A randomized controlled clinical trial of growth hormone in amyotrophic lateral sclerosis: clinical, neuroimaging, and hormonal results.

ABSTRACT

Amyotrophic lateral sclerosis (ALS) is a fatal neurological disease with motor neuron degeneration. Riluzole is the only available treatment. Two-thirds of ALS patients present with growth hormone (GH) deficiency. The aim of this study is to determine if add-on of GH to riluzole, with an individually regulated dose based on Insulin-like growth factor 1 (IGF-I) production, was able to reduce neuronal loss in the motor cortex, reduce mortality, and improve motor function of ALS patients. Patients with definite/probable ALS, in treatment with riluzole, aged 40-85 years, and with disease duration ≤3 years were enrolled. The study was randomized, placebo controlled, and double blind. Before treatment, patients were tested with a GH releasing hormone (GHRH) + arginine test. The initial dose of GH was 2 IU s.c. every other day, and was progressively increased to a maximum of 8 IU. Primary endpoint was N-acetylaspartate/(creatine + choline) (NAA/Cre + Cho) ratio in motor cortex assessed by magnetic resonance spectroscopy performed at months 0, 6, and 12. Secondary endpoints were mortality and ALS functional rating scale revised (ALSFRS-R). The NAA/(Cre + Cho) ratio decreased in all patients who completed the trial. No significant difference was noted between treated and placebo group. At baseline, although IGF-I levels were within the normal range, 73% of patients had GH deficiency, being severe in half of them. Compared with bulbar onset, spinal-onset patients showed more depressed GH response to the GHRH + arginine stimulation test (10.4 ± 7.0 versus 15.5 ± 8.1 ng/mL; p < 0.05). Insulin resistance [homeostasis model assessment of insulin resistance (HOMA-IR)] increased from 2.1 ± 1.0 at baseline to 4.6 ± 1.9 at 12 months (p < 0.001). Insulin-like growth factor (IGF) binding protein 3 (IGFBP-3) decreased from 8,435 ± 4,477 ng/mL at baseline to 3,250 ± 1,780 ng/mL at 12 months (p < 0.001). The results show that GH exerted no effect on cerebral NAA or clinical progression assessed by ALSFRS-R. Two-thirds of ALS patients had GH deficit, with higher levels in the bulbar-onset group. During follow-up, patients showed progressive increase in HOMA-IR and decrease in IGFBP-3 levels.