Health-Related Quality of Life of Week 8 Responders and Non-Responders: Results from the RBX2660 Phase 3 Randomized, Placebo-Controlled Trial in Recurrent Clostridioides difficile Infection (PUNCH CD3)

Introduction: Recurrent Clostridioides difficile infection (rCDI) substantially compromises patients’ health-related quality of life (HRQL). RBX2660, a live biotherapeutic product (LBP), was found to reduce rCDI. Here we report post-hoc HRQL results of responders (with no recurrence) and non-responders (with a recurrence) within an 8-week blinded period of the RBX2660 phase 3 randomized placebo-controlled trial PUNCH CD3 (NCT03244644). Methods: We analyzed the Clostridioides difficile Health-related Quality-of-Life Questionnaire (Cdiff32), a validated disease-specific instrument with three domains (physical, mental, and social) and a total score (all range from 0–100, 100 best possible). Changes in Cdiff32 from baseline to week 8 were summarized for responders and non-responders, respectively, for RBX2660 and placebo (PBO). Per trial protocol, patients experiencing recurrence after blinded treatment received open-label RBX2660 per physician discretion; these participants were excluded. Adjusted regressions were conducted among responders only due to low sample size of non-responders, and controlling for baseline score, age, sex, number of prior CDI episodes, and other covariates. As-observed data were used. Results: Among patients with available Cdiff32 at baseline and week 8, 178 patients (125 RBX2660, 53 PBO) were responders and 7 (3 RBX2660, 4 PBO) were non-responders. Responders were aged (mean±SD) 59.3±16.7 yrs, 68.5% female; non-responders were aged 66.4±16.4 yrs, 85.7% female. Among responders, improvements from baseline to week 8 were statistically significant (all p< 0.001) for both arms and all Cdiff32 scores (Table). Adjusted analyses among responders found a statistical difference favoring RBX2660 vs PBO at week 8 for the mental domain (7.4, 95% confidence interval: [0.33, 14.43], P< 0.05). Among non-responders, numerical improvements in all four scores were observed for RBX2660, while scores remained similar from baseline to week 8 for PBO. Conclusion: HRQL of rCDI patients treated with RBX2660 and with standard antibiotic treatment (PBO) improved significantly among the responders, with a greater magnitude for RBX2660, particularly for mental health. Among the few non-responders, an average improvement in 10 to 20 points on Cdiff32 was observed for RBX2660-treated patients though not among PBO-treated patients. These findings suggest future research is warranted to further evaluate the potential impact of LBP on HRQL of patients with rCDI. Table 1. - Cdiff32 component scores at baseline and week 8 by treatment arm and response status RBX2660 Placebo Component (Mean ± SD) Baseline Week 8 Change from baseline Unadjusted P-value Baseline Week 8 Change from baseline Unadjusted P-value Responders Total 43.7 ± 17.4 75.8 ± 18.3 32.1 ± 21.4 < 0.001 * 42.4 ± 20.3 70.1 ± 23.2 27.7 ± 19.8 < 0.001 * Physical 52.6 ± 21.1 84.2 ± 16.9 31.7 ± 22.8 < 0.001 * 49.9 ± 22.7 79.2 ± 21.6 29.3 ± 23.2 < 0.001 * Mental 32.7 ± 16.8 66.4 ± 22.0 33.7 ± 24.1 < 0.001 * 33.0 ± 20.5 60.1 ± 27.0 27.0 ± 21.4 < 0.001 * Social 53.5 ± 23.6 80.2 ± 21.3 26.6 ± 27.9 < 0.001 * 49.3 ± 26.5 73.6 ± 27.8 24.3 ± 27.3 < 0.001 * Non-responders Total 61.5 ± 16.7 76.3 ± 5.9 14.8 ± 16.4 0.423 59.0 ± 31.0 57.2 ± 35.0 -1.8 ± 25.6 0.854 Physical 78.0 ± 14.9 91.1 ± 4.7 13.1 ± 19.3 0.423 62.5 ± 35.3 62.1 ± 37.8 -0.5 ± 30.8 0.855 Mental 47.0 ± 19.6 60.7 ± 15.6 13.7 ± 16.2 0.181 54.9 ± 27.7 50.0 ± 33.2 -4.9 ± 23.3 0.854 Social 54.2 ± 15.7 79.2 ± 3.6 25.0 ± 16.5 0.181 60.9 ± 33.2 65.6 ± 34.8 4.7 ± 22.5 1.000