To Test or Not to Test? To Treat or Wait and See?

In prostate cancer, seeking two holy grails: better risk assessment and improved treatment
By Aliyah Baruchin | Illustrations by Daniel Bejar

My back hurts.

For Mitchell Benson’77, chair of urology at P&S, those three words are a touchstone. Twenty years ago, before the use of PSA screening for prostate cancer became routine, back pain was a common complaint of men walking in the door at Columbia Urology—and a terrible portent: It meant the men had prostate cancer that hadn’t been detected in its early stages and had metastasized to their bones, including the spine.

That the complaint of back pain is now virtually a thing of the past among newly diagnosed prostate cancer patients describes the evolution of prostate cancer care over the past two decades. Early detection has become the rule, delayed treatment should be common for men with low-risk cancers, and individualized risk assessment has become as crucial in care decisions as improved surgical and radiotherapy treatment techniques.

The medical community’s prevailing points of agreement about prostate cancer are grouped at two opposite poles. On the plus side, the disease is often so slow-growing and nonaggressive that a substantial percentage of patients will die of something else even if they never receive treatment. Where treatment is needed, prostate cancer benefits from a plethora of options, and it is often curable. On the minus side, despite prostate cancer’s reputation as one of the more “benevolent” cancers, treatment in the form of surgery and/or radiotherapy can be worse than the disease, often resulting in incontinence or loss of sexual function or both, and prostate cancer remains the second leading cause of cancer death in men in the United States, claiming some 28,000 lives a year.

That tension between pluses and minuses feeds the central controversy in prostate cancer care today: whether screening asymptomatic men using the prostate-specific antigen test saves lives or increases the risk of harm. The PSA controversy exploded in May 2012, when the U.S. Preventive Services Task Force gave the test a resounding “D” grade, recommending against its use as a screening tool. The independent panel of experts in prevention and primary care (no urology or oncology experts were on the panel) found that the test led to “overdiagnosis” of prostate cancer and, consequently, overtreatment.

For Dr. Benson, as for many urologists across the United States, that line of thinking is unworkable on multiple levels. “There’s no such thing as overdiagnosis; there’s only overtreatment,” he says. “And you can’t decide whom to treat and whom not to treat, you can’t establish risk, without diagnosis. So the concept of overdiagnosis is dangerous, and it will preclude patients from getting life-saving therapy when it’s indicated.”

A test based on a PSA precursor co-discovered by Kevin Slawin’86 became the first new prostate cancer screening instrument since 1998 to receive FDA approval.

In essence, prostate cancer care is still bedeviled by the two most crucial things that remain unknowable—whether some elevated PSAs are a sign of cancer rather than a benign problem such as infection or inflammation and which detected cancers will become aggressive. A man’s PSA comes back high but not conclusively so, depending on age somewhere in the range of 2.5 to 10 ng/mL, and a biopsy is done, risking the kinds of complications, such as infection, that follow up to 5 percent of prostate biopsies. When the biopsy finds prostate cancer that appears to be indolent, or not at a high risk for progression, active surveillance—frequent monitoring of the prostate without cancer treatment—is often recommended. However, some patients request or even insist on treatment. As a result, the patient risks possible impotence or incontinence or both because of a cancer that might never have threatened his life—all, the logic seems to say, because he took a seemingly “unnecessary” test that found it in the first place.

But can it truly make more sense not even to inquire? Making sound decisions about screening involves a matrix of considerations, including a man’s age and factors such as African-American race and first- or second-degree family history, which are known to increase risk.

Dr. Benson stresses that the overall boundaries are clear. “The goal for prostate cancer treatment is death from something else,” he says. “Prostate cancer is, in general, a slow-growing cancer, so to take an 85-year-old man with a PSA of less than 4 and say he needs ongoing PSA screening is insane. But to take an 85-year-old man who has never had a PSA ever in his life and say he shouldn’t get one, just to see where he’s at, is also crazy. I think every man, regardless of age, deserves one PSA. If that one PSA places you in a category where the chance of your dying of prostate cancer is low, then you don’t need to have biopsies and additional PSA testing. This is a $15 blood test. The rub here is not in the PSA; the rub is in what people do with the data.”

The PSA’s limits as a screening tool are part of the problem: Many men undergo unnecessary biopsies when their risk assessment is based on PSA alone. Because of this, improving the specificity of PSA remains an area of significant research. “PSA is prostate-specific, but not cancer-specific,” says Sven Wenske, MD, a urology resident at Columbia. “We use other tools as well, such as a fraction of the PSA which is called the free PSA, or other biomarkers, such as PCA3, a urinary marker which looks at RNA that comes from prostate cancer cells, recently FDA-approved for use in patients with an initial negative biopsy who are at a higher risk, based on their PSA, for prostate cancer.”

