- Finding a needle in a haystack: 1 | 2
What makes a biomarker viable
Many people have heard or read that a CA 125 blood test can check for early signs of ovarian cancer. Unfortunately, this potential biomarker fails to detect the cancer in many cases because only some ovarian tumors – 40 to 50 percent – secrete the CA 125 protein in the early stage of the disease. Part of the problem has to do with volume: A large tumor will secrete a lot of CA 125, and a small tumor, only a little. Because the goal is to find tumors when they are minute, this makes for a poor biomarker. Additionally, the test can read positive when no cancer exists because CA 125 isn't related only to ovarian cancer. Increases in CA 125 levels may be associated with other conditions such as fibroids (benign tumors of muscle and connective tissue that develop within, or are attached to, the uterine wall), pregnancy, pelvic infections, endometriosis (a condition in which uterine tissue grows in other areas of the pelvic region), liver disease, and even congestive heart failure. However, the test is useful to follow ovarian cancer patients after they are diagnosed. If CA 125 levels rise in a patient, it may indicate a recurrence of the cancer.
Four different types of ovarian cancer based on their microscopic appearance are shown below. Dr. Ronny Drapkin is studying potential biomarkers found in two of the major types.
To be a viable biomarker, a protein must be made by the tumor, circulate in the blood or urine, and be specific to the tumor type. For example, the biomarker of an ovarian tumor should not be excreted by breast cancer, normal tissue, or by diseases other than cancer. All three genes that Dr. Drapkin and his team are studying are part of the WAP (whey acidic protein) gene cluster, a group of genes located on the same chromosome and overly active in a large number of ovarian cell samples. Of the trio, HE4 protein is furthest along in studies.
Research shows the HE4 protein circulates in the blood. Moreover, excess HE4 is associated almost exclusively with ovarian tumors, as it was difficult to detect in a collection of non-ovarian tumors, including colon, breast, kidney, thyroid, and some lung cancers. It is also known to be made and secreted by two of the four major subtypes of ovarian cancer. Like breast cancer, there are multiple varieties of ovarian cancer. Distinguishing between the different types helps to design therapies that match the genetic make-up of the tumor. Four major subtypes of ovarian cancer are known: serous, endometrioid, clear cell, and mucinous. It is especially significant that the HE4 protein is found in serous and endometrioid cancers, since these cancers account for the vast majority of malignant ovarian tumors.
The other two potential markers, SLPI (secretory leukocyte protease inhibitor) and elafin, are undergoing similar testing. Thus far, Dr. Drapkin's team has found that both are made and secreted specifically by ovarian tumors. The team is developing tests to determine whether, like HE4, elafin and SLPI can be detected in the blood of cancer patients.
"It is unlikely that we will find one biomarker for ovarian cancer," explains Ross Berkowitz, MD, director of Gynecologic Oncology Services at Brigham and Women's Hospital and Dana-Farber. "To more thoroughly detect ovarian cancer at an early stage, multiple markers in a test panel will provide the best chance of catching the disease when it's still early enough to treat."
For example, combining CA 125 and HE4 in one test gives a 10 percent improvement in distinguishing ovarian cancer from benign disease. Dr. Berkowitz suggests a useful panel may test for several protein markers.
Assisting in this effort are Steven Skates, PhD, a biostatistician, and Patrick Sluss, PhD, a reproductive endocrinologist, both at Massachusetts General Hospital and part of the Dana-Farber/Harvard Cancer Center. They aid researchers who've identified protein candidates for biomarkers in developing serum tests that show whether or not the biomarkers circulate in the blood.
It will take another year to three years of research to find out if all three proteins are valid biomarkers and develop a suitable diagnostic tool, investigators say.
In addition to the biomarker studies, the team is looking at the relationship between these genes and ovarian cancer development. Interestingly, they've found that in laboratory studies, the overproduction of SLPI and HE4 enhances the growth of the cancer cells. Conversely, removing these proteins dramatically slows down their development. While elafin seems to share some of these characteristics, work in progress suggests that it may also function to regulate the ability of the immune system to respond to the tumor cells. These observations imply that the proteins are culprits in the formation and progression of the disease, rather than just innocent bystanders.
There are numerous questions to answer and a few years worth of studies still to come to determine whether these proteins are worthwhile therapeutic targets. If studies show they are, and preliminary tests are encouraging, discussions with drug companies to develop new agents could begin as early as 2008.
"No new therapies have been developed for ovarian cancer in a long time," notes Dr. Drapkin. "While this may be a case of limited resources, we are very excited about the potential impact these new discoveries may have on the disease. Gynecologic research is done by a small community; we are collaborating with our partners in hopes of changing the mortality rate of this disease."
- Page: 1 | 2
