SEARCH

SEARCH BY CITATION

Keywords:

  • PSA;
  • screening interval;
  • prostate cancer

Abstract

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONCLUSIONS
  7. REFERENCES

Routine screening for prostate cancer remains controversial. However, it is very important to show how the optimal rescreening interval should be set for men who want to be screened after informed consent. To solve this issue, the risk of prostate-specific antigen (PSA) increase above 4.0 ng/ml relative to baseline PSA levels and age was investigated. Between 1988 and 2000, 7,757 subjects screened twice or more and also with baseline PSA levels of 4.0 ng/ml or lower were enrolled in our study. All serum PSA levels were measured by E-test Tosoh II PA assay at one center. Interval PSA levels for men undergoing screening with a greater than 1 year interval were calculated on the assumption that PSA levels changed over time in a simple exponential fashion. Then, the cumulative rate of freedom from PSA increase above 4.0 ng/ml was estimated using the Kaplan-Meier technique stratified by baseline PSA ranges of 0.0 to 1.0, 1.1 to 2.0, 2.1 to 3.0 and 3.1 to 4.0 ng/ml and every 10 years of age ranges. Of the 7,757 subjects, 559 (7.2%) were expected to have had PSA levels increase above 4.0 ng/ml within 5 years after the baseline PSA measurements. The cumulative rate of freedom from the PSA increase above 4.0 ng/ml at 5 years was 98.7%, 92.9%, 70.3% and 38.5% in cases of baseline PSA levels of 1.0 ng/ml or lower, 1.1 to 2.0 ng/ml, 2.1 to 3.0 ng/ml and 3.1 to 4.0 ng/ml, respectively. The cumulative rates of freedom from the PSA increase were significantly decreased with the baseline PSA ranges being higher regardless of age range. Re-screening interval should be set stratified by baseline PSA levels, regardless of age and race. Rescreening interval should be set at 1, 1 to 2 and 3 to 5 years for men with baseline PSA ranges of 2.1 to 4.0 ng/ml, 1.1 to 2.0 ng/ml and 0.0 to 1.0 ng/ml, respectively, in individual-based screening. In mass screening system using PSA alone, rescreening interval should be set in the same manner as in individual-based screening, except for men with baseline PSA levels of 1.1 to 2.0 ng/ml, which should be set at 1 year to avoid developing incurable prostate cancer. © 2004 Wiley-Liss, Inc.

As the most frequently diagnosed cancer and the second leading cause of cancer death in most Western countries, prostate cancer represents a significant healthcare problem. One of the best hopes to decrease mortality for prostate cancer has resulted in the wide use of screening regimens based on prostate-specific antigen (PSA) for asymptomatic men. However, screening and management of subsequently diagnosed prostate cancer can be harmful to individuals. In this situation, there is obvious need to provide conclusive information prior to the application of screening tests to men who hope to have information on screening.1 At the same time, it is very important to propose an optimal screening program for well-informed participants on screening.

It is widely known that the probability of prostate cancer is strongly related to the serum PSA levels.2, 3 Therefore, it would be very important to investigate the risk of PSA increase in men with baseline PSA levels of 4.0 ng/ml or lower. In the present study, we investigated the cumulative risk of PSA increase not only relative to baseline PSA levels but also relative to age ranges in order to propose how and when the optimal rescreening interval should be set.

MATERIAL AND METHODS

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONCLUSIONS
  7. REFERENCES

A population-based study of screening for prostate cancer in Gunma Prefecture has been performed since 1981. Between 1981 and 1991, digital rectal examination (DRE) and prostatic acid phosphatase (PAP) were used as screening modalities. After 1992, PSA-based screening has been carried out for all participants. DRE and transrectal ultrasonography (TRUS) were also performed as ancillary tests in most municipalities until 1994. Then, the number of municipalities that conducted DRE and TRUS has decreased each year in the past 5 years because a PSA test had usually been conducted in conjunction with a basic health check-up system using a blood test only, which is conducted by local government and is also spread throughout Japan.

Between 1992 and 2000, 13,021 men, 50 to 79 years old, had PSA levels measured in population-based screening for prostate cancer in Gunma prefecture, Japan. Of these, 12,058 (92.6%) had initial PSA levels of 4.0 ng/ml or less. In addition, 2,447 men in the same age range underwent initial screening using digital rectal examination (DRE) and prostatic acid phosphatase between 1988 and 1991, and had PSA levels of 4.0 ng/ml or less by retrospective PSA measurements using frozen serum (−70°C). Of these 14,505 men with initial PSA levels of 4.0 ng/ml or less, 8,012 (55.2%) had not been rescreened until December 2000 and could not been followed in our population-based screening system. The remaining 6,493 (44.8%), who underwent screening twice or more, were enrolled in our study. The age range at initial screening was from 50 to 78 years old (63.5±6.4; mean±S.D.), and the number of screenings ranged from 2 to 13 (mean; 3.6) during 1 to 14 (mean; 4.1) years of observations.

