Latest News

Researchers Pursue Early Detection of Ovarian Cancer, Foresee Possibility of At-Home Test

By Paul Nicolaus

April 13, 2021 | A new study led by University of Minnesota Medical School researchers suggests that Pap test samples and cervical swabs may be a useful source of tumor-specific biomarkers for ovarian cancer, and the group sees potential for the development of an early detection test. 

Dubbed “the silent killer,” women with early stages of ovarian cancer may experience no symptoms at all or mild symptoms that can be easily confused with common ailments, like bloating, until the disease reaches an advanced stage. While not common, ovarian cancer is responsible for more deaths than other female reproductive cancers.

Although early detection increases the odds of survival, a screening tool sensitive and specific enough to use within the general population is lacking, according to Amy Skubitz, director of the Ovarian Cancer Early Detection Program (OCEDP) at the University of Minnesota Medical School.

She and colleagues are on a mission to develop a clinical test capable of screening women for early markers of the disease and detecting it in its earliest stages, when the survival rate is higher and treatment options include less toxic chemotherapy and less extensive surgery. 

In pursuit of this long-term aspiration, the researchers set out to pinpoint proteins present in Pap test samples and cervical swabs to determine whether the same proteins exist in ovarian cancer tumor tissues. They hypothesized that proteins shed by ovarian cancer cells could be picked up by mass spectrometry-based proteomics during routine Pap tests that are normally used to detect cervical cancer.

In a paper published in Clinical Proteomics (doi: 10.1186/s12014-020-09309-3), they revealed that the Pap test and swab samples contained proteins that were also found in the tumor of a woman with high-grade serous ovarian cancer, and over 2,000 proteins were detected in all three of the sample types collected.  

The research is an early effort to explore the feasibility of using a Pap test or cervical swab to detect biomarkers. “This really is the first step in showing that it may be possible,” Skubitz said, “because the proteins that we identified in a routine Pap test were present in the tumor as well.” More importantly, she added, the cervical swab also contained over 2,000 of the same proteins.  

“Our hope is, looking ahead, that we could actually say go ahead and do your cervical swab yourself at home and send that into the lab, and we’d be able to pick up those proteins that are diagnostic for ovarian cancer.” 

Ron Drapkin, director of the Penn Ovarian Cancer Research Center at the University of Pennsylvania (who was not involved in the research), called the paper “a nice proof of principle study,” but said the biggest limitation is that the sample examined is from just one individual. 

The goal, however, would be to perform this research more thoroughly and rigorously with a much larger cohort of women, he added. 

Moving Closer to the Site of Origin 

The basic premise of this approach, which is not necessarily unique to this study, is the notion that tissue proximal sampling could be a useful way of identifying biomarkers for early detection of ovarian cancer, Drapkin told Diagnostics World.

It is now believed that the majority of high-grade serous carcinomas, which is the most common subtype of ovarian cancer, arise from the very distal end of the fallopian tube—the one closest to the ovary, he said. Most of these tumors start there, form precursors, and eventually reach the ovary. Unfortunately, this is often the point at which they are detected—at an advanced stage, when a surgical cure is no longer possible. 

“So the holy grail of the whole field has been, for decades now, finding a biomarker for early detection,” he explained. “And it’s been challenging.” 

There are currently two FDA-approved biomarkers, CA125 and HE4, but these are generally used for monitoring patients who are already known to have cancer. “The limitation with those markers, especially with CA125, which has been around the longest and is the most commonly used, is that a lot of benign conditions also trigger its expression,” he said.

“It is believed that HE4 adds a certain degree of specificity to that, and so the combination of those two is thought to be better at distinguishing those malignant from benign masses that can develop in the pelvic cavity of some women,” he added. “But we still don’t have one biomarker that is actually sensitive and specific enough to detect ovarian cancer early.” 

That is, at least in part, because it’s a relatively rare disease. It is far less common than breast cancer, for example, but it is also more lethal. And because it is less common, the specificity needs to be quite high to come up with a test that could be used in the general population. 

