|
Cover Story Sidebar |
|
|
Catching Cancer Before it Spreads
Georgia scientists pursue early-stage by JANE M. SANDERS | |
ON A QUEST to develop an early-stage diagnostic test for ovarian cancer is a “virtual” research institute of Georgia-based academic experts in molecular biology, biochemistry and bioinformatics.
|
Led by John McDonald, a former University of Georgia professor of genetics who became chairman of the Georgia Tech
School of Biology
in July 2004, the researchers are studying the same tissue samples from ovarian cancer patients and are comparing their results with each other and with patient data files.
Studying the same tissue samples – obtained from the non-profit, Atlanta-based Ovarian Cancer Institute (OCI) – gives the group a distinctive insight, says McDonald. He is the chief scientific officer for OCI, which was founded in 2001 by noted gynecologist Benedict Benigno of the Southeastern Gynecologic Oncology Group.
“We’re beginning to establish some correlations now,” McDonald notes. “We’re providing the biologists with high-quality tissue to analyze. They are comparing results, and we hope within a year to see some significant outcomes.”
Though much research worldwide is focused on the development of early-stage diagnostic test for ovarian cancer, nothing has come to fruition yet. Ovarian cancer causes almost no symptoms in the early stages and is often misdiagnosed when symptoms – such as weight gain and digestive problems – occur. When symptoms become severe, the cancer has usually progressed to Stage 3 and has spread to surrounding organs. The five-year survival rate at that point is only 25 percent. But if caught before it spreads, the survival rate jumps to 95 percent.
Researchers working under the auspices of OCI are funded by various organizations, including the Georgia Cancer Coalition and the National Institutes of Health. McDonald brings funding to ovarian cancer research on several fronts.
First, OCI researchers are exploring the possibility that every type of tissue contains adult stem cells that regenerate damaged tissue. Some evidence exists that these cells are present in ovarian tissue.
“It is possible that malignant cells may be derived from aberrant adult stem cells and not epithelial cells, which are observed when you examine tumors,” McDonald explains. “If this is true, it’s important because chemotherapy is often used to reduce the size of tumors – some as large as 25 pounds – before surgery. Chemotherapy kills epithelial cells and is effective, but tumors often recur.
This micrograph depicts adenocarcinoma, or ovarian cancer.
“One hypothesis we’re pursuing is that stem cells are not affected by chemotherapy and that they are regenerating more cancer cells after the chemo,” he adds. “So we want to isolate and characterize the stem cells and then determine their response to chemotherapy. Then perhaps we can design a therapy to inhibit the stem cells.”
In a related research effort, McDonald and his team are using molecular technologies, such as microarray analyses, to create molecular profiles of tumors. This information would assist pathologists as they identify ovarian cancer subtypes and assign a stage to a tumor.
“We could remove some of the subjectivity in the pathologist’s decision-making if we determine there are distinguishing molecular profiles between the thresholds of cancer subtypes and stages,” McDonald explains. Researchers have been able to provide such profiles of leukemia and discovered that one classical type was actually two types – one that fully responded to chemotherapy and one that did not respond at all.
“We hope to find a molecular profile of ovarian cancers and identify subtypes,” McDonald adds. “Different subtypes may respond differently to chemotherapy. So it’s a more refined approach. We already have good preliminary data on this.”
Though researchers still need to analyze many more samples, their results to date correspond with the subtype and stage that pathologists identified, McDonald says.
In another interesting finding to be published later this year, molecular profiles of tissue samples from patients treated with chemotherapy prior to surgery either look like the cancer was not treated or they appear to be benign. “One possible interpretation is that the molecular profiles are telling us whether the tumor is responding to chemotherapy,” McDonald explains. “If there’s a positive response to chemo, the tissue looks normal.”
The next research question – which will require years of following patient outcomes – is whether these patterns are accurate predictors of a patient’s prognosis. “If so, then this is a tremendous advance,” McDonald says. “At surgery, a doctor could run a molecular profile to determine whether the pathology suggests the cancer may recur and would require regular follow-up.”
Before he began research on ovarian cancer, McDonald studied retroviruses, stealthy disease-causing agents responsible for many illnesses, including some types of ovarian cancer and other types of cancer. He has filed a patent on a potential retroviral marker for early stages of cancer.
Cancer commonly stems from changes in DNA sequencing that lead to genetic mutations. But the disease can also be attributed to heritable changes in the way DNA is packaged, McDonald says. If proteins are packaged too tightly around a gene, the gene becomes “silent,” as if it is mutated and cannot function. McDonald and his collaborators believe this mechanism of silencing DNA evolved to silence retroviral elements in the human genome. Retroviral-like elements are believed to have been abundant in the primitive genome but more elements have likely been added over evolutionary time.
Of the more than 90 percent of the human genome that does not encode proteins, much of it comprised of retroviral-like sequences. These are potential sources of gene mutation because they can move around and insert themselves in the coding or regulatory regions of genes, effectively disrupting gene function, McDonald explains.
“So the genome developed mechanisms to condense the proteins surrounding DNA (sometimes referred to as epigenetic controls) that recognize and clamp down on these viral-like sequences so they become silent,” McDonald says. “These epigenetic controls evolved initially as a defense against retroviral-like elements…. But over time, these regulatory mechanisms have been co-opted for essential cellular functions, so if you now knock them out, it produces lethality.”
McDonald has discovered that changes in the expression of these viral-like elements reflect changes in epigenetic controls. Specifically, he found changes in the structure of chromatin – a mix of nucleic acids and proteins – in the early stages of cancer. Thus, it appears the activity of viral-like elements could be markers for cancer.
courtesy Drs. Jo Ann Benda & Richard Zaino, Univ. of Iowa
For more information, contact John McDonald at 404-894-3735 or
john.mcdonald@biology.gatech.edu.
Contents
Research Horizons
GT Research News
GTRI
Georgia Tech
Send questions and comments regarding these pages to
webadmin@edi.gatech.edu
Last updated: July 7, 2004