- Email: [email protected]
mL
+
MODEL
Kaohsiung Journal of Medical Sciences (2018) xx, 1e6
Available online at www.sciencedirect.com
ScienceDirect journal homepage: http://www.kjms-online.com
Original Article
Prostate Health Index (PHI) improves prostate cancer detection at initial biopsy in Taiwanese men with PSA 4e10 ng/mL Po-Fan Hsieh a, Chao-Hsiang Chang a,b, Chi-Rei Yang a, Chi-Ping Huang a, Wen-Chi Chen a,b, Chin-Chung Yeh a,b, Eric Chieh-Lung Chou a,b, Kuo-Liang Chen a,b, Chi-Shun Lien a, Guang-Heng Chen a, Po-Jen Hsiao a, Yi-Huei Chang a, Hsi-Chin Wu a,b,c,* a
Department of Urology, China Medical University Hospital, Taichung, Taiwan School of Medicine, China Medical University, Taiwan c China Medical University Beigang Hospital, Yunlin, Taiwan b
Received 6 October 2017; accepted 22 February 2018
KEYWORDS Prostate cancer; Prostate health index; Prostate specific antigen
Abstract In this study, we aimed to validate the Prostate Health Index (PHI) for the detection of prostate cancer (PCa). We prospectively enrolled patients aged 50e75 years with a serum prostate specific antigen (PSA) level of 4e10 ng/mL undergoing transrectal biopsy of the prostate between April 2016 and May 2017. The primary outcome was the diagnostic performance of various PSA derivatives (total PSA, free PSA, %fPSA, p2PSA, %p2PSA, and PHI) to predict PCa. The secondary outcome was comparisons of PSA derivatives between patients with a Gleason score (GS) 6 and GS 7. PCa was diagnosed in 36 of 154 (23.4%) patients, and 26 (16.9%) had a GS 7. The areas under the receiver operating characteristic curves were significantly greater in %p 2PSA and PHI than in PSA (0.76 vs. 0.57, p Z 0.015 and 0.77 vs. 0.57, p Z 0.004, respectively). Patients with a GS 7PCa had marginally higher %p2PSA and PHI than those with a GS of 6 (17.8 vs. 12.73, p Z 0.06; 46.58 vs. 31.55, p Z 0.05). At a PHI cutoff value of 29.6, the sensitivity and specificity were 77.8% and 67.8% in detecting PCa, respectively. In addition, 57.1% of the patients avoided an unnecessary biopsy, while three patients (1.9%) with GS 7 PCa were missed. In conclusion, the ability of %p2PSA and PHI to predict prostate biopsy outcome was better than that of PSA and %fPSA in the initial biopsy in Taiwanese men with serum PSA between 4 and 10 ng/mL. Copyright ª 2018, Kaohsiung Medical University. Published by Elsevier Taiwan LLC. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/ by-nc-nd/4.0/).
Conflicts of interest: All authors declare no conflicts of interests. * Corresponding author. Department of Urology, China Medical University Hospital, No. 2, Yu-Der Rd, Taichung, Taiwan. E-mail address: [email protected] (H.-C. Wu). https://doi.org/10.1016/j.kjms.2018.02.007 1607-551X/Copyright ª 2018, Kaohsiung Medical University. Published by Elsevier Taiwan LLC. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Please cite this article in press as: Hsieh P-F, et al., Prostate Health Index (PHI) improves prostate cancer detection at initial biopsy in Taiwanese men with PSA 4e10 ng/mL, Kaohsiung Journal of Medical Sciences (2018), https://doi.org/10.1016/j.kjms.2018.02.007
+
MODEL
2
Introduction Prostate cancer (PCa) is the most common noncutaneous malignancy and the third leading cause of cancer death among men [1]. The incidence of PCa is lower and the survival rates are better in Asians compared to Caucasians [2,3]. Nevertheless, the incidence and mortality rates of PCa have increased rapidly in East Asia in recent years [4]. In Taiwan, the incidence of PCa is 28.8 per 100,000 persons, which is similar to Japan and Singapore, and PCa is the fifth most common cancer in men [5,6]. Since the introduction of serum prostate specific antigen (PSA), it has become the only molecular marker routinely used in the diagnosis of PCa [7]. However, the low specificity of PSA inevitably leads to many unnecessary biopsies with subsequent complications such as hematuria, rectal bleeding, and urinary tract infection [8,9]. The overdiagnosis and overtreatment of low-risk PCa and worsening quality of life are other concerns with PSA screening [10]. In addition, conflicting results of PSA screening trials have been reported in terms of PCa mortality [11,12]. The latest draft issued by the United States Preventive Services Task Force recommended against PSA screening in men 70 years old (grade D recommendation) [13]. Therefore, there is an urgent need to find a biomarker to better identify patients with clinically significant PCa and avoid the diagnosis of indolent PCa. Free PSA, Prostate Health Index (PHI), 4-kallikrein score, Prostate Cancer Antigen 3 (PCA3), SelectMDx, and Michigan Prostate Score have been shown to improve the early detection of PCa [14], among which PHI is a cost-effective one and was approved by the FDA in 2012. p2PSA is a truncated form of proPSA which contains two pro-leader amino acids [15]. Its derivatives, %p2PSA and PHI, have been shown to improve the prediction of outcomes of prostate biopsies and define cancer aggressiveness [15e18]. p2PSA and PHI have also been applied in Asian populations in several validation studies [19e23]. However, to the best of our knowledge, no study has investigated p2PSA or PHI in Taiwanese men. Therefore, we conducted this study to validate the feasibility of p2PSA and PHI in a Taiwanese cohort.
Methods This is an observational prospective cohort study. Between April 2016 and May 2017, we enrolled patients scheduled to undergo a prostate biopsy for the first time. All patients provided written informed consent before receiving the prostate biopsy. The Institutional Review Board of China Medical University Hospital approved this study, and the protocol number was CMUH105-REC1-123. The inclusion criteria were patients aged 50e75 years and a serum total PSA level 4.0e10.0 ng/mL, with or without an abnormal digital rectal examination (DRE). Patients with urinary tract infection, prostatitis, treatment with 5-alpha reductase inhibitors, or a history of PCa or prostate biopsy were excluded from this study. A transrectal ultrasound-guided prostate biopsy was performed in all patients with at least 12 cores. The biopsy specimens were processed and evaluated by an experienced uropathologist. PCa was graded according to the 2005 consensus
P.-F. Hsieh et al. conference of the International Society of Urological Pathology definitions [24]. PSA parameters including total PSA, free PSA, and p2PSA were determined from serum samples. Blood was drawn before prostate manipulation, centrifuged within 3 h of collection, and frozen at 20 to 80 C until analysis [25]. We used a Beckman Coulter DxI 800 Immunoassay System to measure the PSA parameters, and PHI was determined using the formula PHI Z (p2PSA/free PSA) OPSA [18]. The results of PHI were blinded to the operators before prostate biopsy. We collected clinical variables, PSA parameters, and biopsy reports for analysis. The primary outcome was the diagnostic performance of various PSA derivatives (total PSA, free PSA, %fPSA, p2PSA, %p2PSA, and PHI) to predict PCa. The secondary outcome was comparisons of PSA derivatives between patients with a Gleason score (GS) 6 and GS 7. Quantitative data are represented as mean standard deviation (SD) and categorical data as number (n) and percentage. The demographic and clinical characteristics between the cancer and noncancer groups were compared using the independent sample t-test and ManneWhitney test for continuous normally distributed and skewed variables, respectively. The area under the receiver operating characteristic curve (AUC) was calculated for the various PSA derivatives using SigmaPlot version 12.5 (Systat Software Inc, USA). All statistical analyses were carried out using SPSS version 22 (IBM Corp, Armonk, NY, USA), assuming a two-sided test with a 5% level of significance.
