Role of P‑selectin in thromboembolic events in patients with cancer
- Lúcio Flávio Barbour Fernandes
- José Humberto T.G. Fregnani
- Célia Maria Cássaro Strunz
- Adriana de Andrade Ramos Nogueira
- Adhemar Longatto‑Filho
Published online on: November 2, 2017
The objective of the present study was to evaluate the role of P‑selectin in patients with cancer with suspected thromboembolic events (TEEs). Patients with cancer have a four times greater risk of developing TEEs. P‑selectin is a glycoprotein that has the function of facilitating the interaction (adhesion) of leukocytes with the endothelium, or with platelets. There is a well‑defined relationship between P‑selectin and thrombosis; however, it is likely that the cut‑off value of P‑selectin for patients with cancer should be considered differently from that of the general population. In the present report, a prospective cross‑sectional study was performed with patients of the Cancer Hospital of Barretos who were suspected of having TEEs. Among the 178 study participants, 167 (93.82%) were suspected of having deep vein thrombosis, while 59 of them (35.33%) were confirmed as such; and 11 (6.18%) were suspected of having pulmonary thromboembolism, while 3 of them were confirmed as such (27.69%). The mean results obtained were: P‑selectin, 25.37 ng/ml; and D‑dimer, 2,181.22 ng/ml. The P‑selectin levels averaged 33.60 ng/ml with the confirmed TEE group compared with 20.40 ng/ml with the unconfirmed TEE group, with a standard deviation of 23.35 compared with 6.92 (P<0.001); and the level of D‑dimer was 4,615.38 ng/ml compared with 977.52 ng/ml, with a standard deviation of 6,460.54 compared with 2,145.50 (P<0.001). Multiple logistic regression adjusted for distant metastases and the Eastern Cooperative Oncology Group (ECOG) score (2,3 and 4) were constructed. The cut‑off value of P‑selectin for patients with cancer was identified to be different from that reported in the literature for the general population, and the models using D‑dimer and P‑selectin therefore have been demonstrated to be a potentially useful tool to be used in a panel of tests to predict TEEs, either independently or in a prediction score.