The Evolution of Antibody-Drug Conjugates in Oncology: Have we found our “Magic Bullet”?

Name: The Evolution of Antibody-Drug Conjugates in Oncology: Have we found our “Magic Bullet”?
Uploaded: 2025-01-28T19:58:57.5366667Z
Duration: 1 h 1 min 33 s
Description: Yale Cancer Center Grand Rounds | January 28, 2025 Presented by: Dr. Ian Krop

January 28, 2025

Yale Cancer Center Grand Rounds | January 28, 2025

Presented by: Dr. Ian Krop

Information

ID: 12683
To Cite: DCA Citation Guide

Download Transcript

00:01It's really pretty often that
00:03we have experts
00:04from our own faculty who
00:06go around the world and
00:07give talks, and we never
00:09ask them to give talks
00:10here.
00:11So we're trying to fix
00:13that problem.
00:15And so our speaker today
00:17is Ian Cropp.
00:19I think many of you
00:20know Ian.
00:22Ian came here,
00:24a month after I did,
00:26three years ago from Dana
00:28Farber.
00:30He
00:31spent most of his life,
00:34up until about age,
00:35I don't know, somewhere in
00:37his
00:38mid thirties at, Hopkins where
00:40he went to college and
00:41medical school and got his
00:43PhD and trained,
00:44and then came to Dana
00:45Farber where he was a
00:46fellow
00:47and stayed on the faculty
00:49for quite a number of
00:50years.
00:51He he is a breast
00:53cancer expert,
00:55of,
00:57known around the world.
01:00And for the purposes of
01:01today's talk,
01:02he's also,
01:04an expert in antibody drug
01:06conjugates.
01:08In fact,
01:09he ran,
01:11the first phase one trial
01:14of the first antibody drug
01:16conjugate that was approved in
01:18solid tumors, and that was,
01:20TDM one now back,
01:23a number of years ago.
01:25So,
01:26I'll also mention that Ayan
01:28is the chief,
01:30chief scientific officer for the
01:31translational breast cancer research consortium.
01:34Here at Yale,
01:35he's our,
01:37chief clinical research officer
01:40and runs the the clinical
01:42trials office.
01:43So, Ian,
01:45please.
01:51Good afternoon.
01:53Thank you for that kind
01:54introduction.
01:59So,
02:00we're gonna be talking about
02:02antibody drug conjugates, which, you
02:03know, I think is a
02:04this is a particularly opportune
02:06time to have that discussion.
02:07This is an area that's
02:08moving very quickly. There's actually,
02:11been two FDA approvals of
02:13ADCs just in the last,
02:15like, eight or nine days,
02:16and that's just within breast
02:18cancer.
02:19So there's a lot going
02:19on.
02:21And
02:24what I thought I would
02:25do would be to talk
02:26about ADCs using
02:28HER2 positive breast cancer as
02:29kind of an example, and
02:31and the reasons for that,
02:33selection is because,
02:35HER2 positive disease is,
02:38an area where we've had
02:39the biggest clinical impact of
02:41of targeted therapy in general
02:42and ADCs specifically.
02:45There's interestingly, there's a we
02:47see a benefit of ADCs
02:48across a very wide range
02:49of target expression in in
02:51this, subtype of breast cancer,
02:53which I I think is
02:54worth talking about.
02:55And,
02:57it's a there's a examples
02:59of basically all the key
03:00features of of ADCs are
03:02kind of manifest in HER2
03:03positive disease. And lastly,
03:06this is what I study,
03:07so I have most of
03:08the slides. So it was
03:09easier.
03:10So,
03:12just for those of you
03:12who don't treat breast cancer,
03:15HER2 positive disease makes up
03:17about fifteen to twenty percent
03:18of breast cancers, and these
03:19cancers have dramatic overexpression
03:22of of this cell surface
03:24protein
03:25tyrosine kinase called HER2.
03:27There's typically a million or
03:29even two million copies of
03:30HER2 protein on each cancer
03:32cell.
03:33These cancers are typically high
03:35grade, and they present with
03:36more advanced stage. They recur
03:38more frequently, and they have
03:40resistance,
03:41to,
03:42standard therapy. So these patients,
03:45had very poor prognosis
03:48prior to the realization
03:49that the reason these cancers
03:51were behaving so badly was
03:52because of this dramatic overexpression
03:54of HER2 and all the
03:55downstream signaling,
03:57that was happening because of
03:58that.
04:00So that led to the
04:01development of drugs targeting HER2,
04:03the first of which was
04:04an antibody
04:05called trastuzumab,
04:07and it was shown back
04:08in, I think, nineteen ninety
04:09eight or nineteen ninety nine,
04:11that the addition of paztuzumab
04:14to chemotherapy
04:15led to substantial improvements in
04:17in outcomes, progression free overall
04:19survival.
04:20And,
04:22if you use it in
04:22early stage disease, it increased
04:25the cure rate by about
04:26fifty percent. So it was
04:27a clear breakthrough,
04:30with the with this HER2
04:30targeted therapy introduction.
04:33And it really, in many
04:34ways, met this met the
04:36criteria of,
04:38the Nobel laureate Paul Ehrlich's,
04:41idea of a of a
04:42magic
04:43bullet, a a a drug
04:44that specifically
04:45targets,
04:47a pathogen by while sparing
04:49normal tissues. And, actually, doctor
04:51Ehrlich was thinking,
04:54about chemicals and chemical dyes,
04:56originally, but but we use
04:57it in terms of, thinking
04:59about antibody antibiotics and and
05:01antibodies,
05:02as well. So this feels
05:04like it it met those
05:05criteria.
05:07But the problem,
05:08particularly in HER2 positive disease,
05:11was that
05:13for trastuzumab really to work,
05:14you need to have chemotherapy
05:16around. You need to have
05:17some type of cytotoxic agent,
05:18and that's because one of
05:20the things trastuzumab does is
05:21helps encourage,
05:23apoptosis when you need something
05:25to,
05:25produce that apoptosis, and that's
05:27where chemotherapy comes in. But,
05:30you know, you have this
05:31very targeted
05:32antibody,
05:33the magic bullet, and then
05:34you throw in non targeted
05:36chemotherapy. So you lose some
05:37of your magicness,
05:39when you have to have
05:40chemotherapy around. So this was
05:41a perfect opportunity to take
05:43advantage of this then very
05:45new type of technology called
05:47an antibody drug conjugate,
05:49and,
05:50you know, this is a
05:51very simplified diagram,
05:52obviously,
05:54but the idea of an
05:55ADC is you have an
05:56antibody targeting some cell surface
05:58protein,
05:59and you have a very
06:00cytotoxic,
06:02moiety, typically a potent chemotherapy,
06:04and you attach them,
06:06with a linker that's typically
06:08cleaved once or the idea
06:10is that it's cleaved once
06:11the
06:13the conjugate gets inside of
06:14a cell.
06:15With the overall goal that
06:16you're gonna increase the efficacy,
06:19of of the of the
06:20cytotoxic drug while decreasing the
06:22toxicity
06:23by selectively delivering this to,
06:26the cancer cells. So these
06:27are the conjugates you talk
06:29about,
06:30guided missiles, smart bombs,
06:33whatever warmongering,
06:35terminology you wanna use, but
06:36that's the general idea.
06:39And so
06:40the first of these, as
06:41Eric had mentioned, to really,
06:44become clinically useful in solid
06:46cancers was this one called
06:47trastuzumab emtansine,
06:49and the idea of ADCs
06:51actually had been around for
06:52decades. The problem is none
06:53of them really worked, and
06:54the main problem they were
06:56having in the past,
06:57was that was toxicity because
07:00the linker tended to break,
07:02in,
07:03in in blood. So you
07:05had disassociation,
07:06showed you had lots of
07:07free cytotoxic,
07:09drug floating around, and that
07:10was causing nonspecific toxicity.
07:13And so,
07:15what was different about TDM
07:16one, was that it it
07:18started with this trastuzumab
07:20monoclonal.
07:24Oh, it's a touch screen.
07:25It starts with this trastuzumab
07:27monoclonal,
07:29targeting HER2 that we talked
07:30about. It had a,
07:33microtubule inhibitor payload, and the
07:36linker in this case was
07:37actually not not cleavable. So
07:38I talked about the idea
07:39of general user cleavable. This
07:41one was not.
07:43And
07:44the story is kind of
07:45interesting how that came about,
07:47and I promise this is
07:48the only chemistry I will
07:49be talking about today.
