Background
Hydra study. Open-label, randomised, investigator-initiated noninferiority trial with blinded central adjudication. N=698 patients with active cancer and suspected acute pulmonary embolism (PE), from emergency departments/medical units at 21 hospitals in 6 European countries. Compares the YEARS diagnostic algorithm versus CT pulmonary angiography (CTPA) only to rule out PE.
Results
Interventions and follow up: Arm A: YEARS algorithm — YEARS items + D-dimer + risk-dependent CTPA (n=352)
Arm B: CTPA only (n=346)
Primary endpoint: centrally adjudicated symptomatic VTE or (possible) PE-related death within 90 days after PE ruled out (per-protocol; noninferiority margin 2.6%)
Key secondary: proportion of negative baseline CTPA
mFollow up: 90 days
Results: Primary (per-protocol): 1.8% (5/282) YEARS vs 5.5% (15/273) CTPA-only; risk difference −3.7%, 99.9% CI −8.8 to 1.4, P<.001 for noninferiority.
Intention-to-diagnose: risk difference −2.6%, 99.9% CI −7.5 to 2.4, P<.001 for noninferiority.
CTPA avoided: 22% (77/352) of YEARS patients.
Negative CTPA proportion: no difference (P=.93). PE at baseline in 15% (104/698).
Arm B: CTPA only (n=346)
Primary endpoint: centrally adjudicated symptomatic VTE or (possible) PE-related death within 90 days after PE ruled out (per-protocol; noninferiority margin 2.6%)
Key secondary: proportion of negative baseline CTPA
mFollow up: 90 days
Results: Primary (per-protocol): 1.8% (5/282) YEARS vs 5.5% (15/273) CTPA-only; risk difference −3.7%, 99.9% CI −8.8 to 1.4, P<.001 for noninferiority.
Intention-to-diagnose: risk difference −2.6%, 99.9% CI −7.5 to 2.4, P<.001 for noninferiority.
CTPA avoided: 22% (77/352) of YEARS patients.
Negative CTPA proportion: no difference (P=.93). PE at baseline in 15% (104/698).
Adverse events
90-day symptomatic VTE / PE-related death after negative workup: 1.8% YEARS vs 5.5% CTPA-only (numerically fewer with YEARS).
Imaging burden: CTPA and iodinated contrast avoided in ~1 in 5 YEARS patients.
No safety signal against the YEARS pathway.
Imaging burden: CTPA and iodinated contrast avoided in ~1 in 5 YEARS patients.
No safety signal against the YEARS pathway.
Conclusions
In patients with active cancer and suspected PE, a D-dimer-based YEARS algorithm was noninferior to — and directionally safer than — proceeding directly to CTPA, while avoiding CTPA in about one in five patients. Supports YEARS-based workup rather than automatic CTPA.
Key Limitations
Open-label; noninferiority design with wide confidence intervals; per-protocol primary analysis; the numerically lower event rate with YEARS is unexpected and hypothesis-generating; generalizability across cancer types, thrombocytopenia, and different D-dimer assays uncertain.
Clinical Context
Some prior guidance suggested proceeding directly to CTPA in cancer patients given D-dimer nonspecificity. Hydra provides the first randomised evidence that a D-dimer-based YEARS pathway is safe in active cancer, reducing imaging, contrast exposure, and cost. Builds on the original YEARS study (2017).