Abstract
Background/Aim: BCR-ABL tyrosine kinase inhibitors (TKIs) are exceptionally effective drugs in the treatment of chronic myeloid leukemia, nevertheless, TKIs have also an effect on platelets. We aimed to investigate the effect of a third-generation TKI, ponatinib on platelet functions. Materials and Methods: Collagen-induced platelet aggregation and coated-platelet formation were examined using in vitro and in ex vivo samples of patients on ponatinib therapy. Results: In platelet rich plasma of healthy volunteers, ponatinib at a supra-therapeutic concentration (1,000 nM) significantly impaired collagen induced platelet aggregation (p≤0.01) and reduced the formation of coated-platelets at 150 nM ponatinib concentration (p≤0.05). In addition, upon glycoprotein VI (GPVI) receptor activation, a significantly lower percentage of PAC1 binding platelets (p≤0.05) was observed at 1,000 nM final concentration of ponatinib. Platelets, isolated from patients on ponatinib therapy showed impaired collagen elicited aggregation response, already in pre-dose samples compared to healthy donors. Conclusion: The therapeutic concentration of ponatinib impairs platelet activation processes elicited by GPVI receptor agonists.
Chronic myeloid leukemia (CML) is a clonal bone marrow stem cell malignancy characterized by unregulated and increased growth of myeloid cells due to reciprocal translocation between chromosome 9 and 22 (Philadelphia chromosome). This chromosomal abnormality results in the formation of a chimeric oncogene between the Abelson 1 kinase (ABL1) and breakpoint cluster region (BCR) genes and leads to production of an abnormal tyrosine kinase oncoprotein (BCR/ABL1), which pathologically affects cell differentiation and proliferation. The first generation of tyrosine kinase inhibitors (TKIs) targeted against BCR/ABL1 (imatinib) dramatically improved the overall survival of CML patients (1). However, in the past two decades additional generations of TKIs had to be developed as TKI resistance emerged due to several point mutations in the BCR/ABL1 gene. These second- and third-generation TKIs are more potent inhibitors of BCR/ABL1 (2), they have excellent anti-kinase activity against mutated BCR/ABL1 but they exert multikinase (off-target) activity, also interfering with several important BCR/ABL1 independent signal transduction pathways (3). Hence ponatinib treatment can also be applied effectively in patients with other types of malignancies (4) but can also be associated with more adverse events than patients on first-generation TKI therapy (5).
Extensive research has been conducted regarding the possible pathological mechanisms underlying second- and third-generation TKI therapy-related side-effects, in particular cardiovascular and bleeding complications (5-13). Bleeding tendencies in patients undergoing BCR/ABL1 targeted therapies have been hypothesized to occur as a result of platelet inhibition (9, 11-13) or a decrease in platelet production by megakaryocytes (14). On the contrary serious cardiovascular complications have been observed in CML patients on second- or third-generation TKI therapy (15).
Ponatinib is a third-generation TKI and is highly effective in patients with resistance to first- or second-generation TKIs (16, 17). It is a true multikinase inhibitor and exhibits potent activity against both unmutated and mutated BCR/ABL1, including the prognostically unfavorable T315I mutation. CML patients treated with ponatinib have improved progression-free survival, but they also show a significantly increased risk for vascular occlusive events (15, 18). However, investigation of ponatinib effect on platelet function has revealed that ponatinib indeed inhibits platelet activation, spreading, granule secretion, and aggregation, likely through inhibition of a broad spectrum of tyrosine kinase signaling pathways in platelets (12, 13). On the other hand, no clinically significant bleeding tendencies have been reported in CML patients on ponatinib therapy (19, 20). Additional examinations regarding the vascular complications of ponatinib therapy have revealed that ponatinib exerts anti-angiogenic effects in human umbilical vein endothelial cells (HUVECs) via blocking the vascular endothelial growth factor receptor (VEGFR) signaling pathway (21), triggers inflammatory response in HUVECs by inhibiting extracellular signal-regulated kinase 5 (ERK5) transcriptional activity (22) and induces apoptosis, reduces viability, and inhibits VEGF-induced migration and tube formation of HUVECs (23). In mice treated with ponatinib, von Willebrand factor (vWF)-mediated platelet adhesion and secondary microvascular angiopathy contributes to ischemic wall motion abnormalities (24) and ponatinib at human therapeutic concentrations in mice exhibited deleterious effects on the vessel wall and selectively modulated agonist-induced platelet reactivity (25).