All serum PSA levels were measured using E-test Tosoh II assay with the AIA-600 machine (Tosoh, Tokyo, Japan) at one center (the Department of Urology, Gunma University School of Medicine). The screening for prostate cancer has been performed by local governments, and subjects were invited by letter or by announcement, which included the fact sheet on screening for prostate cancer from the public health care center, and participated in the screening study based on this advertisement.

Figure 1 shows the number of participants enrolled at initial screening and the number of cases with PSA changes from the lower to higher range during observations. Baseline PSA levels were subdivided into 4 groups: 0.0–1.0 ng/ml, 1.1–2.0 ng/ml, 2.1–3.0 ng/ml and 3.1–4.0 ng/ml. Then, baseline PSA was defined as initial PSA levels or PSA levels increased to higher PSA ranges for the first time. For example, when PSA levels have changed in the order of 0.2, 1.2 and 0.5 ng/ml, the first and second PSAs of 0.2 and 1.2 ng/ml are eligible for baseline PSAs, but the third PSA of 0.5 ng/ml is not. Individuals whose PSA levels had changed gradually from lower to higher PSA ranges below 4.0 ng/ml were considered as a different risk set. Therefore, 7,757 cases were enrolled for estimating cumulative risk of PSA increase above 4.0 ng/ml stratified by baseline PSA and age ranges (Fig. 1).

thumbnail image

Figure 1. The number of cases at initial screening in each PSA category, changes in the PSA levels during observation periods and the number of men with PSA levels of 4.1 ng/ml or greater at actual screening.

Download figure to PowerPoint

For cases screened with a rescreening interval of greater than 1 year, interval PSA levels were estimated by yearly units on the assumption that PSA levels had changed over time in a simple exponential fashion. When a case had the most recent PSA level (PSAmost rec.) measured at x years after the baseline PSA (PSAbase) measurement, the interval PSA level (PSAint.) at y year(s) after the PSAbase measurement could be calculated using the following formula:

  • equation image

For estimating cumulative rate of freedom from PSA increase, a time to PSA increase above 4.0 ng/ml was defined as the time between baseline PSA measurement and actual PSA or PSAint. level increased to 4.1 ng/ml or greater for the first time.

PSA data of men 80 years or older in consecutive screening were excluded from our study. The end point of our study was set at the time of PSA increase above 4.0 ng/ml for the first time. PSA distributions were compared by chi-square test. Cumulative rates of freedom from PSA increase were estimated by Kaplan-Meier techniques and were compared using 2-sided log-rank tests. Differences were considered significant when p was <0.05.

RESULTS

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONCLUSIONS
  7. REFERENCES

The detailed number of cases with PSA levels of 4.1 or greater is shown in Figure 1. Of the 6,493 subjects who participated in our study, 469 (7.2%) had PSA levels increased to 4.1 ng/ml or greater. Among the 7,757 in the risk group divided by baseline PSA levels, 667 (8.6%) showed PSA levels had increased above 4.0 ng/ml at the actual screening. Of these 667 cases with the PSA increase, 133, 109, 113, 85, 70 and 157 showed PSA level increases above 4.0 ng/ml at 1, 2, 3, 4, 5 and 6 or more years after the baseline PSA measurements, respectively.

The change in the median and maximum PSA levels during 5 years, stratified by age range and baseline PSA levels, are shown in Table I. The median PSA levels were similar in the same baseline PSA range throughout the observation periods for 5 years. Table I also shows the percentage of cases with PSA levels greater than 4.0 ng/ml during 5 years of observations. In men with baseline PSA levels of 0.0 to 1.0 ng/ml, the percentage of cases with PSA levels greater than 4.0 ng/ml in all age ranges were between 0.0% and 0.7%. The percentage of cases with PSA levels greater than 4.0 ng/ml did not increase with time (p>0.05, χ2 test). In men with baseline PSA levels of 1.1 to 2.0 ng/ml, the percentage of cases with PSA levels greater than 4.0 ng/ml in all age ranges were between 0.6% and 3.9%. There was a significant relation between the percentage of cases with PSA levels greater than 4.0 ng/ml and the elapsed years in the same age range (p<0.05, χ2 test). In men with baseline PSA levels of 2.1 to 3.0 ng/ml, the percentage of cases with PSA levels greater than 4.0 ng/ml in all age ranges was relatively high between 4.0% and 12.8% and increased with time. There was a significant relation between the percentage of cases with PSA levels greater than 4.0 ng/ml and the elapsed time in the same age range (p<0.05, χ2 test). In men with baseline PSA levels of 3.1 to 4.0 ng/ml, the percentage of cases with PSA levels greater than 4.0 ng/ml in all age ranges were high between 20.5% and 32.2%. There was no significant relation between the elapsed time and the percentage of cases with PSA levels greater than 4.0 ng/ml during 5 years of observation (p>0.05, χ2 test).

