Original article
Pretreatment multiparametric MRI is independently associated with biochemical outcome in men treated with radiation therapy for prostate cancer

https://doi.org/10.1016/j.urolonc.2018.07.004Get rights and content

Highlights

  • We conducted a retrospective study of pre-radiotherapy mpMRI for prostate cancer.

  • Adverse imaging features on mpMRI were associated with worse biochemical outcomes.

  • Adverse signs: Extraprostatic extension, seminal vesical invasion, and tumor >15 mm.

  • mpMRI may aid risk stratification and individualize therapy in men treated with RT.

Abstract

Purpose

The purpose of this study was to investigate the utility of pre-treatment multiparametric magnetic resonance imaging (mpMRI) in a modern cohort of intermediate and high-risk prostate cancer patients treated with primary radiotherapy.

Methods and materials

One hundred twenty three men with National Comprehensive Cancer Network (NCCN) intermediate or high-risk prostate cancer were treated with primary EBRT and/or brachytherapy and had evaluable pre-treatment mpMRI with endorectal coil. Images were assessed for the presence of radiographic extraprostatic extension (rEPE), seminal vesicle invasion (rSVI), lymph node involvement (LNI), sextant involvement, and largest axial tumor diameter. Imaging characteristics were analyzed along with clinical risk factors against freedom from biochemical failure (FFBF). Median follow-up time was 50 months.

Results

Fourteen (11%) men developed biochemical failure. The 5-year FFBF was 94% in intermediate-risk patients and 82% in high-risk patients (p < 0.01). mpMRI findings including rEPE (29% vs. 66%, p < 0.01), rSVI (6% vs. 25%, p < 0.01), LNI (1% vs. 30%, p < 0.01), and largest axial tumor size> 15 mm (27% vs. 48%, p = 0.02) were identified in men with intermediate vs. high risk prostate cancer, respectively. mpMRI features associated with 5-y FFBF biochemical failure on univariate analysis included rEPE (80% vs 98%), rSVI (55% vs. 96%), LNI (65% vs. 93%), and largest axial tumor size >15mm (81% vs. 94%, all p < 0.01). Men without any high risk MRI finding had a 5-y FFBF of 100% vs. 81% (p < 0.01). Adverse imaging features (HR 8.9, p < 0.01) were independently associated with biochemical failure in a bivariate model analyzed alongside clinical risk category (HR 3.2, p = 0.04).

Conclusions

Pre-treatment mpMRI findings are strongly associated with biochemical outcomes in a modern cohort of intermediate and high-risk patients treated with primary radiotherapy. mpMRI may aid risk stratification beyond clinical risk factors in men treated with radiation therapy; further study is warranted to better understand how mpMRI can be used to individualize therapy.

Introduction

Risk-stratification for localized prostate cancer remains challenging, in part due to the heterogeneous nature of the disease [1]. Determining the optimal treatment course for an individual patient requires accurate staging and appropriate risk-stratification to balance treatment efficacy with unwanted side effects. Traditional clinical risk features such as Gleason score, prostate-specific antigen (PSA) level, and clinical stage aid in this process, however significant variability in prognosis remains. Furthermore, conventional imaging modalities such as ultrasound and computed tomography have inherent limitations in staging accuracy.

Multiparametric magnetic resonance imaging (mpMRI) is an advanced imaging modality with improved staging accuracy and the potential to improve risk stratification for localized prostate cancer. mpMRI combines anatomic and functional imaging features, with superior soft-tissue delineation as compared to CT [2], [3]. mpMRI may also help identify areas of clinically significant disease and predict tumor aggressiveness [4], as supported by whole mount pathology studies correlating the site of relapse after definitive radiotherapy to the primary tumor focus [5]. The enhanced anatomic and functional information provided by mpMRI could impact radiation treatment planning decisions and potentially identify patients at increased risk of treatment failure [6]. This knowledge may be particularly important for patients receiving primary radiotherapy, as prostate biopsy alone may misclassify risk category compared to surgical staging. Furthermore, there is a desire to optimize cure with primary therapy since salvage local therapy after radiotherapy presents more potential risk for toxicity. mpMRI is increasingly utilized for work-up of localized prostate cancer, and has established applications for initial diagnosis, staging and surgical planning [7]. However, less is known about its utility and prognostic value in men treated with primary radiotherapy. We sought to investigate the prognostic value of pretreatment endorectal mpMRI in addition to commonly accepted clinical risk factors in a modern cohort of intermediate and high-risk prostate cancer patients treated with primary radiotherapy. We hypothesized that adverse imaging features on pretreatment mpMRI would be associated with biochemical failure, and that mpMRI could help tailor therapy for men who undergo radiation therapy to the prostate.

