Hemostatic effects of bortezomib treatment in patients with relapsed or refractory multiple myeloma

Due to increased baseline hypercoagulability, subjects with multiple myeloma (MM) are at increased risk of developing venous thromboembolism (VTE) compared with the general population.1 Both thalidomide and lenalidomide treatment have been associated with an increased thrombogenecity,2 however, this was not associated with shorter overall survival.3 Pre-clinical and clinical reports have suggested a possible anti-thrombotic effect of bortezomib.4,5 We report the first study which prospectively describes in vivo effects of bortezomib on routine tests of blood coagulation and platelet function in MM treated patients. Patients with relapsed /refractory multiple myeloma who failed one line of prior therapy were eligible for the study. After obtaining informed consent approved by the local institutional review board, bortezomib-naïve patients with relapsed/refractory MM were enrolled in the study. Upon enrollment, subjects were required to have a normal prothrombin time (PT) and partial thromboplastin time (PTT) and a platelet count >100×10/L.

Patients that were known carriers of a hypercoagulable state, had a previous history of cardiovascular disease or had platelet abnormalities were excluded. The use of any agents known to interfere with coagulation and platelet function was prohibited. Factor V Leiden and prothrombin gene mutations were measured only at baseline. All the remaining tests were obtained prior to and 1 hour after the first (day 1) and second (day 4) dose of bortezomib therapy. Platelet counts were obtained using a Hemavet® HV950FS (Drew Scientific, Oxford). The PT, activated partial thromboplastin time (APTT), and thrombin times (TT) were measured using a compact coagulometer (Diagnostica Stago). Proteins C, S and anti-thrombin III activities along with fibrinogen, D-dimer levels were also obtained, (Diagnostica Stago, Inc., Parsippany, NJ, USA).

Acquired resistance to activated protein C was determined using an APTT-based assay with pre-dilution in excess of factor V deficient plasma (Chromogenix Instrumentation Laboratory, Milano, Italy). For platelet aggregation studies, blood was collected in 3.2% citrate centrifuged at 200 g for 10 mins. and platelet-rich plasma (PRP) was obtained. Aggregation associated changes in light transmission was recorded for a total time of 5 mins. using a lumi-aggregometer (Chrono-Log Corp, Havertown, PA, USA) with 1 or 10 μg/mL collagen, 10 μM ADP, 1.2 mM arachidonic acid and 5 μM epinephrine. Ristocetin-induced platelet agglutination was also performed. Expression of P-selectin on the platelet surface at baseline and after ADP and collagen exposure was also measured on the collected samples. Whole blood was incubated with FITC-labeled anti-P-selectin antibody and PE-labeled anti-CD41a antibody (Pharmingen, BD Biosciences, San José, CA, C USA, at#555523 and Cat #555467 respectively). Twenty thousand positive events for CD41 were acquired from each sample and analyzed as per standard procedure (FACscan, Becton Dickinson, Franklin Lakes, NJ, USA).6

Table 1.Pre- and post-bortezomib infusion coagulation factor levels.

Ten patients with relapsed/refractory multiple myeloma were enrolled in the study (median age 62 years, range 50–72 years), 80% male. Myeloma subtypes were IgG/IgA/IgD/light chain 40%, 20%, 20% and 20% respectively. No mutations for V Leiden or prothrombin G20210 were identified. The median levels of coagulation factors pre and post bortezomib infusion are shown in Table 1; no statistically significant changes associated to bortezomib exposure were reported. As previously reported, we observed a modest decrease in median platelet count after bortezomib infusion (171×10/L vs. 153×10/L).7 With a median platelet rich plasma (PRP) count of 193×10/L, bortezomib infusion platelet aggregation was decreased after bortezomib infusion with most of the agonists used on both days of treatment. The decrease was statistically significant when comparing ADP induced aggregation in samples collected pre- and post-bortezomib infusion on days one, (20% decrease, p=0.033) and day four (29% decrease, p=0.009) of the cycle. Post-bortezomib samples collected on day one of the treatment cycle showed a significant decrease in epinephrine induced platelet aggregation (p=0.034); ristocetin induced agglutination was also decreased on day 4 after bortezomib infusion (p=0.0077). The percentages of inhibition on platelet-aggregation for each of the agonists are shown in Figure 1. Overall the expression level of the surface protein P-selectin was decreased on the platelets’ surface after bortezomib treatment; geometrical mean fluorescence values were 16.1 and 12.8 respectively before and after bortezomib doses under resting conditions. Overall, when platelets were activated by collagen or ADP, P-selectin expression was decreased after bortezomib treatment. Using 10 μMADP as agonist the geometrical mean fluorescence values before and after bortezomib doses were 81.9 and 73.1 respectively. After activation with 10 μg/mL collagen before and after bortezomib treatment we observed a geometric mean fluorescence value of 25.4 and 21.0. However, no statistical difference was reached after bortezomib exposure with any agonist.

Figure 1.Inhibition of platelet aggregation by bortezomib. For both days 1 and 4, post-treatment are compared with pre-treatment samples. Mean±SEM are shown.

This pilot clinical trial has shown for the first time in vivo that even a short exposure to bortezomib can affect platelet function. In experimental models, proteasome inhibitors are known to prevent cytokine-induced expression of the endothelial cell adhesion molecules E-selectin, vascular cell adhesion molecule-1, intracellular adhesion molecule-1 and P-selectin expression in activated human endothelial cells.8,9 In vitro bortezomib has also shown an inhibitory effect on platelet aggregation induced by ADP in a dose dependent manner.10 PSI, a highly specified peptide aldehyde proteasome inhibitor can prevent arterial thrombus formation in renovascular hypertensive rats; it has been shown to inhibit platelet aggregation and to increase the initial blood flow rate after arterial thrombosis induction.4 We have previously reported a low incidence of deep vein thrombosis during the induction phase of our study combining bortezomib, thalidomide and dexamethasone. 11 In addition, no thromboembolic events were observed in a group of 30 patients with relapsed MM treated with bortezomib, and melphalan, prednisone, thalidomide (MPT) combination despite the absence of any anticoagulant prophylaxis.12 The incidence of VTE reported by the same authors was significantly higher when MPT was given without bortezomib. 13 This manuscript supports strong in vivo evidence that bortezomib exerts anti-thrombotic actions mainly by its effects on platelet function. This effect should be tested after more prolonged bortezomib exposure and related to the clinical events.

Acknowledgments

we wish to thank Ashlie Finlayson for excellent technical assistance. JG has a fellowship from Ministerio de Educación y Ciencia