http://www.clinicaloncology.com/index.a ... e_id=14881Genetic Variation in Metastatic Colorectal Tumors May Guide Therapy
Gabriel Miller
A small study has shown that primary colorectal tumors and corresponding lymph node metastases may have different molecular characteristics. The finding may help explain individually variable responses to standard colorectal chemotherapy agents, and further suggests that physicians may ultimately base adjuvant treatment decisions like chemotherapy on a patient’s individual metastatic tumor characteristics.
“Lymph node metastases display different molecular features compared to their primary tumors, and their characterization is important since chemotherapy agents, including 5-FU [fluorouracil]-based chemotherapy, target metastases, not the primary cancers,” said Craig A. Messick, MD, a general surgery resident involved with the study from the Cleveland Clinic’s Department of Colorectal Surgery, in Ohio. The study, which was led by Matthew Kalady, MD, a colorectal surgeon at the Cleveland Clinic, was awarded the Harry E. Bacon Foundation Award for Best Basic Science Podium Presentation at the 2009 meeting of the American Society of Colon and Rectal Surgeons (scientific abstract 34).
Previous research has established that primary tumors are frequently heterogeneous and arise from different genetic variations and oncogenic pathways. Correspondingly, some studies have indicated that genetically unique primary tumors have different responses to 5-FU-based chemotherapies.
But the understanding of genetic differences between colorectal cancer cell types largely ends at the primary tumor, experts say.
“Of all the genes required to create colon cancer, we know of less than a handful,” said Matthew Mutch, MD, associate professor of surgery at Washington University School of Medicine, in St. Louis. “And we know much more about the primary tumors than the metastases.”
One of the difficulties specific to colon cancer is that, unlike most breast or prostate cancers, for example, which develop through a single oncogenic pathway, colorectal cancers develop through at least three known molecular pathways. There is individual variation between patients, and even individual genetic variations within the same primary tumor.
Understanding how metastases differ from primary tumors may eventually allow physicians to treat them as separate diseases. “If the metastatic tumor is different from the primary, how do we specifically provide treatment for the metastatic tumor?” Dr. Mutch said. “The bottom line is that we can [eventually] identify those at highest risk and appropriately treat them; and in those that are at low risk, avoid toxicity of chemotherapy and minimize the overtreating and undertreating of patients.”
In the study, investigators at the Cleveland Clinic matched 47 stage III primary tumors and corresponding lymph node metastases drawn from a prospectively maintained, frozen tissue bank. From each set of matched tumors, DNA was tested for microsatellite instability (MSI) and CpG island methylator phenotype (CIMP).
Of the 47 matched local and distant tumors, eight (17%) showed genetic differences, either in MSI or CIMP status. Specifically, six (13%) of the primary tumors and seven (15%) of the lymph node metastases exhibited a high level of MSI (MSI-H; P=1.0), while 13 (28%) of the primary tumors and six (13%) of the corresponding lymph node metastases were positive for CIMP (P<0.02).
Significantly, seven (54%) of the 13 primary tumors that were CIMP-positive had lymph node metastases that were CIMP-negative (P<0.02).
The mechanism for these differences is unclear, but Dr. Messick said the literature suggests that metastases stem from a small, selected group of cells within the primary tumor that have been changed by the local environment. Some of these changes provide advantages that allow certain cells to escape and survive in distant sites.
Because the literature is still inconclusive about just how much MSI-H and CIMP status affect the efficacy of 5-FU-based treatment, these differences in local and distant tumor type do not immediately translate to changes in clinical practice. “The only conclusion that you can really draw, at this point, is that there are significant genetic differences between primary and metastatic tumors,” Dr. Mutch said. “The overall meaning of this—as far as [tumor] behavior, prognosis, response to chemotherapy—remains to be seen.”
But Dr. Messick said that studies like his may explain, at least in part, whether treatment with 5-FU is affected by genetic differences. “The debate in the literature exists on whether or not CIMP truly is an independent predictor of survival, and it may be that the disparity that we see may be due to changes that are in the lymph nodes. Perhaps with analysis of survival and chemotherapy data in those patients analyzed in the lymph nodes, a stronger correlation will be gleaned in those patients for CIMP.”
Although studies like these do not yet point to better outcomes, they set the stage for comparing treatments between subsets of patients with genetically unique tumors.
“The thing that’s different with colon cancer is that there are three different pathways, and each has distinct differences,” Dr. Mutch said. “Inherently we’ve known this, but defining it on a true genetic level, and saying these genes are responsible for this or that, has been very difficult. So what we need to be able to do, in a better fashion, is characterize tumors so that you can analyze like tumors with similar tumors, and say, ‘OK, I’m looking at a specific subset.’ ”