To further investigate this controversial activity of ponatinib on platelet function we have studied the in vitro effect of ponatinib on collagen-induced platelet aggregation, adenosine triphosphate (ATP) secretion and coated-platelet formation. Besides, from ex vivo samples the measurement of collagen elicited platelet aggregation was also carried out. The purpose of our study was to utilize these methods to further clarify ponatinib effect in experimental conditions and in ex vivo samples. Our results demonstrate that ponatinib is capable to downregulate the effect of the glycoprotein VI (GPVI) receptor agonists at therapeutic concentrations.
Materials and Methods
Blood drawing from healthy volunteers and patients. Peripheral blood samples were drawn from healthy volunteers into tubes containing 0.105 M sodium citrate. Healthy volunteers were recruited from the staff of the Department of Laboratory Medicine. Patients with chronic phase CML on ponatinib therapy (n=5) were included in the ex vivo study (Table I). Blood was drawn immediately before (pre-dose) and 4 h after (post-dose) witnessed drug administration in case of all five patients. Subjects were recruited from the Hematology Outpatient Clinic at the Department of Internal Medicine, University of Debrecen. At the time the samples were taken, all CML patients were on continuous ponatinib treatment and they did not receive anticoagulant and antiplatelet therapy. Informed consent was obtained from all participants (CML patients and healthy volunteers) in accordance with the local institution review board guidelines. Ethical agreements were provided by the local ethical committee of the University of Debrecen (Permit number: RKEB/IKEB 4875-2017).
Characteristics of CML patients on ponatinib treatment.
Preparation of platelet rich plasma (PRP) and design of in vitro and ex vivo study. PRP was prepared from anticoagulated whole blood of healthy volunteers (for in vitro study) and ponatinib treated CML patients (for ex vivo study) by centrifugation at 170 × g for 15 min at room temperature (RT). Platelet count of PRP was adjusted to 250 G/l by adding platelet poor plasma (PPP). PPP was obtained by centrifugation of the citrated blood sample at 1500 × g for 15 min at RT. In a series of in vitro studies ponatinib (Cayman Chemical; Ann Arbor, Michigan, USA) pretreatment was used at 75, 150 nM and 1,000 nM final concentration for 10 min at 37°C. Firstly, the effect of ponatinib was investigated on collagen and adenosine diphosphate (ADP) induced platelet aggregation, ATP secretion, PAC1 binding (PAC1 antibody binds to the active conformation of integrin αIIbβ3) and coated-platelet formation. From ex vivo samples collagen elicited aggregation tests were carried out.
Platelet aggregation test: in vitro and ex vivo studies. Similarly to our previous study, platelet aggregation and secretion induced by 1 μg/ml fibrillar collagen (Takeda, Linz, Austria) or 5 μM ADP (Labexpert Ltd., Debrecen, Hungary) were tested (9).
