YEARS Algorithm vs CTPA in Active Cancer Suspected PE

Key Takeaways

• The YEARS strategy was associated with noninferior 90-day safety after pulmonary embolism was excluded at baseline.
• CTPA was avoided in 77 of 352 patients, or 22%, assigned to the YEARS pathway.
• Negative baseline CTPA proportions were 83% in both groups, and 3-month all-cause mortality was 21% in each group.

The Hydra randomized clinical trial in JAMA found that the YEARS pathway met noninferiority for 90-day safety in adults with active cancer and suspected pulmonary embolism. After PE was excluded at baseline in the per-protocol analysis, events occurred in 1.8% of patients managed with YEARS and 5.5% of those managed with a computed tomographic pulmonary angiography-only strategy. CTPA use was also lower with YEARS in this cancer population.

This open-label, investigator-initiated, randomized noninferiority trial used blinded central outcome adjudication across 21 hospitals in the Netherlands, Italy, Switzerland, Belgium, France, and Spain. Recruitment ran from August 22, 2019, to May 23, 2025, with final follow-up on August 21, 2025, among adults aged 18 years or older with suspected acute PE and active cancer. Active cancer meant histologically confirmed or strongly suspected cancer within 6 months, ongoing treatment, treatment within 6 months before randomization, or metastatic disease. Among 698 randomized patients, 352 were assigned to YEARS and 346 to CTPA-only management. In the YEARS arm, 3 clinical items guided D-dimer thresholds below 1000 ng/mL with no items or below 500 ng/mL with 1 or more items, with CTPA otherwise, so the trial compared diagnostic pathways rather than treatment strategies.

The primary outcome was centrally adjudicated symptomatic PE, symptomatic upper- or lower-extremity DVT, or possible PE-related death within 90 days after PE was excluded at baseline. The primary per-protocol analysis of the YEARS algorithm used a prespecified 1-sided 99.9% confidence interval framework, with the upper-bound margin adjusted to 2.6% after interim analysis. Five of 282 patients in YEARS and 15 of 273 in CTPA-only met the primary outcome, for an absolute risk difference of -3.7%. The 99.9% confidence interval was -8.8% to 1.4%, and P was 3.4 × 10^-5 for noninferiority. In the intention-to-diagnosis analysis, the absolute risk difference was -2.6%, the 99.9% confidence interval was -7.5% to 2.4%, and P was 5.9 × 10^-4, meeting noninferiority in both analyses.

CTPA was avoided in 77 of 352 patients in the YEARS arm, although most patients in that group still underwent imaging. The proportion of negative baseline CTPAs was 83% in both groups, with P = .93, and the study was not powered to detect differences in diagnostic yield. Three-month all-cause mortality was 21% in each group, and PE was diagnosed at baseline in 104 of 698 randomized patients. Incidental events included 1 PE in YEARS and 1 PE plus 2 DVTs in CTPA-only. One patient in YEARS was lost to follow-up, while none were lost in CTPA-only. Lower CTPA use was paired with similar negative-scan proportions in this population, while applicability may be lower with life expectancy under 3 months and without cancer-site or stage subgroup analyses.