Table I. The Median, Maximum PSA Levels and the Percentage of Cases With PSA Levels Greater Than 4.0 NG/ML at Actual Screening Stratified by Age, Baseline PSA Levels and Years Elapsed After the Baseline PSA Measurements
Years elapsedAge range (years old)Baseline PSA range (ng/ml)
0.0–1.01.1–2.02.1–3.03.1–4.0
PSA levels (ng/ml)Percent cases with PSA >4 ng/mlPSA levels (ng/ml)Percent cases with PSA >4 ng/mlPSA levels (ng/ml)Percent cases with PSA >4 ng/mlPSA levels (ng/ml)Percent % cases with PSA >4 ng/ml
MedianMaximumMedianMaximumMedianMaximumMedianMaximum
150–590.710.00.2%1.36.10.4%2.27.52.6%3.35.415.4%
 60–690.77.40.3%1.49.51.7%2.312.74.8%3.214.820.1%
 70–790.77.30.2%1.46.90.3%2.35.63.0%3.16.222.2%
 All0.710.00.3%1.39.51.1%2.312.74.0%3.214.820.5%
250–590.713.80.9%1.45.20.5%2.33.70.0%2.95.428.6%
 60–690.712.40.6%1.411.70.5%2.57.07.0%3.315.525.4%
 70–790.73.90.0%1.44.71.1%2.35.24.7%3.439.127.5%
 All0.713.80.6%1.411.70.6%2.47.05.6%3.339.126.4%
350–590.72.90.0%1.43.70.0%2.55.212.2%2.77.525.0%
 60–690.814.80.5%1.515.32.1%2.539.112.8%3.512.132.2%
 70–790.73.90.0%1.46.41.7%2.66.610.7%3.57.734.1%
 All0.714.80.3%1.415.31.7%2.539.112.3%3.512.132.2%
450–590.73.20.0%1.45.81.6%2.56.414.8%2.45.712.5%
 60–690.73.90.0%1.539.13.6%2.614.412.9%3.56.426.7%
 70–790.72.80.0%1.61,9281.6%2.78.511.1%3.29.632.1%
 All0.73.90.0%1.51,9283.0%2.614.412.7%3.39.627.1%
550–590.74.80.4%1.46.73.8%2.35.78.3%2.33.50.0%
 60–690.810.40.7%1.613.33.1%2.71,28910.4%2.8150.331.1%
 70–790.722.70.8%1.56.78.6%2.78.123.7%3.38.123.1%
 All0.722.70.7%1.513.33.9%2.71,28912.8%2.8150.328.3%

After calculating interval PSA levels using the exponential model, 559 (7.2 %) were expected to have had PSA levels increased above 4.0 ng/ml within 5 years after the baseline PSA measurements. Of these 559 cases, 206, 147, 107, 54 and 45 cases were expected to have a PSA increase at 1, 2, 3, 4 and 5 years after the baseline PSA measurements, respectively.

The cumulative rates of freedom from PSA increase relative to age and baseline PSA ranges during 5 years of observations are shown in Table II. The overall cumulative rate of freedom from the PSA increase was 89.5% at 5 years after the baseline PSA measurements. The cumulative rates of freedom from the PSA increase in all baseline PSA range of 0.0 to 4.0 ng/ml were significantly different among the age ranges of 50 to 59, 60 to 69 and 70 to 79 years. The cumulative rates of freedom from the PSA increase in all age ranges were significantly decreased with baseline PSA ranges being higher, which were 98.7%, 92.9%, 70.3% and 38.5% in men with baseline PSA levels of 0.0 to 1.0 ng/ml, 1.1 to 2.0 ng/ml, 2.1 to 3.0 ng/ml and 3.1 to 4.0 ng/ml, respectively. The cumulative rates of freedom from PSA increase within the same baseline PSA range were not significantly different among the age ranges of 50 to 59, 60 to 69 and 70 to 79 years, except one subgroup with the baseline PSA range of 1.1 to 2.0 ng/ml and age range of 50 to 59 years. On the other hand, the cumulative rates of freedom from PSA increase within the same age range were significantly different among the baseline PSA ranges of 0.0 to 1.0 ng/ml, 1.1 to 2.0 ng/ml, 2.1 to 3.0 ng/ml and 3.1 to 4.0 ng/ml.

