Testicular cancer incidence predictions in Europe 2010–2035: a rising burden despite population ageing

Abstract

Testicular cancer is the most common cancer among young men of European ancestry, with about one-third of all cases occurring in Europe. With the historically increasing trends in some high-incidence populations reported to have stabilised in recent years, we aimed to assess recent trends and predict the future testicular cancer incidence burden across Europe. We extracted testicular cancer (ICD-10 C62) incidence data from Cancer Incidence in Five Continents Volumes VII-XI and complemented this with data published by registries from 28 European countries. We predicted cancer incidence rates and the number of incident cases in Europe in the year 2035 using the NORDPRED age-period-cohort model. Testicular cancer incidence rates will increase in 21 out of 28 countries over the period 2010–2035, with trends attenuating in the high-incidence populations of Denmark, Norway, Switzerland and Austria. Although population ageing would be expected to reduce the number of cases, this demographic effect is outweighed by increasing risk, leading to an overall increase in the number of cases by 2035 in Europe, and by region (21%, 13%, and 32% in Northern, Western and Eastern Europe, respectively). Declines are however predicted in Italy and Spain, amounting to 12% less cases in 2035 in Southern Europe overall. In conclusion, the burden of testicular cancer incidence in Europe will continue to increase, particularly in historically lower-risk countries. The largest increase in the number of testicular cancer patients is predicted in Eastern Europe, where survival is lower, reinforcing the need to ensure the provision of effective treatment across Europe.

Introduction

Testicular cancer is the most common cancer in young men of European ancestry. Of the 71,000 new cases estimated globally in 2018 (1), over one-third occur in Europe, with the highest age-standardised (world) incidence rates (>10/100,000) observed in Norway and Denmark (2). The vast majority of diagnoses (95%) are germ cell tumours, most of which are seminoma and non-seminoma subtypes, with spermatocytic tumours representing less than 1% of cases (3;4). Seminomas and non-seminomas occur predominantly in males aged between 15 and 44 years, and arise from the same precursor, germ cell neoplasia in situ (GCNIS) (3). The proportions of seminoma and non-seminoma are generally consistent across cancer registries, with seminoma comprising just over half of testicular cancer cases in most populations (5) following a gradual increase in relative frequency during the last two decades (4). The pathogenesis of these tumours is related to developmental disorders (6), particularly cryptorchidism, which has been shown to increase testicular cancer risk by four-fold (7).

Given the relatively young age at onset of testicular cancer and marked geographic variability between and within world regions of the major subtypes, environmental exposures in utero and early-in-life – in particular to endocrine disrupting chemicals – have been postulated as risk factors for both seminoma and non-seminoma (8). Perinatal factors including inguinal hernia, twinning, maternal bleeding, birth order, and sibship size have also been reported as conferring an elevated risk of testicular cancer in meta-analyses (7;9). These factors are likely to be indirectly linked to testicular cancer development via shared intrauterine environmental exposures. However, their role in the reported cohort-specific increases in incidence over the second half of the last century and subsequent stabilization observed in certain high-risk populations over the turn of the century, remains unclear (10–12). The latter attenuation of incidence rates in recent cohorts reported in Denmark and Switzerland contrasts with the rapidly increasing rates seen in lower-risk populations, both in Northern (Finland) and Southern Europe (Slovenia and Croatia) (11–14). A previous study predicted that there would be 23,000 new cases of testicular cancer in Europe by the year 2025 (11;12;15). Given the uniformly increasing rates and possibility to include more recent temporal data across Europe, this study extends the prediction analysis to 2035, taking into account both recent trends in incidence rates alongside demographic projections.

Materials and methods

Data sources

We extracted new cases of testicular cancer (ICD-10 C62) from regional or national population-based cancer registries in Europe from Cancer Incidence in Five Continents (CI5) Volumes VII-XI (4), with at least fifteen consecutive years of data. To improve on the timeliness of this information, the dataset was supplemented with published data from cancer registry websites up to 2015, as well as the extended years available for the Nordic countries using the NORDCAN database (16–20). We also used cancer incidence data from the Russian Federation 1993–2016, and data from the National Cancer Registry of Ukraine for 2000–2002 period that were not published as part of the CI5 series (21;22). We used the UN Population Division estimates as the source of population data for both observed and projection periods (23).

