Latest / EHA Unplugged / MRD Monitoring in MDS Patients (with Prof Arjan van de Loosdrecht)

Transcript

MRD Monitoring in MDS Patients (with Prof Arjan van de Loosdrecht) · EHA Unplugged

• 0:07
  Greetings and welcome to. EHA Unplugged.
• 0:09
  The official podcast. Channel of the European
• 0:12
  Hematology Association EHA. So today we are going to hear
• 0:17
  Professor Iron Vanderlustek from the Amsterdam University Medical
• 0:21
  Center and he will talk to us about MRD measurements in MD's
• 0:27
  but also a little bit in general in myeloid cancers.
• 0:31
  So I and can we use flow for this or how are we going to
• 0:37
  approach MRD measurements in MD's?
• 0:40
  Thank you for this kind interview, Kirsten, and I'm
• 0:44
  happy to discuss with you what is actually meant by measurable
• 0:49
  residual disease. It was known as minimal residual
• 0:52
  disease, but actually it's about measurable.
• 0:56
  We can measure the signal and the the we have to also to know
• 1:01
  what we are doing with that in the sense of that that measuring
• 1:06
  the signal of a disease after courses of chemotherapy or
• 1:10
  immunotherapy of whatever that therapy might be, is a sum of
• 1:15
  intrinsic features of the disease itself as well as
• 1:22
  resistance mechanisms due to whatever the drug might be as
• 1:25
  well as the complex interaction with the microenvironment.
• 1:29
  So after in general measurable disease is a complex sum of all
• 1:35
  kinds of effects. And therefore I think it is of
• 1:38
  interest that you have a kinetics of a disease in the
• 1:43
  treatment and at the end the results and the the signal of
• 1:47
  those measurable radial disease is highly correlated with
• 1:51
  disease outcome. In MD's we have some data, but
• 1:56
  the main data comes from from the AML field.
• 2:01
  And actually already some years ago in the AML guidelines, it
• 2:06
  was recognized that we can define a classical complete
• 2:10
  remission from Mylad disease, mainly AML.
• 2:14
  But on top of that, we can now also define complete remission
• 2:20
  after one or two or more courses of this of induction
• 2:23
  chemotherapy or other therapies with or without the detection of
• 2:29
  measurable residual disease. I think it's important that in
• 2:32
  AML it's already quite established both retrospectively
• 2:37
  as as well as in prospective clinical trials.
• 2:41
  So that's actually what it is. And isn't it not more
• 2:44
  complicated in MD's to to which which cells you should actually
• 2:48
  be looking for or which which signals?
• 2:52
  Yeah. It is a little bit dependent on
• 2:56
  what you use as a signal. If you think that the blasts by
• 3:02
  the microscope or better defined by flow cytometry for instance,
• 3:07
  the progenitor cells or even the leukemic stem cells.
• 3:11
  If you only focus on those very well defined cells as leukemia
• 3:18
  initiating cells, so for instance, then you can use the
• 3:23
  same principle. If you look better to MD's, it's
• 3:28
  really, really a stem cell disease.
• 3:30
  So that therefore I think measure to detect disease on the
• 3:35
  leukemic stem cell might make sense.
• 3:39
  But in upon that if you look to a lower risk MD's for instance,
• 3:42
  there's a whole range of also dysplasia in the more maturing
• 3:47
  cells. So you, you dependent on what
• 3:51
  you think is most important, you can either go to the early,
• 3:56
  early leukemia initiating cell, name it stem cells or more
• 4:00
  progenitor cells together with or without measuring dysplasia.
• 4:07
  And that is a fully different approach from AML towards MD's.
• 4:12
  And now we do not clearly know yet what is most important.
• 4:17
  I think measuring the leukemic stem cell burden, but maybe also
• 4:22
  in therapies concerning more modulatory effects.
• 4:26
  Maybe also apart from aberrant cells in the leukemic stem cells
• 4:31
  together with dysphoresis, which there is a different approach,
• 4:36
  maybe the some of those can detect best the effect of
• 4:43
  therapy. And and can you, can you maybe
• 4:46
  to this audience try to explain how you identify the leukemic
• 4:50
  stem cells? What are the, what are the for
• 4:53
  this? And yeah.
