When it comes to ordering serum marker tests for older women after initial treatment for early-stage cancer, almost half of cases reviewed in a study showed that many oncologists are not adhering to recommendations of the American Society of Clinical Oncology.
The research, online ahead of print in the Journal of Clinical Oncology (doi: 10.1200/JCO.2014.55.5409), found that 42 percent of oncologists had ordered tests for CEA and CA15-3/CA27.29 for women age 65 and over in the period between 2001 and 2007, even though there was not any evidence—and still isn't—that the tests increase overall survival for this group.
“We had heard 10 percent, maybe 20, but not 42 percent,” said the study's first author, Scott D. Ramsey, MD, PhD, Director of the Hutchinson Institute for Cancer Outcomes Research (HICOR) at the Fred Hutchinson Cancer Research Center and Professor of Medicine, University of Washington. “That's a huge proportion of women getting these tests. No one knew.”
Among 39,650 eligible patients identified from Surveillance, Epidemiology, and End Results (SEER) and Medicare records, 16,653 patients (42%) received at least one tumor marker assessment, with an average of 5.7 tests over two years.
And the situation does not appear to have improved much since that study period. The recommendation advising against using biomarkers and advanced imaging for surveillance in patients with breast cancer who are treated with curative intent is No. 4 on ASCO's first “Top Five” list of the “Choosing Wisely” campaign, published in 2012 (asco.org/practice-research/2012-top-five-list), of common, costly procedures in oncology that are not supported by evidence and that should be questioned.
A study reported at the Quality Care Symposium this fall with Ramsey as senior author showed that adherence to the ASCO recommendations about not using biomarkers or advanced imaging following breast cancer treated for cure were followed by only 46 percent of oncologists in western Washington state between 2007 and 2013 (OT 11/25/14 issue).
In an interview, Ramsey explained that he and his colleagues had been interested in the use of tumor markers in part because of the “Choosing Wisely” recommendation against the tests, but also because during discussions at local oncology meetings there was a lot of disagreement among the oncologists there about the value of these tests.
“We sensed there was some difference between the guidelines and what was actually going on,” he said.
Two Factors at Work
Two factors are at work, Ramsey said—patients who want the tests, and physicians who think the tests are valuable or who are following their patients' requests.
“On the physician side, many feel these tests can improve the patient's outcome,” Ramsey said. “Of course, that's without any evidence that they do so.”
He said there is a sense among some oncologists that at the time the recommendations were first made, in the late 1990s, chemotherapy treatment options were not as advanced as they are today. Some therefore believe that catching cancers early through the use of tumor markers in the context of the new treatment options can help women live longer.
“That's an opinion I think many oncologists have, so they order them, feeling that the tests are cheap and they might help.”
On the patient side, he thought that if patients ask for the tests, it's hard for the physician to say no.
“There's a lot of information about these tests on the Internet—a lot of it is inaccurate—but I think women who have been through treatment feel a strong need to try to take control of their risk for recurrence and want to know as soon as possible if there is a chance that the breast cancer is coming back, even though there is no evidence that this is helpful.”
Not a Defensive-Medicine Situation
Ramsey said he did not think physicians are protecting themselves from lawsuits by ordering the tests, because a malpractice suit would have to prove the physician was negligent. Since these tests are not recommended, that complaint might not hold up.
“The issue is whether these tests can pick up a recurrence early and prolong survival, as opposed to picking up a cancer that would have been detected on the next mammogram and would have been curable—that is a question we haven't resolved.”
Ramsey said he and his coauthors on the paper are strong proponents of a clinical trial of tumor markers to try to answer that very question. “It's an important question because if 40 percent of doctors are using these tests, that means either 60 percent of doctors should be, or that those 40 percent should not be.”
Varies by Geographic Area, Age of Clinician
The research showed that those most likely to have the tests were women on the East or West Coasts; relative to the reference region of San Francisco, the odds ratio for women in Connecticut to be tested was 2.1, in New Jersey 3.5, and in Los Angeles 7.7.
Women in rural areas were far less likely to be tested: odds ratios were 0.56 in San Jose, 0.36 in Iowa, and 0.28 in rural Georgia.
Women with more advanced disease were more likely to be tested, which could mean their physicians felt the tests might be useful if the risk of recurrence were relatively high, Ramsey said.
