Urinary engrailed-2 (EN2) levels predict tumour volume in men undergoing radical prostatectomy for prostate cancer

Authors


Hardev Pandha, Oncology, Postgraduate Medical School, Daphne Jackson Road, Guildford GU2 7WG, UK. e-mail: h.pandha@surrey.ac.uk

Abstract

What's known on the subject? and What does the study add?

There are a lot of potential prostate cancer biomarkers being evaluated. All aim to improve on the sensitivity and specificity of PSA. EN2 was recently shown by our group to have better sensitivity and specificity than PSA. EN2 is a simple ELISA test and is not dependent on other parameters, even PSA, unlike all the other current biomarkers under evaluation. To date, no marker correlates with the amount of cancer present – the present study shows this positive correlation with EN2 in men undergoing prostatectomy.

The potential utility of this work is that by knowing that the level of EN2 corresponds to the amount of cancer present, irrelevant of tumour grade and number of cancer foci, we can define an EN2 level corresponding to small cancers, which can then undergo surveillance. We are conducting a further study that is aimed at determining whether the levels of EN2 in urine can indicate ‘significant’ vs ‘non-significant cancer’ using the threshold of 0.5 mL cancer (after Epstein's work).

OBJECTIVES

  • • To evaluate the relationship between levels of a recently described prostate cancer biomarker engrailed-2 (EN2) in urine and cancer volume in men who had undergone radical prostatectomy (RP) for prostate cancer.
  • • To date, prostate-specific antigen (PSA) levels have not reliably predicted prostate cancer volume. Reliable volume indicator biomarker(s) may aid management decisions, e.g. active treatment vs active surveillance.

PATIENTS AND METHODS

  • • Archived patient samples from the Aarhus Prostate Cancer Project, Denmark, were assessed.
  • • Pre-treatment mid-stream urines, without preceding prostatic massage, were collected and stored at –80 °C.
  • • Urinary EN2 levels were measured by a recently published enzyme-linked immunosorbent assay.

RESULTS

  • • In all, 88 of the whole cohort of 125 men (70%) were positive for EN2 in their urine (>42.5 µg/L); 38/58 (65%) men where cancer volume data was available.
  • • There was no statistical relationship between urinary EN2 levels and serum PSA levels.
  • • PSA levels did not correlate with tumour stage, combined Gleason grade, total prostatic weight or cancer volume.
  • • There was a strong statistical relationship between urinary EN2 and prostate cancer volume by linear regression (P= 0.006).
  • • Higher EN2 levels correlated with tumour stage T1 vs T2 (P= 0.027).

CONCLUSIONS

  • • Pre-surgical urinary EN2 levels were associated with increasing tumour stage and closely reflected the volume of cancer in RP specimens.
  • • Given the ease of collection (no prostatic massage required) and the simplicity, low cost and robustness of the assay, EN2 may become a useful biomarker in not only identifying which patients have prostate cancer but may also facilitate risk stratification by indicating the burden of tumour volume.
Abbreviations
AS

active surveillance

RP

radical prostatectomy

ESPRC

European Randomized study of Screening for Prostate Cancer

EN2

engrailed-2

ROC

receiver operator characteristic

PCA3

prostate cancer antigen 3

INTRODUCTION

Prostate cancer remains a significant health problem in the western world. There has been rapid progress in the understanding of the pathobiology of prostate cancer in recent years, but progress in diagnostic and prognostic biomarkers have been less impressive. There is still an urgent need not only to improve diagnostic sensitivity and specificity, but also to determine which cancers may be monitored and which require prompt treatment. PSA has been widely used as a diagnostic, screening and monitoring tool since it was first approved by the USA Food and Drug Administration as an aid to the early detection of prostate cancer in 1986. However, PSA levels at diagnosis do not correlate with tumour volume and clinical outcome [1,2]. The identification and validation of new markers, which robustly predict significant from insignificant disease, would have enormous utility and focus active treatment or active surveillance (AS) appropriately, whilst avoiding unnecessary intervention.

Early studies suggested that a key factor determining the malignant potential of prostate cancer is tumour volume at diagnosis. Indeed, studies early in the PSA era showed a robust correlation between serum PSA levels and tumour volume [3,4]. Later studies failed to support these findings [2]. The designation of a detected cancer as ‘significant’ (suitable for treatment) or ‘insignificant’ (suitable for AS) remains inconsistent amongst clinicians, despite the development of tools such as nomograms [5]. This is probably due to the classification of some men as having indolent disease on TRUS-guided biopsies, although they had actually had significant cancer [6,7]. Although tumour volume is not universally accepted as an independent predictor of prognosis in larger tumours, it is widely acknowledged that a low-grade tumour of <0.5 mL is not clinically significant. i.e. will not give rise to symptoms or shorten life [8]. Recent analysis of patients after radical prostatectomy (RP) within the European Randomized study of Screening for Prostate Cancer (ESPRC) study suggests that the threshold volume for significance may be higher: 1.3 mL and 2.5 mL for index lesion and total volume, respectively [9].

