Defining Mechanisms and Biomarkers of Sensitivity & Resistance To Anti-Cancer Treatments

December 16, 2020

December 9, 2020

Corporate Guest Speaker: Susan Galbraith, PhD, Senior Vice President and Head of Early Oncology R&D, AstraZeneca Panelists from Yale: Megan King, PhD, Mark Lemmon, PhD, Katerina Politi, PhD, Kurt Schalper, MD, PhD, and Qin Yan, PhD

Information

ID: 6012
To Cite: DCA Citation Guide

Download Transcript

00:08everybody. Welcome to our session on behalf
00:13of Yale University and Yale Cancer Center.
00:17I'm pleased to have you with us as
00:20part of the Yale Engage Cancer series.
00:24This session is entitled defining.
00:26Mechanisms and biomarkers of sensitivity
00:29and resistance to anti cancer treatments.
00:31I'll be your moderator.
00:33I'm I'm Barbara burtness.
00:34I'm a medical oncologist and have a interest
00:37in drug development and head neck cancer.
00:40And we have a phenomenal panel of Yale
00:44faculty members and Anna corporate
00:47guest Susan Galbraith from Etsy.
00:49And hope to have a very very
00:53interactive session.
00:53I'd like to start with a
00:57few housekeeping items.
00:59The program format as I said,
01:02is going to be each of our panel
01:05members giving a brief about 5 minute
01:08introduction to the work that they do.
01:11What they see is as key questions.
01:15Will have all of the panel presentations
01:181st and then move on to the discussion.
01:22The question and answer.
01:24We know that to attack cancer
01:27we need team science.
01:28We need collaborations within our
01:31organization and across different sectors.
01:33Academic,
01:33public and industry and Yale engage was
01:36designed to build these connections,
01:39particularly between Yale
01:40scientists and industry leaders.
01:42To keep the discussion lively, we.
01:45We welcome questions.
01:46Some have been submitted ahead of
01:49time and you'll have the ability to
01:52submit them through the Q&A function.
01:55On the Web and R we have an
01:58enormous amount of expertise
02:00among our panelists and will be.
02:04Monitoring those questions as they
02:06come up and and try to get to as
02:09many of them as possible and I I
02:11want you to know that this web,
02:13nor is being recorded,
02:14so now I'm really pleased to be
02:16able to introduce Charlie Fuchs.
02:18He's the secular professor of
02:19medicine and medical oncology and
02:21a professor of chronic disease
02:23Epidemiology here at Yale.
02:24He's the director of the Yale
02:26Cancer Center and Position in chief
02:28at Smilow Cancer Hospital,
02:29Charlie.
02:30Forever thank you and thank you for
02:33your leadership on this and welcome
02:35to all the attendees to what is now
02:38our third Yale Engage cancer event
02:41and it's really been an exciting and
02:43incredibly productive series of forms.
02:46So please it could join us
02:48for this third one.
02:50You know, we we all recognize
02:52that despite the fact that we're
02:54dealing with a global pandemic,
02:56the consistent impact of
02:58cancer on public health.
03:00And the morbidity and mortality
03:02and costs on our population.
03:04Or considerable an it remains one
03:07of the great challenges in medicine.
03:09And it also is one of the
03:12largest investments.
03:13I think that goes on and healthcare
03:16research and drug development and
03:18our our our efforts at Yale is
03:21to really tackle this challenge.
03:24Yeah Liz had a long legacy in
03:27Cancer Research and cell biology,
03:29genetics, pharmacology,
03:30immunology, among other elements.
03:32And I think a lot of the history of success,
03:36including four Yntema therapies,
03:38come out of this University were
03:40privileged to work at one of the
03:43national one of the original
03:45National Cancer Institute,
03:46designated Cancer centers,
03:48and has been a really an area
03:51that research that is as built a
03:54legacy of great innovation as well.
03:56Smilow cancer hospital.
03:58Our clinical center.
04:00Is celebrating its 10th anniversary
04:02and is a robust operation that now
04:05sees about 48% of every newly diagnosed
04:07cancer patient in the state of Connecticut.
04:11And really,
04:11we view that through the science
04:14and through this robust clinical
04:16operation we really are committed to moving,
04:19discovery scientific discovery
04:20into the clinic.
04:22Really pleased with the team
04:24that's been assembled today,
04:25our first and Yale engage cancer was
04:28focused on immunobiology, our second.
04:30Was focused on novel therapeutics,
04:32and the third really ties it all together,
04:35which is to understand now,
04:37given these efforts to develop new drugs,
04:40new targets,
04:41how do we understand resistance?
04:43How do we understand sensitivity?
04:45And how do we further enhance our
04:47approaches to cancer therapy?
04:49Integral to this fight is our
04:51collaboration with industry,
04:52and we're so pleased to have Doctor
04:55Susan Galbraith join us as our
04:57industry partner on the panel,
04:59and we realize that.
05:00So many of you in the audience
05:03come from the biotech and pharma.
05:05An really part of this effort.
05:08Beyond hearing from these experts
05:09in their insights is to really
05:11begin a conversation.
05:13Because one thing we really welcome
05:15here at Yale is to collaborate with you.
05:18We want to build strategic
05:20partnerships with all of you.
05:22Because ultimately this fight against cancer.
05:24Yes,
05:24it requires each of these
05:26domains on the slide,
05:27but it requires a community
05:29focused on every aspect,
05:30and that includes academia
05:31and industry in biotech.
05:33So one thing I want to invite
05:35you today is to ask questions,
05:37but after this form,
05:38please reach out to us,
05:40will reach out to you.
05:41And let's think about ways we
05:43can work together.
05:44I think we have a lot of resources
05:46we can bring here at Yale to
05:48partner with all the great
05:50things you're all doing and we look forward.
05:53To continuing this conversation
05:54long after this form, so again,
05:56thank you for joining and
05:57I'll turn it back to Barbara.
06:02Thank you Charlie. I think that
06:05that's a great introduction to
06:08to what we're trying to do here I
06:11I just had a brief opportunity to
06:14to scroll through the list of 100
06:17participants an it's a formidable group,
06:20including GAIL scientists, industry partners.
06:22Alumni are supporters,
06:23so I think that we can anticipate
06:26some some pretty hard hitting
06:28questions from that group. So we've.
06:31We've tried to arrange these talks so that.
06:34We hope that there's a little
06:36bit of a natural progression
06:38in the scientific questions,
06:40and Dan the approaches that are are
06:43taken to understanding resistance.
06:44As I said,
06:46every speakers been asked to sort
06:48of reflect a little bit on what's her,
06:51his core expertise.
06:52What questions drive the
06:54research and how they hope to,
06:56or Yale hopes to work with industry partners.
07:00To address cancer cancer
07:03treatment resistance.
07:05And what kinds of capabilities and
07:07resources need to be brought to bear?
07:09So each of those speakers has been
07:12asked to go only for about 5 minutes?
07:15I've been told that I should be
07:17ruthless and and cut you off.
07:19If you go over and and that
07:22will be hard to do.
07:23'cause I know the talks
07:25are going to be great,
07:27but let me start by introducing
07:29Doctor Katie Palitti.
07:30She's an associate professor of pathology
07:32and Medicine leader in our Cancer Center.
07:35Through those answering signaling
07:36cancer signaling networks program,
07:38as well as a leader of our
07:40lung spore program and Katie.
07:43Think it away.
07:44Thank you very much, Barbara.
07:46And I'm really delighted to have
07:48the opportunity to speak here today
07:51and tell you about some of the
07:53things that we're interested in.
07:55I have a cancer biology lab here
07:58really with a focus on lung cancer and
08:01one of the areas that we are really
08:04interested in studying is working on
08:07resistance and resistance to various
08:09cancer therapies including targeted
08:11therapies and also immuno therapies and.
08:13Some of the things that we think
08:16about a lot and work on.
08:18I'm really approaches to discover new
08:20mechanisms of resistance were interested
08:22in understanding the relationship
08:24between tumor genotype and drug sensitivity.
08:26We study the influence of the tumor
08:29micro environment on sensitivity
08:30to different therapies and also
08:32investigate mechanisms of drug tolerance.
08:35So why do some cells die when you
08:37apply a therapy and instead other
08:40cells do not die and stick around
08:43and serve as the fertile ground for
08:46the emergence of drug resistance?
08:48And then we also investigate new
08:51approaches based on the science that
08:53we discover to overcome and or to
08:56prevent the emergence of drug resistance.
08:59And we do these studies by really integrating
09:02information from various different systems,
09:05various different models and using
09:07a variety of different technologies.
09:09We use specimens and data from patients,
09:12so we have a very robust biopsy program.
09:16Here,
09:16within the context of the Lung Cancer Group.
09:20Where we.
09:21Can obtain biopsies from patients.
09:26Long sort of the spectrum of
09:28their treatment with therapies,
09:30and we can generate patient derived
09:32models from these biopsies,
09:34but also then analyze the
09:36data and information to really
09:38understand resistance in patients.
09:40We use these models to generate
09:42or these specimens to generate
09:44patient drive Zeno graphs as well,
09:47and also 2D or 3D cultures
09:50from patient specimens,
09:51and we also extensively work
09:53with genetically engineered mouse
09:55models of lung cancer that we can.
09:58I used to study resistance and in
10:00that regard I'd like to tell you today
10:03about some work that we have been
10:05doing in the field of EGF receptor,
10:08mutant lung cancer. Next slide, please.
10:10To really use models to study resistance
10:14to the EGFR tyrosine kinase inhibitor,
10:18also Merton.
10:19If and this is a work that really
10:23illustrates a partnership between.
10:26Academia and investigators in academia
10:28and work that we've done together
10:31with Astra Zeneca and also working
10:33with Garden Technology and work that
10:35was published recently this year and
10:37so EGF receptor mutations are found
10:40in about 15% of lung cancers and can
10:43be targeted with tyrosine kinase
10:45inhibitors and one of the most recent ones.
10:48Is this tyrosine kinase inhibitor
10:50awesome Merton Eben?
10:51So we can take our genetically
10:53engineered mouse models and ask
10:55the question what happens?
10:57If you have mouse models of EGF receptor,
11:00mutant lung cancer,
11:01and you treat them with awesome Merton,
11:03if and so we took my sweet,
11:06treated them till the emergence
11:08of resistance.
11:09And when we looked at resistant
11:11tumors to see what was happening,
11:13we found that almost 50% of the tumors
11:16that emerged had secondary mutations
11:18in EGF receptor that confer resistance
11:20to awesome American if and so.
11:23With that information we can actually
11:25then go ahead using these models.
11:27So we've discovered new mechanisms.
11:29We can now use these models for preclinical
11:32testing and test new therapies.
11:34We can also with this information
11:36go into human specimens and data
11:38and analyze the relevance of the
11:40resistance mechanisms there, and so.
11:42For example, in this study we found
11:44that the mutations that were emerging
11:47were particularly relevant to the L.
11:498:50 at our subset of EGFR mutant tumors,
11:52so there was an allele specificity
11:54that was revealed through our studies
11:56in mouse models and then working
11:59with colleagues like Mark Lemon.
12:00Here, you're going to hear from next.
12:03We can really then study the biochemical
12:06properties in detail of these mutants.