07:51So when Genentech was trying
07:52to develop this drug,
07:54they were really trying to
07:55make for a more stable,
07:57linker. Because, again, the problem
07:59in the past would be
08:00the linkers weren't stable, and
08:01so you were getting toxicity.
08:02So they said, okay. Well,
08:03let's try to make this
08:04super stable. And so they
08:05were testing a number of
08:07different,
08:08chemical structures, looking different,
08:11link different,
08:13cleavable
08:14linker,
08:15chemistries and and distances between
08:17the antibody and the and
08:18the payload.
08:19And they threw in at
08:20the bottom of their experiments
08:22this, MCC
08:23linker, which is basically non
08:25cleavable. And they put that
08:26there as a negative control
08:27because, of course, this wasn't
08:28gonna work,
08:29because if you don't have
08:30a cleavable linker, then you
08:32can't release the drug to
08:33kill the cancer cell. So
08:34they throw that they threw
08:35that in there just as
08:36a control.
08:38And what they found was,
08:39as expected, that the the
08:42non cleavable linker, the MCC
08:44in blue there at the
08:45top, was very stable in
08:47plasma, the most stable, and
08:48it was, stable different ways
08:50looking at it.
08:53And it actually was pretty
08:54well tolerated because,
08:56you know, the there you
08:57weren't having this, disassociation
09:00in blood, but the surprise
09:02was is also was the
09:03most effective.
09:04And so the blue line
09:05at the bottom there is
09:06the is the TDM one,
09:07essentially.
09:14Conjugate, the TDM one binds
09:15to the cell surface,
09:17on the HERT the HERT
09:18two. It gets internalized
09:20into lysosomes where the proteases
09:22actually were
09:24would dissolve the antibody essentially,
09:27and so you you didn't
09:29break the linker.
09:30It's just the the amino
09:31acid, the lysine from the
09:33antibody where it's connected would
09:34just leave,
09:35with the with the with
09:37the drug attached to it.
09:38And so that's how this
09:40drug was working, and that's
09:41how you got this selective
09:42delivery of the DM one,
09:45was because the antibody was
09:46getting just,
09:48proteolyzed.
09:49So, as Eric mentioned, we
09:51were involved in this phase
09:53one trial,
09:54the phase first in human
09:55study of TDM one. We
09:57enrolled patients who had had
09:59already progressed on multiple, HER2
10:01directed therapies,
10:03and despite that and despite
10:05this being just the phase
10:06one, there was a lot
10:07of efficacy seen. So the
10:09forty four percent response rate,
10:12these were durable responses, the
10:13progression free survival is about
10:14ten months, and seventy three
10:16percent of patients had some
10:17benefit.
10:19And at the same time,
10:20not only was it efficacious,
10:23but it was very well
10:24tolerated. So we didn't see
10:26the usual chemotherapy side effects,
10:28people generally didn't get nauseous
10:29or,
10:30have neutropenia or neuropathy or
10:32hair loss,
10:34And so it
10:35it probably and and the
10:37reason why
10:38was because the amount of
10:39the free payload, which is
10:41in black here, was very
10:42low. It was negligible. So
10:44it really wasn't releasing that
10:45payload,
10:46very quickly as
10:48the hope, as as was
10:49hoped.
10:51We then moved to a
10:52phase two trial again in
10:53very heavily pretreated patients, and
10:55I'm just showing this because
10:57it's probably the first trial
10:58that that Eric,
11:00and Pat Larusso and I
11:01did together,
11:03back in the day.
11:05And we saw the same
11:06thing heavily pretreated patients, response
11:08rate was
11:09was was reasonable, and and
11:11and, it was effective.
11:13And then we moved on
11:14to do a a registrational
11:15phase three,
11:17two phase three studies. This
11:18one, in patients who had
11:20already progressed on all standard
11:22therapies,
11:23showing that TDM one was
11:25better than the standard of
11:26care in terms of survival
11:28and PFS and was less
11:30toxic.
11:31And similarly,
11:32this is a earlier line
11:34trial again,
11:36showing better,
11:37efficacy and less toxicity compared
11:40to the standard of care.
11:41So these trials led to
11:42the approval of of PDM
11:44one as you heard, I
11:45think, in two thousand twelve,
11:47as the first ADC,
11:48in solid cancers,
11:52and really,
11:54kind of it did a
11:55couple of things. It validated
11:56HER2 as a as a
11:57target for an antibody drug
11:59conjugate,
12:00but most importantly,
12:02as being the first success
12:03in a common cancer,
12:05it led to the widespread
12:07development of ADCs across,
12:09tumor types.
12:10So, at last check, there
12:12were three hundred and seventy
12:13ADCs in clinical development,
12:16again, spurred on by, by
12:18the success of of this
12:19drug.
12:21And it also inspired people
12:22to start looking at better
12:24technologies for for linkers and
12:25payloads,
12:26and that's probably been best
12:28exemplified by this next generation,
12:31drug targeting HER2, which is
12:33called trastuzumabdoroxican,
12:35which differs in a few
12:36ways. It's no longer it
12:38has a different class of
12:39payload. It's not a microtubule
12:40inhibitor. It's a topoisomerase
12:42inhibitor.
12:43But perhaps the most unique
12:44feature,
12:45was it it evidence this
12:47thing called bystander effect.
12:49And it had and the
12:50reason it did this is
12:51because it had a different
12:52linker that that was cleavable,
12:54by, proteases,
12:57inside the cell and that
12:58led to this bystander effect,
13:00which,
13:01I'll try to depict,
13:03here. So if you the
13:04conjugate now binds, again, binds
13:07to the surface of cell,
13:08gets internalized, the payload's released,
13:10and if you'll excuse my,
13:12grade school animation,
13:15that once that payload is
13:17in the cytoplasm,
13:18unlike TDM one, in this
13:20case there's no lysine attached
13:22to it, it's thus membrane
13:24permeable, and so it can
13:25diffuse outside the cell and
13:27kill neighboring cells, even if
13:29those neighboring cells don't have
13:30HER2 on their surface.
13:33Why is this important? Again,
13:34we're targeting a HER2 positive
13:36cancer. Why is it matter
13:37that this drug could work
13:39in cancers cells that don't
13:40have HER2 on them?
13:42Well,
13:43and I'm sorry. This and
13:44this just shows that this
13:45bystander effect really seems to
13:47work.
13:48So this is a xenograft,
13:50on the left of a
13:51combination of a HER2 positive
13:53cell line and HER2 negative
13:54cell line, HER2 positive stains
13:56brown,
13:57and by IHC.
13:59And if you treat with
14:00t d m one as
14:01shown in the middle,
14:02diagram, you kill off the
14:03HER2 positive cells, but you
14:05leave behind the HER2 negative
14:06cells. Again, because HER2 TDM
14:08one doesn't have this bystander
14:09effect. But on the right,
14:10you use TDXD
14:12with bystander effect, you kill
14:13off both cell lines. Okay.
14:14Again, why does that matter?
14:16This is a xenograph where
14:17you mix two cancers together.
14:19Obviously, people have one cancer,
14:22and it's HER2 positive, so
14:24there should be lots of
14:25HER2 on all the cells.
14:26Well, let's just for the
14:28sake of argument say that's
14:29not the case and that
14:30there are heterogeneous,
14:32expression of HER2 in some
14:34cancers. The concern would be
14:36that if you have this
14:37heterogeneous cancer shown
14:40here where the blue cells
14:41are the HER2 positive and
14:43the red ones are not,
14:44you treat with a very
14:45targeted drug,
14:48you kill off the HER2
14:49positive cells, you leave behind
14:50the HER2 negative cells, and
14:52the HER2 negative cells then
14:53grow up, and now you
14:54have a resistant cancer
14:55similar to what was shown
14:57in in those in the
14:58xenograft I just showed.
15:00Is this clinically relevant? So,
15:03my,
15:05colleague,
15:06Otto Mesker and I did
15:08this IIT,
15:10to ask this seemingly pretty
15:12straightforward question, but at least
15:13as far as we knew
15:14that really hadn't been addressed
15:15before in the clinic.
15:17Are heterogeneous cancers
15:19less sensitive to targeted therapy?
15:22So the way we did
15:23this, we took a hundred
15:24and sixty four patients who
15:25had newly diagnosed HER2 positive
15:27disease,
15:28and before we started treatment,
15:30we did a biopsy in
15:32two different locations of their
15:33cancer.
