In this application of the PURC dataset, the authors focus on utilization of axial imaging in the staging and diagnosis of prostate cancer (PCa). At this point in time, CT scans (Abdomen/Pelvis typically) and MRI (prostate or pelvis) are acceptable imaging modalities used for diagnosis and staging – often in conjunction with Nuclear Medicine Bone Scans. Yet, numerous new imaging modalities are being utilized, though the evidence has not yet caught up with guideline recommendations.
The authors utilize the PURC data to capture demographic and clinicopathologic data for men with clinically localized CaP, with a focus axial imaging modality (CT vs. MRI) utilization. They subsequently looked at predictors of utilization to identifies discrepancies in healthcare administration.
Between January 2015 and May 2018, 6109 eligible patients were enrolled in PURC. Patients with clinically localized CaP that underwent axial imaging and had sufficient information were included, resulting in a total of 1376 men in the cohort of interest. Patients with both CT and MRI (~100 patients) were excluded.
Demographic and clinicopathologic characteristic of the cohort are described in Table below:
Overall, MRI was found to be a more common imaging modality than CT (68.6% vs 31.4%). Caucasian men were more likely to undergo MRI than African-American (AA) men (72.1% vs 57.0%, p < 0.001) on univariate analysis. Men with higher PSA were more likely to undergo CT, likely as a staging study. Of note, when looking at AUA risk stratification, MRI is used more in men with VLR, LR and favorable IR patients – likely in the setting of pre-biopsy evaluation. CT, on the other hand, is more often used in unfavorable IR and high-risk patients – likely in the setting of staging studies.
On multivariable analysis, Caucasian race (OR 1.89, 95% CI 1.30-2.69, p < 0.001) was a strong independent predictor of MRI utilization, along with younger age (p=0.008), lower PSA value (p=0.009), lower Gleason score (HR 0.59, p < 0.001) and treatment at a practice with higher patient throughput (HR 1.05, p < 0.001). They did not look at CT utilization.
Significant variation was observed across individual practice sites:
Similar variation (in AS utilization, renal biopsy utilization) has previously been demonstrated by the MUSIC data.
Limitations / Discussion Point:
- One major limitation to this study was identifying why these images were ordered – for diagnosis, pre-prostate biopsy or for staging? mpMRI and CT are not interchangeable in the PCa diagnostic/staging pathway, and as such, shouldn’t necessarily be compared head to head. Unfortunately, indication was not able to be captured in the dataset.
Presented by: Andrew Higgins MD, Einstein Medical Center, Philadelphia, PA
Co-authors: Jiangtao Gou, Laura McGarry, Bret Marlowe, Claudette Fonshell, Edouard Trabulsi, Mark Mann, David Chen, John Danella, Thomas Guzzo, Thomas Lanchoney, Jay Raman, Adam Reese, Jeffrey Tomaszewski, Marc Smaldone, Robert Uzzo, Serge Ginzburg
Written by: Thenappan Chandrasekar, MD, Clinical Instructor, Thomas Jefferson University @tchandra_uromd, @JEFFUrology at the American Urological Association's 2019 Annual Meeting (AUA 2019), May 3 – 6, 2019 in Chicago, Illinois