12:09Next slide, please.
12:11So we also are working extensively
12:14to take these models that we have
12:16and sort of take them to the next
12:19level to study some of the more
12:22complex mechanisms of resistance,
12:24and we have modified for example this
12:26initial mouse model of EGF receptor,
12:29mutant lung cancer to incorporate
12:31additional genetic alterations that
12:32are also found in humans in EGFR
12:35mutant lung cancer, including,
12:36for example,
12:37tumor suppressor gene alterations using in
12:40vivo CRISPR CAS 9 gene editing and so now.
12:43We can study how those additional
12:45alterations are impacting tumor progression,
12:47sensitivity to therapies,
12:49and the phenotypes of tumors.
12:51As I mentioned in my first slide,
12:54we also have a robust program to
12:57generate patient derived models,
12:58and here is really an illustration
13:01of sort of the different.
13:03PDX is that we've generated across
13:06various different oncogenic subgroups
13:07of lung cancer with different
13:09oncogenic driver alterations,
13:11and so we're using these models.
13:13To really study resistance in human
13:16specimens and really use them to
13:19study heterogeneity of human tumors,
13:21signaling network alterations,
13:22and the molecular profiles that you
13:25can have in these human who tumors
13:28with or without drug treatment.
13:30Thank you.
13:32Thank you so much Katie. I,
13:36I think that there's there's so
13:39much there for the other speakers to riff
13:43off of and and to set up our questions.
13:47Next, let me introduce Mark Lemon,
13:50distinguished Professor of pharmacology.
13:51You see his leadership roles in the Cancer
13:55Center in Cancer Biology Institute.
13:57There, an mark is unique and bringing a.
14:01You know a wealth of expertise in
14:04biology and structural biology to the
14:06very interface with drug development
14:09and and disease based research and so.
14:12Looking forward to your comments, mark.
14:15Thank you very much,
14:17Robert and good afternoon.
14:18So a great pleasure to be here.
14:21I look forward very much to
14:23hearing discussion later on.
14:25As as Barbara mentioned,
14:26I'm really a basic scientist
14:28interested in how molecules work.
14:30My core expertise really is in
14:32biochemistry and structural biology.
14:34The focus of most of our work is is
14:37detailed understanding of of how
14:39molecules and networks involved in
14:41oncogenic signaling actually do
14:43work and do not anatomic detail.
14:45Where we can and quantitatively
14:47understanding how their properties
14:49are changed by oncogenic
14:50and resistance mutations.
14:51As Katy mentioned,
14:52work we're doing with her and how
14:55we can then use that information
14:57to guide mechanistically driven
14:59personalized medicine or put the
15:01biochemistry into personalized medicine.
15:03Those kinds of thoughts.
15:05So our main focus in general is the
15:08class of receptors that Katie discussed.
15:10The growth factor receptors that
15:12have interested Harrison Chinese
15:14domains like EGF receptor.
15:16As you know, and as as key to describe,
15:19these are key targets for cancer therapy,
15:21particularly lung cancer,
15:22and is clear in general in advancing
15:24approaches to controlling their behavior.
15:26So the behavior with drugs dealing
15:28with resistance really requires us to
15:31understand the molecular mechanisms
15:32and understanding well enough
15:34that we can manipulate them in a
15:36predictable way and also manipulate
15:37their complex so the networks.
15:39And I'll give a couple of examples
15:41of things that are driving have
15:44been driving research in my lab.
15:46Recently,
15:46and the first relates to what Katie has
15:49been discussing at the level of growth,
15:51acquired resistance and primary resistance,
15:53and we've actually been working with
15:55Katie quite a bit to understand
15:57details of how secondary mutations
15:59in EGFR cause resistance.
16:01As she mentioned,
16:02with the automotive resistance
16:03mutations and the additional key colon
16:05network is to use that understanding
16:08as it develops to decide when to
16:10use which inhibitor and how to come
16:12up with new and indeed repurposed
16:14inhibitors in resistance situations.
16:16Not going back to two other
16:18working in the lab,
16:19one of our recent first time
16:21has been to identify and target
16:23driver mutations in neuroblastoma,
16:24which is one of the most common
16:27pediatric cancers.
16:27And this is related work we've been
16:29collaborating with the Children
16:30Psychology Group on out another
16:32receptor tyrosine kinase,
16:33a bit like EGFR,
16:35and sequencing out consumers from
16:361600 patients.
16:37That gave us a list with carve
16:39out mutations that we analyzed
16:41biochemically structure.
16:42Real transformation did a full work up
16:44on them and show from that that out.
16:47About 14% of neuroblastoma without dependent,
16:48and we developed a computational
16:50model that you can see in the middle
16:53of the left hand part of the slide
16:55that we can with which we can predict
16:57which mutations are actionable.
16:59Mr Working on that and an in refining
17:01that to identify out dependent
17:02tumors in the clinic and what but
17:05importantly this quickly let us
17:06to understand that some variants
17:08are resistant to 1st generation
17:10out computers result and it does
17:11not work in Europe.
17:12Last over and we also learned that the
17:14stable of 1st generation are contributors.
17:17We're not that different from
17:18one another and and impedes.
17:20In particular, we have one.
17:21We have to be careful to which
17:23drug you choose for the trials,
17:25because there's a limited
17:27number of patients in pediatric,
17:28so more monster pick the right one
17:30and over all those considerations,
17:32using their biochemistry
17:33channel distal mat in it,
17:34which is now looking
17:35promising in neuroblastoma,
17:36overcomes much of the resistance,
17:38although of course we are now experiencing
17:40resistance that we're working on,
17:42and I just want to illustrate that as a
17:43key approach combining biochemistry and
17:45structural biology and computational aspects.
17:47But we could use in principle for
17:49any receptor types in Chinese.
17:51So next slide please.
17:54We also very interested in a new
17:56aspect of getting away from inhibiting
17:58receptors per say as we tend to do
18:01instead correcting their signaling.
18:03So we're all familiar with biased
18:05agonists for G protein coupled receptors,
18:07which can promote different responses to
18:09the same receptors as strength on the left.
18:12The color of signaling,
18:13whether it's orange,
18:14yellow, green or blue.
18:16Many common drugs that we take,
18:18her bias GPS are agonists,
18:20and there's actually a lot of effort,
18:22for example to develop biased agonists of
18:24opiate receptors retaining analgesic effects.
18:26But without the associated
18:28problem problems of the opiates,
18:29we don't do that for receptor
18:31tyrosine kinases.
18:32In the light there traditionally thought
18:34of as being binary signaling systems,
18:36either on or off as an illustrated here,
18:39but we recently showed in the
18:41paper a couple of years ago would
18:43continue to work on that.
18:45Prices have color in their signaling
18:47two and as illustrated on the right,
18:49the same receptor EGF receptor.
18:51Again in this case can give you
18:53can promote self liberation or
18:55differentiation depending in the
18:56same cell depending on which growth
18:59factor is used to activate it,
19:00and this reflects you know
19:02a different dimer structure,
19:03asymmetric or symmetric,
19:04for the two ligands with altered
19:06dimerization and signaling
19:07kinetics that define specificity,
19:09it turns out the mutations in glioblastoma
19:11shift signaling to the right,
19:13making it more proliferative.
19:14That's one of their.
19:16At key issues,
19:17even with small structural changes,
19:18now that we understand the
19:20structural basis for this but through
19:21crystallography and so forth,
19:23we believe that it's possible to
19:25develop biologics that will do the opposite.
19:27Imagine,
19:27for example,
19:28an antibody that could shift EGF
19:30activated in cancer we mutation
19:31allele with living shift signaling to
19:33the left making it differentiative.
19:35This could be a really powerful
19:36approach to signaling,
19:37switching or correcting signaling
19:39from preparations.
19:39Differentiation is actually 1 proof
19:41of principle in that with kit and
19:43stem cell factor that causes that
19:45was been working on.
19:46At Stanford, so next slide please.
19:50And so finally.
19:51We've also been focusing on an
19:54undruggable target the pseudo kinases.
19:56About 10% of the kinases in kind
19:58of is inactive and the blue ones
20:01here on the left in history.
20:03Many of them don't even buy native P,
20:06and these include regions.
20:07Interceptors like Roswick,
20:08PK7 involved in wind signaling
20:10and involved in several councils,
20:12but have been totally
20:13ignored as drug targets.
20:15For the most part.
20:16One hypothesis is that they simply
20:18by switching confirmations to bind
20:20downstream signaling molecules.
20:21We recently determined in this.
20:23Paper in 2022,
20:24referenced here a bunch of
20:25structures and script screen for
20:27small molecule inhibitors to see
20:29if we could bring in principle
20:30drug these in the middle here in
20:32the structure you can see a drug.
20:35It's actually pronounced enable
20:36inhibitor bound to one of these pseudo
20:38kinases that doesn't even bind 80P
20:40and M as shown in the top right.
20:42We've demonstrated using hydrogen
20:44determine change studies that put
20:45out maybe induces conformational
20:47changes in role one as it binds
20:49and so the idea is that this might
20:51inhibit signaling interactions that
20:52naturally there's alot enormous amount
20:53of work to do with selectivity and.
20:56And so forth.
20:57But early studies of signaling effect
20:58suggests that banana can inhibit
21:00went dependent rule one signaling,
21:02and so the idea of sharing centrally
21:03here is that confirmational
21:05disruptors like this could be
21:07valuable tools for understanding.
21:08See Tiffany signaling,
21:09but also targeting them where they play
21:11known roles in cancer and other diseases,
21:14and so far they're all being
21:16hit the articaine ones.
21:17For example, with antibodies.
21:18So that's about my brief summary.
21:20That's all I wanted to say,
21:22so thank you very much for attention,
21:24and I look forward.
21:26To your questions.
21:28Thank you very
21:29much Mark for that wonderful discussion.
21:32Next, I'll be introducing Meghan King,
21:36associate professor of
21:37cell biology and molecular,
21:39cellular and developmental biology.
21:41Program leader in our Cancer Center
21:45and I think notable partly for having
21:47been elected by the her fellow faculty
21:50here at Yale School of Medicine.
21:52As past president of our
21:55faculty Senate equivalent,
21:56the Faculty Advisory Council where she
21:58also showed exceptional leadership.
22:00Sort of in that other realm,
22:03and she's going to be talking to us
22:05about very impactful work regarding
22:08resistance to PARP inhibition.
22:12Alright, so thank you.
22:14I'm also a basic scientist
22:15an over the past decade.
22:17It's really been my interactions
22:19with my colleagues here in
22:21the Yale Cancer Center that is
22:23driven my group with expertise
22:25in genome integrity to really
22:27focus on those aspects that have
22:29impacts for cancer therapies.
22:33So I'm going to start with this classic
22:36example of synthetic lethality,
22:38and that are is specifically PARP
22:40inhibitors in the context of Bracco
22:43Wanan bracket, two mutations,
22:44an, although of course these
22:46therapies have incredible promise.
22:48It's well established now that
22:50the acquired resistance is a major
22:52bottleneck for the durability and
22:54efficacy of these treatments, and so,
22:56how do we tackle this problem and other
22:59opportunities that are presented when
23:02these tumor cells become resistant?
