Toshinaga Tsuji, Naohiro Itoh, Mitsuhiro Ishida, Toshimitsu Ochiai, Shinichi Konno

Duloxetine is efficacious for chronic low back pain (CLBP). This post hoc analysis of a Japanese randomized, placebo-controlled trial (ClinicalTrials.gov, NCT01855919) assessed whether patients with CLBP with early pain reduction or treatment-related adverse events of special interest (TR-AESIs; nausea, somnolence, constipation) have enhanced responses to duloxetine. Patients (N = 456) with CLBP for ≥6 months and Brief Pain Inventory (BPI) average pain severity score of ≥4 were randomized (1:1) to duloxetine 60 mg/day or placebo for 14 weeks. Primary outcome was change from baseline in BPI average pain severity score (pain reduction). Subgroup analyses included early pain reduction (≥30%, 10%-30%, or <10% at Week 4) and early TR-AESIs (with or without TR-AESIs by Week 2). Measures included changes from baseline in BPI average pain severity score and BPI Interference scores (quality of life; QOL), and response rate (≥30% or ≥50% pain reduction at Week 14). Patients with ≥30% early pain reduction (n = 108) or early TR-AESIs (n = 50) had significantly greater improvements in pain and QOL than placebo-treated patients (n = 226), whereas patients with 10%-30% (n = 63) or <10% (n = 48) pain reduction did not; patients without early TR-AESIs (n = 180) had significant improvements in pain at Week 14. Response rates (≥30%/≥50% pain reduction) were 94.4%/82.4%, 66.7%/49.2%, and 25.0%/18.8% for patients with ≥30%, 10%-30%, and <10% early pain reduction, respectively, 74.0%/64.0% for patients with early TR-AESIs, 67.2%/54.4% for patients without early TR-AESIs, and 52.2%/39.4% for placebo. Early pain reduction or TR-AESIs may predict which CLBP patients are most likely to respond to duloxetine with improvements in pain and QOL.