Advancements in imaging technology, such as prostate-specific membrane antigen (PSMA)–targeted positron emission tomography/computed tomography (PET/CT) scans, have proven valuable in diagnosing prostate cancer and monitoring disease progression. Some complementary biomarkers have shown promise in estimating prognosis and predicting treatment response.

Prostate-specific antigen (PSA) remains an important biomarker for prostate cancer. It is not perfect, but it is inexpensive and widely available, and it still gives us insight into the benefit of treatment. In most situations, the definitive way of understanding whether a treatment is working is with repeat imaging. And I think that the decision of when to do a repeat PSMA PET/CT scan depends on the situation. For example, I will only do a scan if I think that the cancer is progressing. Otherwise, if I have a patient whose PSA level is going down on ¹⁷⁷Lu PSMA therapy and he feels great, I do not do another diagnostic PSMA PET/CT scan until the halfway point of treatment (ie, after approximately 3 cycles).

In the past, oncologists looked at the images alongside radiologists, but now I am relying on the radiologist’s report to give me a sense of what we are dealing with in terms of the specific lesions and their location, among other factors. And then I am looking at the mean standardized uptake value (SUVmean) and the maximum SUV (SUVmax) to understand the likelihood that a treatment will be beneficial.

My favorite reports are those that read out every single lesion with size and SUV measurements. These reports are useful, particularly for monitoring. This way, I can describe to the patient whether the treatment is working based on these specific changes. It may not correlate with the most important factors over the long-term, such as survival, but it is a lot like PSA response. Patients receiving therapy want to know that the treatment is benefiting them. So, even if PSA or SUVmean and SUVmax changes are flawed biomarkers, they are what I look at to determine treatment efficacy.

Circulating tumor DNA (ctDNA) fraction is also considered a biomarker in this space, but I will caution that it is only research focused right now. The issue with ctDNA fraction is that we know it is a negative prognostic marker, and questions remain. How often should we be doing this? Do we really need an additional prognostic marker? Unlike PSA, ctDNA tests are not inexpensive, and most of the time, tests such as PSA or repeat imaging will give us the same information. I also have not seen any data that support ctDNA fraction as a valuable predictive marker.

As for specific genes, I do think that this is an interesting area and that the work is just beginning. The idea that patients with DNA damage repair mutations are more likely to benefit from radioligand therapy (RLT) has been around for some time, but—at best—the percentage of patients with these mutations is very low. The vast majority of patients receiving and benefiting from RLT do not have specific genetic markers. This is just a reminder that we certainly do not want to exclude patients who might otherwise benefit from a treatment simply because they do not have identifiable biomarkers. We want to be able to offer these therapies regardless.