Trends in survival of patients diagnosed with male genital cancers in the Nordic countries 1964–2003 followed up until the end of 2006

Abstract

Background. Prostate, testicular and penile cancer constitute about one-third of the cancer incidence burden among Nordic males, but less than one-fifth of the corresponding mortality. The aim of this study is to describe and interpret trends in relative survival and excess mortality in the five Nordic populations. Material and methods. Age-standardised incidence and mortality rates, 5-year relative survival, and excess mortality rates for varying follow-up periods are presented, as are age-specific 5-year relative survival by country, sex and 5-year diagnostic period. Results. The vast majority of male genital cancer incident cases and deaths are prostate cancers, with 5-year and 10-year survival trends resembling the diagnostic-led increasing incidence over the past 25 years. Five-year prostate cancer survival is estimated at 53% in Denmark compared to 78% or above in the other Nordic populations for patients diagnosed 1999–2003. Excess mortality has declined over time, with Denmark having a greater excess of deaths compared to the other countries, irrespective of time of diagnosis. Concomitant with the declines in testicular cancer mortality, testicular cancer survival has increased since the 1970s in each Nordic country. Men diagnosed with testicular cancer in recent decades have had, on average, a continually better prognosis with time, with relative 5-year survival for patients diagnosed 1999–2003 ranging from 88% in Finland to 94% in Sweden. For the few patients diagnosed with cancer of the penis and other male genital organs, survival trends have been rather stable over time, with recent 5-year relative survival estimates ranging from 62% in Finland to 80% in Norway. Conclusions. There are intriguing country-specific and temporal variations in male genital cancer survival in the Nordic countries. Prognosis varies widely by cancer site and relates to both changing diagnostic and clinical practices. The increasing PSA detection and biopsy makes interpretation of the prostate cancer survival trends particularly difficult.

Key Words::

• Neoplasms
• male genital
• relative survival
• excess mortality
• Cancer Registry

The cancers of the male genital organs considered in this study are prostate cancer, the most common neoplasm among Nordic men, testicular cancer, the most frequent cancer in younger men (aged 25–34), and the rarer penile cancer. Together they comprise about one-third of the cancer incidence burden among Nordic males, but less than one-fifth of the corresponding mortality. The vast majority – either in terms of numbers of new cases or deaths – are cancers of the prostate.

Incidence rates of prostate cancer in Western countries have for decades been influenced by the diagnosis of latent cancers via their detection in removed tissue, or latterly, by the widespread use of prostate-specific antigen (PSA) testing followed by multiple biopsies of the prostate gland where the test was positive. The trends in the Nordic countries follow this pattern, with steadily increasing incidence observed from the 1950s through to the 1980s followed by a rapid growth in rates in each Nordic country circa 1990 (Denmark is an exception, but similar increases are observed around 1995). The incidence rises in the 1990s were largely brought about by an increasing use of PSA as a diagnostic tool for detecting asymptomatic cancers [1].

Few prostate cancer patients in the Nordic countries were treated with curative intent before the late-1980s [2] and thus the increasing mortality rates up to the early-1990s primarily reflect the occurrence of clinically-detected cancers with symptoms. In the PSA era, mortality rates stabilised in Sweden and Denmark [3] and declined to levels seen in the early 1980s in Norway and Finland [1]. Survival has been uniformly increasing in the Nordic countries [4], as it has been in Europe, with 5-year relative survival ratios for men estimated at 85% or higher in patients diagnosed in Finland, Iceland, Norway and Sweden [5]. While recent survival increases may partially reflect the downturn in mortality trends, they have also become increasingly inflated following the upsurge in the diagnosis of latent cancers via PSA detection.