Flow cytometric assays. Active conformation of the integrin αIIbβ3 was determined in non activated, 20 μM thrombin receptor activating peptide [TRAP; protease-activated receptor 1 (PAR1)-agonist] or 25 ng/ml convulxin (GP VI agonist) activated samples by binding of fluorescein isothiocyanate (FITC)-conjugated PAC1 and phycoerythrin-Cyanine 5 (PECy5)-conjugated CD41a antibodies (10). For coated-platelets, subsequent gel-filtered (GFP) platelet activation with the mix of convulxin and thrombin was assessed in the presence of biotinylated-fibrinogen. The levels of coated-platelets were determined in percentage of platelet marker positive platelets (10). For the determination of platelet-monocyte aggregates ponatinib pretreated control whole blood samples were incubated with CD14 phycoerythrin (PE) and CD42a FITC. Results were always compared to samples stained with non-immune immunglobulin G (IgG) that served as isotype control. Red blood cells were lysed and the samples were washed twice with phosphate buffered saline (PBS) and fixed with paraformaldehyde (PFA) and subsequently measured by an FC500 flow cytometer and results were analyzed with the Kaluza software (Beckman Coulter).
Statistical analysis. GraphPad Prism version 6.01 program was used for the statistical analysis. Data distribution was evaluated by Kolmogorov-Smirnov test. The statistical significance of the differences between groups of in vitro experiment was analysed by one-way ANOVA in case of Gaussian distribution, and by Kruskall Wallis test in case of non-Gaussian distribution, as appropriate. Differences were considered significant when p-values were below 0.05.
Results
In vitro effect of ponatinib on platelet aggregation and the activation of integrin αIIbβ3. Platelet activation in primary hemostasis involves multiple agonists and their receptors. Adhesion of platelets to extracellular matrix proteins, like collagen, is followed by platelet activation and aggregation. During these processes ADP is released from the dense granules and propagates thrombus formation.We investigated the effect of ponatinib on the aggregation response of platelet rich plasma to collagen or ADP. We observed a decreasing tendency in the aggregation response (Figure 1A) and ATP secretion (Figure 1B) to collagen and this phenomenon became significant at 1,000 nM concentration of ponatinib for both measured parameters (the maximum of light transmission changes (∆Tmax) value of aggregation and ATP secretion). On the other hand, during ADP (5 μM)-induced aggregation neither platelet aggregation nor ATP secretion were affected by ponatinib (data not shown). The activation of integrin αIIbβ3 is essential for platelet aggregation. Bacause of this, we examined the effect of ponatinib on the level of activated αIIbβ3 by PAC1 binding. After ponatinib pretreatment platelets were activated with convulxin and PAC1 binding was measured. We observed that ponatinib pretreatment dose-dependently reduced the percentage of PAC1 bindig platelets in convulxin activated samples and these changes reached the level of significance at 1,000 nM final concentration of ponatinib in convulxin activated samples (Figure 1C).
In vitro ponatinib effect on platelet aggregation and integrin αIIbβ3 activation. Platelet rich plasma (PRP) from healthy donors were incubated with 75, 150 and 1,000 nM ponatinib or vehicle (0.2% DMSO) for 10 min at 37°C and subsequently collagen (1 μg/ml; n=11) induced platelet aggregation and adenosine triphosphate (ATP) release were measured. Pretreatment with different ponatinib concentrations resulted in a dose-dependent inhibition of collagen induced aggregation (A) and ATP secretion (B) and this effect became significant at 1,000 nM concentration of ponatinib. The level of activated integrin αIIbβ3 was detected by fluorescein isothiocyanate conjugated PAC1 (FITC PAC1) binding (n=9). Panel C shows a decreasing tendency of the percentage of PAC1 binding platelets in ponatinib pretreated samples after convulxin (25 ng/ml)-induced activation. A significant difference was observed only at 1,000 nM concentration of ponatinib in convulxin activated samples. (aggregation ∆Tmax: maximal light transmission of sample *p≤0.05; **p≤0.01).
The effect of ponatinib on in vitro formed coated-platelets. Coated-platelets are a subpopulation of strongly stimulated platelets that are formed during activation through the GPVI and PAR1 receptors, thus this platelet subset is highly procoagulant. During in vitro studies, GFPs from healthy controls were activated with a mix of convulxin and thrombin and the quantity of coated-platelets was measured by flow cytometry. Ponatinib pretreatment dose-dependently inhibited the formation of coated-platelets (Figure 2A). After summarizing the results of six parrallel experiments, we concluded that ponatinib pretreatment significantly inhibited the formation of coated-platelets already at a therapeutic concentration (150 nM) and their quantity further decreased using 1,000 nM final concentration of ponatinib (Figure 2B).