• 4:54
  Yeah, actually traditionally we, we we look with several
• 5:00
  techniques. 1 is multi parameter flow cyclometry, that's one.
• 5:06
  And secondly the molecular biology molecular markers of
• 5:10
  disease. Now within AML we only have
• 5:14
  validated for instance MPM one by quadrative PCR, that's a good
• 5:19
  marker and is now in the guidelines we can measure quite
• 5:23
  good that measurements going to the flow approach.
• 5:28
  It's all validated with a specific sets defined by the
• 5:32
  European leukemia net as well as slash EHA platforms, which is
• 5:38
  now robust defined and that concerns parameters within flow
• 5:45
  cytometry considering the progenitor cell compartment, so
• 5:49
  the CD 34 positive cell populations.
• 5:53
  And if you look to aberrances, we look to leukemia associated
• 6:01
  immune phenotype. That's a quite a long name.
• 6:05
  So we look to those aberrant antigen expressions on the level
• 6:10
  of the CD 34 positive cells. Yeah.
• 6:13
  So the progenic cells defined what we think are leukemia
• 6:16
  initiating cells. And if we look to aberrances,
• 6:20
  that means those antigens expressed on leukemia initiating
• 6:25
  cells and not expressed normal progenitor cells.
• 6:30
  And how many markers would you need?
• 6:32
  For that, yes, that can be discussed there.
• 6:38
  We use in general a set of seven or even more well recognized
• 6:44
  validated parameters. And among these are CD seven
• 6:49
  that is a lymphoid marker, but the lymphoid marker on a mild
• 6:54
  progenitor cell is aberrant. So there's a very strong one.
• 6:58
  And also CD2 or CD5 are markers of lymphoid origin normally not
• 7:05
  expressed on normal protein cells but expressed on their
• 7:08
  band cells. And there are some markers are
• 7:11
  more frequently found than others.
• 7:14
  And we should have a panel which can define most phenotypes which
• 7:22
  are most prevalent. And sometimes we have for
• 7:27
  instance in in the whole population of cells which are
• 7:30
  bound by CD7 and the sub sub population with for instance CD2
• 7:34
  or another marker. And then the leading phenotype
• 7:38
  is based on the most frequently existing CD 34 positive cell.
• 7:43
  But that's quite easy. The lymphoid protein cells of
• 7:46
  lymphoid antigens. And and what about if you look
• 7:49
  at MD's so? So will these Abair stem cells
• 7:52
  not be different within at least among the different subtypes of
• 7:56
  MD's or or also within a person is is is a low skinny stem cells
• 8:01
  always the same or how does? Yeah.
• 8:03
  Yeah, yeah, yeah. The comparisons between AML and
• 8:06
  MD's per SE are not yet there really robust data, but I think
• 8:12
  that there are not so much differences at all between these
• 8:16
  entities except from those antigens which are mild
• 8:21
  differentiation antigens such as CD13, CD 33.
• 8:26
  And then come MD's comes in because if you look to maturing
• 8:31
  cells, the expression of normally mild differentiation
• 8:36
  antigens are fully aberrant, are dysregulated.
• 8:41
  So that means for instance that CD 13 which is normally
• 8:46
  expressed intermediate on 34 positive cells is expressed very
• 8:51
  high on the next stage of differentiation to improve
• 8:54
  myelin sites and is then rapidly down regulated.
• 8:58
  So if we look to focus on flow cytometry now and we look to
• 9:04
  what is really aberrances on the progenitor cell level on the CD
• 9:09
  34 positive cells to make it a little bit easier is that you
• 9:13
  can look to antigens which are normally not expressed on myelin
• 9:17
  cells such as lymphoid markers CD7 I mentioned CD2, CD 5:00 but
• 9:23
  you can also look to so-called syndromous expression.
• 9:28
  That means that some differentiation antigens are
• 9:31
  highly dependent expressed on several differentiation stages.
• 9:37
  So that is also a kind of an embarrancy.
• 9:40
  So an an an antigen which is normally not expressed but it's
• 9:45
  well of mild, mild origin. So both the lack or the
• 9:50
  overexpression of those or acinogonias of the different
• 9:57
  differentiation path is all part of the so-called leukemia
• 10:02
  associated immune phenotype. And what I was thinking, you
• 10:06
  probably answered that, but but within an individual, yeah.
