The debate about the usefulness of population-based prostate cancer screening has been ongoing for decades. The most current evidence shows limited benefit of population screening, and significant psychological, physical and financial costs. Efforts to improve screening with better markers and more selective treatment may make population-based screening more effective.
The debate about usefulness of population-based prostate cancer screening has been ongoing for decades, with little consensus among professional medical societies, cancer advocacy groups, or public health professionals. There is no doubt clinicians need to know whether screening is effective, as prostate cancer is associated with significant mortality and morbidity that will likely increase as the population ages.
Recent publication of 2 large-scale randomized studies addressing the issues of population screening for prostate cancer with level I evidence did little to resolve the controversy. The Prostate, Lung, Colorectal, Ovarian (PLCO) cancer screening trial and the European Randomized Study of Screening for Prostate Cancer (ERSPC) trial evaluated screening with prostate-specific antigen (PSA) and digital rectal examination (DRE). These studies demonstrate that PSA-based screening has led to a significant increase in diagnosis of prostate cancer with uncertain mortality benefits. Both studies have limitations that may have led to divergent results, thus leaving clinicians in a quandary regarding screening.
The burden of prostate cancer
In the United States, 192,280 men will be diagnosed with and 27,360 will die of prostate cancer in 2009. A man aged 40 years has a 3% lifetime chance of dying from prostate cancer and 16% chance of lifetime diagnosis of prostate cancer. As medical care and resultant life expectancy improve, the burden of prostate cancer morbidity and the likelihood of prostate-specific mortality are expected to increase because prostate cancer mortality peaks at around age 80 years ( Table 1 ).
| 35–44 Years | 45–54 Years | 55–64 Years | 65–74 Years | 75–84 Years | 85 Years and Older | |
|---|---|---|---|---|---|---|
| Deaths from prostate cancer (n = 28,905) | 24 | 395 (1%) | 2154 (7%) | 5764 (20%) | 11,666 (40%) | 8897 (31%) |
| Overall deaths (men and women, n = 2,448,017) | 84,785 (3%) | 183,530 (7%) | 275,301 (11%) | 398,355 (16%) | 686,665 (28%) | 703,169 (29%) |
In the pre-PSA era, approximately 11% of men were diagnosed with clinically symptomatic prostate cancer and 75% died of prostate cancer or with prostate cancer as a contributing cause. The median length from diagnosis to death was 41 months. Among men with clinically detected prostate cancer in the Scandinavian trial of radical prostatectomy (RP) versus watchful waiting (WW), 20% of men died of prostate cancer in the WW group and 26% developed metastasis at 12-year follow-up. Those with extracapsular disease had 14 times the risk of prostate cancer death compared with those with organ-confined tumors.
Whereas before the PSA era only 27% of prostate cancer cases were clinically localized, in the PSA era 97% to 98% of screening-detected prostate cancer is clinically localized.
Quality of life (QOL) among newly diagnosed men with prostate cancer versus men without prostate cancer is worse, as there are higher rates of depression, anxiety, and worse voiding and bowel function. Moreover, all treatment modalities result in temporary or permanent deterioration in QOL, particularly in the domains of sexual function, incontinence, bowel function, and vitality. Whereas side effect severity and frequency vary, all treatments affect the QOL of patients and their spouses.
Foregoing treatment in favor of WW is not optimal. QOL has been evaluated in the Scandinavian trial of RP versus WW. In that study, patients who had RP and who were alive at the time of evaluation had QOL and sense of well being that was better than that among men undergoing WW. Likewise, patients in the RP group had lower rates of anxiety and depressed mood despite the higher frequency of erectile dysfunction (80% vs 55%) and incontinence (42% vs 25%) 6 to 8 years after study enrollment. Men with bone metastases who were receiving androgen deprivation therapy (ADT) had a significantly reduced QOL, mood, and sense of well being if they were in the WW group.
These findings suggest that even in the absence of treatment, a diagnosis of prostate cancer is associated with significant morbidity. Several explanations for the inferior QOL scores in the WW arm of the study have been offered: (1) patients may regret not getting a potentially curative surgical intervention; (2) the expectations of significant surgical morbidity in the RP group might have lowered their threshold for QOL; (3) the WW group might have greater expectations of QOL than realistically possible. It is well known that expectations can alter perception. It is possible that counseling about expectations at the onset differed for the 2 groups of patients, and may have been the determining factor in poor QOL among the WW study arm if their expectations were higher than the QOL that they experienced.
Overall, these considerations reinforce the need for the urological community to do something about prostate cancer.
Why screen: rationale for screening
The goals of population screening for prostate cancer fall into 3 categories:
- 1.
Reduction of prostate cancer mortality
- 2.
Reduction of morbidity associated with prostate cancer
- 3.
Reduction of financial costs associated with symptomatic prostate cancer.
Although these goals seem self-evident, there are 3 biases worth considering. It is well accepted that population screening is associated with overdiagnosis, which is referred to as detection bias. Overdiagnosis refers to identification of disease in patients in whom it would have never become symptomatic during their lifetime. Screening can also lead to lead-time bias with earlier diagnosis but no effect on mortality. Lead-time bias gives the appearance of longer survival because of earlier detection with no overall improvement in life expectancy. Population screening is also more likely to detect less aggressive disease due to the longer interval before it becomes symptomatic. This factor has been referred to as length bias.
Why screen: rationale for screening
The goals of population screening for prostate cancer fall into 3 categories:
- 1.
Reduction of prostate cancer mortality
- 2.
Reduction of morbidity associated with prostate cancer
- 3.
Reduction of financial costs associated with symptomatic prostate cancer.
