- Top of page
Background: The tendency of the results and quality control of prostate cancer screening serially performed for 10 years in an area of Japan were evaluated.
Methods: A total of 39 213 men over 55 years of age have participated in the mass screening of prostate cancer in the Otokuni District, since 1995. Men whose prostate-specific antigen (PSA) levels were more than 4.1 ng/mL were indicated for the second screening. In the second screening, prostate-specific antigen density (PSAD) was calculated in men whose PSA levels ranged from 4.1 to 10.0 ng/mL.
Results: Secondary screening was indicated in a total of 2428 subjects, of whom 1633 underwent it. Prostate cancer was diagnosed in 267 men. As a result of the evaluation of the indication of prostate biopsy according to the PSAD in 894 who underwent secondary screening for the first time, the procedure was judged to be unnecessary in 269 (35%) of 765 cases. Of these 269 subjects, 23 (8.5%) were found to have cancer. Clinically localized prostate cancer increased by 17%, and locally advanced and metastatic cancers decreased by 12% in the second compared with the first five years of the ten-year period. The exposure rate of PSA screening in the Otokuni District was 65% with the application for the rate of screenees whose PSA level was 4.1 ng/mL or above.
Conclusions: The Japanese basic health screening system allows the determination of high-PSA exposure areas. Serial prostate cancer screening showed a tendency of stage migration in the screened cancer patients. The use of PSAD in secondary screening substantially reduces the necessity of prostate biopsy; however, the encouragement of PSA-positive individuals to periodically receive prostate cancer screening is essential to maintain the quality of the screening system.
- Top of page
The incidence of and mortality from prostate cancer in Japan is still lower than it is in Western countries.1 However, prostate cancer is becoming a major public health concern in Japan. The age-adjusted incidence of this malignancy rapidly increased 6.5 times between 1975 and 1998.2 In addition, the age-adjusted mortality rate also increased 4.3 times between 1980 and 2000.
In conjunction with the recent rapid increases in the incidence of and mortality from prostate cancer in Japan, the percentage of local governments providing prostate cancer screening increased five times during the six years following 2000 (14.7% in 2000, 71.2% in 2006).3 This rapid increase in the number of local governments that started prostate cancer screening reflects the recent increase in concern over prostate cancer among the Japanese.
In Japan, the execution of prostate cancer screening is left primarily to local health administrative organizations, university and core hospitals, medical societies mostly consisting of local practitioners, and public or private screening institutions. A survey of such screening facilities in 2006 showed that 87% of them were local health administrative organizations.3 However, in executing prostate cancer screening in urban areas, the efforts of local health administrative organizations alone are insufficient to efficiently attract primary screenees, and the involvement of general practitioners, most of whom are family physicians, in primary screening is necessary. Also, the designation of core hospitals for secondary screening is necessary for physicians who perform prostate needle biopsy to be able to carry out the detailed management of databases. In Japan, basic health screening consisting of inquiries, body measurements, percussion and auscultation, sphygmomanometry, blood chemistry tests, diabetes tests, and electrocardiogram (ECG) is widely available for people aged 40 years and above as an elderly health protection measure for the promotion of the correct understanding of lifestyle-related diseases and their early detection and treatment.4 In 1995 in the Otokuni District of Kyoto, we first established a primary prostate cancer screening system in which screenees can freely choose between screening by local governmental administration and individual screening at private medical facilities (primarily local clinics) cooperating in basic health screening. The objectives of this study were to examine Japan's original health screening system and to clarify the characteristics of prostate cancer patients detected by screening using the PSA density (PSAD) as an indicational criterion for prostate biopsy during the past 10 years. In addition, we tried to calculate the exposure rate of PSA screening in the Otokuni District.
- Top of page
The Otokuni District is located to the south of Kyoto City and consists of two cities and one town (Nagaokakyo City, Muko City, and Oyamazaki Town). It has a population of 147 500 (2004), of which 22 705 (2004) are males aged 55 years and over. The subjects of this study were those who desired screening for prostate cancer among the males aged 55 years and over who have undergone basic health screening in September to October each year since 1995. Primary screening was made by the examination of the serum PSA level alone with a cut-off level of 4.0 ng/mL. The serum PSA level was determined using a Delfia PSA assay kit in all subjects.