Also, women who had undergone more chemotherapy underwent more marker testing.
“I'm a bit worried that in some cases the chemotherapy was given in response to the test, as opposed to actually finding a tumor—that's been seen in other situations. That's especially worrisome if you're treating a test result–which may be a false positive–rather than documented evidence that the cancer has recurred someplace.”
He said it is important to note that this study does not prove whether tumor markers are helpful or not helpful—“It just shows that there is a lot of testing, and a lot of follow-up testing with CT scans and other tumor markers, and that's all quite expensive. And you can imagine that for a lot of women, it is quite anxiety-producing to go through all of this.”
For some women, getting a low score on a test could give them some reassurance and make them feel less anxious, he said, but then again that may be false reassurance. “Ultimately we want these women to come back for regular exams and regular mammograms—that's what's recommended by the medical societies. We just don't know if getting these tumor marker tests add anything.
“We're spending a lot of money on them and on the follow-up without knowing really what we're getting.”
Still Relevant Today
Although the JCO study covered cases from 2001 to 2007, the findings are still relevant today, said Lyndsay Harris, MD, Professor of Medicine at Case Western Reserve University School of Medicine and Director of the Breast Cancer Program at Seidman Cancer Center, who was first author of ASCO's 2007 Update of Recommendations for the Use of Tumor Markers in Breast Cancer (JCO 2007;25:5287-5312).
“We had already put out guidelines [in 2000] that were recommending against the use of biomarkers in the adjuvant setting, and that's the same setting as in this study,” she said in a telephone interview. “While there may have been some improvements in uptake after the 2007 guidelines, I'm not sure that message has really gotten across because there is definitely evidence of the use of the tests in the adjuvant setting.”
She said she doubted very much that the overuse of serum marker tests is limited to patients age 65 and older as in the study, particularly since patients who come in for a second opinion are generally younger. “There's a lot of pressure on doctors to order these tests, and sometimes it's easier just to say yes,” Harris said. “But the result can be patient anxiety, and the tests potentially lead to an unnecessary biopsy or complication.”
And it's misleading to do a test that is negative and to tell the patient they don't have cancer when we know they well might, even in the absence of a positive test, she continued. Harris said oncologists are relying on ASCO to disseminate the information to providers and consumers—“And we are fortunate in the cancer field that we have patient advocates who can help, who are much better than we are at getting the message out to their constituency.”
Patricia Ganz: Pressure and Time-Consuming
Patricia A. Ganz, MD, Distinguished Professor of Health Policy & Management and Medicine at UCLA Fielding School of Public Health and David Geffen School of Medicine and Director of Cancer Prevention & Control Research at Jonsson Comprehensive Cancer Center, was the Choosing Wisely advocacy representative for the ASCO committee that wrote the current recommendation against serum marker testing.
Asked in a telephone interview if Ramsey's JCO study of data from 2001 to 2007 is relevant today, Ganz referred to a 2013 study on which she was senior author, which found a high rate of non-recommended testing—as well as underuse of mammographic screening—in early-stage breast cancer survivors treated and followed at an academic medical center (Hahn C et al: Cancer 2013;119:4316-4324).
The patterns of over- or under-use were observed in a setting where individual physicians do not have a financial incentive for ordering tests and services, and where physicians would have access to evidence-based guidelines to direct follow-up care.
Ganz said she knows that many women are having PET-CTs for staging and MRIs for staging or follow-up, and that there are inevitably false positives.
“I know women are having double mastectomies because of the MRIs that are being done at baseline,” she said, adding that she is very familiar with the patient who asks for the marker tests.
“It takes about 15 minutes to explain to the patient [that the tests do not improve survival], and you have to do this every time the patient comes in for about the first year after treatment. If you have a good relationship with the patient you can discuss these things. Occasionally they'll twist my arm, and occasionally I'll give in. But if you can explain the facts about the real value of the serum marker tests, they'll eventually understand.”
Ganz agreed with Ramsey about the need for randomized clinical trials to provide up-to-date evidence about the value of serum markers for this group of women.
“There is doubt in a lot of clinicians' minds who feel that with the latest treatments, the early diagnosis of recurrence would today improve overall survival. A randomized trial would be very helpful.”
But it will be a challenge to mount such a trial, she said, because “treatment for early-stage breast cancer is very good today and it would take a large number of patients who would have to be tracked.”