Results One hundred and fifty-four men with a serum PSA level 4.0e10.0 ng/mL underwent a prostate biopsy. PCa was diagnosed in 36 (23.4%) patients, of whom 26 (16.9%) had a GS 7. Table 1 summarizes the demographic and clinical characteristics of the study cohort according to biopsy outcomes. Statistically significant differences between the PCa and noncancer cohorts were found in age (67.9 vs. 64.8, p Z 0.031), %fPSA (0.18 vs. 0.22, p Z 0.013), p2PSA (19.48 vs. 13.66, p Z 0.029), %p2PSA (16.39 vs. 11.04, p < 0.001) and PHI (42.4 vs. 27.1, p < 0.001). The AUC for PSA, free PSA, %fPSA, p2PSA, %p2PSA and PHI were 0.57, 0.39, 0.33, 0.66, 0.76 and 0.77, respectively (Fig. 1). Using PSA as a standard, the AUC were significantly greater in % p2PSA and PHI (p Z 0.015 and 0.004, respectively, Table 2). The probabilities of PCa were 10.26%, 20%, 43.75%, and 77.78% for a PHI of 0e26.9, 27e35.9, 36e54.9 and 55, respectively (Table 3). Using Youden’s index, the best cutoff value of PHI was 29.6, with a sensitivity of 77.8% and specificity of 67.8%. Using a PHI of 29.6 as a cutoff for undergoing a biopsy, 88 patients (57.1%) of this cohort could have avoided undergoing a prostate biopsy. However, eight patients (5.2%) with PCa would have been missed, including five patients with a GS of 6, two patients with a GS of 3 þ 4, and one patient with a GS of 4 þ 3. Only one of these eight patients underwent radical prostatectomy, and his GS was upgraded from 3 þ 4 to 4 þ 3. For the two other patients with a GS of 7, magnetic resonance imaging (MRI) showed a tumor size of 0.2 mL in one and no significant lesion in the other.
Please cite this article in press as: Hsieh P-F, et al., Prostate Health Index (PHI) improves prostate cancer detection at initial biopsy in Taiwanese men with PSA 4e10 ng/mL, Kaohsiung Journal of Medical Sciences (2018), https://doi.org/10.1016/j.kjms.2018.02.007
+
MODEL
PHI improves prostate cancer detection in Taiwanese patients Table 1
3
Demographic and clinical characteristics of the study population.
age, years total PSA, ng/mL free PSA, ng/mL free: total PSA p2PSA, pg/mL %p2PSA PHI Gleason score, n (%) 6 7 8e10
Total (n Z 154)
Cancer (n Z 36)
No cancer (n Z 118)
P
65.5 7.5 6.37 1.64 1.30 0.58 0.21 0.08 15.02 9.47 12.29 6.12 30.68 16.34
67.9 8.2 6.76 1.86 1.21 0.66 0.18 0.07 19.48 14.76 16.39 7.06 42.4 20.8
64.8 7.1 6.25 1.55 1.33 0.55 0.22 0.08 13.66 6.53 11.04 5.19 27.1 12.71
0.031 0.101 0.269 0.013 0.029 <0.001 <0.001
N/A N/A N/A
10 (27.78) 17 (42.22) 9 (25)
N/A N/A N/A
PHI Z Prostate Health Index; PSA Z prostate specific antigen; N/A Z not applicable.
There were no significant differences in PSA, free PSA, % fPSA, and p2PSA between the patients with PCa with a GS 7 and GS of 6 (Table 4). On the other hand, those with a GS 7 had marginally higher %p2PSA and PHI than those with a GS of 6 (17.8 vs. 12.73, p Z 0.06; 46.58 vs. 31.55, p Z 0.05).
Discussion Because PSA screening is not routinely performed in Taiwan, nearly half of the patients with PCa are diagnosed with a locally advanced stage or metastasis [6]. In order to diagnose these patients at an earlier stage and to avoid unnecessary biopsies, a more specific biomarker is needed to identify patients who may harbor PCa. p2PSA is a biomarker which has been studied extensively in recent
years. One meta-analysis reported that it had a sensitivity of 90% for the detection of PCa, and that the specificity of % p2PSA and PHI were 32.5% (95% CI 30.6e34.5) and 31.6% (95% CI 29.2e34.0), respectively [26]. Besides, %p2PSA and PHI were also correlated with tumor aggressiveness [26]. In addition, the European Association of Urology (EAU) guidelines published in 2016 suggested that the PHI could be used as an additional diagnostic option for men with a serum PSA level between 2 and 10 ng/mL and a negative DRE [27]. The PHI has also been validated in several Asian countries. In the first prospective validation study in Asian men, Tan et al. evaluated 157 men with PSA levels of 4e10 ng/mL in Singapore in a first biopsy [21]. The AUC of %p2PSA and PHI were 0.69 and 0.79, and a cutoff PHI value of 26.75 could avoid 49% of unnecessary biopsies. In a prospective study of 569 men with PSA levels of 4e10 ng/mL in Hong Kong, Chiu et al. demonstrated that the PHI had better performance over PSA-based models [22]. In addition, combining the PHI, prostate volume, and age could avoid 38.4% and 55.4% of biopsies at 10% and 20% thresholds of missing PCa, respectively [22]. Among a Japanese population, Ito et al. reported that transition zone volumeadjusted PHI could substantially improve specificity while maintaining sensitivity at 90e95% [19]. In a prospective, multicenter study in Shanghai, Na et al. showed the better predictive performance of PHI over total PSA both in patients with a PSA level of 2.1e10 ng/mL and also in those with a PSA level >10 ng/mL (Table 5) [23].
Table 2 Comparison of AUC between PSA derivatives using PSA as standard.
PSA free PSA free: total PSA p2PSA %p2PSA PHI
Figure 1.
ROC curves of PSA derivatives.
AUC (95% CI)
p
0.57 0.39 0.33 0.66 0.76 0.77
0.001 0.003 0.197 0.015 0.004
(0.46e0.69) (0.28e0.49) (0.24e0.43) (0.56e0.76) (0.67e0.86) (0.68e0.87)
AUC Z area under the curve; CI Z confidence interval; PHI Z Prostate Health Index; PSA Z prostate specific antigen.
Please cite this article in press as: Hsieh P-F, et al., Prostate Health Index (PHI) improves prostate cancer detection at initial biopsy in Taiwanese men with PSA 4e10 ng/mL, Kaohsiung Journal of Medical Sciences (2018), https://doi.org/10.1016/j.kjms.2018.02.007
+
MODEL
4
P.-F. Hsieh et al. Table 3
Performance of PHI according to PHI levels.
PHI Level
Total number
All cancer, n (%)
GS 7, n (%)
No cancer, n
0e26.9 27e35.9 36e54.9 55
78 35 32 9
8 (10.26) 7 (20.00) 14 (43.75) 7 (77.78)
3 (3.85) 6 (17.14) 11 (34.38) 6 (66.67)
70 28 18 2
GS Z Gleason score; PHI Z Prostate Health Index.
Table 4
Comparison of PSA derivatives between GS 6 and GS 7.
total PSA, ng/mL free PSA, ng/mL free:total PSA p2PSA, pg/mL %p2PSA PHI
Cancer (n Z 36)
GS 6 (n Z 10)
GS 7 (n Z 26)
P
6.76 1.86 1.21 0.67 0.18 0.07 19.48 14.76 16.39 7.06 42.4 20.8
6.19 1.68 1.06 0.28 0.18 0.05 12.86 3.4 12.73 4.51 31.55 12.63
6.99 1.94 1.27 0.78 0.18 0.08 22.02 16.89 17.8 7.55 46.58 22.37
0.26 0.41 0.99 0.1 0.06 0.05
GS Z Gleason score; PHI Z Prostate Health Index; PSA Z prostate specific antigen.
The present study is the first validation study of p2PSA and PHI in a Taiwanese population. Consistent with previous reports in the literature, we found significant differences in %fPSA, p2PSA, %p2PSA, and PHI between patients with and without cancer in an initial biopsy in men with a PSA level of 4e10 ng/mL. Among these PSA derivatives, % p2PSA and PHI showed the most promising predictive value. They could help to distinguish PCa from benign prostate tissue, and there was also a tendency to discriminate aggressive (GS 7) from potentially indolent forms of PCa. The AUC of %p2PSA and PHI were similar to other Asian series [20e23]. At a pre-specified sensitivity of around 90%, Tan et al. and Ng et al. suggested a PHI threshold of 27 to achieve the best specificity of 58.3% and 49.8%, respectively [20,21]. In Na’s series of patients with a PSA level of 2.1e10 ng/mL, a PHI cutoff of 35 resulted in a sensitivity of 83.09% and specificity of 70.23% [23]. In the present study, a PHI cutoff Table 5
value of 20.7 resulted in a sensitivity of around 90% but specificity of only 30.5%. Therefore, we used Youden’s index to balance between sensitivity and specificity, and determined that at a cutoff PHI value of 29.6 the sensitivity and specificity were 77.8% and 67.8%, respectively, and up to 57.1% of unnecessary biopsies could be avoided. Three patients (1.9%) with PCa with a GS 7 would have been missed if biopsies were performed only in the men with a PHI 29.6. However, two of them had a small tumor volume (0.2 mL and no identifiable lesion, respectively) on MRI. Because most of the patients with a PHI <29.6 did not undergo radical prostatectomy, we could not analyze the exact number of pathological upgrading. In our series, the AUC of free PSA and %fPSA were significantly lower than that of PSA (0.39 vs. 0.57, p Z 0.001; 0.33 vs. 0.57, p Z 0.003). This may be because of the freezing process of the blood. Free PSA is not thermally stable, and freezing and thawing accelerate the decay of
High-volume studies of PHI in Asian men with serum PSA 10 ng/mL.
References
Ito et al. [19]
Area
Japan
Study design
N
Range of % positive PHI PSA biopsy AUC (ng/mL)
retrospective 239 2.0e10.0 22.2
Ng et al. [20] Hong Kong retrospective 230 4.0e10.0 9.13 Tan et al. [21] Singapore prospective 157 4.0e10.0 19.1 Chiu et al. [22] Hong Kong prospective 569 4.0e10.0 10.9 Na et ala [23]
Shanghai
prospective
660 2.1e10.0 20.6
Present study
Taiwan
prospective
154 4.0e10.0 23.4
N/A
cutoff Sensitivity Specificity % unnecessary (%) (%) biopsy avoided
23.9 24.9 0.781 26.54 0.794 26.75 0.76 N/A N/A 0.87 28 32 35 0.77 20.7 29.6
95 90 90 90 90 80 93.38 88.24 83.09 90 77.8
28 33.3 49.76 58.27 N/A N/A 50.57 62.4 70.23 30.5 67.8
28 33 45.2 49 38.4 55.4 41.52 51.97 59.24 26.6 57.1
AUC Z area under the curve; PHI Z Prostate Health Index; PSA Z prostate specific antigen. a Patients with serum PSA > 10 ng/mL are not listed in this table.
Please cite this article in press as: Hsieh P-F, et al., Prostate Health Index (PHI) improves prostate cancer detection at initial biopsy in Taiwanese men with PSA 4e10 ng/mL, Kaohsiung Journal of Medical Sciences (2018), https://doi.org/10.1016/j.kjms.2018.02.007
+
MODEL
PHI improves prostate cancer detection in Taiwanese patients free PSA [28]. As a result, it is generally recommended to analyze blood within 8 h of venipuncture. Interestingly, we found that p2PSA did not seem to be affected by the freezing process. Further studies are warranted to evaluate the effect of temperature on p2PSA and PHI. To increase the negative predictive value, several tools have been used in conjunction with the PHI. For example, Ferro et al. reported that adding PCA3 to the PHI marginally increased the predictive accuracy of the initial biopsy (AUC 0.82 for the base model plus PHI and 0.83 for the base model plus PHI and PCA3) [29]. Tosoian et al. prospectively assessed the combination of the PHI and MRI in 345 men undergoing radical prostatectomy, and found that none of the men with a PHI <27 and PI-RADS (prostate imaging reporting and data system) score 3 had Gleason group 2 PCa [30]. Further studies are needed to refine the diagnostic accuracy of PHI. There are several limitations to this study. First, it was conducted in a single tertiary referral center and consisted of a small sample size. The results may not be generated to other healthcare settings. Second, not all of the patients in our series underwent radical prostatectomy, and the incidence of Gleason upgrading after prostatectomy has been reported to be up to 44% with a PSA range of 4e10 ng/mL [31]. The possibility of sampling errors with false negative results should also be kept in mind. In addition, we used a GS 7 as the only criteria to define aggressive PCa because the potential of metastasis or cancer-specific death was negligible in patients with PCa with a GS&6 [32]. A more comprehensive assessment may have included the number of tumor foci, tumor size, and percentage of tumor involvement in each biopsy core. Finally, in this study we included patients undergoing prostate biopsy for the first time, and the results may not be extrapolated in those with previous negative biopsy. The usefulness of PHI in repeated biopsy would be addressed in subsequent studies.
Conclusion Similar to other studies in Asian populations, we demonstrated that the ability of %p2PSA and PHI to predict prostate biopsy outcomes was better than PSA and %fPSA in an initial biopsy in Taiwanese men with a serum PSA level between 4 and 10 ng/mL. Using a PHI of 29.6 as a threshold, a significant proportion of unnecessary biopsies could be avoided. Further large-scale studies are needed to validate our findings.
Acknowledgement This study was supported in part by the Taiwan Ministry of Health and Welfare Clinical Trial Center (MOHW106-TDU-B212-113004).
References [1] Siegel RL, Miller KD, Jemal A. Cancer statistics, 2017. CA Cancer J Clin 2017;67:7e30. [2] Hsing AW, Tsao L, Devesa SS. International trends and patterns of prostate cancer incidence and mortality. Int J Cancer 2000; 85:60e7.
5 [3] Robbins AS, Koppie TM, Gomez SL, Parikh-Patel A, Mills PK. Differences in prognostic factors and survival among white and Asian men with prostate cancer, California, 1995-2004. Cancer 2007;110:1255e63. [4] Akaza H, Onozawa M, Hinotsu S. Prostate cancer trends in Asia. World J Urol 2017;35:859e65. [5] Available at External link https://www.hpa.gov.tw/Pages/ List.aspx?nodeid=1415 [Accessed 30 Mar 2018]. [6] NCCN guideline Asia consensus statement prostate cancer version 2. 2013. Available at: External link, http://www.nccn. org/professionals/physician_gls/PDF/prostate-asia.pdf. [Accessed 13 August 2017]. [7] Heidenreich A, Bellmunt J, Bolla M, Joniau S, Mason M, Matveev V, et al. EAU guidelines on prostate cancer. Part I: screening, diagnosis, and treatment of clinically localised disease. Eur Urol 2011;59:61e71. [8] Stamey TA, Yang N, Hay AR, McNeal JE, Freiha FS, Redwine E. Prostate-specific antigen as a serum marker for adenocarcinoma of the prostate. N Engl J Med 1987;317:909e16. [9] Borghesi M, Ahmed H, Nam R, Schaeffer E, Schiavina R, Taneja S, et al. Complications after systematic, random, and image-guided prostate biopsy. Eur Urol 2017;71:353e65. [10] Heijnsdijk EA, Wever EM, Auvinen A, Hugosson J, Ciatto S, Nelen V, et al. Quality-of-life effects of prostate-specific antigen screening. New Engl J Med 2012;367:595e605. [11] Schro ¨der FH, Hugosson J, Roobol ML, Tammela TL, Ciatto S, Nelen V, et al. Screening and prostate-cancer mortality in a randomized European study. N Engl J Med 2009;360:1320e8. [12] Andriole GL, Crauford ED, Grubb 3rd RL, Buys SS, Chia D, Church TR, et al. Mortality results from a randomized prostate-cancer screening trial. N Engl J Med 2009;360: 1310e9. [13] Cooperberg MR. The new US preventive Services Task Force “C” draft recommendation for prostate cancer screening. Eur Urol 2017 May 20;17:30401e3. https://doi.org/10.1016/j.eururo.2017.05.011. S0302e2838. [14] Carlsson SV, Roobol MJ. Improving the evaluation and diagnosis of clinically significant prostate cancer in 2017. Curr Opin Urol 2017;27:198e204. [15] Hori S, Blanchet JS, McLoughlin J. From prostate-specific antigen (PSA) to precursor PSA (proPSA) isoforms: a review of the emerging role of proPSAs in the detection and management of early prostate cancer. BJU Int 2013;112: 717e28. [16] Sokoll LJ, Sanda MG, Feng Z, Kagan J, Mizrahi A, Broyles DL, et al. A prospective, multicenter, National Cancer Institute Early Detection Research Network study of [-2] proPSA: improving prostate cancer detection and correlating with cancer aggressiveness. Cancer Epidemiol Biomark Prev 2010; 19:1193e200. [17] Heidegger I, Klocker H, Steiner E, Skradski V, Ladurner M, Pichler R, et al. [-2]proPSA is an early marker for prostate cancer aggressiveness. Prostate Cancer Prostatic Dis 2014;17: 70e4. [18] Jansen FH, Van Schaik RHN, Kurstjens J, Horninger W, Klocker H, Bektic J, et al. Prostate-specific antigen (PSA) isoform p2PSA in combination with total PSA and free PSA improves the diagnostic accuracy in prostate cancer detection. Eur Urol 2010;57:921e7. [19] Ito K, Miyakubo M, Sekine Y, Koike H, Matsui H, Shibata Y, et al. Diagnostic significance of [22]pro-PSA and prostate dimension-adjusted PSA-related indices in men with total PSA in the 2.0e10.0 ng/mL range. World J Urol 2013;31: 305e11. [20] Ng CF, Chiu PK, Lam NY, Lam HC, Lee KW, Hou SS. The Prostate Health index in predicting initial prostate biopsy outcomes in Asian men with prostate-specific antigen levels of 410 ng/mL. Int Urol Nephrol 2014;46:711e7.
Please cite this article in press as: Hsieh P-F, et al., Prostate Health Index (PHI) improves prostate cancer detection at initial biopsy in Taiwanese men with PSA 4e10 ng/mL, Kaohsiung Journal of Medical Sciences (2018), https://doi.org/10.1016/j.kjms.2018.02.007
+
MODEL
6 [21] Tan LG, Tan YK, Tai BC, Tan KM, Gauhar V, Tiong HY, et al. Prospective validation of %p2PSA and the Prostate Health Index, in prostate cancer detection in initial prostate biopsies of Asian men, with total PSA 4e10 ng ml1. Asian J Androl 2017;19:286e90. [22] Chiu PK, Roobol MJ, Teoh JY, Lee WM, Yip SY, Hou SM, et al. Prostate health index (PHI) and prostate-specific antigen (PSA) predictive models for prostate cancer in the Chinese population and the role of digital rectal examinationestimated prostate volume. Int Urol Nephrol 2016;48:1631e7. [23] Na R, Ye D, Qi J, Liu F, Helfand BT, Brendler CB, et al. Prostate health index significantly reduced unnecessary prostate biopsies in patients with PSA 2-10 ng/mL and PSA >10 ng/mL: results from a multicenter study in China. Prostate 2017;77:1221e9. [24] Epstein JI, Allsbrook Jr WC, Amin MB, Egevad LL, Committee IG. The 2005 international society of urological pathology (ISUP) consensus conference on Gleason grading of prostatic carcinoma. Am J Surg Pathol 2005;29:1228e42. [25] Semjonow A, Kopke T, Eltze E, Pepping-Schefers B, Burgel H, Darte C. Pre-analytical in-vitro stability of [-2]proPSA in blood and serum. Clin Biochem 2010;43:926e8. [26] Filella X, Gime ´nez N. Evaluation of [2] proPSA and Prostate Health Index (phi) for the detection of prostate cancer: a systematic review and meta-analysis. Clin Chem Lab Med 2013;51:729e39. [27] Mottet N, Bellmunt J, Bolla M, Briers E, Cumberbatch MG, De Santis M, et al. EAU-ESTRO-SIOG guidelines on prostate cancer. Part 1: screening, diagnosis, and local treatment with curative intent. Eur Urol 2017;71:618e29.
P.-F. Hsieh et al. [28] Chen KC, Peng CH, Wang HE, Peng CC, Peng RY. Modeling of the pH- and the temperature-dependent deviations of the free to total PSA (prostate specific antigen) ratios for clinical predictability of prostate cancer and benign prostate hyperplasia. Bull Math Biol 2004;66:423e45. [29] Ferro M, Bruzzese D, Perdona S, Marino A, Mazzarella C, Perruolo G, et al. Prostate Health Index (Phi) and Prostate Cancer Antigen 3 (PCA3) significantly improve prostate cancer detection at initial biopsy in a total PSA range of 2-10 ng/ml. PLoS One 2013; 8:e67687. https://doi.org/10.1371/journal.pone.0067687. [30] Tosoian JJ, Druskin SC, Andreas D, Mullane P, Chappidi M, Joo S, et al. Use of the Prostate Health Index for detection of prostate cancer: results from a large academic practice. Prostate Cancer Prostatic Dis 2017;20:228e33. [31] Colleselli D, Pelzer AE, Steiner E, Ongarello S, Schaefer G, Bartsch G, et al. Upgrading of Gleason score 6 prostate cancers on biopsy after prostatectomy in the low and intermediate tPSA range. Prostate Cancer Prostatic Dis 2010;13: 182e5. [32] Kweldam CF, Wildhagen MF, Bangma CH, van Leenders GJ. Disease-specific death and metastasis do not occur in patients with Gleason score 6 at radical prostatectomy. BJU Int 2015; 116:230e5.
Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.kjms.2018.02.007.
Please cite this article in press as: Hsieh P-F, et al., Prostate Health Index (PHI) improves prostate cancer detection at initial biopsy in Taiwanese men with PSA 4e10 ng/mL, Kaohsiung Journal of Medical Sciences (2018), https://doi.org/10.1016/j.kjms.2018.02.007
MODEL
Kaohsiung Journal of Medical Sciences (2018) xx, 1e6
Available online at www.sciencedirect.com
ScienceDirect journal homepage: http://www.kjms-online.com
Original Article
Prostate Health Index (PHI) improves prostate cancer detection at initial biopsy in Taiwanese men with PSA 4e10 ng/mL Po-Fan Hsieh a, Chao-Hsiang Chang a,b, Chi-Rei Yang a, Chi-Ping Huang a, Wen-Chi Chen a,b, Chin-Chung Yeh a,b, Eric Chieh-Lung Chou a,b, Kuo-Liang Chen a,b, Chi-Shun Lien a, Guang-Heng Chen a, Po-Jen Hsiao a, Yi-Huei Chang a, Hsi-Chin Wu a,b,c,* a
Department of Urology, China Medical University Hospital, Taichung, Taiwan School of Medicine, China Medical University, Taiwan c China Medical University Beigang Hospital, Yunlin, Taiwan b
Received 6 October 2017; accepted 22 February 2018
KEYWORDS Prostate cancer; Prostate health index; Prostate specific antigen
Abstract In this study, we aimed to validate the Prostate Health Index (PHI) for the detection of prostate cancer (PCa). We prospectively enrolled patients aged 50e75 years with a serum prostate specific antigen (PSA) level of 4e10 ng/mL undergoing transrectal biopsy of the prostate between April 2016 and May 2017. The primary outcome was the diagnostic performance of various PSA derivatives (total PSA, free PSA, %fPSA, p2PSA, %p2PSA, and PHI) to predict PCa. The secondary outcome was comparisons of PSA derivatives between patients with a Gleason score (GS) 6 and GS 7. PCa was diagnosed in 36 of 154 (23.4%) patients, and 26 (16.9%) had a GS 7. The areas under the receiver operating characteristic curves were significantly greater in %p 2PSA and PHI than in PSA (0.76 vs. 0.57, p Z 0.015 and 0.77 vs. 0.57, p Z 0.004, respectively). Patients with a GS 7PCa had marginally higher %p2PSA and PHI than those with a GS of 6 (17.8 vs. 12.73, p Z 0.06; 46.58 vs. 31.55, p Z 0.05). At a PHI cutoff value of 29.6, the sensitivity and specificity were 77.8% and 67.8% in detecting PCa, respectively. In addition, 57.1% of the patients avoided an unnecessary biopsy, while three patients (1.9%) with GS 7 PCa were missed. In conclusion, the ability of %p2PSA and PHI to predict prostate biopsy outcome was better than that of PSA and %fPSA in the initial biopsy in Taiwanese men with serum PSA between 4 and 10 ng/mL. Copyright ª 2018, Kaohsiung Medical University. Published by Elsevier Taiwan LLC. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/ by-nc-nd/4.0/).
Conflicts of interest: All authors declare no conflicts of interests. * Corresponding author. Department of Urology, China Medical University Hospital, No. 2, Yu-Der Rd, Taichung, Taiwan. E-mail address: [email protected] (H.-C. Wu). https://doi.org/10.1016/j.kjms.2018.02.007 1607-551X/Copyright ª 2018, Kaohsiung Medical University. Published by Elsevier Taiwan LLC. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/). Please cite this article in press as: Hsieh P-F, et al., Prostate Health Index (PHI) improves prostate cancer detection at initial biopsy in Taiwanese men with PSA 4e10 ng/mL, Kaohsiung Journal of Medical Sciences (2018), https://doi.org/10.1016/j.kjms.2018.02.007
+
MODEL
2
Introduction Prostate cancer (PCa) is the most common noncutaneous malignancy and the third leading cause of cancer death among men [1]. The incidence of PCa is lower and the survival rates are better in Asians compared to Caucasians [2,3]. Nevertheless, the incidence and mortality rates of PCa have increased rapidly in East Asia in recent years [4]. In Taiwan, the incidence of PCa is 28.8 per 100,000 persons, which is similar to Japan and Singapore, and PCa is the fifth most common cancer in men [5,6]. Since the introduction of serum prostate specific antigen (PSA), it has become the only molecular marker routinely used in the diagnosis of PCa [7]. However, the low specificity of PSA inevitably leads to many unnecessary biopsies with subsequent complications such as hematuria, rectal bleeding, and urinary tract infection [8,9]. The overdiagnosis and overtreatment of low-risk PCa and worsening quality of life are other concerns with PSA screening [10]. In addition, conflicting results of PSA screening trials have been reported in terms of PCa mortality [11,12]. The latest draft issued by the United States Preventive Services Task Force recommended against PSA screening in men 70 years old (grade D recommendation) [13]. Therefore, there is an urgent need to find a biomarker to better identify patients with clinically significant PCa and avoid the diagnosis of indolent PCa. Free PSA, Prostate Health Index (PHI), 4-kallikrein score, Prostate Cancer Antigen 3 (PCA3), SelectMDx, and Michigan Prostate Score have been shown to improve the early detection of PCa [14], among which PHI is a cost-effective one and was approved by the FDA in 2012. p2PSA is a truncated form of proPSA which contains two pro-leader amino acids [15]. Its derivatives, %p2PSA and PHI, have been shown to improve the prediction of outcomes of prostate biopsies and define cancer aggressiveness [15e18]. p2PSA and PHI have also been applied in Asian populations in several validation studies [19e23]. However, to the best of our knowledge, no study has investigated p2PSA or PHI in Taiwanese men. Therefore, we conducted this study to validate the feasibility of p2PSA and PHI in a Taiwanese cohort.
Methods This is an observational prospective cohort study. Between April 2016 and May 2017, we enrolled patients scheduled to undergo a prostate biopsy for the first time. All patients provided written informed consent before receiving the prostate biopsy. The Institutional Review Board of China Medical University Hospital approved this study, and the protocol number was CMUH105-REC1-123. The inclusion criteria were patients aged 50e75 years and a serum total PSA level 4.0e10.0 ng/mL, with or without an abnormal digital rectal examination (DRE). Patients with urinary tract infection, prostatitis, treatment with 5-alpha reductase inhibitors, or a history of PCa or prostate biopsy were excluded from this study. A transrectal ultrasound-guided prostate biopsy was performed in all patients with at least 12 cores. The biopsy specimens were processed and evaluated by an experienced uropathologist. PCa was graded according to the 2005 consensus
P.-F. Hsieh et al. conference of the International Society of Urological Pathology definitions [24]. PSA parameters including total PSA, free PSA, and p2PSA were determined from serum samples. Blood was drawn before prostate manipulation, centrifuged within 3 h of collection, and frozen at 20 to 80 C until analysis [25]. We used a Beckman Coulter DxI 800 Immunoassay System to measure the PSA parameters, and PHI was determined using the formula PHI Z (p2PSA/free PSA) OPSA [18]. The results of PHI were blinded to the operators before prostate biopsy. We collected clinical variables, PSA parameters, and biopsy reports for analysis. The primary outcome was the diagnostic performance of various PSA derivatives (total PSA, free PSA, %fPSA, p2PSA, %p2PSA, and PHI) to predict PCa. The secondary outcome was comparisons of PSA derivatives between patients with a Gleason score (GS) 6 and GS 7. Quantitative data are represented as mean standard deviation (SD) and categorical data as number (n) and percentage. The demographic and clinical characteristics between the cancer and noncancer groups were compared using the independent sample t-test and ManneWhitney test for continuous normally distributed and skewed variables, respectively. The area under the receiver operating characteristic curve (AUC) was calculated for the various PSA derivatives using SigmaPlot version 12.5 (Systat Software Inc, USA). All statistical analyses were carried out using SPSS version 22 (IBM Corp, Armonk, NY, USA), assuming a two-sided test with a 5% level of significance.
Results One hundred and fifty-four men with a serum PSA level 4.0e10.0 ng/mL underwent a prostate biopsy. PCa was diagnosed in 36 (23.4%) patients, of whom 26 (16.9%) had a GS 7. Table 1 summarizes the demographic and clinical characteristics of the study cohort according to biopsy outcomes. Statistically significant differences between the PCa and noncancer cohorts were found in age (67.9 vs. 64.8, p Z 0.031), %fPSA (0.18 vs. 0.22, p Z 0.013), p2PSA (19.48 vs. 13.66, p Z 0.029), %p2PSA (16.39 vs. 11.04, p < 0.001) and PHI (42.4 vs. 27.1, p < 0.001). The AUC for PSA, free PSA, %fPSA, p2PSA, %p2PSA and PHI were 0.57, 0.39, 0.33, 0.66, 0.76 and 0.77, respectively (Fig. 1). Using PSA as a standard, the AUC were significantly greater in % p2PSA and PHI (p Z 0.015 and 0.004, respectively, Table 2). The probabilities of PCa were 10.26%, 20%, 43.75%, and 77.78% for a PHI of 0e26.9, 27e35.9, 36e54.9 and 55, respectively (Table 3). Using Youden’s index, the best cutoff value of PHI was 29.6, with a sensitivity of 77.8% and specificity of 67.8%. Using a PHI of 29.6 as a cutoff for undergoing a biopsy, 88 patients (57.1%) of this cohort could have avoided undergoing a prostate biopsy. However, eight patients (5.2%) with PCa would have been missed, including five patients with a GS of 6, two patients with a GS of 3 þ 4, and one patient with a GS of 4 þ 3. Only one of these eight patients underwent radical prostatectomy, and his GS was upgraded from 3 þ 4 to 4 þ 3. For the two other patients with a GS of 7, magnetic resonance imaging (MRI) showed a tumor size of 0.2 mL in one and no significant lesion in the other.
Please cite this article in press as: Hsieh P-F, et al., Prostate Health Index (PHI) improves prostate cancer detection at initial biopsy in Taiwanese men with PSA 4e10 ng/mL, Kaohsiung Journal of Medical Sciences (2018), https://doi.org/10.1016/j.kjms.2018.02.007
+
MODEL
PHI improves prostate cancer detection in Taiwanese patients Table 1
3
Demographic and clinical characteristics of the study population.
age, years total PSA, ng/mL free PSA, ng/mL free: total PSA p2PSA, pg/mL %p2PSA PHI Gleason score, n (%) 6 7 8e10
Total (n Z 154)
Cancer (n Z 36)
No cancer (n Z 118)
P
65.5 7.5 6.37 1.64 1.30 0.58 0.21 0.08 15.02 9.47 12.29 6.12 30.68 16.34
67.9 8.2 6.76 1.86 1.21 0.66 0.18 0.07 19.48 14.76 16.39 7.06 42.4 20.8
64.8 7.1 6.25 1.55 1.33 0.55 0.22 0.08 13.66 6.53 11.04 5.19 27.1 12.71
0.031 0.101 0.269 0.013 0.029 <0.001 <0.001
N/A N/A N/A
10 (27.78) 17 (42.22) 9 (25)
N/A N/A N/A
PHI Z Prostate Health Index; PSA Z prostate specific antigen; N/A Z not applicable.
There were no significant differences in PSA, free PSA, % fPSA, and p2PSA between the patients with PCa with a GS 7 and GS of 6 (Table 4). On the other hand, those with a GS 7 had marginally higher %p2PSA and PHI than those with a GS of 6 (17.8 vs. 12.73, p Z 0.06; 46.58 vs. 31.55, p Z 0.05).
Discussion Because PSA screening is not routinely performed in Taiwan, nearly half of the patients with PCa are diagnosed with a locally advanced stage or metastasis [6]. In order to diagnose these patients at an earlier stage and to avoid unnecessary biopsies, a more specific biomarker is needed to identify patients who may harbor PCa. p2PSA is a biomarker which has been studied extensively in recent
years. One meta-analysis reported that it had a sensitivity of 90% for the detection of PCa, and that the specificity of % p2PSA and PHI were 32.5% (95% CI 30.6e34.5) and 31.6% (95% CI 29.2e34.0), respectively [26]. Besides, %p2PSA and PHI were also correlated with tumor aggressiveness [26]. In addition, the European Association of Urology (EAU) guidelines published in 2016 suggested that the PHI could be used as an additional diagnostic option for men with a serum PSA level between 2 and 10 ng/mL and a negative DRE [27]. The PHI has also been validated in several Asian countries. In the first prospective validation study in Asian men, Tan et al. evaluated 157 men with PSA levels of 4e10 ng/mL in Singapore in a first biopsy [21]. The AUC of %p2PSA and PHI were 0.69 and 0.79, and a cutoff PHI value of 26.75 could avoid 49% of unnecessary biopsies. In a prospective study of 569 men with PSA levels of 4e10 ng/mL in Hong Kong, Chiu et al. demonstrated that the PHI had better performance over PSA-based models [22]. In addition, combining the PHI, prostate volume, and age could avoid 38.4% and 55.4% of biopsies at 10% and 20% thresholds of missing PCa, respectively [22]. Among a Japanese population, Ito et al. reported that transition zone volumeadjusted PHI could substantially improve specificity while maintaining sensitivity at 90e95% [19]. In a prospective, multicenter study in Shanghai, Na et al. showed the better predictive performance of PHI over total PSA both in patients with a PSA level of 2.1e10 ng/mL and also in those with a PSA level >10 ng/mL (Table 5) [23].
Table 2 Comparison of AUC between PSA derivatives using PSA as standard.
PSA free PSA free: total PSA p2PSA %p2PSA PHI
Figure 1.
ROC curves of PSA derivatives.
AUC (95% CI)
p
0.57 0.39 0.33 0.66 0.76 0.77
0.001 0.003 0.197 0.015 0.004
(0.46e0.69) (0.28e0.49) (0.24e0.43) (0.56e0.76) (0.67e0.86) (0.68e0.87)
AUC Z area under the curve; CI Z confidence interval; PHI Z Prostate Health Index; PSA Z prostate specific antigen.
Please cite this article in press as: Hsieh P-F, et al., Prostate Health Index (PHI) improves prostate cancer detection at initial biopsy in Taiwanese men with PSA 4e10 ng/mL, Kaohsiung Journal of Medical Sciences (2018), https://doi.org/10.1016/j.kjms.2018.02.007
+
MODEL
4
P.-F. Hsieh et al. Table 3
Performance of PHI according to PHI levels.
PHI Level
Total number
All cancer, n (%)
GS 7, n (%)
No cancer, n
0e26.9 27e35.9 36e54.9 55
78 35 32 9
8 (10.26) 7 (20.00) 14 (43.75) 7 (77.78)
3 (3.85) 6 (17.14) 11 (34.38) 6 (66.67)
70 28 18 2
GS Z Gleason score; PHI Z Prostate Health Index.
Table 4
Comparison of PSA derivatives between GS 6 and GS 7.
total PSA, ng/mL free PSA, ng/mL free:total PSA p2PSA, pg/mL %p2PSA PHI
Cancer (n Z 36)
GS 6 (n Z 10)
GS 7 (n Z 26)
P
6.76 1.86 1.21 0.67 0.18 0.07 19.48 14.76 16.39 7.06 42.4 20.8
6.19 1.68 1.06 0.28 0.18 0.05 12.86 3.4 12.73 4.51 31.55 12.63
6.99 1.94 1.27 0.78 0.18 0.08 22.02 16.89 17.8 7.55 46.58 22.37
0.26 0.41 0.99 0.1 0.06 0.05
GS Z Gleason score; PHI Z Prostate Health Index; PSA Z prostate specific antigen.
The present study is the first validation study of p2PSA and PHI in a Taiwanese population. Consistent with previous reports in the literature, we found significant differences in %fPSA, p2PSA, %p2PSA, and PHI between patients with and without cancer in an initial biopsy in men with a PSA level of 4e10 ng/mL. Among these PSA derivatives, % p2PSA and PHI showed the most promising predictive value. They could help to distinguish PCa from benign prostate tissue, and there was also a tendency to discriminate aggressive (GS 7) from potentially indolent forms of PCa. The AUC of %p2PSA and PHI were similar to other Asian series [20e23]. At a pre-specified sensitivity of around 90%, Tan et al. and Ng et al. suggested a PHI threshold of 27 to achieve the best specificity of 58.3% and 49.8%, respectively [20,21]. In Na’s series of patients with a PSA level of 2.1e10 ng/mL, a PHI cutoff of 35 resulted in a sensitivity of 83.09% and specificity of 70.23% [23]. In the present study, a PHI cutoff Table 5
value of 20.7 resulted in a sensitivity of around 90% but specificity of only 30.5%. Therefore, we used Youden’s index to balance between sensitivity and specificity, and determined that at a cutoff PHI value of 29.6 the sensitivity and specificity were 77.8% and 67.8%, respectively, and up to 57.1% of unnecessary biopsies could be avoided. Three patients (1.9%) with PCa with a GS 7 would have been missed if biopsies were performed only in the men with a PHI 29.6. However, two of them had a small tumor volume (0.2 mL and no identifiable lesion, respectively) on MRI. Because most of the patients with a PHI <29.6 did not undergo radical prostatectomy, we could not analyze the exact number of pathological upgrading. In our series, the AUC of free PSA and %fPSA were significantly lower than that of PSA (0.39 vs. 0.57, p Z 0.001; 0.33 vs. 0.57, p Z 0.003). This may be because of the freezing process of the blood. Free PSA is not thermally stable, and freezing and thawing accelerate the decay of
High-volume studies of PHI in Asian men with serum PSA 10 ng/mL.
References
Ito et al. [19]
Area
Japan
Study design
N
Range of % positive PHI PSA biopsy AUC (ng/mL)
retrospective 239 2.0e10.0 22.2
Ng et al. [20] Hong Kong retrospective 230 4.0e10.0 9.13 Tan et al. [21] Singapore prospective 157 4.0e10.0 19.1 Chiu et al. [22] Hong Kong prospective 569 4.0e10.0 10.9 Na et ala [23]
Shanghai
prospective
660 2.1e10.0 20.6
Present study
Taiwan
prospective
154 4.0e10.0 23.4
N/A
cutoff Sensitivity Specificity % unnecessary (%) (%) biopsy avoided
23.9 24.9 0.781 26.54 0.794 26.75 0.76 N/A N/A 0.87 28 32 35 0.77 20.7 29.6
95 90 90 90 90 80 93.38 88.24 83.09 90 77.8
28 33.3 49.76 58.27 N/A N/A 50.57 62.4 70.23 30.5 67.8
28 33 45.2 49 38.4 55.4 41.52 51.97 59.24 26.6 57.1
AUC Z area under the curve; PHI Z Prostate Health Index; PSA Z prostate specific antigen. a Patients with serum PSA > 10 ng/mL are not listed in this table.
Please cite this article in press as: Hsieh P-F, et al., Prostate Health Index (PHI) improves prostate cancer detection at initial biopsy in Taiwanese men with PSA 4e10 ng/mL, Kaohsiung Journal of Medical Sciences (2018), https://doi.org/10.1016/j.kjms.2018.02.007
+
MODEL
PHI improves prostate cancer detection in Taiwanese patients free PSA [28]. As a result, it is generally recommended to analyze blood within 8 h of venipuncture. Interestingly, we found that p2PSA did not seem to be affected by the freezing process. Further studies are warranted to evaluate the effect of temperature on p2PSA and PHI. To increase the negative predictive value, several tools have been used in conjunction with the PHI. For example, Ferro et al. reported that adding PCA3 to the PHI marginally increased the predictive accuracy of the initial biopsy (AUC 0.82 for the base model plus PHI and 0.83 for the base model plus PHI and PCA3) [29]. Tosoian et al. prospectively assessed the combination of the PHI and MRI in 345 men undergoing radical prostatectomy, and found that none of the men with a PHI <27 and PI-RADS (prostate imaging reporting and data system) score 3 had Gleason group 2 PCa [30]. Further studies are needed to refine the diagnostic accuracy of PHI. There are several limitations to this study. First, it was conducted in a single tertiary referral center and consisted of a small sample size. The results may not be generated to other healthcare settings. Second, not all of the patients in our series underwent radical prostatectomy, and the incidence of Gleason upgrading after prostatectomy has been reported to be up to 44% with a PSA range of 4e10 ng/mL [31]. The possibility of sampling errors with false negative results should also be kept in mind. In addition, we used a GS 7 as the only criteria to define aggressive PCa because the potential of metastasis or cancer-specific death was negligible in patients with PCa with a GS&6 [32]. A more comprehensive assessment may have included the number of tumor foci, tumor size, and percentage of tumor involvement in each biopsy core. Finally, in this study we included patients undergoing prostate biopsy for the first time, and the results may not be extrapolated in those with previous negative biopsy. The usefulness of PHI in repeated biopsy would be addressed in subsequent studies.
Conclusion Similar to other studies in Asian populations, we demonstrated that the ability of %p2PSA and PHI to predict prostate biopsy outcomes was better than PSA and %fPSA in an initial biopsy in Taiwanese men with a serum PSA level between 4 and 10 ng/mL. Using a PHI of 29.6 as a threshold, a significant proportion of unnecessary biopsies could be avoided. Further large-scale studies are needed to validate our findings.
Acknowledgement This study was supported in part by the Taiwan Ministry of Health and Welfare Clinical Trial Center (MOHW106-TDU-B212-113004).
References [1] Siegel RL, Miller KD, Jemal A. Cancer statistics, 2017. CA Cancer J Clin 2017;67:7e30. [2] Hsing AW, Tsao L, Devesa SS. International trends and patterns of prostate cancer incidence and mortality. Int J Cancer 2000; 85:60e7.
