How the CA-125 became a $50,000 blood test

What could be so simple as a blood test?

A quick prick with a needle, a wait of a day or two for results and a discussion with the doctor about those results. In the words of our vaunted politicians, it would be an “up or down vote” on whether there was anything to deal with.

That was the thinking of my wife’s physician when she ordered a test called the “CA-125.” At the time, it was popular as a screening test for ovarian cancer, a disease that had painfully and prematurely ended my wife’s mother’s life.  My wife felt fine, so an abnormally high level could indicate the presence of a small (read curable) cancer. A normal level would signal relief from fear.

As a physician and researcher, I consider myself an expert in the field of diagnostic testing, and was dismayed when my wife came home reporting  as much.

My exact response was “oh, s**t!”

What triggered my outburst was what I knew to be the other side of the story. The CA-125 is a well-meaning idea, that  fails miserably in action. The test is so  inaccurate that  it misses many cancers (in  doctor talk, a false negative) while  labeling many cancer free people as “sick”( a false positive) .

Well, it was too late, the deed was done and we’d have to live with the results. Besides, I was sure that my wife, a supremely healthy woman, wouldn’t be one of the unlucky ones.

A few days later, the first—but not the last—call came.  “Rebecca, your CA-125 is elevated . It’s not very high, but it’s high enough to be of concern. We’ll have to do some additional tests to see why. And that generated the first—but not the last—test.

It began routinely with an ultrasound to check her ovaries. Routine is easy for me to say. Even the least painful or invasive test of the ovaries subjects one to various indignities. They all involve probing  body cavities. The result was, in the words of her doctor, “essentially normal”, which meant that there were a few abnormalities that didn’t look like cancer. That was to become a recurring theme in the coming months.

Her “essentially normal” ultrasound—which showed a cyst or two– generated a second CA-125 , which came back just a little bit higher.  Not the right direction, and enough to make us just the slightest bit queasy. We needed another opinion from an expert—so off to a specialist in women’s reproductive cancers—or gynecologic oncologist.  He reassured us that cancer didn’t seem likely, but that we needed to be sure.

Being sure meant a series of other tests, especially lots of imaging tests.  What we learned is that the more she was tested, the more “things” the doctors found, and the more uncertain the picture became. Some benign looking liver cysts—it turns out that Rebecca, like many other people, has a number of internal cysts– on her CT of the abdomen turned into “can’t rule out metastatic cancer, suggest further evaluation”.  That’s radiologist talk for CYA.

That prompted her primary care doctor to insist on a search for other kinds of cancer, especially  colon cancer.  My wife refused what looked to her to be a series of new tests, designed to function as a distraction instead of a source of certainty.

It felt like her doctor was meandering around my wife’s body on a draconian treasure hunt.

To be sure, we were both becoming a bit frightened. To calm our fears we sought information from multiple sources. I turned to the medical literature  and expert colleagues. Rebecca turned to me and to her doctor. I offered reassurance, but she could see the uncertainty and concern in my words and body language .

Her doctor offered certainty, borne of the conviction that Rebecca was secretly harboring a cancer.  That led to a paradox: the less evidence of cancer her doctor found, the harder she looked to find it. She coupled that conviction with a need to convince Rebecca that it was time to prepare for all of the trappings of cancer treatment. She began to talk about how to deal with the vomiting and hair loss of chemotherapy.

When Rebecca complained to her doctor that such conversation was not helpful, and was frightening her, the response was “My job is to frighten you.”  Rebecca wondered out loud: “I thought your job was to inform and advise me.”

This relationship was not going well.

The combination of my worried uncertainty and the doctor’s dark certainty led Rebecca to note that the house was being hung with crepe.  In retrospect, that was an understatement. She later confessed to a sleep depriving terror that she would repeat her mother’s painful, asphyxiating demise at the hands of the same cancer.

The point was approaching that there weren’t many more tests left to do—and still no definitive evidence of cancer.  But a third CA-125 came back even higher yet. It was still barely above normal, but nowhere near the range found in women with cancer.

But it was getting higher each time.

Could there be a tiny cancer lurking in her ovaries? What else could cause this steady, but slight rise in the test levels?  My research suggested a number of possibilities, but nothing that quite fit Rebecca’s situation.