In contrast to prostate cancer, of which three-quarters of all diagnoses occur in males aged over 65, testicular cancer incidence peaks among the adolescents and young men. The incidence increases (in both seminoma and non-seminoma) are largely a birth cohort phenomenon with increases of 1 to 3% observed in the Nordic countries in generations of men born from at the 1920s and thereafter [6,7]. The age-adjusted (world) incidence rates of over 10 in Norway and Denmark 2004–2006 [3] put them among the highest incidence countries in Europe and worldwide, on a par with those observed in certain Swiss cantons [8]. Mortality rates are very low in comparison, but still reflect the ordering of rates of incidence (Nordic rates ranging from 0.2 in Finland to 0.4 in Denmark). Testicular cancer survival has improved dramatically since the mid-1970s following the introduction of platinum-based chemotherapy schemes [9] and best practice tumour management [10]. Five-year survival is about 95% or greater for patients recently diagnosed with the disease in several of the Nordic countries [5].

Cancers of the penis and other male genital organs are rare in each of the Nordic countries, accounting for 0.3% of all diagnosed cancers [3]. Currently estimates indicate that about as much as half of the incident cases may be attributable to oncogenic types of HPV [11]. Age-standardised incidence varies between 0.4 and 1.7 per 100 000 men in Europe [8], and in the Nordic countries there is a less than twofold variation, rates varying from 0.6 in Finland to around 1.1 in Norway and Iceland [3]. Mortality rates are however similar at around 0.2 in the Nordic populations 2004–2006. Overall survival is estimated at about 70% at five years [12,13], varying from around 80% for patients with localised tumours to about 50% in patients with regional lymph node metastasis [14].

The objective of this study is to describe and compare country and age-specific trends in relative survival and excess mortality among male patients diagnosed with genital cancers from 1964 through to 2003 in each of the five Nordic populations. A further aim is to interpret these in light of the clues gained on examining the observed trends in incidence and mortality, as well as relevant clinical and epidemiologic literature.

Material and methods

The materials and methods are described in detail in an accompanying article in this issue [15]. In brief, the NORDCAN database contains comparable data on cancer incidence and mortality in the Nordic countries, as delivered by the national cancer registries, with follow-up information on death or emigration for each cancer patient available up to and including 2006. This paper studies cancers of the penis and other male genital organs (ICD-10 C60 & C63), prostate (ICD-10 C61), and testis (ICD-10 C62). The number of male patients with prostate cancer included in the survival analyses is large, about 84 000 in the latest period, with rapid but variable increases (three- to nine-fold) over time by country (Table I). There were about 4 300 testicular cancer cases included in the latest diagnostic period 1999–2003 (Table III), with two- to three-fold increases in patient numbers over the consecutive periods. The relative rarity of penile cancer (Table V) – cases are in their hundreds in Denmark, Finland, Norway and Sweden – means that a degree of caution should be applied in interpreting the survival statistics generated. No results are reported for Iceland, given annual patient numbers are in single figures.

Table I. Trends in survival for prostate cancer by country. Number of tumours (N) included and the 5- and 10-year age-standardised (ICSS) relative survival in percent with 95% confidence intervals (RS (CI)). Nordic cancer survival study 1964–2003.

	Denmark		Finland		Iceland		Norway		Sweden
	N	RS (CI)	N	RS (CI)	N	RS (CI)	N	RS (CI)	N	RS (CI)
S-year
1964–1968	3 631	31 (28–34)	2 013	41 (37–46)	105	40 (27–58)	4 547	46 (42–50)	12 301	44 (42–46)
1969–1973	4 254	36 (33–39)	2 981	44 (40–47)	144	39 (30–52)	5 211	51 (47–54)	14 966	48 (46–50)
1974–1978	5 312	36 (33–38)	3 841	54 (51–58)	216	62 (52–73)	6 709	51 (49–54)	16 069	56 (54–58)
1979–1983	6 379	42 (38–45)	5 028	54 (51–57)	327	55 (46–66)	7 669	53 (50–56)	18 208	56 (54–58)
1984–1988	6 824	39 (36–41)	5 983	55 (52–58)	420	64 (57–73)	8 626	53 (50–56)	21 766	60 (59–62)
1989–1993	7 318	38 (35–41)	7 804	60 (57–62)	524	69 (63–77)	10 371	58 (56–61)	25 113	65 (63–66)
1994–1998	7 616	42 (40–45)	13 164	73 (71–75)	709	76 (71–82)	12 817	71 (70–73)	29 116	72 (71–74)
1999–2003	10 471	53 (51–55)	18 291	86 (84–87)	868	81 (78–85)	14 903	78 (77–79)	39 292	81 (80–82)
10-year
1964–1968	3 631	18 (15–21)	2 013	28 (24–33)	105	12 (7–21)	4 547	29 (25–33)	12 301	29 (27–31)
1969–1973	4 254	19 (16–22)	2 981	27 (23–31)	144	21 (14–31)	5 211	33 (29–36)	14 966	32 (30–34)
1974–1978	5 312	19 (17–22)	3 841	36 (33–40)	216	47 (36–61)	6 709	35 (32–38)	16 069	39 (37–42)
1979–1983	6 379	23 (20–26)	5 028	38 (35–42)	327	42 (32–55)	7 669	36 (33–39)	18 208	38 (36–40)
1984–1988	6 824	22 (20–24)	5 983	37 (34–40)	420	46 (37–56)	8 626	36 (34–39)	21 766	42 (40–44)
1989–1993	7 318	20 (18–22)	7 804	43 (40–46)	524	57 (49–66)	10 371	40 (37–42)	25 113	45 (43–47)
1994–1998	7 616	22 (21–24)	13 164	58 (56–60)	709	63 (57–68)	12 817	52 (50–53)	29 116	53 (51–53)
1999–2003	10 471	29 (27–31)	18 291	73 (71–75)	868	68 (63–74)	14 903	58 (57–60)	39 292	60 (58–61)