The health administrative organization or private medical facilities in the Otokuni District informed the screenees in whom the serum PSA level was 4.1 ng/mL or above that they should receive secondary screening at a core hospital (Kyoto Saiseikai Hospital). The second screening including prostate biopsy was performed by urologists in Kyoto Saiseikai Hospital using the health insurance system. Table 1 shows the secondary screening system for the selection of candidates for prostate needle biopsy at the core hospital. Biopsy was indicated according to the PSAD, because it was reported to be promising as an indicator of prostate needle biopsy in individuals with a gray-zone PSA level (4.1–10.0 ng/mL) by a study team on the validity of mass screening for prostate cancer (Watanabe Team) under the Ministry of Health and Welfare (presently Ministry of Health, Labor and Welfare).5
Table 1. Prostate biopsy criteria of second screening
|4.1 ≤ PSA ≤ 10|
|– DRE-positive or TRUS-positive||Biopsy|
|– Both DRE and TRUS-negative|
| – PSAD > 0.15||Biopsy|
| – PSAD ≤ 0.15||Recommend to undergo screening next year|
|PSA ≥ 10.1||Biopsy|
Digital rectal examination (DRE) was performed by one urologist (K.K) who was a voting member of the Japanese Urological Association. Transrectal ultrasound sonography of the prostate was examined using an ultrasound machine equipped with a chair-type scanner (SSD-520, Aloca, Tokyo, Japan), and the prostate volume was obtained by the step-sectioned method. PSA density was calculated as PSA (ng/mL)/prostate volume (ml).6 The cut-off value was defined as 0.15.
In men who were indicated for prostate biopsy, transperineal prostate biopsy was undertaken under local anesthesia. The sextant systematic biopsy (SSB) technique was applied between 1995 and 2001. Since 2002, in addition to the SSB technique, an additional sample has been taken from the far lateral region in each lobe. The clinical stage was evaluated according to the TNM system.7
- Top of page
Figure 1 shows annual changes in the number of prostate cancer screenees. The number increased more than two-fold in 2004 compared with 1995, when prostate screening was started. In 1999, the number of screenees decreased, because only mass screening was performed; individual screening could not be performed because of the lack of cooperation by the local medical society.
A total of 39 213 people attended primary screening using PSA testing during the 10 years between 1995 and 2004. Of these screenees, 8420 (21%) underwent mass screening at health administration organizations in the Otokuni District, and 30 793 (79%) underwent individual screening at private medical facilities. The serum PSA concentration was 4.1 ng/mL or above in 2428 (6%) of all screenees. Of these screenees, 1633 (67%) received secondary screening, and prostate cancer was detected in 267. Of the 1633 secondary screenees, 1439 (88%) were examined at Kyoto Saiseikai Hospital regarding whether or not they should undergo prostate biopsy. Table 2 shows the percentage of the secondary screenees who underwent biopsy on the basis of their PSA level, the number of biopsies performed and the number of cancers detected (Table 2a: Initial screenees only. Table 2b: All secondary screenees). Of the screenees eventually diagnosed with cancer, 248 were diagnosed at Kyoto Saiseikai Hospital.
Table 2. Frequency of patients with indications of biopsy, number of biopsies performed, and number of cancers detected according to the PSA level in those who underwent secondary screening
|PSA(ng/ml)||No. of patients||Biopsy indicated||Biopsy not-indicated||Biopsied||Cancers|
|2-a First-time secondary screenees only|
|2-b All secondary screenees|
| ≤4.0||2||0 (0%)||2 (100%)||0 (0%)||0 (0%)|
| 4.1–10||1193||704 (59%)||489 (41%)||665 (56%)||151 (13%)|
| 10.1–||243||240 (99%)||3 (1%)||223 (92%)||97 (43%)|
| Lack of PSA||1||0||1 (100%)||0||0|
| Total||1439||944 (66%)||495 (34%)||888 (62%)||248 (17%)|
As previously described, the PSAD was calculated to determine the necessity of prostate biopsy. The serum PSA level was 4.1–10.0 ng/mL in 86% of the first-time screenees. Of these screenees, the PSAD > 0.15 or DRE was positive in 496 (65%), of whom prostate needle biopsy was performed in 482 (98%). In the screenees who underwent prostate needle biopsy, cancer was detected in 24% (118/487) and 52% (65/124) of those in whom the PSA level was 4.1–10.0 ng/mL and 10.1 ng/mL or higher, respectively. Of the 269 screenees in whom biopsy was not indicated based on the PSAD in the initial secondary screening (Fig. 2), 147 underwent secondary screening again, and prostate cancer was detected in 23.
Figure 2. Results of the follow-up of screenees in whom prostate needle biopsy was not conducted in secondary screening.
Download figure to PowerPoint
The serum PSA level was 4.1–10.0 ng/mL in 83% (1193/1439) of all who underwent secondary screening. Of these screenees, biopsy was indicated in 704 (59%) but not in 489 (41%), because no abnormality was noted during DRE, and the PSAD was 0.15 or less. Eventually, biopsy was performed in 665 (94%) of these 704 screenees, and prostate cancer was diagnosed in 151 (23%, 151/665). Of the screenees in whom the PSA level was 10.1 ng/mL or higher, prostate cancer was detected in 43% of those who underwent biopsy. Two screenees who showed a PSA level of 4.0 ng/mL or less underwent secondary screening with prostate needle biopsy, because their PSA level had been 4.1 ng/mL or higher in the past, but were thereafter excluded from prostate needle biopsy as no sign of malignancy was noted. Three screenees who showed a PSA level of 10.1 ng/mL or higher but were judged not to have an indication for biopsy had also undergone prostate needle biopsy in the past.
