Accuracy Matters When It Comes to Diagnosing Cancer

Breast cancer has been a part of Katie King Mahan’s family. Her mother, grandmother and great-grandmother survived breast cancer—each of them having dealt with the disease at a relatively young age.

Due to her family’s history of breast cancer, Mahan, a stay-at-home mom and yoga instructor in Marblehead, Mass., was considered at high-risk of developing the disease. “My mom was 29 was she was diagnosed,” Mahan says. “As my 30th birthday approached, I recognized that I would do whatever it took not to be next.”

For Mahan, that meant getting tested for genetic mutations in the BRCA1 and BRCA2 genes. Mutations in either or both genes are known to put women at significantly higher risk of developing hereditary breast and ovarian cancer than women who do not have the mutation. Ideally, genetic testing should start with a family member who has breast or ovarian cancer since unaffected blood relatives may not share the same gene, but that is not always possible.

In fact, as many as 45 to 65 percent of women with BRCA mutations will develop breast cancer, compared with 12 percent of women who do not have the mutations. Similarly, 11 to 39 percent of women who inherit either or both of the BRCA mutations will develop ovarian cancer, compared with 1.4 percent of women without the mutations.

After testing positively for the mutation at age 32, Mahan decided to have both breasts surgically removed.

“For me, it was an obvious choice,” Mahan says of the procedure, which is also called a bilateral prophylactic mastectomy. “I have a husband and child who need me.”

In electing to have the surgery, Mahan reduced her risk of developing breast cancer from as high as 87 percent to less than 1 percent. She also plans to have her ovaries removed within the next five years—a procedure that will reduce her risk of developing ovarian cancer before the age of 70 by as much as 90 percent.

For me, it was an obvious choice. I have a husband and child who need me.

Mahan’s experience demonstrates the potential of using biomarkers to predict cancer risk. A biomarker is a characteristic—in Mahan’s case, a protein that indicates a mutated BRCA2 gene—that may signal, among other things, the presence of disease.

As new molecular technologies have been developed that allow for the examination of thousands of genes and other molecular characteristics, patterns of gene activity, changes to DNA and chromosomal abnormalities have also begun to be used as biomarkers. Many of these biomarkers are used, much like the BRCA genes, for risk stratification—that is, to determine a person’s risk for developing cancer. In fact, risk-stratification biomarkers are now being used for brain, cervical, colorectal, esophageal, liver and pancreatic cancers. Some biomarkers are inherited (particularly those associated with a high risk of developing certain cancers—these can be used to determine the benefits of prophylactic or risk-reducing treatments), while other markers are acquired (only seen in the tumor and not passed on to children—these can be used to estimate the risk of recurrence or to help choose the best therapy).

Based on the success of biomarker testing in determining cancer risk, researchers are also hoping that biomarkers can be used to help screen people for cancer at its earliest, most treatable stages. Unfortunately, says Yang Hubert Yin, an associate professor of chemistry and biochemistry at the BioFrontiers Institute at the University of Colorado—Boulder, “no biomarker today is both sensitive and specific enough to screen for cancer with 100 percent accuracy.”

In fact, Yin says both false-positive and false-negative results are “major shortcomings” of today’s biomarkers. A false-positive result occurs when a person tests positive for cancer even though there is no cancer. A false-negative result occurs when a person tests negative for cancer when in fact cancer is present.

Though testing for CA-125, a protein found on the surface of ovarian cancer cells, has shown potential in detecting ovarian cancer, CA-125 tests are known to deliver both false-positive and falsenegative results. Elevated levels of CA-125 can be caused by many other problems or even normal biologic processes, such as pregnancy. On the other hand, many women with ovarian cancer have normal levels of CA-125.

For these reasons, the U.S. Preventive Services Task Force (USPSTF) does not recommend CA-125 testing in women with an average risk of ovarian cancer. However, CA-125 is still sometimes used to test and monitor women who have a family history of ovarian cancer or women with BRCA mutations who are at higher risk of developing the disease.

