PROBASE study

Prostate cancer is the most commonly diagnosed cancer accounting for approximately 60,000 new cases every year in Germany. However, it is only the third common cause of death from cancer [1,2]. The widespread use of PSA as screening method for prostate cancer has led to an increased incidence of prostate cancer which is accompanied by a shift towards earlier stages of detected cancers [3,4]. With respect to the demographic changes in industrialized countries the incidence of prostate cancer will continue to increase.

PSA screening: reduction of mortality but substantial overdiagnosis and overtreatment

In 2009, first results of the European Randomized Study of Screening for Prostate Cancer (ERSPC) trial have been reported [5]. During a median follow-up of 9 years the relative risk of dying from prostate cancer was 0.8 (95% confidence interval 0.65 – 0.98, adjusted p-value=0.04) in the screening arm. But to prevent one death from prostate cancer 1,410 men had to be screened and further 48 men had to be treated for prostate cancer by surgery and radiotherapy, respectively. Consequently, the realized 20% relative reduction of prostate cancer mortality by PSA screening came along with a considerable risk of overdiagnosis and overtreatment. The relative reduction in the risk of death from prostate cancer in the screening arm was slightly increased to 21% (95% confidence interval 0.68 to 0.91; p-value=0.001) after an additional follow-up of 2 years. After adjustment for noncompliance the relative reduction of prostate cancer mortality was 29%. The number needed to screen and the number needed to treat were 1055 and 37, respectively [6].

Recently, the Swedish part of the ERSPC trial (Göteborg trial) actually reported a 44% relative reduction of prostate cancer mortality in the screening group (95% confidence interval 0.28 – 0.68; p=0.002) with a number needed to treat of only 12 [7]. Despite these encouraging data, general population based PSA screening cannot be recommended at the moment due to concomitant risk of overdiagnosis and overtreatment.

The concept of a risk-adapted PSA screening according to a baseline PSA value

Risk-adapted PSA based screening strategies are potential tools to overcome this dilemma. Several longitudinal studies reported a strong association between early PSA and long-term risk of prostate cancer morbidity and mortality. PSA levels of 2 to 3 ng/ml in young men which frequently are cited as within the normal range, are associated with an increase in odds for prostate cancer of more than 19-fold [8]. Contrary, the probability to develop prostate cancer and more important to experience life-threatening disease during lifetime is markedly decreased if the “baseline” PSA value is very low (<1.6 ng/ml) at age 44-50. For young men with a baseline PSA value below median (0.6 ng/ml), two further PSA tests every 5 years would probably be sufficient to rule out the risk of death from prostate cancer [9].

The first study to prospectively evaluate a risk-adapted PSA screening according to the baseline PSA level

However, the optimal age to begin a risk-adapted PSA screening with screening intervals according to the level of the individual baseline PSA value has never been evaluated in a prospective randomized trial. The present trial is the first randomized trial to prospectively evaluate the concept of a baseline PSA. Furthermore, it is the first trial to define the beginning of a risk-adapted screening either at age 45 or 50. In addition, this trial will evaluate subsequent risk-adapted screening intervals according to the level of the individual baseline PSA value.

The main goal of the present trial is to demonstrate the superiority of a delayed risk-adapted PSA screening beginning at age 50 (study arm B) as compared to a risk-adapted PSA screening beginning at age 45 with respect to the specificity of the screening. The detection of metastatic prostate cancer is the main endpoint in a non-inferiority design (combined primary endpoint). The hypothesis is that young men undergoing a risk-adapted PSA screening at age 50 do not more frequently develop metastatic disease up to the age of 60 as compared to those men beginning a risk-adapted PSA screening at age 45. The screening will end at age 60 with both groups finishing 15 years of screening (those with 50 will have a delay of 5 years and 10 years of screening).

Prospective identification of a low-risk group

By this means the study aims to demonstrate that a delay of 5 years to begin a risk-adapted PSA screening according to the baseline PSA level does not increase the risk of dying from prostate cancer (secondary endpoint). Concurrently, through prospective evaluation of long term screening intervals (5-yearly) in a low-risk group which represents approximately 90% of all subjects and which is defined by the level of the baseline PSA value (<1.5 ng/ml) the rate of screening procedures (PSA tests, biopsies) and consecutive treatment procedures (surgery, radiotherapy) for the whole population could be further decreased. Thus, the present study considers itself as an “antiscreening trial” and aims to reduce overdiagnosis and overtreatment of prostate cancer.

To confirm this main hypothesis a prospective randomized trial comparing the cumulative incidence of metastatic prostate cancer up to the age of 60 in men beginning a risk-adapted PSA screening at age 45 or 50 is warranted. The recruitment of subjects will take place at four study sites and follows an inquiry at the local registers of residents. After obtaining informed consent eligible subjects at age 45 will be randomized into two screening groups following a 1:1 distribution for early first PSA testing at age 45 (study arm A) versus delayed first PSA testing at age 50 (study arm B). At enrolment, subjects of the delayed screening group will only be asked for family history and blood collection but no PSA test will be carried out. These subjects will be contacted yearly by questionnaires. Up to the age of 50 only a yearly digital rectal examination of the prostate will be offered to these subjects (digital rectal examination at age 45 represents the standard of care offered by general health insurances in Germany). The subjects of study arm B are requested to revisit the study site 5 years later for first PSA testing.

