Acute cellular rejection after heart transplantation: validation of HEARTBiT, a transcriptomic biomarker.

BACKGROUND AND AIMS: Acute cellular rejection (ACR) risk is highest within 60 days of heart transplantation (HTx), yet existing minimally invasive surveillance tools lack validation for this high-risk period. This study prospectively validated the diagnostic accuracy of HEARTBiT, a transcriptomic biomarker for detecting moderate-to-severe ACR, including during the early post-operative period. METHODS: This prospective diagnostic accuracy study (NCT03575910) enrolled consecutive adult HTx recipients at four tertiary clinical centres. The diagnostic performance of HEARTBiT was evaluated in 336 visits from 117 participants against centralized expert consensus panel endomyocardial biopsy grades. An exploratory analysis of 246 serial visits from 31 participants evaluated the prognostic potential of tracking individual transcriptomic trajectories over time. RESULTS: HEARTBiT accurately detected moderate-to-severe ACR (ISHLT ≥2R) with an overall area under the curve (AUC) of 0.78 [95% confidence interval (CI) 0.70-0.86]. Performance was stable early post-HTx (≤60 days: AUC = 0.78, 95% CI 0.65-0.90) and beyond 60 days (AUC = 0.80, 95% CI 0.72-0.87). Prioritizing high sensitivity (89% early, 100% late; corresponding to ≤1 missed ACR overall) yielded a specificity of 56% in the early period (≤60 days) and 49% beyond 60 days using the pre-specified cutoff (>0.54). Alternatively, adopting an exploratory time-varying threshold (>0.68) beyond 60 days yielded 83% sensitivity and 76% specificity. Importantly, HEARTBiT scores increased weeks before histological detection of ACR by tissue biopsy (slope-based AUC = 0.78, 95% CI 0.60-0.96). CONCLUSIONS: HEARTBiT provides a reliable, minimally invasive rule-out signal for moderate-to-severe ACR across the first year post-HTx, particularly during the high-risk early post-operative period. Serial tracking of HEARTBiT score trajectories has potential for prognosis of progression to ACR. These findings support the clinical utility of this biomarker in safely sparing more than 50% of protocol surveillance endomyocardial biopsies.