Introduction
Epigenetic factors are appearing to be more prevalent in our
understanding of the causes of many cancers. These factors include such
elements as methylation, long non-coding RNAs (lncRNA), micro RNAs and
acetylation. None of these reflect a fundamental change in the DNA of the
underlying genes, but they do reflect a complex process whereby the way the DNA
is processed and presented functions. Unlike translocations and gene changes
which are difficult to unravel, many of these epigenetic changes may be found
to be reversible in part or in whole. We focus on methylation and methylation
related disorders herein. The details are in a report we have just completed
.
The MDS Therapeutic Paradigm
MDS, the myelodysplastic syndrome, is a multifaceted disease
of the bone marrow cells which leads to the over-production of immature blood
cells; erythrocytes, lymphocytes, platelets and others. It is often indolent in
its early stages but then turns quite virulent and is often fatal, frequently
due to the development of AML, acute myelogenous leukemia. However, recent
understanding of a key driver of MDS, namely hypermethylation, has resulted in
complex therapies which may have proven not only efficacious but curative.
We use this disorder as an example of how methylation causes
potential cancers and further how it can be targeted and treated.
The therapeutic responses to MDS are representative to the
multi-prong attack on various cancers. The fact that MDS is not per se a cancer
but an artifact of a hypermethylation state, and that hypermethylation can be reversed,
as compared to a genetic change such as found in CML, the Philadelphia
chromosome translocation, and that we know how to deal with hypermethylation,
lends MDS to some form of initial treatment. However demethylation does not
always work.
Thus the second attack is more aggressive which is a
modified hematologic stem cell transplant.
That further reduces the aberrant cell load to an almost
miniscule amount. The final hit is using modified T cells called cytokine
induced killer cells specifically targeted for the remaining hypermethylated
cells.
This paradigm has been applied to other malignancies with
substantial success. The classic cases are the childhood leukemias and
Hodgkin’s lymphoma. One would suspect that MDS being substantially of the same
class would fit this paradigm. Our intent here is to examine the literature
across the above spectrum and attempt to make an assessment of progress in this
disease.
Methylation has been know of for decades but it has only
been in the last fifteen years or so that the connection between methylation
and cancers has been somewhat understood. In a 1997 paper by Jones and Gonzalgo
the authors state
:
DNA methylation is a mechanism for changing the base
sequence of DNA without altering its coding function. As a heritable, yet
reversible, epigenetic change, it has the potential of altering gene expression
and has profound developmental and genetic consequences. The methylation
reaction itself is mechanistically complex and involves the flipping of the
target cytosine out of the intact double helix, so that the transfer of the
methyl group from S-adenosylmethionine can occur in a cleft in the enzyme.
Cytosine methylation is inherently mutagenic, which
presumably has led to the 80% suppression of the CpG methyl acceptor site in
eukaryotic organisms, which methylate their genomes. It contributes strongly to
the generation of polymorphisms and germ-line mutations, and to transition
mutations that inactivate tumor-suppressor genes. Despite a 10- to 40-fold increases
in the rate of transitions.
This was somewhat of an opening salvo regarding methylation
and cancers. One should remember that this was almost five years before the
complete reading of human DNA and also at a time when actually reading the
methylated states was complex at best.
The authors hypothesized a mechanism for uncontrolled growth
using the methylation construct. They posited three ways in which methylation
functioned.
First, it caused a gene change. This was the C to T mutation
change.
Second they posited the promoter suppression via methylation
of the promoter. This method is seen
quite frequently in the process.
Third, there may be a chromosome instability resulting from
methylation.
At the same time Robertson and Jones wrote a paper on DNA
methylation and its affects and they also suggested a strong link between that
and cancer
.
They stated:
As with the demethylation and de novo methylation
observed during development, changes in methylation patterns during neoplasia
have been recognized for some time. Initially it was shown that malignant cells
have lower levels of methylation than do normal cells. This global
hypomethylation accompanies a hypermethylation of CpG islands, DNA regions
often associated with promoters of human genes that are normally protected from
methylation.
