Existing
diagnostic criteria for mastocytosis:
·
Over the past four years, well-accepted criteria have been developed
to distinguish the MPDs, a very diverse group, from
each other.
·
Excellent criteria to separate Ph+ from Ph- disease in CML
·
Common links exist among the different criteria.
- e.g., degrees of fibrosis among the MPDs
- e.g., mast cell disorders among the MPDs
- e.g., each disorder can spontaneously progress
to acute myelod leukemia, or "secondary
AML"
SYSTEMIC
MAST CELL DISEASE (SMCD): ETIOLOGY/MOLECULAR BIOLOGY/PATHOLOGY
Overview
of mastocytosis
Dr. Peter Valent posed a list of questions to participants:
1. Do we have
sufficient diagnostic criteria for mastocytosis and its subtypes?
- He believes the WHO diagnostic
criteria are fine and represent a major step forward.
2. What is the exact place of mastocytosis within the group of
myeloid malignancies?
- Over last few years, the hypothesis
that systemic mastocytosis is indeed a MPD
has been strongly supported, particularly by Dr. Cem Akin's data
from Dr. Dean Metcalfe's
lab.
Overview
of mastocytosis.
·
Most patients follow indolent course; but some may have associated
hematologic disorders (MDS, MPDs)
·
Associated with activating mutations in c-kit, the receptor for
stem cell factor, the most important cytokine for
growth and differentiation of mast cells.
Mastocytosis
Consensus Classification (established 2000)
· Published in July 2001 Leukemia Research
·
Adopted by WHO
·
Seven subtypes
·
Most commonly encountered subtypes: cutaneous mastocytosis
and indolent systemic mastocytosis
·
New, provisional addition is "smoldering systemic mastocytosis":
- Patients with indolent systemic mastocytosis
with increased mast cell burden who
may be progressing to more aggressive subtypes.
- Hepatosplenomegaly, increased tryptase levels
and surrogate markers of disease.
- No definite evidence of MDS/MPD.
Similarity
of mastocytosis to MPDs denoted by c-kit mutation in pre-pro-B
cells
Dr. Cem Akin
presented data on clonality in mastocytosis:
· mRNA studies are not sufficient to determine role
of progenitor cells in mastocytosis because mRNA does
not give us information about cell lineage.
·
Study by Dr. Akin's group addressing cell lineage question:
- Detected c-kit mutation in patients with extensive
mast cell disease at single-cell level:
* In peripheral B cells
and monocytes
* Not in peripheral
T cells
- Found that adenine was replaced by thymidine
in heterozygous manner in c-kit.
- Found that Ig gene arrangement suggesting
B cells were polyclonal.
- Conclusions of study:
* c-kit mutation in
B cells in patients with extensive mastocytosis occurred at a
stage earlier
than the pro-B cell stage.
* B cells in patients
with extensive mastocytosis are polyclonal but express the c-kit
mutation.
* Data supported hypothesis
that mastocytosis, especially smoldering subtype of
indolent systemic mastocytosis:
·
Is a disorder of the hematopoietic progenitor cells of B cells
and myeloid cells.
·
Bears striking resemblance to the other MPDs in etiology and pathogenesis.
·
Studies of other MPDs have also supported the hypothesis that
B cells are involved in the clonal process in these
disorders; but, these studies have been conducted only
in a limited number of patients:
- Dr. Ayalew Tefferi: 30% of CML cases in 12-year-old
study had B cells demonstrating
BCR/ABL fusion signal by D-FISH, with some involvement of T cells
as well.
- Dr. Tefferi: X-linked DNA analysis in patients
with ET, PV, etc. showed involvement
of B and T cells, even at cytogenetic level. A very small study
in MF using X-linked DNA analysis
also suggested involvement of B cells.
- In original Ph PV study in Seattle, a population
of clonal B cells was identified.
- Dr. Valent: B cells and myeloid cells in 20-year-old
CML study were clonal, whereas most
of the peripheral T cells were not. Same holds true blast crisis
CML (mostly myeloid, 20-30% B cell,
rarely T cell).