We turned again to the gynecologic oncologist for answers. How long would we keep checking the CA-125?   He offered an alternative. He would take a direct look at her ovaries. Translation: perform surgery, directly visualize everything in the abdomen, look at each ovary and   take samples to examine under the microscope.  To be sure, this offered a definitive answer. It reminded me of a dark joke among physicians. It goes like this: when an internist is offered a gift-wrapped package, he/she shakes it, holds it up to the light, smells it, and listens to it to determine what’s inside. When a surgeon is offered the package, he/she rips the wrapping off, opens the box and looks inside. Despite being an internist, I could relate to that. Now seemed like a good time to take the surgeon’s approach.

Rebecca had the final vote in this discussion. “If you’re going to cut me open, I want this to be the one and only time this happens, so when you’re there take everything out.”  Having passed her childbearing years, she felt little need for all internal  reproductive parts. So it was decided: she would undergo a total abdominal oophorectomy and hysterectomy, removal of the ovaries and uterus.   The operation was scheduled for 3 weeks hence, which allowed us to take a long-planned vacation.

Enter stage left her internist, who called the surgeon and insisted he move up the date. Despite the fact that this whole process had already taken 5 months, she feared that a further 3 weeks could be the difference between life and death. She berated my wife for putting a vacation before the surgery.

My wife’s response  was to fire her and seek another internist with more compatible  values.

So, after months of uncertainty, surgery seemed like a form of relief. No matter the answer, it would provide a sense of finality that we had not enjoyed so far.

Approaching my wife’s bedside, the smile on the surgeon’s face spoke volumes. The surgery was uneventful and there was no sign of cancer.  Oh, and the mystery of the rising CA-125 was solved. The tip of Rebecca’s appendix harbored a pea-sized bit of aberrant uterine lining (called endometriosis), which is known to secrete CA-125.

Rebecca recovered fully from the surgery, just as healthy as she was months earlier, sans several organs, but she never quite recovered her trust in most physicians and their tests (fortunately, I’m her one  exception).

What was the cost of all of this? To paraphrase that credit card ad:

Multiple blood tests and imaging procedures (with attendant exposure to lots of x-rays): $10,000

Major surgery and 4 day hospital stay: $40,000

Five months of mounting worry, loss of several organs, and a simmering distrust of doctors and their tests:  incalculable.

Keith Marton is an internal medicine physician.

Submit a guest post and be heard on social media’s leading physician voice.

Comments are moderated before they are published. Please read the comment policy.

  • Mike Thompson, MDPhD

    Dr. Marton – Tragic story. I always think back to early training about pre- and post-test probabilities

  • Niamh van Meines

    I’m glad the outcome was favorable. Like PSA’s, these tests should not be used at all if the results are variable and so traumatic.

    • Anonymous

       So you are basically saying that no one should be diagnosed with ovarian cancer and we should all be left untreated to die?

  • Candi Scheuermann

    So So true and unfortunately a recurring event for many people.  I hear these stories in groups all the time lately.  I myself under went 4 years of tests and surgeries and implanted devices and seeing many doctors only to be told that Oops they misdiagnosed me and I didn’t need any of the last 4 years of misery.  Then had to have another more complicated surgery to remove the device.  We had to take out loans for tens of thousands of dollars to pay for these procedures and surgeries that I was made fun of for initially refusing.   I am glad your wife’s outcome was so good.  I hope for the same for myself but the wariness of doctors and their “opinions” and the anxiety is really all I got for the money.

  • Manesh Dagli

    I disagree with Dr. Thompson.  This is not a tragic story.  Rebecca does not have ovarian cancer and now no longer has to worry about ever having it.  There’s nothing tragic about being alive and still able to enjoy and experience life.    The most tragic outcome would be to miss a case of ovarian cancer in a patient whose mother died of ovarian cancer.    In addition, if you want to bring up “pre- and post-test probabilities”… this patient’s mother died of ovarian cancer, if she’s not a high risk than who is?  As physicians we see cases that defy probabilities every month; our only choice is to rule out the worst outcome or face 20-30 truly tragic outcomes and malpractice case per year.  Ovarian cancer is a cancer which is often missed and has very few and very vague symptoms, if she was your patient would you not try to perform a screening test? 

    • Anonymous

      It has been proven over and over again that “screening” even high risk patients with CA-125 is of absolutely no benefit.

      If I could have a dollar for every patient that I’ve tried to talk out of this test, usually after reading a spurious e-mail that was forwarded to them . . .

    • Violetta V

      All surgery has risks. What would you say if Rebecca happened to have something happen to her during surgery? Would you call that tragic?
      Also, losing an uterus and ovaries even post menopause may still have a negative effect on her future health.
      The stress she went through is also likely to have had negative effect on her health including her heart.