We have recently reported the potential diagnostic utility of engrailed-2 (EN2), a transcription factor involved in early embryonic development, which is subsequently re-expressed in prostate cancer [10]. We have shown that EN2 was detectable in the urine of patients with prostate cancer by ELISA without the requirement for prostatic massage beforehand. The presence of EN2 in urine was highly predictive of prostate cancer, with a sensitivity of 66% and a specificity of 88.2% (area under the curve of 0.81) [10].

This ‘proof of principle’ study of a retrospective series is the first attempt to examine the potential of urinary EN2 to predict tumour volume. The cohort of patients assessed had relatively high-risk disease. We investigated the relationship between urinary EN2 levels before RP, tumour volume and pathological features in RP specimens in men who had undergone RP for prostate cancer.

PATIENTS AND METHODS

The archived patient samples evaluated were from the Aarhus Prostate Cancer Project biobank, Department of Urology, Aarhus University Hospital, Denmark. The study received institutional approval (ethics study number: 2000 0299); all patients gave written informed consent. Samples were collected between 2006 and 2009. The main criterion for inclusion were planned RP for clinically localised prostate cancer. Patients with non-organ confined disease and histologically confirmed second malignancies of any type were excluded. The diagnosis of prostate cancer was established based on 6–12 trans-rectal prostate biopsies. Moderate – and high-risk patients according to the D'Amico risk stratification [11] routinely completed a bone scan before RP and in case of doubt underwent an additional pelvic MRI to exclude the presence of metastases. All clinicopathological data were systematically and prospectively registered and stored in the Aarhus database (national data authority number: 2007-41-1220). Mid-stream urines (10 mL) without prior prostatic massage were collected 1 day before RP. Urine was collected a minimum of 6 weeks after the prostate biopsy procedure. The urine samples were centrifuged, separated and both supernatant and pellets stored at –80 °C.

All RP specimens were evaluated by a specialist in genitourinary pathology. In brief, the gland was fixed in 10% formalin, paraffin embedded, and sectioned at 3-mm intervals before whole sections were mounted on slides. A complete sampling procedure was performed. The apical portion, the basis, and the neck of the prostate were separated from the rest of the gland and sampled with a cone technique. Tumours were graded using the Gleason grading system (International Society of Urological Pathology revised version) and staged using the TNM classification. Each focus was outlined on the histology slices, and tumour volume was calculated by multiplying the area by the slice thickness.

EN2 levels were determined by ELISA as previously published [10]. Samples were anonymised to laboratory staff undertaking the testing in terms of clinicopathological associations. Briefly, two monoclonal mouse anti-EN2 antibodies, APS1 and APS2 were raised using the synthetically produced C-terminal 100 amino acids (Biosynthesis Inc, Lewisville, Texas, USA) of EN2 as an antigen (Antibody Production Services Ltd, Haywards Heath, Sussex, UK). APS2-biotin was captured onto a 96-well streptavidin-coated plate (Nunc 436014, Rochester, New York, USA) at a concentration of 4 µg/mL. A 100 µL aliquot of urine or a dilution of the EN2 fragment in buffer was used for the assay; all samples were tested in triplicate. The dilution series was used to generate a standard curve by which the concentration of EN2 in each sample was measured.

To test the significance of differences between mean EN2 concentrations in different patient groups (as defined by T-stage or Gleason grade) we used an unpaired t-test with Welch's correction. Correlations between PSA level and tumour volume/prostate weight, and EN2 and tumour volume/prostate weight, were calculated by linear regression. For these, the reported P value is for the slopes being significantly non-zero. Receiver operator characteristics (ROC) curves were generated for the EN2 and PSA values. The area under the ROC curve was tested for significance using an unpaired t-test against the hypothesis that the real area under the ROC curve was 0·5 (i.e. no diagnostic value).

RESULTS

PATIENT COHORT AND CLINICOBIOLOGICAL PARAMETERS

The cohort studied consisted of 125 men. Cancer volume data from the RP specimen was available in 58 cases; the remainder had not been assessed for total tumour volume. The characteristics of the entire cohort and their tumours studies are shown in Table 1. The mean (range) age was 63.9 (47–75) years, and the mean PSA level was 14.3 (4.3–51) µg/L. The mean prostatic weight was 51.8 g. The proportion of men with clinical stages T1c, T2a, T2b and T2c were 8%, 29%, 23%, and 40%, respectively. The mean (median, range) number of positive cores was 2 (2, 0–6). The mean (median, range) percentage of cancer per cores was 84 (33, 10–100)%. The median combined Gleason score was 7 in both biopsy and RP specimens. In all, 11 patients had their Gleason grading upgraded after RP; in 15 individuals the combined Gleason score of the RP specimen was less than the diagnostic biopsies. The RP Gleason scores included 3+3 in 16/58 men, 3+4 in 33/58 men, 4+3 in 6/58 men, 4+4 in 1/58 men, 3+5 in 1/58 men and 5+3 in 1/58 men.

