Chapter 4: Cancer incidence in Canada: trends and projections (1983-2032) – Prostate cancer - HPCDP: Volume 35, Supplement 1, Spring 2015

Chapter 4: Projections by Cancer Site

14. Prostate cancer

Prostate cancer is the most common form of cancer and the third leading cause of cancer death in Canadian males. One in 7 can expect to be diagnosed with prostate cancer in their lifetime, and 1 in 28 males to die from it.Endnote 1 The annual number of new cases was 21 460 in 2003–2007, constituting 13.8% of all new Canadian cases of cancer and 26.6% of those in males (Table 4.14.1, Figure 3.9).

TABLE 4.14.1
Observed (2003–2007) and projected average annual new cases by age and province/territories combined (TC), prostate cancer, Canada, 2003–2032
Period Age New cases
CA BC AB SK MB ON QC NB NS PE NL TC
2003–07 <45 75 5 10 5 0 35 10 0 5 0 0 0
45–54 1570 170 190 50 45 670 295 45 65 10 30 5
55–64 6170 745 620 200 165 2560 1305 190 230 40 105 10
65–74 7935 1060 740 325 245 3385 1435 250 300 55 135 10
75–84 4515 700 395 215 160 1835 840 130 160 25 55 5
85+ 1195 175 105 60 50 390 340 25 35 5 10 0
Total 21 460 2860 2055 850 665 8875 4225 645 790 135 335 25
2008–12 <45 100 5 10 0 0 35 10 0 0 0 0 0
45–54 2160 170 245 65 50 1025 405 55 85 15 30 5
55–64 8275 875 755 295 195 3650 1680 285 340 60 160 15
65–74 9130 1200 660 345 230 4070 1695 325 370 65 165 10
75–84 4460 710 330 175 115 1900 850 130 160 25 55 5
85+ 1370 205 105 45 40 475 430 25 35 5 10 0
Total 25 495 3170 2100 930 630 11 155 5065 825 990 170 420 35
2013–17 <45 95 5 10 0 0 35 10 0 0 0 0 0
45–54 2205 175 245 65 50 1085 400 55 85 15 25 5
55–64 9455 990 915 340 225 4225 1875 305 365 65 170 15
65–74 11 510 1525 880 415 285 5130 2080 415 465 85 210 20
75–84 4920 785 365 175 120 2085 950 155 185 25 65 10
85+ 1740 260 140 50 45 625 560 30 40 5 10 0
Total 29 930 3740 2560 1050 730 13 180 5870 960 1140 200 480 45
2018–22 <45 100 5 10 5 0 35 10 0 0 0 0 0
45–54 2025 165 235 55 45 1005 350 50 70 10 25 5
55–64 10 465 1085 1020 365 245 4830 2020 320 390 70 170 15
65–74 13 830 1855 1135 515 350 6190 2410 490 540 100 245 25
75–84 5975 950 450 200 140 2510 1155 195 225 35 85 10
85+ 2065 305 170 55 50 740 675 35 45 5 15 0
Total 34 460 4375 3025 1190 830 15 310 6620 1085 1280 225 535 55
2023–27 <45 110 10 10 5 0 35 10 0 0 0 0 0
45–54 2000 165 245 55 45 980 350 45 70 10 20 5
55–64 10 480 1120 1020 350 240 4980 1945 305 375 70 165 15
65–74 15 960 2120 1380 595 405 7250 2720 530 600 110 255 30
75–84 7735 1240 615 250 185 3250 1470 260 295 45 110 15
85+ 2400 355 200 60 55 855 800 40 55 10 15 0
Total 38 690 5010 3465 1315 930 17 350 7290 1180 1395 245 570 65
2028–32 <45 115 10 10 5 0 40 10 0 0 0 0 0
45–54 2110 180 260 60 50 1025 375 45 65 10 20 5
55–64 9810 1080 995 315 225 4690 1760 275 335 65 145 15
65–74 17 645 2340 1530 635 435 8285 2910 550 640 120 265 30
75–84 9475 1520 810 315 225 3970 1760 305 350 50 135 25
85+ 3065 450 260 70 70 1075 1040 55 70 10 20 5
Total 42 225 5580 3865 1400 1010 19 085 7855 1230 1465 260 585 75

Abbreviations: AB, Alberta; BC, British Columbia; CA, Canada; MB, Manitoba; NB, New Brunswick; NL, Newfoundland and Labrador; NS, Nova Scotia; ON, Ontario; PE, Prince Edward Island; QC, Quebec; SK, Saskatchewan; TC, All Territories (Yukon, Northwest Territories and Nunavut).

