Introducing a Standard in Prostate Cancer Imaging Evaluation

HomeRadiologyVol. 302, No. 3 PreviousNext Reviews and CommentaryFree AccessEditorialIntroducing a Standard in Prostate Cancer Imaging EvaluationRonald H. Gottlieb Ronald H. Gottlieb Author AffiliationsFrom Sarasota, FL.Address correspondence to the author (e-mail: [email protected]).Ronald H. Gottlieb Published Online:Dec 21 2021https://doi.org/10.1148/radiol.212111MoreSectionsPDF ToolsImage ViewerAdd to favoritesCiteTrack CitationsPermissionsReprints ShareShare onFacebookTwitterLinked In See also the article by Wibmer et al in this issueDr Gottlieb is a professor of radiology. His research has included studying Doppler ultrasound in detecting renal artery stenosis–related hypertension and urinary tract obstruction. His most recent interest has been the assessment of the cost-effectiveness of imaging in a variety of clinical settings. To prepare for his latest interest, he received an MPH degree at the University of Rochester while on the faculty in radiology at this institution.Download as PowerPointOpen in Image Viewer In this issue of Radiology, Wibmer et al (1) evaluate MRI for the initial staging of prostate cancer. They assess how MRI, in isolation, helps to predict patient response to radical prostatectomy including evidence of biochemical recurrence, development of metastatic disease, and prostate cancer–specific mortality rate. Surprisingly, the authors found that MRI, on the basis of its findings of disease confined to the prostate or spread outside the gland, was more predictive in distinguishing good and bad surgical outcomes than actual radical prostatectomy specimens (1).This introduces an area in researching the role MRI may have in determining whether radical prostatectomy is warranted. Our field needs more of this type of work that evaluates patient outcomes by using any of our imaging technologies as a method to confirm the true value in patient care. We usually assume that lesion depiction will be less important for patient outcomes compared with surgery, pathologic confirmation, or clinical decision making. Assessing patient outcomes after imaging is less commonly performed.Outcome analysis has been languishing in our prior research armamentarium since it was first introduced to our specialty by former giants in this area (2). Their work promoted the use of multiple levels of proof to confirm the value of medical imaging. Outcomes were placed in a multilevel hierarchical model of imaging evaluation. The first level represents technical adequacy in improving lesion visibility, the second level represents diagnostic accuracy, the third level represents the effects on physician clinical thinking, and the fourth level represents changes in patient treatment. If all are present, then these surrogate changes should better predict ultimate patient outcomes on a fifth level. But only by measuring true patient outcomes to therapy do we really know this is true. Diagnostic imaging does not directly affect outcomes largely dependent on available treatment, reliability of communications to clinical colleagues, or how referring physicians process those communications. Outcomes analysis is also compromised by difficulties in obtaining medical information on follow-up care because of patient privacy issues or because of the sheer amount of work involved in the effort.The difficulty of obtaining patient outcomes after therapy has been previously discussed in medical decision analyses (3). This type of analysis involves the use of computer software to follow downstream outcomes with a number of branch points, by comparing treatment choices with each other or the “doing nothing” option. Inputs into the model are derived from the best information from the medical literature including that from meta-analyses or other, usually inferior, sources of data. The conclusions regarding which choice leads to the best possible outcomes are not nearly as good as the outcomes derived from real-world actions, as was the case in the current investigation by Wibmer et al (1). The potential outcomes obtained from medical decision analysis are derived often from multiple assumptions and therefore have not been as unequivocally accepted as the type of evaluation used in the current work. Medical decision analysis, however, can suggest future analyses to be performed when patient randomization is not possible because of ethical considerations or when true outcomes cannot be obtained in the short term.Long-term patient outcomes due to slowly progressive pathologic processes such