In vitro effect of ponatinib on coated-platelet formation. Gel-filtered platelets of healthy volunteers were incubated at different ponatinib concentrations (75, 150 and 1,000 nM) for 10 min at 37°C and coated-platelet generation was measured by biotinylated fibrinogen binding (n=6). A representative series of fibrinogen binding dot-plots of ponatinib pretreated coated-platelets are shown (A). The box and whisker plots (B) show the percentage of coated-platelets. The quantity of coated-platelets was significantly reduced already at 150 nM concentration of ponatinib. (*p≤0.05; ***p≤0.001).
Ponatinib treatment of CML patients moderately affects platelet functions. To investigate the clinical relevance of these findings, blood samples of five patients in the chronic phase of CML on ponatinib treatment were examined (Table I). Blood samples were drawn immeditely before (pre-dose) and at 4 h after (post-dose) ponatinib administration. The dosage of ponatinib was variable, so we could not evaluate the data combined, instead we evaluated them as individual results. Similarly to in vitro studies, collagen induced platelet aggregation were measured in PRPs of ponatinib treated patients. In four patients, a decreased collagen induced aggregation responses was measured in the pre-dose samples. In addition, we observed that three patients displayed decreased aggregation response to collagen after 4 h of administration of ponatinib (Table II). Ponatinib significantly inhibited collagen-induced aggregation in post-dose samples of patients taking 30 mg once daily (QD) (p=0.003) compared to control platelets. It needs to be noted, that in one patient on 30 mg QD ponatinib, we carried out the aggregation studies prior to initiation of ponatinib therapy and after 2 weeks on ponatinib treatment. We observed the following ∆Tmax values of collagen-induced aggregation, 86% in the drug-free state, and 73% and 75% during the on-treatment period in the pre-dose and 4 h post-dose samples respectively.
Characteristics of collagen-induced aggregation response of samples of CML patients before and after ponatinib administration.
In vitro effect of ponatinib on the formation of monocyte-platelet heterotypic aggregates. Our results demonstrated that ponatinib had an inhibitory effect on platelet function in PRP and GFP. However previously it has been described that ponatinib exerts pro-thrombotic and pro-inflammatory phenotype in CML patients on TKI treatment. Therefore, the analysis of monocyte-platelet aggregates may be a sensitive indicator of these processes (26, 27). Flow cytometry was used to determine the level of heterotypic aggregates after ponatinib pretreatment in non-activated samples and in samples activated by the GPVI receptor agonist, convulxin. We did not observe any changes in the level of monocyte-platelet heterotypic aggregates when control platelets were pretreated with different concentrations of ponatinib with no subsequent platelet activation (Figure 3A and B). After ponatinib pretreatment and convulxin activation the percentage of monocyte-platelet aggregates increased dose-dependently and this enhancing effect became significant at 1,000 nM ponatinib pretreatment (Figure 3C).
In vitro ponatinib effect on the formation of heterotypic aggregates. Whole blood from healthy donors (n=6) was incubated with 75, 150 and 1,000 nM ponatinib or vehicle (0.2% DMSO) for 10 min at 37°C and subsequently samples were activated by convulxin or were left without activation. The percentage of monocyte-platelet aggregates was determined by fluorescein isothiocyanate (FITC) conjugated CD42a binding (A). The histogram shows percentages of monocyte binding FITC-conjugated CD42a in non-activated (B) and convulxin activated samples (C). There is a general increasing tendency in the ratio of heterotypic aggregates but only 1,000 nM of ponatinib caused significant elevation in convulxin activated samples (*p≤0.05).
