Transcriptome sequencing reveals distinct atypical parathyroid tumor subtypes.

Abstract

Atypical parathyroid tumors (APTs) are a rare subtype of parathyroid neoplasms characterized by diagnostic challenges and an uncertain prognosis. This study aimed to validate the subtypes of APTs using transcriptome sequencing. We applied a clustering model developed for our previous study, in which we had successfully distinguished parathyroid cancer from adenomas using gene expression patterns. Sixteen patients with APT who had undergone parathyroidectomy were enrolled, and we analyzed their baseline data, pathologic reports, and follow-up records and performed transcriptome sequencing of their APT samples. We then used our clustering model to classify tumors as either cancer- or adenoma-type APTs and compared these results with clinical findings. The median age of patients was 48.9 years, with median calcium and parathyroid hormone (PTH) levels of 11.4 mg/dL and 420.0 pg/mL, respectively. Pathologic and immunohistochemical results did not reveal any remarkable differences between adenoma-type and cancer-type APTs. However, clustering analysis classified four of the 16 APTs as being cancer-type and 12 as being adenoma-type tumors. Cancer-type patients had a median age of 30.0 years, with median calcium and PTH levels of 12.6 mg/dL and 800.8 pg/mL, respectively, clinically resembling parathyroid cancer. One patient exhibited a somatic CDC73 two-hit mutation and positive WT1 staining, suggesting a high malignant potential. Clustering analysis through transcriptome sequencing shows promise for risk stratification of patients with APTs. For those classified as having cancer-type tumors, close monitoring and long-term follow-up may be warranted.