Surveying telltale signs of ovarian cancer

Surveying telltale signs of ovarian cancer

March 12, 2015

March 12, 2015

This year, ovarian cancer will claim over 125,000 lives worldwide. The deadly disease remains the fifth leading cause of cancer-related mortality in U.S. women, killing about 15,000 per year. 

While diagnostic screening has long been a watchword, efforts to identify the disease in its early stages often fail. By the time ovarian cancer is detected, it has typically progressed to an advanced phase, where the 5-year survival rate falls below 30 percent. 

Josh LaBaer, MD, PhD., and Karen Anderson MD, PhD., researchers at Arizona State University’s Biodesign Institute, hope to alter these grim statistics. Their work involves the development of biomarkers—blood-borne signals that can reveal the presence of diseases like cancer before a patient displays any outward symptoms. 

In a new study, they investigate novel biomarkers for ovarian cancer by means of high-density microarray technology. LaBaer, director of Biodesign’s Virginia G. Piper Center for Personalized Diagnostics, helped pioneer a powerful advance in this technology, known as NAPPA. The technique was used in successive screening stages to ferret out candidate biomarkers (known as autoantibodies) for ovarian cancer and explore their diagnostic power. 

Three autoantibodies were identified as promising biomarker candidates in the new study. The team’s research findings, which represent the first demonstration of NAPPA technology for the detection of autoantibody biomarkers in ovarian cancer, recently appeared in the Journal of Proteome Research. 

As their name suggests, autoantibodies are antibodies produced by the human immune system targeting one or more of an individual’s own proteins. While autoantibodies are themselves culprits in a broad array of autoimmune diseases including lupus erythematosus, rheumatoid arthritis and multiple sclerosis, they can also act as a surveillance and early warning system for aberrant proteins produced by various cancers. Capturing the activity of cancer-specific autoantibodies therefore holds the promise of significant improvements for diagnosis and effective early-stage treatment. 

“Early detection of ovarian cancer is critical for survival,” Anderson says.  “Right now, most ovarian cancers are caught in late stages, after it has spread in the abdomen.  The immune response to the cancer, as measured by the autoantibodies, may be detected even before clinical diagnosis.”

Autoantibodies are produced by the immune system’s B cells. In some cases, these antibodies arise randomly and in others, in response to the presence of a foreign protein antigen, to which the autoantibody binds. Past research has identified two key antigens implicated in ovarian cancer. One is known as CA 125—a protein found on the surface of many ovarian cancer cells. Another is known as HE4. 

As the authors note, several factors have prevented the use of CA 125 and HE4 as diagnostic biomarkers for the general population. The primary drawbacks have to do with low specificity and/or low sensitivity. 

If a biomarker has low sensitivity, it often fails to detect the presence of the cancer-related protein it is designed to detect, leading to a high false negative rate. A biomarker with low specificity often mistakenly detects non-cancer related proteins in addition to cancer related proteins, leading to a high false positive rate. 

Ovarian cancer is rare in the general population.  The serious consequences of false positives to non-cancerous patients require a prospective biomarker to have very high specificity and sensitivity. The use of several biomarkers in tandem may improve the overall accuracy of diagnostic testing. 

The use of CA 125 as an ovarian cancer biomarker is hampered by low sensitivity (50 percent) for early stage detection at suitably high specificity rates. The need for new, more reliable means of early detection is essential for successfully identifying and treating ovarian cancer patients. 

Autoantibodies as disease biomarkers offer a promising avenue for diagnostic exploration. Often, they are produced in cancer patients as a result of protein overexpression or mutation. Their warning signals are amplified by the immune system, making them easier to detect than some rival biomarkers and they persist in the body even after the disease antigen is no longer detectable. 

But sifting through a body’s storehouse of autoantibodies to identify candidate biomarkers has been notoriously tricky and the hunt for ovarian cancer markers reliable enough for clinical use has, to date, been disappointing.  In the current study, a technique known as NAPPA, (which stands for Nucleic Acid Protein Programmable Array) was used to aid in the search. 

NAPPA is a novel type of protein microarray. The basic idea is to spot a testing slide with thousands of different circular pieces of DNA known as plasmids, which code for various proteins. Once expressed on the microarray, the protein antigens then act like fishing lures for specific autoantibodies. When a small sample of patient blood is exposed to the microarray, autoantibodies in blood serum bind with proteins present on the array, allowing them to be identified. 

For the current experiments, protein arrays were prepared displaying 5177 candidate antigens. Sequential rounds of screening were used in order to limit the false discovery rate. 

In the first screening round, blood from 34 patients with ovarian cancer and 30 healthy controls were tested against the full library of 5177 candidate tumor antigens present on the high-density NAPPA array, (with each slide holding up to 2300 antigens). Results from this initial screening episode were used to select 741 promising antigens for further testing. Round two screening used 60 ovarian cancer cases and 60 healthy controls to test the 741 antigen candidates. This step allowed the researchers to narrow the field to 12 candidate autoantibody biomarkers. 

Finally, the third round of screening exposed the candidate autoantigens to an independent assay known as a bead array. The specificity of the top 7 antigens was evaluated using blood from healthy controls and ovarian cancer cases with low levels of CA 125. (These cases were chosen to determine if the new candidates showed potential as biomarkers beyond CA 125 for accurate cancer detection.)

Of the 12 biomarkers showing encouraging results, three in particular were consistently selective for ovarian cancer, suggesting their possible clinical use to augment CA 125, HE4 or other biomarkers and improve diagnostic specificity and sensitivity. Further, a number of the 12 candidate antigen biomarkers have been implicated in previous research with the development and progression of ovarian cancer tumors. 

These biomarkers are currently undergoing validation in national studies funded by the National Cancer Institute and the Early Detection Research Network.


Written by: Richard Harth
Science Writer: Biodesign Institute