      Most importantly – THERE IS NO EVIDENCE THAT THE TEST SAVES LIVES. Period. Just because something is detected earlier doesn’t mean one’s life is saved. It could be that the cancer is detected early specifically because it was slow growing to begin with and would be curable even if detected later.

      Oh, and a cancer detected earlier could also be an overdiagnosis. So a woman can be treated with chemo/radiation for something that would’ve never threatened her life to begin with. And suffer side effects of the treatment maybe for life.


  • RCK

    I call it the ” the medical merry-go-round”  Once you get on it is hard to get off. CYA is so true, no test ever is definitive now, always suggest more tests.

  • Anonymous

    I have another anecdote. Recently postmenopausal woman was referred for a sonogram after her pcp received her CA-125 level that was slightly elevated for a postmenopausal woman. The ultrasound was entirely unremarkable. Then her Gyn, having read some reports, told her that he wanted the test repeated in 3months to note any interval change. This patient, appropriately anxious, could not wait so she had her blood drawn after 7 weeks. Now the test showed a level in the 1000′s – truly off the charts. Her ultrasound was repeated and the only finding different from the original sonar was a moderate amount of free pelvic fluid. No masses were detected. She was referred to the chief of gyn oncology at the local center who advised laparoscopic surgery for diagnosis/treatment. At surgery, mild ascites and multiple peritoneal and serosal implants were found throughout the abdomen but no definitive primary. This lady had a fortunate response to chemotherapy and thus far is doing well (4 years out.) While I would also not favor indiscriminate use of such tests as CA 125, the test surely saved this woman’s life.

    • Violetta V

      Plural of anecdotes isn’t data. You don’t know what would’ve happened if this cancer had been detected later. You don’t know what will happen later.

      If there really had been a benefit from this test, the studies would’ve shown it. The studies have shown no benefit from ordering this particular test. 

      Oh, and sometimes doctors think there is an elevated risk where there is none. My ex-ObGyn at one point told me that she wants to order a pelvic ultrasound on me every year. When I asked her why, she tells me “we don’t know why your ovaries failed”. Well, pray show me a single shred of evidence that women with spontaneous POF have an elevated risk of ovarian cancer. The information sheet on ovarian cancer shows the opposite – reduced risk with early loss of ovarian function. I changed the doctors. Oh, and I saw a doctor on the web who write that women with POF are at elevated risk of breast cancer compare to general population (as in women who have normal periods and are exposed to natural hormones) because they take HRT. Never mind, that these women start with reduced risk to begin with and that there has never been a single study that showed that HRT until the age of normal menopause extends this risk above that of a women who has normal periods. When I asked him to show a study that compared women with POF on HRT with still-menstruating women of the same age rather than menopausal women, he started writing irrelevant condescending posts.

      • doc99

        The problem with “studies” are that bell-shaped curve. Unfortunately, not all patients read the same book. Moreover, there actually is some evidence that Multimodal screening may be of benefit. Again, this is not to say I advocate indiscriminate use of CA 125 as a screening test, a clinician, when confronted with a patient as this one having had the test performed, should consider a repeat test after a period of time to note interval change.

  • Deep Ramachandran MD

    Thanks, Dr. Marton, well written article and very important message. As physicians we all see abnormal lab results that must be investigated further for the very small chance of meaningful pathology. Until we have meanigful tort reform and change the way we practice such stories will be all too common.  

    • Ailan Medici

      Oh, what comfort to the patient.  Doctors wishing for tort reform so they can freely dismiss patient symptoms without fear of being sued for underdiagnosis.  If your assessment of ‘no further tests needed’ is correct, why would the patient sue you?

    • Anonymous

       I am incredulous at the way you talk. Since probably the majority of women who do have ovarian cancer are misdiagnosed, often for months, and blown off by a succession of doctors until they are diagnosed months later at stage III or IV, I would have to strongly disagree with you. Again, this woman had a first-degree relative with ovarian cancer. I think that under the circumstances, it would have been heinous not to pursue this until there were definitive answers.

  • Dan Scarfo

    I always worry how they keep correct labling to insure that blood test you passed is really test…

  • Anonymous

    Dr. Marton, if after all the exhaustive tests your wife had, she ended up to indeed have ovarian cancer, would you have written the same article?

