Systematic evaluation of levodopa-carbidopa intestinal gel patient-responder characteristics

Abstract

Levodopa-carbidopa intestinal gel (LCIG, carbidopa-levodopa enteral suspension in the United States) is a treatment option for advanced Parkinson's disease (PD) patients with motor fluctuations. The objective of this investigation was to identify the baseline characteristics predictive of treatment response, measured by improvement in motor symptom severity, in advanced PD patients treated with LCIG during a 54-week, open-label phase 3 study. Patients with ≥1 h improvement from baseline in "Off" time were categorized as "Responders"; whereas those with <1 h improvement, any worsening, or no post-baseline assessment were "Non-Responders". A subgroup of Responders with ≥3 h improvement in "Off" time was also examined; this subgroup was identified as "Robust Responders". Baseline demographics and disease characteristics were analyzed and their predictive relationship to change from baseline in normalized "Off" time was assessed. Out of the 324 patients included in the analysis, 272 (84.0%) were categorized as Responders and 52 (16.0%) were Non-Responders. A majority of patients (65.7%) had ≥3 h improvement in "Off" time. In general, baseline characteristics were similar between Non-responders, Responders, and the subgroup of Robust Responders. A conditional tree-structured regression analysis identified baseline "Off" time as the only factor that had significant effect on Responder and Robust Responder status. The safety profile of LCIG was similar between patient groups. Overall, this analysis showed that 84% of LCIG-treated advanced PD patients had ≥1 h improvement in "Off" time and the number-needed-to-treat to observe one patient responder was 1.19 patients. Notably, Responders and Robust Responders to LCIG were observed across the range of baseline demographics and clinical characteristics examined.

Fig. 1

Responder analysis patient disposition. ^aPatients with a baseline and at least one post-baseline PD diary assessment during the PEG-J treatment phase were included in these analyses

Fig. 2

Baseline “Off” time and “Off” time improvement. a Percentage of patients with binned hours of reduction in “Off” time from baseline to last visit. b Percentage of Responders and Non-Responders by hours of baseline “Off” time. ^a Percentage = 37.5%. LV = last visit

Fig. 3

Conditional tree-structured regression analysis of responder rate. a Predictive baseline characteristics of Responders. b Predictive baseline characteristics of Robust Responders. ^aSix patients who did not have baseline “Off” time values could not be included