Note: Totals may not add up due to rounding.

During 2001–2007, the ASIRs for prostate cancer were stable (Figure 3.1). The risk of developing the disease increases with age more than for any other cancer (Table 4.14.2). Between 2003 and 2007, 66% of the overall cases were in men aged 55 to 74, while only 8% were in males younger than 55. A high 5-year relative survival rate of 96% was reported for 2006–2008.Endnote 1

TABLE 4.14.2
Observed (2003–2007) and projected age-standardized incidence rates (ASIRs) by age and province/territories combined (TC), prostate cancer, Canada, 2003–2032
Period Age ASIRs
CA BC AB SK MB ON QC NB NS PE NL TC
2003–07 <45 0.6 0.5 0.7 0.9 0.3 0.8 0.4 0.3 0.8 2.0 0.8 0.6
45–54 62.2 50.1 73.3 67.8 51.0 70.9 47.6 76.3 82.8 108.5 63.1 34.9
55–64 365.2 321.5 412.1 419.6 287.8 408.1 301.5 436.3 420.0 525.9 331.7 225.9
65–74 741.6 700.8 818.2 947.9 655.0 834.9 541.2 945.8 875.5 1041.4 719.0 529.2
75–84 728.5 780.1 773.4 919.9 667.0 764.1 580.0 872.2 858.3 902.1 581.0 753.3
85+ 781.8 728.6 797.7 787.9 681.6 693.4 1067.4 592.6 667.3 717.3 402.0 725.7
Total 123.3 116.6 135.6 149.7 106.0 134.7 100.0 147.8 142.9 169.1 109.9 94.1
2008–12 <45 0.9 0.5 0.7 0.9 0.3 0.8 0.4 0.3 0.8 2.0 0.8 0.6
45–54 76.6 45.8 79.6 80.1 51.6 94.6 60.1 87.8 104.7 108.5 63.1 34.9
55–64 402.7 305.7 376.1 496.1 276.7 478.7 332.0 542.5 522.0 638.1 423.4 225.9
65–74 734.0 680.4 625.8 947.5 545.7 876.2 533.2 1009.2 918.4 1110.4 723.2 529.2
75–84 643.7 700.9 561.0 763.6 465.6 714.5 503.5 826.0 785.7 746.6 542.9 753.3
85+ 652.1 628.4 607.3 547.1 469.8 588.3 933.3 466.5 554.1 582.5 344.1 725.7
Total 123.3 109.5 110.5 149.6 88.3 143.3 99.3 159.6 153.3 176.5 116.1 94.1
2013–17 <45 0.9 0.5 0.7 0.9 0.3 0.8 0.4 0.3 0.8 2.0 0.8 0.6
45–54 76.6 45.8 79.6 80.1 51.6 94.6 60.1 87.8 104.7 108.5 63.1 34.9
55–64 402.7 305.7 376.1 496.1 276.7 478.7 332.0 542.5 522.0 638.1 423.4 225.9