·
Consensus:
- B cell lineage clonality has been demonstrated
in mastocytosis as well as in Ph+
CML.
·
Recommendation:
- Expand number of patients with MPDs in studies
looking for clonality outside of the
myeloid lineage (esp. B cells). These should be confirmative studies.
Dr. Valent commented that this recommendation
should be extended to include a diagnostic
aspect.
General
discussion on c-kit mutation in mastocytosis
·
c-kit mutations in mastocytosis affect codon 816 in the tyrosine
kinase domain of the molecule, most commonly D816Y.
- Cause ligand-independent autophosphorylation
of the tyrosine kinase function of the
receptor and growth factor independence.
- Demonstrated in lesional mast cells, mostly
in adult onset disease.
·
Question: Would the presence of the c-kit mutation be a reliable
indication of subtype diagnosis, e.g., of smoldering
mastocytosis?
- Response: No. Patients with non-mast cell
clonal disorders have had the c-kit mutation
detected in their peripheral blood.
·
Mast cell disorder is the only MPD in which a c-kit mutation
has been detected.
·
It is to early to estimate the proportion of each MPD carrying
the c-kit mutation.
·
c-kit detection is an issue of clonal expansion and disordered
cell migration.
- The larger the clone expands, the more likely
you are to find it in bone marrow, and
then in peripheral blood.
- It really depends on which cells you examine
for the c-kit mutation. Some cells offer
higher sensitivity for detecting the mutation than others (e.g.,
marrow versus skin).
- Collections of mast cells in marrow and skin
are not only the result of clonal expansion,
but also of disordered cell migration.
* CD34+ cells with c-kit
mutation were observed to migrate more rapidly to stem
cell factor.
Opportunities
for chromosomal research in mastocytosis
·
No consistent or distinct chromosomal abnormality has been
found in the mast cell proliferative disorders.
·
Question: Perhaps chromosomal research is an area for further
research in mastocytosis, using current molecular
techniques, such as simultaneous staining of cells
with fluorescent-labeled CD34+ and application of FISH?
- Response: It is difficult to determine abnormalities
in mast cells for two reasons:
* mast cells do not
divide rapidly
* mast cells represent
a very minor fraction of the bone marrow. (<1% of bone marrow
mononuclear cells even in mastocytosis)
- Chromosomal studies on mast cells were nevertheless
urged by one investigator in the room, including
use of interphase FISH using a variety of probes, starting with
c-kit. He said he strongly suspected that the c-kit mutation is
probably a secondary anomaly, and not the BCR/ABL
fusion of CML.
·
Question: Do we have any cytogenetic data on mast cell leukemia?
- Response: It's very rare - only one patient
in last five years. (Dr. Valent)
SMCD THERAPIES
·
Question: Would the c-kit mutation potentially be amenable to
Gleevec and other tyrosine kinase inhibitors?
- Response: No, STI571/Gleevec has been shown
to be ineffective in targeting the specific
c-kit mutation associated with myeloproliferative disorders. This
has been demonstrated both in in
vitro phosphorylation experiments and in clinical studies
in systemic mastocytosis patients. The c-kit mutation is on a
different domain of the molecule
than in GI stromal tumors.
·
Standard therapy in mastocytosis.
- No standard therapy or standard cytoreductive
therapy exists in mastocytosis.
- IFN-alpha 2B and prednisolone are the only
drugs with reported beneficial effect in
systemic mastocytosis of the aggressive type.
- Recommendations:
* We must establish
a standard therapy in mastocytosis, particularly in aggressive
mastocytosis.
* Urgent question: Should
we conduct a trial employing interferon-alpha +/- prednisolone
to establish this treatment as standard therapy?
·
Novel therapies in mastocytosis.
- The subject of anti-angiogenesis/anti-fibrosis
agents was broached, but then tabled
until the session on MF.
- How are mast cells isolated from monoclonals
in the absence of any specific markers
for mast cells?
