World Journal of Urology

, Volume 30, Issue 2, pp 251–256

Biopsy and treatment decisions in the initial management of prostate cancer and the role of PCA3; a systematic analysis of expert opinion

Authors

    • Service d’UrologieCliniques Universitaires Saint-Luc UCL
  • Filip Ameye
    • UZ Gasthuisberg
  • Alexandre de la Taille
    • Hôpital Henri Mondor
  • Theo de Reijke
    • Academic Medical Center
  • Paolo Gontero
    • Ospedale MolinetteUniversità di Torino
  • Alexander Haese
    • Martini Clinic Prostate Cancer CenterUniversity Clinic Eppendorf
  • Paul Kil
    • St. Elisabeth Ziekenhuis
  • Paul Perrin
    • Hôpital Henry Gabrielle
  • Mesut Remzi
    • Medical University of Vienna
  • Jörg Schröder
    • ATURO Uroonkologische Gemeinschaftspraxis
  • Mark Speakman
    • Taunton and Somerset Hospital
  • Alessandro Volpe
    • Maggiore della Carità Hospital
  • Bianca Meesen
    • Ismar Healthcare NV
  • Herman Stoevelaar
    • Ismar Healthcare NV
Original Article

DOI: 10.1007/s00345-011-0721-0

Cite this article as:
Tombal, B., Ameye, F., de la Taille, A. et al. World J Urol (2012) 30: 251. doi:10.1007/s00345-011-0721-0

Abstract

Purpose

The Prostate CAncer gene 3 (PCA3) assay may guide prostate biopsy decisions and predict prostate cancer (PCa) aggressiveness. This study explored the appropriateness of (1) PCA3 testing; (2) biopsy; (3) active surveillance (AS) and the value of the PCA3 Score for biopsy and AS decisions.

Methods

Using the RAND/UCLA appropriateness method, 12 urologists assessed the appropriateness of PCA3, biopsy and AS for theoretical patient profiles, constructed by combining clinical variables. They individually scored the appropriateness for all profiles using a 9-point scale. Based on the median score and extent of agreement, the appropriateness for each profile was calculated.

Results

The PCA3 Assay was mainly considered appropriate in men with ≥1 negative biopsy, PSA ≥ 3 ng/mL and life expectancy (LE) ≥10 years. A LE < 10 years, ≥2 negative biopsies and PCA3 Score <20 decreased biopsy appropriateness, while PSA ≥ 3 ng/mL and PCA3 Score >50 increased it. In men without a prior biopsy, LE ≥ 10 years and a suspicious DRE, PCA3 did not affect biopsy appropriateness. In other men, a PCA3 Score <20 discouraged biopsy, while a value ≥35 supported biopsy. AS was mainly considered appropriate if LE < 10 years, T1c PCa, ≤20% positive cores and PSA < 3 ng/mL. A PCA3 Score <20 pleads for and higher scores (particularly >50) against AS.

Conclusions

These findings illustrate in which men PCA3 can be of additional value when taking biopsy and treatment decisions in clinical practice.

Keywords

Active surveillanceBiopsyExpertOpinionPCA3RAND appropriateness method

Introduction

The decision to perform a prostate biopsy for diagnosis of prostate cancer (PCa) commonly involves digital rectal examination (DRE), serum prostate-specific antigen (PSA), prostate size and family history. Despite the availability of risk calculators and nomograms, the decision to biopsy remains difficult and often uncertain. Because of the low specificity of PSA, about 75% of biopsies in men with a PSA 2.5–10 ng/mL will be negative [1, 2]. Considering that biopsy may cause pain, discomfort, anxiety and other (severe) complications, reducing the number of unnecessary biopsies has become a challenge in PCa diagnosis [3, 4]. Many of these men with, e.g., a persistent elevated PSA level will be confronted with one or more repeat biopsies, of which also only 10–35% will be positive. The PROGENSA® Prostate CAncer gene 3 (PCA3) Assay, alone or combined with traditional tools (e.g. PSA), may aid in guiding biopsy decisions and reducing the number of unnecessary biopsies [58]. In daily clinical practice, the PCA3 Assay is increasingly being used for this purpose [912].