as prostate cancer, which is often indolent, are particularly difficult to obtain because of the sheer length of time required to do so. The retrospective analysis by Wibmer et al overcomes this issue because the authors obtained the outcome data by chart following a sufficiently long time after the initial imaging, with a median follow-up of 11 years. The authors provide us with the actual prostate-specific death rate after radical prostatectomy and, surprisingly, found that MRI was more predictive than surgical pathologic specimens.This study did have limitations. First, it is a single-institution retrospective analysis of a large cohort of patients with prostate cancer obtained from the database at the authors’ institution, a tertiary care academic cancer center. The cohort was part of a previously published analysis by the authors that investigated how patients with prostate cancer fared when treated with either prostatectomy or radiation therapy. Second, the MRI technology used was also rather old and consisted of vintage early 2000s 1.5-T MRI machines. Despite these limitations, this study sets in motion a standard for outcomes analyses in the treatment of prostate cancer. In particular, it has the actual prostate-specific mortality rate as the reference standard, which is the so-called holy grail of outcomes research in this area. Further work comparing the results of the current study with that of other institutions is necessary. Studies that use more state-of-the-art technology are needed. In addition, Gleason scores and quality-of-life evaluation and how they might impact results were not addressed. These are ripe areas for future investigation. Further research in the care of patients with prostate cancer would do well to attempt to mimic or even improve on the methods used in this study.In the past, the majority of radiology literature, including technology development and evaluation, has largely addressed only what radiologists consider valuable and not our clinical colleagues. Assessing how imaging can positively affect patient outcomes when our referring physicians use our technology can prove our value to them. We can then more informatively give them what they need in caring for their patients, making radiologists a more integral part of the medical team.Disclosures of Conflicts of Interest: R.H.G. No relevant relationships.References1. Wibmer AG, Nikolovski I, Chaim J, et al. Local extent of prostate cancer at MRI versus prostatectomy histopathology: associations with long-term oncologic outcomes. Radiology 2022;302(3):595–602. Abstract, Google Scholar2. Fryback DG, Thornbury JR. The efficacy of diagnostic imaging. Med Decis Making 1991;11(2):88–94. Crossref, Medline, Google Scholar3. Phelps CE, Mushlin AI. Focusing technology assessment using medical decision theory. Med Decis Making 1988;8(4):279–289. Crossref, Medline, Google ScholarArticle HistoryReceived: Aug 18 2021Revision requested: Aug 30 2021Revision received: Aug 31 2021Accepted: Sept 1 2021Published online: Dec 21 2021Published in print: Mar 2022 FiguresReferencesRelatedDetailsAccompanying This ArticleLocal Extent of Prostate Cancer at MRI versus Prostatectomy Histopathology: Associations with Long-term Oncologic OutcomesDec 21 2021RadiologyRecommended Articles Multiparametric MRI of the Prostate: Beyond Cancer Detection and StagingRadiology2021Volume: 299Issue: 3pp. 624-625Variation of PI-RADS Interpretations between Experts: A Significant LimitationRadiology2020Volume: 296Issue: 1pp. 85-86Prostate Artery Embolization for Lower Urinary Tract Symptoms Attributed to Benign Prostatic Hyperplasia—Radiology In TrainingRadiology2022Volume: 304Issue: 1pp. 31-37Management of Metastatic Prostate Cancer Biochemical Recurrence with Use of 18F-DCFPyL PET/CTRadiology2022Volume: 303Issue: 2pp. 423-424Local Extent of Prostate Cancer at MRI versus Prostatectomy Histopathology: Associations with Long-term Oncologic OutcomesRadiology2021Volume: 302Issue: 3pp. 595-602See More RSNA Education Exhibits Evaluation of Benign Prostatic Hyperplasia on Magnetic Resonance Imaging: Where Are We?Digital Posters2018Everything a Radiologist Should Know About European Prostate Cancer Guidelines: A Multimodal ReviewDigital Posters202018F–DCFPyL PSMA PET: Typical, Atypical and Unexpected FindingsDigital Posters2020 RSNA Case Collection Peripheral Zone Prostate Cancer with PseudocapsuleRSNA Case Collection2022Parasagittal meningiomaRSNA Case Collection2022Anterior transition zone prostate adenocarcinomaRSNA Case Collection2020 Vol. 302, No. 3 Metrics Altmetric Score PDF download