The paradoxical GH response at OGTT does not predict Pasireotide efficacy but matters for glucose metabolism

Purpose: A paradoxical increase in GH after oral glucose load (GH-Par) characterizes about one-third of acromegaly patients and is associated with a better response to first-generation somatostatin receptor ligands (fg-SRLs). Pasireotide is typically considered as a second-/third-line treatment. Here, we investigated the predictive role of GH-Par in pasireotide response and adverse event development. Methods: we collected a multicenter Italian retrospective cohort of 59 patients treated with pasireotide for at least 3 months, all having GH profile from OGTT. IGF-1 normalization or at least 30% reduction at the last follow-up visit defined a responder patient. Results: Considering the entire cohort, median IGF-1 levels before pasireotide (available in 57 patients) were 1.38 times the upper limit of normal (ULN) in patients with large (median size 18 mm) and invasive (82%) adenomas after failure of fg-SRL treatment. After a 40-month median treatment, pasireotide effectively reduced IGF-1 ULN levels in 41 patients, 37 of whom achieving normalization, and 4 with a ≥ 30% reduction. Thirteen patients were classified as GH-Par. The median pasireotide duration, dosage, and efficacy (9/12 responder in the GH-Par group and 32/45 in the GH-NPar) were similar between groups. However, the occurrence of new-onset or worsening glucose metabolism alterations (GMAs) after pasireotide was more frequent in GH-NPar (from 37 to 80%; p < 0.001) compared to GH-Par patients (from 69 to 76%), likely due to the higher prevalence of pre-existing GMAs in the GH-Par group before starting pasireotide (p = 0.038). Conclusions: The GH-Par does not predict the response to pasireotide in acromegaly but can predict a worse metabolic profile.