Annals of Long Term Care

Key words: Prostate cancer, cancer screening, prostate-specific antigen testing, prostate cancer mortality, PLCO trial, ERSPC trial, informed decision-making, shared decision-making.

Despite significant advances in modern medicine, prostate cancer continues to be a disease that poses a tremendous health burden in the United States. In 2011, the American Cancer Society (ACS) projected that more than 240,000 new cases of prostate cancer would be diagnosed in the United States that year, with an expected mortality figure of approximately 34,000.¹ Further, the ACS estimated prostate cancer to be the second most common cause of cancer death in US men in 2011.¹ 

Prostate cancer screening using the prostate-specific antigen (PSA) test has been controversial and the subject of great debate in the United States and elsewhere for decades. Despite this long-standing controversy, PSA screening for prostate cancer continues to be commonplace in the United States.² In fact, PSA screening has been documented at rates close to 75%, even in men aged 75 years and older.^3,4 Conflicting results from two important randomized cancer screening trials—the US-based PLCO (Prostate, Lung, Colorectal and Ovarian) trial^5,6 and the European-based ERSPC (European Randomized Study of Screening for Prostate Cancer) trial^7,8—have further clouded the debate.

Although the prostate cancer screening debate is of tremendous importance to office-based, primary care physicians and their patients, this issue also has an impact on men residing in long-term care (LTC) settings. This article, presented in a question and answer format, serves to update LTC providers about important and current evidence from randomized trials regarding prostate cancer screening.

Can you remind us about the significance of both the PSA test and the PSA test result?

PSA is a glycoprotein produced exclusively by epithelial cells of the prostate gland. Typically, PSA levels in prostate tissue are low,⁹ but it is well known that conditions that lead to damage of the architecture of the prostate can increase PSA levels. As a result, prostate cancer is not the only condition that may lead to an elevation of PSA levels. Other such conditions include prostatitis, benign prostatic hyperplasia, and conditions resulting from trauma to the urethra or prostate.^9,10

Traditionally, PSA levels of ≥4.0 ng/mL have been considered elevated; however, some experts are now proposing even lower cutoff levels.¹¹ In fact, it is now known that a “normal” PSA level does not exclude a diagnosis of prostate cancer. As a result, several different approaches have been proposed to aid in improving the diagnostic accuracy of the PSA test. For example, PSA velocity and PSA volume tests have been advanced by some to improve PSA’s accuracy, but these and other similar proposals are unproven.¹¹

Is it true that prostate cancer mortality has decreased in the United States? If so, doesn’t that mean that PSA screening has been effective?

In 1994, the US Food and Drug Administration approved the PSA test for use in conjunction with a digital rectal examination (DRE) to help screen for prostate cancer in men aged 50 years and older.¹² Prostate cancer mortality in the United States has decreased since the early 1990s, and that decline has been most prominent in men aged 75 years and older^1,12; however, these mortality improvement statistics cannot necessarily be attributed to PSA screening. Some experts think that the decline in prostate cancer mortality in the United States may be explained by improvements in cancer treatment.^1,12,13 Other potential explanations include the possible early effects of PSA screening among men with more aggressive but asymptomatic prostate cancer and the cause of death being misattributed to conditions other than prostate cancer.^1,12

In addition, the US Preventive Services Task Force (USPSTF) has stated in its recently updated recommendations, after a comprehensive evidence-based analysis, that PSA screening results in little-to-no prostate cancer mortality benefit.^14,15 The USPSTF concluded that the known decade-long mortality reduction (1990 to 2000) is unlikely to be related to PSA screening.¹⁵ (See the next question and answer for more detailed information on the USPSTF’s position on the impact of PSA screening on prostate cancer mortality.) Moreover, a 2006 Cochrane collaboration report concluded, after a meta-analysis of five randomized controlled trials, that prostate cancer screening resulted in no significant mortality improvement.^16,17 BMJ (previously known as the British Medical Journal) published a randomized screening trial by Sandblom and colleagues¹⁸ in 2011 that also found no evidence of mortality benefit. Finally, a 2010 meta-analysis and systematic review of randomized controlled trials on prostate cancer screening failed to report any mortality improvement with PSA with or without DRE.¹⁹ This lack of mortality benefit was evident despite an increase in prostate cancer diagnoses in screened men.

What is the actual, official position of the USPSTF on prostate cancer screening, and what is the rationale for its position?

