Q: For patients with CRPC and no overt metastases, what does optimal care and monitoring consist of?

In the past, we considered advanced therapies only in the setting of castration-resistant metastatic disease. Now we are seeing the value of using these agents earlier in patients with nmCRPC. I think that the real issue here is trying to identify the patients who will achieve benefits despite the risks of side effects and those who may not require any treatment. This is frequently determined by the characteristics of the primary tumor, the time to biochemical recurrence, the timeline of androgen deprivation, and PSA doubling velocity. In an older patient with nmCRPC and a prolonged PSA doubling time beyond 10 to 12 months, the optimal choice may be continued surveillance without additional treatment. I might also consider finding a trial that is adapted to the patient’s disease state and comorbidities. A younger patient who is vibrant with a rapidly rising PSA might be treated differently, with earlier initiation of AR-blocking therapy. The issue of overtreatment exists for all stages of the disease and is particularly important in patients with low-risk disease. The type and frequency of follow-up imaging is quite variable across settings, and this makes it difficult to compare the monitoring of 1 patient with another. In an asymptomatic patient, I think that PSA is your first line of monitoring. In the absence of metastasis, symptoms tend to be treatment related; however, symptoms that emerge, such as bone pain or cognitive changes, must be evaluated because they may represent progression to overt metastatic disease.

When a patient without metastases who is asymptomatic develops symptoms, we all agree that you need to really listen and be very responsive because it is possible that the symptoms represent an underlying manifestation of metastatic disease. When you use PSA to monitor, you are gathering information that was utilized in the clinical trials, but not as a primary end point. In virtually every case that I am aware of, you will have PSA progression before you have radiographic progression, but the interval between the 2 can be quite long. With PSA, the question is not only whether the level is rising, but also how much and how fast is it rising? You may have an asymptomatic patient with a very minor PSA increase (eg, from 0.2 ng/mL to 0.3 ng/mL) who becomes alarmed and begins to seek all types of treatments that would not be justified. How long you continue therapy while monitoring with PSA and imaging is still a bit of an art form. How do I monitor these patients? Well, typically, I would see an asymptomatic patient with M0 CRPC who is on an AR-targeted therapy once every 2 to 3 months, but this can depend on several factors, including the distance that the patient has to travel. I use a combination of PSA, symptom assessment, and imaging that is not done at fixed intervals but is rather dependent, in part, on the patient’s PSA and symptoms. In an asymptomatic individual with declining PSA, imaging rarely uncovers anything fruitful. But if the PSA is rising—particularly if the PSA is rising rapidly—or is accompanied by symptoms such as back pain, new hip pain, and unilateral leg swelling, that patient will be assessed more comprehensively. It is possible that the introduction of newer AR-targeted agents earlier in the natural history of disease may be leading to more PSA-negative tumors in a manner such that current monitoring schemes may no longer be optimal; however, we are in a bit of a transition zone. For now, I am not willing to scan every patient at a fixed interval, but I am interested in learning more about possible triggers for more appropriate scanning.

This is an interesting area because I do think that we are changing the natural histories of these patients, and I think that the optimal imaging frequency for patients with M0 CRPC is a bit of an unknown. By using some of these AR-targeted therapies earlier, we are definitely seeing, for example, more soft-tissue metastases. This is anecdotal, but I recently saw a patient with CRPC who had been on enzalutamide for approximately 1.5 years and was doing really well, was completely asymptomatic, and his PSA was 0; however, we found some incidental new liver lesions during routine scanning. And then, Bryce and colleagues recently reported the results of a post hoc analysis from the PREVAIL trial, in which, as we all recall, patients had confirmed metastatic (M1) CRPC with documented metastases and PSA progression, radiographic progression, or both in bone or soft tissue. Approximately 25% of the patients treated with enzalutamide had disease progression in the absence of a rising PSA level, which was higher than I would expect. I do not want to overreact to my anecdotal patient, but I am seeing more scenarios in which the imaging is discordant with the PSA. If PSA is our only window of observation, we will miss the very poorly differentiated, molecularly different anaplastic neuroendocrine type of cancer cells that do not generate PSA. To the extent that M0 CRPC will eventually progress to M1 CRPC, imaging may be indicated if the PSA has doubled since the prior scan, as suggested by the RADAR III Group, and I am also interested in evolving data to support additional triggers for appropriate scanning.