“The thought has been that if we look for biomarkers in the blood, which may harbor some proteins that are uniquely made by the tumor, that may work,” he said. “But the problem is you’re far removed from the site of the tumor itself.”

And that’s where the concept of tissue proximal sampling comes into play.

If there is a tumor in the fallopian tube, it is possible that some of the cells and proteins secreted by the tumor make their way down through the fallopian tube into the uterus and the cervix. When conducting a Pap smear to look for cervical cancer, some of the proteins in that preparation may reflect abnormalities in the upper reproductive tract. 

“If you can get closer to the site of origin,” he added, “maybe we can find biomarkers that reflect disease versus healthy state.” 

Other Early Detection Efforts 

According to Drapkin, there are two basic types of tissue proximal approaches for finding early detection biomarkers: genomic and proteomic. 

Several groups, including one at Johns Hopkins University (doi: 10.1126/scitranslmed.aap8793), have used pap smear samples to see if they can identify DNA mutations representing ovarian cancers.

The identification of tumor-specific proteins has also been reported. Drapkin pointed to a study (doi: 10.1074/mcp.RA119.001362) that comes from one of his former trainees and involves uterine lavages and the search for proteomic changes associated with ovarian cancer. 

Meanwhile, others have developed a method for sampling the fallopian tube itself, which has been difficult because it is such a narrow tube. “If you can cannulate the fallopian tube, not only can you look and see if there’s something there, but you can also sample cells or possibly even fluid,” Drapkin explained. 

A company called nVision Medical Corporation, which was bought out by Boston Scientific several years ago, developed a device to sample cells from the fallopian tube.

So there are efforts to get not just close to the site of origin, but actually to it. “And I think bioengineering and similar efforts are really ultimately going to get us to that point,” he added, which will be a significant advance in the ability to look at the site of origin in the search for abnormalities. 

One of the ideas brought up in the recently published Clinical Proteomics paper is that the combination of tumor DNA profiles and proteomic profiles could be a more powerful approach than focusing on just one or the other. “That’s something that will be interesting to see as people take it further down the road,” he added. 

Periodic Surveillance Potential 

“I think the promise of a study like this—but I would say it’s not unique to this study—would be that it may offer women an opportunity to do surveillance on a periodic basis,” Drapkin said.  

Because it’s not a very common disease, this would probably boil down to women who have hereditary mutations in the BRCA1 and BRCA2 genes and are at a significantly higher risk of developing ovarian cancer during their lifetime. “In that population of patients, it would be really great to be able to do surveillance and see whether there’s evidence of early worrisome changes or tumor-related changes in their fallopian tubes,” he said. 

Currently, what’s available is essentially prophylactic surgery, he added. The problem, however, is when a pre-menopausal woman is put into surgical menopause, that comes along with the associated morbidity related to cardiovascular disease, osteoporosis, and other physiological manifestations of menopause at a younger age—including infertility. 

“We really want to offer women better options,” he said. 

If it’s true that these tumors are coming from the fallopian tube—and most agree that the vast majority of them are—then one approach is to perform prophylactic surgery by removing the fallopian tube. “If that’s the site of origin, that should give you equal protection,” he said. “But you leave the ovaries in, so you don’t put these women into surgical menopause.” 

There are several international studies currently looking at that in the high-risk population, he pointed out. “I think in a number of years, we’ll know whether just taking the tube out alone is sufficient to confer reduced risk in this population of women.” 

These are the potential positive implications of studies that get closer to the site of origin and may someday enable surveillance and early detection in ways that haven’t been possible when the focus has centered on blood-based biomarkers. 

Although future studies are needed, Drapkin said the possibility of using a test that women are already used to, that isn’t terribly invasive, and that gets conducted routinely in the office setting offers promise. Women are used to doing this, and the approach makes use of the same material—without doing anything different or extra—to look for a different cancer. 

The infrastructure for acquiring the samples and searching for this different form of disease, without adding additional headaches or costs to the patient, is already in place. “I think that’s the power of this study,” he said of the Clinical Proteomics paper.