Not all cancers detected through screening or early detection strategies are life-threatening and need rapid treatment. Indeed, about one-third of detected cancers may harbour features of insignificant PCa [13]. As a diagnosis of PCa causes anxiety, many of these ‘indolent’ cancers will receive active treatment [13]. Current identification of indolent disease using standard markers such as Gleason score, PSA, DRE and number of positive biopsy cores may misclassify aggressive tumours [14]. This may lead to repeat biopsies, often with extended schemes, to detect upstaging or upgrading early in the course of the disease [15, 16]. Therefore, another important challenge in current urological practice is to better identify men in whom active surveillance (AS) may be appropriate. Increasing evidence suggests that the PROGENSA PCA3 Assay may aid in identifying men with features of insignificant PCa, who could be candidates for AS [5, 17, 18].

In daily practice, biopsy and treatment decisions should be based on scientific evidence. However, clinical studies and guidelines are often not sufficiently detailed to provide clear guidance on biopsy or AS decisions for the wide range of patients seen in daily practice. As a consequence, the decision-making process is often also based on the physician’s personal insights and patient preferences. Since thousands of patients are treated each day, this clinical expertise may, at an aggregate level, generate valuable information to complement scientific evidence. This study aimed to systematically examine the clinical expertise of physicians heavily involved in PCa diagnosis and treatment in order to develop patient-specific criteria for biopsy decisions, AS decisions and to explore the potential role of PCA3 in taking these decisions.

Materials and methods

RAND appropriateness method (RAM)

The Research and Development/University of California at Los Angeles (RAND/UCLA) appropriateness method (RAM) [19, 20] was used to systematically canvass clinical expertise, as described later. More information about the RAM method can be found in Online Resource 1. A European panel of 12 urologists heavily involved in PCa diagnosis, and treatment rated the appropriateness of (1) the PCA3 test for the decision to biopsy, (2) biopsy decisions and how knowledge of the PCA3 Score could influence the decision to biopsy, and (3) AS and how knowledge of the PCA3 Score could influence AS decisions in patients with low-risk PCa (PSA < 10 ng/mL, stage T1c-T2a, Gleason sum <7) for a systematically constructed set of hypothetical patient profiles. These profiles were unique combinations of clinical variables predicting PCa risk or considered relevant in AS decisions (Online Resource 2). Variable selection was based on a literature review describing variables taken into account in biopsy and treatment decisions in clinical practice (see Online Resource 3) and was agreed upon by the physicians prior to the rating process. PCA3 Score cut-offs were also assessed based on data from this literature review (Chap. 9 of Online Resource 3). The appropriateness of the PCA3 Assay for guiding biopsy decisions was rated for 108 theoretical patient cases, the appropriateness of biopsy was rated for 540 theoretical patient cases, and the appropriateness of AS was rated for 64 theoretical patient cases (Online Resource 2). An electronic program was used to individually assess the appropriateness on a 9-point scale (1 = very inappropriate; 9 = very appropriate, 5 = uncertain). A procedure was considered appropriate if the expected benefits outweighed the expected risks or negative consequences by a sufficient margin that the procedure was worth doing [19]. Financial costs or other potential constraints had to be disregarded. Two rating rounds were held. The results of the first round were discussed during a meeting leading to refinements of clinical variables for the construction of patient profiles. Thereafter, a second individual rating was performed. Based on the median score and extent of expert agreement, appropriateness statements were calculated according to RAND/UCLA rules [20]. Decisions were deemed appropriate if the median panel score ranged 7–9 and inappropriate if the median ranged 1–3, both without expert disagreement. Disagreement was defined as the situation in which at least four experts scored in each of the Sects. 1–3 and 7–9. All outcomes not categorized as ‘appropriate’ or ‘inappropriate’ were labelled ‘uncertain’.

Statistical analysis

The relationship between PCa risk factors and expert appropriateness ratings was examined using logistic regression analysis.

Results

Appropriateness of PCA3 Assay for biopsy decisions

The appropriateness of the PCA3 Assay for guiding biopsy decisions was rated for 108 theoretical patient cases and was considered appropriate in 40% of cases. Disagreement among panel members was 11%. The PCA3 Assay was particularly considered appropriate in men with ≥1 negative prior biopsy, PSA ≥ 3 ng/mL and life expectancy ≥10 years (Table 1).
Table 1

Appropriateness of the PCA3 test for the decision to biopsy by clinical variables

Variable

Value

% Inappropriate

% Uncertain

% Appropriate

Life expectancy (years)