Table 3 shows the age distributions of all those who underwent secondary screening (n = 1439; 55–96 years; median, 71 years) and those in whom cancer was detected (n = 248; 55–92 years; median, 72 years) at Kyoto Saiseikai Hospital. In both groups, a peak was observed at 70–74 years, and 28% and 29% of the respective groups belonged to this age level. Table 4 shows the distribution of clinical stages in the screenees who were found to have cancer at Kyoto Saiseikai Hospital (n = 248). The disease in 193 (78%) of these patients was clinically localized prostate cancer (T1c-T2bN0M0). In particular, the percentage of patients with clinically T1cN0M0 cancer increased two-fold in the second five years compared with the first. The patients with locally advanced cancer (T3N0M0) numbered 35 (14%) and patients with metastatic cancer comprised 13 (5%) over the whole period. Although the ratio of clinically localized prostate cancer was 68% and the ratios of locally advanced cancer and metastatic cancer were both 27% from 1995 to 1999, the ratio of clinically localized prostate cancer between 2000 and 2004 increased by 15%, while that of locally advanced cancer and metastatic cancer within the same period decreased by 12%.
Table 3. Age distribution of those who underwent secondary screening (n = 1439) and those in whom cancer was detected (n = 248) at a core hospital
|Age||Second screening||Cancer (detection rate)|
|Mean ± SD||71.5 ± 7.0||72.5 ± 7.2|
|Median||71 years||72 years|
Table 4. Distribution of clinical stages in the five-year periods of 1995–1999 and 2000–2004
|T1cN0M0||23 (26%)||86 (54%)||109 (44%)|
|T2aN0M0||23 (26%)||31 (19%)||54 (22%)|
|T2bN0M0||14 (16%)||16 (10%)||30 (12%)|
|Sum of clinically localized cancer (T1c-2bN0M0)||60 (68%)||133 (83%)||193 (78%)|
|T3N0M0||17 (19%)||18 (11%)||35 (14%)|
|TxN1M0||0 (0%)||2 (1%)||2 (1%)|
|TxNxM1||7 (8%)||4 (3%)||11 (4%)|
|Unknown||4 (5%)||3 (2%)||7 (3%)|
In this study, we calculated the total number of those who underwent primary screening during the 10-year period but could not calculate the number of those who were screened for the first time, because there was no system to discriminate first-time screenees and repeaters between 1995 and 1998. Therefore, the cancer detection rate in the true number of primary screenees cannot be calculated. In the Gunma Prefecture, in which prostate cancer screening has been performed annually for more than 10 years, the cancer detection rate was 1.13% (440/38 861).8 To calculate the exposure rate of PSA screening in the Otokuni District, we tired to apply the detection rate of prostate cancer screening in the Gunma Prefecture. If the cancer detection rate in our study is assumed to have been similar to that in Gunma, the true number of primary screenees is estimated from the number of screenees in whom cancer was detected (267) to have been 23 628 (267/0.0113). Since the population of the Otokuni District aged 55 years and over was 22 705 in 2004, as mentioned above, all residents of the target population in this district are considered theoretically to have attended prostate cancer screening. This estimation may, of course, change with the method for the selection of candidates for secondary screening including the cut-off PSA levels for different age levels9 and methods of prostate needle biopsy. Consequently, we tried to calculate the exposure rate of PSA screening using another method. In this study, the PSA level is considered to have been 4.1 ng/mL or above in 6.1% of all primary and secondary screenees during the 10-year period combined, and this value, calculated from the total number of screenees, is considered to reflect the true number of people with an abnormal PSA level in the Otokuni District. Therefore, of the 22 705 males aged 55 years and over in the Otokuni District, the PSA level is considered to be abnormal in 1385 (22 705 × 0.06), since the number of those who were screened for the first time at the core hospital for secondary screening or other facilities was 903 men. In results, 65% (903/1385) of the target population in the Otokuni District is estimated to have undergone the PSA test.
- Top of page
Basic health screenings are undertaken by a higher percentage of the population than other screening systems, and the percentage in the Otokuni District, Kyoto Prefecture (54.6%, 1999) is higher than the national average (44.8%, 2003). The Otokuni District was designated as a prostate cancer screening area, primarily because PSA examination was successfully incorporated in basic health screening, and because the system of referral from other hospitals and general practitioners to the core hospital (clinic-hospital cooperation) has been matured to a functional level.