65–74 734.0 680.4 625.8 947.5 545.7 876.2 533.2 1009.2 918.4 1110.4 723.2 529.2
75–84 643.7 700.9 561.0 763.6 465.6 714.5 503.5 826.0 785.7 746.6 542.9 753.3
85+ 652.1 628.4 607.3 547.1 469.8 588.3 933.3 466.5 554.1 582.5 344.1 725.7
Total 123.3 109.5 110.5 149.6 88.3 143.3 99.3 159.6 153.3 176.5 116.1 94.1
2018–22 <45 0.9 0.5 0.7 0.9 0.3 0.8 0.4 0.3 0.8 2.0 0.8 0.6
45–54 76.6 45.8 79.6 80.1 51.6 94.6 60.1 87.8 104.7 108.5 63.1 34.9
55–64 402.7 305.7 376.1 496.1 276.7 478.7 332.0 542.5 522.0 638.1 423.4 225.9
65–74 734.0 680.4 625.8 947.5 545.7 876.2 533.2 1009.2 918.4 1110.4 723.2 529.2
75–84 643.7 700.9 561.0 763.6 465.6 714.5 503.5 826.0 785.7 746.6 542.9 753.3
85+ 652.1 628.4 607.3 547.1 469.8 588.3 933.3 466.5 554.1 582.5 344.1 725.7
Total 123.3 109.5 110.5 149.6 88.3 143.3 99.3 159.6 153.3 176.5 116.1 94.1
2023–27 <45 0.9 0.5 0.7 0.9 0.3 0.8 0.4 0.3 0.8 2.0 0.8 0.6
45–54 76.6 45.8 79.6 80.1 51.6 94.6 60.1 87.8 104.7 108.5 63.1 34.9
55–64 402.7 305.7 376.1 496.1 276.7 478.7 332.0 542.5 522.0 638.1 423.4 225.9
65–74 734.0 680.4 625.8 947.5 545.7 876.2 533.2 1009.2 918.4 1110.4 723.2 529.2
75–84 643.7 700.9 561.0 763.6 465.6 714.5 503.5 826.0 785.7 746.6 542.9 753.3
85+ 652.1 628.4 607.3 547.1 469.8 588.3 933.3 466.5 554.1 582.5 344.1 725.7
Total 123.3 109.5 110.5 149.6 88.3 143.3 99.3 159.6 153.3 176.5 116.1 94.1
2028–32 <45 0.9 0.5 0.7 0.9 0.3 0.8 0.4 0.3 0.8 2.0 0.8 0.6
45–54 76.6 45.8 79.6 80.1 51.6 94.6 60.1 87.8 104.7 108.5 63.1 34.9
55–64 402.7 305.7 376.1 496.1 276.7 478.7 332.0 542.5 522.0 638.1 423.4 225.9
65–74 734.0 680.4 625.8 947.5 545.7 876.2 533.2 1009.2 918.4 1110.4 723.2 529.2
75–84 643.7 700.9 561.0 763.6 465.6 714.5 503.5 826.0 785.7 746.6 542.9 753.3
85+ 652.1 628.4 607.3 547.1 469.8 588.3 933.3 466.5 554.1 582.5 344.1 725.7
Total 123.3 109.5 110.5 149.6 88.3 143.3 99.3 159.6 153.3 176.5 116.1 94.1

