Journal of the National Cancer Institute Advance Access originally published online on August 11, 2008
JNCI Journal of the National Cancer Institute 2008 100(16):1123-1125; doi:10.1093/jnci/djn259
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Published by Oxford University Press 2008.
EDITORIALS |
SPCG-4: A Needed START to PIVOTal Data to Promote and ProtecT Evidence-Based Prostate Cancer Care
Affiliation of author: Minneapolis VA Center for Chronic Disease Outcomes Research, Minneapolis, MN
Correspondence to: Timothy J. Wilt, MD, MPH, Minneapolis VA Center for Chronic Disease Outcomes Research, 1 Veterans Drive (111-0), Minneapolis, MN 55417 (e-mail: tim.wilt{at}va.gov).
Surgical removal of the prostate for cancer therapy was first performed in 1866 (1), although Young (2) is credited with the first radical surgical procedure in 1904. Prostate cancer deaths were rarely reported in the 1800s, but the relatively high frequency of asymptomatic disease was recognized as early as 1898 (3). Radical prostatectomy has been widely advocated as an effective treatment for localized prostate cancer. However, the only prior randomized trial comparing surgery to watchful waiting (4), begun 40 years ago, failed to demonstrate a survival difference even at 23 years of follow-up. Twenty years ago Whitmore (5) asked these questions regarding the natural history and treatment of prostate cancer: "Is cure necessary in those for whom it is possible and is cure possible in those for whom it is necessary?" Recent systematic reviews have been unable to address Dr Whitmore's questions, concluding there was little high-quality evidence on comparative treatment effectiveness and harms (6,7).
Uncertainty surrounding early treatment of localized prostate cancer has taken on greater importance today. Life expectancy for men exceeds 75 years and has increased by 20 years since the 1930s. The average life expectancy for a man aged 65 is 17 years (8). Prostate cancer is now the most common visceral malignancy and second leading cause of male cancer mortality (9). Widespread use of the prostate-specific antigen (PSA) blood test to detect prostate cancer is common even among men with very limited life expectancy (10). Physicians are diagnosing prostate cancer earlier and more often. Thresholds of PSA values that label an individual abnormal have been lowered (11). The number of tissue specimens obtained at biopsy and the frequency of repeating biopsies among men with initially negative biopsies have increased. Twice as many men are diagnosed with prostate cancer today, many with very low volume tumors, than in the pre–PSA era. Grade inflation, the tendency of pathologists to assign a higher grade now than in the past, has led to an upward shift in the percentage of tumors assigned an aggressive histologic score (12). Surgical and radiation techniques have been refined and new therapies developed and marketed. These factors have contributed to a markedly increased use of potentially effective but unproven early intervention options. In the United States, Medicare expenditures related to prostate cancer are $8 billion annually, and urologists perform 60 000 radical prostatectomy procedures (13). Radical prostatectomy is performed in approximately 60% of men with prostate cancer who are younger than age 75 (14).
A report in this issue of the Journal (15) helps answer some of the late Dr Whitmore's questions. Bill-Axelson et al. (15,16) provide long-term follow-up of a randomized trial comparing radical prostatectomy with watchful waiting in 695 men with clinically localized prostate cancer. The SPCG-4 study was of high methodological quality. It was large, long-term, and designed to assess prespecified clinically relevant outcomes. Investigators maximized treatment adherence, minimized loss to follow-up, and used intention-to-treat analyses. They rigorously assessed prostate cancer death and nonmortality outcomes.
Their primary findings indicated that radical prostatectomy reduced death from prostate cancer and distant metastases compared to watchful waiting by 5.4% and 6.7%, respectively, at 12 years (numbers needed to treat = 18 [95% confidence interval = 9 to 500] and 15 [95% confidence interval = 8 to 500]). Differences remained stable after 10 years. Nonmortality outcomes, including local recurrence and/or progression, institution of hormonal therapy, and use of other palliative therapies, also favored radical prostatectomy. Overall mortality was not statistically significantly different between groups (P = .09) but favored surgery (13.5% vs 19.5%). The authors assessed whether tumor and patient factors modified treatment outcomes. The presence of extracapsular tumor extension strongly increased the risk of prostate cancer death among men undergoing surgery, and only two prostate deaths occurred among 151 men whose prostatectomy specimens indicated no extracapsular disease. The effect of radical prostatectomy was surprisingly not modified by baseline PSA level or Gleason score. However, it was highly dependent on age (P for interaction on overall mortality = .015). Surgery improved overall mortality, disease-specific mortality, and distant metastases in men younger than 65 years but not in men aged 65 and older. Prior SPCG-4 findings indicated that erectile dysfunction and urinary leakage were more common after radical prostatectomy, but general well-being and functional status were similar between treatments at 4 years among a subgroup of respondents (17).
These results demonstrate that among men younger than 65 years whose prostate cancer is detected by methods other than PSA testing (eg, due to a digital rectal examination to evaluate urinary or other symptoms), cure with radical prostatectomy is possible, may be necessary, and should generally be recommended. Benefits were large in relative and absolute terms. Results are less certain for men older than 65 years or with limited life expectancies due to comorbidities. Age group findings were exploratory, driven by a statistical (median age of enrollees) rather than a biologic premise, require confirmation, and lack ready explanation. Reported values for tumor prognostic factors did not reveal large differences across age groups. The interaction between age and treatment did not change when the model was adjusted for PSA level, Gleason score, and tumor stage, indicating that these factors did not account for age-related findings. Age differences between the two subgroups were large (8.2 years). This difference likely contributes to differing competing risk of death or cause-of-death attribution. Older men are likely to have sociodemographic factors associated with an increased risk of death (eg, lower education and income and not being married) as well as more prevalent and severe comorbidities. The authors did not provide patient characteristics and comorbidities that may help explain outcomes. Interestingly, in men younger than 65 years, radical prostatectomy reduced nonprostate cancer mortality in addition to its large protective effect on disease-specific mortality. I do not have an explanation for this finding, but it was not observed in men aged 65 and older.
The SPCG-4 study has limitations. Enrollees may not be representative of US men with prostate cancer. Only 5% of SPCG-4 prostate cancers were detected by PSA testing. The mean PSA level (13 ng/mL) was higher than in contemporary US series. Tumor volume was greater than among men currently diagnosed in the United States. Three-quarters of SPCG-4 enrollees had T2 tumors, and extracapsular extension was noted in nearly half. It is unlikely that many men of black race were enrolled. I am not surprised that the risk of prostate cancer death among men having a radical prostatectomy was associated with extracapsular disease and Gleason score, as determined by examination of prostatectomy specimens. I am encouraged that only two deaths occurred in men without extracapsular disease. However, this does not prove that surgery is more effective than watchful waiting in this subgroup or among individuals with the small-volume organ-confined tumors that are commonly detected by PSA testing. The authors cannot identify an identical comparison group in the watchful waiting arm. The risk of prostate cancer death in men without extracapsular disease is likely low, regardless of treatment.
Despite these limitations, the SPCG-4 findings are immensely important. SPCG-4 is the first of a series of randomized trials currently evaluating primary treatments of clinically localized prostate cancer. Bill-Axelson et al. demonstrate the feasibility of conducting difficult but important studies. The SPCG-4 authors note that although the relative risk reduction they found was large, absolute risk reductions in men with PSA detected tumors may be small and require many years to accrue. The potential for overtreatment is real, and reliable tumor markers to predict comparative treatment effectiveness are still lacking. Unless future studies show differently, SPCG-4 also provides evidence against routinely recommending radical prostatectomy to reduce overall or disease-specific mortality or metastases in men older than 65 years. Because of overdiagnosis and long lead-time bias associated with PSA testing (approximately 12 years) (18), as well as histological grade inflation, PSA testing in men age 65 or older (and perhaps even in younger men) is unlikely to result in effective treatments, is associated with harms, and should be discouraged.
Ongoing treatment trials will complement SPCG-4. The US Prostate Cancer Intervention Versus Observation Trial (PIVOT) expands SPCG-4 findings to include PSA-detected tumors and African-American men. PIVOT began in 1992, is nearing completion, and will also compare radical prostatectomy to watchful waiting (n = 731). Three-quarters of PIVOT enrollees have tumors detected by PSA testing. Median PSA values are 9 ng/mL, and 45% of enrollees have T2 tumors. One-third are African-American, and 86% self-reported their baseline health status as good, very good, or excellent [(19); Timothy J. Wilt, unpublished]. Overall mortality is the primary outcome. Prostate cancer–specific mortality is adjudicated similar to the way it was in SPCG-4. PIVOT is assessing treatment-related adverse effects and quality of life.
SPCG-4 demonstrates that surgery may be effective and necessary in some but not most men with localized prostate cancer and that it causes treatment-related harms. Therefore, many patients and clinicians will wonder about options between the treatment extremes addressed by SPCG-4 and PIVOT. The Prostate Testing for Cancer and Treatment (ProtecT) and Standard Treatment Against Restricted Treatment (START) trials address a useful compromise in the therapeutic spectrum. Both studies are attempting to compare early interventions to active surveillance but use different inclusion criteria and intervention approaches. ProtecT, begun in the United Kingdom in 2001, will soon complete enrollment of more than 1500 men aged 50–69 (mean age = 61 years) with clinically localized prostate cancer detected through a PSA testing program (PSA level of 3–20 ng/mL and any histologic grade) ( Jenny Donovan, Freddie Hamdy, David Neal, personal communication). ProtecT is randomly assigning patients judged to be fit for any one of three treatment arms: conformal radiotherapy, prostatectomy, or active monitoring following a treatment plan decided by patients and their health professionals. The plan includes a PSA test and clinical examination every 3–6 months. Decisions to continue active monitoring, undergo restaging, or receive interventions are made jointly by patients and clinicians based on discussions incorporating PSA changes and clinical examination. Survival at 10 and 15 years is the primary outcome. ProtecT investigators have addressed patient and physician barriers to recruitment and devised a pragmatic simplified protocol. These efforts, combined with a greater acceptance of the scientific and financial equipoise within the UK across these three treatments, have likely contributed to their outstanding enrollment (20). The START trial is in its feasibility phase in Canada. Enrollment criteria are less inclusive, and the active surveillance protocol is more prescriptive, than those of ProtecT. The full trial will enroll 2130 men in Canada, the United States, and the United Kingdom with low-risk localized prostate cancer defined by PSA 10 ng/mL or less and Gleason score 6 or less. START will compare early interventions (surgery, external beam radiation, or brachytherapy) determined by patient or provider preference to active surveillance with delayed intervention based on biochemical, clinical, or patient or provider factors. Men who were randomly assigned to active surveillance receive PSA and digital rectal examination every 3 months for 2 years, then every 6 months for life. Biopsies are performed at years 1, 3, 4, 7, 10, and then every 5 years, with radiological exams as clinically indicated. Criteria for intervention include biochemical progression, PSA doubling time less than 3 years; histologic and/or grade progression, Gleason 7 (4 + 3); clinical progression, local progression, distant metastasis; patient choice; and clinical judgment. Disease-specific survival is the primary outcome (Laurence Klotz, personal communication). Estimates of effectiveness that preserve randomization will be between the overall tested strategies (active surveillance vs early intervention) rather than the comparative effectiveness of specific treatments. Cost–effectiveness analyses of the differing strategies should consider the number of early treatments avoided vs biopsies and other monitoring and treatment costs incurred with active surveillance.
In summary, SPCG-4 is the first of many randomized trials needed to accurately assess long-term outcomes of modern prostate cancer care. PIVOT will soon contribute additional information in African-American men and men with disease detected early in the PSA era. ProtecT and START have the potential to expand our knowledge to active surveillance regimens and early interventions that include external beam radiation and brachytherapy among younger men diagnosed later in the PSA era. Taken together, these studies and hopefully other large randomized trials of emerging technologies will provide information long lacking in localized prostate cancer. SPCG-4 has not yielded all the answers to Dr Whitmore's questions. However, after more than 100 years of uncertainty, SPCG-4 provides a needed START to PIVOTal data to promote and ProtecT evidence-based prostate cancer care.
REFERENCES
1. Kuchler J. Uber Prostatavergroserungen. Deutsch Klin (1866) 18:458–465.
2. Young HH. The early diagnosis and radical cure of carcinoma of the prostate. Being a study of 40 cases and presentation of a radical operation which was carried out in 4 cases. Johns Hopkins Hosp Bull (1905) 16:315.
3. Herbst RH, Polkey HJ. Prostatic malignancy. In: History of Urology. Vol. II—Ballenger EG, Frontz WA, Hamer HG, et al, eds. (1933) Baltimore, MD: Williams and Wilkins. 187–208.
4. Iversen P, Madsen PO, Corle DK. Radical prostatectomy versus expectant treatment for early carcinoma of the prostate. Twenty-three year follow-up of a prospective randomized study. Scan J Urol Nephrol Suppl (1995) 172:65–72.[Medline]
5. Whitmore WF. Overview: historical and contemporary. NCI Monogr (1988) 7:7–11.[Medline]
6. Thompson I, Thrasher JB, Aus G, et al. Guideline for the Management of Clinically Localized Prostate Cancer: 2007 Update. (2007) American Urological Association Education and Research. http://auanet.org/guidelines Accessed June 18, 2008.
7. Wilt TJ, Shamliyan, Taylor B, MacDonald R, Tacklind J, Rutks I, Kane RL. Systematic review: comparative effectiveness of treatments for clinically localized prostate cancer. Ann Intern Med (2008) 148(6):435–448.
8. US Department of Health and Human Services Center for Diseases Control and Prevention. United States life tables, 2004. http://www.cdc.gov/nchs/agingact.htm, http://209.217.72.34/aging/TableViewer/tableView.aspx?ReportId=438.
9. Jemal A, Siegel R, Ward E, et al. Cancer Statitistics, 2008. CA Cancer J Clin (2008) 58(1):71–96.
10. Walter LC. PSA screening among elderly men with limited life expectancies. JAMA (2006) 296(19):2336–2342.
11. Welch HG, Schwartz LM, Woloshin S. Prostate specific antigen levels in the United States: implications of various definitions for abnormal. J Natl Cancer Inst (2005) 97(15):1132–1137.
12. Albertsen PC, Hanley JA, Barrows GH, et al. Prostate cancer and the Will Rogers phenomenon. J Natl Cancer Inst (2005) 97(17):1248–1253.
13. Healthcare Cost and Utilization Project (U.S.). United States Agency for Healthcare Research and Quality. http://www.ahrq.gov/data/hcup/Accessed June 16, 2008.
14. Konety BR. Patterns of primary and secondary therapy for prostate cancer in elderly men: analysis of data from CaPSURE. J Urol (2008) 179(5):1797–1803.[CrossRef][ISI][Medline]
15. Bill-Axelson A, Holmberg L, Filen F, et al. Radical prostatectomy versus watchful waiting in localized prostate cancer: The Scandinavian Prostate Cancer Group-4 randomized trial. J Natl Cancer Inst (2008) 100(16).
16. Bill-Axelson A, Holmberg L, Ruutu M, et al. Radical prostatectomy versus watchful waiting in early prostate cancer. N Engl J Med (2005) 352(19):1977–1984.
17. Steineck G, Helgesen F, Adolfsson J, Dickman PW, Johannson J-E, Norlen BJ. Quality of life after radical prostatectomy or watchful waiting. N Engl J Med (2002) 347(11):790–796.
18. Draisma G, Boer R, Otto SJ, et al. Lead times and overdetection due to prostate-specific antigen screening: estimates from the Euorpean randomized study of screening for prostate cancer. J Natl Cancer Inst (2003) 95(12):868–878.
19. Wilt TJ, Brawer MK. The prostate cancer intervention versus observation trial (PIVOT). Oncology (1997) 11(8):1133–1139.[Medline]
20. Donovan JL, Hamdy F, Neal D, et al. Prostate testing for cancer and treatment (ProtecT) feasibility study. Health Technol Assess (2003) 7:1–39.[Medline]
Related Articles in JNCI
CiteULike 
Connotea 
Del.icio.us What's this?
J Natl Cancer Inst 2008 100: 1144-1154.
J Natl Cancer Inst 2008 100: 1119.
J Natl Cancer Inst 2008 100: 1119.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||