The combined data covered 28 countries and 88% of the male population of Europe in the year 2010, of which 22 countries had national cancer registration (Table 1). Analyses for Cyprus, Estonia, Iceland and Malta are presented in the relevant Table, but not included graphically, given the inherent random variation associated with these relatively small populations (all with <1.5 million inhabitants).

Table 1:

Testis cancer incidence among men in 2010 and in 2035 in 28 European countries, Age standardized rate (W) per 100,000.

		Annual population (million)		period use in model	Number of new cases (Annual)		Age standardized rate (W)		Peak in ASR	Change in number of cases (%)	Change in number of cases due to population (%)	Change in number of cases due to risk (%)	Annual percent change of ASR (%)
country_label	coverage	2010	2035		2010	2035	2010	2035
Central and Eastern Europe
Belarus	100	4413836	4179189	1988–2012	123	154	2.3	3.2	2035	25.6	−15.8	41.4	1.3
Bulgaria	100	3627840	3002146	1998–2012	214	269	5.2	8.8	2035	25.3	−28.5	53.8	2.2
Czech Republic	100	5147709	5127491	1993–2012	465	440	7.6	8.8	2035	−5.4	−19.5	14.1	0.6
Poland	4	18529195	17208584	1998–2012	891	1678	4.3	9.5	2035	88.3	−25.3	113.6	3.3
Russian Federation	100	65867211	63931524	1993–2015	1400	1572	1.7	2.2	2035	12.3	−12.6	24.9	0.9
Slovakia	100	2633406	2574300	1988–2012	269	284	8.6	10.9	2035	5.5	−21.5	27	1
Ukraine	100	21138694	18470502	1998–2012	543	761	2.2	4.2	2035	40.1	−20.9	61.1	2.6
Northern Europe
Denmark	100	2747106	3042573	1993–2015	295	229	10.2	7.6	2010	−22.3	4.7	−27	−1.2
Estonia	100	619813	583894	1988–2012	23	36	3.3	6	2035	58.3	−18.5	76.8	2.5
Finland	100	2633496	2874204	1993–2015	135	315	5.1	10.5	2030	134	2.8	131.2	2.9
Iceland	100	161132	188567	1993–2015	10	14	5.5	7.9	2035	46.9	9.9	37	1.5
Ireland	100	2260064	2671816	1993–2012	177	194	6.8	7.1	2025	9.3	0.6	8.7	0.1
Latvia	100	943831	782338	1993–2010	38	44	3.4	4.6	2035	16.8	−23.9	40.7	1.2
Lithuania	100	1483139	1225478	1988–2012	44	72	2.6	5.3	2035	62.4	−26.3	88.8	2.9
Norway	100	2445841	3156234	1993–2015	301	335	11.5	10.3	2010	11.2	21.7	−10.5	−0.4
Sweden	100	4676566	5508767	2003–2015	330	340	6.9	5.7	2020	3.3	8.5	−5.2	−0.7
United Kingdom	95	31099747	35750115	1995–2012	2222	2733	6.6	7.5	2030	23	5.7	17.4	0.5
Southern Europe
Croatia	100	2100603	1837793	1993–2015	172	219	7.6	13.2	2035	27.7	−22.6	50.3	2.2
Cyprus	100	407664	657261	1998–2012	30	66	6.6	9.6	2035	123	52.6	70.3	1.5
Italy	6	29016922	28255167	1998–2012	1979	1574	6.5	5.9	2015	−20.5	−17.7	−2.8	−0.3
Malta	100	206859	220473	1993–2012	13	37	5.6	16.7	2035	180.3	−7.8	188.1	4.5
Slovenia	100	1014319	1015426	1993–2015	104	130	8.8	14.6	2035	25.1	−23	48.1	2
Spain	11	23132790	22404281	1998–2012	1347	967	5	4	2015	−28.2	−28.6	0.4	−0.8
Western Europe
Austria	100	4075960	4432736	1998–2012	354	277	7.6	5.7	2010	−21.9	−7.6	−14.2	−1.1
France	9	30933311	33907724	1998–2012	2354	2581	7.3	6.8	2020	9.6	0.8	8.9	−0.3
Germany	4	39656447	40537003	1998–2012	4040	4800	9.1	11.6	2035	18.8	−8.3	27.1	1
Netherlands	100	8219181	8843722	1989–2012	704	886	8.5	9.5	2025	25.8	−2.7	28.5	0.5
Switzerland	29	3856024	4678836	1998–2012	383	343	8.8	6.7	2010	−10.5	5.8	−16.3	−1.1

