Plasma Androstenedione Concentration Can Discriminate Frail versus Non-Frail Men with Prostate Cancer under Androgen Deprivation Therapy

Abstract

Background: Androgen deprivation therapy (ADT) is a mainstay of prostate cancer in both adjuvant and palliative settings. Since androgens are crucial for functional status and psychological functions, we evaluated whether blood testosterone, androstenedione, or DHEA concentrations were associated with functional status and psychological alterations in patients with localised (PCa) or metastatic prostate cancer (mPCa) receiving ADT with analogues of luteinising hormone-releasing hormone (LHRH).

Methods: The five Fried criteria were considered to identify frailty syndrome. In addition, complementary evaluations were carried out to measure other variables of interest. Sleep quality was assessed using the Athens Insomnia Scale, cognitive functions were assessed using the Mini-Mental State Examination, and symptoms of depression were measured using the Yesavage Geriatric Depression Scale. Logistic regression analysis was performed to determine if the androgens level could be related to frailty syndrome, sleep impairment, depressive symptoms, and cognitive functions.

Results: The results of the multivariate analyses show that high concentrations of androstenedione were significantly associated with frailty syndrome in both groups (p = 0.018; odds ratio = 4.66, 95% confidence interval [1.30-16.6]). There were significant relationships between frailty syndrome and the systemic concentration of androstenedione (p = 0.01), but not the concentration of testosterone (p = 0.60) or DHEA (p = 0.42). In addition, the results of the non-parametric tests show significant results between a decreased gait speed in the two groups (metastatic and localised) and the concentration of androstenedione (p = 0.015). High androstenedione levels were associated with a slow walking speed in the mCaP group (p = 0.016), while high testosterone levels were associated with a better walking speed in the localised CaP group (p = 0.03). For the concentration of androstenedione in plasma, the area under the curve was 0.72, with a 95% CI of 0.55-0.88 with acceptable values, and with a cut-off point of 4.51 pg/mL, a sensitivity of 82.9%, and specificity of 53.8%. No relationships between the concentration of androgens in plasma and sleep quality, cognitive functions, or symptoms of depression suggest that the changes were specific to frailty syndrome.

Conclusions: Further research into the role of androstenedione should be evaluated in follow-up studies in order to recommend its use as a suitable biomarker of frailty syndrome in prostate cancer patients.

Keywords: DHEA; LHRH analogues; androstenedione; frailty syndrome; geriatric assessment; localised prostate cancer; metastatic prostate cancer; testosterone.

Figure 1

(A) Plasma androstenedione concentration in the three frailty groups in men with metastatic and localised prostate cancer. (B) Plasma androstenedione concentration and dichotomised frailty (robust/prefrail-frail) in men with metastatic and localised prostate cancer. * means significant differences compared with robust individuals.

Figure 2

(A). Plasma androstenedione concentration and gait speed in men with localised and metastatic prostate cancer. (B) Plasma androstenedione concentration and gait speed in men with metastatic prostate cancer. (C) Plasma androstenedione concentration and gait speed in men with localised prostate cancer. * means significant differences compared with normal weight speed.

Figure 3

Receiver operating characteristic (ROC) curve for androstenedione. The red line represents a hypothetical ROC curve of a perfect classifier. The blue curve represents the actual ROC curve classifier. * indicates the point selected as a cut-off with the best sensitivity and specificity.