5 [3] Robbins AS, Koppie TM, Gomez SL, Parikh-Patel A, Mills PK. Differences in prognostic factors and survival among white and Asian men with prostate cancer, California, 1995-2004. Cancer 2007;110:1255e63. [4] Akaza H, Onozawa M, Hinotsu S. Prostate cancer trends in Asia. World J Urol 2017;35:859e65. [5] Available at External link https://www.hpa.gov.tw/Pages/ List.aspx?nodeid=1415 [Accessed 30 Mar 2018]. [6] NCCN guideline Asia consensus statement prostate cancer version 2. 2013. Available at: External link, http://www.nccn. org/professionals/physician_gls/PDF/prostate-asia.pdf. [Accessed 13 August 2017]. [7] Heidenreich A, Bellmunt J, Bolla M, Joniau S, Mason M, Matveev V, et al. EAU guidelines on prostate cancer. Part I: screening, diagnosis, and treatment of clinically localised disease. Eur Urol 2011;59:61e71. [8] Stamey TA, Yang N, Hay AR, McNeal JE, Freiha FS, Redwine E. Prostate-specific antigen as a serum marker for adenocarcinoma of the prostate. N Engl J Med 1987;317:909e16. [9] Borghesi M, Ahmed H, Nam R, Schaeffer E, Schiavina R, Taneja S, et al. Complications after systematic, random, and image-guided prostate biopsy. Eur Urol 2017;71:353e65. [10] Heijnsdijk EA, Wever EM, Auvinen A, Hugosson J, Ciatto S, Nelen V, et al. Quality-of-life effects of prostate-specific antigen screening. New Engl J Med 2012;367:595e605. [11] Schro ¨der FH, Hugosson J, Roobol ML, Tammela TL, Ciatto S, Nelen V, et al. Screening and prostate-cancer mortality in a randomized European study. N Engl J Med 2009;360:1320e8. [12] Andriole GL, Crauford ED, Grubb 3rd RL, Buys SS, Chia D, Church TR, et al. Mortality results from a randomized prostate-cancer screening trial. N Engl J Med 2009;360: 1310e9. [13] Cooperberg MR. The new US preventive Services Task Force “C” draft recommendation for prostate cancer screening. Eur Urol 2017 May 20;17:30401e3. https://doi.org/10.1016/j.eururo.2017.05.011. S0302e2838. [14] Carlsson SV, Roobol MJ. Improving the evaluation and diagnosis of clinically significant prostate cancer in 2017. Curr Opin Urol 2017;27:198e204. [15] Hori S, Blanchet JS, McLoughlin J. From prostate-specific antigen (PSA) to precursor PSA (proPSA) isoforms: a review of the emerging role of proPSAs in the detection and management of early prostate cancer. BJU Int 2013;112: 717e28. [16] Sokoll LJ, Sanda MG, Feng Z, Kagan J, Mizrahi A, Broyles DL, et al. A prospective, multicenter, National Cancer Institute Early Detection Research Network study of [-2] proPSA: improving prostate cancer detection and correlating with cancer aggressiveness. Cancer Epidemiol Biomark Prev 2010; 19:1193e200. [17] Heidegger I, Klocker H, Steiner E, Skradski V, Ladurner M, Pichler R, et al. [-2]proPSA is an early marker for prostate cancer aggressiveness. Prostate Cancer Prostatic Dis 2014;17: 70e4. [18] Jansen FH, Van Schaik RHN, Kurstjens J, Horninger W, Klocker H, Bektic J, et al. Prostate-specific antigen (PSA) isoform p2PSA in combination with total PSA and free PSA improves the diagnostic accuracy in prostate cancer detection. Eur Urol 2010;57:921e7. [19] Ito K, Miyakubo M, Sekine Y, Koike H, Matsui H, Shibata Y, et al. Diagnostic significance of [22]pro-PSA and prostate dimension-adjusted PSA-related indices in men with total PSA in the 2.0e10.0 ng/mL range. World J Urol 2013;31: 305e11. [20] Ng CF, Chiu PK, Lam NY, Lam HC, Lee KW, Hou SS. The Prostate Health index in predicting initial prostate biopsy outcomes in Asian men with prostate-specific antigen levels of 410 ng/mL. Int Urol Nephrol 2014;46:711e7.
Please cite this article in press as: Hsieh P-F, et al., Prostate Health Index (PHI) improves prostate cancer detection at initial biopsy in Taiwanese men with PSA 4e10 ng/mL, Kaohsiung Journal of Medical Sciences (2018), https://doi.org/10.1016/j.kjms.2018.02.007
+
MODEL
6 [21] Tan LG, Tan YK, Tai BC, Tan KM, Gauhar V, Tiong HY, et al. Prospective validation of %p2PSA and the Prostate Health Index, in prostate cancer detection in initial prostate biopsies of Asian men, with total PSA 4e10 ng ml1. Asian J Androl 2017;19:286e90. [22] Chiu PK, Roobol MJ, Teoh JY, Lee WM, Yip SY, Hou SM, et al. Prostate health index (PHI) and prostate-specific antigen (PSA) predictive models for prostate cancer in the Chinese population and the role of digital rectal examinationestimated prostate volume. Int Urol Nephrol 2016;48:1631e7. [23] Na R, Ye D, Qi J, Liu F, Helfand BT, Brendler CB, et al. Prostate health index significantly reduced unnecessary prostate biopsies in patients with PSA 2-10 ng/mL and PSA >10 ng/mL: results from a multicenter study in China. Prostate 2017;77:1221e9. [24] Epstein JI, Allsbrook Jr WC, Amin MB, Egevad LL, Committee IG. The 2005 international society of urological pathology (ISUP) consensus conference on Gleason grading of prostatic carcinoma. Am J Surg Pathol 2005;29:1228e42. [25] Semjonow A, Kopke T, Eltze E, Pepping-Schefers B, Burgel H, Darte C. Pre-analytical in-vitro stability of [-2]proPSA in blood and serum. Clin Biochem 2010;43:926e8. [26] Filella X, Gime ´nez N. Evaluation of [2] proPSA and Prostate Health Index (phi) for the detection of prostate cancer: a systematic review and meta-analysis. Clin Chem Lab Med 2013;51:729e39. [27] Mottet N, Bellmunt J, Bolla M, Briers E, Cumberbatch MG, De Santis M, et al. EAU-ESTRO-SIOG guidelines on prostate cancer. Part 1: screening, diagnosis, and local treatment with curative intent. Eur Urol 2017;71:618e29.
P.-F. Hsieh et al. [28] Chen KC, Peng CH, Wang HE, Peng CC, Peng RY. Modeling of the pH- and the temperature-dependent deviations of the free to total PSA (prostate specific antigen) ratios for clinical predictability of prostate cancer and benign prostate hyperplasia. Bull Math Biol 2004;66:423e45. [29] Ferro M, Bruzzese D, Perdona S, Marino A, Mazzarella C, Perruolo G, et al. Prostate Health Index (Phi) and Prostate Cancer Antigen 3 (PCA3) significantly improve prostate cancer detection at initial biopsy in a total PSA range of 2-10 ng/ml. PLoS One 2013; 8:e67687. https://doi.org/10.1371/journal.pone.0067687. [30] Tosoian JJ, Druskin SC, Andreas D, Mullane P, Chappidi M, Joo S, et al. Use of the Prostate Health Index for detection of prostate cancer: results from a large academic practice. Prostate Cancer Prostatic Dis 2017;20:228e33. [31] Colleselli D, Pelzer AE, Steiner E, Ongarello S, Schaefer G, Bartsch G, et al. Upgrading of Gleason score 6 prostate cancers on biopsy after prostatectomy in the low and intermediate tPSA range. Prostate Cancer Prostatic Dis 2010;13: 182e5. [32] Kweldam CF, Wildhagen MF, Bangma CH, van Leenders GJ. Disease-specific death and metastasis do not occur in patients with Gleason score 6 at radical prostatectomy. BJU Int 2015; 116:230e5.
Appendix A. Supplementary data Supplementary data related to this article can be found at https://doi.org/10.1016/j.kjms.2018.02.007.
Please cite this article in press as: Hsieh P-F, et al., Prostate Health Index (PHI) improves prostate cancer detection at initial biopsy in Taiwanese men with PSA 4e10 ng/mL, Kaohsiung Journal of Medical Sciences (2018), https://doi.org/10.1016/j.kjms.2018.02.007