    • Keith Marton

      Emily, For sure the story would have had a different ending. Based on what we know, Rebecca would most likely be dead now–from Ovarian cancer. Since the CA-125 has not been shown to save lives my story would have most likely focused on something called “lead bias” . That is, a poorly performing screening test may well discover a cancer earlier but not really change overall life expectancy. The sad result: the patient gets to know they have cancer for a longer period of time than they would have if they had not been screened, but they don’t live any longer. 

    • Gregory Pawelski

      What if Dr. Marton’s wife had ended up indeed to have ovarian cancer? They did the very best (first) thing: the reason for better survival for patients who undergo complete resection without any tumor left behind is that these tumors are biologically less aggressive and would do better regardless of the trial-and-error (physician’s choice) treatment they received. Surgery is an integral part of the multimodality treatment of many cancers. Kudos Dr. and Mrs. Marton!

  • Anonymous

    That $50,000 may have save your wife’s life, especially with her family history. The mystery to me is why a prophylactic BAH/O/S wasn’t suggested at the same time as a CA125. Those of us living with ovarian cancer-and those of us living with it are the fortunate ones-have had to educate ourselves, and some of us very quickly. Most OvCa isn’t diagnosed until a much later stage, and some gyn/oncs speculate some forms can go from stage 1 to stage 4 in a matter of weeks. OvCa is the poor sister of the female cancers; it is relatively rare, and appearance isn’t a factor as in Save the Boobs. Women find little knowledge, and little support. Once treatment starts, most are eligible for SSI, if that gives any indication of the severity of the side effects from being gutted and dosed. Mucinous ovarian cancer, “my” cancer, is believed to originate in the appendix. I hope that little piece of endometriosis on your wife’s removed appendix was assayed and tested and bx to hell and back. Thank heavens for the elevated CA125 and the direction it took you, even if it was a false elevation. 

  • Anonymous

    While the downsides of cancer testing are clear, I think your post is misleading and in fact, harmful. Given that your wife’s mother had ovarian cancer, and given that ovarian cancer is nearly always found in a late stage and has a poor prognosis, I don’t think it was wrong to err on the side of caution. (That having been said, the internist sounds like someone with no idea of how to talk to a patient.)

    Your statement that your wife’s CA-125 “was still barely above normal, but nowhere near the range found in women with cancer” is just plain wrong. There is no such number. Firstly, some women with ovarian cancer never have an elevated CA-125, and secondly, you can have ovarian cancer with a slightly elevated CA-125. I was diagnosed with stage IA ovarian cancer with a CA-125 of 75, which is not very high.

    Bottom line: There are real problems with overtesting for cancer (read Should I Be Tested for Cancer? Maybe Not, and Here’s Why). But I don’t think this was a good example of the problems with overtesting, given the ramifications if it had turned out that your wife had cancer.

    • Anonymous

      The doctor ordered a test that was not indicated and did not inform the patient of the consequences of the test. Those actions cannot be defended.

    • S

      Completely agree with southern doc. I have yet to see any evidence in support of screening CA125 in this setting. Indeed in the June 4th edition of JAMA (305(22)2295-2303:2011) The PLCO trial using transvaginal U/S and CA125 testing specifically did NOT show any survival benefit in average risk woman and invasive testing was associated with complications. Evidence-based medicine has shown NO BENEFIT…..PERIOD. For all of you naysayers, DO A LITTLE RESEARCH. I get frustrated with these type of referrals. Read the damn evidence supporting a test before ordering the test.

      • Anonymous

        Do you consider Dr. Marton’s wife, a woman with a history of ovarian cancer in the family (her mother), to be of “average risk”?

        • S

          The recommendation presently is to treat woman who are at higher risk but DO NOT have familial ovarian cancer syndromes (Lynch syndrome or BRCA) as average risk. You are asking the wrong question and to answer your incorrect question, yes I would treat this woman as average risk until I have more information (ie. NO CA125).  The right question is did the internist evaluate to determine if this patient had a familial ovarian cancer syndrome. That involves simply a family history and an understanding as to when to consider further testing. Again, NOT a kneejerk CA125 or kneejerk BRCA testing.

    • Anonymous

      Why did it have to lead to hysterectomy?  What was wrong with laparoscopy to directly visualize the ovaries or even just removing the ovaries?  Who was so consumed by the desire to eliminate the possibility of a future malignancy that they would undergo potentially unnecessary surgery? The internist?  The gyn oncologist?  The patient?  I’m glad that ultimately your wife didn’t have cancer, but it seems like a case where everyone got consumed with panic.

