Sperm cryopreservation incidence in men with testicular cancer: towards a stabilization in testicular cancer incidence? Results from the CECOS network

Over the past quarter century and particularly in the last decade, the incidence of testicular cancer (TC) has been rising rapidly while at the same time major advances in therapeutic management have led to improved prognoses and survival rates [1–4]. Although it is a relatively rare disease which accounts for approximately 1% of all new male cancer cases, TC remains the most frequent cancer in adolescents and young adults in industrialized countries [5–8].

An important issue for these young men is how TC and its treatment will affect, transiently or permanently, their future fertility [9, 10]. So far, banking samples of semen before treatment is considered as the most appropriate approach [11]. In France, a unique public network of sperm banks (the Centre d’Etude et de Conservation des Oeufs et du Sperme humain, CECOS) was established in 1973 and covers the whole country through 24 affiliated regional sperm banks. We recently published the trends for 1973–2007 in sperm cryopreservation for TC throughout this French network, showing an increasing use of sperm banking during this period [12].

Assessment of TC incidence rate is a worldwide key public health objective in order to predict future trends towards increase or stabilization of this disease, the rate of incidence increase according to country, and regional trends. Several important facts must be highlighted. Firstly, the highest TC incidence rates are observed in North European countries: Norway (12.7 per 100,000 person-years), Denmark (13.4 per 100,000), but rates are also high in East European countries (9.4 per 100,000 in Slovenia) [13]. Secondly, the steepest increases of incidence are often observed in countries with a formerly lower incidence (especially southern European countries: 5.5% per year in Spain), and affecting countries with the highest TC incidence rates (for example Switzerland with an annual increase of 1.8%, and an incidence rate of 12.7 per 100,000) [13]. Thirdly, during the last decade the rise in incidence rates has stabilized in the United Kingdom, Denmark and Austria [13].

In France, the TC incidence rate was 5.7 per 100,000 in 2000–2004 [8]. However, estimations of testicular cancer incidence were based on specific cancer registries in some regions (six registries). No countrywide cancer register exists in France. Thus, registration of annual TC sperm bank procedures, especially in countries such as France which have a single standardized sperm bank network, provides an opportunity to assess indirectly global TC incidence, and consequently to formulate hypotheses regarding TC incidence trends.

The aim of our study was to estimate testicular cancer incidence trends from time trends in use of sperm cryopreservation over the period 1990–2013 by men with testicular cancer.

In the Fig. 1, two models were performed using the incidence of men with TC who used sperm cryopreservation (gray point) between 1990 and 2013: the Poisson regression model (blue line with confidence intervals) and the Verhulst model (single orange line). The estimated incidence of men with TC in France (INCa data) is shown in green (dashed line with triangle).

The overall incidence of sperm cryopreservation rose from 1.73 to 5.57 per 100,000 between 1990 and 2013. The maximum incidence was observed in 2011 with 6.14 per 100,000.

The Poisson regression model predicted an incidence rising from 2.64 to 6.76 per 100,000 between 1990 and 2013, and suggested that the incidence would reach 9.00 per 100,000 [95% confidence interval CI = 8.66–9.34] in 2020. The annual percent change was estimated at 4.16% [95% CI = 3.98–4.39]. The Poisson regression model suggested that the incidence would increase after 2013. However, the Figure shows that the predictions performed by this model do not seem to fit the trend and were above the observed incidence of sperm cryopreservation. After 2005, the data are in favor of a pause in the increase of sperm cryopreservation incidence.

The Verhulst model was more in line with the observed data, showing attenuation in the incidence of sperm cryopreservation after 2005. Predictions performed by this model showed an incidence of sperm cryopreservation rising from 1.96 to 5.81 per 100,000 between 1990 and 2013, and to 5.98 per 100,000 in 2020. The Verhulst model estimated that the maximum incidence of sperm cryopreservation will be 6 per 100,000 after 2020, which will represent 1100 men visiting a sperm bank.

The trends estimated by the Verhulst model are much closer to the estimated TC incidence based on the National Cancer Institute data than the Poisson regression trends.

Our results showed strong evidence for increasing use of sperm cryopreservation by men with TC between 1990 and 2013, rising from 1.73 to 5.57 per 100,000 person-years. Interestingly, this annual increase is in agreement with the TC incidence observed in other studies worldwide [8, 13, 19].

However, we observed a slight attenuation beginning after 2005, as the incidence decreased from 5.58 in 2005 to 5.57 in 2013. A similar decrease in the incidence rate of TC has recently been reported in other countries in Northern Europe, in the United States and Australia [8, 20]. Although this study does not show attenuation in France, the testicular cancer data were only collected up to 2004. The same remark could be made regarding the study of Le Cornet et al. [19]. They showed that the greatest increase of TC incidence was in France, with over 2500 new cases estimated in 2025. However, this estimation was calculated for the period 1988–2007. In the light of these findings, our observed trends and attenuation need to be strengthened by prospective studies on TC.

In order to explain this transition, Znaor et al. studied generational transitions in 38 countries and showed that a deceleration in risk occurred in younger generations born in the 1960s and 1970s, but only in the most developed countries [20]. This generational effect had already been observed for men born during World War II. It has been suggested that modifications in lifestyle exposure to etiological factors for TC could account for these two changes.

In our study, we used two methods to model the number of sperm cryopreservations for TC between 1990 and 2013. Poisson regression is a commonly used method for modeling count data in cancers such as TC (rare events with few new cases per year [16]). However, in this method, the estimated coefficient is a constant rate parameter which cannot take into account changes in data trend. Sperm cryopreservation observations throughout the CECOS network showed a higher increase before 2005 and an attenuation afterwards. The second method, the Verhulst model, assumes that the growth rate (or incidence) will tend to the value K, which is here the maximum sperm cryopreservation incidence. This model could therefore take into account trend changes in incidence and therefore the attenuation in sperm cryopreservation before TC that was recorded by the CECOS sperm bank network.

Sperm cryopreservation incidence in TC confirms a trend toward a slight attenuation or plateau in the TC incidence rate. With regard to TC incidence, the Verhulst model seems more efficient than the Poisson model. As a matter of public health policy, it is important to follow up TC incidence and sperm cryopreservation so that the decrease in this disease can be confirmed or refuted. In the meantime, it remains and will always remain important to identify the causes of TC and trends in the incidence of the disease, and to protect the health and future fertility of young men.