Document resource
Introduction The ABRA trial was a multicentre, phase 2, double-blind, double-dummy, placebo-controlled clinical trial. Adults with asthma or COPD exacerbations and blood eosinophils ≥300 cells/µL were randomised 1:1:1 to: (1) benralizumab + placebo (BENRA), (2) placebo + prednisolone (PRED), or (3) benralizumab + prednisolone (BENRA+PRED). Benralizumab was superior to prednisolone (standard of care) at reducing treatment failures, but treatment failures were still 45% at 90 days (compared to 74% with prednisolone).Aim Describe the characteristics of patients who had a treatment failure following an eosinophilic exacerbation.Results The characteristics of the patients who did and did not have any treatment failures are shown in table 1.In the multivariable logistic regression model, steady-state% predicted FEV1 was a statistically significant predictor of treatment failure (OR = 0.971, 95% CI: 0.945–0.999, p=0.042), indicating that higher baseline lung function was associated with reduced odds of repeat exacerbation. Additionally, treatment with prednisolone alone was associated with a more than threefold increase in the odds of treatment failure compared to benralizumab or benralizumab plus prednisolone (OR=3.16, 95% CI: 1.38–7.25, p=0.005). Other variables, including age, smoking status and eosinophil count and FeNO at time of exacerbation, were not significantly associated with treatment outcome.Abstract M16 Table 1Patient characteristics for those patients who had treatment failure vs those who did not in the ABRA trial Did have a treatment failure (n=87) Did not have a treatment failure (n=71) Women, n (%) 49 (55.1) 37 (52.1) Age, mean (SD) 58 (13.84) 56.3 (16.85) Asthma exacerbation, n (%) 48 (53.9) 40 (56.3) COPD exacerbation, n (%) 39 (43.8) 31 (43.7) Smoker, n (%)Ex Smoker, n (%) 11 (12.4)44 (49.4) 4 (5.6)36 (50.7) FEV1% predicted at steady state, mean (SD) 68.38 (23.29) 75.9 (24.2) Blood eosinophils at exacerbation, mean (SD) 0.7 (0.41) 0.68 (0.42) Fraction exhaled nitric oxide at exacerbation, mean (SD) 54.67 (46.86) 59.5 (43.91) Benralizumab alone treatment, n (%) 25 (28.1) 28 (39.4) Benralizumab and prednisolone treatment, n (%) 23 (25.8) 29 (40.8) Prednisolone treatment, n (%) 39 (43.8) 14 (19.7) Conclusion Reduced baseline lung function and monotherapy with prednisolone were key predictors of treatment failure in the ABRA study.