The effort to improve PSA screening is taking a range of forms. An August 2012 guideline from the American Society of Clinical Oncology suggests different parameters for the test’s use, recommending that the PSA be used to screen for prostate cancer only in men with a life expectancy of more than 10 years and only after close consultation with their doctors about the possible risks and benefits of the test and its overall appropriateness for them. And a February 2013 study in the Annals of Internal Medicine looked at 35 “alternative screening strategies” for deploying the PSA more effectively, using different start and stop ages, screening intervals, and thresholds for referring patients for prostate biopsies. 

'If you could do a prostatectomy that had no side effects and you treated the cancer—whether you think the cancer was a threat to your life or not—we would hear fewer arguments about overtreating.'

Other researchers are working to develop entirely new screening tests that aim for greater specificity than the PSA. In June 2012, a new test based on a PSA precursor co-discovered by Kevin Slawin’86, director of the Vanguard Urologic Institute at Houston’s Memorial Hermann-Texas Medical Center, became the first new prostate cancer screening instrument since 1998 to receive FDA approval. The test combines measurement of the precursor, [-2]proPSA—which is more elevated in prostate cancer patients and can more accurately identify the disease—with measurements of PSA and free PSA. The resulting “prostate health index,” or phi, can be used to help a man 50 years of age or older, with a borderline PSA level of between 4 and 10 ng/mL and a negative digital rectal exam, decide whether he should have a biopsy. 

Results of the multi-center clinical study of the phi test showed a 31 percent reduction in unnecessary biopsies over standard PSA screening and preferential identification of more aggressive, potentially life-threatening cancers. “The FDA approval of phi—based on the discovery of [-2]proPSA—may have reopened the door on effective screening and offers hope to patients who may need treatment for a disease they otherwise wouldn’t know they had,” says Dr. Slawin. “It represents a significant step forward in the prostate cancer screening debate and has the potential to rebalance the scale toward screening—preventing us from losing the considerable ground we’ve gained since PSA was first introduced.”

At the other end of the spectrum, one of the PSA’s most ardent detractors is Anthony Horan’65, author of “The Big Scare: The Business of Prostate Cancer” (reissued in 2012 as “How to Avoid the Over-diagnosis and Over-treatment of Prostate Cancer”), who disagrees with much of the customary practice in screening, diagnosis, and treatment of prostate cancer. Dr. Horan, who has a urology practice in California, agrees with the U.S. Preventive Services Task Force’s recommendation and leaves the PSA test last in his chain of diagnostic events. “Screening for prostate cancer with a blood test and treating the cancer, discovered in the absence of a palpable nodule, offer no measurable good that outweighs the measurable harm,” he wrote in a post on the Health Care Blog. “Prostate cancer is a very slow moving disease with estimates showing that 94 percent of the cancers detected with the routine PSA blood test would not even cause death before the age of 85. More men die in accidents than of prostate cancer.”

Caught in the crossfire of the PSA screening controversy are African-American men, who are at greater risk than men of other ethnicities at every stage of the prostate cancer screening, diagnosis, and treatment process. African-American men are more likely to develop prostate cancer, more likely to be diagnosed with high-grade tumors, more likely to present with metastatic disease, less likely to receive aggressive treatment or to get care from high-volume surgeons at high-volume hospitals, and more likely to die of prostate cancer—twice as likely, in fact, as white men. Crucially, African-American men also are less likely than other men to be screened in the first place.

Brian A. Stone, MD, who served on the P&S urology faculty until returning to his native Alabama in 2009, was a resident at Jacobi Hospital and Montefiore Medical Center in the Bronx when he started a free prostate cancer screening program in partnership with local black churches after seeing the alarming disparities in death and advanced disease between low-income African-American patients at Jacobi and their better-insured, often white counterparts at Montefiore. “I remain very concerned about the poor and the underserved, men who are much less likely to seek out the care of a physician for PSA,” says Dr. Stone. “You’ve already got a reluctant group of individuals who don’t want to do this. We have made so much progress in educating high-risk men since my screening program started in 1993. And I think the position the task force took saying that PSA was a useless test could cause harm in this population.”

As a member of the legislative committee of the American Urological Association, Dr. Stone is working with members of Congress on legislation that would require the task force to retract the recommendation pending further study and limit Medicare’s ability to deny coverage for prostate cancer screening and care. In the meantime, he continues to recommend the PSA to his patients. “You should educate men about the PSA blood test, the implications of getting the test done, the implications of finding a cancer that may potentially not be one that’s going to pose a threat to you, and then proceed with the test so patients work from a position of knowledge. The alternative is to put your head in the ground like an ostrich, then you’ll have patients showing up like they did before PSA.”

