Surveillance with dual noninvasive testing for acute cellular rejection after heart transplantation: Outcomes from the Surveillance HeartCare Outcomes Registry.

BACKGROUND

Molecular testing with gene-expression profiling (GEP) and donor-derived cell-free DNA (dd-cfDNA) is increasingly used in the surveillance for acute cellular rejection (ACR) after heart transplant. However, the performance of dual testing over each test individually has not been established. Further, the impact of dual noninvasive surveillance on clinical decision-making has not been widely investigated.

METHODS

We evaluated 2,077 subjects from the Surveillance HeartCare Outcomes Registry registry who were enrolled between 2018 and 2021 and had verified biopsy data and were categorized as dual negative, GEP positive/dd-cfDNA negative, GEP negative/dd-cfDNA positive, or dual positive. The incidence of ACR and follow-up testing rates for each group were evaluated. Positive likelihood ratios (LRs+) were calculated, and biopsy rates over time were analyzed.

RESULTS

The incidence of ACR was 1.5% for dual negative, 1.9% for GEP positive/dd-cfDNA negative, 4.3% for GEP negative/dd-cfDNA positive, and 9.2% for dual-positive groups. Follow-up biopsies were performed after 8.8% for dual negative, 14.2% for GEP positive/dd-cfDNA negative, 22.8% for GEP negative/dd-cfDNA positive, and 35.4% for dual-positive results. The LR+ for ACR was 1.37, 2.91, and 3.90 for GEP positive, dd-cfDNA positive, and dual-positive testing, respectively. From 2018 to 2021, biopsies performed between 2 and 12-months post-transplant declined from 5.9 to 5.3 biopsies/patient, and second-year biopsy rates declined from 1.5 to 0.9 biopsies/patient. At 2 years, survival was 94.9%, and only 2.7% had graft dysfunction.

CONCLUSIONS

Dual molecular testing demonstrated improved performance for ACR surveillance compared to single molecular testing. The use of dual noninvasive testing was associated with lower biopsy rates over time, excellent survival, and low incidence of graft dysfunction.