The above statement is a clear description of what we now
know to be correct; namely hypomethylation globally but hypermethylation of the
CpG islands. The hypomethylation allows expression of a wide variety of
proliferation genes while the CpG Island silencing via hypermethylation
deactivates control genes. They continue:
The mechanism by
which these regions remain unmethylated in the normal cell is not known, but it
may be mediated by the binding of certain transcription factors. In malignant
cells, these CpG-island regions become methylated and expression of the
associated gene is silenced. In the case of a tumor-suppressor gene, this may
result in a growth advantage for the cell.
DNA methylation– mediated transcriptional inhibition has
thus been proposed as a mechanism that is alternative to mutation and deletion,
in the removal of tumor suppressor– gene function. Examples of such genes
include the two cell-cycle regulators p16 Ink4a and p15 Ink4b, the von
Hippel–Lindau gene VHL in some renal carcinomas, the retinoblastoma gene
product Rb, BRCA1, the angiogenesis inhibitor thrombospondin, and the
metastasis-suppressor gene E-cadherin.
… Chuang et al. have shed new light on how methylation
patterns are maintained and how they may of the associated gene is silenced. In
the case of a tumor-suppressor gene, this may result in a growth advantage for
the cell. DNA methylation– mediated transcriptional inhibition has thus been
proposed as a mechanism that is alternative to mutation and deletion, in the
removal of tumor suppressor– gene function.
Examples of such genes include the two cell-cycle
regulators p16 Ink4a and p15 Ink4b, the von Hippel–Lindau gene VHL in some
renal carcinomas, the retinoblastoma gene product Rb, BRCA1, the angiogenesis
inhibitor thrombospondin, and the metastasis-suppressor gene E-cadherin (Graff
et al. 1995). In a recent issue of Science …has shed new light on how
methylation patterns are maintained and how they may become altered in cancer.
It was shown that the DNA methyltransferase is targeted to newly replicated DNA
by the replication associated protein PCNA (proliferating cell nuclear
antigen).
PCNA is the polymerase-processivity factor for the d and
e DNA
polymerases, is homologous to the E. coli b subunit,
and is required for DNA replication becomes altered in cancer. It was shown
that the DNA methyltransferase is targeted to newly replicated DNA by the
replication associated protein PCNA (proliferating cell nuclear antigen). PCNA
is the polymerase-processivity factor for the d and e DNA polymerases, is
homologous to the E. coli b subunit, and is required for DNA replication
There are slightly more than 10,000 new cases of MDS each
year. There may be a little difficulty in determine them because they can often
go un-noticed until they convert to AML at which point the diagnosis would be
clear. There may be a slight anemic, thrombocytopenia, and the presence of
blasts, immature hematopoietic cells. A true diagnosis requires a bone marrow
biopsy. The MDS patient may have one of many variants which we shall discuss
latter.
However what seems common is the presence of
hypermethylation resulting in the suppression of cell growth and proliferation
control genes on the lineage of hematopoietic cells first affected. Thus the
thrombocytes may be the initial ones affected and we see a drop in platelets
and a presence of blasts. But in all cases it is the hypermethylation. There is
as of yet in the process no genetic change, the excess immature growth is due
solely to hypermethylation. Thus the control is simply control of
hypermethylation via drugs which block the process. It is a somewhat simple
model for developing a therapeutic.
Thus why study MDS? The answers are:
1. MDS is not a full blown cancer. It lacks the genetic
breakdown.
2. MDS is a hypermethylation disease. Hypermethylation can
be reversed. Thus there is an opportunity to seek a “cure”.
3. MDS does lead to cancer, most likely AML. The process
that results in that change is worth of study as a means to seek both
prevention and cure.
4. MDS can be monitored both genetically as well as via
hypermethylation measurements.
In this report we examine several factors in depth.