* Presence of high-affinity
IgE receptor (not unique: basophils have this also)
* Presence of kit on
surface (not unique: other cells are kit +)
* Presence of tryptase
within mast cell granules (not specific either)
* CD34- (exclude the
CD34+ cells since these cells co-express c-kit)
* CD45+
* Neoplastic mast cells
can be isolated from normal mast cells by:
·
Applying antibodies against CD25+
·
Applying antibodies against CD2+
·
Developing response criteria for trials of anti-mastocytosis
agents is crucial.
- Clinical response criteria will be an important
component of these criteria.
- Drs. Akin and Valent are involved in developing
response criteria for studies of anti-mastocytosis
agents.
DIAGNOSTIC CRITERIA IN ESSENTIAL THROMBOCYTOSIS (ET):
TRUE VS. FALSE ET
Dr. Juergen
Thiele presented evidence from slide reviews showing superiority
of WHO's histologically strong diagnostic criteria over the histologically
weak PVSG criteria in distinguishing true ET from false ET (usually
early fibrosis/IMF).
·
Proportion of true ET patients among those diagnosed with
ET depends on which criteria are used :
- PVSG (weak histological emphasis)
- WHO (strong histological emphasis)
·
Review of slides of bone marrow pathology and clinical records
from retrospective, multi-center observational study
on Ph- chronic MPD patients with platelet count of at
least 600,000.
- Per updated PVSG: only 76.6% of patients diagnosed
with ET were true ET
- Per WHO criteria: 33.5 % of patients diagnosed
with ET were true ET patients
·
Megakaryocyte nuclear morphology is a hallmark WHO criterion
for distinguising true ET from false ET (usually IMF
with thrombocythemia) :
- False ET: Bulky, bulbous, atypical MK nuclei.
- True ET: Nicely lobulated MK nuclei.
·
Clustering is also a hallmark criterion:
- False ET: dense clustering
- True ET: loose clustering.
·
Review of 403 slides of patients diagnosed with ET in ongoing
British UKPT1 study.
- Per Dr. Thiele's blinded review: only 21%
true ET.
- Per pathologist re-review using histological
criteria: 19% true ET.
- Per second pathologist's re-review using histological
criteria: 20% true ET.
- Most of the patients were IMF.
·
Histology, esp. via CD61 staining, greatly facilitates correct
diagnosis of true ET.
·
Diagnosis-specific outcomes in true vs. false ET :
- It is very important to distinguish between
true and false ET when examining outcomes.
- Myelofibrosis as outcome:
* Most true ET patients
do not progress to MF.
* Most false ET patients
(pre-fibrotic IMF) progress to MF.
- Survival as outcome:
* PVSG vs.WHO criteria
predict a tremendous difference in survival.
* Per PVSG criteria:
·
16.5% loss of life expectancy in Ph- ET.
* Per WHO criteria:
·
Negligible loss of life expectancy in true ET.
·
Real loss of life expectancy in false ET.
·
Dr. Thiele's recommendations:
- For any trial, ongoing or retrospective, include
histopathology as one of the major points
of diagnosis. This means you must perform pre-treatment representative
bone marrow biopsies as part of
diagnosis.
- Use WHO diagnostic criteria, which emphasize
histopathology to arrive at a more
accurate classification of the myeloproliferative disorders.
- You can teach these differences (e.g., MK
nuclei morphology and clustering) to normal
pathologists and technicians interested in morphology on H&E
slides. It's very learnable. You
must, however, use some kind of fiber staining.
·
Dr. Bennett's recommendations:
- Three things are needed to distinguish true
ET from false ET pathologically:
* Very good, readable
bone marrow biopsy.
* Good visualization,
in order to recognize morphological differences of nuclei in megakaryocytes,
not cytoplasm (cytoplasm will be positive for markers anyway).
* The degree/presence/absence
of fibrosis as determined by reticulin fiber stain, as
well as amount of fat involved.
·
Recommendations from discussion of using histopathology in
diagnosing and prognosticating ET:
- There was general consensus to develop some
sort of publishable syllabus/manual
with pictures to describe to pathologists and technicians the
minimum criteria for performing
the above recommended histopathology (bone marrow
biopsy, megakaryocyte morphology, fiber staining, etc.).
* Dr. Heike Pahl mentioned
that German pathologists, headed by Dr. Thiele, are now
developing such a consensus pathology.
