase because they can take things a
- 25:26step further because they can tell you about the biology of
- 25:31the cancer cell without the need for sampling via biopsy cancer
- 25:35cells themselves, but looking at what they are, what they're
- 25:38releasing into the bloodstream, because what they're releasing
- 25:41to the bloodstream kind of cancer progression and stage
- 25:45being one example of that is dependent on the biology of the
- 25:49of the cancer. So as the cancer evolves, some
- 25:53incarnates do change, respond in response to that.
- 25:56And so you can basically be the powerful AI sitting on top of
- 26:02Incarnate patterns, be able to kind of continuously monitor how
- 26:06the tumor is evolving and changing.
- 26:09But going back to the point of for example, when it comes to
- 26:13metastasis, right, again like liquid biopsies is really the
- 26:16only and best way of kind of getting information even about
- 26:20the presence of them. But also kind of what state are
- 26:22they in that there are just, you know, you don't have biopsies at
- 26:27that point. So it's kind of a unique source
- 26:30of information that we have access to that if we really like
- 26:33invest properly into kind of collecting the kind of data that
- 26:37we need and developing the kind of AI models that we need to
- 26:42really be able to to change the game here in terms of how we
- 26:45approach this problem. Well, I just think this is
- 26:50extraordinary work. It's it's very exciting to think
- 26:54about the different, the different approach that you are
- 26:58presenting. And I hear you talking about the
- 27:01biology of the cancer cell itself rather than just a global
- 27:06perspective. So I really appreciate is there
- 27:10anything else that you are are interested in sharing or what?
- 27:14I just I I just yeah, I mean, I just wanted to be very clear
- 27:18that what IA lot of what I excite left our lot, you know,
- 27:224-4 years ago. So a lot of what we know about
- 27:25kind of the Balaji on Carneys and the power of power of that
- 27:28included some of the developing this, this machine learning
- 27:31models has been in collaboration with that been very open to
- 27:35publishing. So we have published a couple of
- 27:37models that we have been using and kind of the modeling
- 27:41approach that I mentioned earlier around like how we can
- 27:43leverage on Krnase using these AI models.
- 27:47And then they have a newer foundation model, AI foundation
- 27:50model called XI One. And both of these models are
- 27:53kind of, I think I see them as kind of foundations of liquid
- 27:58BIOS about oncoronase. But in general, anything that
- 28:02you want to look at in the context of blood and DNA
- 28:07methylation patterns or fragmentation patterns or even
- 28:11protein levels and all that, I think they can all go through
- 28:14this similar kind of modeling approach to really take the most
- 28:19out of information that is, that is embedded in, you know, in one
- 28:22mill of blood. Well, I just think this is
- 28:25really interesting and so I want you to have it done by next
- 28:28week. OK.
- 28:32Thank you very much for your work.
- 28:34We really appreciate it and we will let you know when we get
- 28:38this podcast launched and just looking forward to it.
- 28:42Thank you for spending your time with me.
- 28:44I really appreciate that. My pleasure and I so wanted to
- 28:48really also emphasize that that support from earlier in early
- 28:52days of this project in our lab was was very much instrumental
- 28:56and you know, kind of gave us the boost that we needed to
- 28:59continue to realize that we are even like on the right path.
- 29:02Because as I mentioned, liquid biopsies and cancer detection
- 29:05and all that was not a direction that we're really truly thinking
- 29:08about. So a lot of kudos goes to the
- 29:10folks who actually kept us going.
- 29:13They're both at UCSF, members of the I Spy team and also outside
- 29:19all the funders that who helped us push this forward.
- 29:22Well, you're very, very sweet to do say that.
- 29:25I appreciate all of our donors and I appreciate our medical
- 29:29advisory committee. I would never presume to decide
- 29:33where these grants go. So I just said, oh, I'll take
- 29:36over because I, that's a lot for me to do.
- 29:38So thank you so much for recognizing them because they
- 29:42have done remarkable work too. So thank you for for everything
- 29:46and I'll look forward to next week when it's all fixed.
- 29:51Yes, you and I both. OK, thank you so much.
- 29:54Awesome. OK, bye.
- 29:55Bye, goodness. Bye.
- 29:57This episode of Isn't That Interesting was made possible
- 30:00thanks to volunteers and donors just like you.
- 30:03It was hosted by Martha Kaley, founderofearlier.org and edited
- 30:07by Liam Mcburney. I'm Valerie Amsterdam, reminding
- 30:11you that for the last 30 years, this is North Carolina based
- 30:14organization has helped fund groundbreaking research all over
- 30:18the world from molecular imaging to a sensitive blood test for
- 30:22breast cancer. Plain and simple, earlier
- 30:25detection saves lives. And you can help host a
- 30:28fundraising event, make a donation for volunteer your
- 30:31time. You can find us not only on the
- 30:33web, but on Facebook and Instagram too.
- 30:36Let's change the picture of breast cancer together.