Ibrutinib-associated bleeding; pathogenesis, management, and risk reduction strategies.

Abstract

Ibrutinib is an irreversible inhibitor of Bruton’s tyrosine kinase (Btk) that has proven to be an effective therapeutic agent for multiple B-cell mediated lymphoproliferative disorders. Ibrutinib, however, carries an increased bleeding risk compared to standard chemotherapy. Bleeding events range from minor mucocutaneous bleeding to life-threatening hemorrhage, due in large part to the effects of ibrutinib on several distinct platelet signaling pathways. There is currently minimal data to guide clinicians regarding the use of ibrutinib in patients at high risk for bleeding or on anticoagulant or antiplatelet therapy. In addition, the potential cardiovascular protective effects of ibrutinib monotherapy in patients at risk for vascular disease is unknown. Patients should be cautioned against using nonsteroidal anti-inflammatory drugs, fish oils, vitamin E, and aspirin-containing products, and consider replacing ibrutinib with a different agent if dual antiplatelet therapy is indicated. Patients should not take vitamin K antagonists concurrently with ibrutinib; direct oral anticoagulant should be used if extended anticoagulation is strongly indicated. In this review, we describe the pathophysiology of ibrutinib-mediated bleeding and suggest risk reduction strategies for common clinical scenarios associated with ibrutinib.

Introduction:

Ibrutinib is an irreversible inhibitor of Bruton’s tyrosine kinase (Btk), which is approved for the treatment of several B-cell malignancies.[1] A side effect of ibrutinib is an increased rate of clinically significant bleeding as compared to standard chemotherapy. This has been attributed in large part to the platelet-specific effects of inhibiting the non-receptor tyrosine kinases Btk and Tec, as well as drug induced thrombocytopenia. The relatively recent introduction of ibrutinib and lack of precise data on its risk in combination with other antiplatelet agents and anticoagulants has created a lack of consensus on the proper use of ibrutinib in patients at risk of bleeding and in those treated with other drugs with adverse bleeding effects. Guidelines on the management of ibrutinib in the setting of active bleeding or surgery are also poorly defined. Lastly, ibrutinib is associated with an increased risk of atrial fibrillation which itself often mandates anticoagulation, leading to complex decisions regarding stroke versus bleeding risk.[2] In this review we describe the pathophysiology of ibrutinib-mediated platelet inhibition and suggest management and risk-reduction strategies for common clinical scenarios associated with ibrutinib.

The history of Btk and ibrutinib:

The discovery of Btk dates back to the 1950s. In 1952 Colonel Ogden Bruton, a pediatrician at Walter Reed Army Hospital, reported the first case of congenital agammaglobulinemia in an 8 year-old boy.[1] This X-linked disease was later coined X-linked agammaglobulinemia (XLA) or Bruton’s agammaglobulinemia.[3] In 1993, the molecular mechanisms of the disease were discovered to be related to mutations in a specific member of the Tec family of protein tyrosine kinases, later named Bruton’s tyrosine kinase.[1] Btk is responsible for a variety of intracellular signaling pathways in B-lymphocytes and is required for B-cell precursor maturation; the absence of Btk in those with XLA leads to a paucity of peripheral B-cells and immunoglobulins.[4] Btk is also necessary for B-cell receptor signaling in the adaptive immune response and lymphocyte trafficking, among other functions.[5] In 2009, the first human Btk inhibitor, ibrutinib, was developed.[6] Since then, several clinical trials have shown efficacy in treating B-cell malignancies, leading to FDA approval for the treatment of chronic lymphocytic leukemia (CLL), mantle cell lymphomas and lymphoplasmacytic lymphoma, otherwise known as Waldenström’s macroglobulinemia.[1]

Current clinical data on ibrutinib-associated bleeding.

After three years of follow up, more than half of patients on ibrutinib will have a bleeding event. [7] The majority of reported bleeding events are low grade (Common Terminology Criteria for Adverse Events 4.0 (CTCAE) Grade I-II bleeding) (Table 1). The most common bleeding phenotypes are subcutaneous or mucosal bleeding, including contusions, epistaxis, petechial bleeding, hematuria or ecchymosis.[7] Major hemorrhage (Grade III-IV) has been reported with rates varying from 4% to 8% in trials that followed patients for over a year,[7–11] with fatal hemorrhage occurring in less than 1% of patients (0.6–0.7%).[9, 11] Subdural hematoma is the most commonly reported form of central nervous system (CNS) bleeding,[7, 10–12] though hemorrhagic conversion of ischemic stroke, subarachnoid hemorrhage after a fall and vitreous hemorrhage have also been reported.[12]

Table 1.

