Phase 2b, randomized, double-blind 12-week studies of TZP-102, a ghrelin receptor agonist for diabetic gastroparesis

Background: TZP-102, a potent, oral, ghrelin receptor agonist, improved diabetic gastroparesis symptoms in Phase 2a.

Methods: Patients with type 1 or 2 diabetes, delayed gastric half-emptying (T(1/2)), and ≥3 months gastroparesis symptoms randomized 1 : 1 : 1 to double-blind placebo, 10-mg, or 20-mg TZP-102 once daily for 12 weeks (Study TZP-102-CL-G003). Study TZP-102-CL-G004 patients randomized 1 : 1 to 10-mg TZP-102:placebo three-times-daily. Primary endpoint was change-from-baseline through Weeks 11-12 in Daily Diary of Gastroparesis Symptoms Questionnaire (GSDD) via electronic Patient Recorded Outcome device: worst severity of nausea, early satiety, bloating, and upper abdominal pain in 24 h (0 = none-to-5 = very severe). GSDD Composite Score for eligibility was ≥2.5 (Day-14-to-baseline). Patient Overall Treatment Evaluation (OTE) provided an anchor-based minimal clinically important difference (MCID) for GSDD Composite Score.

Key results: Study TZP-102-CL-G003 enrolled 201 outpatients: females 72%; Caucasians 87%; type 2 diabetes 61%; insulin-dependent 65%; age mean ± SD 53 ± 11.3 years; HbA1c 7.8 ± 1.5%; GCSI 3.4 ± 0.7; GSDD Composite 3.6 ± 0.6; gastric T1/2 131 ± 32 min; n = 69 (10-mg), n = 66 (20-mg), n = 66 (placebo). Primary endpoint (GSDD): significant improvement in all arms, although not for TZP-102 vs placebo: mean change-from-baseline -1.7, -1.4, -1.5 (10-mg, 20-mg, placebo); Gastroparesis Cardinal Symptom Index -1.8, -1.6, -1.5, respectively. The OTE (all patients) at Week-12 was: Patient 3.7 ± 3.2 and Physician 3.6 ± 3.0 with median score for both of 5.0 = important on scale of improvement; individual MCID was 1.61 and 0.94 for group analyses, greater than expected. Study TZP-102-CL-G004 with similar demographic/disease characteristics was prematurely terminated for efficacy futility (n = 64 with Week-4 assessments).

Conclusions & inferences: Efficacy of TZP-102 was not demonstrated compared with placebo in diabetic gastroparesis; however, there was substantial symptom improvement in all arms (ClinicalTrials.gov NCT01452815/NCT01664637).