15:34And then they treated with
15:36TBM one, we threw in
15:37another HER2 antibody,
15:38called pertuzumab,
15:40and and then they had
15:41surgery. And so we we
15:43took advantage of the fact
15:44that we did these two
15:45different biopsies, and so we
15:46had, and each of those
15:47biopsied had three different sections
15:49that we looked at. And
15:50we looked at whether there
15:52was heterogeneity,
15:53in terms of HER2 amplification
15:55in those different sections.
15:57And we scored those cancers,
16:01using a standard ASCO,
16:04cap definition of heterogeneity.
16:07And then we looked at
16:07how the heterogeneity
16:09played out in terms of,
16:11benefit to this treatment.
16:12And what we found was
16:13that in the cancers that
16:15were not heterogeneous, those cancers
16:17where there's pretty homogeneous expression
16:19or amplification of HER2,
16:21fifty five percent of those
16:22patients had complete eradication of
16:24their tumor just from the
16:26HER2 targeted therapy. So in
16:28in that situation,
16:30this HER2 target therapy is
16:31highly effective,
16:32and and well tolerated.
16:34But then when you looked
16:35at the cancers that were
16:36heterogeneous,
16:37none of them had a
16:38pathologic complete response. And so
16:40this was highly statistically significant,
16:43and it confirmed that actually
16:44heterogeneity matters, that if you
16:46have a heterogeneous
16:47cancer, you're not as likely
16:49to respond as completely to
16:51HER2 targeted therapy, or at
16:53least a HER2 targeted therapy
16:54that doesn't have this bystander
16:56effect, which TDM one does
16:58not.
16:59We actually
17:00went on to look a
17:01little bit more in-depth in
17:03collaboration with,
17:05a lab at Dana Farber
17:07led by Neli Polyak and
17:08and Francisca,
17:10Mihor,
17:11looking at at the single
17:13cell level by HER2 amplification,
17:15trying to understand what was
17:16actually driving,
17:18the resistance.
17:19And interestingly,
17:21at least interesting to me,
17:23it was actually the fraction
17:24of the non amplified cells
17:26within,
17:27the population
17:28that was strongly correlated with
17:30PACER. And it so, originally,
17:32we kind of had the
17:33idea, okay, a heterogeneous cancer
17:35is one where, you know,
17:36there's a big c of
17:37HER2 positive cells and then
17:39there's this little cluster of
17:40HER2 negative cells that was
17:42gonna stay behind and grow
17:43out. But, actually, that's not
17:44what we saw. It's actually
17:47fairly
17:47oftentimes relatively uniform distribution of
17:50HER2 positive and HER2 or
17:52HER2 amplified and HER2 non
17:53amplified cells,
17:54that was that was leading
17:56to heterogeneity. So you didn't
17:57actually have to do those
17:58six,
17:59biopsy sections. If you just
18:01look at one section and
18:02you quantitate,
18:04how,
18:05much how many of these
18:06HER2 non amplified cells are
18:08there, you can actually predict
18:10PCR,
18:11very strongly,
18:13which which is
18:15which is kind of interesting.
18:16And,
18:17we actually that our data
18:19was subsequently replicated a few
18:21years later, in a in
18:22a larger trial where they
18:23went back and looked at
18:24the same population patients treated
18:26with TDM one and pertuzumab.
18:28And, again, those who had
18:30heterogeneous
18:31positivity for HER2 had no
18:33pass ERs, whereas those that
18:34were homogeneous had a a
18:36over fifty percent pass ER
18:37rate.
18:38Okay. So
18:41getting back to TDXD,
18:43it's got this
18:45bystander effect, at least in
18:47vivo.
18:48Does that matter in terms
18:49of improving efficacy?
18:53So it does.
18:54Or at least somehow it
18:56has much better efficacy, whether
18:57how much of that's bystander
18:58effect or some of the
18:59other,
19:01aspects that we don't really
19:02know at this point. But
19:03this was the,
19:04phase two single arm study
19:05that that,
19:07we were involved with,
19:10that demonstrated
19:11of in heavily pretreated patients,
19:13patients who had already had
19:14all the standard HER2 therapies,
19:16I think, the median of
19:16six prior lines. So these
19:18were seventh line metastatic
19:20patients, and the response rate
19:21to TDXD alone was over
19:23sixty percent. It was very
19:24durable. The patients,
19:26had a, a were on
19:27study for over twenty months,
19:29and virtually a hundred percent
19:31of patients had some benefit
19:32as shown in this waterfall
19:33plot.
19:36The trade off was that
19:37there was more toxicity. So
19:38unlike TDM one,
19:40with this drug,
19:42most patients get some nausea,
19:43there's fatigue, there's,
19:47hair loss in some patients,
19:49and
19:50in ten to fifteen percent
19:51of patients,
19:53they get a serious complication
19:54called interstitial lung disease or
19:56pneumonitis,
19:58which is typically manageable, but
20:00it's definitely something that you
20:01have to pay attention to
20:02because it can be fatal.
20:04So,
20:06because of the incredible efficacy
20:08in very refractory patients,
20:10these data led to the
20:12the accelerated approval of of
20:13TDXD,
20:15in in this,
20:17kind of refractory setting.
20:20But then we went on,
20:21there were several other trials.
20:22This was a phase three
20:24style
20:24phase three trial looking specifically
20:26at patients who had already
20:27had the other ADC, TDM
20:29one, and comparing TDXD versus
20:32standard,
20:33HER2 therapy,
20:34and TDXD was much better.
20:38And this demonstrated,
20:41that you actually can
20:43have benefit from one antibody
20:45drug conjugate followed by another,
20:46even though they have the
20:47same target, but they have
20:48different payloads. So by switching
20:49payloads, you're able to,
20:51provide,
20:52more efficacy.
20:54And at least in my
20:55mind, this supports
20:56the paradigm,
20:58that you could treat patients
20:59with sequential ADCs with different
21:01payloads, and we'll we'll talk
21:03more about that,
21:04later.
21:08This went on to now
21:09look head to head at
21:11TDXD
21:11versus PDM one, so two
21:13ADCs against each other. Probably
21:15the only trial that's done
21:16that.
21:17And again, TDXD was was
21:19far superior to to TDM
21:21one,
21:22fourfold greater
21:24progression free survival, so almost
21:26twenty nine months progression free
21:27survival,
21:28a level that I don't
21:29think had ever been seen
21:31before in pretreated
21:32patients with breast cancer.
21:34So very effective
21:36survival was also beneficial,
21:38and this established tDxD as
21:40the standard
21:41care for patients,
21:43with her trophosid metastatic disease.
21:47But
21:49there was a question at
21:50that time of what about
21:51patients with progressive brain mets?
21:52And, unfortunately, that's a big
21:54problem in HER2 positive disease
21:56because,
21:57for reasons that aren't completely
21:59clear, there's a strong predilection
22:00for HER2 positive breast cancer
22:02to go to the brain.
22:04It may partly, it may
22:05be because,
22:06or part of it may
22:07be
22:09the the conventional wisdom that
22:11that drugs like that we
22:12use in HER2 positive disease,
22:14like antibodies, antibody drug conjugates,
22:16don't get into the brain
22:17into into the the brain
22:19because of the blood brain
22:20barrier. So it's kind of
22:21a a sanctuary site, and
22:22that's why we see so
22:23much of it. But there's
22:24also some biology involved that
22:26these cancers are just have
22:27a tropism to the brain.
22:29But as I said,
22:30you know, the the idea
22:32was antibodies don't get into
22:33the brain, therefore, antibody drug
22:35conjugates don't get into the
22:36brain, so how are we
22:37gonna how could a drug
22:38like tDxD
22:40work in this very common
22:41situation?
22:43But it turns out that
22:45actually antibodies can get into
22:46the brain at least somewhat.
22:48This is a pet label,
22:50trastuzumab
22:51study, and you can see
22:52on the bottom there,
22:54that actually the antibody does
22:57get to the to the
22:58brain metastases at least to
22:59some level.
23:01Probably because
23:02the blood brain barrier breaks
23:03down a little bit when
23:04you have a cancer there
23:06and it's disrupting, you know,
23:07causes dysregulation of of vascular
23:09genesis, so the blood brain
23:11barrier isn't quite as intact.
23:13But regardless,
23:15some antibody can get there.
23:17And so we actually went
23:18back and looked at,
23:20the the large studies I
23:22just had shown you, and
23:23there were a small number
23:24of patients on those studies
23:25that actually had progressive brain
23:27metastases
23:27at baseline.