In October 2011, the USPSTF issued a draft recommendation against the use of the PSA test for prostate cancer screening. The analysis concluded, “Prostate-specific antigen–based screening results in small or no reduction in prostate cancer–specific mortality and is associated with harms related to subsequent evaluation and treatments, some of which may be unnecessary.”¹⁴ More recently, the USPSTF has formally issued an update to its 2008 prostate cancer screening recommendation, and has definitively adopted a position against PSA prostate cancer screening in men in the general US population, regardless of age, issuing a grade D recommendation (“discourage the use of this service”) to such screening efforts.¹⁵ The USPSTF’s rationale for this recommendation resulted from its analysis of questions relating to PSA’s effect on overall prostate cancer mortality, harms of PSA screening, early stage or PSA-screening–detected prostate cancer treatment benefits, and untoward treatment harms for early stage prostate cancer or prostate cancer found via PSA screening.¹⁵

After an extensive evidence-based analysis, the USPSTF rated the aforementioned PLCO trial and the ERSPC as “fair quality” trials. These two randomized trials yielded conflicting results.^2,5-8,15,20 The PLCO trial’s randomized design studied more than 77,000 men aged 55 to 74 years; patients were assigned to annual screening with PSA and DRE, and were compared with men with usual care. No improvement in either prostate cancer–specific or all-cause mortality was evident in the PLCO trial after a follow-up period of 7 to 10 years. There was an identified nonstatistical trend favoring a higher prostate cancer mortality in the screened group after 7 years (risk ratio, 1.14; confidence interval [CI], 0.75-1.70).^5,15 The recently published PLCO follow-up study evaluated mortality results after an extended 13 years of follow-up.⁶ No improvement in mortality was evident, even after the expanded follow-up period.

The ERSPC trial, a seven-country multinational trial, evaluated more than 180,000 men between the ages of 50 and 74 years via randomized assignment to regular PSA testing versus usual care.^7,8,21 The USPSTF concluded that, while the “prostate cancer incidence was higher in the screened group,” no “statistically significant difference in prostate cancer–specific mortality” was identified.¹⁴ However, a subgroup analysis of more than 160,000 men between the ages of 55 and 69 years showed an improvement in prostate cancer mortality in the screened group compared with the nonscreened group (rate ratio, 0.80; 95% CI, 0.65-0.98 after a median follow-up of 9 years). Similar ERSPC findings were reported after 11 years (rate ratio, 0.79; CI, 0.68-0.91).⁸ It is important to note, however, that only two of the seven countries in the ERSPC analysis exhibited mortality improvements that were statistically significant (Netherlands and Sweden) after 11 years of follow-up.¹⁵ Based on the 55-to-69–year subgroup analysis, close to 1410 men would have to be screened and 48 would need to be treated for one prostate cancer–specific death to be prevented.^7,14

The USPSTF stated that the overall prostate cancer mortality improvement reported in the ERSPC trial was very likely to be the result of the improvement seen only in Sweden and the Netherlands. In addition, none of the studies reviewed by the USPSTF revealed improvements in all-cause mortality.¹⁵ Other USPSTF concerns include the issues relating to overdiagnosis and overtreatment, described in greater detail below. Additionally, the USPSTF analysis pointed to the lack of consensus—and the difficulty of assessing efficacy—regarding treatment of localized prostate cancer.^15,22-25 Just as important, the USPSTF recommendation statement provides documentation of the real and potential harms of screening (see the question and answer on page 19).¹⁵ The USPSTF evidence-based analysis concluded that “the benefits of PSA-based screening for prostate cancer do not outweigh the harms” (Table 1).¹⁵

Even if a clinician agrees that routine PSA prostate cancer screening is unproven, does it not make intuitive sense to concentrate such screening efforts on populations deemed to be at high risk?

There are clearly identified risk factors for prostate cancer, including older age, a known family history of the disease, and African-American and African-Caribbean descent.^2,15,26,27 Exposure to Agent Orange (a combination of tactical herbicides that was sprayed by the US military to remove trees and dense tropical foliage providing enemy cover from 1962 to 1971 during the Vietnam War) is also thought to be a risk factor for prostate cancer.¹⁵ It has been well established that black men have the highest incidence of prostate cancer in the United States.²⁶

The black racial prostate cancer predominance has led some researchers to hypothesize that men of West African descent share either “genetic and/or lifestyle factors” that likely explain this elevated disease risk.²⁷ US data also indicate that black men exhibit a much higher mortality rate, which is at least twice that of other nonblack US men (Table 2).^10,28

Although higher prostate cancer detection rates and higher positive predictive values have been reported in higher-risk populations undergoing PSA screening for prostate cancer, there is no definitive evidence supporting the notion that such screening has resulted in mortality improvement from the disease.¹⁵ Clearly, there are insufficient published data on the question of PSA screening in higher-risk populations, and as a result, it is unknown whether targeting higher-risk groups for PSA screening will result in prostate cancer–specific mortality reduction.¹⁵ Moreover, early results from PIVOT (Prostate Cancer Intervention Versus Observation Trial) have not identified significant differences in prostate cancer treatment outcomes in black versus white men. A reported 30% of PIVOT study participants were black.^15,29

How do age, comorbidities, and remaining life expectancy influence decisions regarding prostate cancer screening?