Statistical analysis

Age-standardised incidence rates (ASR) were estimated by the direct standardisation method using the world standard population [4]. For countries with subnational data, we included the registries with 25 years of data available. We applied the regional age-specific incidence rates to the national populations to obtain the estimated numbers of cases by age-group to enable computation of the national ASRs.

We fitted NORDPRED age-period-cohort models in R by country to predict the national number of new cases and rates of testicular cancer in 2035, partitioning the changes in numbers between those due to changing rates (risk), and those due to changing demographics (population structure) (24). The rates were predicted by extrapolating three- to five-year periods, and the number determined by model goodness-of-fit. We extrapolated the most recent 10-year linear trend, attenuating the slope by 25% and 50% in the second and third prediction periods, and by 75% in the fourth and fifth periods, as recommended by the authors following empirical validation (24). The numbers of new cases in 2035 were predicted by applying the predicted age-specific rates for 2033–2037 to the UN population forecasts for 2035 for each country (24). We present the annual numbers and ASRs for the five-year period centred on the year 2010, and compare these with those in the future period centred on the year 2035. We also present the predicted changes according to the respective risk and demographic change components.

Data availability statement

Data are derived from the public domain resources quoted above; see references 4 and 16–22.

Results

Observed and predicted incidence trends by region 1980–2035

Figure 1 shows the observed (starting from 1980) and predicted (up to 2035) testicular cancer incidence ASR for all ages, with Supplementary Figure 1 detailing incidence trends and predictions for age groups 0–34, 35–44 and 45+.Table 1 describes the changes in population, number of new cancer cases and age-standardised rates by country between 2010 and 2035. The predicted temporal patterns varied between, and to a lesser extent, within different regions in Europe. Overall, the ASR will rise in 20 of the 28 studied countries from 2010 to 2035, with the most notable increases observed in Eastern Europe (Figure 2).

Figure 1.

Trends in age-standardised (World) incidence rates of testicular cancer in European countries, 1980–2035: observed rates (solid line) vs. predicted rates (dashed line)

List of regional registries: France (Calvados, Doubs, Haut-Rhin, Manche, Herault, Isere and Somme), Germany (Hamburg and Saarland), Italy (Parma, Ragusa, Romagna and Modena, Varese), Poland (Kielce), Spain (Basque Country, Tarragona, Granada, Girona), Switzerland (Geneva, Neuchatel, St Gall-Appenzell, Vaud and Valais)

Figure 2.

Predicted changes in age-standardised (World) incidence rates of testicular cancer between 2010 and 2035 in Europe, by country and region.

Rates for 2010 and 2035 represent average rates for the five-year period centred on the respective year.

List of regional registries: France (Calvados, Doubs, Haut-Rhin, Manche, Herault, Isere and Somme), Germany (Hamburg and Saarland), Italy (Parma, Ragusa, Romagna and Modena, Varese), Poland (Kielce), Spain (Basque Country, Tarragona, Granada, Girona), Switzerland (Geneva, Neuchatel, St Gall-Appenzell, Vaud and Valais)

In Northern Europe, the peak year of the incidence rate was attained in 2009 in Norway (12.3/100,000) and in Denmark (11.3/100,000), and slightly later, in 2013 in Sweden (7.6/100,000), with subsequent declines in the ASR predicted up to 2035. In the U.K. and Ireland, incidence rates of <7/100,000 in the period 2008–2012 were predicted to modestly increase or stabilise, whereas the lower incidence rates in the Baltic countries (<3.5/100,000 in 2010), were predicted to increase up to 2035. The steepest increases in rates in Northern Europe were predicted for Finland, with an annual percent increase of almost 3% per annum estimated from 2010 to 2035.

