Dr. Aparicio began by emphasizing that decades of research have demonstrated that the predominance of advanced prostate cancers, including those that are castration resistant, are driven by the androgen receptor pathway. However, approximately 20% of tumors may be so-called androgen indifferent and are driven by pathways other than androgen signaling though the pathways have yet to be well characterized.
One of the issues in studying androgen indifferent cancer is the lack of accepted definition or nomenclature: small cell neuroendocrine prostate cancer is one example characterized by histologic appearance though the underlying biology extends beyond those tumors with the typical histologic appearance and includes a variety of tumor types with a variety of definitions. Some investigators have relied upon neuroendocrine markers while others have utilized a phenotype including disease biology, markers, and morphology. Some relevant clinical characteristics including low PSA levels relative to tumor burden, exclusive visceral metastasis, predominantly lytic bony metastasis, bulk tumor mass, elevated non-PSA markers (including CEA and LDH), short (<6 months) response to androgen deprivation therapy, and others. Molecularly, Dr. Aparicio described a characteristic pattern of alterations in two of TP53, RB1, and PTEN. However, none of these are unanimously used across definitions though there is some overlap.
One of the relevant questions is whether this androgen indifferent biology has increased in frequency with the increasing utilization of advanced androgen inhibition with second-generation androgen signaling inhibitors. Indeed, work from Blumin et al. in an autopsy series suggested just this trend over time. Androgen receptor active prostate cancer decreased in frequency over time. Further, the double negative type increased dramatically in the decade that followed the approval of abiraterone acetate and enzalutamide.
Further, genomic studies using paired samples of primary hormone-sensitive prostate cancer and specimens collected from the same patients once they had developed metastatic castration resistant disease demonstrated enrichment in AR amplifications as all as mutations including TP53, RB1, PTEN, FANCA, MYC, and AKT1 from the samples collected at mCRPC, compared to baseline. Similarly, in the mCRPC setting, treatment with enzalutamide resulted in a significant increase in the frequency of genomic alterations. This could also be demonstrated using assessment of circulating tumor cells rather than biopsy specimens. However, in studies using ctDNA, some alterations seen at baseline were no longer seen following treatment with either enzalutamide or abiraterone. This supports that the genetic evolution of prostate cancer is non-linear.
Dr. Aparicio then emphasized that there is evidence to support that androgen indifference may be both present de novo and induced by treatment. This, therefore, suggests different treatment approaches, first, considering directed therapies upfront and second, considering anticipatory strategies to prevent the development of this phenotype. Each of these depends on robust biomarkers to identify these disease characteristics.
In terms of treatment options, one approach for androgen indifferent disease has been to use platinum-based chemotherapy. There are a number of single-arm phase II studies that explored a variety of combination treatment regimes, also relying on a number of different eligibility criteria.
One observation of these trials is that high levels of neuroendocrine markers were not associated with the benefit from the use of platinum-based chemotherapeutic approaches. However, the AVPC molecular profile (comprising 2 or more alterations in TP53, RB1, and PTEN by immunohistochemistry and or genomic sequencing) is associated with benefit from platinum-based chemotherapy. This trial used the molecular profile to stratify patients who were then randomized to cabazitaxel with or without carboplatin, finding that in those with the biomarker positive disease, there was an overall survival benefit to the addition on carboplatin where no difference was seen in those who were biomarker negative.
Other studies have examined aurora A kinase inhibition with alisertib, based on elegant biochemical studies. However, median progression-free survival in a single-arm study was only 2.2 months. Further, work has examined Delta-like 3 (DLL3) which is highly expressed in neuroendocrine carcinoma. Rova-T is a DLL3-targeting antibody-drug conjugate that, unfortunately, did not demonstrate antitumor activity in 21 patients with small cell neuroendocrine prostate cancer.
As with many tumor sites, there has been interest in immune checkpoint inhibition in patients with androgen indifferent prostate cancer. However, across a handful of studies including basket trials of nivolumab and nivolumab as well as a prostate cancer-focused study of avelumab, there doesn’t appear to be significant activity of this approach.
However, there are many ongoing trials focusing on this disease subset, building often on the backbone of platinum-chemotherapy.
Overall, she emphasized that the best treatment for these patients is a clinical trial.
Presented by: Ana Aparicio, MD, Associate Professor, Department of Genitourinary Medical Oncology, The University of Texas MD Anderson Cancer Center, Houston, Texas
Written by: Christopher J.D. Wallis, Urologic Oncology Fellow, Vanderbilt University Medical Center Contact: @WallisCJD on Twitter at the 2021 American Society of Clinical Oncology (ASCO) Annual Meeting, Virtual Annual Meeting #ASCO21, June, 4-8, 2021