Amid the discussion about overdiagnosis and overtreatment, urologists at Columbia have put a priority on active surveillance, in which men with low-risk prostate cancers are monitored and rebiopsied annually for the first two to three years and then biopsied at increasing intervals and moved into treatment if their cancer appears to become more aggressive. Active surveillance serves two simultaneous purposes: It insists on diagnosis to establish risk but fights overtreatment while still protecting patients and preserving the possibility of cure if treatment is ultimately needed. “Treating without overtreating is exactly where active surveillance plays a very important role,” says Dr. Wenske. “Urologists need to learn how to interpret PSA values but also biopsy results in an intelligent way, look at the patient as a whole, then make an individual recommendation together with the patient about whether immediate treatment is necessary or whether it is safe for the patient to go on active surveillance.”

One of the department’s most interesting areas of current research is in confirmatory biopsies to decide eligibility for active surveillance among men who come to Columbia after being diagnosed elsewhere. “I have greater confidence in our ability to thoroughly biopsy the prostate,” says Dr. Benson, who performs biopsies with 24 to 30 cores rather than the more typical 12. To test the accuracy of initial biopsies at other facilities and consequent eligibility of patients for active surveillance, Dr. Benson selected 60 incoming patients and, before enrolling them in active surveillance, repeated their biopsies to be sure that the original tests hadn’t missed prostate cancer that might be of greater risk. 

The results: 30 percent of the men had a more aggressive or higher-volume cancer than had first been diagnosed and, therefore, were not good candidates for active surveillance; by contrast, only 5 percent of Dr. Benson’s own patients who underwent the 24-30 cores at initial diagnosis were failing surveillance at their first repeat biopsy. “The patients from the outside who were failing surveillance were not people whose cancer went from good to bad; they were people who always had bad cancer but it hadn’t been found at the time of their initial diagnosis,” says Dr. Benson. “The vast majority of people who fail surveillance never should have been on surveillance to begin with.” The department now has an IRB-approved protocol to give confirmatory biopsies to all patients in active surveillance before their planned 12-month repeat biopsies. 

Dr. Benson also stresses the importance of a multidisciplinary approach to detecting and accurately diagnosing prostate cancer, particularly with the use of advanced imaging. Columbia uses magnetic resonance imaging before each 12-month biopsy and often at the start of the process as well, several weeks after the initial biopsy. “We believe that the selection of appropriate therapy is multidisciplinary,” he says.” It involves expert pathologists, expert radiologists in prostate MR, and expert urologists. I wouldn’t say we’re unique, but we’re a leader in that.”

Advanced imaging, particularly functional imaging, can go far beyond pinpointing the location, volume, and extent of prostate cancer, says Lawrence Schwartz, MD, chair of radiology at P&S and a specialist in the imaging of pelvic malignancies. “With functional imaging techniques that look not only at anatomy but at biologic and cellular activity, we’re able to potentially judge things like the aggressiveness of a tumor. We’re able to evaluate the tumor micro-environment; we’re able to look at a tumor’s metabolism.” Also, he says, the scope of imaging offers an implicit advantage over biopsy. “In a biopsy, you’re sampling a small amount of the prostate. Imaging has the potential to be transformative in the sense that we can interrogate and evaluate the entire gland in a non-invasive manner.”

This level of collaboration in diagnosing prostate cancer was not viable as recently as five years ago. “The clinical implications of the findings of these advanced imaging modalities are just now becoming evident,” Dr. Schwartz says. “Expanding the clinically relevant role of functional imaging is going to be key, and this requires an intense multidisciplinary approach.”

One persistent challenge in prostate cancer treatment is that some men with low-risk cancer will decline active surveillance and insist on treatment because they can’t stand walking around with cancer inside their bodies. “Approximately a third of patients going off active surveillance do so because they are too anxious to pursue it. And they won’t necessarily go off it because of their own anxiety, but it could be family-member anxiety,” says Dr. Benson. “For that patient, therapy may be medically unnecessary but may be psychologically necessary in order to maintain a normal life. If you say to someone, ‘You have cancer, you need to have your prostate removed,’ they understand that, but if you say, ‘You have cancer, you don’t need to be treated,’ they don’t.”

In the treatment arena, several issues—for doctors, for patients, and for residents—have accompanied the rise of minimally invasive prostatectomy, or surgical removal of all or part of the prostate. While some 90 percent of prostatectomies are done robotically at Columbia—a rate only slightly higher than in the United States as a whole—Ketan Badani, MD, director of robotic and minimally invasive surgery at NewYork-Presbyterian/Columbia, stresses that the type of operation should not be the deciding factor for patients weighing surgical options and trying to avoid possible postsurgical side effects such as incontinence or impotence.