Table II. The Cumulative Rates of Freedom from PSA Increase Stratified by Age Range and Baseline PSA Levels
Age range (years old)Baseline PSA range (ng/ml)Number of cases at riskNumber of event observedCumulative rate of freedom from PSA increase above 4.0 ng/ml (standard error)Statistical significance
12345
  1. PSA: prostate specific antigen, Age (50–59): age range of 50–59 years, Age (60–69); age range of 60–69 years. Age (70–79): age range of 70–79 years, PSA (0.0–1.0): baseline PSA range of 0.0–1.0 ng/ml, PSA (1.1–2.0); baseline PSA range of 1.1–2.0 ng/ml, PSA (2.1– 3.0); baseline PSA range of 2.1–3.0 ng/ml, PSA (3.1–4.0); baseline PSA range of 3.1–4.0 ng/ml

50–590.0–1.0979899.9% (0.1%)99.4% (0.3%)99.4% (0.3%)99.4% (0.3%)98.5% (0.6%)Age (50–59) p = 0.03: PSA (0.0–1.0) > PSA (1.1–2.0) p < 0.0001: PSA (0.0–1.0). PSA (1.1–2.0) > PSA (2.1–3.0) > PSA (3.1–4.0) Age (60–69) p < 0.0001: PSA (0.0–1.0) > PSA (1.1–2.0) > PSA (2.1–3.0) > PSA (3.1–4.0) Age (70–79) p < 0.0001: PSA (0.0–1.0) > PSA (1.1–2.0) > PSA (2.1–3.0) > PSA (3.1–4.0) All Age ranges p < 0.0001: PSA (0.0–1.0) > PSA (1.1–2.0) > PSA (2.1–3.0) > PSA (3.1–4.0) PSA (0.0–1.0) p = 0.91: Age (50–59) = Age (60–69) p = 0.32: Age (50–59) = Age (70–79) p = 0.23: Age (60–69) = Age (70–79) PSA (1.1–2.0) p = 0.04: Age (50–59) > Age (60–69) p = 0.04: Age (50–59) > Age (70–79) p = 0.98: Age (60–69) = Age (70–79) PSA (2.1–3.0) p = 0.14: Age (50–59) = Age (60–69) p = 0.21: Age (50–59) = Age (70–79) p = 0.72: Age (60–69) = Age (70–79) PSA (3.1–4.0) p = 0.89: Age (50–59) = Age (60–69) p = 0.82: Age (50–59) = Age (70–79) p = 0.85: Age (60–69) = Age (70–79) All PSA ranges p < 0.0001: Age (50–59) > Age (60–69), Age (70–79) p = 0.0081: Age (60–69) > Age (70–79)
 1.1–2.0423999.8% (0.2%)99.4% (0.4%)99.1% (0.6%)97.0% (1.2%)95.6% (1.5%)
 2.1–3.01161598.3% (1.2%)96.0% (2.0%)85.3% (4.2%)77.2% (5.4%)77.2% (5.4%)
 3.1–4.0381668.4% (7.5%)58.2% (8.4%)53.3% (9.0%)53.3% (9.0%)53.3% (9.0%)
 All1,5564899.0% (0.3%)98.1% (0.4%)96.9% (0.5%)96.2% (0.6%)95.1% (0.7%)
60–690.0–1.02,1242099.8% (0.1%)99.4% (0.2%)99.1% (0.2%)99.1% (0.2%)98.6% (0.3%)
 1.1–2.01,5207199.0% (0.3%)98.1% (0.4%)96.0% (0.6%)94.4% (0.7%)92.6% (0.9%)
 2.1–3.063612895.9% (0.8%)88.0% (1.4%)80.1% (1.8%)74.1% (2.1%)70.1% (2.4%)
 3.1–4.032815475.0% (2.4%)54.5% (3.0%)47.2% (3.3%)41.9% (3.5%)37.7% (3.7%)
 All4,60837397.2% (0.2%)94.4% (0.4%)91.9% (0.5%)90.4% (0.5%)88.9% (0.6%)
70–790.0–1.0619299.8% (0.2%)99.8% (0.2%)99.8% (0.2%)99.4% (0.5%)99.4% (0.5%)
 1.1–2.05311899.6% (0.3%)98.3% (0.7%)97.4% (0.9%)93.7% (1.7%)90.7% (2.4%)
 2.1–3.02734097.8% (0.9%)91.4% (2.0%)82.3% (3.1%)76.2% (3.9%)65.0% (5.4%)
 3.1–4.01707869.4% (3.5%)56.7% (4.1%)45.9% (4.8%)43.8% (5.0%)35.9% (6.5%)
 All1,59313896.2% (0.5%)93.3% (0.7%)90.5% (0.9%)88.0% (1.1%)84.7% (1.4%)
All0.0–1.03,7223099.8% (0.1%)99.5% (0.1%)99.2% (0.2%)99.2% (0.2%)98.7% (0.3%)
 1.1–2.02,4749899.2% (0.2%)98.3% (0.3%)96.8% (0.4%)94.8% (0.6%)92.9% (0.7%)
 2.1–3.01,02518396.7% (0.6%)89.8% (1.1%)81.0% (1.5%)75.4% (1.8%)70.3% (2.0%)
 3.1–4.053624872.8% (1.9%)55.9% (2.4%)47.5% (2.6%)42.4% (2.8%)38.5% (3.0%)
 All7,75755997.3% (0.2%)94.9% (0.3%)92.6% (0.3%)91.1% (0.4%)89.5% (0.5%)