* Dr. Bennett said it
would be great to have that consensus translated for use in studies
in the U.S.
* Dr. Tefferi did caution
that we first need to reproduce Dr. Thiele's findings before
developing such a syllabus.
* Dr. Richard Silver
maintained that clinical hematologists have a duty to teach our
colleagues to use this biopsy technique. Dr. Bennett asserted
that the only
way to address the question of megakaryocyte morphology in diagnosing
ET would
be the establishment of a specific hematopathology committee.
* Committee would seek
to reproduce Dr. Thiele's results via a formal, prospective
study involving exchange of slides.
* This study would form
part of any clinical trial involving ET.
* Committee would include
clinicians.
- Dr. Silver urged that Dr. Thiele's method
should made part of the diagnosis of every
patient with an MPD. (Thiele's criteria)
·
Other discussion of using histopathology in diagnosing and prognosticating
ET:
- Widespread resistance and skepticism of examining
bone marrow biopsy in U.S. and Europe
for diagnosis of the MPDs.
* Histology in the U.S.
is still suboptimal, despite innumerable tutorials.
* Unfortunately, this
bone marrow biopsy technique is rarely and poorly done in the
U.S. (e.g., big Novardis Gleevec trial, in which <50% of sophisticated
investigators
did bone marrow biopsies to evaluate marrow cellularity).
* The vast majority
of American hematologists were trained using the PVSG criteria,
which explicitly said that the details of the bone marrow biopsy
were irrelevant.
Therefore, expect to encounter resistance when requesting this
level of
care from pathologists in examining bone marrow biopsies. It is
very difficult to
convince hematopathologists to reproduce this kind of work.
* In Europe, the importance
of bone marrow biopsy morphology was greeted with much
skepticism four years ago at a European MPD conference, until
a clinician
(Dr. Silver) vouched for the veracity of Dr. Thiele's findings.
* At conference in Italy
five years ago, however, pathologists agreed that one diagnostic
criterion for MF should be megakaryocyte morphology of bone marrow,
in particular the presence of clusters of MKs.
- Drs. Giovanni Barosi and Thiele discussed
the importance of establishing a histological
pattern based on several criteria.
* Dr. Barosi warned
that, of 300 patients in Italian MF registry, sensitivity of the
MK clusters
is no more than 60%.
* Dr. Thiele said yes,
there are many other criteria. Clustering is only one of 20 criteria.
He further pointed out that the density of the clustering is also
important,
with loose clustering of megakaryocytes characteristic of true
ET.
* Dr. Thiele did stress
that one of the most important hallmarks is the bulky nuclei
of the megakaryocytes in false ET.
- Retrospective vs. prospective studies
* A retrospective study
would require a different format of funding, i.e., submission
of a grant proposal.
* Dr. Bennett pointed
out that reproducing results regarding value of megakaryocyte
morphology must be done prospectively. Otherwise, we lose the
ability to know who is truly ET from onset.
* Others maintained
that prospective studies are a good idea, especially given the
multitude of other questions that need addressing.
* Dr. Tefferi expressed
concern that prospective studies would take too long, since
this disease takes 10-20 years to progress. He urged that there
is still value
in retrospective studies, that the bone marrows do not change.
- Dr. Silver urged that the diagnostic bone
marrow be examined.
DIAGNOSTIC
CRITERIA IN POLYCYTHEMIA VERA (PV): PRV1 AS MARKER
Dr. Heike
Pahl briefed participants on the progress of investigations by
the Germans into polycythemia rubra vera 1 (PRV1), a gene whose
protein travels on the granulocyte and whose mRNA is a potential
marker for PV.
·
The Germans have established a quantitative RT-PCR assay demonstrating
overexpression of the PRV1 mRNA in PV patients:
·
PRV1 is present in:
- All patients with PV diagnosed per the PVSG
or newer WHO criteria.
- Half of patients with a clinical diagnosis
of ET who form endogenous erythroid colonies.
- Half of patients with IMF who form endogenous
erythroid colonies.
- ICU patients with severe sepsis.
