Q: How important is MFS as an end point in nmCRPC?

This is a complex question because these are asymptomatic patients, but the transition from nonmetastatic CRPC (nmCRPC) to detectable metastatic disease is an event that can be associated with pain and illness and can result in the need for additional interventions. Further, we all objectively recognize that the development of metastasis is upsetting to patients and can lead to other consequences. All 3 of the trials of next-generation androgen receptor (AR) antagonists for nmCRPC (ie, PROSPER, SPARTAN, ARAMIS) show a roughly 2-year delay in MFS, and I think that 2 years of delaying visible bone metastasis is significant and clinically meaningful. 

The field is evolving very rapidly with advances in molecular imaging. In nonmetastatic patients with rising prostate-specific antigen (PSA)—whether biochemically relapsed without androgen deprivation therapy or in the CRPC setting—it is not that they are truly nonmetastatic, but rather that we do not detect the metastases using conventional imaging. ¹⁸F-fluciclovine positron emission tomography (PET) is available and has proven very useful, and there is great interest in gallium-68 prostate-specific membrane antigen (PSMA)–PET imaging, which is not commercially available in the United States but might be much more sensitive and specific than fluciclovine. In any case, the question that arises is: When we do find metastases using next-generation imaging, what can we do about them? We do not know, for example, whether radiating lesions that we may find in the lymph nodes or in bone will change the outcome for these patients. We do not know whether detecting microscopic metastases will improve patient outcomes either. Should small lesions on next-generation PET scans be treated with an agent to delay progression to metastasis? Or should that patient get some type of targeted radiation directed at the metastasis? Or perhaps both? These are all unanswered questions. What we do know now is that this relatively narrow population of patients with nmCRPC is at risk of metastatic disease, particularly those with rapid PSA doubling time. This is where the next-generation AR antagonists such as enzalutamide, apalutamide, and darolutamide can come into play, by prolonging the metastasis-free interval.

There are issues associated with the use of an operational definition of metastases and how we currently define metastatic disease, but I think that MFS should correlate with overall survival. MFS is a valuable end point that provides us with earlier insights into the disease than we would otherwise have. The detection of metastatic disease is an area that has been revolutionized with the availability of advanced imaging techniques and newer forms of molecular characterization, including circulating tumor cells and cell-free DNA. We are now better able to calculate the likelihood of being disease free with any form of local therapy and we can predict which patients are more likely to benefit from post-surgical radiation. In the past, our treatment was based on a more limited understanding of the disease. There was a time when our approach was primarily PSA based, with less knowledge of the exact extent of disease, and it was classified as either localized or distant. But we are now characterizing the extent of disease much more accurately. With the use of PET imaging, we can better identify both local/regional disease and more limited presentations of metastatic disease. If I see a patient with CRPC who has never had advanced imaging, we turn to PSMA-PET. The question now is whether advanced imaging can provide an oncologic benefit. We are seeing unusual patterns of metastasis to the lungs, cervical lymph nodes, and other locations that were not detected on standard imaging. We still must demonstrate whether treating these patients with stereotactic body radiotherapy, salvage lymph node dissection, or other interventions, for example, actually impacts oncologic outcomes.

Indeed, molecular imaging is revolutionizing our understanding of “nonmetastatic” prostate cancer. There are 5 types of PET scans that are potentially relevant in this space, and it is important that we specify which type of imaging we are referring to. Much of the excitement today is within the context of gallium-68 PSMA-PET imaging, which would locate disease in almost all of these patients who were defined as nonmetastatic on the basis of bone scan, computed tomography, or magnetic resonance imaging. We are finding that many of these PSMA-positive lymph nodes, including those that are pathologically proven to be positive, were actually of normal size. The nonmetastatic disease space is revolutionary right now, and we are beginning to redefine the nomenclature for advanced disease. In the future, what we now refer to as nonmetastatic may shift to oligometastatic, and the optimal treatment for oligometastatic disease remains to be defined in prospective clinical trials. PSA doubling time is a fabulous tool that is inexpensive and easy to use, but, as we are molecularly defining the tumors, we are creating new scenarios that affect our understanding of the natural history of the disease. In addition, we are also treating these patients earlier and with different drugs in the investigative setting, and the issue of cross-resistance can arise.