Next Step: Targeting a Shorter List of Proteins 

Skubitz and colleagues at the University of Minnesota Medical School plan to build upon their work by making use of a quantitative mass spectrometry technique called TMT, or tandem mass tag, where it is possible to handle 10 samples (rather than one) at a time. 

“The beauty of that is you can do, say, five ovarian cancer samples,” she said, “and then three from benign ovarian disease, and then another two normal Pap test samples.” 

The protein can be digested into peptides and labeled with different mass tags. From there, they can be combined into one. It is then possible to run high-performance liquid chromatography (HPLC) to fractionate them before running mass spectrometry.  

“You can see the differential expression between the cancer samples and the normal samples and the benign samples,” she added. “That’s where we come up with profiles that would indicate which ones are going to be ovarian cancer-specific versus in normal or benign cases.” 

Skubitz and her team now have a limited number of proteins that they are going after with another mass spec technique called selected reaction monitoring (SRM). 

Once they determine which peptides they are interested in, those can be synthesized, she said, “and you can spike them into your samples.” It is possible to analyze hundreds of samples, and it can have a mixture of many of these peptides found to be specific.

“So that’s kind of where we’re at right now,” she said. Instead of all 2,000 proteins in this paper, they have whittled it down to a targeted list of 30 and plan to see if they can develop an algorithm that would indicate which of those proteins are most important.

Validation Challenges and Other Hurdles

One of the biggest hurdles remaining, Skubitz said, is obtaining samples to validate their findings.

Whereas Pap tests used to be conducted annually, guidelines changed when HPV testing became more prevalent. Pap tests are currently recommended every three to five years, and few cytopathology labs or research labs actually collect these tests.

After a liquid-based Pap test sample is gathered, it is sent off to a cytopathology lab. The cells are put onto slides and analyzed for abnormalities. They are also used to isolate DNA for HPV testing. After a diagnosis is made, though, the excess fluid and vials are thrown out. (Some hospitals save vials temporarily in case they need to validate an irregular result, but typically they are discarded.)

The University of Minnesota cytopathology lab handles diagnosis on roughly 60,000 Pap tests every year, she estimated, but they all get discarded. And few come from women with ovarian cancer.

When Skubitz and colleagues first began their studies, they turned to the university’s cytopathology lab and experimented with “normal healthy” samples that were going to be discarded so that they could optimize a protocol to isolate the proteins and use mass spectrometry to identify them. 

Because ovarian cancer is so rare, they would need to gather thousands of samples to have enough to perform experiments using samples from women with ovarian cancer. To increase their odds, they came up with a protocol to obtain “Mock” Pap tests from women in the Gyn/Onc clinic suspected of having ovarian cancer.

Now that they are in the process of validating their discovered biomarkers, the researchers will need thousands of Pap tests to determine whether their biomarker panel will be able to detect those Pap tests that came from the ovarian cancer patients.

The factor that makes this project so difficult is that they would need to ensure that some or many of those thousands of Pap tests come from women with ovarian cancer. “Otherwise,” she explained, “it will be impossible to validate our biomarker panel.”

For research purposes, they have been gathering cervical swabs at the same time as the “Mock” Pap tests, so they have accumulated swabs from over 600 women with a wide range of gynecological conditions, including ovarian cancer.

But getting others to use a cervical swab every time they perform a Pap test will be a major challenge, Skubitz said.

Some researchers have collected swabs for their studies of the cervical microbiome, and others for HPV testing. But most research labs that focus on ovarian cancer or other gynecologic conditions tend to work with other biospecimens, such as blood, urine, or tumor tissue. And to her knowledge, there isn’t a biobank that collects cervical swabs.

The endeavor may have to go into a clinical trial, she pointed out, which could take plenty of time.  

Eventually, after discovering the proteins that would make up a panel of markers found in either the Pap test fluid or cervical swabs used at home, there would be a need to translate that into something like an ELISA (enzyme-linked immunosorbent assay) that would be more commonly used in clinical lab settings than a mass spectrometer, she added.


Paul Nicolaus is a freelance writer specializing in science, nature, and health. Learn more at

Load more comments