Although these goals seem self-evident, there are 3 biases worth considering. It is well accepted that population screening is associated with overdiagnosis, which is referred to as detection bias. Overdiagnosis refers to identification of disease in patients in whom it would have never become symptomatic during their lifetime. Screening can also lead to lead-time bias with earlier diagnosis but no effect on mortality. Lead-time bias gives the appearance of longer survival because of earlier detection with no overall improvement in life expectancy. Population screening is also more likely to detect less aggressive disease due to the longer interval before it becomes symptomatic. This factor has been referred to as length bias.
Lead times and overdiagnosis
With high discrepancy rate between clinical diagnosis of prostate cancer and histologic prevalence of prostate cancer, population-screening initiatives carry a significant risk of overdiagnosis of prostate cancer. Overdiagnosis refers to identification of latent disease that would not have otherwise caused symptoms or been identified during the patient’s lifetime.
In the case of prostate cancer screening, overdiagnosis of subclinical disease is often associated with significant overtreatment due to the current inability to distinguish between prostate cancers that have a high potential for symptomatic progression during a patient’s lifetime and those that are destined to have no clinical manifestations. Overtreatment leads to unnecessary costs to the health care system, and significant morbidity and possible mortality to some patients exposed to curative treatment.
Estimates for overdiagnosis of screen-detected prostate cancer in the United States range from 23% to 42%. Welch and Albertsen estimated that more than 1 million men were overdiagnosed by PSA-based screening between 1987 and 2005. Modeling of SEER data suggests that uncensored lead time ranges from 7.2 to 10 years whereas nonoverdiagnosed lead time ranges from 5.4 to 6.9 years. Based on reports of 11% prevalence of symptomatic prostate cancer by age 80, with 16% lifetime risk of diagnosis of prostate cancer for a man in the United States, a crude calculation suggests that the overdiagnosis rate of 45% is consistent with Draisma’s 42% overdiagnosis rate.
Prostate cancer screening tools
Digital Rectal Examination
The sensitivity of DRE for prostate cancer is dependent on the tumor stage and observer bias. At present, the sensitivity of DRE is poor and depends on the degree of DRE abnormality. DRE detects 15% of cancers that would go undetected by PSA screening with a threshold of 4 ng/mL. DRE is likely to detect higher risk tumors than PSA alone; 20% of tumors detected by DRE with PSA less than 2 ng/mL are nonorgan confined. Positive predictive value (PPV) for abnormal DRE can be as high as 50%. Men with abnormal DRE typically have more advanced disease, and cT2 disease is often pathologically upstaged to pT3 disease. Because DRE screening tends to detect nonorgan-confined disease undetected by PSA screening, the opportunity for cure with DRE testing alone is limited.
Prostate-Specific Antigen
The operating test characteristics of PSA depend on several factors, including the PSA threshold used for biopsy and patient age. For patients with mean age of 63 years, detection rate of prostate cancer when combined with DRE for PSA of 2.5 to 3.5 ng/mL is 28%; for PSA of 4 to 9.9 ng/mL it is 46%, and for PSA greater than 10 ng/mL, the detection rate is 60% with extended 12-core or saturation biopsy. Specificity improves at higher PSA thresholds while sensitivity declines significantly. Among men with mean age of 70, 46% had prostate cancer at PSA levels of 10 to 20 ng/mL, 76% had prostate cancer with PSA between 20 and 50 ng/mL, and 93% had prostate cancer with PSA greater than 50 ng/mL.
The Prostate Cancer Prevention Trial (PCPT) is unique in that men underwent sextant biopsy irrespective of their PSA level. Prostate cancer was found in 6.6% of men with PSA of less than 0.5 ng/mL (12% high grade), in 10% of men with PSA 0.6 to 1 ng/mL (10% high grade); in 17% of men with PSA 1.1 to 2 ng/mL (12% high grade); in 24% of men with PSA 2.1 to 3 ng/mL (19% high grade); and in 27% of men with PSA 3.1 to 4 ng/mL (25% high grade). PSA sensitivities and specificities vary by age, especially for PSA values less than 4.0 ng/mL ( Table 2 ). PSA sensitivity is better for high-grade disease (Gleason ≥7 vs <7 or no cancer) but even at PSA cutoff of 2.1 ng/mL, the sensitivity is only 76%, dropping to 57% at PSA 3.1 ng/mL or more (see Table 2 ).
| PSA (ng/mL) | Sensitivity <70 y | Specificity <70 y | Sensitivity >70 y | Specificity >70 y | Sensitivity Gleason ≥7 vs Gleason <7 or No Cancer | Specificity Gleason ≥7 vs Gleason <7 or No Cancer |
|---|---|---|---|---|---|---|
| 1.1 | 82.6 | 43.2 | 81.4 | 37.6 | 92.9 | 34.6 |
| 2.1 | 54.8 | 72.8 | 53.9 | 68.5 | 76.2 | 65.5 |
| 3.1 | 37.3 | 85 | 34.3 | 85.2 | 57.6 | 82.3 |
| 4.1 | 27.7 | 91.7 | 21.1 | 92.9 | 40.4 | 90 |
| 6.1 | 5.7 | 97.5 | 5.0 | 98.6 | 13.2 | 97.8 |
| 8.1 | 2.5 | 99.1 | 1.5 | 99.1 | 4.8 | 99 |
| 10.1 | 1.3 | 99.4 | 0.7 | 99.7 | 2.4 | 99.5 |
Related posts:
Chemoprevention of Prostate Cancer
Update on Prostate Imaging
Primary and Salvage Cryotherapy for Prostate Cancer
Neoadjuvant and Adjuvant Therapies in Prostate Cancer
Current Topics in the Treatment of Prostate Cancer with Low-Dose-Rate Brachytherapy
Immunotherapy for Prostate Cancer: An Emerging Treatment Modality
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