23:04So the approach that we've been
23:06taking is first to start by really
23:09trying to define the genetic basis
23:11of resistance in this context,
23:13and so we know that there has been
23:16real value in crisper screens.
23:18But I think increasingly we're very
23:20excited about the possibility of
23:22circulating tumor DNA sequencing as
23:24well as potential for serial biopsies,
23:26particularly along this axis.
23:27As tumors gain resistance to combine
23:30with genome sequencing as well as
23:32gene expression analysis to provide
23:34new insights into therapy resistance.
23:36And we use a range of models,
23:38although from model organisms to mouse
23:40models to really get the mechanisms,
23:42and of course the ultimate goal
23:44is always to really be able to
23:46leverage the mechanism of resistance,
23:48ideally to come up with new
23:50therapies and so awhile.
23:51Of course we'd like these to be
23:53actionable were really particularly
23:54would like to go beyond that,
23:56and to be sure to consider based
23:58on our mechanistic studies,
24:00what can we bring to the table
24:02in terms of stratification?
24:03And today I'll talk about an example
24:05where we really think that we have
24:07to consider Bracco one patient
24:09separately from bracket two patients.
24:11Of course it would be best really
24:13if we can develop new biomarkers
24:15that will further help us stratify
24:17patients based on the mechanisms
24:19underlying resistance,
24:20and I think one real potential
24:23there is for example,
24:24circulating tumor DNA may allow
24:26us to identify patients who have
24:28a so called reversion allele.
24:30That'll now will make them
24:32insensitive Department of Therapy
24:33and that baby one cohort,
24:35but there may be other patients
24:37where resistance is arising through
24:39a secondary mechanism that maybe.
24:41Therapeutically actionable and so
24:42I just wanted to take you through
24:45the work that we've been doing,
24:47just not just my lab,
24:48but across our team to look at
24:51the genetic basis of resistance.
24:53So much of again,
24:54these crisper screens have been published.
24:56The work that's been going on here
24:58at Yale really has taken advantage
25:00of a partnership that we already have
25:03between Astra Zeneca and our team,
25:05particularly Ryan Jensen,
25:06and Ryan has been modeling reversion
25:08alleles that are arising from
25:10patient derived DNA sequencing.
25:11And testing really,
25:12is there still an actionable approach that
25:15we could use in these contexts or not?
25:17By functionally characterizing
25:18the reversion alleles?
25:19What I'm particularly excited about
25:21at the moment is that paleru so has
25:23been leading a trial along with Kurt Shopper,
25:25who you'll hear from in a moment where
25:27she is and acquiring these serial biopsies.
25:29Along this progression to relapse.
25:31And this allows us now to go in and
25:33really look not just a genome changes,
25:36but gene expression changes.
25:37And so these sequencing
25:38is ongoing at the moment,
25:39and we're really excited about the
25:41new targets that it may reveal.
25:44So I just want to give you one
25:45vignette of what was really originally
25:47motivated by these in vitro screens,
25:50and some work that my group has
25:51done and the possibilities that we
25:53can see for this going forward.
25:55So it's well established that Braca
25:571 one of its key roles is to promote
25:59what's called double strand break and
26:01resection through the EXO 1 pathway.
26:03And this is a critical step in
26:05the HR pathway and so it was came
26:08out of these screens.
26:09That loss of either 50 BP one or Rev
26:117 can drive therapy resistance in
26:13the context of Graco one mutations.
26:15Well, my group discovered is that
26:17these are negative regulators of the
26:19bloom helicase acting with DNA 2,
26:21which is an alternative and
26:22resection mechanism.
26:23So this is a way where these tumor
26:25cells have essentially rewired reception
26:26so they're no longer dependent on
26:29bracca one and instead can use this
26:31bloom pathway and so as examples
26:32of what that mechanism has brought
26:34about in terms of the way we're
26:36thinking about future therapeutics,
26:38the first is that identifies
26:39the bloom helicases,
26:40a really novel target that we have
26:42already shown in vitro is also synthetic
26:45lethal with Bracco one on its own.
26:47Particularly if we think in the short term,
26:50maybe more actionable input
26:51ways in which this has changed.
26:53Our thinking is that it highlights
26:55also the potential for combinations
26:57of PARP inhibitors in ATR inhibitors,
26:59and that's because the other thing we
27:01discovered is that this blue mediated
27:04helicase is driving resection at very
27:06high rates and this leads not just
27:08to functional reception to do repair.
27:10It actually leads to hyper resection,
27:12and ATR is an important negative
27:14regulator of resection,
27:15and so we think that this combination
27:18of treatments will push this.
27:20Hyper resection even further,
27:21and this is a really good rationale for why.
27:24Initially patients with RK one
27:26mutations may not respond well to a
27:28combination with an ATR inhibitor,
27:30but when there is a mechanism that down
27:32regulates these particular proteins
27:33that that will make these tumors
27:35very sensitive to the combination,
27:37and so along those lines we're
27:39currently just submitting anello I
27:41with paleru so where we are proposing
27:43to do a trial specifically in Bracco
27:45in patients because this is not a
27:48mechanism that's relevant for the bracket,
27:50two patients.
27:50Hoping to really test this idea clinically,
27:53so thank you and I look forward
27:55to the questions.
27:55I also just like to highlight that
27:57much of this work as I mentioned
27:59was a collaboration with Astra
28:00Zeneca and is also supported very
28:02generously by the Great Foundation.
28:07Thank you Megan. That was terrific.
28:10Next, I'd like to introduce.
28:14Jinyoung he's an associate professor
28:17of pathology and director of
28:20our Epigenetics program here at
28:22Yale and will be talking about
28:25epigenetic mechanisms of resistance.
28:28Thank you, Barbara. On So, uh,
28:33my expertise in the menu on cancer genetics,
28:36and as you all know I project
28:38magnet is critical for cancer
28:40initiation and progression.
28:42Especially my laptop is interested in
28:44understanding how epigenetic regulators,
28:46also called reader writer and erasers of
28:48being an maceration histone modification.
28:51How regulate different steps
28:52of cancer progression?
28:53My number to your interest
28:55in a couple different areas?
28:57One is resistant mechanism
28:59to anti cancer drugs,
29:01which is the main topic today.
29:03Cancer metastasis and tumor in valuation,
29:05which is one of the areas that
29:08I could show but will talk
29:10more about later on and next.
29:13My #2 is also very interesting,
29:15developing different epigenetic drugs
29:16and and we have done some work with
29:19your Center for molecular discovery,
29:21which is our in house training center
29:23and I have done some work with
29:25the NCI Experimental therapeutics
29:27program and right now I'm also
29:29collaborating some about tech and
29:31pharmaceutical company in this
29:33area as well and and in the next 2
29:36slides I'm going to tell you some
29:38of the examples that we have done
29:41to look at the resistant mechanisms.
29:44Next please.
29:45One which is targeted therapy,
29:47and in this case the transaction number
29:50one called Herceptin for breast cancer,
29:52and we can generate those resistant
29:54cells in tissue culture.
29:55And we found that those resistant cells
29:57actually are do not have genetic mutations.
30:00They actually resistant mechanism
30:01is actually reversible if you take
30:04the drug away from the cells for
30:06short period time and they are
30:08still maintain resistant.
30:09But if you take it away for
30:11a long period time,
30:12for example about months and those
30:15cells becomes those so called watch out.
30:17And those cells become sensitive
30:21to just over again.
30:24To local internal mechanism next piece.
30:27We profile the expression of the
30:29expression profile of the reason
30:32cells compared to the sensitive cells.
30:34We can see that those resistant cells
30:37have increased oxidative phosphorylation
30:39or called off force and remarkable need.
30:42Those cells are very sensitive
30:44to ox force inhibitor.
30:46As you can see the tumor regression if
30:49you combine traditional Antonio Massenet
30:51which is 1 nautical force inhibitor.
30:54You can see regression of those.
30:57Resistant tumors Next place as I
31:00mentioned that this app is genetic
31:03mechanism that contributes to resistance,
31:06so we are one of the mechanism we
31:08found is that Arcadian 5 histone
31:10demethylase are critical for this
31:13formation of those resistant cells
31:15we can combine with the target
31:17therapy and Kaden 5 inhibitor which
31:19this is one of the early generation
31:22inhibitor and four to prevent the
31:25formation of the recent sales for
31:27both breast cancer which is beating
31:29for some report cells.
31:31And non cancer cells on PC 9 cells.
31:35And next race.
31:36So we are also very interested
31:39in understanding how.
31:42Resistant happens to our email
31:44checkpoint blockade and this is
31:46our version of the cancer immunity
31:49cycle and and as you can see,
31:51there's actually 2 steps are the
31:54critical for email checkpoint
31:55to work is the trafficking and
31:58infiltration of the immune cells to
32:00the tumor and apparently some of the
32:03epigenetic modulators have been shown
32:05to be critical for those processes,
32:08and then I will just show example
32:10in our laboratory next please.
32:12Where we found the Canadian
32:15Fire B or history history.
32:18You must nice file B is critical off for.
32:22Infiltration and trafficking
32:24of the T cells to the tumors.
32:27And if not colocating 5B,
32:29I I in those Yamaha 1.7 cells,
32:32which is more smaller,
32:34generated by Markus Persson.
32:35Book idea we can see that if you
32:39knockout account info be those
32:41cells are unable to form tumors.
32:44And if we re challenge,
32:46those are two mice with control sales,
32:48which normally grow very well.
32:50You can see they cannot grow and
32:53meaning that those might have gained
32:56immunity against those younger cells.
32:58If you look at the the pony IMO Genic
33:01young one point cells down in the policy,
33:05you can see those cells are not
33:07responsive to PD one blockade at all.
33:10And if we do need killing
33:12file before those cells,
33:14you can see the slowdown of the growth
33:16of cells and if you combine with PD
33:19one blockade you can significantly
33:21extend the lifespan of those miles.
33:24To my very mice.
33:26So this suggests that can you
33:28invite me is that very good target
33:30to overcome resistance to email,
33:33check one blockade and I would just
33:36want to mention that this is done in
33:39collaboration with multiple laps and yell,
33:41including archical,
33:42even sucking and much boesenberg snap.
33:45So team science is one of the same idea.
33:48We workout together or not.
33:51Thank you.
33:54Thank you that is such a terrific story.
33:57Now I'm pleased to introduce Curt Shopper.
34:00He's an assistant professor
34:02of pathology and medicine.
34:03An recent rooms at the
34:05end of an NCI Merit Award.
34:08He conducts really cutting edge
34:10immuno profiling studies and
34:11look forward to your talk Kurt.
34:15Thank you, Barbara. Next slide please.
34:18So I I trained clinical molecular
34:21diagnostics that I've been working in
34:23cancer immunology for about 10 years now,
34:25and it's unquestionable that immuno
34:27oncology has really revolutionized
34:29the treatment of cancer.
34:30But there are major challenges
34:32still to overcome,
34:33so I'll cover a few of the challenges
34:35that I think are critical to
34:38potentially move the few forward,
34:40one of which is that I think there
34:42have been conceptual limitations
34:44of in both in drug development
34:46and identification of biomarkers.