Abbreviations: AB, Alberta; BC, British Columbia; CA, Canada; MB, Manitoba; NB, New Brunswick; NL, Newfoundland and Labrador; NS, Nova Scotia; ON, Ontario; PE, Prince Edward Island; QC, Quebec; SK, Saskatchewan; TC, All Territories (Yukon, Northwest Territories and Nunavut).

Overall incidence rates of prostate cancer increased steadily until 1993–1997, showed additional but smaller increases to 1998–2002 and then levelled off (Figure 4.14.1).

FIGURE 4.14.1
Age-standardized incidence rates (ASIRs) by region, prostate cancer, 1983–2032

figure 4.14.1

[Click to enlarge]

[FIGURE 4.14.1, Text Equivalent]

Based on a two-step approach of the short-term modelling projection followed by the long-term constant-rates projection, British Columbia and the Prairies are likely to have a drop in prostate cancer incidence rates, Ontario and the Atlantic region will experience an increase, and Quebec and Canada will remain stable in the first 5 prediction years. The rates for each region will remain unchanged thereafter.

This pattern was observed in the Prairies, the Atlantic region and Ontario. The rates in British Columbia reached their first peak one period sooner than in other regions, then gradually levelled off and decreased thereafter. Quebec experienced much lower rates in the last 10 observation years than other regions. The trends of age-specific ASIRs show that recent incidence rates of prostate cancer in Canada increased in the younger age groups (<75 years) and decreased in the older age groups (75+) (Figure 4.14.2).

FIGURE 4.14.2
Age-standardized incidence rates (ASIRs) for prostate cancer by age group, Canada, 1983–2032

figure 4.14.2
[Click to enlarge]

[FIGURE 4.14.2, Text Equivalent]

The trends of age-specific ASIRs show that recent incidence rates of prostate cancer in Canada increased in the younger age groups (<75 years) and decreased in the older age groups (75+). Based on a two-step approach of the short-term modelling projection followed by the long-term constant-rates projection, these observed trends are expected to continue in the first 5 prediction years. The rates for each age group will remain unchanged thereafter.

The Nordpred method produces extreme increases in prostate cancer incidence rates and counts, so we used a two-step approach of the short-term modelling projection followed by the long-term constant rates projection (method ADa + AVG, see Chapter 2 for the definition). Consequently, British Columbia and the Prairies are likely to have a drop in prostate cancer incidence rates, Ontario and the Atlantic region will experience an increase, and Quebec and Canada will remain stable in the first 5 prediction years (Figure 4.14.1). The rates for each region will remain unchanged thereafter.

From 2003–2007 to 2028–2032, the ASIRs for prostate cancer in Canada are expected to be stable at about 123.3 per 100 000 (Table 4.14.2). Despite this trend in ASIRs, the aging and growth of the population mean that the annual number of new cases is projected to increase by 97%, from 21 460 to 42 225 (Table 4.14.1).

Comments

The lower prostate cancer incidence rates observed in Quebec are likely artefactual, as a result of possible underreporting of cases (see details in Chapter 5: Data quality issues).

A range of medical investigations, including digital rectal examination, transrectal ultrasonography, PSA level, fine-needle aspiration biopsy (FNAB), and magnetic resonance imaging have been considered for the early detection of prostate cancer. Endnote 195 Part of the rise in incidence in the 2 decades prior to 1990 has been attributed to detection of cancers following transurethral resection of the prostate for benign prostatic hypertrophy.Endnote 196 The 1993 and 2001 peaks in incidence mirror the 2 waves of increased PSA screening activity.Endnote 1 In 2003, the percentage of men reporting a screening PSA test in the past 12 months was highest in the 60–69 age group at just over 35%.Endnote 197

When a slowly developing cancer is detected through screening, over-diagnosis may occur. This is concurrent with the fact that these cancers are most frequent at older ages when competing causes of death are more frequent.Endnote 198 The observed increase in prostate cancer incidence rates may be biased by such over-diagnosis from PSA screening. The benefits and harms of prostate cancer screening by PSA test is still being debated,Endnote 199 and it is important that men know the arguments in order to decide whether to screen for the disease.

Similar to England, recent incidence rates of prostate cancer in Canada increased in the younger age groups (<75 years) because of increasing uptake of the PSA test, and decreased in the older age groups (75+) from increasing pharmacological treatment for obstructive uropathy caused by benign prostatic hypertrophy and the resulting decrease in use of transurethral resection, which had previously led to detection of many prostate cancers (Figure 4.14.2).Endnote 28, Endnote 200 This pattern resulted in relatively stable crude rates in the last 2 observation periods. Therefore, the recent trends could be used for a model of short term projections. However, such a model is less suitable for long-term projections because it would be expected that the prevalence of screening would plateau in the future. Figure 4.14.3 shows the projected ASIRs in the first 10 projection years derived from using the current 2-step approach (ADa + AVG), and using the age-specific trend model (ADa) only to project for all the 10 years. There appears to be no substantial advantage to a 10-year projection for using model ADa only. Similarity of the estimated national ASIRs between using ADa + AVG and ADa may be because the decreases in rates in older age groups were partly cancelled out by the increases in younger ages, and may also be a result of offsetting provincial differences. The projected increases in the 10-year incidence rates in the Atlantic region and Ontario, from using ADa only, seem less likely. Thus current trends were only extended to the first 5 projection years through ADa based on yearly data. The age-specific average rates of the predicted 5-year incidence data were then assumed to remain constant in the future years. Current projections of prostate cancer rates will be an overestimate if recent decreases in the rates (based on the more recent observed data, which were not available when present study was undertaken) continue.Endnote 1