• 10:11
  Will will this this clone sort of be the same or or or will
• 10:14
  there be more different? I mean the heterogeneous how,
• 10:17
  how does it come out? Actually within that this
• 10:21
  compartment named CD34 positivity, we can find several
• 10:26
  labs, several leukemia associated about immune
• 10:29
  phenotypes. And what we now are doing in the
• 10:35
  standardised European leukemia platforms is that the most
• 10:40
  prevalent LAP leukemia associated phenotype is the
• 10:46
  phenotype we will follow. But it's not fully true because
• 10:50
  you are several clones maybe which are aberrant.
• 10:55
  So what we now do is we at the start we have an 5.0 phenotype
• 11:02
  and we define those phenotypes, the cephalophenotypes and then
• 11:08
  we follow them during the treatments and in early days and
• 11:14
  they can change. So the shift in
• 11:16
  immunophenotypes, what's called the shift immuno shifts come up
• 11:22
  or come down. Some several phenotypes go away
• 11:25
  which might might be represented for a specific clone and others
• 11:30
  come and like molecular clone definition.
• 11:34
  So DMT 3A or Ted 2 or others can occur within a specific clone or
• 11:41
  a part of several clones. And what the research direction
• 11:45
  is going on that we separate that by flow fax machine.
• 11:48
  So by separating several flaps and put them into a molecular
• 11:54
  analysis and then look whether or not those clones really
• 11:59
  represent different molecular defined clones or not.
• 12:03
  So is this research or is this actually?
• 12:05
  That's research. That's research.
• 12:06
  That is research because it sounds.
• 12:07
  A little bit complicated for for everything for.
• 12:12
  Routine, well validated platforms.
• 12:15
  Those platforms are available, but they focus mainly on the CD
• 12:20
  34 compartment. Yet, no, they're also CD34
• 12:24
  negative leukemias. And then we do have another
• 12:28
  approach. I mean we have the 117 as a
• 12:32
  marker, but not in a probyocytic stage, but in the United States.
• 12:36
  And in addition to that, but that is also then experimental
• 12:40
  is going to a level deeper to the CD 34 positive, CD 38
• 12:45
  negative, yeah, published. And that is the what you call
• 12:49
  now the stem cell compartment. Yeah, the very early ones.
• 12:52
  The very early and that's of interest that upon classical
• 12:56
  MRD, the leukemia stem cell MRD and you combine that both then
• 13:04
  you have more discriminative power in the prediction of
• 13:10
  relapse. So and that that is in line with
• 13:15
  what what you was asking. Can we combine several
• 13:18
  technologies. So within the traditional flow
• 13:22
  we can add now leukemia stem cell flow and there's just a
• 13:28
  paper or several papers are already published on that
• 13:31
  subject, but not routinely used yet.
• 13:34
  But then you can ask what can we do with for instance
• 13:38
  quantitative PCRMDA like the MPM one?
• 13:41
  Yeah, you should, yeah. Yeah, yeah.
• 13:43
  And, and, and that is well established.
• 13:44
  And if you combine those technologies together, they have
• 13:49
  an increase in predictive power. Yeah, so, so, so the so you
• 13:54
  know, because what what the hospital administrators always
• 13:58
  ask you, can you not take out one of these?
• 14:00
  I mean it's expensive to flow and NTS but but but the answer
• 14:04
  is it's better to combine, right?
• 14:05
  Yeah, at the end and and, and and and if you look to NGS for
• 14:10
  instance and excluding the tattoo and the the chip like
• 14:14
  backbone mutations, you have to rule out them.
• 14:18
  But also within the whole from prospective studies you may be
• 14:21
  aware, but also from other studies from the UK, it was
• 14:24
  shown that putting upon the flow well established platforms on
• 14:29
  next generation sequencing, you can better discriminate patients
• 14:36
  with the predictive power of incidence of relapse.
• 14:39
  Yeah, yeah, that makes sense, of course.
• 14:42
  Yeah, but. But the main question is,
• 14:44
  Kirsten is OK if you know that we have a predictor, but if you
• 14:49
  treat them differently, is the outcome outcome also then
• 14:54
  better? Yeah, that's absolutely the most
• 14:58
  important question, right. There is a treatment because I
• 15:02
  think in MD's we have rather limited I, I guess in AML there
• 15:05
  are more options, right. But yeah.
