MON-699 - The Growth Hormone Secretagogue LUM-201 Induces Insulin Sensitivity Index-Dependent Reorganisation of GH Chronobiology, Driven by Higher-Order, Multi-Timepoint Interactions

Background
The growth hormone (GH) secretagogue LUM-201 restores pulsatile GH secretion and increases fat-free mass in older adults1, but aggregate endpoints obscure how diurnal GH time series architecture reconfigures under treatment. This includes changes to the Matsuda insulin sensitivity index (ISI)2.
Objective
To determine how LUM-201 treatment alters 24-hour GH secretion dynamics after 6 months and whether these changes differ by insulin sensitivity.
Methods
Participants (43 healthy adults [60 – 81 years of age], 27 females & 16 males) underwent GH profiling (10min samples starting at 08.00 for 24hr) and were grouped into Low, Medium, and High ISI tertiles based on ISI. Functional principal component analysis (fPCA) on GH-secretion trajectories yielded eigenfunctions capturing amplitude-phase structure. Hypergraphs were used to measure higher order interactions between time points; nodes are sampling time points (144) and hyperedges encoded multi-point co-activation (subjects with GH secretion > median). Random-walk analyses (quantifying the hypergraph) combine diffusion and centrality measures; mean curves were split into four phases: Phase 1 (early morning), Phase 2 (late-morning), Phase 3 (afternoon), Phase 4 (evening and night); above baseline defined as positive, below baseline as negative in response to 6-months treatment. Non-parametric tests (Wilcoxon within tertiles; Kruskal–Wallis across tertiles) assessed transition behaviour.
Results
ISI in the low tertile decreased over 6 months (p=0.001), in the medium tertile was unchanged and in the high tertile increased (p=0.003). GH concentrations increased by a similar amount in each ISI tertiles (p < 0.003). fPCA after 6-months treatment revealed leading eigenfunctions in Low ISI had greater impact across the profile in amplitude and phase compared to medium and high tertiles. Hypergraph analysis showed higher-order, time-dependent interactions intensified under LUM-201, within all ISI tertiles. Random-walk centrality was lower in the evening and night phase in all tertiles most prominently in the high ISI tertile (p-value [p < 1x10-10] rank - high < medium < low) indicating larger, connected GH-secretion bursts. A different pattern was found in late-morning secretion, with greater higher-order impact (p-value [p < 0.05] rank - medium << high < low).
Conclusions
Integrating time series analytics reveals LUM-201 provokes an ISI-dependent change in GH chronobiology, governed by multi-timepoint interaction patterns. Lower ISI associates with greater reconfiguration and diffusion of burst structure across 24 hours. These insights have implications for understanding the interactions between GH and metabolism.
1. Nass, R. et al. Ann Intern Med 149, 601–611 (2008).
2. Matsuda, M. & DeFronzo, R. A. Diabetes Care 22, 1462–1470 (1999).

*Unless otherwise noted, all abstracts presented at ENDO must not be released to the press or the public until the date and time of presentation. For oral presentations, the abstracts are embargoed until the session begins. The Endocrine Society reserves the right to lift the embargo on specific abstracts that are selected for promotion prior to or during ENDO.*