≥10

28

24

48

<10

28

41

31

DRE

Normal

33

33

34

Suspicious

22

32

46

Prior biopsy

Not performed

56

44

0

1 negative biopsy

14

28

58

≥2 negative biopsies

14

25

61

Volume (cc)

<30

22

42

36

30–60

30

28

42

>60

30

28

42

PSA (ng/mL)

<3

50

39

11

3–10

8

36

56

>10

25

22

53

Total

 

28

32

40

DRE digital rectal examination, PSA prostate-specific antigen

Appropriateness of biopsy and impact of the PCA3 Score on biopsy decisions

The appropriateness of biopsy was rated for 540 theoretical patient cases and was considered appropriate in 42% of cases. Disagreement among panellists was 1%. Logistic regression analysis showed that nearly all clinical variables had a significant impact on the biopsy appropriateness (Fig. 1). A life expectancy <10 years, ≥2 negative prior biopsies and a PCA3 Score <20 significantly decreased biopsy appropriateness, while a PSA ≥ 3 ng/mL and a PCA3 Score >50 strongly increased it (Fig. 1). The PCA3 Score had no impact on biopsy appropriateness in men with no prior biopsy, a life expectancy ≥10 years and a suspicious DRE. Figure 2 gives an overview of the 108 hypothetical patient cases in which knowledge of the PCA3 Score changed the appropriateness of biopsy decisions; this was mainly the case when biopsy was considered inappropriate or uncertain. For example, if the decision to biopsy was considered inappropriate, a PCA3 Score >50 would change the decision in favour of biopsy in men with a PSA < 3 ng/mL and/or a normal DRE. In uncertain biopsy decisions, a PCA3 Score <20 changed this into an inappropriate decision in many cases, e.g., for no prior biopsy, life expectancy <10 years, normal DRE and PSA 3–10 ng/mL.
https://static-content.springer.com/image/art%3A10.1007%2Fs00345-011-0721-0/MediaObjects/345_2011_721_Fig1_HTML.gif
Fig. 1

Relationship between clinical variables and appropriateness of biopsy (logistic regression analysis)

https://static-content.springer.com/image/art%3A10.1007%2Fs00345-011-0721-0/MediaObjects/345_2011_721_Fig2_HTML.gif
Fig. 2

Diagram displaying how knowledge of the PCA3 Score changed the appropriateness of the decision to biopsy and in which patient profiles

Appropriateness of AS and impact of the PCA3 Score on AS decisions

The appropriateness of AS was rated for 64 theoretical cases of men at low risk of significant PCa. It was considered appropriate for 22% of cases, and disagreement among panellists was 6%. AS was considered particularly appropriate in men with a life expectancy <10 years, stage T1c PCa, ≤20% positive cores and a PSA < 3 ng/mL (Table 2). A PCA3 Score <20 supported the decision for AS in men with a life expectancy <10 years, T2a PCa, and >20% positive cores and/or a PSA of 3–10 ng/mL as well as in men with a life expectancy ≥10 years, T1c PCa and ≤20% positive cores. A PCA3 Score >20 (and in particular >50) advocated against AS in men with a life expectancy ≥10 years, T1c PCa and >20% of positive cores or T2a PCa and ≤20% positive cores.
Table 2

Appropriateness of active surveillance by clinical variables

Variable

Value

% Inappropriate

% Uncertain

% Appropriate

Life expectancy (years)

≥10

56

38

6

<10

3

59

38

Clinical stage

T1c

25

47

28

T2a

34

50

16

% Positive cores

≤20

19

53

28

>20

40

44

16

PSA (ng/mL)

<3

25

50

25

3–10

34

47

19

PCA3 Score

Not performed

19

56

25

<20

6

31

63

20–50

38

62

0

>50

56

44

0

Total

 

