- Top of page
- Materials and Methods
- Conflict of Interest
Widespread adoption of PSA screening has led to a significant downward risk migration in prostate cancer (CaP) over the past two decades . Many of these cancers are diagnosed in older men and may have little impact on patient survival. As a result, active surveillance (AS) has emerged as an effective management option allowing many of these men to delay or possibly avoid definitive treatment and its associated side effects .
Despite careful patient selection, up to 35% of men in AS will experience biopsy reclassification during follow-up [3, 4]. Non-invasive methods of cancer surveillance, including PSA and PSA kinetics assessments, are not considered reliable triggers for intervention in this population, and annual prostate biopsy is currently recommended by some for monitoring men undergoing AS ; however, serial prostate biopsies are associated with potentially serious infectious and quality-of-life sequelae [6, 7]. Improved patient selection and less invasive methods of cancer surveillance are needed to improve the safety of AS in the management of low-risk CaP.
Recently, MRI has shown promise in identifying men entering AS who are at risk of immediate pathological upgrading after repeat biopsy or radical prostatectomy (RP) [8, 9], but the MRI characteristics of men within AS programmes who are at low risk of disease reclassification have not been fully described. Identification of these men through MRI has the potential to reduce a man's exposure to serial prostate biopsies without compromising cancer surveillance. In the present study, we report our experience with multiparametric MRI in identifying pathological index (path-index) lesions, defined as cancer present in the same prostate sextant in two separate surveillance biopsies, in an AS cohort with extended follow-up.
- Top of page
- Materials and Methods
- Conflict of Interest
The use of MRI in AS protocols for men with low-risk CaP has not been widely accepted, in part because of conflicting reports on its usefulness in this population. In the present study, there was a low prevalence of MRI-index lesions (20%) as would be expected in men who have remained in a selective AS programme for a median follow-up of 47.5 months. In this population, multiparametric MRI demonstrated high specificity and NPV in correctly classifying those without path-index lesions. Furthermore, when compared with those without a MRI-index lesion, those with a MRI-index lesion were more likely to have experienced biopsy reclassification. These results are among the first to describe the findings of multiparametric prostate MRI in an AS cohort with extended follow-up.
The interpretation of these results must begin by fully understanding the study population, which was a cohort of men at very low risk of cancer reclassification given their acceptance into and maintenance within a highly selective AS programme. In these men, in whom a negative MRI would be expected, the absence of a MRI-index lesion provided reassurance on the indolent nature of their disease. Whether MRI results can be used to follow these men is an interesting question which requires further prospective study, but, even within this population, significant tumours were identified on MRI (Fig. 2). Indeed, these men had higher rates of disease reclassification, underscoring the need for vigilant pathological follow-up in men with a visible tumour on MRI despite favourable disease characteristics on biopsy. In total, these results suggest multiparametric MRI may have an important role in the management of men in AS for low-risk CaP.
Figure 2. MRI of a 71-year-old man followed using AS for >10 years. He had had multiple positive biopsy cores from the location of the tumour visualized by MRI. Although his disease was reclassified owing to the number of biopsy cores involved with cancer (3) he elected to remain on AS because of his age, comorbidities, and low grade cancer. Multiparametric MRI performed at 3 tesla, with the endorectal coil, shows a MRI-index lesion in the right mid posterior peripheral zone. The lesion is 11 mm and shows suspicious features on T2-weighted imaging, DW imaging, DCE imaging and spectroscopy. A, Axial T2-weighted image shows an ill-defined T2 hypointense lesion (arrow). B, Axial ADC map calculated from the DW image data set shows a low signal lesion with restricted diffusion corresponding to A. (arrow), ADC value 0.783 × 10−3 mm2/s. C, Post-contrast axial subtraction image shows a hyperperfusing lesion (arrow) corresponding to A. D, DCE-MRI color-coded map shows a lesion with high vascular permeability (arrow) in the location corresponding to A. E, MR spectroscopy shows high Choline peak (arrow) in the region corresponding to A.
Download figure to PowerPoint
The use of MRI for the diagnosis and staging of CaP has increased over the past 5 years. Using RP specimens as a reference, early reports demonstrated excellent performance of various MRI sequences in identifying tumours of the peripheral and transition zones as well as anterior tumours [15-17]. Application of this technology has been extended to identify and target tumours for biopsy in men with previous benign biopsies and persistently elevated PSA levels [18, 19]. Despite the increasing use of MRI in CaP diagnosis and management the use of such technology in AS populations remains undefined. The present report adds to a growing body of literature demonstrating that MRI may have an important role in screening patients for entry into AS as well as monitoring their cancer for progression.
There is much controversy concerning the ability of MRI to predict disease reclassification among men in AS. Studies by Margel et al.  and Fradet et al.  showed that, among men undergoing AS, a suspicious lesion on MRI conferred a significant risk of disease reclassification on repeat biopsy. Conversely, among men undergoing RP, who would have otherwise qualified for AS, Guzzo et al.  did not find any correlation between tumour identification on MRI and adverse pathological features. In the present study, we found a significantly greater rate of biopsy reclassification in men who had a MRI-index lesion when compared with those without. Although the present study was not designed to analyse MRI as a prognostic tool, its results suggest that MRI may play an important role in counselling patients regarding both entry into surveillance and cancer monitoring; for example, a non-suspicious MRI may indicate that surveillance is safe for those men with tumour features that do not fit all AS criteria, thus broadening enrolment in the programme. Furthermore, a non-suspicious MRI together with previous favourable biopsy features may allow an increased interval between biopsies for some men.
The present study has several limitations. First, MRI was typically performed after the patient had been enrolled in AS for several years, therefore, only limited conclusions on the prognostic value of MRI can be made without further prospective data. Second, defining a path-index lesion with serial biopsy data provides an estimation of true pathological tumour characteristics. We have previously shown, however, that having two positive biopsies in the same location is associated with an index cancer in the same location at RP , so the present results may provide a robust estimate of the performance of MRI in identifying true index lesions which could only be determined after prostatctomy. Nevertheless, such post-prostatectomy results are also inherently biased by different selection of patients. Third, the men included in the study had undergone multiple biopsies at the time of MRI. Despite waiting at least 6 weeks from the last biopsy to perform MRI, biopsy artifact and/or scar may have still influenced image interpretation. Fourth, only one experienced radiologist interpreted the imaging studies and therefore interobserver variability cannot be reported. Finally, the study is limited by its small population size and its retrospective design. Larger, prospective studies are needed to determine the long-term impact of MRI findings in AS populations and are planned by our group.
In conclusion, a non-suspicious multiparametric MRI is associated with an absence of a path-index lesion in an AS population. Furthermore, biopsy reclassification appears to be more common in men with a MRI-index lesion. MRI in conjunction with other patient characteristics may ultimately be useful in selecting patients for AS and for guiding the need for repeat prostate biopsies during surveillance.