Jose Manuel Cózar, Servicio de Urología, H. Virgen de las Nieves, Avda. de las Fuerzas Armadas, 2, 18014 Granada, Spain. e-mail: email@example.com
Study Type – Therapy (practice pattern evaluation)
Level of Evidence 2b
What's known on the subject? and What does the study add?
Prostate cancer (PCa) accounts for 12% of newly diagnosed cases of cancer in Europe. It is one of the most frequently diagnosed tumours in the developed world. Since the introduction of prostate specific antigen as a test for early detection of PCa, the rate of diagnosis has increased significantly and specific mortality has reduced in most western countries.
Most of the data on the incidence of PCa are obtained from population-based cancer registries which frequently do not cover the whole population. This first national hospital-based PCa registry aims not only to estimate the incidence of the disease but to ascertain the clinical profile of newly diagnosed PCa patients, a useful tool for evaluating the impact of the disease and its socio-health management.
• To estimate the 2010 incidence of prostate cancer (PCa) in Spain.
• To describe the clinical profile of newly diagnosed cases using a nationwide hospital-based registry.
PATIENTS AND METHODS
• This was a national epidemiological observational study in 25 public hospitals with a specific reference population according to the National Health System.
• Sociodemographic and clinical variables of all newly diagnosed, histopathologically confirmed PCa cases were collected in 2010, in the area of influence of each centre. Cases diagnosed in private practice were not collected (estimated nearly 10% in Spain).
• Data monitoring was external to guarantee quality and homogeneity.
• The age-standardized PCa incidence was determined based on the age distribution of the European standard population.
• In all, 4087 new cases of PCa were diagnosed for a reference population of 4 933 940 men (21.8% of the Spanish male population).
• The estimated age-standardized PCa incidence was 70.75 cases per 100 000 men.
• Mean age at diagnosis was 69 years; 11.6% of patients presented with tumour-related symptoms and 39.5% with LUTS. Median PSA was 8 ng/mL. Gleason score was ≤6 in 56.5%, 7 in 26.7% and >7 in 16.8% of patients. At diagnosis, 89.8% had localized, 6.4% locally advanced and 3.8% metastatic disease.
• This study on PCa incidence in Spain, a western country with intensive opportunistic PSA screening, shows that PCa is a high incidence tumour, diagnosed close to 70 years, usually asymptomatic.
• Almost 40% of cases have low risk disease with a risk of over-diagnosis and over-treatment.
• Around 55% of patients with intermediate or high risk disease are candidates for active therapy which may result in a reduction of cancer-specific mortality.
Cancer of the Prostate Strategic Urologic Research Endeavor.
Prostate cancer (PCa) accounts for 12% of newly diagnosed cases of cancer in Europe . It is one of the most prevalent tumours in Spain, and one of the most frequently diagnosed tumours in the developed world [2,3]. Since the introduction of PSA as a test for early detection of PCa, the rate of diagnosis has increased significantly and specific mortality has reduced in most western countries [4,5]. Despite the reduction of PCa-specific mortality, there is no consensus-based recommendation with regard to population screening due to the proven risk of over-diagnosis (diagnosis of indolent disease) and over-treatment in a considerable number of patients . This is due to the high mean age at which PCa is diagnosed and to the migration of the newly diagnosed cases towards earlier low and intermediate risk tumour stages .
Correct information on PCa incidence is decisive in estimating the impact of the disease and in establishing action priorities at policy and healthcare level, as well as for research.
Most of the data on the incidence of PCa are obtained from population-based cancer registries which frequently do not cover the whole population. The effect of a non-homogeneous geographical distribution on national estimates has already been addressed in the literature [8,9].
Studying actual incidence from direct estimations of a large proportion of the population of a country involves a major effort. However, in view of the adequacy of current clinical practice guidelines in force, it can provide additional data that are not obtained through population registries, such as descriptive epidemiological information on the characteristics of newly diagnosed patients and the quality of diagnostic care. There are very few initiatives that systematically collect descriptive data of registered patients. One example is the EUROCARE-4 project . In order to improve patient survival estimates, it includes data on tumour typology and morphology according to the third revision of the International Classification of Diseases for Oncology.
The Spanish National Health System divides the country into well-defined health areas, all with a known reference population . Based upon this organizational structure, the Spanish Urology Association set out a hospital-based PCa registry with the aim of both evaluating the incidence of the disease and ascertaining the clinical profile of newly diagnosed patients with PCa.