In Western Europe, the highest ASRs in 2010 were seen in the Netherlands (8.5/100,000), Switzerland (8.8/100,000) and Germany (9.1/100,000). While the incidence rates in Germany are predicted to modestly increase, uniform declines in the ASR are predicted for Switzerland and Austria. In France and the Netherlands, the peak year of incidence rates is predicted to occur in the 2020s, followed by a decline in the years that follow.

The testicular cancer incidence rates in the Eastern European countries are of lower magnitude, except Slovakia (8.6/100,000) and the Czech Republic (7.6/100,000). Throughout the region, uniformly rising trends in incidence rates are predicted, with no indication of a stabilisation in the period up to 2035; the rises are most pronounced in Poland, with a 3% year-on-year increase predicted from 2010 to 2035. Different patterns are observed between South-East and South-West parts of Europe. While steeply increasing incidence rates have been predicted for Croatia and Slovenia, the ASR are predicted to decline in Spain and Italy.

Observed and predicted numbers of testicular cancer cases by region – 2010–2035

Changes in the number of testicular cancer cases by country are described in Table 1 and Figure 3. As the predicted number of new testicular cancer cases in 2035 in a given country reflect both changing rates and national population projections, evidently the countries with the largest populations will have the proportionally largest number of future testicular cancer patients e.g. Germany (21% of all cases in the 28 countries in 2010), France (12%), United Kingdom (12%), Italy (10%) and Spain (7%). Even in the Russian Federation, where rates are relatively low, the large population size means that 7% of all European cases in 2035 are predicted to occur in that country.

Figure 3.

Predicted changes in numbers of new cases of testicular cancer between 2010 and 2035 in Europe, partitioned into population and risk change.

Numbers for 2010 and 2035 represent average numbers for the five-year period centred on the respective year.

Due to population change: predicted change in numbers of cases due to population growth and ageing

Due to risk change: predicted change in numbers of cases due to changes of ASRs

List of regional registries: France (Calvados, Doubs, Haut-Rhin, Manche, Herault, Isere and Somme), Germany (Hamburg and Saarland), Italy (Parma, Ragusa, Romagna and Modena, Varese), Poland (Kielce), Spain (Basque Country, Tarragona, Granada, Girona), Switzerland (Geneva, Neuchatel, St Gall-Appenzell, Vaud and Valais)

Overall, an increasing burden in terms of the future number of cases is predicted in 20 of the 28 European countries by 2035, with the overall number of patients predicted to increase by 13%. This temporal shift occurs in Southern and Eastern Europe as well as the Baltic countries, despite the observation that the increasing rates are partly offset by the ageing of their respective populations, which serves to diminish the future number of males at risk given the disease occurs at younger ages from ages 20 to 44. Conversely, in Northern Europe, apart from the Baltic countries, the predicted demographic changes will contribute to an increased number of new cases (Figure 3). The overall number of testicular cancer cases is predicted to increase by approximately 21%, 13%, and 32% from 2010 to 2035 in Northern, Western and Eastern Europe respectively, while a decline in the future number of cases of around 12% is predicted in Southern Europe.

Discussion

While confirming previously-documented increases in both the incidence rates and numbers of testicular cancer cases in Europe over the next decades (15), this study reports the novel finding of a recent peak and a predicted decline in incidence rates in the four high-risk countries of Denmark, Norway, Switzerland and Austria. Despite this finding and the marked variability in the general patterns and trends among European populations that we report here, the aetiology of testicular cancer remains elusive, in part due to difficulties in obtaining robust exposure data related to the perinatal period (7;9). Recent postulated risk factors for the disease include environmental exposure to endocrine-disrupting chemicals (8), consistent with the predominance of cohort effects in testicular cancer incidence observed globally (12), marijuana use (25) as a risk factor for non-seminoma. The prevalence of the latter exposure is however relatively low and unvarying across Europe, relative to equivalent differences in the disease rates.