30

48

22

PCA3 Prostate CAncer antigen 3, PSA prostate-specific antigen

Discussion

Clinical studies have shown that the PCA3 Assay may be used as an additional tool to guide biopsy decisions [58] and could also aid in selecting men for AS [5, 17, 18]. In daily practice, the PCA3 Assay is increasingly being used [912]. Since thousands of patients are treated each day, aggregated clinical expertise from physicians using the PCA3 Assay in their daily clinical practice may provide additional valuable information to evidence-based information on how this test can be integrated in current initial management of PCa. This study systematically investigated the clinical expertise of physicians involved in PCa diagnosis and treatment. Overall, there was a considerable extent of agreement between experts, particularly regarding biopsy decisions. The PCA3 test was mainly considered appropriate in men with ≥1 negative prior biopsy, a PSA ≥ 3 ng/mL and a life expectancy ≥10 years. Biopsy was particularly considered appropriate in case of a PSA ≥ 3 ng/mL and a PCA3 Score >50 and considered inappropriate in case of a life expectancy <10 years, ≥2 negative biopsies and a PCA3 Score <20. Knowledge of the PCA3 Score did not change biopsy decisions in men with no prior biopsy, a life expectancy ≥10 years and a suspicious DRE as these men are obvious biopsy candidates, independent of PSA level and prostate volume. However, knowledge of the PCA3 Score can make a difference in biopsy decisions, particularly when these are considered inappropriate or uncertain. In many uncertain decisions, a PCA3 Score <20 or <35 avoided a biopsy, while a PCA3 Score ≥35 (and in particular >50) often supported the biopsy decision.

AS was particularly considered appropriate in men with a life expectancy <10 years, stage T1c PCa, ≤20% positive cores and a PSA <3 ng/mL. A PCA3 Score <20 may support the decision for AS, while a PCA3 Score >20 and particularly >50 advocated against AS. In this respect, the PCA3 Score can be considered useful for patient counselling, but this has to be tested in well-designed trials.

In this study, various cut-offs of the PCA3 Score were used, which were based on clinical studies [5, 6, 17, 18]. However, the most appropriate PCA3 Score cut-off for guiding biopsy and AS decisions is still under discussion and should be further evaluated in well-designed prospective trials [21].

It should also be noted that the patient profiles used in this study were hypothetical cases of which the prevalence may vary in clinical practice. For example, if a biopsy was considered inappropriate, a PCA3 Score >50 would change this decision to appropriate in men with a PSA <3 ng/mL and/or a normal DRE. However, the prevalence of these men in daily urological practice is likely to be low, and the chance that these men would need a PCA3 test can be considered small. Furthermore, in men with no prior biopsy, a life expectancy <10 years, a normal DRE, and a PSA 3–10 ng/mL, a PCA3 Score <20 changed the decision from uncertain to inappropriate independent of prostate volume. In this study, this concerned only few hypothetical cases, but this patient profile is likely to be very prevalent in clinical practice. The prevalence of different patient profiles in daily clinical practice should be further evaluated.

A limitation of this study is that the applicability of the results in daily clinical practice may be complicated by the large number of patient profiles. This can be resolved by the development of a computer program or algorithm as a decision aid allowing urologists to compare their own decisions to the expert recommendations. This will be further evaluated.

The appropriateness ratings reflect the expert panel decisions on the likelihood that the subject may be a good candidate for the PCA3 test, biopsy or AS and the potential impact of the PCA3 Score on biopsy and AS decisions. The study is thus also limited by the selection of panel members. However, it has been demonstrated that different expert panels, even from different countries, often reach similar outcomes [22]. In clinical practice, this expert-based evidence can be an aid in clinical decision making and be combined with scientific evidence and patient preferences to make final therapeutic decisions.

Conclusions

The results of this systematic analysis of expert decisions may be useful in complementing evidence-based information in the grey/uncertain area of biopsy and AS decisions in daily clinical practice. In cases where biopsy is considered uncertain or inappropriate, the PCA3 Score can provide additional information for guiding biopsy decisions.

Acknowledgments

This study was supported by an unconditional educational grant of Gen-Probe Incorporated. The authors are grateful to Ismar Healthcare NV for their assistance in the development of the study and the analysis of the results and in the writing of the manuscript. The publication of this article was supported by Gen-Probe Incorporated.

Conflict of interest

The following authors have declared conflicts of interest: B. Tombal: consultant to Gen-Probe, A. de la Taille: investigator for Gen-Probe, A Haese: Investigator for Gen-Probe, M. Speakman: consultant and investigator for Astellas, GSK and Pfizer, investigator for Bayer and Gen-Probe, the other authors have no conflicts of interest to declare.

Supplementary material

345_2011_721_MOESM1_ESM.docx (12 kb)
Supplementary material 1 (DOCX 12 kb)
345_2011_721_MOESM2_ESM.pdf (6.1 mb)
Patient profiles (PDF 6196 kb)
345_2011_721_MOESM3_ESM.pdf (562 kb)
Literature Review: Initial Management of Prostate Cancer (PDF 561 kb)

Copyright information

© Springer-Verlag 2011