The prostate specific antigen (PSA) is a protein made by the prostate gland that may signal prostate cancer. Until recently, most doctors encouraged yearly PSA screening for all men beginning at age 50, or between the ages of 40 and 45 for men at higher risk of the disease. A PSA test measures the level of PSA in the blood.

However, the usefulness of PSA-testing as a prostate cancer screening tool remains a matter of debate because most men who have an increased PSA level do not actually have prostate cancer, or may have a slow growing or indolent cancer that will not hasten death. False-positive results often lead to additional medical procedures, such as a prostate biopsy, that can cause potential complications and harmful side effects.

In 2012, the USPSTF issued a recommendation against routine PSA-testing. Other professional organizations have also adopted the policy but others have not. The American Cancer Society, for example, recommends that men make an informed decision about whether to be tested after learning about the both risks and benefits of PSA-testing.

Being able to more accurately find aggressive cancers and reduce the number of unnecessary biopsies, will help make it possible to retain the benefits of prostate cancer screening while reducing possible harms of over-detection and overtreatment.

Soon, a more accurate alternative may also be available. Known as the Prostate Health Index (PHI), the new blood test is more than 2.5 times better at correctly ruling out cancer in patients who do not have it than existing PSA tests. The PHI can also help distinguish aggressive tumors from slow-growing ones.

“Being able to more accurately find aggressive cancers and reduce the number of unnecessary biopsies will help make it possible to retain the benefits of prostate cancer screening while reducing possible harms of over-detection and over-treatment,” says Martin Sanda, chief of urology at Emory University in Atlanta.

The test received Food and Drug Administration approval in 2012. It should be noted, however, that a cancer diagnosis in and of itself often leads patients to seek treatment that could actually do more harm than good. For example, prostate surgery can result in side effects, such as erectile dysfunction and urinary incontinence, in patients who may have otherwise lived a full, productive life and die of something other than cancer. This calls for additional biomarkers to distinguish truly dangerous cancers from those that could be observed or treated much less aggressively.

[What Makes a Screening Test Useful?]

In some cancers, tests that reveal the presence of certain biomarkers can help confirm a cancer diagnosis. When used in combination with other tests, such as biopsies and imaging scans, biomarkers can be especially useful in distinguishing one type of cancer from another. For example, doctors increasingly rely on biomarkers to classify lymphomas and leukemias, as well as soft tissue sarcomas. This is important because the cancers have vastly different prognoses and require completely different treatment approaches.

Biomarkers can also help doctors understand whether a tumor is the first site of cancer or whether cancer has spread (metastasized). To make this distinction, researchers test cancer cells obtained through a tissue biopsy for chromosomal alterations; then, they compare them with cancer cells from other areas of the body. If the alterations match, it means the cancer has spread from its original area; whereas if the alterations differ, the cancer cells may be from more than one type of cancer.

Marty Pagel, an associate professor of medical imaging at the University of Arizona Cancer Center in Tucson, says imaging biomarkers—detected through X-rays, magnetic resonance images, computed tomography (CT) and positron emission tomography (PET) scans, for example—are a “fantastic tool” for diagnosing cancer.

Pagel says PET/CT scans are particularly effective in detecting large or aggressive forms of cancer and “determining where cancer is hiding.” They may also be able to provide an early sign of response to treatment.

Before a PET/CT scan, a person receives an injection of sugar and a small amount of radioactively labeled sugar. Cancer cells absorb sugar differently than other tissues in the body, causing the scans to “light up” where cancer is present, he explains.

Circulating tumor cells (CTCs), another cancer biomarker, can also help diagnose cancer. CTCs are cells that are shed by the tumor and circulate in the blood. The number of CTCs detected are strongly linked to prognosis, and research studies are ongoing to obtain more detailed genetic and protein information from CTCs so that they could serve as a “liquid biopsy” in the future.