DISCUSSION

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONCLUSIONS
  7. REFERENCES

The age-adjusted incidence of prostate cancer in Japanese males is very low, approximately 10 in 100,000 compared to 147.3 in Caucasian Americans and 222.9 in African-Americans4 by epidemiological studies. Furthermore, during the period of 1988–1992, the average age-adjusted incidence rates of prostate cancer in native Japanese and Japanese-Americans in Los Angeles were 9.0 and 47.2 per 100,000, respectively.5 The difference in the age-adjusted incidence rate between the 2 groups has been considered to be related to the different lifestyle characteristics in USA and Japan. However, there may be differences in the cancer registry systems and screening systems among different countries. The increasing incidence in Americans may reflect the more intensive screening for prostate cancer in the USA than that in Japan.

Alternatively, serum PSA measurement has been widely accepted to detect early stage of prostate cancer, and levels in the serum are strongly related to the possibility of prostate cancer. There was only one comparative study on serum PSA levels among Japanese-American and native Japanese males.6 They compared the age-specific PSA levels and estimated the prevalence of undetected prostate cancer in the 2 populations. They suggested that estimated cancer prevalence in males who were 75 years old was 10.0% and 5.4% in American-Japanese and native Japanese, respectively. They asserted that Japanese-American males showed increased risk to have prostate cancer in comparison with native Japanese in Japan but less than the findings reported by some epidemiological studies.4, 5

It is of interest to compare the cumulative rate of PSA increase above 4 ng/ml in men with baseline PSA levels of 4.0 ng/ml or lower. The 5 years cumulative risks of PSA increase above 4 ng/ml in Americans were 1.6%, 7.6%, 34.6% and 83.0% in men with baseline PSA levels of 0.0 to 1.0 ng/ml, 1.1 to 2.0 ng/ml, 2.1 to 3.0 ng/ml and 3.1 to 4.0 ng/ml, respectively.7 In the present study, the cumulative rates of PSA increase above 4.0 ng/ml at 5 years also increased when baseline PSA ranges were higher, which were 1.3%, 7.1%, 29.7% and 71.5% in men with baseline PSA levels of 0.0 to 1.0 ng/ml, 1.1 to 2.0 ng/ml, 2.1 to 3.0 ng/ml and 3.1 to 4.0 ng/ml, respectively. The future risk of PSA increase in the same baseline PSA range may be similar to that between Americans and Japanese men.

However, there were some limitations and flaws in the present study. First, almost half of the first screened men did not undergo consecutive screening. This low percentage of men undergoing screening twice or more may lead to self-selection bias. The mean age of men screened only once were significantly older and the baseline PSA levels of those men were significantly lower than those of 6,493 men screened twice or more (data not shown). However, we analyzed the cumulative rate of freedom from PSA increase stratified by baseline PSA and age ranges, so the flaw from self-selection bias in our study may be ignored.

Second, the lack of uniformity in screening interval may lead to a serious flaw about changes in PSA. Therefore, we estimated interval PSA levels for all men who underwent screening with greater than a 1 year interval on the assumption that PSA levels changed over time in a simple exponential fashion. Therefore, the percentage of cases with PSA levels that were greater than 4.0 ng/ml in each year may be corrected more precisely.