• 15:07
  And if you, if you, if you translate it to AML 1st and then
• 15:09
  come to the MD's question more specific, if you look to some
• 15:15
  data how you can guide MRD for clinical decision of you, how
• 15:21
  you can guide treatment decisions based on MOD.
• 15:25
  There are data that especially in the intermediate risk group
• 15:29
  in AML. If you have a intermediate risk
• 15:33
  according to the European Leukemia Risk Classification and
• 15:36
  you are MOD negative after two courses, you can then better go
• 15:45
  to an orthologous stems of transportation.
• 15:48
  Yeah. Yeah, rather than an allogeneic
• 15:50
  transplantation, because then the outcome is much more better
• 15:55
  because you do not lose the patience of the toxicity of the
• 15:58
  transplant, which is. Interesting to me because I
• 16:02
  think we hardly ever use the Toluca's transplant in AML.
• 16:05
  Yeah. That, that, that, that is yeah,
• 16:08
  it's a standard in in the hope on the group and and because we
• 16:14
  lose, we lose patients due to an allotransplant.
• 16:19
  Yeah, yeah. Yeah.
• 16:22
  Right, but then it's not true for the MD's.
• 16:24
  I think MD's is really another disease in AML.
• 16:26
  No, MD's, yeah, I don't think you can find the problem in MD's
• 16:29
  is to find a healthy stem cell, isn't it?
• 16:31
  I mean, that's. Yeah, yeah, yeah, yeah, yeah.
• 16:34
  What is it actually? But if you now go to the to the
• 16:36
  really the MD's setting and we have the data now and also from
• 16:40
  other groups which are a mixture of AML as well as MD's and
• 16:45
  treated accordingly with high risk of high dose chemotherapy.
• 16:50
  And if we look to the data produced by by the whole phone
• 16:54
  group for instance, in general we see that patients with high
• 16:58
  risk MD's versus AML with the same induction scheme, it's the
• 17:03
  same consolidation regiments MD's is always worth.
• 17:09
  But then it is interesting if you then separate those patients
• 17:13
  within the MD's, the high risk MD's by definition and we
• 17:18
  separate those MD's, patients with MRD positive after two
• 17:24
  cycles of classical chemo induction chemotherapy, they do
• 17:28
  even worse than MRD positivity in the AML setting, what is
• 17:33
  expected. Yeah, yeah.
• 17:35
  But if you are MRD negative in MD's you do as bad as the MRD
• 17:42
  positive in AML. So can you follow me?
• 17:47
  So normally if you are MDMDI negative, in general you do
• 17:52
  better. But if you are MD negative and
• 17:56
  MD's as a background disease, you do as bad as AML which are
• 18:02
  MD positive. So although these data should be
• 18:05
  confirmed and still we are busy with the analysis to cut into
• 18:08
  more detail. It is, but fortunately it's
• 18:12
  it's, it's linked to a molecular backbone, I think.
• 18:15
  Yeah, the molecular backbone of bone of MD's is different,
• 18:18
  right? From yeah, which dictates the
• 18:21
  outcome I think. But but but that can already be
• 18:24
  measured by MRD status. So isn't it a problem that I
• 18:27
  mean when, if you if in in the ideal setting, when should you
• 18:31
  use this MRD measurements in MD's because with the number of
• 18:36
  options we have for treatment and so how do you see?
• 18:39
  This if if you look to to high risk MD's for instance for the
• 18:43
  first then in general high risk MD's which will go to an
• 18:49
  allotransplant and with an excess of blast have
• 18:52
  conventional induction there induction chemotherapy in most
• 18:56
  countries then I think as we do and the and the data of other
• 19:01
  groups support is that we use the classical induction and we
• 19:06
  follow the MID status according to AML protocols.
• 19:10
  Also before or you mean or after?
• 19:15
  Before going to a transplant even we have although the
• 19:19
  prevalence is very low, you you have MD's patients which are
• 19:26
  relatively good risk AML like. Yeah.
• 19:30
  And those patients are still going to autologous transplants
• 19:35
  in hopeful study. Wow.
