Broad discussion on the role of biomarker testing in prostate cancer, particularly how it has helped to inform therapy selection and patient management.
Scott Tagawa, MD, MS, FACP: There are a wide range of biomarkers for prostate cancer and multiple diseases in general, but I’d lump them into clinical factors, which are quite important. Those have to do with symptoms, different sites of metastasis that we can pick up with imaging biomarkers, comorbidities, and performance status. Most people think of biomarkers as either blood tests or tissue tests, ie, biopsies, and those are quite important. Probably the furthest along beyond general histology has been genomics, whether that’s DNA or RNA or the combination. We’re into proteomics, metabolomics, etc.
We’ve used imaging biomarkers for a very long time. People don’t always recognize that. For instance, CT or MRI cross-sectional imaging has been quite important. If you see prostate cancer that’s in the liver or the brain, that’s clearly a prognostic biomarker. That’s an example. It’s the same thing with technetium-99m bone scan, a very classic prognostic biomarker that’s maybe predictive as well for certain therapeutics, such as radium-223. That’s an example.
We’re now in the age of more molecular imaging. The most common molecular imaging in 2022 is PSMA [prostate-specific membrane antigen] PET [positron emission tomography] imaging, which is probably the most sensitive imaging biomarker that we have. It can pick up volume of disease that’s too small to be picked up on other conventional types of imaging, as well as the second generation of PET imaging, such as choline or fluciclovine. There may be some phenotypic characteristics that can be picked up with PSMA PET. For instance, PSMA expression is very common and is expressed in the vast majority of tumors, but not all. There’s somewhat of a link with the AR [androgen receptor] pathway. The classic dogma is that when there’s no AR presence, such as metastatic disease with a PSA [prostate-specific antigen] that is 0 ng/mL or very low—some people would call that neuroendocrine or small cell type—then PSMA also tends to be low, although that’s a generalization and not a strict 1:1 link.
Oliver Sartor, MD: The diagnosis of prostate cancer isn’t hard. It’s done with a biopsy. But we’ve known for some time that there’s significant molecular heterogeneity. If we look at the genomics, we’ve been able to discover that certain alterations, such as TP53; alterations in DNA repair genes, such as BRCA2; and alterations in the PTEN pathway, particularly PTEN loss or PTEN mutated, are relatively common in prostate cancer. One of the things that’s interesting is they’re not necessarily consistent over time. You can have certain mutations that arise after courses of therapy and other mutations that may disappear with courses of therapy. In particular, alterations in the androgen receptor pathway—I’ll mention things like the androgen receptor mutations at position 875, 878, 702—are often induced by treatments such as abiraterone or enzalutamide, which affect the androgen axis.
If we move forward a little, we have also begun to understand issues about what I’m going to call phenotypic heterogeneity. Phenotypic heterogeneity has been characterized predominantly through PSMA uptake. Most of us are familiar with PSMA PET scans. It turns out that the vast majority of people, particularly in early-stage disease, can have PSMA increased in their cancer relative to other tissues, and even normal prostate. As you treat the cancer over time, some degree of heterogeneity arises within this PSMA biomarker. The reason that PSMA is particularly important is we use it to select therapies when we’re using agents such as PSMA lutetium. It turns out that if we’re going to be using something like PSMA lutetium, we have to do a PSMA PET first to verify that you have the PSMA marker expressed on the cancer cells.
I can also mention other areas of heterogeneity that apply as predictive biomarkers. A little earlier, I mentioned things like BRCA1 and BRCA2. These are homologous recombination repair pathways. When they’re mutated, these particular tumors can become sensitive to things like PARP inhibitors. I should also mention mismatch repair and MSI [microsatellite instability]–high. These are alterations that can predispose to a positive effect from agents such as pembrolizumab and PD-1 inhibitors. Overall, we have a variety of biomarkers that we can utilize. Some are measured on the surface of cells, like PSMA, and we use PET imaging. We’re looking at others more molecularly with BRCA1, BRCA2, and mismatch repair. But the implications for this heterogeneity are important because it can help clinicians like myself choose better therapies.
Transcript edited for clarity.