Third, the lack of information on screening outcomes was also a limitation in our study. The cumulative rates of PSA increase may overestimate the true risk of developing prostate cancer in the future because an unknown part of the PSA increase may have originated from benign prostatic disease. On the other hand, the cumulative rate of screen detectable prostate cancer may underestimate the true risk of developing prostate cancer because of the likelihood of missing cancer. Recently, Ito et al.8 demonstrated the natural history of PSA increase in men with and without prostate cancer. The probability of noncancer-related PSA increase was high at about 90% if cases had baseline PSA levels of 2.0 ng/ml or less and also had PSA increase within 2 years. On the other hand, the probability of noncancer-related PSA increase was relatively low at about 50% for cases who had baseline PSA levels of 2.1 to 4.0 ng/ml or had PSA increase after 3 years of baseline PSA measurements.

Finally, men with a gradual PSA increase from lower to upper baseline PSA category were counted multiple times in the present study. Of those men, some patients with PSA increase above 4.0 ng/ml may have been counted multiple times; then, this methodology might inflate the risk of PSA increase in the PSA range of 1.1 to 4.0 ng/ml. Therefore, we compared the risk of PSA increase above 4.0 ng/ml between cases classified into each baseline PSA range at initial screening and those done in consecutive screening. However, the cumulative rate of freedom from PSA increase in men with baseline a PSA range of 1.1 to 2.0 ng/ml or with 2.1 to 3.0 ng/ml at initial screening was not significantly different from that in those with the same baseline PSA category in consecutive screening (data not shown). The 5-year cumulative rate of freedom from PSA increase was significantly lower at 35.1% in men with baseline PSA levels of 3.1 to 4.0 ng/ml at initial screening than those with in consecutive screening, which was 44.0% (p=0.02) (data not shown). Therefore, persons who got counted multiple times according to the inclusion criteria may not inflate the estimates in the risk of PSA increase above 4.0 ng/ml in the baseline PSA reflex range of 1.1 to 4.0 ng/ml.

The risk of PSA increase at 1 year after the baseline PSA measurement was relatively high at 3.3% and high at 27.2% in men with baseline PSA levels of 2.1 to 3.0 ng/ml and 3.1 to 4.0 ng/ml, respectively. Furthermore, the probability of cancer related PSA increase in men with baseline PSA levels of 2.1 to 4.0 ng/ml was high at 1 year after baseline PSA measurements.8 Therefore, we recommend annual screening using PSA measurements and DRE to minimize the delay in prostate cancer detection for individual-based screening. On the other hand, several studies demonstrated the validity of lower PSA cut-offs. Catalona et al.9 demonstrated that the rates of organ-confined cancer were 81%, 70%, 71% and 53% for PSA reflex ranges of 2.6 to 4.0 ng/ml, 4.1 to 5.0 ng/ml, 5.1 to10 ng/ml and greater than 10 ng/ml, respectively, in 676 patients treated with radical prostatectomy. Furthermore, regardless of the DRE findings, the positive predictive value of PSA levels of 2.5 to 4.0 ng/ml10 was unexpectedly high at 25%. Catalona et al.11 also demonstrated that the positive predictive value of prostate cancer in subjects with PSA levels of 2.5 to 4.0 ng/ml and without abnormal findings on DRE was relatively high at 15%. They also demonstrated the possibility of using cut-off points for the free/total PSA ratio to identify men at relatively high risk of prostate cancer in this reflex range of PSA. Therefore, when a mass screening for prostate cancer is conducted using PSA measurement alone, lowered cut-offs of PSA can detect more organ-confined cancers compared to a cut-off of 4.0 ng/ml. However, lowered cut-offs for all participants require more biopsies, and they may detect more insignificant cancers.12, 13 Then, we should use lower cut-offs for PSA in combination with other serum markers like free/total PSA ratio11 or should set cut-offs for PSA in an age-specific manner.14

In men with baseline PSA levels of 1.1 to 2.0 ng/ml, the cumulative risk of PSA increase above 4.0 ng/ml was relatively low at 0.8% and 1.7% at 1 and 2 years after the baseline PSA measurement. In this baseline PSA range, the probability of noncancer-related PSA increase was high at 86% within 2 years after baseline PSA measurements.8 On the other hand, the 3 years cumulative risk of rising PSA increased to 3.2%, and the probability of noncancer-related PSA increase lowered to 55% between 3 and 4 years after the baseline PSA measurement.8 Furthermore, a previous study also suggested that the proportion of clinical T3N0M0 disease in prostate cancer cases detected after a rescreening interval of 3 years was high in the baseline PSA range of 1.1 to 2.0 ng/ml.15 Therefore, the rescreening interval using the PSA measurement should be set within 2 years.

