Free alpha-subunit is superior to luteinizing hormone as a marker of gonadotropin-releasing hormone despite desensitization at fast pulse frequencies.

A pulsatile pattern of GnRH stimulation is essential for normal secretion of luteinizing hormone (LH), while both continuous and fast-frequency GnRH stimulation result in a paradoxical decrease in gonadotrope responsiveness known as desensitization. Under physiological conditions there is striking concordance between the pulsatile secretion of LH and the glycoprotein free alpha-subunit (FAS). The aims of this study were to determine whether the FAS response to GnRH is also decreased at fast frequencies of GnRH stimulation and whether FAS is superior to LH as a marker of GnRH secretory activity at fast-pulse frequencies. The model of GnRH-deficient men was chosen to permit precise control of the dose and frequency of GnRH stimulation of the gonadotrope. The frequency of i.v. administration of GnRH to 5 GnRH-deficient men was progressively increased from every 120 to every 60 min, from 60 to 30 min, and from 30 to 15 min during three 12-h admissions, 1 week apart. The bolus dose of GnRH remained constant and was set at that dose previously shown to produce physiological concentrations and amplitudes of LH secretion and normal testosterone levels. As the frequency of GnRH stimulation was increased, a progressive rise in mean FAS levels was noted (353 +/- 13, 448 +/- 42, 466 +/- 50, and 698 +/- 85 ng/L [mean +/- SEM] for 120, 60, 30, and 15 min intervals; P < 0.005). However, normalization of mean FAS levels to account for the increase in total GnRH delivered with increasing frequencies revealed a progressive decrease in pituitary responsiveness to each GnRH bolus with increasing frequency of stimulation (353 +/- 13, 224 +/- 21, 117 +/- 13, 87 +/- 11 ng/L; P < 0.001). The decrease in normalized mean levels was supported by a decrease in the FAS pulse amplitude with increasing frequency (517 +/- 53, 365 +/- 50, 176 +/- 29 ng/L for 120, 60, and 30 min intervals, respectively; P < 0.005). At interpulse intervals of 120 and 60 min, there was complete concordance of LH and FAS pulses in response to GnRH. However, at the 30-min frequency FAS proved to be a better marker of GnRH with a higher true positive rate and lower number of false positives than LH (P < 0.05). At all frequencies, the number of false positive pulses detected tended to be lower for FAS than for LH (P = 0.06). From these data we conclude that FAS is subject to desensitization in response to increasing frequencies of GnRH administration in GnRH-deficient men, but is superior to LH as a surrogate marker of GnRH pulse generator activity at fast pulse frequencies.