PATIENTS AND METHODS
In order to estimate the national PCa incidence, a nation-wide epidemiological study was designed within the National Health System setting. All new PCa cases diagnosed in 2010 were collected in a number of health areas, together covering >20% of the Spanish population.
The study protocol was approved by the Clinical Investigation Ethical Committee of the Hospital Virgen de las Nieves de Granada, and all patients signed the informed consent form.
A total of 25 public hospitals, each with a specific health area, participated in the study; these hospitals were selected in a way such that the age distribution of the reference population of each hospital was very similar to the national mean. This was verified by calculating the average weight of the population in each age group in the provinces where the selected centres are and comparing with the average of the whole country, as estimated by the current population of Spain in the provinces .
According to their regular clinical practice, each centre included the data of all patients with newly diagnosed histopathologically confirmed PCa in any stage, between 1 January and 31 December 2010. Cases diagnosed in the centre but not belonging to the hospital's reference population (i.e. patients living in another geographical/health area) were excluded. Thus, the reference population included 4 933 940 men belonging to the health area of the study hospitals, which represents 21.8% of the male population of Spain.
A study-specific case report form was used to collect patient sociodemographic variables (age, ethnic group, geographical area), clinical variables (weight, height, body mass index, familial history of PCa, patient symptoms, comorbidities, DRE, serum total PSA, prostate volume by transrectal ultrasound) and histopathological variables for confirmation of the PCa diagnosis (number of cores, Gleason score and clinical stage TNM) according to each centre's regular clinical practice. The data were obtained from the medical history and histopathological confirmation records from Anatomical Pathology Units. Data were monitored by an independent person in all centres to guarantee data quality. The 2009 TNM staging system was used for classification of the clinical stage  and the D’Amico classification  was used for distribution according to risk of progression.
The data from the case report form were collected in a database created specifically for the study using the SPSS 14.0 statistics software. Internal consistency rules and ranges were applied to the data to control inconsistencies and/or corrections in data collection and tabulation. Absolute and relative frequencies were used in the statistical analysis of the qualitative variables, and median and interquartile range or mean and standard deviation for the quantitative variables.
For the calculation of PCa incidence at national level, the weight of each area according to its reference population (obtained from the most recent hospital report available, 2009 or 2010) was taken into account. The distribution of men in the area of influence was assessed by assimilating it to the estimates of the corresponding province . The data were standardized in accordance with the standard European distribution .
A total of 4087 new cases of histologically confirmed PCa were diagnosed in the 25 centres during the study period for a reference population of 4 933 940 men. This results in an estimated incidence rate (95% CI) in Spain of 70.75 (68.71–73.17) cases per 100 000 men (age-adjusted based on the European standard population).
The mean age (standard deviation) of the newly diagnosed patients was 69 (8.15) years; 14.6% were under the age of 60 years, 41.3% were between 60 and 70, 20.3% were between 70 and 75 and 23.1% were aged above 75. Almost all cases (98.6%) were Caucasian. Some 232 patients (5.7%) had a familial history of PCa and in 93% of these cases the relatives were parents and/or siblings.
The distribution of the cases with regard to sociodemographic and clinical characteristics at anamnesis is displayed in Tables 1 and 2. In total, 475 patients (11.6%) presented with one or more tumour-related symptoms at diagnosis and 39.5% of patients had LUTS. A total of 2541 patients (62.2%) had one or more comorbidities.
A description of the histopathological variables at PCa diagnosis is given in Table 3. In 34.5% of the patients, DRE was abnormal. Diagnosis was reached by transrectal prostate needle biopsy in 4035 cases (98.7%), with eight or more cores obtained in 81.8% of the cases. In the other cases, tissue for histological confirmation was obtained by transurethral resection, adenomectomy or radical cystectomy. In terms of Gleason score, 56.5% of patients had a score of ≤6, 26.7% had a score of 7 and 16.8% had an index higher than 7.