23:29And we looked at the
23:30intracranial response of TDXD,
23:33and, actually, there there was
23:34some response. It was about
23:35a forty something percent response
23:36rate in the brain with
23:38this ADC, but the sample
23:39size was was pretty small.
23:42But just recently presented,
23:45at ESMO a few months
23:46ago was a prospective study
23:48of over two hundred and
23:49fifty patients with brain metastases,
23:52treating with tDxD.
23:54And as you can see,
23:56response rate in the brain
23:57with active brain metastases was
23:59over sixty percent. So I
24:00think we now have pretty,
24:02definitive data,
24:04that these ADCs actually are
24:05quite active in the brain.
24:07I think that's important to
24:08know since we obviously
24:10have a lot of cancer
24:11types that that have, brain
24:12metastases as a major problem.
24:15And so this idea that
24:16you have to use small
24:16molecules,
24:18is probably not true.
24:21So as you might expect,
24:24with the efficacy of of
24:26of these conjugates in patients
24:27with metastatic disease, there was
24:29interest in seeing whether these
24:30conjugates could also,
24:33work in preventing recurrences in
24:35patients with early stage disease.
24:37And so there originally, there
24:38was a large trial that
24:39looked at patients who had
24:40a high risk early stage
24:42disease because their cancers did
24:43not respond all that well
24:44to neoadjuvant
24:46therapy,
24:47neoadjuvant HER2 therapy, and randomized
24:49those patients to either the
24:50standard back then, which was
24:51just continuing trastuzumab
24:53or using TDM one,
24:56and the TDM one showed
24:58about almost a fifty percent
24:59reduction in recurrences,
25:02compared to trastuzumab. So that's
25:04now the standard of care.
25:07One issue was that the
25:08brain metastases,
25:10actually was not significantly reduced
25:12with t d m one
25:13compared to trastuzumab, and I
25:14know that goes against a
25:15little bit about what I
25:16just said about brain metastases.
25:18And maybe we can talk
25:18about why that might be,
25:20in our
25:21very interesting question answer period
25:23that's gonna follow this talk.
25:27But it has led us
25:28to do this study in
25:30in the alliance
25:31where we're taking
25:33those patients who had,
25:36residual disease after neoadjuvant therapy
25:37and randomizing them to TDM
25:39one or TDM one plus
25:41this, potent HER2 tyrosine kinase
25:43inhibitor called tucatinib. And this
25:44study is underway
25:46here, at Yale, so you
25:48can put patients on this,
25:49and I will designate that
25:50by the handsome Dan icon
25:51as you'll see for the
25:52rest of the talk here.
25:55We also were looking at
25:56whether you can,
25:57use the the very well
25:59tolerated nature of TDM one
26:01to deescalate
26:02therapy in patients with earlier,
26:05or lower risk HER2 positive
26:06disease.
26:08So my then colleague at
26:10Dana Farber, Sarah Talaney, and
26:11I, did this, investigator initiated
26:14trial
26:14looking at patients with stage
26:16one HER2 positive cancers,
26:17randomizing them to,
26:19the previous standard that was
26:20actually established by Eric of
26:22paclitaxel and trastuzumab or TdM
26:25one,
26:26and, the TdM one was
26:27associated with incredibly good outcomes.
26:29There was,
26:31less than
26:32one percent distant recurrence at
26:34five years.
26:35So clearly this was effective.
26:37We had assumed it was
26:38gonna be much better tolerated
26:40than the the the taxane,
26:42trastuzumab
26:43regimen,
26:44and it turned out it
26:45it had less some of
26:47the standard chemotherapy toxicities,
26:49but people were discontinuing the
26:51TDM one, which was given
26:52for a year in this
26:53study,
26:54earlier,
26:55than they were discontinuing the
26:56trastuzumab in the in the
26:57other arm of the study.
26:59So, that's led,
27:02Sarah to go on to
27:03do this second version of
27:05the study,
27:06comparing just six cycles of
27:07TDM one, because we think
27:09that might be all you
27:09need,
27:10versus,
27:11the same
27:13control arm. Again, this is
27:14a study that's ongoing at
27:16at Yale,
27:17and it's actually a very
27:18good study for these patients,
27:20in my opinion.
27:22There's also studies going looking
27:24at TDXD, this more potent
27:25ADC. This is a study,
27:27actually comparing TDM one to
27:29TDXD in this adjuvant setting,
27:30and then there's a neoadjuvant
27:31file as well.
27:32Okay.
27:33So,
27:35switching gears a little bit,
27:36and this, I think, is
27:37where it gets really interesting.
27:41All the data I've showed
27:42you before is for these
27:43HER2 amplified cancers. These are
27:45the cancers that have incredibly
27:47high levels of HER2.
27:49In breast cancer, there's actually
27:51a continuum of HER2 expression.
27:52So you got these super
27:53high
27:54amplified cancers with a million
27:56or two million copies of
27:57HER2, and then you've got
27:59everything in between,
28:00moderate, lowish levels of HER2,
28:02a hundred thousand, fifty thousand
28:04HER2 proteins. And we call
28:06those HER2 low by immunohistochemistry.
28:08They're called one plus or
28:09two plus, but they're not
28:10amplified.
28:11And then you have the
28:12very negative,
28:13cancers, which we're gonna call
28:15HER2 negative.
28:18And it turns out that
28:19these lowish levels of HER2
28:21are actually very common. In
28:22fact, the majority of breast
28:24cancer has some level of
28:26HER2 expression.
28:29So given that, and given
28:30that we have this monoclonal
28:31antibody called trastuzumab
28:33that we know works,
28:35with chemotherapy,
28:37The NSABP,
28:39did this very large trial
28:41where they took patients with
28:42HER2 low early breast cancer,
28:43and they randomized them to
28:45chemotherapy with or without trastuzumab,
28:47the same thing that had
28:48shown to be very effective
28:49in HER2 amplified cancers.
28:52Unfortunately, this was completely not
28:53effective, so adding trastuzumab for
28:55these HER2 low cancers did
28:57absolutely nothing,
28:58and you can see the
28:59IDFS has a ratio is
29:01point nine eight shows randomization
29:03was very effective,
29:04there,
29:05but no benefit.
29:07So then
29:11because trastuzumab works by, at
29:13least in part, by inhibiting
29:14HER2 signaling,
29:16it suggests that HER2 signaling
29:18really isn't important in these
29:19HER2 low cancers, so blocking
29:21it doesn't do anything.
29:23But it's still there. The
29:24HER2 is still there, and
29:25we have an antibody drug
29:26conjugate, which is basically looking
29:27for a target. And we
29:29use these antibody drug conjugates
29:31to
29:32to use the HER2 that's
29:33on the surface of these
29:34low cancers
29:35just as an address, a
29:36place, you know, a way
29:37to deliver our cytotoxic agent.
29:42So we now we had
29:44TDXD, and in the phase
29:45one trial of TDXD, we
29:47did have some cohorts
29:48of HER2 low cancers,
29:50and it actually looked like
29:51there was some activity in
29:52these HER2 low cancers.
29:54That prompted this very large
29:56trial
29:57of
29:57patients with metastatic HER2 low
29:59breast cancer. Again, the most
30:01common kind of breast cancer,
30:03there is sixty at least
30:04sixty percent of breast cancers,
30:06and randomized them to tDxD
30:07or chemotherapy because chemotherapy was
30:09a standard for these non
30:11HER2 amplified cancers,
30:13and tDxD was much better
30:14than chemotherapy in terms of
30:16survival,
30:17progression, response rate.
30:20And that led to the
30:21approval of tDxD in these
30:23HER2 low cancers, the first
30:24approval for of anything in
30:27HER2 low cancers,
30:28because it really wasn't a
30:29thing before the drug worked
30:31there.
30:32And it said, well, okay,
30:33if if it worked in
30:35these HER2 low cancers, and
30:36in fact, you know, I
30:37talked about that there's these
30:39one plus level and two
30:40plus levels.
30:42Two plus is more than
30:43one plus.
30:45If it in the trial,
30:46it actually the efficacy was
30:47pretty similar between the one
30:49plus and two plus. So
30:49that kind of begged the
30:51question, okay, well, can you
30:52go? How how low can
30:53you go?
30:54And in about twenty percent
30:55of cancers, there's, like, really
30:57marginal levels of HER2. So
30:59not even one plus, it's
31:01just like you can, you
31:01know, if you look real
31:02close, you can see a
31:03little smidgen of HER2 on
31:05the surface, but they're not
31:06completely
31:07stone cold zero.