34:48Relative to drug development,
34:49I think the focus of many people developing
34:52targets has been on immuno stimulation,
34:55but that doesn't necessarily consider
34:57correcting alterations in the tumor
34:59and this is critical because if we're
35:02only stimulating T cells we are and
35:04there is not a clear gradient towards
35:06activating it more in the tumor.
35:08It's likely that the therapeutic index
35:10is smaller and the potential benefit
35:13and toxicity balances is affected.
35:15So I think the concept is that we shouldn't
35:18focus only on stimulating T cells everywhere.
35:21We should probably look for.
35:23Signals that have a gradient favoring
35:25the tumor in relative to the development
35:27of biomarkers for resistance.
35:29I think there have been a little bit
35:31of confusion in the field because it
35:34mean a therapy has been used so widely
35:37that people are calling every patient
35:39that don't respond as a resistance.
35:41And conceptually I think that's probably
35:43not accurate because patients without PD
35:45L1 expression tumor mutational burden,
35:47any biology should not respond to start with,
35:50so I think there is a confusion between.
35:53Any patient that Blacks benefit
35:55versus true resistance,
35:56which in my opinion are the patients
35:58that should have responded but didn't.
36:01I think this is critical to design
36:03programs and biomarker plans.
36:05The second important concept that
36:06it's connected with the previous
36:08one is that it's probably necessary
36:10to identify dominant immunization
36:12pathways that are well represented.
36:14The tumor and this is for the same
36:17reason because we need to have
36:19this gradient and strong biology
36:21in the tumor to be able to.
36:23Achieve a meaningful anti cancer response
36:26and then another major need in the
36:28field is trying to identify potential
36:30targets that are beyond the T cells.
36:33So to have complementary effort and
36:35not have only redundant mechanisms,
36:37another important observation is that
36:38we as I follow you know when we look at
36:42the tumors we realize how difficult and
36:44how complex is the tumor micro environment.
36:47Where where is most interactions between
36:49tumor and immune cells are happening.
36:51And I think the suffering.
36:53The tumor micro environment and how different
36:56is across tumors and across patients.
36:58It's a major need to really drive better
37:01biomarkers and better immunotherapy.
37:03Then also I think we need to do a better
37:05work at understanding the interactions
37:08between major dominant oncogenic
37:10signals and immune evasion pathways.
37:12This has been somehow being revealed
37:14in EGFR mutant tumors that are
37:16less sensitive and less inflamed,
37:18but they I think there's a whole
37:21world to discover.
37:22What alterations in the tumor,
37:24somatic alterations are able to manipulate.
37:26It means an immune response.
37:28And then finally,
37:29I think there are limitations of traditional
37:32studies as we just solve from Jane.
37:34Many alterations are not the genomic level.
37:36Which is the favorite way we used to
37:39analyze the tumor site of the interaction.
37:41So I think by just doing genomic analysis,
37:44we're missing a lot of alterations
37:46that the immune system and this.
37:48I think it's it's something we can
37:50overcome and finally think that most of
37:52the studies are focusing on both ends on
37:55the very early discovery type of work,
37:57with crisper screens and other strategies.
37:59And then there is a huge effort
38:01on the clinical development,
38:03but I think there is room to improve some
38:06studies in more sort of human real context.
38:09Next slide, please.
38:12So this is an example of the approach
38:14that we have taken in my group where
38:17we generate hypothesis using discovery
38:19in biology and then we actually have
38:22generated assays to screen for pathways,
38:24cell types in tumor cell indicators
38:26in the same issue.
38:27So we can actually do both genomic
38:30analysis to understand the
38:31genomic context during drivers,
38:33but then we can also look at the
38:35immune contexture and pathways
38:37that are potentially actionable.
38:39We have become pretty good
38:40at looking at multiple.
38:42High throughput methods to
38:43detect protein level and then we
38:45can do single cell analysis,
38:47spatial analysis and really try
38:49to understand the tumor micro
38:50environment to prioritize what
38:52signals are dominant or relevant,
38:54we usually use aggressive analysis using
38:56outcomes and response to treatment.
38:57So that way we can identify which
39:00signals are relevant from the ones
39:03that are not next slide please.
39:05This is important because ultimately
39:06those signals are the ones with.
39:08Then we can validate in vitro to
39:10demonstrate that these are not just
39:12epiphenomenon's or correlations,
39:14but they are mechanistically relevant
39:15and then ultimately we can go back
39:18and look at this in the context
39:19of human clinical trials,
39:21and I'll show you an example
39:23of that next slide, please.
39:26So just for to illustrate how this
39:28cycle works, this is a story that it's,
39:32uh,
39:32have published this year where we
39:34identify Interleukin 8 and local
39:36neutrophils in the tumor micro
39:38environment as dominant immunization
39:39pathway and resistant mechanism.
39:42So the story started a few years
39:44ago where we look at inside too
39:47aren't expression for Interleukin 8,
39:49and we found that it was producing
39:51tumor cells and highly associated with
39:54resistance to immune checkpoint blockers.
39:56So to advance this further,
39:58we look at the relationship between
40:01Interleukin 8 and neutral fields as
40:03shown in the upper side of the slide,
40:05and then we found a fraction of
40:08tumors that had up regulation of
40:10Interleukin 8 and an unfavorable
40:12micro environment characterized by
40:14increased deals in fewer T cells.
40:16We also did genomic analysis to
40:17understand that this was independent
40:19from tumor mutational burden
40:21and major genomic alterations,
40:22and then we finally were able
40:24to demonstrate that the
40:26production of Interlake in the tumor.
40:28Was actually associated with
40:30interleukin 8IN serum in circulation,
40:32so we that we conducted an studying
40:35over 1200 cancer patients from three
40:38phase three pivotal trials using immune
40:41checkpoint blockers and we found that
40:43about 1/3 of a patients across tumors
40:46have up regulation of interleukin Aiden.
40:49They have low sensitivity to
40:51immune checkpoint blockers.
40:53Next slide, please.
40:56Then to further demonstrate this,
40:58we need another study in which we
41:00cultured neutrophils and my Lord arise
41:02suppressor cells to show the mechanism
41:04behind and we were able to demonstrate
41:07that formation of Nets was involved
41:09in affective response suppression,
41:10and then ultimately we're working
41:12with the clinical trial where
41:13patients are being treated with an
41:16antibody and targeting Interleukin 8,
41:17and to understand if this pathway can
41:20actually be action in real patients,
41:22and hopefully we can use the biology
41:24that we figure out to drive.
41:26The biomarker plant next slide please.
41:30So finally we have gotten a little bit
41:33more sophisticated now and generated
41:35models or in in vitro tumor treatment.
41:37And this is just an example of what we're
41:40doing where we can culture primary tumors,
41:44treat them in vitro but intact so that
41:46we can then generate preparations and
41:48analyze the tumor micro environment.
41:51Change now perturbing these tumors
41:53with immunostimulatory or other
41:54anti cancer agents and we are
41:56incorporating new technologies such
41:58as single cell transcriptomics.
42:00Another analysis to do more unbiased studies.
42:03Thank you.
42:05Thank you Kurt. I mean I think probably
42:08everybody can see the incredible
42:10power of that of that approach.
42:13Well, we said at the outset,
42:15Yale engage is focused on building bridges
42:18and and collaboration with industry,
42:20and in each of these seminars,
42:23we've invited an industry
42:24partner to speak to us,
42:27and I'm really thrilled that today,
42:29it's Susan Galbreath she's a senior
42:32Vice president and head of early
42:34oncology R&D and Astra Zeneca.
42:36She's been there about 10 years and.
42:39In the early development program,
42:42there brought 7 compounds into phase three.
42:47The story with. PARP inhibition,
42:51the third generation EGFR inhibitor.
42:54Awesome Merton if that our colleague,
42:57Roy Herbst, was involved in presenting
43:00very impactful angemon trial this year.
43:02Megan met inhibitors selective
43:04estrogen receptor directed agents.
43:06Really phenomenal portfolio and a
43:09phenomenal track record of success.
43:11So Suzan,
43:12we look forward to hearing your thoughts.
43:17Thank you, Barbara Ann.
43:19It's a it's a pleasure to be here
43:21with you and just a bit introduction.
43:23I'm a clinical psychologist by
43:24training MD PhD and I've been,
43:26as Barbara said, Astra Zeneca for 10
43:28years and before that I was in the
43:31US with Bristol Myers Squibb also
43:32in the early Development Group and
43:34and stayed there for about 9 years.
43:36Just go on to the next slide.
43:38I want to talk a little bit to build on
43:41some of the thoughts we've got about,
43:43you know, understanding resistance
43:44and one of the challenges that we've
43:47got about understanding resistance is
43:48really having access to the samples.
43:50That would enable us to understand
43:51the clinical resistance.
43:52So Katie Elite is already talked to you
43:54about some of the models that we can
43:56use pre clinically to model resistance.
43:58One of the challenges we've got
44:00with those techniques though,
44:01is that it doesn't always predict
44:03what the true prevalence of the
44:05resistance mechanisms is going to
44:06be in in the clinical setting.
44:08So if you start off with a PC 9
44:09so when you look at the mechanisms
44:12of resistance to that,
44:13you don't necessarily understand
44:14what the true prevalence of all the
44:17things are when patients are starting
44:18with their with their own set of.
44:20Wiring diagrams in their EGFR
44:22mutant lung cancer.
44:24The other challenge that you've got is
44:26tried for number of years to actually
44:28get biopsies from patients on at the
44:30time of progression in clinical trials,
44:32or must we concluded that you know
44:34typically has to be as an optional biopsy.
44:37At that time of progression,
44:39we've actually heard across
44:40the range of clinical trials.
44:42Relatively few of those
44:43actually materialized,
44:43and so that means that our mechanisms
44:46of understanding resistance during
44:47the development of certain IP,
44:48you know, have been somewhat limited.
44:50We started right the beginning by
44:52looking at circulating tumor DNA,
44:53it right from the phase one
44:55trials with awesome antonym,
44:57and we have some understanding of
44:58actually published some of the
45:00data from the first line study
45:02with a semantic that flora trial
45:04looking at those CT DNA mechanisms,
45:05but really actually one of the
45:07things that comes out of that is,
45:09we could only explain.
45:11I am just over 1/3 of the patients
45:13resistance mechanisms through looking
45:15at city DNA and the patterns that
45:17we saw there was we saw their city.
45:20The emergence of the Sistine
45:21797 S mutation met amplification
45:23PSP KEARNEYS pathway mutation.
45:25An activation fee 10 losses and
45:26in some cases and MEK pathway
45:28activation as well in some cases.
45:31But the really the majority of patients
45:33we still had a question mark over
45:35what the resistance mechanisms worth.
45:37So that led us to design that
45:39this kind of study.
45:41It's called the Orchard and platform study.
45:43This takes patients that
45:45we're progressing on.
45:46First line automotive,
45:47and it offers them something that
45:49is potentially of potentially
45:50of benefit to them,
45:52which is to take a biopsy to
45:53look at what
45:54the data says on next generation sequencing.
45:57From that biopsy and then to allocate
45:59them to a range of different potential
46:02arms and this biomarker matched arms
46:04which you can see above depending on the
46:07mechanism that that is seen with resistance.
46:10And then there's also non biomarker match on.
