In the absence of level 1 evidence, there are multiple options for sequencing. We are starting to try to incorporate molecular targeting into treatment sequencing, which has been a very positive development, and we are making some progress with poly (ADP-ribose) polymerase inhibitors. None of us have simple solutions, or one-size-fits-all algorithms, to therapeutic sequencing, but, ultimately, it will all be based on our understanding of the disease biology, together with testing in prospective clinical trials.

In prostate cancer, there are multiple drugs available with many different mechanisms of action. We also have multiple drugs available within a single class. This presents challenges to treatment sequencing in prostate cancer, and some of these challenges are distinct from those of breast and lung cancers, for which there is an established, evidence-based sequence of therapies in a number of clinical settings.

In contrast to other solid tumors, there is more diversity in the clinical backgrounds of the physicians who treat patients with advanced prostate cancer, which can lead to slightly different treatment patterns or trends. We have urologists, medical oncologists who focus on genitourinary oncology, medical oncologists in the community, radiation oncologists, and, in some settings, nuclear medicine physicians. At times, there may be different approaches in therapeutic sequencing in different clinical settings. For example, community oncologists who practice in rural areas may have less access for their patients to treatments such as sipuleucel-T, radium-223, and lutetium Lu 177 vipivotide tetraxetan. Thus, in addition to factors such as the disease biology, prior lines of therapy, and economics, the clinician’s background and their extent of experience with prostate cancer can also drive therapeutic decision making.

We do have more treatment options than in the past, and, although this is a positive development, it also presents certain challenges. We do not always have a single, formalized, standard sequence, but we do have helpful data in some instances. For example, there are several lines of evidence indicating that switching between novel antiandrogens such as abiraterone and enzalutamide may not be the best strategy. In the CARD study, we saw that switching to chemotherapy (ie, cabazitaxel) improved a number of clinical outcomes vs switching from one androgen receptor (AR)–targeted agent to the other. Additionally, the PROfound study showed that, for men with homologous recombination deficiency, switching to olaparib is much more effective than using the other AR-targeted agent. Further, cross-resistance has been observed with the novel antiandrogens. Thus, I think that it can be helpful to think in terms of maximizing the patient’s access to all different lines of treatment, including chemotherapy, which should not be considered as only a last resort.

Regarding aspects of the disease biology, I agree with Dr Dreicer that poly (ADP-ribose) polymerase inhibition is a good example of molecular targeting. AR-targeted therapy and prostate-specific membrane antigen (PSMA)–targeted radioligand therapy are other examples of treatments that are directed toward specific molecular phenotypes. Importantly, both AR sensitivity and PSMA expression are phenotypes that may change in later disease stages.

Fortunately, we have an opportunity to target the AR and PSMA in the majority of our patients; however, with regard to sequencing, I would emphasize that obtaining a biopsy may be important in patients who have metastatic castration-resistant prostate cancer that is progressing, due to the possibility of neuroendocrine dedifferentiation. When prostatic adenocarcinoma differentiates into a neuroendocrine-type cancer, the traditional prostate cancer treatments are relatively ineffective. We might suspect neuroendocrine dedifferentiation when a patient develops a liver metastasis, but sometimes it is not that obvious. Rapid progression should certainly raise the suspicion for neuroendocrine differentiation, and a biopsy later in the disease course can be critical.

Throughout the prostate cancer literature, there are certain words or phrases that are used over and over again. For example, we often hear that androgen deprivation therapy is the mainstay, cornerstone, or foundational aspect of advanced prostate cancer treatment. Another word that is frequently used in prostate cancer is heterogeneity. Indeed, there is a lot of biological heterogeneity in prostate cancer. We are now—very importantly—moving more toward precision-based therapy, or what some would call personalized or tailored therapy, and that is often not a one-size-fits-all undertaking.

As relates to genomic profiling, there is a hereditary cancer risk from germline alterations, and detecting these alterations can be extremely informative. Further, through cascade testing, we can identify individuals within the patient’s family who have inherited a pathogenic variant and who may be at risk for a different disease. Somatic testing, where the tumor itself is the source of the genetic alteration, can be ascertained through either tissue- or blood-based testing. 

As Dr Dorff alluded to, we now have PSMA-based imaging and the ability to target PSMA-expressing disease. Advanced imaging allows us to visualize disease that cannot be seen on conventional imaging. Detecting genetic alterations and employing advanced imaging–based strategies can further allow us to optimize therapeutic sequencing based on a patient's individual biology. Finally, artificial intelligence, deep machine learning, and pattern recognition based on histopathological slides are also adding to the knowledge base and may help to answer some important questions related to therapeutic sequencing in advanced prostate cancer.