Discussion
Cancer treatment has undergone a tremendous improvement as more and more carefully constructed molecules are incorporated into the clinical practice. TKI and monoclonal antibodies have been designed to specifically target overexpressed receptor tyrosine kinases and subsequent downstream molecules in malignant cells. The third-generation TKI ponatinib was designed to overcome TKI resistance in CML treatment. This was the first effective drug in patients carrying the T315I mutation (28). However, after its accelerated approval by the Food and Drug Administration in 2012 it was abruptly withdrawn from the market in less than a year later due to the increasing number of vascular events that were recorded in phase I and in subsequent phase II clinical trials (16, 17). Taken these initial data, obviously a considerable amount of research dealt with the prothrombotic effects of ponatinib. These studies found that ponatinib induces a thromboinflammatory response as was demonstrated in a collagen and FeCl3 injury model (29). Other authors have proven in animal experiments that ponatinib treatment predisposes to platelet activation as was detected by αIIbβ3 activation analysing JON/A expression and P-selectin expression in mice. The differences were, however only evident when lower concentrations of the agonists (collagen or thrombin) were used (25).
Contrary to these studies, a few reports have dealt with the antiplatelet effect of ponatinib. Recent pharmacokinetic studies report that the half-life of ponatinib is between 27-34 h and the mean ponatinib C0 values in patients following administration of daily doses of 15, 30 or 45 mg ponatinib were 26.2 nM, 56.1 nM and 64.3 nM, respectively. In these studies it was shown that ponatinib reached Cmax at 4-6 h after drug administration and the Cmax values were 77 nM and 145 nM with 15 mg or 45 mg daily regimen, respectively (16, 30, 31). In our experiments we have investigated the effect of ponatinib at clinically relevant concentrations (75 and 150 nM) and at a supra-therapeutic level (1,000 nM) on platelet agonist induced changes.
Our group has previously shown that the second-generation TKI dasatinib inhibits the formation of procoagulant and clot retracting platelets (10). In that report, we also demonstrated that the Lck/Yes novel tyrosine kinase (Lyn), proto-oncogene tyrosine-protein kinase Fyn (Fyn) and proto-oncogene tyrosine-protein kinase cellular-Src (Src) are inhibited by dasatinib. Ponatinib is capable of inhibiting these kinases too, which may explain its attenuating effect to platelet aggregation and ATP release assays. The significant inhibitory effect of ponatinib on collagen elicited aggregation response was observed only at supra-therapeutic concentration (1,000 nM). Our findings were consistent with a previous study that utilized purified platelets. This might be the reason why ponatinib affected collagen related peptide (CRP) elicited platelet aggregation already at 100 nM concentration (13). Loren et al. found that in vitro addition of ponatinib to washed platelets inhibited GPVI activation pathways and ponatinib treated platelets spreaded less on fibrinogen or collagen surfaces similar to platelets treated with a specific Src family kinase-inhibitor. Furthermore, ponatinib treated platelets attached to fibrinogen or collagen have reduced phosphorylated GPVI pathway kinases similar to the specific Src family kinase inhibitor treated platelets (13).
When the GPVI receptors were activated after ponatinib pretreatment a decreasing tendency in the percentage of PAC1 binding platelets was observed and a significant attenuating effect of ponatinib was detectable at 1,000 nM concentration of ponatinib. In line with our observations, Deb et al. found, that ponatinib can decrease the mean fluorescence intensity of PAC1 binding platelets upon the activation of PAR1 and PAR4 receptors at 145 nM final concentration but they did not find any effect of ponatinib upon the GPVI receptor activation in whole blood samples (8). Based on our previous study with dasatinib, we hypothesized that coated-platelet detection might be a more sensitive method to identify the inhibitory effect of ponatinib than the platelet aggregation test. Coated-platelets are dual-agonist activated platelets and they are highly procoagulant (32). In numerous pathological states, elevated (33-35) or decreased (36-38) coated-platelet levels were observed. When GFP samples were co-activated via the GPVI (convulxin) and the PAR1 (thrombin) receptor the effect was evident and the decrease in coated-platelets became significant already at 150 nM ponatinib and their number was halved at 1,000 nM. We consider the effect on coated-platelets pathologically relevant as previously we found it more sensitive than platelet aggregometry and thus it might predict more efficiently bleeding in CML patients on second-generation TKI therapy (9).