Table 1.  Patient, tumour and marker characteristics. A threshold level for EN2 of 42.5 µg/L was used. The proportion of men with clinical stages T1c, T2a, T2b and T2c were 8%, 29%, 23%, and 40%, respectively
VariableMeanMedianRange
Age, years63.96547–75
PSA level, µg/L14.310.54.3–51
EN2 level, µg/L8042160–9332
Tumour volume, mL21.4151–80
Total prostatic weight, g51.847.528–156

In the RP specimens the pathological stages were Gleason 6 (27%), 7 (67%), and 8 (6%), respectively. The mean (median, range) tumour volume was 21.4 (20, 1–80) mL. Of the total 125 patients in the original cohort, 70% were positive for EN2 in their urine using the threshold level of 42.5 µg/L [10]. In the 58 patients where cancer volume data was available, 65% of patients had a positive urinary EN2 result. In these men the mean (median, range) EN2 level in urine was 917 (312, 0–9332) µ/L. Figures 1–7 reflect only the 58 patients where cancer volume was available from the RP specimen.

Figure 1.

Relationship between serum PSA and urinary EN2 levels. There was no significant correlation, P= 0.912.

Figure 2.

Relationship between urinary EN2 levels and tumour stage. T2 cancers were associated with significantly higher levels of EN2 compared with T1, P= 0.027.

Figure 3.

EN2 levels across pathological tumour stage. Although an increasing EN2 level was evident, the differences between each pathological stage were not statistically significant.

Figure 4.

Relationship between EN2 levels and combined Gleason grading in biopsy vs RP specimens. There were lower levels of EN2 with higher Gleason 8 but there were far fewer cases with Gleason 8 disease.

Figure 5.

Relationship between serum PSA levels and combined Gleason grading in biopsy vs RP specimens. There was no statistical differences across different combined Gleason grading in either biopsy or RP specimens.

Figure 6.

Relationship between EN2 level, PSA level and cancer volume in the RP specimens. a, Urinary EN2 levels were significantly associated with cancer volume by linear regression, P= 0.006 (95% CI for slope 9.344–52.77). b, There was no correlation between serum PSA levels and cancer volume, P= 0.107 (95% CI for slope 0.0785–0.7829).

Figure 7.

Relationship between EN2 level, PSA level and total prostate mass. There was no correlation between either EN2 or PSA levels and total prostate mass by linear regression.

CORRELATION BETWEEN SERUM PSA LEVEL, URINARY EN2 AND PATHOLOGICAL FINDINGS IN RP SPECIMENS

We investigated any possible relationships between serum PSA level, pathological data from the resected prostate glands, clinical and pathological stages and EN2 levels in the urine. Consistent with our previous study, there was no correlation between serum PSA level and levels of urinary EN2 (P= 0.912, Fig. 1). Higher levels of EN2 protein were detected in more advanced clinical and pathological stages (Figs 2,3). Comparing T1 vs all T2 cancers combined, a statistically higher level of EN2 was detected in the T2 cancers (P= 0.027). In contrast, although T1c cancers were associated with higher PSA levels, there were no statistical differences across the T and pT stages and when T1 cancers were compared to all T2 cancer combined.

The relationship between urinary EN2 and serum PSA level and Gleason grading of biopsies and RP specimen was examined. In the biopsies, higher levels of EN2 were associated with Gleason 7 disease, and much lower levels in Gleason 8 (although there were far fewer cases of Gleason 8; Fig. 4). This trend was not seen in the RP specimens, with equally higher levels in Gleason 6 and 7, and again lower levels in Gleason 8. Differences did not reach statistical significance in either group. PSA levels varied to a lesser extent than EN2 levels in relation to Gleason grading in biopsy and RP groups: in both cases this did not reach statistical significance across either group (Fig. 5).

CORRELATION OF URINARY EN2 LEVELS AND SERUM PSA WITH TUMOUR VOLUME

We compared the relationship between urinary EN2 levels and serum PSA level (both before RP) with tumour volume as calculated from the RP specimen. There was a highly statistically significant correlation between EN2 levels and tumour volume (P= 0.006), in a linear regression analysis. There was no such relationship between serum PSA levels and tumour volume, (P= 0.107; Fig. 6). Notably, we also evaluated the relationship of both PSA and EN2 levels with total prostate gland weight and there was no statistical correlation (Fig. 7).

