2 Clarke Drive
Suite 100
Cranbury, NJ 08512
© 2024 MJH Life Sciences™ and OncLive - Clinical Oncology News, Cancer Expert Insights. All rights reserved.
James Knight, MD, discusses a case study of a patient with brain metastases from prostate cancer and the importance of multidisciplinary, comprehensive care.
The rareness of brain-metastatic prostate lesions underscores the importance of broadening radiologic differentials to detect and recognize these metastases and accordingly determine optimal treatment options for patients with prostate adenocarcinoma, according to James Knight, MD.
“[More research in this area is needed] on a multi-institutional level. If this kind of disease occurrence is so rare, we need to pull together data from multiple institutions to get good, robust findings,” Knight said during an interview with OncLive® at the 2024 ACRO Annual Meeting.
In the interview, Knight, a GME resident in the College of Radiation Medicine at the University of Kentucky College of Medicine in Lexington, discussed details from a case study involving a patient with brain-metastatic prostate cancer, potential future research directions that may amplify awareness for cases such as this one, and the importance of relying on multidisciplinary teams to deliver accurate and comprehensive care for each individual patients.
Knight: We had a patient, a 70-year-old gentleman, who had a history of high-risk prostate cancer that had been treated with radical prostatectomy who had biochemical recurrence within the prosthetic fossa that was treated with radiation. However, when we did a Gallium-68 prostate-specific membrane antigen [PSMA] scan to survey for additional metastases, we didn’t find anything in the spine or in the pelvis. However, he did have a single hypermetabolic focus within the left parietal lobe.
Brain-metastatic prostate adenocarcinoma tends to be rare. It [comprises approximately] 1.6% of all metastatic prostate cancer cases. Additionally, to have a brain metastatic lesion without intervening metastases is particularly rare. The other challenge with this patient was where [the tumor] was located; it was in close proximity to the dura. With PSMA, there are other tumors, such as meningiomas or renal cell carcinomas, that also have an affinity for this.
[This finding] raised some questions about our differential. Was this a rare brain-metastatic prostate cancer, or was it a meningioma? The ways in which they are managed are quite different. We presented this case to demonstrate the need to have a combination of radiographic, clinical, and histopathological data to guide diagnoses and to highlight that PSMA is not always specific for prostate cancer; other differentials must be considered.
This was a case [study] with 1 patient whom we had evaluated. We would like to evaluate other patients [with this disease]. The nuclear medicine department [at the University of Kentucky College of Medicine] has at least 3 or 4 other patients with brain-metastatic prostate lesions. I would like to do even more digging into these data to see whether they also have PSMA data. Do we also have histopathological confirmation on these patients? Because [brain-metastatic prostate lesions are] so rare, if we have just a few more patients, perhaps we could turn this into a case series.
We would also want to find out: Are there other ways to parse out whether we can improve our differentials for prostate [cancer] compared with meningioma using other biomarkers? Are there ways of determining this based on apical vs basal membrane expression of PSMA? There are several other ways in which we could take this kind of study.
Another question we would like to highlight is: Just exactly how prevalent [are brain-metastatic prostate lesions] within patients [with prostate cancer]? Do we need to make changes with our imaging? Rather than just doing a skull base–to-mid–thigh scan, if you have high-risk patients, should you consider doing a whole-body scan or doing MRIs more frequently to determine whether there are these metastases? This [case] raises the possibility [for patients to] have asymptomatic metastases that we have not seen. A lot more studies, both on an individual level institutionally and multi-institutionally, would need to be done to answer those kinds of questions.
Continue to keep a broad differential. Have those conversations regularly with your radiologists, surgeons, and medical oncologists. Based upon what you’re finding from clinical, histopathological, and radiological data, [you should] use all [those findings] to guide treatments appropriately. Having that understanding will help early residents and medical students [identify] good treatment options and select the most appropriate treatments for their patients.