The probability of developing prostate cancer is known to increase with age. In fact, older age is accepted as the strongest risk factor for the disease.¹⁵ Men aged 60 to 69 years have a 6.62% probability (1 in 15) of developing prostate cancer. That percentage increases to 12.6% in men aged 70 years and older (1 in 8 men).¹ In addition, prostate cancer exerts a significant burden of disease in older men. The median age of death from the disease was 80 years from 2003 to 2007, and a reported 71% of prostate cancer deaths occurred in men over age 75 years.^15,30 These statistics notwithstanding, the heterogeneous clinical presentation of prostate cancer indicates that most die with the disease, not because of it.^3,15 Specifically, autopsy data reveal that up to 75% of men older than 85 years of age have histologic evidence of prostate cancer, yet most of these cases are unlikely to be clinically significant.^15,31

Despite the contentious nature of the prostate cancer screening debate, there is consensus that men who are unlikely to live an additional 10 years because of advanced age and/or multimorbidities should not be screened.^2,15 Researchers agree that screening such men will lead to overdiagnosis (see the next question).^2,15 LTC providers should be aware that there are tools available to help estimate life expectancy based on age and the status of one’s health.^2,32 Various medical life expectancy calculators and tables can be found online, including the estimating prognosis for elders calculator, which was developed by geriatricians at the University of California at San Francisco.^33,34 Despite the near cross-organizational recommendation unanimity to avoid PSA prostate cancer screening in men with significant comorbidities, one Veterans Affairs study has documented PSA screening rates of 45% in men aged 80 years and older.^2,35

PSA screening in older men will undoubtedly result in increased prostate cancer detection rates.² Yet, many of these cancer diagnoses will not affect morbidity over the course of a lifetime.² LTC clinicians should be aware that prostate cancer screening and treatment trials have failed to demonstrate benefit for men older than 70 years.¹⁵

Even if clinicians and patients accept the contention that PSA prostate cancer screening may not improve mortality, are the potential harms of PSA screening really that significant?

An understanding of the concepts of overdiagnosis and overtreatment is crucial to the discussion of potential harms of prostate cancer screening.^2,19,36 Overdiagnosis occurs when patients receive a cancer diagnosis before the screening-detected cancer becomes clinically evident.^36-39 Overtreatment is linked to overdiagnosis, and refers to the treatment of screening-detected cancer in patients who would not ordinarily have received the diagnosis during their lifetimes.^2,36 Thus, overtreatment results in the treatment of indolent cancer, or cancer unlikely to be harmful.³⁶ Because it is well known that most men with prostate cancer actually die from other causes, overdiagnosis and overtreatment are problematic. Overdiagnosis estimates have been reported to be as high as 44% in blacks and 23% or higher in whites.^2,26,40,41

Abnormal PSA results often lead to prostate biopsies; such biopsies sometimes result in infection and bleeding, the two primary risks of this intervention. Prostate biopsies may also cause hematuria or urinary tract infections.^2,15,42 A significant—but often overlooked—biopsy risk is the potential that a biopsy may actually miss a patient’s prostate cancer. Anxiety is another frequently reported risk associated with PSA testing.²

Treatment of prostate cancer, either by surgical or radiation therapies, is also not without potential deleterious effects. Urinary incontinence and other urinary side effects are not uncommon. Sexual dysfunction may occur, even with nerve-sparing prostatectomy.^2,15,43 Additionally, radiation therapy may cause erectile dysfunction or bowel dysfunction.^2,15,26

Since there are so many unanswered and complex questions regarding PSA prostate cancer screening, how should LTC clinicians approach this important topic with their patients?

Despite the USPSTF’s recommendation, there are still discordant recommendations regarding PSA screening for prostate cancer.^2,15 As previously noted, screening is not without potential harms; thus, it should not be performed without the patient’s consent or understanding.^26,44-46 Because of this, some experts recommend that clinicians not undertake PSA screening without informed decision-making or shared decision-making (Table 3).^2,26,44-46 Informed decision-making should include a comprehensive discussion regarding the patient’s values and preferences, as it relates to screening. Braddock and colleagues⁴⁵ proposed distinct elements as criteria for informed decision-making. These elements include, but are not limited to, discussion regarding the patient’s role in medical decision-making, a comprehensive discussion of the stated clinical condition, discussion of the uncertainties involved in the decision-making process, and exploration of the preferences of the patient. 

Unfortunately, evidence suggests that informed and shared decision-making discussions regarding PSA screening are constrained by time and the controversy surrounding the issue of PSA screening.⁴⁴ It is recommended that prostate cancer screening “decision aids” be used to assist patients in making important decisions regarding screening. Decision aids are available in multimedia format, extending from print to web-based (Table 4). There is solid evidence supporting increased prostate cancer screening knowledge via the use of prostate cancer decision aids.^2,47 Because information on the PSA debate is now available in the lay media, patients are more aware of the issue and are more likely to bring it up during appointments; thus, there is an increasing need for clinicians to have a good understanding of the debate to facilitate discussion with their patients.