FIGURE 4.14.3
Comparison of projected age-standardized incidence rates (ASIRs) derived from the two modelsa by region, prostate cancer, 2008–2017

figure 4.14.3

[Click to enlarge]

[FIGURE 4.14.3, Text Equivalent]

Figure 4.14.3 shows the projected ASIRs in the first 10 projection years derived from using the current 2-step approach (ADa + AVG), and using the age-specific trend model (ADa) only to project for all the 10 years. There appears to be no substantial advantage to a 10-year projection for using model ADa only. Similarity of the estimated national ASIRs between using ADa + AVG and ADa may be because the decreases in rates in older age groups were partly cancelled out by the increases in younger ages, and may also be a result of offsetting provincial differences. The projected increases in the 10-year incidence rates in the Atlantic region and Ontario, from using ADa only, seem less likely. Thus current trends were only extended to the first 5 projection years through ADa based on yearly data. The age-specific average rates of the predicted 5-year incidence data were then assumed to remain constant in the future years.

Note: Abbreviations: CA=Canada, BC=British Columbia, PR=the Prairies, ON=Ontario, QC=Quebec, AT=the Atlantic region.

a Model ADa: Using ADa (see Methods) to project for the first 10 future years (2008–2017).
Model ADa +AVG: Using ADa to project for the first five projection years (2008–2012), and then using the age-specific average rates of the predicted 5-year data to estimate counts for the future years (2013–2017).

The established risk factors, which are all non-modifiable, are age, family history and ethnicity. Having a first-degree relative with prostate cancer more than doubles a man's risk of developing this tumour.Endnote 201, Endnote 202, Endnote 203, Endnote 204 Risks to sons appeared to be lower than in brothers. Risk is further increased by early age at onset in relatives and multiple relatives with the disease.Endnote 201 A number of studies may suggest an overall contribution of inherited genes or a shared environment in the development of this disease.Endnote 201, Endnote 202, Endnote 203, Endnote 204 Genes and family history account for about 5% to 9% of all prostate cancers.Endnote 205 White males have a lower risk of prostate cancer than Black males worldwide but have a higher risk than males of Asian ethnicity.Endnote 56, Endnote 206, Endnote 207 However, the risk for Asian Americans is higher than that for males of a similar background living in Asia.Endnote 208 Ethnicity may be a surrogate for genetic, environmental or socioeconomic factors.Endnote 208

Although a definite modifiable risk factor has not been identified, a number have been implicated in prostate cancer initiation. Potential preventative factors include physical activity and frequent intake of soy and foods containing lycopene.Endnote 209 Lycopene may reduce prostate cancer risk by preventing oxidative DNA damage in prostate tissue by mitigating exposure to cellular free radicals.Endnote 209, Endnote 210 A meta-analysis shows that high soy consumption reduces the risk by 26%; however, this inverse association appears to be confined to the Asian population.Endnote 211 Possible explanations for this observation include different types or amounts of soy products consumed in Asian and Western societies.

Factors that may increase prostate cancer risk include high intakes of dairy products and meat.Endnote 209 Higher intake of calcium has been associated with a 39% increased risk,Endnote 212 possibly because of the suppression of 1,25-dihydrovitamin D, which may inhibit cancer cell invasion.Endnote 213 IARC stated that evidence of an increased risk of prostate cancer in relation to exposure to thorium-232 and its decay products is limited, as is exposure to X-radiation and gamma-radiation, use of anabolic androgenic steroids, exposure to cadmium and cadmium compounds, or exposure to arsenic and inorganic arsenic compounds and the rubber production industry.Endnote 47

Page details

Date modified: