Prostate health index density predicts aggressive pathological outcomes after radical prostatectomy in Taiwanese patients

Abstract

Background:

There are models to predict pathological outcomes based on established clinical and prostate-specific antigen (PSA)-derived parameters; however, they are not satisfactory. p2PSA and its derived biomarkers have shown promise for the diagnosis and prognosis of prostate cancer (PCa). The aim of this study was to investigate whether p2PSA-derived biomarkers can assist in the prediction of aggressive pathological outcomes after radical prostatectomy (RP).

Methods:

We prospectively enrolled patients who were diagnosed with PCa and treated with RP between February 2017 and December 2018. Preoperative blood samples were analyzed for tPSA, free PSA (fPSA), percentage of fPSA (%fPSA), [-2]proPSA (p2PSA), and percentage of p2PSA (%p2PSA). Prostate health index (PHI) was calculated as (p2PSA/fPSA) × √tPSA. Prostate volume was determined by transrectal ultrasound using the ellipsoid formula, and PHI density was calculated as PHI/prostate volume. The areas under the receiver operating characteristic curve were estimated for various PSA/p2PSA derivatives. Aggressive pathological outcomes measured after RP were defined as pathological T3 or a Gleason score (GS) >6 as determined in RP specimens.

Results:

One hundred and forty-four patients were included for analysis. Postoperative GS was >6 in 86.1% of the patients, and pT stage was T3a or more in 54.2%. Among all PSA- and p2PSA-derived biomarkers, PHI density was the best biomarker to predict aggressive pathological outcomes after RP. The odds ratio of having an aggressive pathological outcome of RP was 8.796 (p = 0.001). In multivariate analysis, adding %fPSA to base model did not improve the accuracy (area under curve), but adding PHI and PHI density to base model improved the accuracy by 2% and 16%, respectively, in predicting pT3 stage or GS ≥ 7. The risk of pT3 stage or GS ≥ 7 was 20.8% for PHI density <1.125, and 64.6% for PHI density >1.125 (sensitivity: 74.6% and specificity: 88.9%).

Conclusion:

PHI density may further aid in predicting aggressive pathological outcomes after RP. This biomarker may be useful in preoperative counseling and may have potential in decision making when choosing between definitive treatment and active surveillance of newly diagnosed PCa.

1. INTRODUCTION

Prostate cancer (PCa) is the cause of death worldwide,¹ and the incidence and associated mortality rates have been increasing over the past decades, and it has become an increasing burden on the healthcare system in Taiwan.² Since the adoption of prostate-specific antigen (PSA), there has been a stage migration toward earlier or more localized disease status at the diagnosis of PCa worldwide, including Taiwan. Standard treatment options for localized PCa include active surveillance (AS), radical prostatectomy (RP), and definite radiotherapy (RT).³

Currently, the decision to manage localized PCa with AS vs RP or RT relies on risk stratification systems such as the D’Amico/NCCN⁴ and EAU³ systems. However, significant upgrading or staging after RP has been reported, and this may limit the usefulness of these risk models.^5–8

RP has become the standard of care for localized PCa. It is generally regarded to be oncological surgery with significant potential functional consequences such as incontinence or erectile dysfunction. To optimize the likelihood of being continent and potent postoperatively, nerve sparing prostatectomy is desirable when feasible.⁹ To this end, assessing the capsule status preoperatively is important to safely preserve bilateral cavernosal nerves while attempting to achieve optimal oncological outcomes. Several look-up tables and nomograms have been developed based on a sizable amount of data;^10,11 however, they were developed more than a decade ago and are limited by the biomarkers and biopsy methods that were available, which may limit their generalizability to the present day.

PSA has a 17-amino acid leader sequence (preproPSA), and it can be cleaved to generate an inactive precursor protein (proPSA) with seven additional amino acids. Several truncated forms of ProPSA were generated by partial removal of the leader sequence of the preproPSA. Of all the forms of proPSA, p2PSA is a stable form.^12,13

Previous studies have shown that p2PSA and its derivatives, percentage of p2PSA to free PSA (fPSA) (%p2PSA; p2PSA/fPSA ng/mL), and the prostate health index (PHI; [p2PSA/fPSA] × √PSA) (Beckman Coulter; Brea, CA, USA) outperform PSA and other PSA derivatives in predicting PCa after transrectal ultrasound-guided prostate biopsy (TRUSP Bx).^14–21 Furthermore, a higher PHI has been associated with higher Gleason score (GS) in TRUSP Bx specimens.^16,18,20,21 PHI and %p2PSA have also been more significantly correlated with pathological outcomes after prostatectomy than PSA in European and Chinese populations.^21–25 PSA density has also been shown to be a better predictor of tumor upgrade and pathological features after prostatectomy.^26,27 Therefore, we conducted this prospective study to investigate whether PHI, %p2PSA, and PHI density (PHI/prostate volume) are correlated with pathological features after prostatectomy.