• 19:38
  Wow. And that is different and
• 19:41
  whether or not that is fully correct, the data see the same
• 19:45
  results. So I think we in that context we
• 19:48
  are doing well. But in general the most patients
• 19:51
  are high risk, high risk. So I have a high European
• 19:55
  leukemia risk profile or a IPSS molecular risk profile which is
• 20:00
  high. So they should go to a
• 20:01
  transplant, but does not necessarily exclude.
• 20:06
  If you go a little bit further with the with this MRD in MD's
• 20:13
  and we go to a lower risk, the data are coming and what we can
• 20:19
  see. For instance, if you look to
• 20:23
  dysplasia in MD's by flow at the concept that you can can
• 20:32
  objectify quite objective the aberrances in the maturational
• 20:39
  profile of lowest MD's. And if you then look to for
• 20:45
  instance in the treatment with Lenalidomide, for instance in
• 20:47
  the Dell 5Q or even in a non Dell 5Q, you can see that if
• 20:52
  patients respond to Len for instance by hematological
• 20:59
  improvement, we in general see that also the amount of
• 21:04
  appearances on maturing cells is going to normal.
• 21:13
  Then you see the appearances in in in expression profiles.
• 21:16
  So yeah. Right.
• 21:18
  So the dysplasia objectified by flow is normalized to do what's
• 21:25
  normal hematopresses mainly in those patients who also do
• 21:30
  response with the current response criteria like
• 21:34
  hypothetical improvement or blood cell transfusion
• 21:37
  independency. So that means that there is a
• 21:41
  correlation of restoration of towards normal and the known
• 21:50
  improvement data if we can really use that in clinical
• 21:55
  practice for instance, Yeah, to to ask the question in patients
• 22:01
  who are not responsive to Len for instance, yeah, However they
• 22:07
  normalise the spastic hematomyces.
• 22:12
  What is the outcome of those patients in terms of event free
• 22:16
  of OVALSA 5 or duration of transfusion in dependency that's
• 22:21
  only a little bit hypothesis different yet.
• 22:24
  So we don't not yet exactly What we do know is the normalization
• 22:32
  of hematopresses in the ASA treated patients group for in
• 22:36
  the azocytidine. There are two papers published
• 22:41
  showing that patients under ASA so high risk MD's patients which
• 22:50
  do not respond clinically at six months.
• 22:54
  However, the amount of aberrances are significantly
• 23:00
  lower control to baseline, we have a better overall survival
• 23:08
  and that's interesting because should.
• 23:10
  We continue, should we not continue that right?
• 23:12
  Exactly, exactly, exactly so. You can be guided by that
• 23:17
  because that would be. Prospective data are not yet
• 23:20
  there, only data suggesting that patients who are stable but do
• 23:25
  not really approve regarding transfusion dependency or other
• 23:30
  parameters, but do have normalization of the abnormal
• 23:35
  hematopoiesis do better. And then you can decide, OK, we
• 23:38
  have no progression this one. We are stable but not better
• 23:42
  regarding transfusion dependency.
• 23:44
  However, the dysplastic features are better then we are allowed
• 23:52
  to continue ASA treatment longer.
• 23:55
  I think that would be. Very, very important basis
• 23:58
  because this is always a challenge.
• 24:00
  You know when to stop them. So that doesn't that this could
• 24:03
  be have a direct effect on on clinical decision making.
• 24:07
  Yeah. And that's but the data are
• 24:10
  maturing and actually you should have an in, in the randomized
• 24:14
  trial in which you have that stable patients you stop and the
• 24:18
  other group you continue and then etcetera, etcetera.
• 24:21
  Then that's the really final proof.
• 24:24
  But the concept is of interest. Yeah, it's very interesting.
• 24:28
  But you can also say why not buy molecular markers?
• 24:31
  But the molecular markers are not so indicative to stop or not
• 24:36
  to stop. I guess in most, in most
• 24:40
  situations they're not. And which ones should we then
• 24:43
  pick? Right, because.
• 24:44
  Yeah, right. Not the chip.
• 24:45
  The chip alterations, but I mean even among the others, they seem
• 24:49
  to be going a little bit up and down, you know?
• 24:51
  Yeah, so. What the transplant?
