Prospective study of anamorelin in pancreatic cancer cachexia: Clinical and translational insights into response heterogeneity.

BACKGROUND & AIMS: Clinical evidence for anamorelin in pancreatic cancer is extremely limited, despite its approval in Japan. This study provides the first prospective evaluation of anamorelin specifically in pancreatic cancer, aiming to assess real-world efficacy and safety and to identify factors contributing to treatment-response heterogeneity through integrated biomarker analyses.

METHODS

This prospective, single-center, observational study enrolled 24 patients with unresectable or metastatic pancreatic cancer who developed cachexia. Efficacy was evaluated in patients who received anamorelin for >1 month. The primary endpoint was change in LBM from baseline, and secondary endpoints included the LBM-based response rate. Responders were defined as those who maintained or increased LBM during treatment. Safety assessment focused on treatment-related adverse events, particularly hyperglycemia.

RESULTS

Seventeen patients were included in the efficacy analysis (median age, 68 years; median weight loss, 8.3 %). The mean LBM increased by 0.9 and 1.4 kg at 1 and 2 months, respectively. Quality-of-life scores related to appetite, weight gain, and total scores improved significantly at 1 month. Ten patients (58.8 %) were classified as responders, showing significant LBM gains from baseline (+2.4 kg at 1 month, +3.4 kg at 2 months, p < 0.001). Handgrip strength also improved in responders compared with non-responders at 2 months (+1.7 kg vs -2.0 kg, p < 0.01). Serum levels of insulin-like growth factor-1, inflammatory cytokines, ghrelin, and leptin levels did not differ significantly between baseline and 1 month. However, lower baseline body mass index (BMI) was strongly associated with response (sensitivity 85.7 %, specificity 90.0 %, area under the curve [AUC] 0.886, cutoff 20.4 kg/m2; p = 0.008). In the safety analysis (n = 23), 34.8 % experienced hyperglycemia of any grade, and 26.1 % developed grade ≥2 hyperglycemia-higher than in NSCLC trials. Median time to onset was 4.5 days (range, 2-18). Baseline diabetes was significantly associated with grade ≥2 hyperglycemia. This event was highly predictable by low pre-treatment ΔC-peptide levels (6-0 min; sensitivity 100.0 %, specificity 91.7 %, cut-off 1.03 ng/mL, AUC 0.967; p = 0.0032).

CONCLUSIONS

Anamorelin effectively improved LBM and appetite/QOL domains in pancreatic cancer, particularly in patients with low BMI. However, hyperglycemia-especially in those with impaired insulin secretion-requires careful monitoring. Baseline BMI and insulin secretion capacity should be evaluated before initiating therapy, and these findings provide preliminary insight into treatment response heterogeneity in pancreatic cancer cachexia.