46:12And this has been an important component
46:15of many platform trial designs because
46:17it means that every patient whose
46:19given a consent to have a biopsy
46:22gets the offer of something.
46:23I can't guarantee that than what they're
46:26getting offered is necessarily going to work,
46:28but it gives them that,
46:30and that has driven really
46:31quite a good uptake in terms of
46:34enrollment and accrual in this.
46:35And actually, what one of the things
46:38that we've already learned now is,
46:40we've now got, you know,
46:42data and over 60 patients.
46:43You know,
46:44with with tissue available at the time of
46:47progression in in the in the Orchard study,
46:50and now that we can we have an
46:52identifiable resistance mechanism now,
46:54in the in nearly 2/3 of patients,
46:56as opposed to just a third.
46:58We've increased the detection and
47:00some of the amplification mechanisms
47:02which can be under estimated using
47:04CT DNA would increase the detection
47:06of some of the Fusion mechanisms,
47:08which can also be difficult to
47:10detect using the CT DNA techniques.
47:12And we've got a better sense.
47:14With the prevalence,
47:15there's still some work to be done here,
47:17and I still think we need to look at
47:19the epigenetic mechanisms that are
47:21driving resistance in this setting,
47:23but I just wanted to illustrate this
47:25as a as an example of one way that we
47:28need to look at in terms of understanding,
47:30documenting resistance and moving on
47:32from it so we can go to the next slide.
47:35The similar approach has been
47:36taken in the understanding.
47:38Resistance to checkpoint inhibition,
47:39and I completely agree with shoppers comment,
47:42but not everybody who progress
47:44is on a checkpoint inhibitor
47:46is necessarily truly resistant,
47:48but I think we need to understand
47:51some of those mechanisms, and again,
47:54this is a mechanism where you can get
47:57the biopsies from these patients.
47:59Also,
48:00some peripheral blood sampling
48:01and look at ways in which we
48:04can potentially offer them.
48:06Treatments that may have the opportunity
48:08to to make it make a difference.
48:11So again,
48:11I just want to share with you a
48:14couple of observations from this.
48:16We're going to the next slide.
48:19So first of all,
48:20there are some mechanisms that we
48:22might anticipate seeing based on,
48:25you know,
48:26really good data that's already emerged,
48:28and this is about the loss of
48:31her Psycho City for HLA or MHC
48:33and we are seeing as expected.
48:36But after treatment or one
48:38of these checkpoints,
48:39inhibitors and increased
48:40prevalence of loss of HLA or MHC.
48:43In the inability of the of the tumors
48:46to be seen by an an an an effective
48:49by at the adaptive immune mechanisms
48:52of if the antigen can't be presented
48:56effectively, it's like.
48:58Other things that we're doing.
49:00We've seen a range of different
49:02mechanisms that we have.
49:04We have looked at in this setting.
49:06Wilson mentioned the fact that
49:08obviously we're looking at the
49:09ATR combination with a lap robe
49:11in terms of part resistance,
49:12but in fact actually one of the
49:14observations that we made earlier phase
49:16one with our selected slot assertive,
49:18which is, uh, ATI inhibitor,
49:20is that we were seeing some unusual
49:22responses in patients that had a
49:23prior checkpoints in innovation.
49:25In some other trials,
49:26and so that led to some further
49:28investigation and so so there are
49:30certain underbelly map is one of the
49:32arms in the in the Hudson study and
49:34some of the data that we're seeing
49:36is quite interesting in seeing that.
49:38Getting a decrease in exhausted T cells,
49:41exhausted NK cells and an increase in
49:43antigen presentation in patients that
49:45have both got primary resistance to
49:47checkpoint inhibition and subsequently
49:48had some degree of response and
49:50subsequently progressed as well.
49:52And we're also seeing it not just in the
49:55ATM mutant patients that are selected,
49:57but also more more broadly, so.
49:59This is just an interesting observation.
50:01There's a lot more mechanistic data that
50:04is required and that will be followed
50:06up in order to understand this better.
50:08But I do think that these kinds of trials
50:10are really helpful in trying to understand
50:12the clinical prevalence of resistance.
50:14Mechanisms get a lot more
50:15data that can feedback,
50:17and you know,
50:17back with the preclinical work
50:19that we can do to them to then
50:21understand what we might do next.
50:23So I'm going to stop there and
50:24I'm very happy to address any
50:26questions that you might might have.
50:28Thank you.
50:29That was
50:30fabulous. Thank you very much.
50:33I am now going to ask that all of the
50:37panelists turned on their audio and video
50:42and will now go into the the full discussion.
50:47And I'm going to ask the attendees to
50:52please continue to post questions we we are
50:56monitoring these and the first one, I think.
51:01Basically immediately follows that the
51:04last slide that we saw and so maybe I'll
51:08ask Susan and Kurt both to address this.
51:12How critical is it to overcome the
51:15mechanical functional barriers to immune
51:17checkpoint inhibitors and the question
51:20relates specifically to HLA loss,
51:22although I can think of other
51:25mechanisms related to hypoxemia and
51:28and vascular alterations as well,
51:30but can you please comment on?
51:33Potential pathways and targets to overcome
51:35mechanical and functional barriers to
51:37immune checkpoint inhibitors and Susan.
51:38Do you want to go first and then
51:41kick it to to Kurt?
51:43Yeah well, the the the the Council
51:45you think of when I think of 1st when
51:48you're talking about mechanical barriers
51:50potentially is is of pancreatic cancer.
51:53Cause at the high level of you know
51:56Disney plastic streamer that you see that
51:58you see there that has been discussed
52:01as not just having actually a physical
52:04potential barrier to treatment but also
52:06the presence of the constituents of that.
52:09Desmond plastics.
52:09German may also have a you know,
52:12biochemical effects that reduce
52:13the likelihood of sensitivity to.
52:15Of the tumor cells that are
52:17adjacent about two treatment,
52:19and I think there are a lot of data
52:21suggesting that understanding the
52:23components of the micro environment,
52:25the distribution and types of you
52:28know cancer associated fibroblasts,
52:29for example,
52:30and not in that disease,
52:32and their feelings that might be
52:34absolutely critical to understanding
52:35mechanisms of resistance and sensitivity.
52:38I think in the context of loss of HLA.
52:41It it's you know that you know lots
52:44of HLA may increase the sensitivity
52:46potentially to other mechanisms like
52:49inducing the innate immune system
52:51rather than the adaptive immune system
52:54to NK cell enhancement potentially.
52:56Then you know so.
52:58So there are things that then
53:00creates a formability I suppose.
53:03I think the issue from my perspective is
53:06it you know you wouldn't be expecting.
53:09No high likelihood of subsequent
53:11response to something that requires
53:13HLA antigen presentation.
53:14If you've got lots of HP laser
53:17fundamental mechanism,
53:18so we should be segmenting patients by
53:21an understanding of these mechanisms
53:23in order to identify the populations
53:25that might best be subsequently treated
53:28with different kinds of therapies.
53:30Cut any thoughts from you.
53:33Yes, I agree with all the comments.
53:35I think there is more biology emerging
53:37suggesting that the mechanical
53:38barriers may not be so mechanical.
53:41You know some of these fibroblast basic
53:43read inhibitory molecule so it may be
53:46also an active immunity victory component
53:48to that and that I think is driving.
53:50I think they were going to see a lot of
53:52new studies showing active mechanism of
53:55rejection of immune cells in the tumor bed
53:58and relative to the empty in presentation.
54:00We have actually a study under review
54:02that should see the light soon.
54:04When we look at large
54:06cohorts of tumor mapping,
54:08different parts of the antigen presentation
54:10pathway in a Long story short where we've
54:13learned is that when we look at the genomics,
54:16we don't see that.
54:18So the majority of alterations are non
54:20genomic meaning non mutation related.
54:22In the second interesting lesson is
54:25that depending on what molecule is
54:27lost in the tumor cell meaning HAHABCV,
54:30A2M or other proteins,
54:31the immune contexture changes.
54:33So so I think.
54:34Understanding that part will be critical
54:36to understand how to treat those patients,
54:38we do see upregulation of natural
54:40killer service in in certain loss.
54:42Eventually molecules,
54:43but not in everyone,
54:44and each of them has sort of a certain
54:47different balance between T cells,
54:49NK cells, and other cells.
54:51So I think it will be critical to do
54:53those studies to understand how granular
54:55disease and if we can lump the antigen
54:58presentation defect into one category.
55:00Or maybe it will be more than that.
55:02I think that's to be figured out.
55:05So just continuing on with this theme
55:08in in a question for Chin can HLA loss
55:12be overcome by epigenetic modification?
55:15Or what is epigenetic role in HLA loss?
55:19So this is not an area I have been
55:23working on very well having it,
55:25but I could just mention another
55:28with those changes are non genetic
55:30changes so we have different
55:32tools to execute those jeans.
55:34Reactivate those jeans and
55:36to make them successful too.
55:38Make make them to be sensitive
55:41to our treatment.
55:43So email checkpoint blockade will
55:46work if you re reactivate those.
55:51Terrific terrific, I have a question
55:54that was submitted earlier, but I think.
56:00Could probably be answered extensively or
56:02exhaustively by each one of the panelists,
56:05but maybe I'll ask Katie and
56:07Mark to start on this one.
56:09How does the mutational landscape of a
56:11tumor affect resistance and sensitivity?
56:13And I'm interpreting that the questioner
56:16means the other mutations besides
56:18the one in your target molecule.
56:22Thank you sure I can.
56:24I can get started with that.
56:27I think this is really an area that
56:30we are starting to learn more about
56:33as we have learned more about the
56:36mutational profiles of tumors and of
56:39different genetic subgroups of tumors.
56:41So now one of the things that
56:45we've been able to look at,
56:47for example, are in if we think
56:50about lung cancers in different.
56:52Oncogenic driver subgroups.
56:53We can look at the pattern of Co occurring
56:57genetic alterations that happened,
56:59so I'm thinking about for example,
57:02in K Rasputin lung cancers,
57:04these can Co occur with P53 mutations.
57:07They can Co occur for example
57:10with mutations in STK 11,
57:12also known as Elchibey one.
57:14And we're really beginning
57:16to learn about what it means.
57:18If the tumor has Akira's mutation and
57:21a P53 mutation versus ACARAS mutation.
57:24And then Elchibey one mutation for example.
57:27And what and that the LKB one meeting
57:31tumors seem to have a different or
57:35reduced sensitivity to immunotherapy
57:37treatment, for example, and.
57:39In parallel,
57:41I think similarly with targeted therapies,
57:43we're really starting to scratch
57:46the surface and really beginning to
57:48start to understand how different Co
57:51occurring alterations also impact
57:53response to targeted therapies.
57:55So for example,
57:56some of the work that we've been
57:59doing recently looking at different
58:01tumor suppressor gene alterations
58:04in EGFR mutant lung cancer and
58:06how they affect sensitivity to
58:09tyrosine kinase inhibitors.
58:10One of the things that has emerged
58:13from our studies in animal models,
58:15an also is emerging from studies of patients.
58:18Patient specimens is that if you have
58:21EGFR mutant tumors that also have
58:23mutations in the keep one access,
58:25so the keep 1 NRF 2 access that
58:28is important for the antioxidant
58:30response of a tumor cell.