  • Violetta V

    The study you reference talks about the sensitivity, specificity, and positive-predictive values of the combination of tests. It doesn’t tell anything about the test’s ability to save lives. Just because a cancer is detected earlier doesn’t mean it is going to save someone’s live – only studies that compare ovarian cancer mortality in screened vs non-screened groups show if the test has a benefit. The study about sensitivity is interesting, but it doesn’t prove anything. Maybe the cancers that are detected by the test are slow-growing to begin with which is why they are detected on time.

  • Anonymous

    My maternal grandmother died of ovarian cancer.  My mother got it in her 50s.  We’re all only children, so the odds weren’t looking good…..I felt like a timebomb.  

    Two of three doctors consulted said “if you were my wife/daughter, I’d want you to have your ovaries out after you’ve had your kids”.  The third was doing research on waiting/watching so suggested that route….no thanks.  I had to be very assertive with my insurance company to cover my “prophylactic” laparoscopic hyst/oopherectomy.  

     I’m glad your wife didn’t have cancer…..but sorry for the high price that was paid.

  • Suzanne

    My mother also died of ovarian cancer.  It scares me that there is really no test to positive diagnose other than going in.  I too am contemplating the same procedure just to get rid of the uncertainty.

  • Anonymous

    As an ob/gyn, I’m wondering why in the world did you keep repeating the CA-125 and going down this path that inevitably led to a surgery for benign disease?  What happened to repeat imaging to demonstrate stability in the size of structures over time, which would be inconsistent with a diagnosis of cancer?  I’m not sure who lost perspective here — your internist, your wife, or both, but it ultimately led down a path of chasing a very imperfect test, putting the results of that test in the wrong perspective, and leading to a surgery that ultimately was unnecessary (but probably bought you some peace of mind).

  • Michael Crespo

    Fortunately, help may be in the horizon to prevent the ordeal that your wife went through. Arrayit Corporation is developing OvaDx®, the market’s first large panel biomarker screening test for ovarian cancer. Upon FDA approval, Arrayit Diagnostics will offer OvaDx® as an elective test for women seeking greater wellness and for women in the elevated risk category for ovarian cancer. In studies OvaDx® has shown high sensitivity and extremely high specificity for all types and stages of ovarian cancer.  So false positives may be a thing of the past.OvaDx® is in late stage development. Hopefully the test  will be in the market in the near future.

  • Jay F. Schamberg

    False negatives will never be a thing of the past.  There are no “perfect” tests  — not even histopathologic diagnoses. 

    • Michael Crespo

      You are correct.  Higher specificity (less false positives) comes at the cost of somewhat lower sensitivity (false negaitves). Per Arrayit Corp website, studies with OvaDx®, the test I mentioned earlier, has shown ~80% sensitivity for Stage I ovarian cancer. So the test is catching 8 out 10 earliest stage cancer when treatment is highly successful. Although they have not published the specificity for OvaDx®, other articles I have read point to very close to 100%. For a reliable screening test for ovarian cancer experts recommend that sensitivity should be at least 75% and specificity at least 99.6%.

  • DrJoe Kosterich

    We place far too much faith in tests. And the co-lateral damage done in the rush to “early diagnosis” with these tests is huge in economic and human terms. we like to think that screening for cancer helps people but the facts (prostate breast ovary and lung) show it does not

  • dave d

    If you do enough tests on enough healthy people, you will find something abnormal. 

  • Anonymous

    If a doctor did post this article, please don’t go too him…..I mean really how dumb can you be. I’m reading this thing thinking. It’s probably Endometrois. Notice he never states the CA-125 result. With a little research he would have found this out. The ultrasound was find as the next step. Believe me not all that different from a pap. His childbearing wife has certainly gone through worst. Since she did not have any other systems. Abdominal bloating, pain, bowel problems. Then they should have waited 6 months and retested the CA-125, and if they were that concerned do a lapo, take a biopsy and call it a day. It’s an outpatient procedure.
    Also, for a physican not too have insurance seems well pretty unbelievable. $40K it cost him. Wow, buy some insurance buddy.
    The CA-125 can have a high false positive. It should be thought of like the PSA test for men. As a baseline. You should NOT panic unless it goes up drastically. Like mine did. From a CA-125 of 10 to a CA-125 of 740 in one year. Which prompted an ultrasound that found an 18cm tumor. Which was CCC Ovarian Cancer. After a complete Hysto, appendectomy, tumor removed along with 3 nodes, and 6 months of Chemo, the CA-125 is used to monitor cancer reoccurance. You have too use a little common sense here folks. Going up a few points is not something to panic about, in regards to this tests.

Most Popular