Does a screening digital rectal exam provide actionable clinical utility in patients with an elevated PSA and positive MRI?

Abstract Objective To define the value of a digital rectal exam (DRE) in the prostate‐magnetic resonance imaging (MRI) era. Prostate MRI is increasingly used in men with elevated prostate‐specific antigen (PSA) prior to biopsy. Methods A retrospective study was performed in men with elevated PSA undergoing MRI followed by MRI fusion with systematic biopsy and men with elevated PSA/active surveillance with negative MRI followed by biopsy. Baseline clinicopathologic characteristics and DRE findings were collected. We examined performance of a positive DRE on sensitivity and specificity of diagnosing clinically significant prostate cancer (CSPC). Results A total of 339 patients had elevated PSA and positive MRI followed by MRI fusion guided with systematic biopsy. Pre‐biopsy DRE was documented in 286/339 patients, who were included in further analysis. About 81.6% positive, 78.7% questionable, and 55.8% negative DRE patients had CSPC. Positive DRE had 21.8% sensitivity and 91.3% specificity for CSPC. Positive or questionable DRE had 42.1% sensitivity and 81.5% specificity. Among 148 men with non‐CSPC (GG1)‐targeted biopsy, 28 had systematic biopsy with CSPC. About 5/28 had positive DRE and 8/28 had positive or questionable DRE. Twenty‐seven patients were included who had elevated PSA/on active surveillance with negative MRI and biopsy done within 2 years. About 77.8% had negative, 7.4% had questionable, and 14.8% men had positive DRE. About 7.4% had CSPC and all had a negative DRE. Conclusions Our study provides limited evidence for the value of a DRE. However, it does show occasional benefit in detecting GG2 or higher disease and given the lack of cost and side effects, should still be considered.


| INTRODUC TI ON
The digital rectal exam (DRE) has been a routine part of prostate cancer diagnosis and screening for decades, providing important prognostic information to clinicians. As time has passed, the DRE has remained a mainstay of clinical management and guidelines, 1,2 even when tools such as prostate-specific antigen (PSA) testing 3 and prostate multiparametric magnetic resonance imaging (MP-MRI), 4 have been integrated into clinical practice. DRE adds both to risk assessment and stratification following a diagnosis of prostate cancer and is notably efficient in terms of cost (low, although does require an in-person visit) and ease of performance. 5,6,7 However, multiple studies have demonstrated variability in both the consistency in performance of a DRE 8 and its inter-and intra-observer reliability. 9,10 Given the issues with clinical benefits, some question the utility of subjecting patients to this uncomfortable exam. In addition, other studies have pointed out the DRE exam as a potential barrier to prostate cancer care, especially for underrepresented minorities. 11,12 Data suggest that prostate MRI may improve the detection of CSPC. The PROMIS study evaluated men with elevated PSA up to 15 ng/mL and no previous biopsy and found MP-MRI could be used as a triage test to allow men to avoid unnecessary transrectal ultrasound (TRUS) biopsy and diagnose fewer non-CSPC. 13 The PRECISION randomized trial demonstrated that among patients with an elevated PSA or abnormal DRE, MP-MRI-targeted biopsy improved the diagnosis of CSPC and avoided detection of low-risk disease compared to standard systematic biopsy. 14 While the use of MRI-targeted biopsy alone, without systematic cores, is a current topic of debate, 15,16,17 its suggestion by these data 15,16,17 raises the question as to whether a DRE has a role in screening algorithms in which only targeted biopsy of positive MRI lesions is pursued. Therefore, we sought to define the value of a DRE in the detection of clinically significant (defined as Gleason grade group [GG] 2 or higher) prostate cancer in the setting of pre-biopsy MRI testing.
We aimed to determine the clinical utility of DRE in the setting of an abnormal MRI and explore its use in a subgroup of men with negative MRI.

