Circulating MicroRNA-23a-5p Is a Potential Biomarker for Recovery of Bone Mass After Parathyroidectomy in Primary Hyperparathyroidism

Abstract

Circulating microRNA (miRNA), a non-coding RNA molecule, is the potential biomarker for bone and mineral metabolism. Primary hyperparathyroidism (PHPT) is one of the leading causes of secondary osteoporosis. After parathyroidectomy, BMD improves as parathyroid hormone (PTH) decreases in patients with PHPT but degree of recovery in BMD vary greatly among patients. Biomarkers which are able to predict postoperative bone response may facilitate individualized approach to PHPT treatment. In this study, we aimed to investigate whether changes of miRNA related to PHPT can predict postoperative BMD recovery. A total 12 patients (10 women) who underwent parathyroidectomy for PHPT at Severance hospital from 2016 to 2019 were analyzed. A total of 15 selected miRNAs (miR-19b-3p, miR-21-5p, miR-23a-3p, miR-152-3p, miR-335-5p, miR-375, miR-532-3p, miR-24-3p, miR-100-5p, miR-125b-5p, miR-24-5p, miR-93-5p, miR-122-5p, miR-124-3p, and miR-148a-3p) from literature review were measured from pre- and pot-operative serum samples. Among pre-, post-operative level, and fold changes [the normalized gene expression(2-ΔCt) in postoperative period divided the normalized gene expression (2-ΔCt) in pre-operative period] of each miRNA, potential predictors of femoral neck BMD changes (%) as primary outcome were selected by univariable regression. The mean age of patients during parathyroidectomy was 56±16 years. PTH level and calcium level restored to normal range after parathyroidectomy in all subjects. (pre to postoperative: PTH, median 113 to 42; calcium, mean 10.6 to 9.2). Circulating level of miR-19b-3p, miR-21-5p, miR-100-5p, miR-125-5p, miR-152-3p and miR-335-5p dropped significantly after parathyroidectomy (p<0.05 for all). A fold change in miR-23a-5p showed a negative correlation with changes (%) in femur neck BMD (β coefficient=-10.58, 95% CI=-19.84 to -1.31, p=0.029). The association remained robust (β coefficient=-7.25, 95% CI=-14.17 to -0.34, p=0.042) even after adjustment for gender, age, and preoperative PTH level in multiple linear regression model. Preoperative miR-124-3p showed a positive correlation with changes in total hip BMD (β coefficient=-9.748, 95% CI= 3.740 to 15.755, p=0.005), but there was no association after adjustment for gender, age, and preoperative PTH level. In conclusion, a fold change in circulating miRNA-23a-5p predict the degree of improvement for femoral neck BMD. MiRNA have potential to serve as novel biomarker to predict treatment response in patients with PHPT, which merits further investigation.