This analysis used a decision-analytic model to compare sulbactam-durlobactam with colistin for the treatment of carbapenem-resistant Acinetobacter baumannii (CRAB) infections in critical care settings in the United States. Clinical inputs derived from the Acinetobacter Treatment Trial Against Colistin (ATTACK) trial informed survival and clinical cure outcomes. The model explicitly incorporated the incidence and downstream costs of acute kidney injury (AKI), a safety outcome more commonly observed with colistin, and evaluated economic and clinical outcomes from a per-patient perspective.
Key clinical inputs were survival probabilities and clinical cure status (complete or incomplete), taken from the ATTACK trial where patients received either sulbactam-durlobactam or colistin alongside imipenem-cilastatin. The model included the higher frequency of acute kidney injury (AKI) associated with colistin and accounted for the costs of managing AKI events. Cost inputs included the treatment acquisition and associated care costs; specific unit costs and model parameter values were reported in the manuscript.
Deterministic sensitivity analysis was planned to examine how changes in individual parameters would affect incremental cost-effectiveness. The authors also ran two scenario analyses to test alternative assumptions around clinical cure and other drivers of cost-effectiveness.
In the base-case analysis the model estimated the expected per-patient cost for sulbactam-durlobactam at $57,887 compared with $55,157 for colistin. Despite higher treatment costs, patients treated with sulbactam-durlobactam demonstrated greater survival (0.66) than those treated with colistin (0.46), corresponding to an additional 0.19 life saved per patient.
The incremental cost-effectiveness ratio (ICER) for sulbactam-durlobactam versus colistin in the base case was $14,097 per additional life saved. This ICER reflects the incremental cost divided by the incremental survival benefit observed in the modeled cohort when AKI costs are included.
Deterministic sensitivity analysis identified several parameters with the greatest influence on the ICER. The ICER was most sensitive to:
Scenario analyses demonstrated that assumptions about clinical cure outcomes materially affected cost-effectiveness. Across different scenarios the ICER ranged up to $69,758 per additional life saved. Under certain scenario assumptions (noted in the manuscript), sulbactam-durlobactam was dominant, meaning it provided greater effectiveness at lower or equivalent overall cost when accounting for AKI management and other modeled factors.
The authors conclude that, although sulbactam-durlobactam incurs higher upfront treatment costs compared with colistin, its improved survival outcomes and the higher rate and costs of AKI with colistin result in a favorable cost-effectiveness profile when evaluated per additional life saved. The base-case ICER of $14,097 per additional life saved indicates a relatively modest incremental cost for the observed survival benefit in the modeled critical care population.
The sensitivity of results to clinical cure rates and drug costs underscores the importance of real-world effectiveness, pricing, and management of adverse events in economic assessments of therapies for CRAB infections. Decision-makers in critical care and antimicrobial stewardship programs should consider both clinical and renal safety outcomes when comparing these treatment options.
The authors report that all data produced for the analysis are contained within the manuscript. Competing interest disclosures listed that three authors (AO, SC, SK) declared no conflicts of interest, while one author (ARC) reported research funding from Pfizer unrelated to this study.
Note: This report summarizes findings presented in the medRxiv preprint. Specific parameter values, detailed cost components, and full scenario definitions were reported in the manuscript and are the source of the figures and ranges summarized here. Any additional granular model inputs or subgroup results not presented in the preprint are not available in the source document.