58:32If you have mutations that Co occur in
58:35that path where you have a decreased
58:38sensitivity to tyrosine kinase inhibitors,
58:40so the tumors will shrink
58:42less on treatment with.
58:44These targeted therapies,
58:45and so that begs the question,
58:47is that a subset of patients who you could,
58:51for example,
58:52select initially for treatment
58:54with different therapies,
58:55or for combination therapies
58:57together with a tyrosine kinase
58:59inhibitor so that you could.
59:01Improve outcomes in patients
59:03with that disease.
59:04I think of course,
59:05this these types of landscapes
59:07also this studying these landscapes
59:09really requires a lot of mechanistic
59:11investigation to understand exactly
59:13what is happening in those tumors.
59:15Finally,
59:16I think one of the other things to
59:19think about in terms of the genetic
59:21landscape also has to do with the
59:24overall mutation burden and the
59:26overall tumor mutation burden, which.
59:28You know we talk a lot about it
59:31in the context of immuno therapies
59:33and where you know we've.
59:35We've heard about a lot about it
59:37in in recent years.
59:38I'd say also there's some evidence that
59:40in the context of targeted therapies,
59:43the overall genetic landscape or the
59:45tumor mutation burden can have an
59:47effect on the response to targeted therapy.
59:49So again in EGFR mutant lung
59:51cancer tumors that seem
59:52that have that are in the highest
59:54tertile of tumor mutation burden,
59:56which is generally lower
59:58than most other lung cancers.
59:59But in that highest circle seemed
01:00:01to do worse on treatment with
01:00:03targeted therapies with tyrosine
01:00:05kinase inhibitors and the ones with
01:00:07the lower two mutation burden.
01:00:09So there are lots of different
01:00:11aspects to consider.
01:00:12The specific mutation.
01:00:13So qualitatively but also quantitatively.
01:00:16Yep,
01:00:16I was just at the office or at a kind of.
01:00:22Broad conceptual thought to
01:00:24that which is ultimately,
01:00:26I think, with all of these,
01:00:29with all of the therapies.
01:00:31We're talking about, one is really
01:00:33trying to correct the signaling network.
01:00:36However you define network,
01:00:38whether its intracellular intra tissue,
01:00:40Inter intra Organism.
01:00:42Once regular network and in a sense
01:00:45if you think about the fact that
01:00:48cancers are really caused by the
01:00:50networks losing robustness and
01:00:52kind of careering out of control to
01:00:56uncontrolled proliferation so far.
01:00:57It's almost surprising actually.
01:00:59The targeted therapy can work,
01:01:01and indeed, actually,
01:01:02if you create models where you
01:01:04just mutated something,
01:01:06we're hitting with a targeted
01:01:08therapeutic and nothing else.
01:01:10You don't actually.
01:01:11But that's not enough to cause cancer,
01:01:14so the context is key,
01:01:16and the targeted,
01:01:17the target that we're trying to correct is.
01:01:21It's really just kind of an
01:01:23Achilles heel in the sense for the
01:01:25rather plastic tour in some sense,
01:01:27so I think I think that the answer
01:01:30the answer to the question is that we
01:01:32need to think about these things as networks.
01:01:35We need to get into considering
01:01:37the systems biology of this.
01:01:39I think there are two ways
01:01:41of thinking about that one,
01:01:43and you'll be aware of this as the
01:01:45enormous effort put into machine learning,
01:01:47AI types of approaches,
01:01:49whereas we collect more and more data.
01:01:51For the mutational landscape to
01:01:52try to understand their with with
01:01:54various their principle components,
01:01:56analysis and what have you, what.
01:01:58How we can correlate combinations
01:02:00of mutations with sensitivity
01:02:02and so on so forth.
01:02:03But there's another element I think
01:02:05we have to consider the a variety
01:02:08of systems biologists are taking,
01:02:09which I think is is really key.
01:02:12And actually I think RAF inhibitor
01:02:14resistance illustrates this very nicely.
01:02:16Is that we we can actually learn an awful
01:02:18lot about how the networks operate,
01:02:21you know?
01:02:21A classic example is if you
01:02:23ever ask mutation,
01:02:24then the graph inhibited
01:02:26does the wrong thing,
01:02:27you know,
01:02:28but the bottom line is I think that
01:02:30we really we need to start thinking
01:02:32beyond the targets to the networks
01:02:34and what the effect of the targeted
01:02:37therapeutics is on the networks and
01:02:38that that of course is going to hold
01:02:41in the immune context too because
01:02:43again what you actually correcting
01:02:44as as as curtain Susan pointed out
01:02:47what you're actually dealing with,
01:02:48that is,
01:02:49is trying to restore balance in an incredibly
01:02:51complicated interstellar the network.
01:02:53So I think there's a couple of
01:02:55perspectives I would like to.
01:02:57Could you could you just answer
01:02:59that by introducing people a
01:03:01little bit to what's going on at
01:03:03the Systems Biology Institute.
01:03:05Not all of the audience
01:03:07may may know about the scale of the effort.
01:03:10Yeah indeed. So actually at yeah we
01:03:12have a system colleges too that's
01:03:15actually headed up by Andra Chanco,
01:03:17who's the Director of Vietnam.
01:03:19Very is also a Pi on one of the
01:03:21NCI cancer Systems biology centers.
01:03:24There's a NCI has a physical
01:03:26Sciences long college center that.
01:03:28Actually, that initiative that you're
01:03:30involved with one of those at Penn and and
01:03:33also the system cancer systems biology,
01:03:35can sort centers consortia
01:03:36that 100 grand sent,
01:03:37and so the Systems Biology Institute
01:03:39here is really very council focused.
01:03:41It has a lot of interactions
01:03:44with the Cancer Biology Institute
01:03:45on also on West Campus at Yale,
01:03:48and Andre is an integral part
01:03:50of the Cancer Center.
01:03:51Of course, the Council biologist is too,
01:03:54and so are most members of the Systems
01:03:56Biology Institute and the kinds of things.
01:03:59Being looked at there,
01:04:00which is actually related to this,
01:04:02so that Barbara it's a good point
01:04:04are for example Andre is very
01:04:07interested in looking at.
01:04:08But it sells from brain tumors
01:04:12in particular and asking.
01:04:15Adam and looking at at migration versus
01:04:18proliferation in those in those cells,
01:04:20and the epigenetic difference between
01:04:22between those in terms of what's
01:04:25defining the signaling networks that
01:04:27cause the cells to behave very differently.
01:04:30And he can do that with some
01:04:33microfabricated devices where you can
01:04:35separate cells based on how rapidly
01:04:37they can migrate and then go in and
01:04:41look with various single cell and
01:04:43other technologies to look at their.
01:04:46Transcriptome and obviously
01:04:47gentleman and epic transcriptome etc.
01:04:48So that's really very exciting.
01:04:50There's another element which is
01:04:52really at another #2 elements of
01:04:55that is one of city channel that
01:04:57many of you will come across.
01:04:59City is doing an awful lot of work in
01:05:02terms of it in the system bars Institute.
01:05:05Two in terms of trying to understand
01:05:08using using in vivo CRISPR technologies.
01:05:10Origins of resistance to terminate
01:05:12therapies and another therapeutic approaches.
01:05:14And so that's really very exciting again.
01:05:16Is is plugged into the network
01:05:19consideration of what's going on,
01:05:20and then one other aspect,
01:05:22which I which I think is really cool.
01:05:25Actually as a project in our in EU 54
01:05:28that Andre Valances that Gunter Wagner,
01:05:31who's an evolutionary biologist is
01:05:33is very interested in why certain.
01:05:35Mammals don't tend to get meta
01:05:37static cancer cows in particular.
01:05:39As an example that caused him
01:05:41not to die of cancer,
01:05:43they just they carry them
01:05:44around with the carry.
01:05:46The tumors around with him
01:05:47without metastasize Ng,
01:05:48and pretty much we don't see them here
01:05:51because they all killed before they
01:05:53get to that stage for other purposes.
01:05:56But but and so that's fascinating.
01:05:58And so again,
01:05:59it's looking at the network context.
01:06:01The network differences between the
01:06:02mammals that do and don't suffer from
01:06:05metastatic cancer, and actually.
01:06:06It's kind of related in some senses
01:06:09to placental invasiveness in some.
01:06:11In these of these organisms,
01:06:13which kind of makes sense in some ways.
01:06:16And so. So there's a cost, if any.
01:06:19There was a cost in some ways of of having
01:06:22placenta that more interdigitate ihd,
01:06:25which is that you more susceptible to
01:06:27metastatic metastasis in your cancer.
01:06:29So so so these different perspectives,
01:06:32whether it's immune.
01:06:33Immunology approaches targeted therapeutics.
01:06:34I think there's a a burgeoning
01:06:37and is very strong at.
01:06:38Yeah, I'm really very exciting.
01:06:40I think a burgeoning effort
01:06:42and understanding of how to put
01:06:44this into the quantitative,
01:06:46and I think it needs to be quantitative
01:06:48in the sense of biochemical
01:06:50networks and pathways contexts.
01:06:54Wonderful. Oh, it's perfect.
01:06:56I have a question here asking
01:06:59the panel to comment on the role
01:07:02of proteomics and studying tumor
01:07:05resistance, and maybe you'll
01:07:07you'll take that first mark in
01:07:10the market. Only relates
01:07:11exactly to what I was saying.
01:07:14I mean, you know you as a biochemist.
01:07:16My view is that. Biochemist,
01:07:19but my chemistry is defined by the
01:07:21component by the combination of
01:07:23components that you have and so
01:07:25and a lot of those are proteins.
01:07:28Of course, it's not just proteins and
01:07:31that there are lots of other things too,
01:07:34but oh mix of various sorts are crucial
01:07:36for really getting a quantitative handle
01:07:38on an understanding signaling itself.
01:07:41Response to therapeutic have
01:07:42course therefore resistance,
01:07:43and I think there are two components of it.
01:07:47What is just at and I would just want
01:07:49to stress this one is just the I guess
01:07:53fingerprinting approach that one
01:07:54often sees with pretty and say what
01:07:57proteins are there in the snapshot?
01:07:59What metabolites are there in
01:08:00Tableau makes sense in the snapshot.
01:08:02That's one aspect and that can give you
01:08:05a lot of information but but looking
01:08:07at changes in the proteome changes in
01:08:10the metabolon with time is really a
01:08:12crucial aspect is really that's what what?
01:08:14What gives us a picture of the networks
01:08:17that we're trying to correct and corral.
01:08:19When we're targeting them in with all
01:08:22of therapeutics that that we discuss
01:08:24it and I just point out that there's a
01:08:26quite a lot of activity on this at Yale,
01:08:29and one of the people recruited
01:08:31into the Cancer Biology Institute,
01:08:32for example, is yeah, Shane Lou,
01:08:34who has been doing a lot of work looking at,
01:08:37for example.
01:08:38Proteomic Lee,
01:08:39both snapshots and time evolution
01:08:41of protein contents.
01:08:42It is considered an employee,
01:08:44and you know it's kind of interesting
01:08:47you really.
01:08:48If you look at any point,
01:08:50sells the effects on the protein
01:08:52were really not what you would have
01:08:55predicted based on on what you've lost
01:08:58in terms of of gene copies or gain.
