Clinical Investigation
Percentage of Positive Biopsy Cores: A Better Risk Stratification Model for Prostate Cancer?

Presented in part at the 51st Annual Meeting of the American Society of Radiation Oncology, November 2009, Chicago, IL, and the 2011 American Society of Clinical Oncology Genitourinary Symposium, February 2011, Orlando, FL.
https://doi.org/10.1016/j.ijrobp.2011.09.043Get rights and content

Purpose

To assess the prognostic value of the percentage of positive biopsy cores (PPC) and perineural invasion in predicting the clinical outcomes after radiotherapy (RT) for prostate cancer and to explore the possibilities to improve on existing risk-stratification models.

Methods and Materials

Between 1993 and 2004, 1,056 patients with clinical Stage T1c-T3N0M0 prostate cancer, who had four or more biopsy cores sampled and complete biopsy core data available, were treated with external beam RT, with or without a high-dose-rate brachytherapy boost at William Beaumont Hospital. The median follow-up was 7.6 years. Multivariate Cox regression analysis was performed with PPC, Gleason score, pretreatment prostate-specific antigen, T stage, PNI, radiation dose, androgen deprivation, age, prostate-specific antigen frequency, and follow-up duration. A new risk stratification (PPC classification) was empirically devised to incorporate PPC and replace the T stage.

Results

On multivariate Cox regression analysis, the PPC was an independent predictor of distant metastasis, cause-specific survival, and overall survival (all p < .05). A PPC >50% was associated with significantly greater distant metastasis (hazard ratio, 4.01; 95% confidence interval, 1.86–8.61), and its independent predictive value remained significant with or without androgen deprivation therapy (all p < .05). In contrast, PNI and T stage were only predictive for locoregional recurrence. Combining the PPC (≤50% vs. >50%) with National Comprehensive Cancer Network risk stratification demonstrated added prognostic value of distant metastasis for the intermediate-risk (hazard ratio, 5.44; 95% confidence interval, 1.78–16.6) and high-risk (hazard ratio, 4.39; 95% confidence interval, 1.70–11.3) groups, regardless of the use of androgen deprivation and high-dose RT (all p < .05). The proposed PPC classification appears to provide improved stratification of the clinical outcomes relative to the National Comprehensive Cancer Network classification.

Conclusions

The PPC is an independent and powerful predictor of clinical outcomes of prostate cancer after RT. A risk model replacing T stage with the PPC to reduce subjectivity demonstrated potentially improved stratification.

Introduction

The clinical T stage, Gleason score, and initial pretreatment prostate-specific antigen (iPSA) have been shown to be independent predictors of biochemical failure (BF) after definitive radiotherapy (RT) (1) and constitute the most commonly used risk-classification systems for prostate cancer (PCa), including that recommended by the National Comprehensive Cancer Network (NCCN; available from: www.nccn.org). However, those models do not perfectly predict the final outcome and include significant heterogeneity (2). The percentage of positive biopsy cores (PPC) and the presence of perineural invasion (PNI) as determined by transrectal ultrasound-guided biopsy have been suggested as potential prognostic factors to enhance the standard risk stratification for PCa patients treated with external beam RT (EBRT) 3, 4, 5. However, there are conflicting reports on their independent prognostic value, in addition to the established factors 6, 7. Recently, a multi-institutional study showed that the PPC would add minimal improvement to conventional stratification systems in predicting BF (8). Those previous studies have mostly concentrated on the correlation of the PPC or PNI to BF, which represents a surrogate for clinical outcomes (9). The potential prognostic value of PPC and PNI on potentially more clinically relevant outcomes such as locoregional recurrence (LRR), distant metastasis (DM), and overall survival (OS), require longer follow-up and have not been well quantified. The purpose of the present study was to retrospectively review a large cohort of patients with extended follow-up to clarify the clinical predictive value of PPC and PNI and to devise a new risk classification to improve on the existing stratification systems.

Section snippets

Methods and Materials

The eligible patients for the present study had clinical Stage T1c-T3N0M0 (1993 American Joint Committee on Cancer staging system) prostate adenocarcinoma (PCa) treated with definitive external beam radiotherapy (EBRT) with or without a high-dose-rate (HDR) brachytherapy boost at William Beaumont Hospital between 1993 and 2004. All cases were required to have the results of the transrectal ultrasound-guided biopsy available, with a minimum of four biopsy cores sampled. Patients with incomplete

Results

The characteristics of the entire cohort and comparisons between low (≤50%) and high (>50%) PPC are listed in Table 1. The median PPC was 33%, and 18% of patients had PNI. A high PPC was associated with younger age, more risk factors (T stage, Gleason score, iPSA, and PNI), and more aggressive treatment (increased use of high-dose RT and AD therapy). As seen in supplemental Table e1, a high PPC was associated with younger age for the intermediate- and high-risk groups and increased use of

Discussion

In the present study, we analyzed 1,056 PCa patients who were treated with modern RT techniques and found that the PPC is a powerful and independent predictor of highly relevant clinical outcomes. The association of the PPC with DM was especially robust and exists regardless of the use of AD therapy or high-dose RT. Furthermore, the prognostic effect of the PPC on DM appears to translate to a correlation with OS, especially for higher risk patients. This is remarkable considering patients with

Acknowledgments

The authors would like to thank Michelle Wallace, Kimberly Marvin, and Christina Mitchell for support with data management.

References (20)

There are more references available in the full text version of this article.

Cited by (48)

  • Impact of percent positive biopsy cores on cancer-specific mortality for patients with high-risk prostate cancer

    2020, Urologic Oncology: Seminars and Original Investigations
    Citation Excerpt :

    However, for subsets of high-risk patients with a very high risk of PCSM, such as those with ≥50% PBC, who have a 2-fold increased risk compared to patients with <50% PBC, intensification of therapy may be appropriate to maximize cancer-specific outcomes. Other investigators have also attempted to further risk stratify patients with high-risk disease using clinical [17–22] and genomic data [23]. Our study represents the largest cohort of high-risk patients in which the role of PBC has been studied, though a few limitations of our analyses should be noted.

  • Health-related quality-of-life changes due to high-dose-rate brachytherapy, low-dose-rate brachytherapy, or intensity-modulated radiation therapy for prostate cancer

    2015, Brachytherapy
    Citation Excerpt :

    After obtaining investigational review board approval, the records of 1294 men with clinically localized prostate cancer who underwent HDR brachytherapy, iodine (I)-125 LDR brachytherapy, or IMRT between January 2002 and September 2013 were reviewed. Patients were excluded if they had unfavorable intermediate-risk (19, 20) or high-risk (21) prostate cancer, or if they were treated with androgen deprivation therapy (ADT) (22, 23), combined IMRT and HDR brachytherapy or LDR brachytherapy, three-dimensional conformal radiation therapy (3DCRT), IMRT to a total dose other than 7400–8100 cGy in 37–45 fractions (7–10), HDR brachytherapy to a total dose other than 2700–2800 cGy in two fractions, or I-125 LDR brachytherapy to a total dose other than 14,500 cGy in one fraction (24). Seven patients treated with palladium-103 LDR brachytherapy to 12,500 cGy were excluded because of the small patient numbers.

View all citing articles on Scopus

Conflict of interest: none.

View full text