In our study, 1 case who did not undergo DRE at the baseline PSA measurement had a huge PSA increase from 1.6 ng/ml to 1,927 ng/ml after a rescreening interval of 4 years. The interval PSA levels were expected to be 6.6 ng/ml and 43.9 ng/ml at 1 and 2 years after the baseline PSA measurement, respectively. Therefore, setting of a 2-year rescreening interval for men with baseline PSA range of 1.1 to 2.0 ng/ml may result in increasing the risk of incurable prostate cancer in mass screening system using PSA alone. On the other hand, there was no case with a huge PSA increase within 5 years in men with the same baseline PSA range and also with normal baseline DRE findings (data not shown). Therefore, rescreening interval can be set at 1 or 2 years in individual-based screening using both PSA and DRE. Alternatively, rescreening interval in mass screening using PSA alone should be set at 1 year to avoid developing incurable prostate cancer.

In men with baseline PSA levels of 1.0 ng/ml or lower, the cumulative rate of PSA increase above 4.0 ng/ml was low during 5 years of observations. The probability of cancer-related PSA increase may be extremely low within 2 years after the baseline PSA measurement.8 However, the 3-year cumulative risk of PSA increase was 0.8% in the present study, and the probability of cancer related PSA increase among men with increasing PSA above 4.0 ng/ml at 3 years after the baseline PSA measurement increased to 20%.8 Furthermore, the cumulative risk of PSA increase was 1.3% during 5 years of observation, and the probability of cancer related PSA increase at 5 years after the baseline PSA measurement was relatively high at about 40%.8 Therefore PSA test should be conducted every 3 years or more but no more than 5 years. In this low baseline PSA range, DRE may still have an important role for detecting prostate cancer because only 14% of prostate cancer cases detected within 5 years after the baseline PSA measurement had PSA abnormality.15 Therefore, both PSA and DRE should be conducted in individual-based screening.

At present, screening and management of subsequently diagnosed prostate cancer can be harmful to individuals. To solve controversies regarding the screening for prostate cancer, prospective randomized controlled trials are currently on going in the USA16 and Europe.17, 18 In this situation, there is obvious need to provide informed consent prior to the application of screening tests to men who hope to undergo screening.1 At the same time, we should construct an optimal screening system for well-informed participants on screening. Furthermore, when we evaluate the usefulness of screening for prostate cancer with regard to mortality rate, quality of life and cost effectiveness, the establishment of an optimal screening system must also be important. The setting of an optimal rescreening interval is considered one of the most important issues not only to influence the efficacy of screening for prostate cancer but also to be informative for participants in the present screening system. Rescreening intervals have been proposed between 1 and 4 years by several organizations and clinical trials.16, 17, 18, 19, 20 More studies are necessary to confirm the optimal screening intervals. However, the present findings could demonstrate some important suggestions that the future risk of prostate cancer may be similar in men with the same baseline PSA levels between Americans and Japanese.

CONCLUSIONS

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONCLUSIONS
  7. REFERENCES

The probability of PSA increase was not significantly related to age range and may be similar between Americans and Japanese in the same baseline PSA ranges. Therefore, the rescreening interval should be set based on baseline PSA levels. In individual-based screening using both PSA and DRE, the rescreening interval should be set at 1 year and 1 to 2 years in men with baseline PSA ranges of 2.1 to 4.0 ng/ml and 1.1 to 2.0 ng/ml, respectively. The rescreening interval in mass screening system using PSA alone should be set at 1 year for men with baseline PSA ranges of 1.1 to 4.0 ng/ml. The risk of rapid PSA increase may be low in men with baseline PSA levels of 0.0 to 1.0 ng/ml, so the rescreening interval should be set between 3 and 5 years.