·
PRV1 is not present in:
- Patients who do not form endogenous erythroid
colonies. Due to this 100% correlation,
the quantitative RT-PCR assay renders colony-forming assays unnecessary
in patients meeting clinical criteria for diagnosis.
- Patients with secondary erythrocytosis or
reactive thrombocythemia.
- Patients with CML.
·
The German Society of Internal Medicine, in its new guidelines
published at end of March 2002, has recommended the
following:
- Patients with suspected myeloproliferative
disorder should be tested for their status
of PRV1 expression.
- A diagnosis of PV should only be given to
patients positive for the over expression
of the PRV1 marker.
·
Dr. Pahl invited American investigators to reproduce the German
results in future U.S. trials (Dr. Prchal has done
this).
·
Dr. Pahl stressed that PRV1 is a convenient surrogate marker
for diagnosis and useful epiphenomeonon, not the molecular
aberration leading to the disease.
·
Italian study finding that 100% of ET patients also overexpressed
PRV1 could not distinguish between positive and negative
because it did not use quantitative
RT-PCR.
·
Dr. H. Phillip Koeffler explained how the PRV1 marker was detected:
- Nanogram scale analysis was used, due to the
paucity of mRNA, on CD34+ cells
purified from patients with agnogenic myeloid metaplasia (AMM)
and counterpart cells purified from
normal patients. Confirmation analysis is being done
using real-time RT-PCR
- Cell populations in addition to CD34 are being
considered for gene chip analysis.
·
Dr. Pahl explained her group's theory, now being tested, on what
is happening to the PRV1 protein:
- Hypothesis: More PRV1 protein is formed in
patients over expressing the PRV1 mRNA;
but the protein is soluble and thus is not inserted in the membrane,
already full of protein.
- Theory is based on:
* Dr. Stroncek's work
showing that PRV1 and NB1 CD177 are in fact the same antigen.
* NB1 CD177 is likely
soluble, since it can cross the placenta (demonstrated by
formation of anti-NB1 antibodies by NB1- mothers of NB1+ children).
·
One expert cautioned that PRV1 is the same in PV progenitors
and normal progenitors, and after all the disease
starts in stem cells/progenitors.
DISEASE
CLASSIFICATION VIA MICROARRAY IN THE MPDS
·
Existing MPD assays are not specific enough.
·
Recommendations:
- Dr. Valent and Tefferi urged the development
of highly specific assays for diagnosing
the MPDs.
- Since most cells are maturing cells with no
obvious malignant markers, it would probably
be best to isolate cells in the bone marrow, not peripheral blood.
(Dr. Bennett)
- Dr. Tefferi proposed that a prospective study
should be done by the Cooperative Groups
using innovative microarray analysis to arrive at a more precise,
molecular classification of these
now crudely classified diseases.
- Dr. Valent agreed, stressing that a panel
of antibodies and genes/gene products will
likely be what is needed to define the type of diseases and various
prognostic subgroups precisely.
Precise classification will likely not be based on a single marker.
- Dr. Pahl agreed that such a national initiative
would be a good idea.
- Dr. Tefferi urged that, in order to obtain
sufficient mRNA for analysis, we must focus
on easily accessible tissue, such as the granulocytes, to gather
enough cells, and improve our nanogram
scale analysis ability.
- Dr. Valent questioned the specificity of using
the number of peripheral CD34+ cells
for diagnosing MF in Ph- chronic MPDs patients, since polyclonal
patients with MF also have a large
number of CD34.
- Dr. Bennett urged that distinguishing differences
in molecular abnormalities within
the MPDs is more important than contrasting normal vs. abnormal
(e.g., Ohio State study finding
huge differences in gene array patterns between trisomy 8
and non-trisomy 8 AML).
- Dr. Bennett suggested developing guidelines
to establish a separate Cooperative Group
focusing on either individual MPDs or selected MPDs.
* Group would develop
its own studies and export them through mechanisms that
now exist like the CTSU.
* Or, we could go directly
to one or more of Cooperative Groups to solicit advice.