Sex-specific 5-year relative survival was calculated for each of the diagnostic groups in each country for eight 5-year periods from 1964–1968 to 1999–2003. For the last 5-year period, so-called hybrid methods were used. Country-specific population mortality rates were used for calculating the expected survival. Age-standardisation used the weights of the International Cancer Survival Standard (ICSS) cancer patient populations [16]. We present age-standardised (world) incidence and mortality rates, 5-year relative survival, and excess mortality rates for the follow-up periods of within one month, 1–3 months and 2–5 years following diagnosis, as well as age-specific 5-year relative survival by country and 5-year period.

Results

Prostate cancer

Incidence and mortality. Incidence trends have been increasing in each of Nordic countries over the last 50 years (Figure 1). Since around 1990, there has been a substantial increase in prostate cancer incidence rates in each population excluding Denmark, although a rise can be seen among Danish men about five years later. While incidence rates varied twofold in the mid-1960s, with rates highest in Sweden and lowest in Denmark, by 2003, rates in Iceland, Finland, Norway and Sweden rapidly converged. In contrast, rates have been substantially lower in Denmark over time, and remain about half of those observed in the other Nordic countries.

Figure 1. Trends in age-standardised (World) incidence and mortality rates per 100 000 and age-standardised (ICSS) 5- and 10-year relative survival for prostate cancer by country. Nordic cancer survival study 1964–2003.

In absolute terms, the magnitude of prostate cancer mortality rates can be seen to be much more similar than for incidence, irrespective of year of death (Figure 1). Time trends in prostate cancer mortality in the Nordic countries resemble those of incidence up to the late-1980s, although the rate of change is more modest. Overall mortality rates flattened off in Iceland and Sweden during the early-1990s, while the rates in Norway and Finland have steadily declined since the mid-1990s. The most recent mortality rates in Norway are of a similar magnitude to those in Denmark, Sweden and Iceland, whilst Finnish rates are substantially lower. Recent mortality trends in Denmark and Sweden appear rather stable.

Survival. The 5-year and 10-year survival trends in Figure 1 resemble the incidence curves over the past 25 years. Survival has been increasing with time, first in Iceland and Sweden in the early-1980s, in Finland and Norway from the early 1990s, and in Denmark circa 1995. Five-year survival in the latest study period is estimated at 53% in Denmark but around 80% or above in the four other Nordic populations (Table I), and is consistently lower in Denmark (Table II). The absolute differences between Denmark and the other countries have increased; there was a 15 percentage point difference in survival between populations during the diagnostic period 1964–1968, compared with a net difference of 33 by 1999–2003. Survival is poorer among younger ages (0–49) than in patients aged 50–69, although a lower survival among Danish prostate cancer patients persists irrespective of age at diagnosis. Excess mortality has declined over time but Denmark has a greater excess of deaths compared to the other countries, irrespective of time of diagnosis (Figure 2). Elsewhere, excess deaths are very close, particularly 1–3 months and 2–5 years after diagnosis. The survival increases are seen in each age group, and are of a greater magnitude in the latest 15-year period (Table II).