Section snippets

Methods and materials

Using a prospectively maintained institutional database, we retrospectively identified 257 men with National Comprehensive Cancer Network (NCCN) intermediate or high-risk localized prostate cancer treated with primary external beam radiation therapy (EBRT) and/or brachytherapy between 2006 and 2013 [8]. Patient demographics, treatment information, and follow-up information were prospectively recorded. Review of patient outcomes was conducted with approval from our institutional review board.

Results

Baseline characteristics were similar between men with and without pretreatment mpMRI (Table1). A greater proportion of men with mpMRI were treated with brachytherapy. Thirty seven out of 257 (14%) patients developed biochemical failure during the follow-up period. The estimated actuarial 5-year FFBF was 91% in intermediate-risk vs. 81% in high-risk patients in the overall population of men with or without mpMRI (P < 0.01). The median time to biochemical failure was 47 months (50 months in the

Discussion

In this modern cohort of intermediate and high-risk patients receiving dose-escalated RT with conformal, image-guided techniques, mpMRI identified several advanced features of disease in a significant proportion of men, especially in men with high-risk disease. These radiographic findings were associated with biochemical failure, independently of the clinical risk classification. These results endorse a potential role for pretreatment imaging as a complement to traditional clinical risk

Conclusions

Pretreatment mpMRI is associated with biochemical outcomes in a modern cohort of intermediate and high-risk patients treated with primary radiotherapy, and adds prognostic value in addition to traditional clinical risk factors. Further study is warranted to confirm these findings and determine whether pretreatment mpMRI can help inform treatment decisions for localized prostate cancer.

References (23)

Cited by (10)

  • Can Pre-treatment Quantitative Multi-parametric MRI Predict the Outcome of Radiotherapy in Patients with Prostate Cancer?

    2022, Academic Radiology
    Citation Excerpt :

    Similarly, quantitative T2 values were not associated with biochemical outcome either. Previous studies have found pre-treatment MRI parameters such as higher PI-RADS score and lesion size (>15mm) (7), extra-prostatic extension (EPE) presence, lymph node invasion, seminal vesicle invasion (SVI) (8) and even prostate shape (9) to be strongly associated with biochemical failure. However, there is high inter-observer variability in the qualitative interpretation of prostate mpMRI which remains a major concern (10,11).

  • PI-RADS score is associated with biochemical control and distant metastasis in men with intermediate-risk and high-risk prostate cancer treated with radiation therapy

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

    mpMRI has been demonstrated to improve detection of clinically significant disease, prostate biopsy targeting, and staging of local disease [4,5]. mpMRI is also useful for optimizing therapy and predicting response to treatment; adverse features on pretreatment mpMRI, including radiographic extraprostatic extension (rEPE), radiographic seminal vesicle invasion (rSVI), lymph node involvement (LNI), and largest axial tumor dimension >15 mm, have been shown to correlate strongly with biochemical outcomes after primary radiation therapy (RT) [6–10]. While pretreatment mpMRI may provide useful information for clinical decision making, a potential limitation is difficulty interpreting mpMRI.

  • Biochemical outcomes and predictive factors by risk group after permanent iodine-125 seed implantation: Prospective cohort study in 2,316 patients

    2019, Brachytherapy
    Citation Excerpt :

    Riaz et al. reported that the percentage of high-risk patients with stage T3+ was 33.3% and their 5-year bFFF rate was 78% (26). Kauffmann et al. reported that percentage of high-risk patients with stage T3+ was 42.1% and their 5-year bFFF rate was 82% (21). In our study, younger age was significantly associated with biochemical failure only in low-risk patients.

View all citing articles on Scopus
View full text