According to the individual level of the baseline PSA value subjects of both screening groups will be submitted to the same risk-adapted screening intervals. Subjects with a baseline PSA value <1.5 ng/ml undergo 5-yearly screening intervals as long as the PSA value remains <1.5 ng/ml in the following screening rounds. Subjects with a baseline PSA value of 1.5-2.99 ng/ml undergo 2-yearly screening intervals. Subjects having or exceeding the cut-off PSA value of 3.0 ng/ml at baseline or in one of the following screening rounds will be submitted to a multiparametric magnetic resonance imaging (MRI) and to a stereotactically guided targeted biopsy combined with a random biopsy of the prostate. The primary endpoint of this study is incidence of metastatic prostate cancer up to the age of 60 as judged by imaging and confirmatory biopsy of metastases.

Multicenter trial over 20 years with 50,000 participants to be enrolled

Within 15 years the estimated cumulative incidence of metastatic prostate cancer in men undergoing a risk-adapted PSA screening at age 45 is 0.13-0.15% [10,11]. With respect to sensitivity at most a 20% relative increase of metastatic prostate cancer is accepted for non-inferiority of a delayed risk-adapted PSA screening strategy beginning at age 50. Assuming a study power of 80%-90% the number of cases to be randomized at age 45 is approximately 50,000.

Complete recruitment of the study will be finished within 5 years. Thus, the whole study duration including 15 years of screening and follow-up is estimated to be 20 years (without data cleaning and write up).

This “antiscreening” trial aims to establish a standardized and risk-adapted prostate cancer early detection

Already in the first years the study will annually report the distribution of baseline PSA levels in young men and the tumor characteristics of detected cancers at a PSA cut-off value of 3.0 ng/ml. The study will give deep insights into the time-dependent natural course of PSA levels in healthy young men. This will allow clear identification of a low-risk group which needs no intense surveillance. From the 6th year of the study, when subjects of study arm B have their first screening and the low-risk subjects of study arm A undergo their first following screening round the study will be able to make first statements if young men with very low baseline PSA values can safely undergo 5-yearly screening intervals without having a risk for at least locally advanced or high grade cancers.

Accordingly, from the 11th year of the study the same applies to men beginning with a risk-adapted screening at age 50. Reported data in the course of the study could already have influence on general recommendations for prostate cancer early detection. Due to an enrolment period over 5 years the data derived from the study will become more solid with each passing year. Between the 16th and the 20th year of the study the participants will reach step by step the end of the study.

Final analysis of the primary endpoint will then clarify if a risk-adapted PSA screening can begin at age 50 instead of age 45. This is 10 years later as compared to a beginning at age 40 which is currently discussed in the German guideline (S3-Guideline). Additionally, by establishment of longterm screening intervals for the majority of men risk-adapted PSA screening would have major impact on patient´s anxiety, and socioeconomical considerations such as health care costs for prostate cancer screening.

Prospective evaluation of targeted prostate biopsies according to standardized multiparametric MRI examinations based on a simple scoring system

In a translational approach the present study pursues the goal to prospectively evaluate multiparametric MRI, and consequently, biopsy strategies based on these MRI findings. The main goal of this translational research project is to establish a standardized evaluation and reporting of multiparametric MRI examinations of the prostate. This study is the first to prospectively evaluate a simple and standardized scoring system as decision criterion for or against prostate biopsy within a multicenter study.

Establishhment of a national biobank for prostate cancer

Finally, repeated collection of body fluids (blood, urine) from the study subjects at each study visit as well as storage of tissue samples (prostate biopsies, biopsies of metastases, formalin-fixed paraffin embedded and fresh frozen tissue from radical prostatectomies) will allow prospective evaluation of molecular biomarkers in the diagnosis of prostate cancer. Early begin of sample collection – in most cases before initiation of prostate cancer – will give improved understanding of molecular pathomechanisms involved in initiation, progression, and clinical behavior of prostate cancer. Due to the high number of subjects within this trial and the repeated collection of samples the study will allow establishment of a large prostate cancer biobank in Germany.

PROBASE study – English Synopsis

References

[1] Jemal A, et al. Cancer statistics, 2010. Cancer J Clin. 2010;60:277-300

[2] Bray F, et al. Estimates of cancer incidence and mortality in Europe in 1995. Eur J Cancer. 2002;38:99-166

[3] Etzioni R, et al. Overdiagnosis due to prostate-specific antigen screening: lessons from U.S. prostate cancer incidence trends. J Natl Cancer Inst.2002;94:981-90

[4] Pashayan N, et al. Mean sojourn time, overdiagnosis, and reduction in advanced stage prostate cancer due to screening with PSA: implications of sojourn time on screening. Br J Cancer. 2009;100:1198-204

[5] Schröder FH, et al. Screening and prostate-cancer mortality in a randomized European study. N Engl J Med. 2009;360:1320-8

[6] Schröder FH, et al. Prostate-cancer mortality at 11 years of follow-up. N Engl J Med. 2012;366(11):981-90

[7] Hugosson J, et al. Mortality results from the Göteborg randomised population-based prostate-cancer screening trial. Lancet Oncol. 2010;11:725-32

[8] Lilja H, et al. Long-term prediction of prostate cancer up to 25 years before diagnosis of prostate cancer using prostate kallikreins measured at age 44 to 50 years. J Clin Oncol. 2007;25:431-6

[9] Lilja H, et al. Toward a rational strategy for prostate cancer screening based on long-term risk of prostate cancer metastases and death: Data from a large, unscreened, population-based cohort followed for up to 30 years. J Clin Oncol. 29: 2011 (suppl; abstr 4512)

[10] Distribution of PSA levels and different types of events from prostate cancer at different follow-up times in unscreened men at ages 45-49. Unpublished data from the Malmö Prevention Project With kind permission of Andrew J. Vickers and Hans Lilja, Memorial Sloan-Kettering Cancer Center, New York, USA

[11] Schröder FH, et al. Screening for prostate cancer decreases metastatic disease – findings from the European Randomized study of Screening for Prostate Cancer (ERSPC). Eur Urol. 2012;62(5):745-52