Cancer Screening Based on Unprovoked Venous Thromboembolism

Alimited cancer screening strategy, including basic blood testing without a CT scan, is adequate for detection of occult cancer in patients with a first unprovoked venous thromboembolism (VT), according to a study published June 22 in the New England Journal of Medicine. An unprovoked VT (as opposed to a deepvein thrombosis or pulmonary embolism tied to a transient risk factor like trauma, surgery, prolonged immobility, or pregnancy) may be an early sign of cancer. Previous studies estimate that as many as 10 percent of patients with an unprovoked VT receive a cancer diagnosis in the year following VT and unprovoked cases represent more than 40 percent of all VTs. Clinicians and payers have struggled with how aggressively to screen for occult cancers in these patients with unprovoked VT.

In the current multicenter, Canadian trial, researchers randomized 854 patients with a new diagnosis of first unprovoked symptomatic VT to receive either limited occult cancer screening (including medical history, physical exam, complete blood counts, serum electrolyte and creatinine levels, liver-function testing, and a chest x-ray) or extensive screening with the addition of a CT of the abdomen and pelvis. Additional sex-specific testing (mammography, Papanicolaou, and prostate-specific antigen) occurred for those not up-to-date with recommended screenings.

The investigators found that following the initial screening strategy, just over 14 percent of patients in both groups underwent additional testing for a potential cancer diagnosis. A total of 3.9 percent of all patients were diagnosed with cancer within 1-year of randomization, with no significant difference in numbers diagnosed between the groups—a lower than expected rate. Of patients over the age of 50 years, 6.7 percent in the limited-screening group and 10.2 percent in the extensive screening group underwent colon cancer screening (fecal occult-blood testing, sigmoidoscopy, or colonoscopy), which led to diagnosis of three colorectal cancers in the extensive screening group.

Limited screening missed four of 14 occult cancer cases, while extensive screening missed five of 19 occult cancer cases. Acute leukemia (two cases), gynecologic tumors (two), and colorectal tumors (two) were equally missed by the two screening strategies. The mean time to cancer diagnosis was similar between the strategies (4.2 months for limited versus 4.0 months for extensive screening).

"The risk of subsequent cancer was also quite low, and 'doing more' did not lead to earlier cancer detection," writes Alok A. Khorana, M.D., from the Cleveland Clinic (Ohio), in an accompanying editorial.

Takeaway: The risk of diagnosis with an occult cancer in the year following an unprovoked VT may be lower than previously thought, particularly among patients in their 50s. A limited cancer screening strategy is adequate to catch cancers in this population.