Table 3. Histopathological variables at the time of PCa diagnosis (N= 4087)
Presence of defined limits according to DRE
Prostate mobility according to DRE
Number of cores by prostate biopsy (n= 3931)
7 (3 + 4)
The results of the prostate transrectal ultrasound showed a mean volume (sd) of 44.4 (23.0) mL. The median PSA was 8 ng/mL (interquartile range 5.63–13.55; total range 1.01–1.3%). Distribution according to PSA level showed that 62.9% of the cases had a level of PSA ≤ 10 ng/mL and 36.2% > 10 ng/mL (Fig. 1).
In addition, 65.4% of the cases were T1, 25.7% T2, 7.2% T3 and 1.5% T4 (Table 4). Distribution according to risk groups shows 37% of patients were low risk, 23% intermediate risk and 28.6% high risk for PCa (Table 5).
Table 4. Clinical stage of prostate cancer (N= 4087)
Primary tumour (T) (n= 4053)
Regional lymph nodes (N) (n= 4071)
Distance metastases (M) (n= 4070)
Table 5. Distribution according to risk groups
D’Amico classification (n= 4025)
Localized low risk
Localized intermediate risk
Localized high risk
Lymph node affection without metastasis
This is the first national registry in Spain assessing PCa incidence, in a regional registries setting where PSA screening is widespread nowadays. This is a public health registry in a country in the fifth position in EU population. The incidence of PCa in Spain adjusted to the age distribution of the European standard population is 70.75 cases per 100 000 men, as calculated by direct estimation from a hospital-based registry in the public health setting which, overall, covers a very considerable proportion of the country's male population. The numbers obtained could be translated into 16 040 new cases of PCa diagnosed every year. The organization of the Spanish National Health System into health areas allows for reliable incidence estimates. In addition, this national incidence study includes the clinical profile of the patients diagnosed, a useful tool for evaluating the impact of the disease and its socio-health management. Our estimate for the incidence of PCa (adjusted to the age distribution of the European standard population) is lower than that obtained by other authors based on population-based registries (86.6 per 100 000 men and year in 2004 and 77.2 in 2006) [2,8] and the indirect estimates based on available mortality and survival data (86 per 100 000 men and year for 2006 and 98 for 2012) . This may be due to differences in the estimation methods, as well as the different geographical distribution of the participating centres with regard to the population-based registries and the fact that cases diagnosed in the private healthcare setting were not collected in this study.
Although population-based registries should in principle provide more thorough information than hospital-based registries, the population covered by the centres in this study accounts for 21.8% of the total male Spanish population, including important population areas such as the provinces of Barcelona and Madrid that are not covered in population-based cancer registries. Moreover, and as a study limitation, it must be mentioned that the data here do not take into account cases diagnosed in the private healthcare setting, although the National Health System has universal coverage.
Another limitation could be the difference in diagnostic methods or biopsy techniques between the hospitals, although a high percentage of centres use the standards of the European Association of Urology guidelines. To our knowledge, there are no national prospective studies that offer information on the clinical profile of newly diagnosed patients. However, these data are of great importance in making forecasts and planning healthcare resources, as well as in the development of research policies.
At the time of diagnosis, 71% of patients were over 65 years old, and a scant percentage of cases were diagnosed in under 60s (14.6%). This is in accordance with what has been described in previous studies [17,18]. The mean age at diagnosis in this study is 69 years and hence, by age, the patients diagnosed would be potential candidates for some type of therapeutic approach. The mean age is similar to that recorded in the USA in the Surveillance, Epidemiology and End Results (SEER) and Cancer of the Prostate Strategic Urologic Research Endeavor (CaPSURE) databases [19,20].
The percentage of patients with localized disease (89.8%) is slightly higher than in the USA for the SEER database (81%), while the percentage of metastatic cancer is similar (close to 4%) .
With regard to tumour-related symptoms present at diagnosis, the low frequency of haematuria (3.62%), bone pain (2.40%), spinal cord compression (0.17%) and constitutional syndrome (2.42%) is remarkable, but can be explained by the high percentage of cases diagnosed in early stages of the disease (65% in stage T1). Furthermore, similarly to the general population, the most frequent comorbidity in newly diagnosed patients with PCa is cardiovascular disease (48.2%) such as hypertension, ischaemic heart disease, arrhythmia, valvulopathy, followed by endocrine-metabolic disease (24.8%).
Since 37% of newly diagnosed patients present with clinically low risk disease and a significant number suffer from comorbidities, there is a risk of over-diagnosis and over-treatment; the potential survival advantage provided by treatments with curative intent is minimized by the impact of life expectancy/competing mortality risks in this group. The percentage of low risk patients found is somewhat below the percentage found in the USA in the CaPSURE database (37% vs 46%), possibly the result of a less intensive screening .