2. METHODS

2.1. Study design

We prospectively enrolled patients who had been diagnosed with PCa and had undergone RP from February 2017 to December 2018 at a single medical center. The aim of this study was to assess the accuracy of p2PSA-derived biomarkers in predicting aggressive pathological outcomes after RP. The study was approved by the Institutional Review Board/Ethics Committee of Taipei Veterans General Hospital (approval code: 201708017A), and all patients were informed of the procedures and possible complications and signed informed consent forms before entering the study.

2.2. Subjects

Patients who had been diagnosed with PCa and had undergone RP at our hospital were included. All cancers were confirmed by at least a 12-core TRUSP Bx. Patients who received any dosage of androgen deprivation therapy or 5-α-reductase inhibitors before collecting blood sample were excluded from our study cohort.

2.3. Methods

Clinical data including age, body mass index, and digital rectal examination (DRE) findings were collected. Blood samples were collected before surgery to measure tPSA, fPSA, percentage of fPSA (%fPSA), [-2]proPSA (p2PSA), and percentage of p2PSA (%p2PSA). PHI was calculated as (p2PSA/fPSA) × √tPSA. Prostate volume was calculated by transrectal ultrasound using the ellipsoid formula as width × height × length × 0.523, and PHI density was calculated as PHI/prostate volume. The blood samples were processed using a Beckman Coulter DxI800 Unicel Immunoassay system (Beckman Coulter, Taiwan Inc.). Pathologic outcomes including pathologic tumor stage, pathologic Gleason score (pGS), and pathologic nodal status were evaluated by experienced genitourinary pathologists, who were blinded to the blood sample results. An aggressive pathological outcome was defined as pT3 or pGS ≥ 7. PCa was staged and graded according to the 2010 version of the American Joint Committee on Cancer.

2.4. Statistical analysis

Descriptive statistics of categorical variables were reported as mean, median, and interquartile range (IQR). The Mann-Whitney U test was used to compare continuous variables, and the χ² test was used for categorical variables. The area under the receiver operating characteristic (ROC) curve was used to assess the accuracy in predicting final aggressive pathologic outcomes of each variable. Univariate and multivariable logistic regression models were used to predict aggressive pathologic outcomes. The defined base model in multivariate analysis included age, tPSA, biopsy GS, and abnormal DRE.

All statistical analyses were conducted using SPSS software for Windows version 22 (IBM, Armonk, NY, USA). A two-sided p value of <0.05 was considered to be statistically significant.

3. RESULTS

The baseline demographics and prostate evaluations including biomarkers and sonographic findings are listed in Table 1. A total of 144 patients were included in this study with a mean age of 65.5 years. Table 1 shows the clinical characteristics of the study population.

Table 1.

Demographic and baseline characteristics of the patients receiving radical prostatectomy

The pathological characteristics of the cohort are listed in Table 2.

Table 2.

Pathologic characteristics of the patients receiving radical prostatectomy

p2PSA-derived biomarkers were analyzed for their predictive ability with regards to pathological T3 or GS ≥ 7. The ROC curves of preoperative PSA- and p2PSA-derived parameters including %p2PSA, PHI, and PHI density are shown in Fig. 1. The area under the curve was biggest for PHI density as shown in Table 3. Further, univariate and multivariate analyses of these parameters are also shown in Table 3. Among the currently established parameters, GS as determined by TRUSP Bx had the strongest predictive ability of adverse pathological outcomes after prostatectomy (pT3 or GS ≥ 7), followed by preoperative PSA, age, and abnormal DRE findings. The odds ratios (ORs) of p2PSA-derived parameters including %p2PSA, PHI, and PHI density were then evaluated after adding them to the base model. As shown in Table 3, the OR was highest for PHI density (OR = 8.796, confidence interval 2.533-30.546; p = 0.001). In multivariate analysis, adding PHI and PHI density to base model improved the accuracy (area under curve) from 73.1% to 75.1% and 89.1%, respectively, in predicting pT3 stage or GS ≥ 7, but adding %fPSA to base model did not improve the accuracy. The risk of pT3 stage or GS ≥ 7 was 20.8% for PHI density <1.125, and 64.6% for PHI density >1.125 (sensitivity: 74.6%, specificity: 88.9%).

Fig. 1.

Prediction of pT3 stage or Gleason score ≥ 7. fPSA = free PSA; %fPSA = free PSA to total PSA; PHI = prostate health index; ROC = receiver operating characteristic curve; tPSA = total prostate-specific antigen.

Table 3.

