This article is more than 5 years old. Information may no longer be current.
Deadly subtype makes up 17% of metastatic prostate cancers
Treatment-emergent small cell neuroendocrine prostate cancer, a particularly deadly subtype of the disease, occurs in nearly one-fifth of all cases of metastatic, castration-resistant prostate cancer, study data showed.
Researchers suggested that the subtype should be treated with novel targeted therapies that are currently in the development or testing phase.
“Think of advanced, hormone treatment-resistant prostate cancers as a pie,” Rahul Aggarwal, MD, assistant professor of medicine at University of California, San Francisco, said in a press release. “Instead of treating these advanced cases homogenously as we do with today’s standard treatments, we want to split the pie according to tumor characteristics, and develop treatment protocols tailored to individual slices, based on the cancer’s distinctive growth-driving genetic mutations and gene expression patterns.”
Androgen receptor-targeting therapies have provided significant clinical benefit for metastatic castration-resistant prostate cancer. However, a subset of patients who experience therapeutic resistance to this therapy develop a new histologic subtype — treatment-emergent small cell neuroendocrine prostate cancer — that resembles the highly aggressive de novo small cell subtype.
Aggarwal and colleagues evaluated the prevalence and characteristics of this treatment-emergent subtype to inform tumor classification, clinical recommendations and future therapies.
The researchers performed metastatic tumor biopsies on 202 patients with castration-resistant prostate cancer. All biopsy specimens were independently reviewed and underwent DNA/RNA sequencing.
Seventy-three percent of patients (n = 148) experienced progression after being treated with abiraterone (Zytiga, Janssen) and/or enzalutamide (Xtandi, Astellas).
Seventy-nine percent of biopsies were evaluable.
The overall incidence of detected treatment-emergent small cell neuroendocrine prostate cancer was 17%.
Androgen receptor amplification occurred in 67% of treatment-emergent small cell neuroendocrine prostate cancer biopsies, and androgen receptor protein expression was present in 75%.
Similar proportions of the disease subtype occurred in bone, node and visceral organ biopsies.
Alterations in DNA repair pathway genes — such as BRCA1, BRCA2, ATM, CDK12, RAD51, PALB2, FANCA, CHEK2, MLH1, MSH2, MLH3 and MSH6 — were “almost entirely mutually exclusive” with treatment-emergent small cell neuroendocrine prostate cancer tumors (8% vs. 40%; P = .035).
Treatment-emergent small cell neuroendocrine prostate cancer was associated with a shorter OS among men who had received previous androgen receptor-targeting therapy (HR = 2.02; 95% CI, 1.07-3.82).
Researchers used unsupervised hierarchical clustering of the transcriptome to identify “a small cell-like cluster” that appeared associated with worsened survival (HR = 3; 95% CI, 1.25-7.19).
Aggarwal and colleagues developed a transcriptional signature for treatment-emergent small cell neuroendocrine prostate cancer and validated the signature with greater than 90% accuracy.
The researchers identified several transcriptional regulators of the disease subtype, identifying the pancreatic neuroendocrine marker PDX1 as the most active.
“With novel therapies in clinical development, there is the potential to improve disease outcomes for this high-risk and increasingly prevalent subset of metastatic castration-resistant prostate cancer,” the researchers wrote. – by Andy Polhamus
Disclosures: The authors report no relevant financial disclosures.
Perspective
Back to Top
It is increasingly well recognized that neuroendocrine variants of prostate cancer may occur de novo as small cell carcinoma, but much more commonly arise following systemic therapy directed against the androgen receptor. However, there is a paucity of prospective studies to assess this phenomenon.
The study by Aggarwal and colleagues is important as a prospective study among patients with metastatic castration-resistant prostate cancer with both tissue and clinical data collection, enriched with patients with tumors resistant to abiraterone and/or enzalutamide. There was a high rate of having evaluable tissue — the majority coming from bone (55%) — and researchers demonstrated prognostic value for histology consistent with small cell/neuroendocrine carcinoma, as well as a prognostic transcriptional gene signature.
Multiple terms are used to designate this clinical/pathologic disease state. The “t” prefix was initially used to signify the emergence of this phenotype/genotype following systemic therapy and by no means implies causality of androgen receptor-targeted drugs — and, therefore, appropriate patients should still receive those therapies. Differences in terminology, such as treatment-emergent neuroendocrine prostate cancer (tNEPC), treatment-emergent small-cell neuroendocrine prostate cancer (t-SCNC), or aggressive-variant prostate cancer, demonstrate overlapping phenotypes of treatment-resistant disease with a poorer prognosis than those without these labels, but also heterogeneity of the disease with different dominant pathways.
Trends for neuroendocrine differentiation and NMYC, TP53 and RB1 alterations are seen to varying degrees across different series. However, patient and tumor selection are likely responsible for some of the differences, with prior studies of tNEPC selecting for tumors with neuroendocrine and low androgen receptor activity by definition; this study demonstrates persistence of androgen receptor activity in many samples. Relevant genomic and transcriptomic alterations give credence to the current early-phase clinical trials examining the safety and efficacy of drugs targeting EZH2 and DLL3, among others.
A bit of conflicting data in the Aggarwal study compared with other data sets is the minimally overlapping presence of DNA repair pathway alterations in the setting of t-SCNC, whereas other groups have demonstrated no exclusivity for these genotypes — and possibly some enrichment for tNEPC in the setting of DNA repair alterations. It is unclear if this has to do with different patient populations or histologic/genomic definitions of t-SCNC vs. tNEPC.
Most — or all — of the patients included in this series with prior potent androgen receptor-targeted therapy received these drugs in the setting of metastatic castration-resistant prostate cancer. As we are increasingly using abiraterone, enzalutamide and apalutamide (Erleada, Janssen Oncology) — plus additional investigational agents — in the initial treatment of hormone-sensitive metastatic disease or for nonmetastatic (M0) castration-resistant prostate cancer, it is unknown whether first-line treatment for these patients will reflect similar clinical features, histologies, genomics and outcomes as demonstrated in this mostly second- and third-line study with a 17% incidence of t-SCNC. However, it seems likely and clinicians should maintain clinical vigilance.
The role of serum neuroendocrine markers remains unknown. This study points toward differences in median neuron specific enolase (NSE) among patients with t-SCNC, but absolute numbers are not that different and median values for both groups were in the normal range for the assay; median chromogranin-A (CgA) levels were no different. The authors comment that a very low serum NSE or CgA — less than approximately one-third of the upper limit of normal for the assays utilized — was associated with a very low chance of biopsy identifying t-SCNC. This finding should be assessed in an independent data set as a clinically available biomarker with potential negative predictive value.
The authors should be applauded for completion and publication of this prospective study. Unfortunately, treatment-resistant disease remains an unmet need with subgroups such as t-SCNC and tNEPC comprising a substantial minority. Studies such as this one may identify mechanisms of resistance and physicians should refer such patients to clinical trials.