High-risk localized and locally advanced disease is a broad, operationally defined category, one that will require much more definition in the future. Risk is currently defined by factors such as grade, prostate-specific antigen, and T stage—individually or in combination. Additionally, we are starting to incorporate molecular imaging biomarkers and genetic information. 

The first abstract in this section, abstract 14, dealt with genetic drivers of prostate cancer and potential racial differences. We know that African American men suffer a disproportionate burden of prostate cancer in this country and around the world, and we suspect that disparities may reflect social determinants, genetic factors, or a combination of both. The investigators conducted a real-world analysis of genetic factors and found a high frequency of mutations in key prostate cancer drivers (eg, AR, TP53, SPOP, and homologous recombination repair pathway gene mutations) in African American men relative to men of other ancestries. This is a reasonable approach, provided that we remain mindful that the reasons for disparities in outcomes can be very complex (eg, the authors noted that genetic findings could be a reflection of the more advanced disease presentation of African American men due to socioeconomic factors/health care access). Interestingly, a recent thoughtful editorial by Ioannidis et al in JAMA on this very topic (ie, race in medical research) made the point that race may be a poor surrogate of social constructs, let alone biology. So, I think that we need to understand that complexity in more detail to understand disparities.

Next-generation imaging was the focus of several analyses. The authors of abstract 32 showed that, in the phase 2/3 OSPREY trial, prostate-specific membrane antigen (PSMA)–targeted ¹⁸F-DCFPyL positron emission tomography (PET)/computed tomography (CT) imaging improved staging with high specificity. Similarly, abstract 33 reported favorable data from the phase 3 CONDOR study with ¹⁸F-DCFPyL PET/CT in patients with biochemically recurrent prostate cancer. There is no question that PSMA-based imaging improves our knowledge over conventional imaging and can affect staging and treatment in a significant proportion of cases. What is not known is whether this will affect long-term outcomes. We commonly use other forms of imaging in a wide variety of patients without knowing its impact, and I favor the wider use of PSMA PET/CT (especially now that it has undergone US Food and Drug Administration review); still, patient selection and appropriate use is important.

A growing body of work supports the use of genomic scores to predict risk and guide treatment. Abstract 195 reported the ability of the combined clinical cell-cycle risk (CCR) score to determine the risk of metastasis in those receiving dose-escalated radiation therapy with or without androgen deprivation therapy. The CCR score combines the cell cycle progression score with the University of California, San Francisco-Cancer of the Prostate Risk Assessment (UCSF-CAPRA) score, and, in this study, patients with a favorable CCR score did not significantly reduce their 10-year risk of metastasis with the addition of androgen deprivation therapy. This trial shows that combining genomics with clinical pathologic features improves the ability to predict metastasis. Thus, this could be used to individualize treatment and may help to avoid additional hormonal treatment in those who would not benefit from such therapy. In the past, whether it was Oncotype DX, Prolaris, or Decipher, these genomic classifiers were used as prognostic markers, but now there is great interest in predictive applications.

Additionally, new approaches to impact the primary tumor prior to radical prostatectomy continue to be investigated. Abstract TPS264 described the open-label phase 1/2 LuTectomy study design, which evaluated whether ¹⁷⁷Lu-PSMA-617 (Lu-PSMA), administered before radical prostatectomy in men with high-risk clinically localized disease, will deliver high doses of radiation to the prostate and lymph nodes. If the trial results demonstrate that lutetium has low toxicity and can clear disease from the prostate and regional lymph nodes, then Lu-PSMA might be able to eliminate the need for prostatectomy in the future. So, researchers are going to test this hypothesis. Right now, most of the work on Lu-PSMA is in the high-risk metastatic castration-resistant setting.

Elsewhere at the symposium and in the recent literature, we continue to see that there may be pathologic predictors of response to hormonal therapy (eg, with luminal and basal subtyping). In the past, risk assessment was being driven primarily by knowledge of T stage, grade, prostate-specific antigen, and percent positive biopsy. All of these factors are included in the UCSF-CAPRA score. Today, we appreciate that we are dealing with a spectrum of disease, and we should try to quantitate it as best we can. This could be through PSMA, genomic classifiers (commercialized and new classifiers that are being developed), multivariable risk profilers such as the UCSF-CAPRA score, or, more likely, a combination of all 3.