31:09And so,
31:10we just so there are
31:12just another trial that was
31:13just presented over the summer.
31:16Same almost the same trial
31:17as I just showed you,
31:18but now,
31:20slightly different setting,
31:21but now this trial included
31:23a population of these ultra
31:25what we're now calling ultra
31:26low cancers. So just the
31:28smallest amount of HER2,
31:30not enough to be one
31:31plus.
31:33And what was seen, surprisingly
31:35somewhat,
31:37was that,
31:38actually, tDxD was much better
31:40than chemotherapy in these ultra
31:41low cancers. Seemed like the
31:42benefit was pretty similar to
31:43what we saw with the
31:44low cancers.
31:46Survival also seemed to be
31:47trending in the right direction,
31:48although immature.
31:49But I think what was
31:50particularly important or what was
31:52striking was that response rate
31:53in these ultra low cancers
31:55was sixty two percent. These
31:56were pretreated patients,
31:58and,
31:59you can see that it's
32:00actually the response rate in
32:02the ultra low is pretty
32:02similar to the HER2 low.
32:04And again, we can kinda
32:05talk about why that that
32:07might be,
32:08towards the end.
32:11It's funny, you know, when,
32:13in some of this original
32:14steering committee meetings of of
32:15TDXD,
32:16when the original ultra low
32:18date I'm sorry, when the
32:19original HER2 low data came
32:20out,
32:21people that I remember one
32:22specific person raising their hand
32:23and say, why don't we
32:24look at HER two zero
32:25cancers? And it was like,
32:26everybody kinda laughed because it's
32:27a HER two targeted drug.
32:29Of course, it's not gonna
32:29work in HER two zero
32:30cancers.
32:32But it seems to work
32:33in these ultra low cancers,
32:34and now there's a question
32:35of could it even work
32:36with pretty undetectable levels of
32:38HER2?
32:39And to test that, Adrianna
32:41Khan is doing,
32:43this IIT
32:44looking specifically at HER2 zero
32:46cancers,
32:47treating with TDXD,
32:48and then using some of
32:50David Rymm's,
32:51you know, very sophisticated,
32:53assays for HER2 to see
32:55if you can really identify
32:56whether there really is a
32:57threshold of HER2 expression below
33:00which you don't see activity.
33:01So that study hopefully will
33:03open, very soon.
33:06So what have we learned
33:07about HER2 ADCs?
33:09So clearly,
33:10they're superior to trastuzumab and
33:13chemotherapy,
33:14both in early stage and
33:15late stage disease.
33:17They've,
33:18TDXD is better, more efficacious
33:20at least than TDM one,
33:21but it also has more
33:22toxicity. And I think that
33:23like likely reflects the trade
33:25off, for these cleavable linkers
33:28versus non cleavable linkers.
33:30You get the bystander effect,
33:32but the bystander effect also
33:33can,
33:35hit normal tissue, not just,
33:38to other tumor cells.
33:41And, you know, this very
33:42interesting finding of tDxD being
33:43affected even in minimal levels
33:45of HER2, which may be
33:46because of this bystander effect,
33:48although we haven't proven that.
33:51So I just wanna take
33:52a slight,
33:57divergence here and and and
33:59bring up this question,
34:01just because I think it's
34:01really cool,
34:03of the fact that given
34:04that we have these very
34:06effective HER2 therapies, you know,
34:07progression free survival of thirty
34:09months and, you know, very
34:11long durations of response. And
34:12we actually have there's actually
34:14eight drugs now approved for
34:15HER2 positive disease, different mechanism
34:17of action.
34:19So given all of these
34:20highly effective drugs,
34:23can we really move the
34:24needle of treating patients with
34:26metastatic disease and go away
34:27from treating in a non
34:29curative setting, which is the
34:30way we do it now,
34:31and move it, to a
34:32curative setting?
34:34And by for those of
34:35you who don't treat patients,
34:37with metastatic disease,
34:39nowadays, we treat with one
34:40treatment. We wait for the
34:41cancer to become resistant, and
34:42then we switch to the
34:43other drug. And we're we
34:44try to string along our
34:46treatments,
34:47to keep patients with disease
34:49control as long as possible
34:50because we know we can't
34:51cure them. So there's no
34:52use giving a lot of,
34:53you know, kind of piling
34:55on your therapies. You wanna
34:56stretch them out so they
34:57last as long as possible.
34:59But by doing that, generally,
35:01you're you're gonna get resistance
35:03because you're only giving one
35:04drug, and, eventually, the cancer
35:05is gonna learn to become
35:06resistant. And that's why metastatic
35:07disease is typically felt to
35:09be not curable.
35:10So maybe that's not true,
35:11though, given the fact that
35:13we have these highly effective
35:14drugs.
35:15And so,
35:18a trial that's gonna be
35:20launched here at Yale shortly
35:21that's being run-in this consortium
35:23called the TBCRC,
35:25is trying to address, can
35:26we cure HER2 positive metastatic
35:28disease?
35:29So to do this, we're
35:31gonna deviate from the normal
35:32practice and take newly diagnosed
35:34patients and treat them with,
35:37twelve weeks of of ataxane
35:39and and trastuzumab, and then
35:41give them TDXD for eighteen
35:43weeks, and then give TDM
35:44one with the kinase inhibitor,
35:46and then give more kinase
35:47inhibitor,
35:48for about a year, and
35:49then just stop treatment and
35:51just follow patients with c
35:53tDNA and c and CAT
35:54scans with the idea
35:56that we're gonna try to
35:58improve the percentage of patients
35:59who don't have progression after
36:01four years essentially are cured.
36:03You know, this is way
36:04leukemias are treated, lymphomas are
36:06treated, you pile on mass
36:08you know, lots of different
36:09drugs,
36:10you know, in a very
36:11intensive way,
36:12even though kind of leukemia
36:13is kind of metastatic,
36:15to begin with, but it
36:16it works. Can we do
36:18that for a solid cancer?
36:19In the past, we really
36:20just didn't have the effective
36:22therapies to do that. Now
36:23that we do,
36:25can we change the paradigm?
36:26So this is a trial
36:27that should open soon here,
36:28and and, again, I think
36:29it's really worth exploring. It
36:31may be wrong, may not
36:32work, but it's worth trying.
36:35Okay.
36:36Enough of HER2.
36:37Other HER2 there other targets.
36:40So I guess I didn't
36:41realize I have a slide
36:42on HER2 here. So when
36:45when,
36:46you know,
36:47when we were, you know,
36:48those of us in the
36:49field were working on HER2
36:51ADCs, we said, hey. This
36:52is great. These drugs are
36:53working really well,
36:55but it's probably just because
36:57HER2 is just this amazing
36:58target for an ADC.
37:00Why is it amazing for
37:01ADCs? Well, first, you have
37:02tons of it on the
37:03surface. And so the more
37:05protein you have on the
37:06surface, the more ADCs combined,
37:08and therefore, the more the
37:10more ADC you can get
37:11inside the cell.
37:13The normal tissue tended to
37:14have very low amounts of
37:15HER2.
37:16The internalization of HER2 is
37:18very fast, and it doesn't
37:19down regulate. And you can
37:20see in this photomicrograph,
37:22if you coat the cell
37:23with a fluorescent HER, trastuzumab
37:25and then wait a few
37:26hours, all of it gets
37:28inside the cell. So all
37:29those HER2s are getting internalized,
37:31which if there's a ADC
37:32attached, it'll bring it with
37:34it. So internalization is important
37:35for ADCs, and the tumors
37:37are addicted to their HER2.
37:39As I said, it's the
37:40HER2 that's driving these cells,
37:41so they really need the
37:43signaling. So it's really hard
37:44for them to down regulate
37:45as a way to escape
37:46the effects of a d
37:47of the ADC.
37:48So we thought, hey. It
37:49all makes sense that these
37:50drugs are gonna work in
37:51HER2,
37:52but they're probably not gonna
37:53work other with other targets
37:55because the other targets don't
37:56have all these great characteristics.
37:58Fortunately, I was wrong, as,
38:00as often is the case.
38:02We now have the twelve,
38:03I think, roughly twelve ADCs,
38:09that are FDA approved, and
38:10you can see across a
38:11wide range of targets.
38:14And I forgot to mention,
38:16TDXD,
38:17just got approved yesterday
38:19for treating those ultra low
38:20patients, so that's kind of
38:22exciting. Another group of cancers
38:23to be treated.