REFERENCES

  1. Top of page
  2. Abstract
  3. MATERIAL AND METHODS
  4. RESULTS
  5. DISCUSSION
  6. CONCLUSIONS
  7. REFERENCES
  • 1
    Ito K, Schröder FH. Informed consent for PSA-based screening: European view. Urology 2003; 61: 202.
  • 2
    Catalona WJ, Richie JP, Ahmann FR, Hudson MA, Scardino PT, Flanigan RC, deKernion JB, Ratliff TL, Kavoussi LR, Dalkin BL, Waters WB, MacFarlane MT, et al. Comparison of digital rectal examination and serum prostate specific antigen in the early detection of prostate cancer: results of a multicenter clinical trial of 6,630 men. J Urol 1994; 151: 128390.
  • 3
    Crawford ED, DeAntoni EP, Etzioni R, Schaefer VC, Olson RM, Ross CA. Serum prostate-specific antigen and digital rectal examination for early detection of prostate cancer in a national community-based program: The Prostate Cancer Education Council. Urology 1996; 47: 8639.
  • 4
    Greenlee RT, Murray T, Bolden S, Wingo PA. Cancer statistics, 2000. CA Cancer J Clin 2000; 50: 733.
  • 5
    ParkinDM, WhelanSL, FerlayJ, RaymondL, YoungJ, eds. Cancer incidence in five continents, Vol VII. Lyon: IARC Science Pub, No. 143, 1997.
  • 6
    Shibata A, Whittemore AS, Imai K, Kolonel LN, Wu AH, John EM, Stamey TA, Paffenbarger RS. Serum levels of prostate-specific antigen among Japanese-American and native Japanese men. J Natl Cancer Inst 1997; 89: 171620.
  • 7
    Crawford D, Chia D, Andriole GL, Reding D, Gelmann EP, Gohagan JK, Pinsky P, Hayes RB, Levin DL, Fagerstrom RM, Kramer B. PSA testing interval, reduction in screening intervals: data from the prostate, lung, colorectal and ovarian cancer (PLCO) screening trial. J Urol 2002; 167: 99100.
  • 8
    Ito K, Yamamoto T, Ohi M, Takechi H, Kurokawa K, Suzuki K, Yamanaka H. Natural history of PSA increase with and without prostate cancer. Urology 2003; 62: 649.
  • 9
    Catalona WJ, Ramos CG, Carvalhal GF, Yan Y. Lowering PSA cutoffs to enhance detection of curable prostate cancer. Urology 2000; 55: 7915.
  • 10
    Babaian RJ, Johnston DA, Naccarato W, Ayala A, Bhadkamkar VA, Fritsche HA Jr. The incidence of prostate cancer in a screening population with a serum prostate specific antigen between 2.5 and 4.0 ng/ml: relation to biopsy strategy. J Urol 2001; 165: 75760.
  • 11
    Catalona WJ, Partin AW, Finlay JA, Chan DW, Rittenhouse HG, Wolfert RL, Woodrum DL. Use of percentage of free prostate-specific antigen to identify men at high risk of prostate cancer when PSA levels are 2.51 to 4 ng/mL and digital rectal examination is not suspicious for prostate cancer: an alternative model. Urology 1999; 54: 2204.
  • 12
    Carter HB, Epstein JI, Partin AW. Influence of age and prostate-specific antigen on the chance of curable prostate cancer among men with nonpalpable disease. Urology 1999; 53: 12630.
  • 13
    Woolf SH. Screening for prostate cancer with prostate-specific antigen: an examination of the evidence. N Engl J Med 1995; 333: 14015.
  • 14
    Ito K, Yamamoto T, Kubota Y, Suzuki K, Fukabori Y, Kurokawa K, Yamanaka H. Usefulness of age-specific reference range of prostate-specific antigen for Japanese men older than 60 years in mass screening for prostate cancer. Urology 2000; 56: 27882.
  • 15
    Ito K, Yamamoto T, Ohi M, Takechi H, Kurokawa K, Suzuki K, Yamanaka H. Possibility of re-screening intervals for more than one year in men with PSA levels of 4.0 ng/ml or less. Prostate 2003; 57: 813.
  • 16
    Prorok PC, Andriole GL, Bresalier RS, Buys SS, Chia D, Crawford ED, Fogel R, Gelmann EP, Gilbert F, Hasson MA, Hayes RB, Johnson CC, et al. Prostate, lung, colorectal and ovarian cancer screening trial project team.: design of the prostate, lung, colorectal and ovarian (PLCO) cancer screening trial. Control Clin Trials 2000; 21: 273S309S.
  • 17
    Schröder FH, Kranse R, Rietbergen J, Hoedemaeke R, Kirkels W. The European Randomized Study of Screening for Prostate Cancer (ERSPC): an update. Members of the ERSPC, Section Rotterdam. Eur Urol 1999; 35: 53943.
  • 18
    Standaert B, Denis L. The European Randomized Study of Screening for Prostate Cancer: an update. Cancer 1997; 80: 18304.
  • 19
    Prostate-specific antigen (PSA) best practice policy. American Urological Association (AUA). Oncology (Huntington). 2000; 14: 26772, 277–8, 280.
  • 20
    Ito K, Kubota Y, Yamamoto T, Suzuki K, Fukabori Y, Kurokawa K, Yamanaka H. Long term follow-up of mass screening for prostate carcinoma in men with initial prostate specific antigen levels of 4.0 ng/ml or less. Cancer 2001; 91: 4451.