The geographical landscape of testicular cancer incidence in Europe appears to be changing. With some exceptions, we noted that, while rates stabilised in the highest-risk countries, the largest increases in rates were observed in lower-risk countries, including Finland, the Baltic countries and some countries in Eastern Europe. While the burden of most epithelial cancers is increasing as a result of the demographic effects of population ageing, testicular cancer presents a counterexample; the disease occurs mainly in young men aged 15–44, a population group that is diminishing in numbers in many European countries as population ageing continues. If incidence rates were to remain unchanged from the present year to 2035, we would anticipate a net reduction in the future number of testicular cancer patients. Instead, the rapidly rising incidence rates observed in a number of European populations offset the negative demographic impact, resulting in an overall increase of the number of predicted cases, estimated at 13% in the 28 European countries from 2010 to 2035. The converse scenario is observed in Norway and Denmark, where the incidence rates started to decline after a peak incidence of >12/100,000 accompanied by a growth in the male population aged 15–44; this is also observed in Switzerland and Sweden, although rates are somewhat lower. Several studies from the Nordic countries have documented lower cancer risk in first generation immigrants, overall and for the testicular cancer in particular (26–28). Thus, a possible explanation for the observed stabilization in incidence rates in these countries is the increased proportion of the first-generation immigrant population from lower-risk countries, as hypothesised by a study from Denmark (28).

The main strength of our study was the use of the recorded incidence data from longstanding population-based cancer registries in Europe to predict future trends; a limitation however, was the lack of data from a number of South-Eastern European countries. While few countries in Europe have zero registry coverage and their relative contribution to the overall testicular cancer burden is rather modest, clearly greater insight into the disease magnitude and trends in these countries would have been achieved were data more complete. Some of the long-established registries in Western European countries, such as France, Germany, Spain or Italy, cover only a proportion of the total population, increasing the uncertainty with regards to the representativeness of the estimates extrapolated to the whole country. In addition to the CI5 data – considered of high quality given the longstanding process of quality control – we also included data from the Russian Federation and Ukraine that were not included in CI5 (4). The rationale for their inclusion was the large contribution of these populous countries to the overall European cancer burden, as well as low risk of misclassification or under-diagnosis of testicular cancer in these countries, relative to other cancer sites.

An additional limitation of our study was the unavailability of data by histological type and stage. In a study of testicular cancer incidence in the US, the observed (1999–2012) and predicted (up to 2026) increases of testicular cancer incidence have largely been determined by non-seminoma (29). In contrast, a recent global analysis reported parallel trends in seminoma and non-seminoma incidence in most European countries, while in Denmark, Italy and the UK, the incidence of seminoma was increasing at a greater rate than non-seminoma (30). While different trends by histology type and age-group reported from some countries worldwide would merit further analysis by age-period-cohort modelling, this was outside the scope of our current study. Given the poor completeness and comparability of population-based data on stage at incidence in Europe (31), we did not attempt to collect these data across our study population.

Overall, our results highlight that the testicular cancer epidemic in Europe is far from over, and in the absence of prevention prospects, there remains an overwhelming need to optimise cancer care. While the disease is considered a “curable” cancer with a five-year relative survival mostly exceeding 90% in Europe, large disparities in the average prognosis following a testicular cancer diagnosis have been reported in different countries on the continent (32;33). Prognosis is particularly poor in the Baltic countries and in Eastern Europe. According to EUROCARE-5 results for 2005–2007 period, age-standardised relative five-year survival in Eastern Europe was 82%, compared to 90–93% in other parts of Europe (32). In studies from Estonia and Lithuania (included in Eastern Europe in EUROCARE), poorer survival was attributed to lesser access or greater delays to diagnosis and treatment, and to the unavailability of appropriate treatment (34;35). The factors associated with optimal testicular cancer survival are complex, including access to accurate and rapid diagnosis, treatment and patient follow-up (36). For successful treatment and long-term survival of testicular cancer stage at presentation, centralisation and treatment in high volume centres are of great importance. While in the early stage surgery is often the only necessary treatment modality, higher stages require chemotherapy, and in some patients also radiotherapy or advanced salvage surgery (37). Since the introduction of platinum-based chemotherapy in 1970s, the survival of testicular cancer has significantly increased (32;38), with even the poorest prognosis group patients reaching a 5-year survival of approximately 48% (37). Even in an optimal treatment scenario, the burden of disease is further increased by undesirable effects of treatment related to absence of one or both testicles, including sub-fertility, hypogonadism, sexual dysfunction, metabolic syndrome and osteoporosis, as well as secondary cancers, cardiovascular disease, peripheral neuropathy, ototoxicity and hepatotoxicity in patients treated by radiotherapy and chemotherapy, providing an additional incentive to prioritise early diagnosis (39;40).

