NEW ORLEANSColorectal cancer appears to be two distinct
diseases at the molecular and natural history level, but what exactly
does this mean in terms of clinical practice? Steven Gallinger, MD,
of the Univesity of Toronto, attempted to answer that question in his
overview of Dr. Elsalehs presentation on microsatellite
There are distinct differences between tumors with chromosomal
instability and microsatellite instability, Dr. Gallinger said.
These distinct differences appear at the histologic level, at
the genetic level, at the anatomic level in terms of right side/left
side, and certainly theres now new evidence that there are
significant differences at the chemotherapy-response levels as well.
While genomic instability in cancer is a well-established field now,
he said, the specific mechanisms by which these instabilities
contribute to growth abnormalities are still not well understood. It
is known that in leukemias and lymphomas, major interstitial
deletions and translocations occur at the gross chromosomal level. In
various epithelial tumors such as colon cancer, the changes occur
differently, as major chromosomal arms have been shown to be lost.
Further, he said, we now know of a new genome instability
pathway called microsatellite instability whereby small insertions
and deletions and point mutations have been shown to be important in
various tumor systems.
In colorectal cancer, there are two molecular pathwaysthe
chromosomal instability pathway and the microsatellite instability
pathway. Thats the message of what were hearing
today, he said. Tumors seem to progress down one pathway
or the other but not both.
The mechanisms of this process however, are still not known, although
colorectal cancer serves as one of the best paradigms for
understanding the progressive accumulation of changes from the
adenoma to carcinoma sequence.
In 100 colorectal patients, about two will have germ-line mutations
in mismatch repair genes and tumors that demonstrate high-frequency
microsatellite instability, Dr. Gallinger said. About 85% of them
will have tumors that appear to progress down the traditional
chromosomal instability pathway.
Of increasing importance is the knowledge that there is a
fairly large and meaningful fraction of colon cancer patients who
seem to have developed disease in a sporadic fashion, but whole
tumors developed on the basis of the MSI pathway, Dr. Gallinger
explained. This fraction represents about 10% to 15% of all colon
Once this was recognized, he said, investigators began looking for
phenotypic differences between high-frequency MSI tumors (MSI-H) and
the traditional chromosomal instability tumors known as MSI-stable (MSS).
It would be naïve for us to consider that patients whose
tumors arose from two different pathways would have the same natural
history, and thats actually one of the points of todays
talk by Dr. Elsaleh and his group, he said.
Dr. Gallinger described two main questions addressed by Dr.
Elsalehs study and others: Do stage II and III colorectal
cancer patients with MSI-positive tumors require adjuvant
chemotherapy and radiotherapy? Do MSI-positive and MSS colorectal
cancers actually respond similarly to standard adjuvant chemotherapy?
The strengths of the study by Elsaleh et al, Dr. Gallinger pointed
out, are the large sample size, the fact that they all had Dukes
C colon cancer, and there was a long follow-up.
The limitations were that the patients were not randomized, and the
trial was not population based. While the latter is not of great
significance, the fact that the patients were not randomized begs the
question of how the oncologists chose the patients for adjuvant
therapy. So, the question of bias is raised.
The use of a single marker for detecting MSI could be another
limitation. According to the criteria set forth by an international
committee in 1997, MSI analyses require the use of five MS markers
two mononucleotide repeats and three dinucleotide repeats.
However, the researchers used the BAT26 marker, which has high
sensitivity (92%) and specificity (100%), so according to Dr.
Gallinger, it is probably adequate to define MSI. This actually could
be considered a new finding of Elsalehs study, he said.
Elsaleh et al showed that the MSI phenotype seems to be primarily
confined to the right colon. However, there are a few left-sided MSI
tumors, Dr. Gallinger said.
But, the most interesting, albeit controversial, aspect of the study,
Dr. Gallinger said, is the fact that it shows that the MSI phenotype
appears to predict an improved response to chemotherapy.
In fact, patients with the MSS tumors, the traditional
chromosomal instability tumors, did not seem to derive much survival
benefit from the use of 5-FU when compared with the MSI
phenotype, he said. According to Dr. Gallinger, this finding is
both supported and not supported in the literature.
While the findings appear strong, Dr. Gallinger feels it might be
premature to change clinical practice at this time. Probably
the most important research agenda is to examine ongoing and
prospective studies, he said. Im well aware that
some of the current trials are including MSI and some of the
molecular markers in their analysis. This will certainly provide a
lot of information in the coming years. If more studies
validate Dr. Elsalehs study, colon cancer treatment can be
further refined, he said.