NCCN Endorses PSA Testing In Absence of Better Alternatives
NCCN Endorses PSA Testing In Absence of Better Alternatives
Fine tuning, not abandoning PSA testing, is the approach
The National Comprehensive Cancer Network’s revised guideline for early detection of prostate cancer continues the standoff on PSA testing. The NCCN says it “supports the continued use of PSA testing for the early detection of prostate cancer in informed, healthy men in certain age groups.”
That directly contradicts a 2012 recommendation from the United States Preventive Services Task Force (USPSTF) not to use the test, period.
The USPSTF was able to make a simplistic, blanket recommendation because its mandate does not require offering an evidence-based alternative approach for diagnosing prostate cancer. The key tools for diagnosing prostate cancer are a family history, PSA test, and digital rectal exam, but a family history and DRE by themselves don’t get the job done.
In the real world, between 23 and 25 million PSA tests are performed annually, and it is unlikely PSA testing will be discontinued. One reason: Prostate cancer is the second leading cause of cancer deaths in men.
The USPSTF’s recommendation stems from the universally recognized fact that PSA testing is the dominant cause of overdiagnosis and overtreatment that result in unnecessary harms.
Positive PSA tests lead to biopsies and up to 33% of men will have pain, fever, urinary difficulties, and infections. Following positive biopsies, 90% of diagnosed men are treated and they are at risk for erectile dysfunction and urinary incontinence, which occur in 29 and 18 of 1,000 men, respectively. In terms of benefits, while 28,000 men die annually from prostate cancer, PSA testing will prevent only 1 death in 1,000, or 28 deaths total, say the USPSTF and the NIH.
The NCCN acknowledges the harms of PSA testing; its perspective is that PSA testing is the starting point for diagnosing prostate cancer and that steps should be taken to improve its value.
“The Achilles’ heel of PSA testing is overdetection, and the USPSTF got that right — you do not need to treat all patients immediately to save lives,” says Peter Carroll, MD, chair of the NCCN’s prostate cancer early detection panel. “But the USPSTF did not interpret the data on PSA testing correctly. Overlooked data says that PSA testing reduces mortality and the justification for PSA testing is randomized trials that demonstrate a reduction in prostate cancer-specific mortality,” says Carroll.
“The USPSTF did not interpret the data on PSA testing correctly,” says NCCN’s Peter Carroll, MD. “The justification for PSA testing is randomized trials that demonstrate a reduction in prostate cancer-specific mortality.”
The NCCN’s revision is an attempt to move beyond squabbling about PSA testing. “The guideline homes in on identifying cancers that are potentially fatal while trying to avoid diagnosing the indolent cancers that cause anxiety,” says James Mohler, MD, chair of the NCCN Prostate Cancer Guidelines Panel and chair of urology at the Roswell Park Cancer Institute. But some of its recommendations may inflame PSA supporters.
“I tell patients that if a smart urologist recommends watching your prostate cancer, it is extremely unlikely you will ever die from the disease,” says James Mohler, MD, at Roswell Park Cancer Institute.
A core element in the guideline is an initial PSA test that serves as a baseline for repeat testing at future age-specific intervals. In addition, “One item that has received attention is that we recommend considering a baseline PSA test in men between the ages of 45 and 49. We point out that a baseline PSA trumps ethnicity or family history as an indicator of risk. If you really want to identify men at higher risk, then get a PSA at age 45 to 49,” says Carroll.
Age 50 has been the common starting point for PSA testing for men with normal risk, and results from the landmark trials indicate the highest benefit of testing occurs at ages 55–69.
Health plans commonly cover PSA testing and some plans begin coverage at age 40 for high-risk men. Medicare covers it at age 50. However, “under the Affordable Care Act, PSA testing is not a required preventive service because it has not received a positive recommendation from the USPSTF,” reminds Donald Liss, MD, senior medical director at Independence Blue Cross.
Many of the NCCN’s recommendations are revised approaches to biopsies — more would be conducted — and incorporation of second-line tests such as PSAV, PCA3, and the prostate health index. The cutpoint that indicates a biopsy is PSA >3.0 ng/mL, whereas >4.0 ng/mL has been a common point for triggering a biopsy.
“Under the Affordable Care Act, PSA testing is not a required preventive service because it has not received a positive recommendation from USPSTF,” says Donald Liss, MD, senior medical director at Independence Blue Cross.
In contrast to the NCCN’s lowered biopsy cutpoint and the common cutpoint of >4.0 ng/mL, the USPSTF recommends investigating a higher, more conservative cutpoint for biopsies as a potential means to reduce overdiagnosis.
While the NCCN recommends PSA testing, it also emphasizes that shared decision making and active surveillance are answers to the dangers of overdiagnosis and overtreatment. The NCCN and all other credible guidelines say a conversation about the risks and benefits of PSA testing plus decisions about biopsy and treatment should occur before PSA testing. Active surveillance may be appropriate for up to 50% of diagnosed patients.
“The rationale for active surveillance is that prostate cancer is a different type of cancer, it is very slow growing, and the chance of death is lower than many other cancers. Patients have to understand that prostate cancer is different and the risk of death is different. The answer is an informed discussion with shared decision making about what action to take,” says Charles Cutler, MD, chair of the American College of Physicians’ Board of Regents and a practicing internist.