38:25But you can see we
38:25have, activity of ADCs across
38:28liquid tumors and solid tumors,
38:31with a broad range of
38:33of of, targets.
38:34So I'll talk of just
38:36a brief, in the last
38:37few minutes, some of the
38:38other targets.
38:40Trope two is being tested
38:42and is is is been
38:43validated in breast cancer. It's
38:44being tested in other cancers.
38:48So COP two is a,
38:49cell surface protein that's pretty
38:51widely expressed in breast cancer
38:52associated with the worst prognosis.
38:55And there's a drug called
38:56sacituzumab
38:57gobletikin, which is an ADC,
38:59also with the topoisomerase
39:01pay payload, and it it's
39:02set up its linker a
39:04little differently. So in addition
39:05to being cleavable inside the
39:07cell, it's also cleavable
39:09by,
39:10low pH environments outside the
39:12cell, so it can it
39:13can release the payload both
39:14extracellularly
39:15and intracellularly.
39:18And it's been tested in
39:19triple negative breast cancer compared
39:21to chemo where it's much
39:23better, and it's approved in
39:24that setting.
39:26And its toxicity profile is
39:28quite is different,
39:29than PDXDs.
39:31It's all myelosuppression
39:33and a little GI toxicity.
39:35And interestingly,
39:37datapodimab
39:38daroxetine, which is another trope
39:40two ADC,
39:41same antibody as sacituzumab,
39:43same class of payload as
39:45sacituzumab,
39:46but with a daroxican
39:48linker, which is a little
39:49different.
39:50It's it's
39:51purely protease cleavable. It's not
39:53pH cleavable.
39:55And this is an active
39:57drug. In our in the
39:58phase one trial, we showed
39:59that it was active and,
40:01even in patients who had
40:02already progressed on sacituzumab, there
40:04was some activity.
40:06But interestingly, again, same payload
40:08essentially, same antibody. The toxicity
40:10is completely different. So there's
40:12virtually no myelosuppression
40:14with dapotumab,
40:16stomatitis is the toxicity showing
40:18how important those linkers are
40:20in driving the the characteristics
40:21of these ADCs.
40:23Dapotumab
40:24actually just got approved last
40:25week,
40:26for hormone receptor positive breast
40:28cancer based on another study.
40:30And then lastly,
40:31HER3,
40:32which is another,
40:34tyrosine kinase. Actually, it's not
40:36a tyrosine kinase. It's it's
40:37related to the other tyrosine
40:38kinase, the HER2 tyros HER
40:41family tyrosine kinases. It itself
40:43doesn't have an active kinase,
40:45but it's important in in
40:46signaling,
40:47and is overexpressed in a
40:49number of breast cancers.
40:50And there's a a conjugate
40:52called HER3 DXD or pertitumab
40:54daroxetin,
40:55which we showed also has
40:56activity across breast cancers. But
40:58not really sure how this
40:59one's gonna develop get developed
41:01because,
41:02the field is getting crowded,
41:04lots of,
41:05antibodies with the same payload.
41:07So where this one's gonna
41:08fit in is unclear.
41:12So there are other conjugates
41:14being developed.
41:16There are ones with fancy
41:18new protein structures. So there's
41:20biparotropic
41:21ADCs that bind two different
41:22epitopes of the same,
41:24molecule. There's bispecific ADCs binding
41:26two different molecules.
41:27There's probody conjugates that get
41:29activated,
41:31in the microenvironment.
41:32There's new payloads
41:34beyond very potent cytotoxic drugs.
41:36There are targeted,
41:39therapy kinds of payloads like
41:40kinase inhibitors and apoptosis promoting
41:43drugs. There's immunomodulatory
41:45payloads,
41:46radionuclides,
41:47and,
41:49there are new antigens not
41:50targeting the tumor anymore, but
41:51actually targeting the the microenvironment.
41:53All of these things are,
41:55currently,
41:56in development. So
41:57lots to more to come.
41:59I just wanted to close
42:00by bringing up a few
42:01what I think are important
42:02unanswered questions.
42:04One is and this is
42:05a little
42:06wonky, I appreciate.
42:08Should we be making more
42:10ADCs with non cleavable linkers?
42:12Right now, TDM one is
42:13the only approved ADC with
42:15a non cleavable linker. All
42:16the other ones have different
42:17types of cleavable linkers.
42:21And,
42:23you know, there's reasons for
42:24the cleavable linkers. You get
42:26the bystander effect,
42:27but you also get more
42:28toxicity. And I always wondered
42:30whether if you took a
42:31really potent cytotoxic drug and
42:33made it with a noncleavable
42:34linker, whether you could get
42:35efficacy and still keep the
42:36toxicity down. Because
42:38TDM one is still really
42:39the only drug with that
42:40really favorable,
42:42toxicity profile, which, again, was
42:43one of the original visions
42:44of an ADC.
42:47How should we sequence ADCs
42:49with different targets? So I
42:50I mentioned,
42:51we had a trial where
42:52we use the same target,
42:54but two different payloads, one
42:55ADC after another. What about
42:57the kind of the opposite?
42:58And now we have the
42:59tools to do that. So
43:00in the trade trial, which
43:02is gonna open here, hopefully
43:03soon,
43:05Patients are gonna be randomized
43:06to either tDxD or dapodimab,
43:09the trop two ADC,
43:10and then when they progress,
43:11they'll switch to the other
43:12one to see and then
43:14we'll try to figure out
43:15using biomarkers which is the
43:16best sequence for each individual
43:19answer. We don't know how
43:20to do that yet.
43:22And then,
43:23lastly,
43:25and this kinda gets back
43:26to the point I made
43:26at the beginning,
43:28can we get rid of
43:30conventional chemotherapy altogether?
43:33Ideally, why would you use
43:34conventional chemotherapy, which goes everywhere
43:36in the body and causes
43:37nonspecific toxicity? Why would you
43:39use that when you can
43:40link it to an antibody
43:41and deliver it to the
43:42cancer cell?
43:44To do that, you're gonna
43:45need anti ADCs with different
43:47payloads.
43:48Just like, you know, in
43:50practice with metastatic disease, we
43:51use chemotherapy a. Patients progress,
43:54we use chemotherapy b, and
43:55so on.
43:56You could do ADC a,
43:58ADC b, each just switching
44:00payloads.
44:01The problem is we don't
44:03have those drugs other than
44:04TDM one and TDXD I
44:06showed you. We don't have
44:07a lot of different payloads,
44:08and that's partly because the
44:10success of trastuzumab daroxetine
44:12was so high
44:13that everybody's jumping on the
44:15bandwagon of these topoisomerase
44:16inhibitor payloads.
44:18And you can see this
44:19is,
44:20from a review that just
44:21came out,
44:23showing, the different payloads that
44:25are being used.
44:26There's a hundred and seven
44:28in clinical development using,
44:30camptothecans or or basically topoisomerase
44:32inhibitor payloads.
44:33Almost all of them as
44:34shown in this graph are
44:36topoisomerase inhibitors.
44:37So they work,
44:39but we're getting very crowded,
44:41and you can imagine, and
44:42we have data now developing,
44:44that cancers can become resistant
44:45to the payload by,
44:47mutating topoisomerase.
44:51And then once you've got
44:53resistance to the payload, it
44:54doesn't matter which antibodies you
44:55hook up to it. It's
44:56it's not gonna work. So
44:58we need to diversify our
44:59payloads.
45:00So with that, I will
45:01stop. I'm happy to take
45:03questions.
45:10Oh,
45:11yep. Dan.
45:17Excellent talk, and I have
45:19two questions.
45:20Number one,
45:21I'm fascinated by this,
45:23observation about responses in brain
45:25metastases.
45:27It's known that in hyperprolisular
45:29vascular endothelium that HER-twoneu is
45:31expressed. Have you looked at
45:32the HER-twoneu expression in the
45:33vascular endothelium
45:35in addition to the tumor
45:36cells in those brain mets?
45:37Because it would certainly make
45:38sense if they have a
45:40bystander effect,
45:41that may that may be
45:42the actual mechanism of action.
45:44And the second question is
45:45about interstitial lumenitis. This is
45:46something we've seen with enfortumab
45:48vedotin.
45:49It's not really clear whether
45:50that's because of the interaction
45:51between checkpoints and enfortumab.
45:53But what do you think
45:54is the mechanism,
45:55with the HER2 targeted agents
45:57with that? Yeah. So both
45:58good questions.