Specifically:
MDS: We present an overview of MDS and its various forms.
This is a complex disease and it is almost as if no one patient is identical to
any other patient. We consider the cause of methylation at the DNA level but we
can at best speculate on the ultimate initiator. We know that many MDS patient
had pre-existing malignancies for which the received both chemotherapy and
radiation therapy. The nexus there seems to somewhat clear. However, many, if
not most, MDS patients have no clearly defined initiating event.
Methylation: We explore methylation and examine how it
occurs, and what it does to the functioning of the DNA expression. In many of
our cancer models we often just look at gene, RNA and protein flow. As we have
indicated before we often look at the epigenetic factors as noise. However it
has become clear that the epigenetic elements are integral parts of a cells
expression of its genetic capabilities and thus should be included in any
model.
Demethylating Therapies: We examine the various demethylating
therapies. The specifics are discussed in some detail as well as the efficacy
of the therapeutics.
Acetylation: The histones around which the DNA is wound also
exhibit acetylation. We examine this phenomenon and relate it to methylation.
Immunotherapy: We discuss immunotherapy focusing on the use
of CIKs, cytokine induced killer cells, primed T cells directed at the
remaining methylated hematopoietic cells.
We conclude with observations relevant to combined
therapies.
Now the histones may also be acetylated and drawn together.
When histones are drawn closer the genes in between cannot be read and thus
they are not expressed. Now we can summarize this as follows:
The third general step is the use of CIK, or cytokine
induced killer cells. These are somewhat akin to NK cells and have been
developed specifically for cancers of these type. We briefly discuss how they
are prepared. The efficacy is yet to be fully determined but there is a large
base of Phase I and II Trials demonstrating efficacy
Lin and Hui provide a definition for CIK cells
:
Cytokine-induced killer (CIK) cells are polyclonal T
effector cells generated when cultured under cytokine stimulation. CIK cells
exhibit potent, non-MHC-restricted cytolytic activities against susceptible
tumor cells of both autologous and allogeneic origins. Over the past 20 years,
CIK cells have evolved from experimental observations into early clinical
studies with encouraging preliminary efficacy towards susceptible autologous
and allogeneic tumor cells in both therapeutic and adjuvant settings. …
we anticipate that the continuous therapeutic application
of CIK cells will likely be developed along two major directions: overcoming
the challenge to organize large prospective randomized clinical trials to
define the roles of CIK cells in cancer immunotherapy and expanding its
spectrum of cytotoxicity towards resistant tumor cells through experimental
manipulations.
Jiang et al add to this description as follows
:
The number of immune cells, especially dendritic cells
and cytotoxic tumor infiltrating lymphocytes (TIL), particularly Th1 cells, CD8
T cells, and NK cells is associated with increased survival of cancer patients.
Such antitumor cellular immune responses can be greatly enhanced by adoptive
transfer of activated type 1 lymphocytes.
Recently, adoptive cell therapy based on infusion of ex
vivo expanded TILs has achieved substantial clinical success. Cytokine-induced
killer (CIK) cells are a heterogeneous population of effector CD8 T cells with
diverse TCR specificities, possessing non-MHC-restricted cytolytic activities
against tumor cells. Preclinical studies of CIK cells in murine tumor models
demonstrate significant antitumor effects against a number of hematopoietic and
solid tumors. Clinical studies have confirmed benefit and safety of CIK
cell-based therapy for patients with comparable malignancies.
Enhancing the potency and specificity of CIK therapy via
immunological and genetic engineering approaches and identifying robust
biomarkers of response will significantly improve this therapy.
The preparation and creation of CIK cells is done as
described by Jakel et al
:
CIK cells are generated by culturing peripheral blood
lymphocytes (PBL) with
1. interferon-γ (INF-γ) monoclonal
2. antibody against CD3 (anti-CD3) and
3. IL-2 in a particular time schedule.
The cytokines INF-γ and IL-2 are crucial for the
cytotoxicity of the cells and anti-CD3 provides mitogenic signals to T
cells for proliferation. Most of these CIK cells (87%) are positive for CD3 and
for one of the T-cell coreceptor molecules CD4 (37.4%) or CD8 (64.2%),
respectively.