01:09:01And Moreover,
01:09:01it's important to note that
01:09:03reaction is really pioneered.
01:09:05This too that RNA seek data and proteomic
01:09:07data have substantial discrepancies,
01:09:09and so and so that means also
01:09:12appreciate the proteomics.
01:09:13Is really an important thing
01:09:15to add to all of this code.
01:09:19Yes,
01:09:19so stay think I would like just to comment
01:09:22about the frustration of analyzing tissue
01:09:24level data without spatial resolution.
01:09:26This is critical because as a smart pointed,
01:09:29we see a striking difference between
01:09:31our name protein which is becoming
01:09:33the rule more than the exception.
01:09:35But second, the protein measurements and
01:09:37any other like really is context dependent.
01:09:39So for example, just to give you
01:09:42a rough example measurement of K
01:09:4467 protein in any given sample,
01:09:46if it's in the tumor cell,
01:09:48it means that you know.
01:09:50Tumors are proliferating if it's
01:09:51in the immune cells means good
01:09:53T cells are expanding,
01:09:54so really I think having the possibility
01:09:56of looking at the proteins in the in
01:09:59the context of tissue organization,
01:10:00it's critical for understanding
01:10:02what's going on,
01:10:02and I think that's a little bit when I
01:10:05tried to reflect in my presentation,
01:10:07we're still early.
01:10:08We're getting more quantitative
01:10:09than throughput.
01:10:10It's coming up to speed,
01:10:11but but I think it's a.
01:10:13It's an important dimension
01:10:14of the protein and any
01:10:16other light data. Absolutely perfect
01:10:17and Megan. Do you want to just jump
01:10:20in on this one as well? I just wanted
01:10:23to add just also OK, great. You
01:10:25know, as a first, just to say
01:10:27I think this is exactly right.
01:10:29When we asked us why,
01:10:30we know that just doing the genome
01:10:32sequencing is not going to be sufficient.
01:10:34We can start with the gene
01:10:36expression analysis because there
01:10:37are these epigenetic changes,
01:10:38but it's very clear that
01:10:39that's also in not sufficient.
01:10:41And one thing I just wanted to point
01:10:43out is that as a cell biologist,
01:10:45I think we we know understand
01:10:46really well that for example,
01:10:48translational capacity is something
01:10:49that's highly affected by stress,
01:10:51and so when we think about what's going
01:10:52on in a particular tumor environment,
01:10:54how that might be affecting the.
01:10:56Relative efficiencies of translation,
01:10:57which will never be fully
01:10:59reflected in an RNA seek data set,
01:11:00is going to be really important.
01:11:02And you know,
01:11:03one of the things, for example,
01:11:04that we think about a lot because
01:11:06appeared lasers work is thinking about
01:11:07hypoxia as a good example of this,
01:11:09and then the protein turnover aspects,
01:11:11right?
01:11:11And so these are all the factors that are
01:11:13that are contributing to what we might
01:11:15see different in a in a podium data set.
01:11:17And I was going to exactly
01:11:19the same point about,
01:11:20you know, aneuploidy,
01:11:21because you know,
01:11:22we know,
01:11:22and this is particularly relevant
01:11:23also for DNA repair factors that one
01:11:25of the ideas of why aneuploidy causes
01:11:27such changes in the proteomes that.
01:11:29We have these large protein
01:11:30complexes which are very codependent
01:11:32and they become kind of out of
01:11:34titration with regard to the
01:11:35components and that can you know,
01:11:37most of us are working on
01:11:39complex molecular machines where
01:11:40that's going to have an impact,
01:11:41and so that's going to
01:11:43really require detailed,
01:11:44the kind of mechanistic analysis.
01:11:45But we might get pointed to the
01:11:47fact that we need to do that work
01:11:49only if we go actually looking for
01:11:51proteome wide data instead of just
01:11:53the genomics. Thanks great Katie.
01:11:55Yes, so I just wanted to add to the
01:11:58conversation that because of the
01:12:01challenges of studying the pathways
01:12:03in in patient specimens directly,
01:12:06all of the things that were
01:12:08brought up the patient arrived.
01:12:10Models actually represent a really
01:12:12useful system to look at signaling and
01:12:15how signaling changes with treatment.
01:12:18And it's one of the reasons for
01:12:21which we did engage in this effort.
01:12:24In developing these models and also.
01:12:27Allows us to really explore how
01:12:29heterogeneous these samples are
01:12:31across different patient tumors so
01:12:32we can take tumors with a specific
01:12:35alterations or just across the
01:12:37wear resistant to specific therapy,
01:12:38and we can look at specific things
01:12:41in terms of at the protein level
01:12:44in those and we can look if we
01:12:46apply other therapies,
01:12:48what changes and so that really,
01:12:50I think,
01:12:50is a very valuable system in
01:12:52which to study what's happening at
01:12:55the protein level and signaling.
01:12:58Susan, could I just ask you to
01:13:01comment on how the the the different
01:13:04omics are approached from within
01:13:06an organization like yours and and
01:13:09to what extent leveraging systems
01:13:12biology approaches is is practical
01:13:15within your organization? Yeah, sure,
01:13:17and so I mean, I agree with the
01:13:20comments that have been made.
01:13:22First of all that we do need to
01:13:25look at these different mechanisms
01:13:27and it is possible to do that.
01:13:30Increasingly, you know where we are
01:13:32looking with things like Multiplex
01:13:34immediate fluorescence at the
01:13:36spatial organization of the tumors,
01:13:38and in doing that in in patient samples now.
01:13:41So I think the technologies are advancing to
01:13:44enable you to do that single cell sequencing.
01:13:47Is also helping.
01:13:48I think you know what I would say is that you
01:13:52can't do that intensively on on many child,
01:13:54so you have to choose the trial
01:13:56setting and the context for that.
01:13:58And it does have to be complemented
01:14:01by the kinds of things that
01:14:03Katie was talked about as well,
01:14:05so I think you know you can.
01:14:07You can see some sense of the overall
01:14:09picture emerging from some of the
01:14:11clinical trial data you really
01:14:12need to understand the mechanism,
01:14:14and for that you need a
01:14:16different setting and.
01:14:17Environment to do that in the
01:14:20different techniques and then you
01:14:21know the the PDX models have the
01:14:23have some challenges the Gen models
01:14:26have their own set of challenges.
01:14:28The humanized models for IO have
01:14:30their own set of challenges and I
01:14:32think what we can try and do is
01:14:35by looking collectively at at at,
01:14:37you know the clinical sample data and
01:14:39these range of preclinical models and
01:14:41backwards and forwards across that divide,
01:14:44that that's how you build up the
01:14:46bigger picture of of understanding.
01:14:48But you know,
01:14:49I think it's like it is like trying to
01:14:52sort of workout the overall picture from
01:14:54having several pieces of the jigsaw together,
01:14:57which is great,
01:14:58but you know nothing,
01:14:59that holistic view is absolutely
01:15:01critical to understanding,
01:15:02so I you know my comment would be
01:15:05that I think that the technology
01:15:07advances are now in place to enable
01:15:10us to see so much more than we
01:15:13were able to see 510 years ago.
01:15:15We need to bring that together,
01:15:17but have an integrated plan
01:15:19that goes across preclinical.
01:15:20Translational and clinical trial environment.
01:15:22Then I think that's critical,
01:15:25so we spend a lot of time in what
01:15:27is called early stage oncology
01:15:30without translation of medicine
01:15:32group actually working with drugs
01:15:34and programs that are in late phase
01:15:37development already on the market
01:15:39like awesome Internet and you know.
01:15:42And if I'm out.
01:15:43The other drugs that we have there
01:15:45because the two reasons one is
01:15:47that I think that's absolutely
01:15:49critical to understanding how to
01:15:51develop those in and continue that.
01:15:53But Secondly that understanding
01:15:55their feedback into the discovery
01:15:56organization for new opportunities.
01:15:58And I think the final piece I would say
01:16:01is that we can't do it all internally.
01:16:04Collaborations with organizations
01:16:05like Yale is absolutely critical.
01:16:07You've already heard a number of examples
01:16:09of the kinds of collaborations that
01:16:11that that we have that really helped to.
01:16:14And to feed and stimulate
01:16:15the work that we're doing
01:16:16internally. So you know really
01:16:18appreciate the work that that Megan,
01:16:20a team of doing that Katie Ability
01:16:22and her team are doing because
01:16:24that that sees what we're doing
01:16:26internally and we can't do all of it.
01:16:29Great good, there's a question here
01:16:31asking can we share examples of
01:16:33partnership with other academic centers
01:16:35that we may have for precision medicine
01:16:38efforts and adaptive combination
01:16:40treatment to overcome resistance and?
01:16:43And I'll talk about that maybe a
01:16:46little bit from the medicine side.
01:16:48Pet Larusso here leads experimental
01:16:51therapeutics clinical trials
01:16:52network U M1 grant that has.
01:16:55Many consortium members and she and
01:16:58her colleagues are leaders in taking
01:17:01molecularly driven questions and
01:17:04actually molecular selection strategies
01:17:07forward in the Umm in the ETCTN network.
01:17:11Jeff Sklar, here, runs one of the.
01:17:15Lamps that did the precision medicine
01:17:19sequencing for the match trial.
01:17:22We have investigators here
01:17:25leading match sub trials.
01:17:28We have, I think within our spores,
01:17:31collaborations across other cancer
01:17:33centers that are molecularly
01:17:35driven clinical trial questions.
01:17:37So I think from the disease based and
01:17:41clinical arena Ann from the Phase
01:17:44one arena there is quite a rich.
01:17:47A network of of these types of interactions,
01:17:51I don't know if anybody wants to
01:17:53address more from the preclinical.
01:17:56Level.
01:17:59I could just just comment that the
01:18:02anaplastic lymphoma kinase work
01:18:04that I mentioned at work.
01:18:05That's all guns are long term
01:18:07collaboration with people at Children's
01:18:09Hospital in Philadelphia and a new
01:18:11pen where much of the computational
01:18:13modeling is done actually through that.
01:18:16And then another approach that
01:18:17another aspect that we're working on,
01:18:19which is a collaboration of many.
01:18:22Out of many academic medical centers
01:18:25actually in the US and abroad,
01:18:27we actually have through the Alex
01:18:30is now in H Town Foundation and
01:18:33approach to targeting Myc signaling.
01:18:35Let me say it's not targeting
01:18:37Nick Per saver comes relates to
01:18:40combination therapies and so forth,
01:18:42the idea being that multigroup approach.
01:18:45With the idea that that make aberrations,
01:18:48particularly making neuroblastoma
01:18:49affect the network,
01:18:50and in principle one could rescue the
01:18:52network with appropriate combinations,
01:18:54and I think with the technologies,
01:18:56as Susan pointed out,
01:18:58advancing,
01:18:58I think the time is is is is is
01:19:01right to get to ask that question
01:19:04in that type of way.
01:19:09Anybody else wanna OK?
01:19:11There's a question here in the
01:19:13context of overcoming resistance.
01:19:15Can panelists share with their
01:19:17most excited about in terms
01:19:19of combination modalities?
01:19:21And maybe I'll just ask him to go first?
01:19:24'cause I love the KTM Ivy story.