Public awareness campaigns, such as Movember (41), can influence stage at presentation, shifting the distribution towards more localised and treatable disease. A study from UK reported a reduction of the mean initial size of tumour on pathologic specimen from 7.2 cm in 1975–85 to 4.1 cm 2007–12 (39), while regional and national population-based data from the Netherlands have documented large increases in the incidence of localised disease in recent years (42;43).

Adherence to treatment guidelines presents an important aspect to achieve and maintain excellent treatment results and to reduce long-term toxicity (40). Current guidelines from the European Association of Urology (44) support a uniformity of effective treatment pathways across European countries, at least for the lower stage tumours. However, several studies on adherence to guidelines from several European countries have reported varying results, and the need for knowledge dissemination among the wider urological community (45–47).

The large migration trends in Europe cannot be ignored. Although it has been reported that the immigration of populations from lower-risk countries can serve to stabilize incidence rates of testicular cancer, it is important to note that second generation immigrants commonly have similar risks to the host population (27;28). There is also marked migration of medical practitioners from lower- to higher-income countries, with, for example, around 10% of practicing physicians in the UK from European Economic Area (EEA) countries (48). A consequence is that lower-income countries can be faced with deteriorating number of clinicians, leading to limited access to health care and longer average waiting times (49).

To conclude, the burden of testicular cancer incidence in Europe is continuing to rise, particularly in historically lower-risk countries. The predicted increase in the number of patients by 2035 will be greatest in Eastern Europe, where cancer survival estimates are currently among the lowest. Better adherence to guidelines, ensuring the organisation and availability of treatment in centres of excellence, and the implementation of public awareness campaigns can all positively impact in improving treatment results and lowering the disease burden in the near future. In parallel, improving the coverage and quality of population-based cancer registries – currently covering only 60% of Europe (2) – is a priority that will further enhance the monitoring of the population-level patterns and trends of testicular cancer incidence between and within European countries, including the benchmarking of survival of this highly treatable disease.

Supplementary Material

Supplementary figure 1

Supplementary figure 1. Trends in age-standardised (World) incidence rates of testicular cancer in European countries, 1980–2035: observed rates (solid line) vs. predicted rates (dashed line) by age-group

List of regional registries: France (Calvados, Doubs, Haut-Rhin, Manche, Herault, Isere and Somme), Germany (Hamburg and Saarland), Italy (Parma, Ragusa, Romagna and Modena, Varese), Poland (Kielce), Spain (Basque Country, Tarragona, Granada, Girona), Switzerland (Geneva, Neuchatel, St Gall-Appenzell, Vaud and Valais)

Novelty and impact:

Over the period 2010–2035, testicular cancer incidence rates are predicted to rise in 21 of 28 European countries studied. The effect of population ageing is outweighed by increasing risk, leading to an overall 13% increase in the number of cases by 2035. The largest increase in the number of testicular cancer patients (32%) is predicted in Eastern Europe, where survival is lower, reinforcing the need to ensure the provision of effective treatment throughout Europe.

List of abbreviations:

ASR

Age-standardised incidence rates

CI5

Cancer Incidence in Five Continents (CI5)

GCNIS

germ cell neoplasia in situ