Metastatic prostate cancer diagnosed by fine‐needle aspiration: Contemporary cytopathologic and biomarker assessment with clinical correlates

INTRODUCTION: The diagnosis of metastatic prostatic cancer (MPC) by fine needle aspiration (FNA) can usually be rendered by typical cytomorphologic and immunohistochemical (IHC) features. However, MPC diagnosis may be complicated by transformation to atypical phenotypes such as small cell carcinoma, typically under pressure from androgen deprivation therapy (ADT). Predictive and prognostic biomarkers can also be assessed by IHC. This study illustrates how careful assessment of cytologic and biomarker features may provide therapeutic and prognostic information in MPC. DESIGN: We reviewed our anatomic pathology archives for MPC diagnosed by FNA from January 2014 to June 2021. Clinical histories, cytology slides, and cell blocks were reviewed. Extensive IHC biomarker workup was performed, including markers of prostate lineage, cell‐cycle dysfunction, Ki‐67, neuroendocrine markers, PDL1, and androgen receptor splice variant 7. Cases were reclassified into three categories: conventional type, intermediary type, and high‐grade neuroendocrine carcinoma (HGNC). RESULTS: Eighteen patients were identified. Twelve had conventional MPC, including six of six ADT‐naive patients. Six of twelve (50%) with prior ADT were reclassified as intermediary or HGNC. Four intermediary cases included two with squamous differentiation and two with pro‐proliferative features. Two HGNC cases had typical small cell carcinoma cytomorphology. Expression of PDL1 was identified in two cases and ARv7 in three cases. Five of five intermediary and HGNC patients died of disease versus six of eleven with with conventional type. CONCLUSIONS: Aggressive cytomorphologic variants were commonly identified in patients with prior ADT. Identification of nonconventional cytomorphology and increased proliferation can provide important prognostic information. Recognition of these changes is important for an accurate diagnosis, and the identification of high‐grade variants can affect therapeutic decision‐making. Clinically actionable biomarkers such as PDL1 and ARv7 can be assessed by IHC.