Introduction: D-dimer has been well characterized as a prognostic biomarker for venous thromboembolism (VTE) in both the general population and cancer patients. However, it is unclear if D-dimer has prognostic value after hematopoietic transplantation (HCT) in the context of regimen-related toxicity, immune dysregulation, and infectious complications. In the current study, we examined the utility of biomarkers of coagulation activation and fibrinolysis (D-dimer, PAI-1, and TPA) as prognostic and diagnostic biomarkers for VTE post-transplant.
Methods: We performed a prospective cohort study of adult allogeneic or autologous HCT recipients over 4 years at the Fred Hutchinson Cancer Research Center. Plasma samples were collected weekly from pre-transplant to discharge from HCT. Plasma samples were rapidly thawed and concentrations of D-dimer, PAI-1 activity, and TPA antigens were tested by commercially available immunoassays. VTE was defined as radiology confirmed pulmonary embolism (PE), lower extremity (LE) or upper extremity (UE) deep vein thrombosis (DVT) within 1 year after the date of transplant. This outcome was abstracted through a combination of ICD codes and individual patient record review. Cumulative incidence was estimated using the Kaplan Meier failure method. We used the Cox regression model to determine the association between baseline biomarkers and the development of VTE. The prediction accuracy of the model was assessed by the Harrell's C statistic. To explore the utility of D-dimer testing post-transplant, we selected all patients that had clinical suspicion for VTE, underwent radiographic evaluation (chest computed tomography or compression ultrasound), and had available sample for D-dimer testing within a two-week window. We defined the age-adjusted threshold as 500 ng/mL if at or below age 50 and 10 times the age if above age 50. We also repeated the analysis using a higher cut-off of 1000 ng/mL. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were generated from the 2x2 table.
Results: We identified 112 HCT adult recipients (97 allogeneic and 15 autologous) who were enrolled in the prospective cohort study and provided baseline plasma samples for biomarker testing. Patients had a mean age of 48, were predominately male and white, and had acute myeloid leukemia (AML) as the most common diagnosis (Table 1). Approximately 8% (9 patients) had a history of prior VTE. During the 1-year follow-up period, a total of 8 patients developed VTE (3 PE, 3 LE-DVT, 2 UE-DVT) with a cumulative incidence of 11.8% (5.8-23.2) by 12 months. There was no association between baseline TPA and PAI-1 and VTE; however, baseline D-dimer was associated with VTE with a HR of 1.91 (p=0.007) in the unadjusted model and a HR of 1.95 (p=0.007) in the model adjusted for history of VTE (Table 2). The predictive model with baseline D-dimer and history of VTE had a Harrell's C statistic of 0.75 for discrimination. For the exploratory analysis of the D-dimer post-transplant, we used a subset of 21 patients that had clinical suspicion of VTE, radiology confirmation, and available plasma sample for D-dimer testing. The sensitivity and NPV for both age-adjusted cut-off and 1000 ng/mL were 100%; however, the sample size was small (Table 3).
Conclusions: In this cohort study, we found that baseline D-dimer, but not PAI-1 or TPA, was associated with the development of VTE during the first-year post-transplant. A multivariable model with D-dimer and history of VTE had a modest discrimination for VTE prediction. Similar to many clinical settings, D-dimer testing post-transplant may aid clinicians with prediction and diagnosis of VTE. Larger cohort is needed for validation of the findings.
No relevant conflicts of interest to declare.
This feature is available to Subscribers Only
Sign In or Create an Account Close Modal