Concerning reports on prostate cancer screening in Japan sponsored basically by local administrative organizations, Kuwahara et al. reported that 2212 were screened in Natori, Miyagi Prefecture during a 7-year period,10 and Terai et al. reported that 1995 were screened in Okayama Prefecture during an 8-year period.11 In the Otokuni District, the number of residents who attended primary screening increased, because about 80% of the screenees were screened individually as a result of the incorporation of the PSA test in basic health screening. Recently, PSA has also been examined as a part of basic health screening in other areas of Japan.12
In promoting screening for prostate cancer in a particular area, the percentage of the population in the area previously exposed to the PSA test (exposure rate) must be estimated to set the target age level of the screening. We tried to calculate the exposure rate of PSA screening using two methods. Despite the difference between the above two values, the exposure rate in the Otokuni District is considered to be high as a value of annual screening in local municipalities compared with 5.3% in the Gunma Prefecture,8 and comparable to the value in the United States.13
To improve the detection rate of prostate cancer, the efficient selection of screenees for secondary screening including prostate needle biopsy is necessary. In the Otokuni District, people aged 55 years and over are screened. In Japan, the cancer detection rate in the population aged 50–54 years is 0.10%, which is lower than 0.25–2.55% in other age levels, but the evaluation of whether the screening age should be lowered to 50 years may become necessary in the future in consideration of the importance of early detection in younger patients.14
For the efficient selection of screenees for prostate needle biopsy, the use of the age-specific PSA reference,9 PSA velocity,15 and free/total PSA ratio16 as well as PSAD, which we are using, has been reported. Our evaluation was negative regarding the usefulness of the age-specific PSA reference for prostate cancer.17 However, Ito et al. reported that the age-specific PSA reference range cut-off value in this setting demonstrated a better diagnostic efficiency than the standard cut-off value of PSA and the age-specific PSA reference range determined by the 95% confidence interval.9 Annual calculation of the PSA velocity in screenees with an initial PSA level of 1.0–4.0 ng/mL has been reported to have improved the diagnostic accuracy of prostate cancer.15
While various factors have been proposed by different institutions for the proper selection of candidates for prostate biopsy, no conclusion has been reached as to which is the optimal parameter for the evaluation of the indication for the procedure. On the basis of the evidence shown by the Watanabe Team,5 we have evaluated the indication for biopsy according to the PSAD for 10 years. There has been no report on the use of the PSAD for mass screening, but, of the screenees undergoing secondary screening for the first time, prostate cancer was detected in 24% (118/487) and 52% (65/124) with PSA levels of 4.1–10.0 ng/mL and 10.1 ng/mL or higher, respectively. The cancer detection rate based on the PSA range was comparable to the average detection rate in Japan.
Of the 1436 screenees in whom the PAS level was 4.1 ng/mL or higher, biopsy was indicated in 944 (66%), and it could be circumvented in 492 (34%). Of the screenees who underwent secondary screening for the first time and were exempted from biopsy, because their PSA level was below the cut-off value, only 16% (23/147) were later diagnosed to have prostate cancer, and a considerable part of the screenees were repeatedly screened. An appropriate number of screenees should be selected for biopsy in consideration of the ability of the pathologists handling biopsies in the area, but this should not allow cancer to be overlooked. While the PSAD is useful for avoiding unnecessary biopsies, the fact that 45.3% (122/269) of the screenees did not undergo secondary screening thereafter suggests that the education of PSA-positive individuals to serially attend prostate cancer screening is necessary to maintain the reliability of the screening system.
Ito et al.18 detected prostate cancer in 440 screenees by prostate cancer screening in the Gunma Prefecture between 1992 and 2001, and reported that the median age of patients was 69–71 years and that the percentage of patients with T1c/T2N0M0 disease was 56.3–76.9% during those 10 years. The age of patients and the clinical stage of the disease were very close to our results. The European Randomized Study of Screening for Prostate Cancer (ERSPC) also reported that the stage of the disease in 84.4% of the cancers detected in 1269 patients by screening was T1c/T2N0M0, and prostate cancer was detected in an early stage by screening.19 In both the Otokuni District and the Gunma Prefecture,18 where screening has been performed annually, prostate cancer tends to be detected at a progressively earlier stage. The American Urological Association sets a life expectancy of 10 years or longer as a criterion for the screenee selection for prostate cancer, and, of the randomized controlled trials reported in the past, the Quebec Study20 set 80 years, and Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening21 set 74 years as the upper limit of screenees' age. Therefore, comparison of the age of patients at the time of detection of prostate cancer between our and the above studies is impossible.
Since the mean age of prostate cancer patients registered in Japan was 71.8 years at diagnosis,22 no young age migration was suggested in the prostate cancer patients detected by screening in Gunma or Otokuni. For the future, comparison of the prostate cancer mortality rate between the whole of Japan and the Otokuni District is indispensable to examine whether the high PSA exposure and early detection rates by this screening system are causing lead time bias.