Prostate cancer incidence and mortality trends among elderly and adult Europeans

Abstract

Prostate cancer is a common malignancy primarily of elderly men, with incidence rates rapidly increasing, owing to the population ageing and the introduction of more sensitive diagnostic procedures. Although the effectiveness of a screening test remains controversial, the decreasing mortality rates, which recently emerged in the USA, may be partly attributable to the changes of patterns of care, thus suggesting a potential effect of preventive measure. The object of this study is to examine time trends in incidence and mortality from prostate cancer in European Union (EU) countries, with particular attention to possible differences between the elderly (65 years old or over) and younger or middle age adults (35–64 years old). EUROCIM, the data base created by the European Network of Cancer Registries, provided the incidence and mortality data for the 12 EU Countries analysed (namely: Finland, Denmark, Scotland, England and Wales, Ireland and The Netherlands in Northern Europe; Austria, Germany and France in Central Europe; Italy, Spain and Portugal in Southern Europe), for the 1978–1994 period. Incidence and mortality time trends, expressed as mean difference per cent (MD%) per year, were estimated by a Poisson log-linear regression model. Higher resolution analyses were also carried out to check differences in time trends by age class within the two groups under study. Upward mortality trends occurred in several countries, excepting Ireland, Austria and Southern Europe, but only for younger and middle aged adults. Rates increased more rapidly in older age groups; a clear north–south gradient appeared both in the elderly and in younger adults; for the elderly, MD% higher than +1.5 for most countries of Northern Europe, MD% around +1 for Central Europe, and MD% less than +1 for Southern Europe were registered, with lower values for younger people. Incidence rates rose across the period considered, almost in all countries both for elderly and for younger and middle age adults, increasing more rapidly in younger age. Incidence trends showed a less clear geographic pattern than for mortality. In the younger group, high MD%, ranging in Northern Europe from +3.2 in Finland and England and Wales to +5.7 in The Netherlands, were observed, while in the South values ranged between +4.2 and +5.0. In Central Europe, very high MD%, ranging between +8.4 in France and +16.6 in Austria, were noted. No significant trends were observed for Denmark, Ireland and Portugal. For the elderly the increase was generally lower and no significant trend was observed in Germany and Portugal. Interaction between age and calendar period in the older group was observed for most of the considered countries. With reference to mortality, the MD% showed a tendency to rise, with increasing age, while no consistent pattern emerged for incidence. The observed incidence trends are probably a consequence of the different times in which the more recent detection methods were introduced in each country, and of the different policies adopted by each health care system towards the elderly. A comparison with the USA data suggests that in the next future a favourable downward mortality trend could be expected also in some EU Countries and, particularly, for younger age groups, even though prostatic cancer in old patients will remain a great burden, which National Health Care Systems will have to face in the next decades.

Introduction

In the late 1980s prostate cancer became the most diagnosed cancer in the USA and the second leading cause of neoplastic death in males, after lung cancer [1]. In EU countries it ranks, according to estimates by Ferlay et al. [2] for 1990, third as number of new cases per year, following lung and colorectal cancers (about 87 000, or over 12% of overall incident cases in males), and it represents the second most frequent cause of death, together with colorectal cancer and after lung cancer (about 50 000 deaths or 10% of overall neoplastic mortality).

When attention is focused on the elderly population over 75 years of age, prostate cancer becomes the most common neoplasm diagnosed in males with 44 353 incident cases, representing 21.4% of all new cases, estimated in 1990 [2].

Notwithstanding these figures and such a large health care burden, this cancer site is still characterised by some uncertainties and issues: not well defined risk factors, highly variable clinical course and controversial treatment options.

Furthermore agreement on screening effectiveness does not exist and randomised clinical trials may not be able to provide evidence for a positive role played by prostate-specific antigen (PSA) measure, digital rectal examination (DRE), transrectal ultrasonography (TRUS) and ecoguided biopsy (EGB) in secondary prevention, due to the already widespread use of these methods in the population [3].

However, in the meantime, descriptive epidemiology, analysing time trends in survival, incidence and mortality, could provide some useful indications for health care providers [3]; indeed mortality trends now occurring in the USA would seem to foresee a rather encouraging situation.

As a consequence of the introduction of new therapeutic technologies such as transurethral resection of prostate (TURP) [4] and of diagnostic modalities such as PSA, EGB and TRUS [5], [6], in the USA in the late 1980s a sharp increase in incidence rates occurred, followed, after a few years (between 1992 and 1993), by a drop of newly diagnosed cases [7], [8].

A similar pattern was not observed for mortality which, on the contrary, grew more slowly [9] and then reversed its trend, beginning recently to decrease: between 1990 and 1995 prostate cancer mortality declined from 26.5 to 17.3 per 100 000, with a percentage of decrease greater for younger than for older men [10], [11].

The purpose of this paper is to analyse incidence and mortality trends for prostatic cancer over the 1978–1994 period in EU countries, where, at least to our knowledge, the epidemiology of such disease is not much studied [12], in order to understand if, presently or by a next future, the findings reported for the USA are or will be observed also in Europe.

These analyses have been performed with a particular emphasis on possible differences existing between the elderly (65 years of age or older) and younger and middle aged adults (between 35 and 64 years of age) and, within the elderly group, among different quinquennial age groups, in order to test the hypothesis, already confirmed by North-American sources, of less favourable trends in the elderly [11].

Indeed the rapidly population ageing and the decrease of available financial resources, require a deeper knowledge of ‘cancer problem’ in these patients, with the purpose to provide health care planners with specific data.