Table II. Trends in 5- and 10-year age-specific relative survival in percent after prostate cancer by country. Nordic cancer survival study 1964–2003.

Age	5-year survival						10-year survival
	0–49	50–59	60–69	70–79	80–89	90 +	0–49	50–59	60–69	70–79	80–89	90 +
Denmark
1964–1968	19	31	41	33	16	0	20	17	22	16	11	0
1969–1973	45	40	40	35	21	30	22	24	22	18	6	77
1974–1978	21	42	41	37	27	7	13	25	23	20	9	0
1979–1983	54	46	44	40	25	13	32	28	25	21	9	0
1984–1988	22	40	44	42	35	28	9	19	27	26	17	0
1989–1993	27	31	42	43	37	32	6	17	24	23	23	0
1994–1998	36	38	48	43	39	18	15	20	26	24	20	0
1999–2003	39	58	61	54	44	29	28	31	32	28	22	0
Finland
1964–1968	56	48	45	38	19	0	52	36	30	24	4	0
1969–1973	28	40	52	49	35	28	22	29	34	27	15	0
1974–1978	52	54	59	55	46	21	30	38	43	33	34	0
1979–1983	42	50	62	59	42	42	30	35	45	44	27	0
1984–1988	27	52	61	64	52	18	20	39	42	42	29	57
1989–1993	38	54	65	66	60	38	36	37	49	48	34	0
1994–1998	47	80	82	76	65	58	36	64	67	60	49	56
1999–2003	75	90	91	87	77	68	57	78	80	75	64	66
Iceland
1964–1968	31	31	45	23	81	*	26	26	17	*	*	*
1969–1973	15	15	51	56	33	*	37	37	39	*	*	*
1974–1978	68	68	74	53	44	133	58	58	54	35	34	*
1979–1983	44	44	68	57	48	*	48	48	48	37	26	*
1984–1988	55	55	79	62	59	*	38	38	64	39	41	*
1989–1993	69	69	61	75	73	56	62	62	54	59	49	*
1994–1998	74	74	77	81	71	74	65	65	63	65	53	*
1999–2003	83	83	83	83	72	54	71	71	71	67	59	*
Norwtay
1964–1968	51	53	53	42	26	8	40	32	36	21	18	0
1969–1973	51	57	57	49	35	7	34	40	38	31	15	0
1974–1978	45	52	57	53	41	50	30	41	41	33	25	0
1979–1983	30	54	59	56	51	40	19	41	43	37	33	0
1984–1988	35	51	60	56	49	37	29	37	44	37	27	44
1989–1993	39	61	64	63	50	34	30	39	46	44	28	0
1994–1998	73	77	78	69	57	33	56	60	59	48	31	15
1999–2003	76	86	86	76	60	34	65	70	67	53	34	0
Sweden
1964–1968	44	47	50	42	31	6	35	34	34	26	16	0
1969–1973	55	54	54	46	31	17	39	38	37	28	17	0
1974–1978	54	62	60	55	42	27	38	45	43	39	29	0
1979–1983	32	61	63	59	52	45	21	41	45	39	34	22
1984–1988	42	62	65	64	58	51	25	46	47	44	39	0
1989–1993	49	63	69	68	63	48	36	47	51	47	37	28
1994–1998	62	78	79	73	62	65	39	60	59	52	41	27
1999–2003	78	88	87	79	70	74	52	72	67	57	44	36

Numbers in italics indicate that two or more cells had to be combined to get sufficient number of patients to calculate survival.

*Too few patients to calculate survival, see ref. [15].

Figure 2. Trends in age-standardised (ICSS) excess death rates per 100 person years for prostate cancer by country and time since diagnosis in Nordic cancer survival study 1964–2003.