46:00You know, when you look
46:01at the
46:02you know, I
46:03don't know
46:05I haven't seen data on
46:06looking at at the HER2
46:08expression on the vasculature,
46:11around tumors specifically in solid
46:13cancers. It's a great question
46:14and
46:15should be able to be
46:16looked at. We do enough,
46:17you know, resections of these
46:18cancers.
46:20You know, it's,
46:22you know, the the alternative
46:24you know, so if it's
46:24not just breakdown of the
46:25blood brain barrier,
46:27you know,
46:28is it released in the
46:29microenvironment?
46:31There are some data that
46:32I didn't have time to
46:33show you.
46:36But I think very provocative
46:37data that was presented at
46:38our San Antonio Breast Cancer
46:40meeting last month, just as
46:41a poster because I don't
46:42think people appreciated
46:44the impact,
46:46suggesting that it's actually
46:48cathepsins in the microenvironment.
46:50They're actually cleaving these conjugates,
46:52and that's why they work
46:53in, you know, potentially HER2
46:56null or HER2 very low,
46:58because they're really they're hurt
46:59it's you don't need the
47:00HER2. It's just the microenvironment
47:02has enough cathepsins there,
47:04that you get selective cleavage
47:06around the tumor.
47:07So that could be a
47:08alternative explanation. But,
47:11also, if you had anchoring
47:12because of HER2 overexpression in
47:14the vasculature,
47:15you could imagine,
47:17having efficacy there
47:20and perhaps
47:21causing more disruption by causing
47:24some apoptosis of the endothelium,
47:26causing more disruption, allowing more
47:27ADC in there. So I
47:28think that's a great question,
47:30and I don't have a
47:30great answer for that. And
47:31in terms of the ILD,
47:34we we we don't know
47:36the the mechanism. It it,
47:39actually,
47:40Adriana Khan is looking at
47:41trying to do lung biopsies
47:43in patients who get pneumonitis
47:44or or who are getting
47:45these drugs and get pneumonitis,
47:47to try to kinda clarify
47:49that.
47:50You know, certainly, there is
47:51HER2,
47:52in in in some lung
47:54tissue,
47:57whether it's through direct target
47:59mediated, although as you pointed
48:00out, there are multiple targets
48:02that are that are seeing
48:03ILD.
48:04There are,
48:06you know, with TDM one,
48:08which is HER2 target, you
48:09don't you know, the ILD
48:10rate is is is very,
48:12very low.
48:14And so we have ILD
48:15in some drugs where you
48:17change the payload or change
48:18the linker and you you
48:19you change the ILD rate
48:21substantially.
48:23So we don't know. You
48:24you know, is it macrophage
48:25uptake because of FC receptors?
48:27I think there's a lot
48:28of hypotheses, but I don't
48:29think there's any definitive,
48:31data.
48:32And, you know, it is
48:33certainly a problem for some
48:34of these drugs that limits
48:36their their, you know, applicability.
48:39Thanks.
48:41Yeah.
48:46Okay.
48:47Thank you for your talk.
48:49I had a question about
48:50the brain mets.
48:52Have you seen any coexpression
48:53of cell adhesion molecules that
48:55you could then use as
48:57a, like, HER2 bispecific
48:59to increase internalization
49:01in those brain meds specifically
49:03or the potential Yeah. So,
49:04I mean, again, this gets
49:05back to, Dan's question and
49:07and then,
49:08you know,
49:09whether we can learn from
49:11some of these
49:13datasets where people are doing
49:14resections of of brain metastases
49:16and and looking at, dysregulation
49:18of adhesion molecules and as
49:20a way of potentially doing
49:21bispecifics. I mean, there have
49:22been some
49:24ADCs looking at at those
49:27at at adhesion molecules to
49:28try to get at the
49:29at the, microenvironment, but I
49:31I don't know of any
49:32of that with her too.
49:34But it's a good idea
49:35and and kinda looking at
49:36that. Because, again, the BRAINMET
49:37issue is a is a
49:38real problem. And, where we've
49:40seen we see benefit with
49:42ADCs in the brain as
49:43I showed you, but they're
49:44not,
49:45you know, they're not,
49:46eliminating the brain metastases. And,
49:48generally, what we see with
49:50our patients, once a patient
49:51has brain metastases,
49:53that's progressed after radiation,
49:55they always progress in the
49:57brain. And so it becomes
49:58the rate limiting step for
50:00a lot of patients. So
50:01we we need new treatments,
50:02but that's a good question.
50:04Yeah. Great talk, Ian. I'm
50:05one of the best molecular
50:07pathologist.
50:08The trial you presented, was
50:09AutoMedigar when you predicted the
50:11pathologic complete response by the
50:13heterogeneity
50:14Yeah. In HER2.
50:15I we encounter heterogeneity on
50:17IHC all the time. And
50:18sometime when we go for
50:19fish, it doesn't translate to
50:21heterogeneity in fish in this
50:23area. But also on FISH,
50:25any FISH I review, there
50:26is negative cells in the
50:28FISH.
50:29Can how can we make
50:30this clinically applicable?
50:32And can this patient go
50:33instead of going getting new
50:35adjuvant antibody drug conjugate? Because
50:37they're not gonna achieve BCR
50:39with the regular regimen. Thank
50:40you.
50:41Yeah. So, I mean, with
50:43TBM one, we've definitely seen
50:46in in
50:47in both of the studies
50:48I showed you and and
50:49pretty much every study of
50:50TBM one, which, again, non
50:52cleavable link are very dependent
50:54on HER2 expression,
50:55that you see less substantially
50:58less activity as you go
50:59down to either lower expression
51:01levels of HER2 or heterogeneity.
51:03But in truth, we don't
51:04use TDM one in the
51:05neoadjuvant setting. It's not it's
51:07it's not clinically
51:08used.
51:10And
51:11as Eric has been potting
51:13me for years
51:14to go back and redo
51:15that trial of heterogeneity with
51:17one of these conjugates that
51:18has by standard effect to
51:20see whether we eliminate that
51:23disparity.
51:24You would expect we would
51:25if the hypothesis was correct,
51:28but we haven't proven it.
51:29But one of the as
51:30I said, there's a big
51:31trial that's just been it's
51:32been completed. We're waiting for
51:34the results. We should get
51:35it sometime this year of
51:36neoadjuvant
51:37TDXD,
51:38the one with the payload,
51:40spreading, the one with,
51:42bystander effect.
51:44And, hopefully,
51:47I'm not involved in that
51:48trial, but, hopefully,
51:49they'll look at that question
51:50and and and hope and,
51:52hopefully, we won't see that
51:53just big disparity
51:55by HER2 level,
51:56because of the unique features
51:58of this conjugate.
51:59But if we do, then,
52:00again, it brings up your
52:01point and says, hey. You
52:03know, a a one size
52:04fits all approach is not
52:05right, and the heterogeneous cancer
52:07should be treated in different
52:08ways,
52:10which probably is
52:12you know? I didn't have
52:13time to talk about resistance
52:14or biomarkers,
52:16but we're not good at
52:18figuring out either one of
52:19those,
52:20areas right now. We need
52:21a lot more work on
52:22identifying
52:23whether there are biomarkers that
52:24predict benefit of any of
52:25these drugs. We haven't been
52:26able to figure that out
52:27yet. Eric, do you have
52:28a follow-up?
52:30Question from online.
52:32So this is the limitation
52:34of PCR
52:35versus long term.
52:37Because just because someone doesn't
52:39achieve a PCR
52:40doesn't necessarily mean for going
52:42to have a good
52:44answer.
52:45So some of those cells
52:46that may be first and
52:47negatives
52:48may also be ER positive
52:50in response to anything therapy,
52:52maybe biologically
52:53less aggressive.
52:55So it's it's not
52:57absolutely the case when you
52:58have to eradicate every single
53:00one of their
53:01cell. The question
53:05online,
53:06was,
53:07do you think it would
53:08be possible using antibody drug
53:10conjugate
53:11technology to deliver non chemotherapy
53:14agents like,
53:17immune checkpoints
53:18or such?
53:20Yeah. So, yeah, so it's
53:21a great question, and then
53:22there's been a lot of
53:23interest in in in delivering
53:25everything with that,
53:27antisense,
53:29you know, as I said,
53:30DNA damage,
53:32inhibit you know, repair inhibitors,
53:36I think you name it.
