IJE Advance Access originally published online on April 23, 2007
International Journal of Epidemiology 2007 36(2):287-289; doi:10.1093/ije/dym056
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Commentary: The ubiquity of prostate cancer: echoes of the past, implications for the present
"What has been will be again, what has been done will be done again,; there is nothing new under the sun." ECCLESIASTES 1:9
1 Department of Urology, University of Texas Health Science Center at San Antonio, 7703 Floyd Curl Drive, San Antonio, TX 78229, USA.
2 Department of Pathology, University of Colorado Health Science Center, Denver, Colorado.
3 Fred Hutchinson Cancer Research Center, Seattle, WA, USA.
* Corresponding author. Department of Urology, University of Texas Health Science Center at San Antonio, 7703 Floyd Curl Drive, San Antonio, TX 78229, USA. E-mail: thompsoni{at}uthscsa.edu
Accepted 1 March 2007
"What has been will be again, what has been done will be done again,; there is nothing new under the sun."It was over 70 years ago that Rich made a simple observation: prostate cancer can be found surprisingly often in asymptomatic men who succumb to other diseases.1 The prostate cancer student of today is humbled by the prescience of Rich. The context is important: Rich was a member of the faculty of the home of the discipline of Urology in the USA at Johns Hopkins. Hugh Hampton Young, MD, the ‘father of American Urology’, was the Chair and had previously described treatments for the cure of prostate cancer. Young was the first to suggest that early detection (at the time, with digital rectal examination) could identify the disease prior to metastasis, allowing surgical intervention for cure.2 Rich's observation was startlingly simple: upon pathological examination of a very small sample of the prostate from men who had expired, most without a previous diagnosis of prostate cancer, the disease was commonly encountered. He observed as well that the disease developed on the periphery of the prostate and often invaded the prostate capsule; he furthermore implied that the disease might even be more common if it was sought more assiduously.ECCLESIASTES 1:9
Since Rich's landmark observation, others have looked carefully at the autopsy risk of prostate cancer. Prevalence rates of prostate cancer have been found to range from 31% for men aged 31–40, 44–46% for men aged 51–60, to over 80% for men aged 71–80.3 Despite the observations of Rich and subsequent authors, it has been generally felt that if a small sample of the prostate is obtained at biopsy and then examined pathologically, the disease found must be significant in size and clinically consequential. Bolstering this impression was the observation that even among men with a small focus of cancer in one biopsy core, over 90% were found to have what was called ‘significant’ disease at radical prostatectomy.4 As a result, when prostate cancer is detected, fearing it represents large-volume and potentially lethal disease, over 92% of men found with low-risk disease in the United States receive active treatment; of men over age 75, three-quarters are treated.5
Said the sage: ‘things change’. In the time of Rich as well as until the early 1980s, using digital rectal examination, if men were screened with the only available test—a rectal examination of the prostate, between 0.8% and 1.4% were found to have prostate cancer.6,7 Diagnosis of prostate cancer was established at that time by what might be considered a poorly representative sample of the gland: as few as two to four 18-gauge needle biopsies of the gland with the biopsy needle directed by the physician's finger either through the rectum or perineum. Using this technique, it was not possible to ensure that the needle sampled the peripheral zone of the gland where most tumours develop. What has changed? Regarding diagnosis, prostate specific antigen (PSA) is now the most common cause of biopsy. With its introduction for screening in the late 1980s, the rate of diagnosis of prostate cancer in the general population increased dramatically. Instead of 2–4 biopsy cores, the average number of cores obtained initially increased to six and currently is most commonly 10–12. If biopsy is negative, it is not unusual for a physician to recommend a ‘saturation’ biopsy with as many as 24 cores. Ultrasound is now used to ensure that the peripheral zone of the prostate is targeted with the biopsy needle.
The result of these changes has been dramatic: an 8% lifetime risk of diagnosis in 1985 is now 17%. This rate reflects the prevalence rate of screening with PSA: about 75% of men over 50 have had at least one PSA test and about 50% have the test regularly.8 A variety of factors may further increase the lifetime risk of diagnosis including, (i) lower PSA thresholds below the traditional 4.0 ng/ml cutoff for biopsy, (ii) an increasing fraction of the population screened, (iii) further aging of the population, (iv) new biomarkers for prostate cancer to increase screening sensitivity and (v) increases in the number of biopsy cores obtained.
