Elsevier

The Lancet Oncology

Volume 16, Issue 16, December 2015, Pages 1667-1676
The Lancet Oncology

Articles
Prostate cancer screening in men aged 50–69 years (STHLM3): a prospective population-based diagnostic study

https://doi.org/10.1016/S1470-2045(15)00361-7Get rights and content

Summary

Background

The prostate-specific antigen (PSA) test is used to screen for prostate cancer but has a high false-positive rate that translates into unnecessary prostate biopsies and overdiagnosis of low-risk prostate cancers. We aimed to develop and validate a model to identify high-risk prostate cancer (with a Gleason score of at least 7) with better test characteristics than that provided by PSA screening alone.

Methods

The Stockholm 3 (STHLM3) study is a prospective, population-based, paired, screen-positive, diagnostic study of men without prostate cancer aged 50–69 years randomly invited by date of birth from the Swedish Population Register kept by the Swedish Tax Agency. Men with prostate cancer at enrolment were excluded from the study. The predefined STHLM3 model (a combination of plasma protein biomarkers [PSA, free PSA, intact PSA, hK2, MSMB, MIC1], genetic polymorphisms [232 SNPs], and clinical variables [age, family, history, previous prostate biopsy, prostate exam]), and PSA concentration were both tested in all participants enrolled. The primary aim was to increase the specificity compared with PSA without decreasing the sensitivity to diagnose high-risk prostate cancer. The primary outcomes were number of detected high-risk cancers (sensitivity) and the number of performed prostate biopsies (specificity). The STHLM3 training cohort was used to train the STHLM3 model, which was prospectively tested in the STHLM3 validation cohort. Logistic regression was used to test for associations between biomarkers and clinical variables and prostate cancer with a Gleason score of at least 7. This study is registered with ISCRTN.com, number ISRCTN84445406.

Findings

The STHLM3 model performed significantly better than PSA alone for detection of cancers with a Gleason score of at least 7 (p<0·0001), the area under the curve was 0·56 (95% CI 0·55–0·60) with PSA alone and 0·74 (95% CI 0·72–0·75) with the STHLM3 model. All variables used in the STHLM3 model were significantly associated with prostate cancers with a Gleason score of at least 7 (p<0·05) in a multiple logistic regression model. At the same level of sensitivity as the PSA test using a cutoff of ≥3 ng/mL to diagnose high risk prostate cancer, use of the STHLM3 model could reduce the number of biopsies by 32% (95% CI 24–39) and could avoid 44% (35–54) of benign biopsies.

Interpretation

The STHLM3 model could reduce unnecessary biopsies without compromising the ability to diagnose prostate cancer with a Gleason score of at least 7, and could be a step towards personalised risk-based prostate cancer diagnostic programmes.

Funding

Stockholm County Council (Stockholms Läns Landsting).

Introduction

Levels of prostate-specific antigen (PSA) are widely used as an initial screening test for prostate cancer and is largely credited with the reduction in prostate cancer mortality reported during the past two decades.1, 2 Even with evidence that PSA-based screening has reduced prostate cancer mortality, no governmental body has recommended structured PSA testing because of the potential harms of overdiagnosis.3, 4 Nonetheless, opportunistic screening is frequent, resulting in 1 000 000 prostate biopsies annually in the USA.5 There is growing concern about the increasing incidence of serious infections caused by multidrug-resistant bacteria and related infectious complications (eg, 2% of patients develop septicaemia in Stockholm) after prostate biopsy.6, 7

Alternative plasma protein biomarkers have been proposed to address these concerns; however, none has been prospectively assessed in screening studies.8, 9 At least 100 single nucleotide polymorphisms (SNPs) have been identified, accounting for about 30% of the inherited risk for prostate cancer.10 Combination of PSA with a genetic score based on these SNPs has been suggested to increase the specificity of prostate cancer testing.11, 12

We report a risk-based model for prostate cancer screening that combines PSA, SNPs, clinical variables, and established and novel plasma protein biomarkers (the STHLM3 model). We aimed to assess whether the STHLM3 model could increase the specificity of detecting men with high risk (Gleason score of at least 7) prostate cancer and thus substantially reduce the proportion of men undergoing prostate biopsy, while maintaining the same sensitivity to detect high risk prostate cancers as the PSA test alone (which uses a threshold of at least 3 ng/mL).