Testicular cancer

Incidence and mortality. While incidence rates have increased continuously in the Nordic countries, mortality began to decline in the mid- to late-1970s. Figure 3 indicates that the increasing trends in incidence vary both in terms of their magnitude and level. Historically, Norway and Denmark have had higher incidence levels and a higher rate of increase than seen in Finland and Sweden. In Denmark however, the increase in incidence has attenuated and there is evidence of a decline beginning towards the end of the 1990s. Trends in incidence vary with age (data not shown) and a negative trend is seen among men above age 60 in Finland, contrary to the other countries, where trends are either constant or, as in Norway, slightly increasing. Mortality is low relative to incidence, and the declining trends have reduced rates to levels seen in the early-1980s, across all Nordic populations.

Figure 3. Trends in age-standardised (World) incidence and mortality rates per 100 000 and age-standardised (ICSS) 5-year relative survival for testis cancer by country. Nordic cancer survival study 1964–2003.

Survival. Concomitant with the decline in mortality, increases in survival are observed in all the Nordic countries. Men recently diagnosed with testicular cancer have a quite favourable prognosis, with a relative 5-year survival ranging from 88% in Finland to 94% in Sweden (Figure 3 and Table III).

Table III. Trends in survival for testis cancer by country. Number of tumours (N) included and the 5-year age-standardised (ICSS) relative survival in percent with 95% confidence intervals (RS (CI)). Nordic cancer survival study 1964–2003.

	Denmark		Finland		Iceland		Norway		Sweden
	N	RS (CI)	N	RS (CI)	N	RS (CI)	N	RS (CI)	N	RS (CI)
S-year
1964–1968	602	57 (52–62)	115	49 (40–60)	7	*	328	64 (56–73)	501	66 (61–71)
1969–1973	784	59 (55–64)	173	57 (48–67)	15	*	425	55 (50–59)	588	67 (61–72)
1974–1978	991	74 (69–79)	206	59 (53–66)	17	*	492	71 (64–78)	656	69 (65–74)
1979–1983	1 075	83 (79–88)	206	73 (65–83)	15	*	611	82 (76–88)	769	88 (83–93)
1984–1988	1 217	84 (80–89)	295	80 (74–87)	38	*	772	87 (83–92)	902	90 (86–94)
1989–1993	1 366	89 (85–93)	348	84 (78–90)	45	*	957	87 (83–92)	1 075	93 (88–97)
1994–1998	1 464	91 (87–96)	377	87 (79–96)	44	*	1 032	91 (86–97)	1 140	92 (88–96)
1999–2003	1 338	93 (89–97)	482	88 (83–94)	41	*	1 184	92 (88–97)	1 249	94 (90–97)

*Too few patients to calculate survival, see ref. [15].

The excess mortality in relation to time from diagnosis has decreased, although there are larger variations in Finland and Norway during the first month after diagnosis (Figure 4). In the period 2–5 years after diagnosis, the excess mortality has been constant and at a low level during the entire period of observation. Studying the age-specific survival reveals that the improvements have taken place across all age groups. The most notable shift is seen among men in the age-span 0–29, where the 5-year survival rose from 40–60% to 90–100% in all countries (Table IV).

Figure 4. Trends in age-standardised (ICSS) excess death rates per 100 person years for testis cancer by country and time since diagnosis in Nordic cancer survival study 1964–2003. No Icelandic curves. Too few partients to calculate survival in Iceland.

Table IV. Trends in 5-year age-specific relative survival in percent after testis cancer by country. Nordic cancer survival study 1964–2003.