• 24:55
  Have you any experience with that?
• 24:57
  I mean after use measuring the transplant.
• 25:01
  We do not have, we do measure after allotransplant also MRD
• 25:06
  levels not in clinical trial setting.
• 25:11
  But the use of MRD after transplant is of importance
• 25:16
  because you can taper earlier even suppression for instance,
• 25:20
  or you can add open on the ally in the absence of of professor's
• 25:24
  host for instance. However, prospective studies are
• 25:28
  not there, but you know that in a Scandinavian group, mainly
• 25:32
  Sweden and you are part of that, they try to to molecular define
• 25:40
  the post anal transplant setting and then also in the same
• 25:44
  concept direct immune suppression or DLI or other
• 25:50
  immune modulatory strategies upon MOD assessments.
• 25:54
  Yeah, yeah, but but. I guess in this in that
• 25:58
  situation do would you then go for the for the molecular, the
• 26:02
  the flow or what would you, what do you?
• 26:04
  Not know not not and that is the the difference between the
• 26:10
  molecular beliefs and the flow beliefs I think.
• 26:13
  And at the end you should combine it.
• 26:15
  Yeah, I think you're right. Also, according to what you've
• 26:18
  shown earlier, I think you know like.
• 26:21
  Yeah, that's good. In clinical practice a lot, a
• 26:24
  lot of groups already use the the GAIMERISM as well as the MOD
• 26:30
  assessment post LO to optimize their post LO treatment
• 26:35
  strategies. And I think that is true.
• 26:37
  That's in some extent clinical practice, but there are no
• 26:40
  randomized trials yet to to make it really clear.
• 26:44
  Yeah. I know read the mice trials at
• 26:47
  all about adult transplants so. Now it's interesting how this is
• 26:54
  the most dangerous treatment in a way we have and still there.
• 26:58
  Are a lot of discussions about the best mode?
• 27:01
  Yes. So if I can summarize, you may
• 27:05
  allow me. I think it is important to to to
• 27:11
  have the MID on board of treatment decisions in the
• 27:14
  future. In the AML field, it is well
• 27:17
  established at note the the response criteria for MD's, high
• 27:23
  risk MD's are just defined by the working group.
• 27:28
  But note MID is not in the response criteria in high risk
• 27:33
  MD's yet. So that is the next step.
• 27:36
  But on the other hand, I think that the translation from AML to
• 27:41
  higher MD's is quite easy to go, but should be confirmed in in
• 27:49
  the in the next years. In addition to the concept of MD
• 27:54
  by measuring the CD 34 positive cell compartment or or or the
• 27:59
  molecular stuff upon that, you can also think about the balance
• 28:06
  in restoration of the spastic hematopoiesis.
• 28:10
  What we discussed and that is a new field and it is not yet
• 28:14
  established. Can we really use a direct
• 28:17
  therapy based on normalization of the spasia And they give you
• 28:21
  some examples of in the ASA fields in the high risk not
• 28:26
  going to a transplant on the HMA.
• 28:29
  Those data are quite convincing and they're also in the low risk
• 28:33
  MD's in the Lenalidomide. There are upcoming new data
• 28:37
  which show that the hematology improvement is at least
• 28:40
  correlated to the normalization of dysplastic unit decreases,
• 28:45
  but how to use that in the clinical decision making is only
• 28:48
  hypothesis driven yet. But this would really, I think
• 28:52
  that's really an interesting way forward where I think it could
• 28:55
  make an impact on on our clinical practice.
• 28:58
  I think that. Yes as well because the patients
• 29:00
  are all with low risk Mdas and thereby we have to carefully
• 29:07
  treat them all regarding toxicities.
• 29:10
  And therefore you can make your decisions maybe earlier, but or
• 29:14
  not that it makes sense or not to continue with at the end and
• 29:17
  inappropriate therapy at all. So that that is the interest of
• 29:21
  the balance. Thank you so much.
• 29:23
  I think we came around a lot of very interesting topics.
• 29:27
  So thank you very much for your time, Aaron.
• 29:30
  And I think we now know a little bit more about MRD in India.
• 29:35
  It it was my pleasure to to have this conversation with you.
• 29:38
  Thank you. Thank you also very much.
• 29:40
  Thank you.