“Active surveillance is treatment. In studies of active surveillance, we are now up to more than 4,000 North American and European men without a single death from prostate cancer. So I now tell patients that if a smart urologist recommends watching your prostate cancer, it is extremely unlikely you will ever die from the disease,” says Mohler.
“Somewhere between 20% and 30% of individuals in active surveillance will eventually be converted to active treatment. About two thirds of those are because there is some indication that the cancer might be growing. About one third of men revert to treatment because of anxiety. The men who convert from surveillance to active treatment have a higher cure rate than a matched cohort of men who enter treatment immediately upon identification of a tumor. That should reassure men about active surveillance,” says Mohler.
“There’s a lot more talk about active surveillance but I’m not sure there’s a lot of uptake,” says Ballentine Carter, MD, chair of the American Urological Association’s guideline panel for early detection.
Active surveillance may be the praiseworthy approach to managing prostate cancer, but it flies in the face of reality. Economic ardor often prevails over good medicine in that the income from self-referral to treatment services is a major cause of overtreatment. An article in the Oct. 24, 2013, issue of the New England Journal of Medicine reported that nearly all of the 146% increase in intensity-modulated radiation therapy (IMRT) for Medicare patients with prostate cancer among urologists with an ownership interest in the service was due to self-referral. The study corroborated a similar report by Government Accounting Office.
“This study provides clear, indisputable evidence that many men are receiving unnecessary radiation therapy for their prostate cancer due to self-referral,” said Colleen A.F. Lawton, MD, chair of the American Society for Radiation Oncology (ASTRO), in a written statement. “Unfortunately, the continuous stream of data indicate that patient choice is being restricted — patients are being steered to the treatment that provides the most profit for the urologist. As a result, patients are subjected to unnecessary treatment and side effects, and millions of dollars are wasted.”
ASTRO historically has been a full-bore advocate for the economic interests of radiotherapists.
The study also reported that participation rates in active surveillance decreased during the ownership period among self-referring urologists in private practice.
The study covered growth in 2005–10 but urologists continue to install IMRT machines in their practices. Philadelphia’s Fox Chase Cancer Center operates a program that actively supports this arrangement. It provides radiation oncologists who are located in urologists’ practices.
“Physicians are perennially interested in new gizmos like IMRT and research since 1991 has shown that utilization of new technologies goes up when they have an economic incentive like referral to services they own. This reflects both human nature in a capitalist system as well as serious flaws in a health care payment system that rewards doing more rather than less without regard to outcomes,” says Alice Gosfield, JD, a health care lawyer and member of Managed Care’s editorial advisory board.
“Physicians are perennially interested in new gizmos like IMRT,” says Alice Gosfield, JD. This, she says, “reflects human nature in a capitalist system as well as serious flaws” in a system that rewards volume without regard to outcomes.
No good test
Ultimately, the resolution of bickering about PSA testing is tied to better diagnostic tests and treatments. “Every urologist, oncologist, and personal physician knows that we need a really good test to distinguish life-threatening cancers,” says Mohler. “The problem is we don’t have a perfect way of identifying them.”
Approximately 80% of positive PSA tests are false positives when PSA levels are 2.5–4.0 ng/mL, the USPSTF says. Biopsies are also a problem. “The rate of overdiagnosis of prostate cancer increases as the number of men subjected to biopsy increases. The number of cancer cases that could be detected in a screened population is large; a single study in which men eligible for PSA screening had biopsy regardless of PSA level detected cancer in nearly 25% of men,” the USPSTF says.
“The use of biopsies to monitor disease progression has not been proven. The necessity for and frequency of biopsies is not known. There is every different schedule you can imagine and it may take several years to get better information on the frequency of biopsies,” says Mohler.
Mohler concludes, “Some people believe that we can put all of the available tests and measures together into a nomogram and tell indolent and aggressive cancers apart, but we’re imperfect.”
Yet there has been some progress. “The advances in testing have been around markers of specificity. These are tests which are better designed to tell you who might have cancer or benign disease in those with elevated PSA levels. These tests are free PSA, PCA3, and prostate health index,” says Carroll.
“These, however, are not first-line screening tests. They are tests for PSA levels between 4 and 10 and when you are considering whether or not they might consider a second biopsy,” says Carroll.
Imaging, focal therapy, and molecular diagnostics are the hoped-for new interventions that will identify only aggressive tumors and reduce harms.
MRI is being used to identify the exact location of tumors and rule out any aggressive disease. A very sophisticated version, multiparametric MRI, covers anatomic, metabolic, diffusion-weighted, and contrast imaging, requiring an hour-long confinement inside the magnet.
Focal therapy is an alternative to conventional surgery and radiation. It includes less-invasive procedures, including laser therapy, cryoablation and high-intensity focused ultrasound — all intended to preserve a man’s continence and potency. Focal therapy is still in early clinical development.
“There’s a new generation of molecular tests, but they haven’t been tested thoroughly,” says Mohler. “These molecular tests are being used in conjunction with the biopsy results to identify gene signatures that distinguish aggressive tumors. The average prostate cancer has 5-7 individual tumors and one problem is identifying which tumor is the one that controls disease progression. Another problem is whether or not the genetic profile or mutations change over time, in other words, how does prostate cancer evolve and when should you sample it?” says Mohler.