Age	5-year survival
	0–29	30–39	40–49	50–69	70–89	90+
Denmark
1964–1968	50	69	69	70	27	*
1969–1973	53	74	75	69	25	*
1974–1978	73	86	83	75	49	*
1979–1983	88	94	93	80	53	*
1984–1988	93	94	91	83	46	*
1989–1993	93	96	94	92	63	*
1994–1998	96	97	97	87	75	*
1999–2003	98	97	98	88	77	*
Finland
1964–1968	43	71	59	51	17	*
1969–1973	48	81	66	53	38	*
1974–1978	59	78	80	72	0	*
1979–1983	80	84	66	77	46	*
1984–1988	89	88	98	81	34	*
1989–1993	94	95	99	91	26	*
1994–1998	94	95	96	90	48	*
1999–2003	95	97	96	85	60	*
Norway
1964–1968	57	73	76	67	48	*
1969–1973	52	66	83	69	0	*
1974–1978	71	79	82	78	42	*
1979–1983	94	91	95	85	28	*
1984–1988	96	97	92	86	54	*
1989–1993	96	97	96	99	32	*
1994–1998	96	99	93	93	65	*
1999–2003	97	98	95	92	71	*
Sweden
1964–1968	58	74	79	77	43	*
1969–1973	60	86	71	60	56	*
1974–1978	71	73	81	69	52	*
1979–1983	91	90	88	80	89	*
1984–1988	95	94	94	87	74	*
1989–1993	97	98	97	88	80	*
1994–1998	97	99	96	91	66	*
1999–2003	98	98	97	94	76	*

*Too few patients to calculate survival, see ref. [15]. Iceland not shown for same reason.

Penile cancer

Incidence and mortality. The incidence and mortality of penile cancer and other male genital organs is low and stable over time in the Nordic countries (Figure 5). Age-standardised incidence has been approximately 0.9 per 100 000 since the 1960s and mortality just above 0.2 per 100 000 since the 1970s. In absolute terms, there are larger differences between countries with respect to incidence than mortality. Finland has a lower incidence (0.5–0.6 per 100 000) than Denmark, Norway and Sweden where rates are around 1.1–1.2 per 100 000. Incidence among Icelandic men was low before 1980 but rapidly increased subsequently. Mortality is stable and rather homogenous across the Nordic countries.

Figure 5. Trends in age-standardised (World) incidence and mortality rates per 100 000 and age-standardised (ICSS) 5-year relative survival for cancer of the penis and other male genital organs by country. Nordic cancer survival study 1964–2003.

Survival. Age-standardised survival ratios for patients diagnosed with cancer of the penis and other male genital organs are rather stable over time in Finland and Sweden, with lower survival observed in Finland. The age-standardised relative 5-year survival is 80% in Norway, 74% in Denmark, 70% in Sweden and 62% in Finland for patients diagnosed 1999–2003 (Table V). The time trends are rather stable but appear to have increased in Norway and Denmark during the 1980s (Figure 5). Iceland has too few cases to provide reliable estimates of survival.

Table V. Trends in survival for cancer of the penis and other male genital organs by country. Number of tumours (N) included and the 5-year age-standardised (ICSS) relative survival in percent with 95% confidence intervals (RS (CI)). Nordic cancer survival study 1964–2003.

	Denmark		Finland		Iceland		Norway		Sweden
	N	RS (CI)	N	RS (CI)	N	RS (CI)	N	RS (CI)	N	RS (CI)
S-year
1964–1968	191	63 (54–73)	62	64 (53–79)	5	*	114	61 (50–74)	269	70 (62–79)
1969–1973	189	65 (56–76)	69	64 (53–79)	3	*	120	68 (56–83)	320	69 (61–77)
1974–1978	188	72 (64–83)	72	64 (49–83)	2	*	127	88 (77–100)	288	73 (66–81)
1979–1983	219	72 (63–81)	65	67 (53–84)	5	*	130	76 (65–89)	302	69 (62–76)
1984–1988	229	70 (62–79)	72	65 (50–86)	5	*	136	69 (58–80)	315	70 (63–78)
1989–1993	196	63 (55–72)	97	65 (52–83)	10	*	125	74 (64–86)	325	71 (64–78)
1994–1998	235	74 (66–83)	98	65 (54–79)	9	*	196	79 (71–88)	324	70 (63–77)
1999–2003	265	74 (67–81)	110	62 (51–74)	9	*	176	80 (73–88)	425	70 (65–75)

*Too few patients to calculate survival, see ref. [15].

Studying excess mortality reveals large variations within and between countries over time, although the results should be interpreted with caution due to the inherent large random variation (Figure 6). Denmark and Finland have the highest number of excess deaths per 100 person years during the first month after diagnosis. Two to five years after diagnosis, patients from Finland have a higher excess mortality than the other countries irrespective of calendar period. The lowering of excess mortality in Norway is also notable during the first month after diagnosis.