53:37I think the the problem
53:39the the concern people have
53:41with going that direction,
53:43not that it's not being
53:44tried, but the concern is
53:45just gonna be, you know,
53:49the amount of ADC that
53:50actually gets the tumor cell
53:52is very low,
53:53because they get taken up.
53:54Even though,
53:56you know, we talk about
53:57it being a guided missile,
53:59most of the drug actually
54:00ends up in just
54:02random tissue just because it
54:04just antibodies get stuck places.
54:06So, you know, data suggests
54:08that, like, one percent of
54:09the dose that you give
54:10actually gets to the tumor.
54:11So you don't really getting
54:12a whole lot of of
54:13the payload to the cancer
54:15cell. And so
54:17there's been a push to
54:18get very high potency
54:21payloads
54:22with the idea that you
54:23can get away with that
54:23because they're being somewhat selectively
54:25delivered,
54:26because you you we're not
54:27getting a whole lot into
54:28the cancer, so you you
54:29want what you get in
54:30there to be very potent.
54:32And the problem with most
54:33of these small molecules,
54:36is that they're not as
54:37potent as as they need
54:39to be or the concern
54:40is they're not as potent
54:41as they need to be.
54:42We'll see. I mean, again,
54:42there's a lot of them
54:43in development, but that's the
54:45problem with,
54:47with your payload is if
54:48it's not
54:49quite potent,
54:51it may not be effective
54:52enough even if
54:53the general
54:54hypothesis is a good one
54:56of of of doing that.
54:57So but, you know, none
54:58of them have been approved
54:59as far as I know.
55:00We'll have to we'll have
55:01to see as they pan
55:02out. It's a it's a
55:03good idea.
55:04Did did you have a
55:05question?
55:06Yeah.
55:17Is is autoimmune, like is
55:19autoimmune
55:20toxicity?
55:22So well, autoimmune meaning anti
55:24antibody.
55:25Yeah. So so,
55:32antihuman antibodies or anti
55:35conjugate antibodies
55:36don't seem to be a
55:37big problem with these. It's
55:39it's a good question because
55:40you would expect you've, you
55:41know, you've got a a
55:42humanized antibody. You've got derivation
55:44of that antibody.
55:46Could that be immunogenic?
55:48It's it hasn't been a
55:50a problem.
55:53You don't see much hypersensitivity
55:54with these.
55:56I mean, you can't at,
55:57you know, low percentages.
55:59But but antibody you know,
56:01anti ADC antibodies haven't been
56:03clinically significant,
56:04and autoimmune disease really hasn't
56:06been. You know, whether the
56:07the pneumonitis that we see
56:08could be an autoimmune reaction
56:09is certainly possible, but but
56:11we other than that, we
56:12really haven't seen it.
56:25Yeah. So, you know, meaning
56:26if the ADC bot is
56:28is encounters an immune cell,
56:30it kills it because of
56:31the cytotoxic
56:32moiety. Yeah. I think that
56:33that's that's certainly possible.
56:37But,
56:38and then that's a good
56:39thought.
56:40Fortunately, I said clinically, it
56:41just really hasn't been an
56:42issue.
56:50Pan,
56:51I had, like, twenty questions,
56:52but I'll Absolutely. Eliminate them.
56:55So so
56:56with with the ultra low,
56:58you're you're assuming that you're
56:59dealing with with
57:00probably a small percentage of
57:02cells that are actually expressing.
57:04By definition. And I guess
57:06the I understand the idea
57:07of cathepsins doing this. But
57:08I guess the other question
57:09is for the spreading effect,
57:11do we think this is
57:11apoptotic cell death, or do
57:12we think this is cell
57:13death that's actually messier? And
57:15if so,
57:16has anyone thought of immune
57:19in combination,
57:20you know, like p one?
57:22So I didn't talk about
57:23biomarkers. I'm sorry. I was
57:24just last night. I don't
57:25wanna forget. Forget. Brady and
57:26Anna's thing. I know David
57:27Brim is gonna be doing,
57:29QIF. Are you also gonna
57:31be doing IHC at the
57:31same time? Because I think
57:32you sort of I go
57:33on and do both
57:35because I mean, at least
57:36David's not here. Right? He's
57:37been testing,
57:40you
57:43know, Yeah. We're we're actually
57:45see if it's covered because
57:45Yeah. So the actual correlate.
57:47Yeah. Yeah. The actual analysis
57:49plan is to do multiple
57:50HER2 assays, both
57:52protein and genomics
57:53to try to come up
57:54with, you know, the best
57:55predictor.
57:57So good question.
58:00In terms of of immunogenic
58:02cell death, it certainly looks
58:03like these are causing immunogenic
58:04cell death.
58:06You know, again, the payloads
58:08are standard chemotherapies, essentially, and
58:10you're getting a lot of
58:11it. You know, they're they're
58:12potent. So there's there are
58:13actually some interesting papers that
58:14you get particularly good immunogenic
58:17cell death for reasons that
58:18I I don't know how
58:19well they they've been validated.
58:21And there there are combinations
58:23of,
58:24ADCs plus checkpoint inhibitors,
58:27which have shown some promise.
58:30In smaller studies, there's some
58:32studies that have shown incredible,
58:33you know, response rates, eighty,
58:34ninety percent,
58:37and with and there have
58:38been some small randomized trials
58:40of of TDM one plus,
58:42checkpoint inhibitors, which show
58:44a signal, but not overwhelming
58:46signal. So I think,
58:50we'll have to wait,
58:52for
58:53there are randomized trials right
58:54now with the FOC two
58:56ADCs ADCs with or without,
58:58checkpoint inhibitors,
59:00being,
59:01being conducted, and we should
59:02have the results actually pretty
59:03quickly.
59:04The hope is that those
59:05are gonna be, you know,
59:07really,
59:08impressive combinations,
59:09both because we know that,
59:11you know,
59:12checkpoint inhibitors plus chemotherapy work
59:14well in triple negative breast
59:15cancer, but particularly because of
59:18the the the having an
59:19antibody there, maybe you're getting
59:21more,
59:22antigen presentation, maybe you're getting
59:24more immunogenic cell death that
59:25it's really gonna be,
59:27truly synergistic. So it's a
59:29great question, and we'll have
59:30data by the end of
59:31this year
59:32on that.
59:33Barbara?
59:39Wonderful talk. And I'm so
59:42fascinated by the HER2 low
59:44because, I guess,
59:45I think we have some
59:46head and neck cancers that
59:47are like that. But so,
59:49you know, in the in
59:50the amplified,
59:51you always have the same
59:52target, and you're just changing
59:53your payload, you're changing your
59:54link area, and it keeps
59:55working for years.
59:56And in the HER2
59:58low,
59:59I'm assuming that to the
01:00:00extent that that HER2
01:00:02is doing something biologically,
01:00:04it's heterodimerizing
01:00:05with EGFR or HER3. Right?
01:00:08And so I wondered either
01:00:10with a panHER kinase inhibitor
01:00:12or,
01:00:13you know, cetuximab or something
01:00:15like that, have there been
01:00:16attempts to kinda cotarget
01:00:18what HER2 is is hanging
01:00:20out with in the in
01:00:21those cancers? Yeah. So great
01:00:23question as always, Barbara.
01:00:25So as I was gonna
01:00:27tell Mike, you know, so
01:00:27we didn't talk about biomarkers
01:00:29or resistance. We also didn't
01:00:30talk about combinations.
01:00:33Combining things with ADCs has
01:00:34been more complicated than we
01:00:36would like. The the checkpoint
01:00:37inhibitors actually are an exception.
01:00:40There have been studies looking
01:00:42at kinase inhibitors for the
01:00:43reason you talked about.
01:00:46So far, the data don't
01:00:47look great.
01:00:51It's funny you bring that
01:00:52up. Jingde and I are
01:00:53just,
01:00:54in the process of submitting,
01:00:57a
01:00:58concept
01:00:59for,
01:01:01the combination of a HER2
01:01:02ADC plus the HER3 ADC
01:01:04that you're working on,
01:01:06for that very reason. Also,
01:01:07knowing that HER3 gets upregulated
01:01:09when you block HER2,
01:01:10can you, you know, can
01:01:11you leverage that by combining
01:01:13those two ADCs?
01:01:15But just
01:01:16from lots of unfortunate personal
01:01:18experience, combining ADCs with other
01:01:20drugs has not been as
01:01:21easy as we would have
01:01:22thought.
01:01:23Even TDM one, which, you
01:01:24know, is such a good
01:01:25toxicity profile, it's been hard.
01:01:28So
01:01:31Alright. Thank you very