Univariate and multivariate analyses for pT3 stage or Gleason score ≥ 7

According to the National Comprehensive Cancer Network criteria, patients with very low or low-risk disease were analyzed in upgrading and upstaging subgroups. There were 23 patients with low-risk disease and 17 patients with very low risk. A total of 19 out of 23 low-risk patients had upgrade of GS, with upgraded patients having significantly higher PHI values (56.9 vs 38.3; p < 0.001) and PHI density (1.91 vs 0.89; p < 0.001). Nine out of 17 very low-risk patients had an upgrade of GS, but the difference in PHI values and PHI density did not reach statistical significance. Besides, 19 (82.6%) of 23 low-risk patients had disease upstaged to pT2b or above, and 13 (76.4%) of 17 very low-risk patients had disease upstaged to pT2b or above. The difference in PHI values and PHI density of two groups did not reach statistical significance.

4. DISCUSSION

We conducted this prospective study to evaluate the power of biomarkers including PSA, fPSA, PHI, %p2PSA, PSA density, and PHI density to predict adverse pathological outcomes, and especially pT3 stage or GS ≥ 7 obtained after RP. The results showed that among these markers, PHI density had the strongest predictive ability for a high stage or high grade. To the best of our knowledge, this is the first study to show that PHI density performed better than current biomarkers and could increase the power of current predictive models.

Since PSA was officially approved by the FDA in 1994 for PCa screening, the diagnosis of PCa has experienced a dramatic stage migration toward an earlier stage at initial presentation in countries adopting PSA screening. However, the decrease in PCa mortality has also been controversial due to issues such as overdiagnosis and overtreatment. Taiwan does not officially support PSA screening, and therefore the likelihood of having stage IV at the initial diagnosis of PCa is twice as high in Taiwan as in the U.S.^28,29 In general, PCa is diagnosed at a later time in its natural course in Taiwan, as reflected by the higher PSA level (11 ng/mL) and higher proportion of pT3 after RP (54.2%), which are significantly higher than those reported in European and Hong Kong series,^21–23,25 suggesting that PCa treated with RP in this study was of a high stage and potentially a high grade. Different pathological characteristics may affect the performance of biomarkers or models to predict post-RP results preoperatively.

One way to attenuate the overdiagnosis/overtreatment of earlier or indolent PCa detected by PSA is to employ active surveillance combined with deferred treatment. However, selecting those with truly indolent PCa is challenging. The current gold standard to identify indolent PCa is the use of models incorporating clinical parameters to stratify the risk of biological behavior, such as the D’Amico/NCCN and EAU risk stratification systems or Partin’s table. Although these risk models are derived from large datasets and so have good statistical power, they were all developed more than a decade ago when newer biomarkers such as PHI and PCA3 were not available. Thus, incorporating newer biomarkers, which have been shown to perform better than PSA, seems to be attractive and may assist in more accurately identifying indolent PCa.

In the current study, neither PHI nor %p2PSA identified high-stage/high-risk PCa as previously reported.^22,23,25,30 There may be multiple reasons for this finding. We do not believe it is due to systemic errors caused by the assay results as we analyzed our cohort using the same system to predict a diagnosis of PCa by TRUSP Bx and the performance of PHI and %p2PSA was consistent with previous reports. Another plausible explanation may be due to the tumor characteristics in our cohort. In our cohort, the mean PSA level at treatment was 11.0 ng/mL, which is higher than that reported in other series (typically between 5 and 7 ng/mL), and consistently, pT3 disease accounted for 54.2% of our cohort. It is therefore possible that the current model utilizing GS and PSA performed better in the patients with a relatively high stage compared to those with a relatively low stage. In addition, PHI or %p2PSA alone may not have been able to show a significant effect in our series.

As with PSA density, including prostate volume in the predictive model would appear to be beneficial to further differentiate benign enlargement from cancer. We therefore further analyzed PHI density (PHI/volume) as a marker to predict worse pathological outcomes even in patients with a high stage cancer. The OR of adding PHI density to the currently used popular markers was 8.796 (Table 3). This is consistent with a recent report by Tosoian et al,³¹ who found that PHI density performed better than PHI and %p2PSA in detecting clinically significant cancer as defined by clinicopathological features after TRUSP Bx.

There are there main strengths to the current study. First, the prospective collection of plasma samples before RP and adherence of product recommendation. Second, the pathological review was conducted by experienced and dedicated genitourinary pathologists. Third, the use of PHI derivatives including PHI density greatly enhanced the predictive power of adverse pathological outcomes with RP.

There are also several limitations to this study. First, the sample size was relatively small, and a higher stage may have reduced the power of predicting a pathological high stage PCa postoperatively compared to current predicting models including GS, PSA, or Partin’s table. Second, the lack of oncological follow-up data may have reduced the robustness of the biomarker derivatives. However, this was mitigated by the strong association between higher stage/grade PCa with a worse prognosis with conservative treatment such as AS. Third, there may be operator-dependent bias of prostate volume when different operators performed TRUS. Lastly, without tumor volume parameters, we could not incorporate the Epstein criteria for comparison.

In conclusion, P2PSA derivatives, and especially PHI density, greatly enhanced the power of predicting adverse pathological outcomes. The use of such factors may help in preoperative counseling with patients, and may have potential in decision making when choosing between definitive treatment and active surveillance.