Figure 6. Trends in age-standardised (ICSS) excess death rates per 100 person years for cancer of the penis and other male genital organs by country and time since diagnosis in Nordic cancer survival study 1964–2003. No Icelandic curves. Too few partients to calculate survival for Iceland.

Discussion

This study has presented the main trends in survival of patients diagnosed with male genital cancer, and has noted intriguing country-specific variations in each of the three major groupings under investigation. The small absolute differences in prostate cancer mortality contrast with the considerable variations in prostate cancer incidence and survival in the Nordic populations noted prior to the PSA era [17], and certainly this variability has widened during the past 15 years [1]. For this reason, prostate cancer survival estimates and trends must be interpreted with considerable caution. The steady increases in secular survival trends can be partly attributed to lead time bias, plus an increasing detection of nonlethal tumors and subsequent length bias up to the late-1980s. The effect has certainly intensified during the PSA testing era, while early diagnosis and curative treatment may also have had some impact in improving survival of patients with aggressive tumours.

A recent decline in prostate cancer mortality has become apparent in Norway and Finland, while there are signs of a stabilisation in Denmark and Sweden [3]. The present level of mortality rates does not however correlate terribly well with the levels of PSA testing in each country. There is however a close relation between use of PSA and cancer incidence in Sweden and Norway [1], and one can assume that the incidence trends (and to an increasingly greater extent, survival trends) largely follow PSA testing in the Nordic countries. PSA screening entails detection of prostate cancer in many men who would not have had been diagnosed during their lifetime in the absence of screening [18], and overdiagnosis in turn implies a considerable risk of overtreatment. Tretli et al. had already commented on the possibility of unnecessary treatment prior to the PSA upsurge, given the considerable variations in incidence and survival yet minor differences in mortality [17].

Through estimation of screening lead time and assumptions regarding the recent effects of curative treatment, mortality trends can be related to the effectiveness of PSA testing, but this is not the case for survival. The inherent country-specific effects of increasing detection makes interpretation of the prostate cancer survival estimates and trends presented herein very difficult. The steady increases in secular survival trends can be attributed to a number of factors including increasing awareness of prostate-related symptoms, better access to health care, as well as the more frequent use of surgical treatment for benign prostate hyperplasia [2], where early and non-lethal prostate cancer can be detected in asymptomatic males. The diagnostic pressure effect is intensified in the PSA testing era, a period where early diagnosis and curative treatment may also have had some impact in improving survival of patients with aggressive tumours [1].

The ranking of excess deaths within 30 days following diagnosis is in keeping with an inverse ordering of prostate cancer incidence rates in each Nordic country during the period 1999–2003. Excess deaths within one month may therefore be a marker of diagnostic intensity, although overdiagnosis will have an effect on decreasing excess hazard rates in later periods too. That Denmark continues to have a higher excess mortality up to five years after diagnosis may in part relate to a treatment policy that, in relative terms, is historically conservative [19]; the recent declines in excess mortality both 1–3 months and 2–5 years after treatment, would tend to support evidence that clinical management in Denmark has changed within the last decade [20].

Long-term survival in patients diagnosed with testicular cancer has been increasing since the early-1970s in each of the Nordic countries. The sustained increases in survival (and concomitant declines in mortality) were first observed in Denmark and Norway in the early-1970s, and soon after in Sweden and Finland [6]. The starting times and the substantive changes in the survival trends can in part be attributed to the improvements in survival of young and middle-aged patients following the introduction of cisplatin as a therapeutic agent for advanced germ-cell tumours [9,21,22]. As well as platinum-based chemotherapy, the current treatment modalities for germ cell tumours – radiotherapy and surgery – lead to high clinical cure rates of 99% or above in early stage disease and 90%, 75–80% and 50% in advanced disease with ‘good’, ‘intermediate’ and ‘poor’ prognostic criteria, respectively [23]. Power and colleagues have noted that a combination of further developments of cisplatin-based regimens, improvements in tumour imaging and surgical interventions of residual disease, together with a multidisciplinary approach to cancer care, have all probably contributed to the continuation of the declining mortality trends in the past three decades [24].

