The first randomized trial of its kind concludes that an algorithm widely used to rule out pulmonary embolism (PE) can safely be extended to patients with cancer.

As it stands, patients with cancer and suspected PE usually go straight to CT pulmonary angiography (CTPA) for evaluation. But in the new trial, the so-called YEARS algorithm worked just as well as immediate CTPA in ruling out PE — and it allowed 22% of patients to skip the scans.

“We have undoubtedly demonstrated in a randomized clinical trial the use of the YEARS algorithm is as safe as using CTPA directly,” said senior researcher Menno Huisman, MD, of Leiden University in Leiden, Netherlands.

The findings, published online in the Journal of the American Medical Association, highlight one way to limit unnecessary CT scans. But experts not involved in the study had caveats — notably around using the algorithm in patients whose cancers place them at especially high risk for PE.

Until more data are available, they believe first-line use of CTPA is still warranted in those cases.

Nearly a decade ago, Huisman and his colleagues published the original YEARS study, establishing the algorithm as an effective tool for ruling out PE in the general patient population. The algorithm uses three clinical questions along with a D-dimer blood test: If the D-dimer level is below a certain threshold (determined by the clinical questions), PE can be safely ruled out; if exceeds the threshold, CTPA is used.

There has been hesitancy, however, to use YEARS in evaluating patients with cancer, who have a much higher-than-average risk for blood clots. Standard practice is to send those with suspected PE straight to CT.

However, given the cost, inconvenience and radiation exposure associated with CTPA, Huisman’s team wanted to put YEARS to the test specifically in patients with cancer.

Their trial involved 698 patients (median age, 65 years) with active cancer and suspected PE, recruited from 21 hospitals throughout Europe. Half were sent to CTPA only, while the other half were first assessed by the YEARS protocol.

Among all patients in which PE was excluded, 1.8% of those screened with YEARS went on to experience a venous thromboembolism or possible PE-linked death within 90 days. That compared with 5.5% of those evaluated by CTPA alone.

There was an absolute risk difference between YEARS and CTPA of -3.7%, well within the noninferiority margin set by the researchers of 2.6%.

Overall, 22% of patients screened via YEARS were able to skip CTPA. That’s below the 48% reported in the original YEARS study but similar to the rate observed in a recent meta-analysis involving patients with cancer and suspected PE.

Extrapolating to the US, Huisman said, using the YEARS algorithm in patients with cancer with suspected PE could avoid 70,000-80,000 scans per year.

An accompanying editorial called the trial “a big step forward.”

However, important questions remain, according to co-author Joshua Beckman, MD, of the University of Texas Southwestern Medical Center in Dallas.

He pointed out that the trial aimed to recruit 1,500 patients, but after an interim analysis determined the noninferiority of YEARS, recruitment was halted at less than half of that goal.

That’s unfortunate, Beckman told Medscape Medical News, because a larger trial would have allowed a “more thorough and more detailed understanding of this approach.”

Importantly, he said, bigger numbers might have helped clarify the safety of using YEARS according to the type and stage of a patient’s cancer. Beckman noted that the risk for venous thromboembolism ranges widely based on cancer type — from a 12-month incidence of 1% among patients with prostate cancer to 12% among those with pancreatic cancer.

He also pointed to another issue: D-dimer levels also vary based on the cancer — with levels being generally higher with metastatic disease vs earlier stages.

For his part, Huisman acknowledged that it was impossible to conduct sub-analyses of patients with different cancer types and stages. But he believes that, given the robust non-inferiority demonstrated in the trial, the algorithm is “safe for all cancer types.”

Beckman had a more cautious take. He said that for patients with cancers associated with a low PE risk, using YEARS as a first-line tool may make sense.

But, he said, “I’m not sure I would feel comfortable yet in the cancers that may elevate the D-dimer by itself or have a very high rate of blood clots.”

The trial was funded by the INVENT 2021 Kickstarter Award Grant. Huisman and Beckman reported having no relevant disclosures.