IFN-γ, added at day 0, activates monocytes providing
crucial signals to T cells via interleukin-12 (IL-12) and CD58 (LFA-3) to
expand CD56+ cells.
After 14 days of culture, 37.7% of cells are
CD3+CD8+CD56+. These cells are referred to as natural killer T (NK-T) cells and
represent the cell type with the greatest cytotoxicity in the CIK cell
population.
Interestingly, these CD3+CD56+ double positive CD8+ T
cells do not derive from the rare CD3+CD56+ cells in the starting culture but
from proliferating CD3+CD8+CD56− T cells.
Their cytotoxicity is nonmajor histocompatibility complex
(MHC)-restricted and they are able to lyse a variety of solid and hematologic
tumors. Cell lysis is not mediated through FasL but through perforin release.
CIK cell cytotoxicity depends on NKG2D recognition and signaling.
Jiang et al propose the following:
Jiang et al prepare their cells as follows:
CIK cells have been evaluated as an adoptive cell
immunotherapy for cancer patients in a number of clinical trials.
Peripheral blood mononuclear cells (PBMC) were isolated
by apheresis.
T cells were then activated, expanded, and differentiated
by
1. anti-CD3 in the presence of cytokines including
2. IFN-γ,
3. IL-1α, and
4. IL-2
for 14 to 21 days to generate CIK, which were
subsequently infused into patients.
There are no significant clinical results for this in MDS
but there are many Trials underway. One could suppose that this is a
substantial third step after a BMT procedure. Logically it could be curative.
Observations
We now want to make some general and specific observations.
WE shall discuss each as a separate topic.
Epigenetics has become as significant a factor in cancer as
the pathway and immunological approaches. The impact of miRNA, lncRNA,
methylation, acetylation, and other epigenetic elements are now understood as
causative. However the drivers initiating many of these are not clearly
understood. The methylation in MDS for example is understood as a cause but
what leads to the methylation is still speculative. For example in melanoma one
could speculate that backscatter X-rays in full body airport scans provide just
the driver for methylation if it is applied at the right time. However that is
also speculative and no studies have been done. It is speculated that excess
radiation, excess CAT scans or radiation for cancers can cause the methylation
seen in MDS. Clinical proof is lacking however.
Methylation treatment with DNMT suppressors is known to
drive down the blast percentage. However it is a broad based therapeutic and
demethylates many other cell. This may also give rise to secondary neoplasia,
by activating proliferation genes in other cells in the body. It is not known
how significant this is. It might result in sequelae as is found in Hodgkin’s
lymphoma but the sequelae there are often found 20-30 years later. Thus since
MDS occurs at 70 years of age that well exceeds any life expectancy.
The problem with MDS is known to be hypermethylation. But
there are many cases of hypomethylation as well. One then wonders if the
approach taken herein applies to those cases as well.
Limited survival data is clinically available using the CIK
approach. Koreth et al present data based upon a Markov model but we have
considerable concerns about the approach
.
The results are shown below.
It should be noted that the top graph is for low to low
intermediate and the bottom graph is high intermediate to high, using IPSS
scoring. These Kaplan Meir curves show that for the high case we have a rapid
drop and then a slow decline with about 20% at 10 years. Since the average age
is about 70, the average life expectancy is 14 years and so 20% seem to have
reached average life expectancy. In contrast the opposite is the case for the
more indolent forms.
The problem that we see is the initial conditions. Perhaps
one would expect most patients have initial health conditions which would bias
them against a BMT survival. Perhaps other health conditions are also a
concern. The problem is that MDS is so complex and given the patients initial
health status conditions it is expected that any case is different and thus any
generalized result is problematic.
Posts