The general pattern with respect to cancers of the penis and other male genital organs is stable whether one examines incidence, mortality or survival. Survival varies from 80% in Norway for patients diagnosed 1999–2003 in Norway to just above 60% in Finland. It is not clear whether the poorer survival and higher excess mortality observed in Finland is real, a result of artefact, or generated by the inherent random variation. Certainly, interpretation is hampered by the small numbers of patients observed.

Table VI. Trends in 5-year age-specific relative survival in percent after cancer of the penis and other male genital organs by country. Nordic cancer survival study 1964–2003.

Age	5-year survival
	0–49	50–59	60–69	70–79	80–89	90+
Denmark
1964–1968	71	71	70	61	35	*
1969–1973	86	70	58	64	60	*
1974–1978	83	81	67	70	69	*
1979–1983	83	92	74	59	58	*
1984–1988	82	72	66	75	57	0
1989–1993	71	73	57	79	25	0
1994–1998	88	75	66	72	80	*
1999–2003	73	75	74	73	75	59
Finland
1964–1968	73	74	68	73	23	*
1969–1973	73	74	68	73	23	*
1974–1978	72	64	59	98	0	*
1979–1983	68	82	76	83	0	*
1984–1988	72	68	54	75	60	*
1989–1993	79	26	66	78	73	*
1994–1998	81	58	91	60	24	*
1999–2003	76	64	82	49	34	*
Norway
1964–1968	71	66	76	47	46	*
1969–1973	43	70	77	67	73	*
1974–1978	93	85	89	76	106	*
1979–1983	91	73	75	74	72	0
1984–1988	74	78	84	55	52	*
1989–1993	83	84	74	53	98	*
1994–1998	83	88	87	71	67	*
1999–2003	80	86	78	87	66	99
Sweden
1964–1968	82	79	67	64	66	*
1969–1973	81	68	70	71	53	*
1974–1978	85	83	81	75	33	*
1979–1983	84	81	74	61	48	*
1984–1988	79	77	72	61	71	0
1989–1993	73	82	79	64	55	0
1994–1998	69	78	63	74	65	71
1999–2003	78	72	65	65	76	70

Numbers in italics indicate that two or more cells had to be combined to get sufficient number of patients to calculate survival.

*Too few patients to calculate survival, see ref [15]. Iceland not shown for same reason.

In conclusion, the three cancer types studied here vary in terms of incidence, mortality and survival patterns, and the prospects for their prevention. The absolute incidence and relative trends of prostate cancer are aligned to the changing prevalence of PSA testing in each country, although a correlation with mortality rates is much less apparent. It is not clear whether GPs and urologists will change practice with respect to PSA testing in light of the results of European and US [25,26] randomised trials. The European Randomized Study of Screening for Prostate Cancer (ERSPC) showed that PSA-based screening reduced the rate of death from prostate cancer by 20% (up to 31% after adjusting for contamination effects [27]), after a median follow-up of nine years, but was associated with a high risk of overdiagnosis [26]. In absolute terms, the unadjusted risk estimates indicated about 1 410 men would need to be screened and of greater concern, 48 additional cases of prostate cancer would need to be treated to prevent a single death from the disease. Beyond diagnosis and treatment, the utility of risk-reduction strategies for prostate cancer is still under investigation [28].

The high level of sensitivity of germ cell cancer to a variety of chemotherapeutic agents means that tertiary prevention, while highly successful – Nordic 5-year survival is 95% or mDore – remains the solitary approach to testicular cancer control. Other strategies remain unfeasible at the present time, given the paucity of risk factors uncovered that are amenable to prevention.

For penile cancer, 5-year survival at 60–80% may be considered fairly reasonable, but prospects for primary prevention will depend on whether prophylactic vaccines for HPV are efficacious in men. Recent estimates suggest that such vaccines could potentially reduce the incidence burden of invasive penile cancers by one-third [11].

Acknowledgements

The Nordic Cancer Union (NCU) has financially supported the developement of the NORDCAN database and program, as well as the survival analyses in this project. Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.

Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.