The implications of these changes are enormous and were, to an extent, foretold by Rich. The first regards early detection of prostate cancer based on any ‘marker’ of the disease and biopsy sampling of the prostate. PSA as a biomarker actually has reasonable performance characteristics with an area under the receiver operating characteristic curve (AUC or C-statistic) of 0.68.9 The ‘problem’ with PSA is actually the disease prevalence. If only six biopsy samples are obtained, the most common cut-point of PSA, 4.0 ng/ml, has only 20.5% sensitivity. To achieve an 83.4% sensitivity, a threshold value of PSA of 1.1 ng/ml must be used. With this recognition and the patient's and physician's desires to not ‘miss’ prostate cancers, lower levels of PSA are increasingly used, increasing the likelihood that smaller tumours will be detected. Additionally, as a prostate biopsy incorporates an increasing number of biopsy samples, the risk of detection of these ‘autopsy’ tumours increases. This phenomenon has become manifest with the 17% lifetime risk of prostate cancer today, a value that has increased over the years since the introduction of PSA and which should be contrasted against the 3% lifetime risk of death from the disease. The end product of this dramatic increase in detection is the increasing risk of ‘overdetection’ of the disease. In this context, overdetection refers to the detection (and potential treatment) of prostate cancers that are destined to remain quiescent during the patient's lifetime. The implications of identifying one man in six in the U.S. with the disease in his life are profound including the costs (economic, social and psychological) as well as the potential side effects of treatment including sexual, urinary and bowel complications.
Rich's and others' observations of the high prevalence of prostate cancer and our subsequent proof that it can be detected clinically have had other implications for sciences related to prostate cancer risk and detection. Widespread efforts are ongoing to identify new biomarkers for prostate cancer.10 Probably the most common study design in these efforts is the case-control study. Risk factors or biomarkers are then compared between these two groups with the goal of improving the performance of current detection tests. Against the backdrop of Rich's observation, who should be selected as a control? Is a PSA <4.0 ng/ml sufficient? Clearly not, as 15% of these men have cancer if sampled with only six biopsy cores. Should a requirement be a negative biopsy? Perhaps, however, in samples of convenience, one must ask why the negative biopsy was performed in the first place: an ‘elevated’ or rising PSA or abnormal digital rectal examination? How does this selection skew the sample? (Are the investigators, for example, simply selecting subjects with large prostates by virtue of the higher PSA values?)11 Finally, it must be understood that a prostate biopsy is only a sampling of the prostate and that the somewhat random placement of the biopsy needle can allow relatively large or, even more concerning, high-grade tumours to be missed. Prior studies have shown that if a 6-core biopsy is simply repeated, it will increase the number of cancers detected by 30%.12
We should also reflect on what constitutes a ‘case’ in these studies. Recall that there is an approximate 3% lifetime risk of prostate cancer death. With only 50% of the US population having regular screening, the lifetime risk of diagnosis is now 17%. If all at-risk men were screened and lower PSA bounds, perhaps 2.5 ng/ml were used to recommend a biopsy, the lifetime risk would certainly increase, perhaps to rates of 25–30%. Should all these tumours be considered a case and therefore worthy of diagnosis? One would hope not. Ideally, the prostate cancer ‘case’ is a tumour that is destined to cause either morbidity or mortality. Our greatest challenge at this time is to reliably differentiate inconsequential and consequential tumours. The most commonly used measure is Gleason tumour grade; higher-grade (Gleason 7–10) tumours are generally felt to be almost always consequential while Gleason 6 and lower tumours are potentially inconsequential. Unfortunately, there are two challenges to this approach. First, as many as a third of prostate cancer deaths in one study occurred in men whose initial biopsy showed a Gleason 6 or lower grade.13 The second challenge relates to a focus on detecting only those high-grade tumours. If this is the goal, one presumes that a secondary goal is to then implement treatment sufficiently early to prevent the growth, spread and complications of the tumour. Unfortunately, among high-grade tumours, disease recurrence after treatment is seen in as many as 64–70% of subjects who are found with clinically organ-confined disease.14,15
The conundrum of the unpredictable natural history of prostate cancer combined with the ubiquity of the disease may seem almost unsolvable. Fortunately, ongoing efforts to understand the molecular signature of aggressive tumours, to image these lesions, and to potentially treat them with minimally invasive techniques hold tremendous promise. The use of array technologies in which an enormous range of features can be examined in prostate cancers that remained indolent over time vs those that exhibited disease progression should allow a more accurate discrimination of the ‘cases’ and ‘controls’ of the future. Imaging, heretofore unreliable, is rapidly taking advantage of new technologies including 3 Tesla endorectal coils potentially fused with magnetic resonance spectroscopy or even PET scanning.16 Finally, with validated, reliable imaging modalities, image-guided biopsy with molecular assessment of tumour prognosis can be followed by a range of ablative techniques in which only the tumour itself is destroyed, resulting in minimal impact on sexual, urinary and bowel function.
The challenges in prostate cancer are enormous and, as described by Rich in the increasing prevalence of disease with age, will continue to grow as the US population ages. To identify cancers, to then segregate unimportant from biologically aggressive tumours, and then implement effective treatments while maintaining a high quality of life will require the careful collaboration of scientific disciplines ranging from the most basic of sciences, to biostatistics, epidemiology, imaging, with ultimate validation in properly designed clinical trials.
 |
References |
|---|
 Top References |
|---|
1 Rich ARR. On the frequency of occurrence of occult carcinoma of the prostate. J Urol (1935) 33:215–23. Reprinted Int J Epidemiol.[Web of Science]
2 Young HH. The early diagnosis and radical cure of carcinoma of the prostate. Bull Johns Hopkins University (1905) 16:315–21.
3 Delongchamps NB, Singh A, Haas GP. The role of prevalence in the diagnosis of prostate cancer. Cancer Control (2006) 13:158–68.[Medline]
4 Lee AK, Doytchinova T, Chen MH, et al. Can the core length involved with prostate cancer identify clinically insignificant disease in low risk patients diagnosed on the basis of a single positive core? Urol Oncol (2003) 21:123–27.[Web of Science][Medline]
5 Cooperberg MR, Lubeck DP, Meng MV, Mehta SS, Carroll PR. The changing face of low-risk prostate cancer: trends in clinical presentation and primary management. J Clin Oncol (2004) 22:2141–49.
6 Thompson IM, Ernst JJ, Gangai MP, Spence CR. Adenocarcinoma of the prostate: results of routine urological screening. J Urol (1984) 132:690–92.[Web of Science][Medline]
7 Chodak GW, Schoenberg HW. Early detection of prostate cancer by routine screening. JAMA (1984) 252:3261–64.
8 Sirovich BE, Schwartz LM, Woloshin S. Screening men for prostate and colorectal cancer in the United States: does practice reflect the evidence? JAMA (2003) 289:1414.
9 Thompson IM, Ankerst DP, Chi C, et al. Operating characteristics of Prostate specific antigen in men with an initial PSA level of 3.0 ng/mL or lower. JAMA (2005) 294:66–70.
10 Semmes OJ, Feng Z, Adam BL, et al. Evaluation of serum protein profiling by surface-enhanced laser desorption/ionization time-of-flight mass spectrometry for teh detection of prostate cancer: I. Assessment of platform reproducibility. Clin Chem (2005) 51:102–12.
11 Roehrborn CG, McConnell J, Bonilla J, et al. Serum prostate specific antigen is a strong predictor of future prostate growth in men with benign prostatic hyperplasia. PROSCAR long-term efficacy and safety study. J Urol (2000) 163:13–20.[CrossRef][Web of Science][Medline]
12 Levine MA, Ittman M, Melamed J, Lepor H. Two consecutive sets of transrectal ultrasound guided sextant biopsies of the prostate for the detection of prostate cancer. J Urol (1998) 159:471–75.[CrossRef][Web of Science][Medline]
13 Thompson KE, Hernandez J, Canby-Hagino ED, Troyer D, Thompson IM. Prognostic features in men who died of prostate cancer. J Urol (2005) 174:553–56.[CrossRef][Web of Science][Medline]
14 Donohue JF, Bianco FJ, Kuroiwa K, et al. Poorly differentiated prostate cancer treated with radical prostatectomy: long-term outcome and incidence of pathological downgrading. J Urol (2006) 176:991–95.[CrossRef][Web of Science][Medline]
15 Do TM, Parker RG, Smith RB, Kagan AR. High-grade carcinoma of the prostate: a comparison of current local therapies. Urology (2001) 57:1121–26.[CrossRef][Web of Science][Medline]
16 Futterer JJ, Scheenen TW, Huisman HJ, et al. Initial experience of 3 tesla endorectal coil magnetic resonance imaging and 1H-spectroscopic imaging of the prostate. Invest Radiol (2004) 39:671–80.[CrossRef][Web of Science][Medline]
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  S. Ebrahim Prostate cancer, Cochrane, Durkheim and Beer Lao Int. J. Epidemiol., April 1, 2007; 36(2): 263 - 264. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||