Research in context

Evidence before this study

We searched PubMed between Jan 1, 2000, and 31 Dec, 2009, using search terms related to prostate cancer, biomarkers, and screening for articles written in English.

Prostate-specific antigen (PSA) is widely used as an initial screening test for prostate cancer and is largely credited with the reduction in prostate cancer mortality reported during the past two decades. The European Randomised Study of Screening for Prostate Cancer (ERSPC) showed a 21% reduction in prostate cancer mortality with structured PSA screening after 13 years. However, the poor specificity of PSA translates into many unnecessary prostate biopsies and overdiagnosis of low-risk prostate cancers. Subsequently, no governmental body has recommended structured PSA screening because of the potential harms of overdiagnosis.

PSA is the only biomarker that has been assessed prospectively and in randomised controlled trials. Alternative plasma protein biomarkers other than PSA have been proposed to address these concerns. In addition to total PSA, the other plasma protein biomarkers used in the STHLM3 study were selected by a systematic scientific literature search during 2010 and two subsequent validation studies using the STHLM2 cohort (appendix). Five additional biomarkers were selected based on their associations with the presence of prostate cancer with a Gleason score of at least 7. Urine-based markers were excluded because they would require a prostate massage, which makes them unsuitable to use in a population screening setting. On the basis of the literature and genetic assessment of previous studies, 254 SNPs were selected based on their association with prostate cancer risk.

Added value of this study

We showed that the STHLM3 model, a combination of plasma protein biomarkers, genetic polymorphisms, and clinical variables, can significantly improve prostate cancer screening specificity with the same sensitivity compared with the PSA testing using a cutoff of at least 3 ng/mL. The STHLM3 model can identify cancers with a Gleason score of at least 7 in men aged 50–69 years and identify clinically significant prostate cancers in the PSA concentration range of 1–3 ng/mL. The STHLM3 model also includes two novel biomarkers and genetics markers that have not been previously included in a prospective diagnostic study. Additionally, STHLM3 is the first large-scale diagnostic study in prostate cancer where biopsy decision is prospectively based on the results from the predefined STHLM3 model. Finally, STHLM3 is population based, thus minimising selection bias and increasing the generalisability.

Implications of all the available evidence

Together with the results from the European Randomized Study of Prostate Cancer, the findings from the STHLM3 study indicate that prostate cancer mortality can be reduced but with substantially fewer biopsies and reduced overdiagnosis.

Section snippets

Study design and participants

STHLM3 was a prospective, population-based, diagnostic study following a paired, screen-positive design, in which we compared the STHLM3 model with PSA in men aged 50–69 years from Stockholm, Sweden. Men, irrespective of any comorbidity except prostate cancer, were randomly selected by date of birth from the Swedish Population Register kept by the Swedish Tax Agency and invitations were posted to them. The two screening methods, PSA and the STHLM3 model, were both tested in each study

Results

Of 260 000 men aged 50–69 years with a residential address in Stockholm, 145 905 were randomly selected and invited to the study. 1633 men with prevalent prostate cancer at recruitment were excluded (figure 1). Between May 28, 2012, and May 31, 2013, 11 130 men were recruited to the STHLM3 training cohort. The results from the training cohort are summarised in the appendix.

113 082 men aged 50–69 years were invited to participate in the STHLM3 validation cohort and 1263 men were excluded because

Discussion

We have shown that a combination of plasma protein biomarkers, genetic polymorphisms, and clinical variables can improve the specificity of prostate cancer screening significantly compared with PSA in men aged 50–69 years. Use of the STHLM3 model in structured screening could reduce the number of prostate biopsy samples taken by about a third compared with the use of PSA screening. Importantly, this can be achieved without compromising the number of high risk cancers diagnosed. We identified an

References (36)

Cited by (0)

View full text