Comparative efficacy and safety of monotherapy and combination pharmacotherapies for idiopathic pulmonary fibrosis: a network meta-analysis of randomized controlled trials

BACKGROUND: Idiopathic pulmonary fibrosis (IPF) is a fatal, progressive fibrosing lung disease. While novel agents and combinations are emerging, head-to-head trials are scarce, underscoring the need for robust comparative evidence to guide treatment prioritization. METHODS: We conducted a systematic review and network meta-analysis (NMA) of IPF RCTs in adults, searching PubMed, Embase, Cochrane Library, and Web of Science through 1 December 2025. Eligible studies compared pharmacotherapies (mono/combination) with placebo or active comparators and reported prespecified outcomes. Primary outcomes were forced vital capacity (FVC) change and disease progression; secondary outcomes included all-cause mortality, adverse events (AEs), serious AEs (SAEs), and DLCO change. Risk of bias was assessed via RoB 2, with frequentist random-effects NMA for continuous/binary outcomes and a random-effects model for disease progression. Treatment rankings used SUCRA, and evidence certainty was evaluated via GRADE with CINeMA. RESULTS: Thirty-five reports (8,983 participants, 21 strategies) were included. RoB 2 ratings were low (19 studies) or raised some concerns (16). For FVC preservation, recombinant human pentraxin-2 (RHP) (SMD = 0.72, 95% CI: 0.23-1.20), rentosertib (SMD = 0.63, 95% CI:0.04-1.22), and nintedanib (SMD = 0.50, 95% CI:0.27-0.73) showed the highest statistical probability of benefit versus placebo, though these rankings are based primarily on low-certainty indirect comparisons. For disease progression, pirfenidone, pamrevlumab, and nerandomilast showed favorable but imprecise effects. No regimen significantly reduced all-cause mortality or showed clear AE/SAE differences versus placebo, with wide uncertainty in comparisons. Evidence certainty was mostly low to very low, driven by imprecision and indirect comparisons in star-shaped networks. CONCLUSIONS: RHP, rentosertib, and nintedanib showed consistent signals for preserving FVC versus placebo. However, comparative effects on progression, mortality, and safety remain uncertain due to sparse data and limited head-to-head evidence. Large, well-designed trials with harmonized endpoints and longer follow-up are needed to validate these exploratory rankings and define optimal treatment sequencing and combinations. REGISTRATION: PROSPERO CRD420261289653.