| ME THODS
This is a retrospective study consisting of two cohorts of men. The first cohort consisted of consecutive patients who underwent MRI fusion biopsy at our institution beginning in 2014, who were prospectively entered into an Institutional Review Board approved database. We included men from this database who, due to an elevated PSA (defined by the treating urologist) and a positive MRI, underwent a transrectal prostate biopsy with both targeted ultrasound fusion biopsies and systematic cores. Of note, four men had biopsy following an initial diagnosis of GG1 prostate cancer using systematic transrectal biopsy and subsequent enrollment on active surveillance. Prior to biopsy, a prostate MRI was performed with a 1.5-Tesla or 3-Tesla Signa HDx MR scanner (GE Healthcare, Waukesha, WI, USA) or Siemens MR scanner (Siemens, Erlangen, Germany) using an 8-channel abdominal array coil was used for mpMRI. For the majority of patients, an endorectal coil was used. 18 MP-MRI sequences   performed included T1-weighted, T2-weighted, diffusion-weighted   imaging including calculation of the apparent diffusion coefficient maps, and dynamic contrast enhancement. Two dedicated genitourinary radiologists reviewed and performed MRI interpretation, segmentation, and contouring of region of interests using ProFuse software. Suspicious lesions were graded on a Likert scale from 1 to 5 and any lesion graded ≥ 3 was targeted for biopsy. For targeted, ultrasound fusion biopsies, all biopsies were performed with monitored IV conscious sedation using the Artemis TM system (Eigen, CA). In the same setting, all patients underwent an additional systematic biopsy from sextant regions that did not overlap previously biopsied areas. Patients were excluded if they did not have a DRE result documented.
Baseline clinicopathologic characteristics and DRE findings for this first cohort were then manually extracted from a chart review, which included age, PSA, PSA density, and ethnicity. The DREs obtained from the chart review were either done at the initial encounter or preoperative visit prior to the planned MRI guided fusion and systematic biopsies. DRE was performed by either the attending physician, fellow, resident, or advanced practice practitioner. The DRE was divided into three categories: "negative," "questionable," and "positive." Negative meaning: cT1c, "symmetrical," "no nodules," and "benign/smooth." Questionable meaning: "asymmetry," "unsure enlargement/engorgement," "firm findings," "induration," and "unable to access due to body habitus." Positive meaning: anything other than cT1c, "nodules," "abnormal," or "concerning." Due to interest in evaluating the role of DRE in setting of a negative MRI, we additionally performed an exploratory analysis in a second cohort of patients. These patients were extracted from a retrospective chart review of consecutive patients seen in a MD Anderson institutional active surveillance clinic from 11/2018 to 2/2020, which consisted of patients with an elevated PSA and those undergoing active surveillance for a previously diagnosed very lowrisk prostate cancer. Patients included in this second cohort from this clinic had a negative MRI (defined as PIRADS 1 or 2, Likert scale 1 or 2, "no dominant lesions," "no definitive tumors," and "no suspicious lesions"), and subsequent prostate biopsy within 2 years of the MRI. Of note, some patients underwent different biopsy techniques; these included systematic 12 cores using ultrasound guidance alone or transperineal biopsy. Clinicopathologic characteristics and DRE findings were extracted. DRE results were similarly evaluated for sensitivity, specificity, and negative/positive predictive value in relation to CSPC.
Our primary outcome of interest was the performance of a DRE in detecting CSPC, which was defined as GG2 or higher disease, as defined by the 2014 International Society of Urological pathology (ISUP) Consensus Conference. 19 Dedicated genitourinary pathologists reviewed all slides. We evaluated the sensitivity and specificity of a negative, questionable, or positive DRE for detection of GG2 disease. We calculated descriptive statistics for clinicopathological characteristics of both groups, including median and interquartile range of age, PSA, PSA density, and numbers of targeted and systematic biopsies. ANOVA and Chi-square test calculations were used to assess baseline differences between groups. All calculations, including determination of sensitivity and specificity, were obtained using Microsoft Excel.

| RE SULTS
A total of 339 patients underwent MRI fusion biopsy with systematic cores due to elevated PSA and positive MRI findings. Pre-biopsy DRE was documented in 286/339 (84.3%) patients. Therefore, 53/339 (15.7%) did not have a documented DRE and were excluded. Table 1 lists clinical characteristics of the 286 patients included in further analysis. Median age of this group was 66 years (IQR 60-71) and median PSA was 6.1 (IQR 4.8-8.9). About 78% were Caucasian, 10% were African American, 4% were Asian, 4% were Latin American, 1% was Middle Eastern, and 3% classified as "other." The median number of targeted biopsies was 4 (IQR 3-6) and the median number of systematic biopsies was 10 (IQR [8][9][10][11]. DRE results are shown in Figure 1 and are summarized as follows: 49 (17.2%) positive, 47 (16.4%) questionable, and 190 (66.4%) negative.   prostate MRI and incorporation into risk-assessment algorithms, future studies with increased power will have the ability to better determine DRE performance in the current era.

ACK N OWLED G M ENTS
Overall, this study provides evidence that DRE may not add value in detecting CSPC in men with an elevated PSA and positive MRI who are undergoing fusion and systematic prostate biopsy. However, DRE may be useful in a scenario where targeted biopsy only is planned.
Limited data suggest that DRE is not helpful in setting of a negative MRI, though further studies are needed to confirm these findings.