“If you’re comparing open and robotic prostatectomy and you’re looking at outcome, you have to realize that the No. 1 determinant of outcome is not the tool you use; it’s the experience you have doing it the way you’re doing it,” Dr. Badani says. “Because prostatectomy is such a technical operation, the more you do of a technical thing, the better you’re going to be at it.” He suggests that patients ask their surgeons not just what type of surgery they recommend, but how many surgeries of each type they have performed and their personal outcome experience. 

One concern that has dogged robotic surgery is its cost/benefit ratio, but Dr. Badani sees little conflict there. “If you just look at operating room costs, a robotic operation is more expensive than the open procedure, hands down,” he says. “But when you start looking at recovery time, faster return to normal activities, faster return to work, and you incorporate that, there are numerous studies recently that have shown that there’s actually a cost benefit to doing it robotically.”

As robotic surgery has become more popular and more widely available, concerns have arisen at medical schools about whether students and residents are receiving enough training in open procedures, to which they might have to convert at any time during robotic surgery. “The young urologists coming up now have become very adept at technology,” says Nicholas Romas’62, chair of urology at St. Luke’s-Roosevelt Hospital Center, a P&S affiliate. “But some patients have had multiple operations in the pelvic area, and the scar tissue is such that it’s impossible to use a robot or laparoscopy. Are the younger urologists adept enough to go in and do the surgery with open techniques? Who’s got the experience to do the open cases? The older generation is retiring, so what’s going to happen to the new generation coming forward? All of us are concerned about that.”

In fact, residents themselves are now concerned, says Dr. Badani. “A few years ago, when we were interviewing medical students for residencies, all the questions revolved around, ‘Do you do robotics?’” Dr. Badani recalls. “Now, the question is the opposite: It’s, ‘How much open do you do? How much open experience is there?’ The trainees are worried that they’re not going to get enough open experience. I think that they realize that most big places are doing enough robotic surgery, but maybe they’re not doing a lot of open—and that’s true.”

The issue of robotic surgery procedure volume also has implications for racial disparities in access to prostate cancer care. A recent study from the Mayo Clinic of nearly 30,000 men who had radical prostatectomy between 2006 and 2008 found that African-American and Latino men, as well as Medicaid patients, had lower odds of being treated at high-volume, high-experience hospitals offering robotic surgery than did their white or privately insured counterparts. 

Benjamin Spencer’94, assistant professor of urology at P&S, focuses his research on racial disparities in prostate cancer, and he notes that prostate cancer care could also benefit, as the breast cancer community has, from greater use of patient navigation, in which trained nonclinician navigators assist patients as they make their way through the logistics of cancer care—choosing providers and facilities, talking with insurers, making appointments, obtaining transportation, and getting information on diseases and treatment options. “Prostate cancer is a complicated disease for patients to understand, and I think patient navigation is an important area,” he says. “The whole concept of trying to facilitate patients through the system and getting timely consults and treatment—you definitely need it for a disease like this that often involves multiple modalities and providers. There are more and more options for treatment, so it’s going to get harder and more complicated for patients. I think that we’re just starting to try to bring resources together to help men through what are complicated and difficult decisions that have a huge impact on quantity and quality of life.”

Looking down the road, prostate cancer care is focused on pursuing both of the disease’s holy grails at once—making risk assessment more accurate and continuing to refine and improve surgical technique to further minimize side effects. Dr. Badani approaches the overdiagnosis controversy from a surgeon’s perspective. “The biggest concern is figuring out whom to treat and whom not to treat. That is what’s happening now, and I think that is what’s going to continue in the future,” he says. “But if you could do a prostatectomy that had no side effects and you treated the cancer—whether you think the cancer was a threat to your life or not—we would hear fewer arguments about overtreating.” 

For Dr. Benson, his current touchstones show the way forward. “There are two telling statistics,” he says. “One is that before PSA screening, the most common presentation was a patient with metastatic disease. And the second thing is that metastatic disease is now rare and the death rate has been reduced by 40 percent. What we have to do is find ways of continuing to have a death rate reduced by 40 percent while not treating people who don’t need treatment. That has to be the goal.” 

As this issue was in production, findings of new genetic variants that increase risk for prostate, breast, and ovarian cancers were announced. Two resources for the latest developments in prostate cancer research and care are Prostate Cancer InfoLink, a website co-founded by Arnon Krongrad’84  (, and the Prostate Cancer Foundation (