Ponatinib therapy is restricted to a small proportion of CML patients who have failed other TKI treatments or harbor the rare T315I mutation of BCR-ABL1. Currently 22 patients are on ponatinib therapy in Hungary. We had access only to a limited number of these patients (n=5) to evaluate the effect of ponatinib treatment on platelet functional assays. Maximal platelet aggregation (∆Tmax value) as investigated by collagen was considerably suppressed already in pre-dose samples and stayed below the reference range or decreased to a subnormal value. Thus both the in vitro and ex vivo results underscore the inhibitory effect of ponatinib on platelet activation elicited by classical platelet agonist via the GPVI receptor. In vitro studies were done under controlled conditions on PRP or GFP samples of drug-free volunteers. In CML patients several simultaneously added drugs may replace the highly protein bound ponatinib and thus can affect concentrations of the free drug. Also these patients may have comorbidities and the ponatinib is exerting its effect in the flowing blood and not in a static PRP or GFP samples. We assume that the inhibitory effect of ponatinib on platelets may be related to the inhibition of Sarcoma family kinases (Lyn, Fyn and Src). In a previous study the degree of inhibition was found to be similar in case of dasatinib and ponatinib (39), which may explain its attenuating effect to platelet activation via the GPVI receptor. As Lyn is constitutively active, the lower ponatinib dose may have produced the open conformation of p−Lyn by blocking only the higher inhibitory site at p−LynY507. These combined studies indicate that ponatinib treatment alters both vessel wall and platelet functions, making the latter hyperactive (39). It was also demonstrated that ponatinib therapy in mice can result in vWF-mediated platelet adhesion to the microvascular endothelium (24). Platelet adhesion was mediated by GPIbα binding of the exposed A1 domain on endothelial or leukocyte–platelet complex associated vWF. These processes occurred in both large arteries and in the peripheral microcirculation. This latter can produce ischemic changes in ventricular function (24).
The cardiovascular events in patients taking ponatinib may be a result of an effect on platelets and on other cells in the circulation. In our purified experimental setup we explored the inhibitory effect of ponatinib on platelets. However, the results of the formation of monocyte-platelet aggregates in whole blood samples demonstrated a possible activating effect of ponatinib on platelets and/or monocytes.
The limitation of our studies is the small number of samples of ponatinib treated patients. On the other hand, we verified the inhibitory effect of ponatinib in purified experimental settting.
In conclusion by using lumi-aggregation studies and a flow cytometric technique for the detection of coated-platelets during in vitro experiments and in ex vivo samples we demonstrated an inhibitory effect of ponatinib on human platelets where the resultant effect depends largely on the actual measuring conditions.
Acknowledgements
The technical assistance of Erzsébet Nagy and Csaba Antal are acknowledged. This work was supported by the GINOP-2.3.2-15-2016-00043 – Ironheart and OTKA K16 120725 project.
Footnotes
Authors’ Contributions
Conceptualization, G.M., J.K. and I.B.D.; methodology, I.B.D.; formal analysis, I.B.D. and J.K.; investigation, I.B.D., K.L., G.M. and P.B.; data curation, I.B.D.; writing-original draft preparation, I.B.D., G.M and P.B.; writing-review and editing, J.K.; visualization, I.B.D.; supervision, J.K. and I.A.; funding acquisition, J.K.
Conflicts of Interest
The Authors state that they have no conflicts of interest.
- Received July 22, 2021.
- Revision received August 27, 2021.
- Accepted August 31, 2021.
- Copyright © 2021 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.