01:19:28But I think everybody probably has
01:19:31their own favorite combination too,
01:19:32and you just
01:19:34want to, yeah, so of course I
01:19:36think because I work samples next,
01:19:38my my opinion might be any
01:19:41bias. Modulation is
01:19:42critical for that
01:19:43and not only the case.
01:19:45So one of the example I have showed
01:19:48this Acadian 5B where we can show pretty
01:19:51synergistic effect that you're going to
01:19:54check my blanket in multiple models.
01:19:56I only showed one in breast cancer.
01:19:59Will also see that as well.
01:20:02In addition, an and there's other
01:20:04modalities that you can actually modulate
01:20:06the tumor micro environment and,
01:20:08for example, someone that can recognize
01:20:10this as we are working on one,
01:20:13then it's called the CCR two,
01:20:15we can by inhibiting that we can change
01:20:17the macrophage population and by that,
01:20:19by doing that we can change
01:20:21the T cell activity.
01:20:23So, but basically it's just moderating
01:20:25the whole tumor micro environment
01:20:27and make it sensitive for email,
01:20:29checkpoint blockade and and this.
01:20:30This works well an intimate asks.
01:20:32Setting so I'm quite excited about
01:20:35that and that many of you probably
01:20:37know in other other institutions
01:20:39have studied with DMT inhibitors,
01:20:42HVAC inhibitors,
01:20:42and is it still inhibitors
01:20:44and those actually,
01:20:46I showed him assuming have showed strong
01:20:48efficacy in many different models,
01:20:50so I'm quite excited about this.
01:20:53This kind of combination.
01:20:56Yeah, I I think in Karen
01:20:58Anderson and Eli are broken.
01:21:01Working on demethylating therapy
01:21:02to uncover immune silencing
01:21:04and virally associated cancers.
01:21:06And I think there are a lot of examples.
01:21:10Coming on that so so
01:21:12I I just wanted to come and I think
01:21:15that's a very hard question, you know,
01:21:17because I think ultimately there the
01:21:19best combination is not going to
01:21:20be one combination that works every
01:21:22time I think it is so clear now that
01:21:25the tumors evade immunity through
01:21:27different dominant pathways and and
01:21:28more advanced tumors tend to have
01:21:30multiple pathways that I think the
01:21:32question has to do with where in an
01:21:34immunization pathways dominant and
01:21:35where more than one is dominant.
01:21:37And I think that should drive the
01:21:39combination not think the opposite.
01:21:41Wait and think that one combination will
01:21:43fix tumors with different problems.
01:21:47Yeah, so I think it's very interesting
01:21:49that you heard that question as an
01:21:52immune resistance question. It was.
01:21:54It was a very broad resistance question,
01:21:57but that's an interesting perspective.
01:21:58I think one of the things that
01:22:01that I think about sometimes is how
01:22:03some of the same mechanisms that.
01:22:06Have generated resistance to conventional
01:22:08therapies are now also generating
01:22:10resistance to immunotherapy and
01:22:12how you know our relentless focus
01:22:14on target instead of environment,
01:22:16which you know?
01:22:17I think it's something I've heard.
01:22:19You speak about alot.
01:22:21Kurt, you know we may have the
01:22:23same Achilles heel over and over
01:22:26again and in head neck cancer.
01:22:28A clear example of this is hypoxia,
01:22:31which leads to resistance
01:22:32to DNA damaging agents.
01:22:34It leads to resistance to.
01:22:37Radiation therapy.
01:22:38Prime example,
01:22:39but is now increasingly linked to resistance,
01:22:42demeanor,
01:22:43therapy as well.
01:22:45Yeah,
01:22:46and I say even taking it down to the simplest
01:22:50level of talking therapeutics it the
01:22:53answer is it depends because for example,
01:22:56just thinking about RAF inhibitor resistance.
01:22:59Actually David Stern and Marcus Bosenberg
01:23:02and others did a nice study that they
01:23:06published it a few years ago of combination
01:23:09combinations of drugs in a variety of
01:23:12cell lines for Melanoma and elsewhere.
01:23:15And showed that the combination which
01:23:18combinations work in which cells is
01:23:20very valuable and actually one of
01:23:22the things I'm quite excited about.
01:23:24The moment we're working as a group with
01:23:27systems biology island and equipment
01:23:29Los Alamos in the group and trying to
01:23:32understand that in terms of of the
01:23:35signaling networks around RAF and MEK
01:23:37and and and rest in different cells,
01:23:39some cells from different cancers,
01:23:41and innocence.
01:23:42Which combinations work depends
01:23:44say on the level of KSR 1.
01:23:46And that's a key determinant,
01:23:48and so it just depends on so much
01:23:50on how the network is wired,
01:23:52which of course goes back to
01:23:54the question about proteomics,
01:23:56because ultimately that you you're
01:23:58trying to control the system and the way
01:24:00the system is set up by a chemically,
01:24:02it defines on how it will define how
01:24:05we respond to different combinations.
01:24:07And so I think we're going to want
01:24:09to get into things at that kind of
01:24:11level to understand where we should
01:24:14use which combination,
01:24:15and I think it occurs .2.
01:24:17That's that's exactly.
01:24:18Got it that the case is going to be
01:24:20the case in the more complex systems
01:24:22of intercellular communication too.
01:24:25And can I just add?
01:24:26There's also the kinetic component, right?
01:24:28So I think when things were still
01:24:30not clear on is are you better
01:24:32off with this combination early
01:24:33on or is this going to be better
01:24:35once you get initial resistance?
01:24:36And actually that's may seem trivial,
01:24:38but I really don't think it is and
01:24:40requires modeling to think about
01:24:41just how the kinetics is playing out.
01:24:44Great, related, related to that,
01:24:46one of the things that I was going to say
01:24:50is one of the things I'm excited about.
01:24:53Sort of going forward and looking at
01:24:56the field over the next few years is
01:24:59is really what we can learn about the
01:25:01tumor from the get go that can tell us
01:25:05how we would want to treat it to stave
01:25:07off certain mechanisms of resistance.
01:25:10I think we're starting to see some
01:25:12examples also of clinical trials
01:25:14that are starting to subset.
01:25:16Patients with certain tumor Gina
01:25:18types or whether tumors have certain
01:25:20features and sort of put them and and
01:25:23into trials with specific combinations,
01:25:25and I think that's going to be really
01:25:28interesting approach in the next few years.
01:25:32Terrific, thanks, just
01:25:33a couple comments on this,
01:25:35so there's a few things that
01:25:38you that you've heard there
01:25:40that I just like to to build on.
01:25:42A completely agree that we need
01:25:44to understand what's the right
01:25:46combination for the set of what
01:25:49the adaptive mechanisms have
01:25:50been in that particular tumor,
01:25:52and I think to that end,
01:25:54looking at the adaptation at an
01:25:57earlier point than we typically
01:25:59do is a key part of this strategy,
01:26:01so there's some really interesting.
01:26:03Concepts here that you will be
01:26:06working with Gordon Mills from
01:26:07OHSU on and looking at the adaptive
01:26:10rewiring that goes on which really
01:26:12happens quite quickly.
01:26:13And of course one element of
01:26:15that that we need to understand
01:26:17is undoubtedly the epigenetic
01:26:19mechanisms that that that come in,
01:26:22'cause they're quite commonly involved
01:26:23in some of the resistance mechanisms,
01:26:26and as I pointed out,
01:26:28we don't really have good techniques for
01:26:30looking for those if we haven't got a biopsy.
01:26:34Early on,
01:26:34so I think that's that's absolutely key,
01:26:37and then I think that also informs,
01:26:39but the potential for how we do
01:26:41combinations 'cause one of the limiting
01:26:43factors of actually getting these
01:26:45to work has been the Taler ability
01:26:47of the combinations clinically.
01:26:49And for that I think there's a couple
01:26:52of chinks of light of what we can do.
01:26:55First of all,
01:26:56we're starting to develop some better
01:26:57tolerated therapies inherently,
01:26:59so I think that gives us a bit
01:27:01more headroom for some of the
01:27:03combinations that we need to do.
01:27:06Things like Adcs,
01:27:06better ways of you know,
01:27:08delivering some of the mechanisms of
01:27:09killing give you a bit more headroom
01:27:11and understanding what drives total ability,
01:27:13and then looking at the sequencing
01:27:15rather than trying to do it does
01:27:17everything at the same time is
01:27:19another innovation that I think will
01:27:20come in that will help us with that.
01:27:22And but again I think we're going to
01:27:25need to apply all of these tools that
01:27:27we've got and the modeling of that too,
01:27:29to reduce down the number of options
01:27:31that we actually bring into the
01:27:33clinic and increase the likelihood
01:27:34of each one of those being.
01:27:36Being successful with in combination
01:27:38therapy is going to keep us
01:27:40occupied for a little while yet
01:27:41before we solve that problem.
01:27:44Terrific great well this I,
01:27:46I hope the attendees have enjoyed
01:27:48this exchange of opinion and
01:27:50knowledge as much as I have.
01:27:52I'd like to turn it back now to
01:27:55Charlie Fuchs and just ask him to
01:27:58share a couple of concluding remarks.
01:28:00Well, Barbara,
01:28:01thank you and all the panelists.
01:28:03It was a fantastic discussion and
01:28:06really provided so much insight
01:28:08in terms of how we continue to
01:28:10move grade science into the clinic
01:28:12and frankly how we learn more.
01:28:14About the tests in clinical
01:28:16trials that were actively doing
01:28:18as part of our investigation,
01:28:20this is I mentioned is the third
01:28:22of our Yale engage cancer forms
01:28:24and I hope that our attendees
01:28:27enjoyed it and benefited from it.
01:28:29And as I mentioned,
01:28:30the work continues and we very much
01:28:33want this to be the beginning of
01:28:35the conversation and so hopefully
01:28:37what will what you'll do and what
01:28:40we'll do is engage each other in
01:28:42thinking through how we partner,
01:28:44how we work strategically together.
01:28:46To think of these,
01:28:47the ideas that are panels are brought
01:28:49up an develop new initiatives,
01:28:51so people should feel free to
01:28:53reach out to me or any of the
01:28:56panelists to think about.
01:28:57These collaborations will be contacting you
01:29:00really appreciate your taking the time.
01:29:01You know when I listen to
01:29:03discussions like this,
01:29:04I think it gives all of us hope
01:29:07and excitement about the years
01:29:09ahead of of cancer investigation.
01:29:11So let me just turn it back,
01:29:13turn it back to Barbara for
01:29:15some final thoughts.
01:29:18You know, so I've been at Yale about 6
01:29:221/2 years and the conversation today.
01:29:26Sort of reminds me of of the
01:29:28excitement that I felt when I started
01:29:30going to seminars around here.
01:29:32I mean, there's just unbelievable
01:29:34scale of work of this quality
01:29:36going on at this institution,
01:29:37and a lot of people thinking about
01:29:40how to make cancer treatment better.
01:29:42So thank you all for joining us today.
01:29:45Please stay in touch and I want
01:29:48to thank Susan and curtain,
01:29:50Katie and Mark and Megan Inch in for
01:29:53their their wonderful presentation.
01:29:55Thank you very much.
01:29:59Keyboard.