Early diagnosis and/or opportunistic screening seem to be effective in view of the reduction in mortality in countries with these practices . In fact, our study shows that 51.6% of newly diagnosed patients belong to the intermediate risk (23.1%) and high risk (28.6%) groups in which the disease is potentially life-threatening if life expectancy is long enough. The percentages of intermediate and high risk patients found in the CaPSURE database are 27% and 25%, respectively , so our data are similar. The data on incidence together with the related sociodemographic and clinical data allow us to estimate the consumption of resources and related costs. For example, it could be estimated how many newly diagnosed patients would be potential candidates for hormone therapy as appropriate treatment . If, on the basis of current recommendations, we include high risk clinically localized tumours, locally advanced and metastatic tumours in this group, around 39% of the newly diagnosed patients would be potential candidates.
Finally, the results on the practice of indications and biopsy methods reflect a good adaptation to current European PCa guideline recommendations [23,24].
This study was supported by an unrestricted grant from Astellas Pharma S.A. The authors would like to acknowledge all the participating investigators: C.H.U.A.C (Juan Andrés González Dacal), Complejo Hospitalario Regional Carlos Haya (Victor Baena González, Pedro Morales Jiménez), Fundaciò Puigvert (Joan Palou Redorta), H. de Basurto (Miguel Unda Urzáiz), H. Clínic i Provincial (Antonio Alcaraz Asensio, Alex Ciudin), H. Clínico Univ. de Valencia (José Ma Martínez Jabaloyas, Cristina Ferrandis Cortés), H. Clínico Univ. Lozano Blesa (Francisco Javier Romero, Jorge Subirá Ríos, José Gabriel Valdivia Uría), H. General Univ. de Alicante (Juan José Lobato Encinas, Jesús Jiménez Navarro, Luís Pérez Llorca), H. General Univ. de Valencia (Emilio Marqués Vidal, Joaquín Ulises Juan Escudero), H. General Univ. Gregorio Marañón (Carlos Hernández Fernández, Felipe Herranz, Adrián Husillos Alonso, Juan Tabares Jiménez,), H. General Univ. Morales Meseguer (Emilio Izquierdo Morejón), H. Univ. 12 de Octubre (Juan Passas Martínez, Laura Díez Sicilia), H. Univ. Central de Asturias (Miguel Ángel Hevia, José Ma Abascal), H. Univ. de Bellvitge (José Francisco Suárez Novo, Manel Castells Esteve), H. Univ. de Canarias (Pedro Rodríguez Hernández, Tomás Concepción Masip, Ana Cristina Plata Bello), H. Univ. Infanta Cristina (Simón Asuar Aydillo, Eduardo Laguna Álvarez, Juan Alonso Cabo), H. Univ. Puerta de Hierro-Majadahonda (Joaquín Carballido, Claudio Martínez Ballesteros), H. Univ. Ramón y Cajal (Javier Burgos Revilla, Rafael Rodríguez Patrón, Vital Hevia Palacios, Sara Álvarez Rodríguez), H. Univ. Reina Sofía (Ma José Requena Tapia, Rafael Prieto Castro, José Luís Carazo Carazo, Beatriz Santiago Agredano), H. Univ. Río Hortega (José Amón Sesmero, Consuelo Conde Redondo, Luís Álvarez Buitrago, Verónica Rodríguez Tesedo, Alberto Rivero Cárdenes, Juan Francisco Sánchez García), H. Univ. Virgen del Rocío (Rafael Medina López, Manuel Conde), H. Univ. Virgen Macarena (Jesús Castiñeiras Fernández, Antonio Carlos González Baena, José Manuel Carmona Soto), H. Universitari Vall d'Hebron (Juan Morote Robles, Carles Xavier Raventós Busquets), H. Virgen de la Salud (Antonio Gómez Rodríguez, Elena Buendía, Natanael García Betancourth). We would also like to acknowledge Nieves Pérez for monitoring of the study and Marta Prieto for her effort and support in the initial stage of the project.
CONFLICT OF INTEREST
Arancha Cantalapiedra and Emilio Pedrosa are Employees of Astellas Pharma.