DISCUSSION

There is an unmet need for prostate cancer biomarkers that have diagnostic and prognostic utility, and which can aid treatment decisions. The present ‘proof of principle’ study explored the relationship between pre-surgical levels of EN2 in urine and total tumour volume at RP. Although not all studies in the PSA era identify tumour volume at RP as an independent predictor of prognosis [12], it is nevertheless considered so important that all pathological definitions of clinically significant vs potentially harmless prostate cancers incorporate tumour burden [13,14], and current guidelines on pathological reporting of prostate cancer recommend an assessment of tumour volume [15,16]. Consistent with our previous study, we found that 65–70% of patients with prostate cancer had a high level of EN2 in their urine, as detected by a simple ELISA-based test. There was a strong correlation between EN2 and tumour volume (P= 0.006), which was not evident with overall prostate weight. Notably, PSA level did not correlate with tumour volume or prostate weight. This volume association has of course been reported previously; the lack of association in the present study may just be reflective of the small cohort. In previous studies, PSA level appeared to correlate with tumour volume and risk of progression [17], but recently the same group reported erosion in the association of serum PSA level with tumour volume, which in the 4–20 µg/L range appears now to correlate better with total prostate volume [1]. PSA kinetics also have not been shown to reliably indicate disease progression [18,19].

There are several limitations of the present study that are being addressed currently in three ongoing prospective studies including a comparative study to assess the diagnostic utility of urinary EN2, prostate cancer antigen 3 (PCA3) and TMPRSS2:ERG biomarkers, with clinicopathological correlation in men undergoing RP. Firstly, we evaluated retrospective archived samples, but have previously determined that urinary EN2 is stable and does not degrade during –80 °C storage (unpublished data). Secondly, the cohort represents men with high-risk disease in view of the tumour volumes, higher than previously described by Stamey et al. [1] 2004) and with wider PSA distributions. Prospective studies of lower risk patients are being conducted in the European Union and the USA so will allow for differences in patients selection for RP. We will examine the relationship of EN2 with index lesion and total tumour volume, clinical and pathological T stage again and also tumour grade, which was not correlative in the present study. The observation of reduced EN2 levels in men with higher grade disease may reflect the biology of this marker. EN2 is secreted by prostate cancer cells into acini and ducts. The de-differentiation with increasing tumour grade may affect this secretion process. We are investigating the biological aspects of EN2 secretion in prostate cancer cells lines and fresh tissue currently.

The reliable identification of significant prostate cancer remains a key clinical objective. Low-grade prostate cancer (<0.5 mL) is still considered to have a low malignant potential [20] and the extent of cancer detected on biopsy cores is regarded as an indication of tumour stage and risk of recurrence after treatment [21]. A recent analysis of the ESPRC cohort confirmed the original value of the index tumour volume threshold of 0.5 mL for ‘insignificant’ prostate cancer, and suggested that clinically insignificant prostate cancer may include index Gleason score 6, pT2 tumours with volumes up to at least 1.3 mL [9]. Therefore a pre-treatment tumour marker that provides an accurate assessment of cancer volume would have considerable clinical utility in assessing the risk of prostate cancer progression. Our prospective studies encompass much lower risk individuals than the present cohort, to allow assessment of pre-treatment EN2 levels to guide treatment vs AS decision making.

In recent years, there has been a plethora of potential diagnostic prostate cancer biomarkers but none, apart from PCA3, accurately reflect tumour volume. The data for PCA3 is inconsistent: some groups have described a significant association between low PCA3 scores and low tumour stage and grade [22,23], while others have found a correlation between high PCA3 scores and extracapsular extension [24]. Using a threshold PCA£ score of 25, Plousaard et al. [22] found urinary PCA3 predicted low volume (0.5 mL) cancer but a threshold level to identify high-risk disease was not apparent. Other groups have not found a correlation between PCA3 and tumour volume or between PCA3 and advanced pathological stage or intermediate and high RP Gleason score [25,26]. In the largest current prospective study, the PCA3 level correlated with low-volume prostate cancer but not with higher volume and stage disease [27].

In conclusion, we have shown a similar prevalence of urinary EN2 in a retrospective series of patients as our original study, and that pre-treatment urinary levels of EN2 correlated with disease volume as determined at RP. Ongoing prospective studies of lower risk individuals will address the utility of EN2 levels in identifying significant vs insignificant prostate cancer.

ACKNOWLEDGMENTS

We would like to thank Francesca Launchbury for assistance with the assay, and Hashim Udin for critical review of the manuscript.

CONFLICT OF INTEREST

None declared. Source of funding: University of Surrey and Aarhus University Hospital.

Ancillary