More Aggressive Prostate Cancer in Elderly Men

More than one-fifth of men who are diagnosed with prostate cancer are age ≥ 75 years.¹ However, prostate cancer screening and treatment are both controversial in elderly men. In 2008, the US Preventive Services Task Force recommended against prostate-specific antigen (PSA)-based screening in men aged ≥ 75 years. This recommendation, which has been expanded to include all men,² generated controversy by focusing on chronologic age rather than health status.^{3, 4} More recently, the American Urological Association issued new guidelines stating that they do not recommend routine PSA screening in men aged > 70 years or those with less than a 10- to 15-year life expectancy, although some older men who are in excellent health may benefit from prostate cancer screening.⁵ This review describes two recent studies on the tumor features and outcomes of prostate cancer in elderly men.^6,7

Prostate Cancer in Men Aged 70 Years Old or Older, Indolent or Aggressive: Clinicopathological Analysis and Outcomes

Brassell SA, Rice KR, Parker PM, et al. J Urol. 2011;185:132–137.

Brassell and colleagues reported on the clinicopathologic features and the survival outcomes in 12,081 men diagnosed with prostate cancer between 1989 and 2009, from the Prostate Disease Research database. The study population was further stratified by age and race. Prediagnosis PSA velocity (PSAV) was defined as rapid for a change > 0.35 ng/mL/year at PSA < 4 ng/mL, or > 0.75 ng/mL/year at PSA > 4 ng/mL. Analysis of radical prostatectomy (RP) specimens was performed by the Armed Forces Institute of Pathology. Biochemical recurrence was defined as a PSA > 0.2 ng/mL after RP and as PSA nadir +2 ng/mL after radiotherapy. Mean overall follow-up was 6.3 years.

Men aged ≥ 70 years (n = 3350, 30.2%), had a significantly higher clinical stage and biopsy Gleason grade. Older men also had higher prediagnosis PSAV (P < .0001), which has previously been shown to be a marker for more aggressive prostate cancer.⁸ Among patients aged ≥ 70 years, 49.4% had external beam radiation therapy (EBRT), 24.6% had RP, 18.7% received primary hormonal therapy, 6% had brachytherapy, and 1.2% had cryotherapy. Among patients who underwent RP, pathologic stage, upgrading, and positive surgical margin rates were all significantly higher in older men.

Biochemical recurrence (BCR) occurred in 49.3% of men aged ≥ 70 years, compared with 44.2%, 42.1%, and 29.6% of men aged 60 to 69 years, 50 to 59 years, and < 50 years, respectively (P < .0001). On multivariate analysis age > 70 years was a significant predictor for BCR with hazard ratio 1.45 (confidence interval, 1.115– 1.873). Although there was no difference in PSA doubling time at recurrence by age, secondary treatment was significantly more likely in older men compared with younger men (11.6%, 20.9%, 20.8%, and 23.7% for age < 50 years, 50 to 59 years, 60 to 69 years, and > 70 years, respectively).

As expected, among men treated surgically, overall survival was lowest in men aged ≥ 70 years. However, Kaplan-Meier survival analysis showed improved survival for men aged ≥ 70 years who received RP or EBRT compared with expectant management.

Overall, Brassell and colleagues showed that aggressive disease was more common as men age, even in this heavily prescreened population. Proposed explanations for these findings include the natural progression of undiagnosed prostate cancer or changes in hormones with age⁹; however, the study did not provide data that could help elucidate the underlying mechanism. Another important limitation was that the study was retrospective, and treatment selection was not randomly assigned. As such, the results suggesting improved survival with definitive treatment in elderly men should be viewed with caution because there are numerous confounding factors. Despite these limitations, the study provides useful information from a large study of men in different age groups showing that elderly men quite frequently have clinically significant prostate cancer, and that some of these men would benefit from definitive therapy.

Prostate Cancer in the Elderly: Frequency of Advanced Disease at Presentation and Disease-specific Mortality

Scosyrev E, Messing EM, Mohile S, et al. Cancer. 2012;118:3062–3070.

Scosyrev and colleagues used the Surveillance, Epidemiology, and End Results database to determine the frequency of metastatic (M1) disease and prostate cancer death in different age groups. A total of 464,918 patients diagnosed with prostate cancer between 1998 and 2007 were categorized into 10 age groups ranging from < 50 years to ≥ 90 years. The largest group was aged 65 to 69 years (n = 87,568).

Information on comorbidities was not included in the analysis. Tumor features, including Gleason score and stage at presentation were recorded, and men were followed for a median of 4.5 years (range < 1–10 years). Expected survival was obtained from the general population, and the Gray method was used to assess the cumulative incidence of death from prostate cancer in different age groups. The frequency of metastatic prostate cancer ranged from 3% (group aged < 75 years) to 17% (group aged ≥ 90 years). Overall, men aged > 75 years represented 52% of all M1 cases.

Similarly, the 5-year cumulative incidence of death from prostate cancer increased from 3% to 4% at age < 75 years to 30% at age ≥ 90 years. Cumulative incidence of death from other causes also increased with age, from 9% in men aged 65 to 69 years up to 48% in men aged ≥ 90 years. Despite the greater risk of mortality from competing causes, men aged > 75 years still contributed to 47% of prostate cancer deaths. Although older patients did not lose as many years of life as younger patients, men aged > 75 years still lost approximately 75% of their remaining years of life (range, 74% to 78%). Hence, the proportion of remaining life lost in the elderly still was very high.

The authors suggested less frequent PSA testing and diagnostic evaluation as probable reasons for age-specific differences in the frequency of aggressive disease and prostate cancer mortality. For example, it is possible that biopsies were performed less frequently to evaluate PSA elevations among elderly men. However, they did not describe the prior screening or biopsy history, precluding the ability to examine the presence and impact of delays in diagnostic evaluation by age. Other limitations of this study include missing data on comorbidities and the use of androgen deprivation therapy, which could also contribute to survival differences.

Discussion

Several differences between the study by Brassell and the study by Scosyrev should be highlighted, including the sample size (12,081 vs 464,918), years of the study (1989 to 2009 vs 1998 to 2007), and age distribution (30.2% vs 44% who were aged ≥ 70 years). Despite these differences, both studies provide the consistent message that elderly men have a greater risk of diagnosis with aggressive prostate cancer.

Nevertheless, Level 1 evidence on routine use of PSA testing in the elderly is lacking because men aged > 75 years were not included in major screening trials.^10,11 According to the 2011 US National Vital Statistics, the average man is expected to live an additional 14.3 years at age 70, and 11.0 years at age 75.¹² Although overdiagnosis and overtreatment clearly can cause harm in this expanding population, our challenge is to also avoid undue morbidity and mortality from underdiagnosis and undertreatment of elderly men.