·
Current, promising analyses using gene chip/microarray:
- Gene chip studies looking at 5,000-20,000
different genes are demonstrating major
differences between AML and MDS. They are using whole-cell lysates,
rather than just CD34. (Dr. Bennett)
- Oligonucleotide microarray analysis is being
applied to neutrophils of MPD patients.
(Dr. Tefferi)
- We have used granulocytes and cDNA chips of
12,000 cDNAs, complementing Dr.
Tefferi's work with oligonucleotides. (Dr. Pahl)
- Microarray analysis in lymphoma has shown
some promise. (Dr. Tefferi)
·
Dr. Koeffler addressed our ability to examine CD34+ cells
using nanogram scale technology:
- It is fair - not great.
- Nanogram scale technology is not as effective
as using the typical, regular scale.
- But nanogram scale technology it is better
than a lot of other techniques, such as
RDA, whose yields are low.
- Amount of mRNA is important.
* Typically, you label
7-10 mg of RNA to create and label the cDNA.
* We are currently using
500 ng, which requires two rounds of amplification.
* With our little amount
of mRNA, our hit rate is about 60%, not bad.
* With a large amount
of RNA, your hit rate is higher: Using the regular-scale, you'd
have hit rate of 80%.
- Use of real-time RT-PCR in nanogram scale
technique is important.
* Verifying your hits
using real-time RT-PCR is very important, since what amplified
well during the first round of amplification will amplify even
better during
the second round.
* After setting the
criteria of what to amplify, making it fairly robust, you perform
real-time
RT-PCR, confirming what you saw in chip.
·
Dr. Bennett reminded participants of material not yet covered
in etiology and pathology:
- Thrombopoietin
- PPO
- Vascular growth factors
PURPOSE
OF MEETING REVISITED
·
Question: Dr. Silver asked Dr. Cheson about the purpose of the
meeting, NCI's sudden interest in the MPDs and the
fiscal feasibility of the proposals heard this morning.
- Response: Dr. Cheson replied that the NCI's
position is that:
* Cooperative Group
leukemia chairs felt the myeloproliferative disorders were underserved
since termination of the PVSG.
* The objective of this
meeting is to arrive at a series of action items, the realistic
ones of which NCI will do its best to see implemented.
* Studies resulting
from this meeting may have to be performed in conjunction with
NHLBI.
ETIOLOGY
BEFORE THERAPY?
Dr. Bennett
asked if studies on the etiology of the MPDs were warranted, and,
if so, how should they be accomplished? Debate then ensued on
whether emphasis should be placed on elucidating etiology first,
or determining therapy.
·
Arguments against strongly focusing on etiology:
- Dr. Silver urged that epidemiology is as important
as etiology, stressing that the epidemiology
of the MPDs has never been looked into systematically.
- Drs. Gordon Dewald and Tefferi agreed that
we do not necessarily need to know the
exact etiology to attack the MPDs. What is important is to identify,
particularly with genomic studies,
pathways that can be altered with novel therapeutics,
perhaps small molecules.
- One clinician in attendance warned that if
we wait to find etiology of disorders, we are
not doing service to patients. Physicians and patients very frustrated
by absence of data on how to treat
MPD patients. There has never been a trial in MPD
focusing on how best to treat.
·
Arguments for focusing on etiology:
- Before any progress is made, we must agree
on a databank to be made available to
basic investigators. There is no point in investing money in therapeutic
trials when we do not know what
the disease is.
- The major focus should be on attempting to
define molecular defects. Besides mastocytosis,
we do not have any molecular targets for therapy development.
(Dr. Valent)
·
Argument resolved by comments from Drs. Bennett and Prchal:
- The beauty of Cooperative Group trials is
that they allow examination of epidemiology,
etiology, and molecular at the same time they seek to determine
which therapies are best.
- This simultaneous study of therapy and natural
history through Cooperative Group
trials is a point that should be emphasized with NCI.
·
Comment: We should get away from the term "orphan" disease.
The incidence of PV is actually greater than CML.
- Rebuttal: The term "orphan" is useful
for fast-tracking new agents through FDA bureaucracy.
Furthermore, the incidence or prevalence of any of the MPDs is
unknown, and can only be determined
through well-designed studies. (Dr. Bennett)