Incidence of bleeding events with ibrutinib.^*

Trial	Phase	Comparison	Median follow up	Any bleeding (%)	Grade 1–2 hemorrhage (%)	Major hemorrhage Grade III-IV (%)	Fatal hemorrhage (%)	Major hemorrhage on concomitant AC, antiplatelet, or Vitamin E (%)	Median time to major hemorrhage	Thrombocytopenia Grade 1–2 (%)	Thrombocytopenia Grade 3–4 (%)	Held for procedures /surgery	Excluded warfarin?	ASA during study (%)	Clopidogrel during study (%)	AC during study (%)
Chanan- Kahn, Lancet 2016	Phase III	Ibrutinib R- Benda (n=287) Vs. R-Benda (N=287), CLL	17 mo	31% vs 15%	28% vs 9%	4% vs. 2%	0.6% vs 0	54% (6/11) vs 60% (3/5)	4.2 mo vs 2.3 mo	16% vs 9%	15 vs 15	Not reported	YES
Burger, NEJM 2015	Phase III	Ibrutinb (N=136 )Vs. Chlorambacil (N=133), CLL	18.4 mo	Not Reported	Not Reported	4% vs. 2%	0 vs 0	50% (3/6) vs not reported	Not Reported	Not Reported	2% vs 6%	YES (14 days)	YES
Byrd, NEJM 2014	Phase III	Ibrutinib (N=195) Vs. Ofatumumab (N= 196), CLL	9.4 mo	44% vs 12%	27% vs 10%	1% vs 2%	0 vs 0	Not Reported	Not Reported	11% vs 8%	6% vs 4%	YES	YES
Dreyling, Lancet 2016	Phase III	Ibrutinib (N=139) Vs. temsirolimus (N= 141), MCL	20 mo	Not Reported	Not Reported	10%vs 6%	0 vs 0	Not Reported	Not Reported	9% vs 14%	9% vs 42%	Not reported	YES
Byrd, Blood 2015	Phase II	Ibrutinib (N=132) CLL	36 mo	61%	53%	8%	0.7%(1)	Not Reported	Not Reported	Not Reported	8%	Later in protocol. 3 days for minor, 7 days for major	Later in protocol	33%	2%	22%
Wang, Blood 2015	Phase II	Ibrutinib (N= 111 ) MCL	26.7 mo	50%	44%	6%	0	85% (6/7) (includes NSAID) Bleeding events were higher in those on AC/anti plt: (69% any grade, 8% grade 3–4) vs those not receiving these treatments (28% any grade, 4% grade 3–4).	Not Reported (41% were in the first 6mo)	9%	13%	Not reported	Later in protocol	26%	6%	24% (14% LMWH, 10% Warfari n)

^*Phase III and selected phase II trials with over 100 patients, reporting bleeding events.

Post-procedural bleeding accounted for a large amount of bleeding events in initial clinical trials,[11, 13] leading to amendments requiring discontinuation of the drug for a period of time before and after invasive procedures. A protocol of holding ibrutinib for three days prior to minor procedures and seven days prior to major surgery was used in some clinical trials, with re-initiation of the drug 1–3 days later.[11, 13] Other trials held the drug for 14 days (one week prior and one week after the procedure).[10]

Initial ibrutinib clinical trials allowed the inclusion of patients on warfarin. These were later amended, however, to exclude vitamin K antagonists given evidence of excessive bleeding.[7, 11] Ibrutinib is currently contraindicated in the setting of warfarin use. Clinical trials did allow other anticoagulants including heparin, low molecular weight heparin (LMWH) and direct oral anticoagulants (DOACs).[14] Clinical trials generally did not restrict the use of other antiplatelet agents concurrently with ibrutinib including aspirin, clopidogrel, and nonsteroidal anti-inflammatory drugs (NSAIDs). At least one clinical trial reported that concurrent use of antiplatelet agents or anticoagulants resulted in notably increased rates of bleeding (69% any grade bleeding, 8% grade 3–4 in those on concurrent anticoagulation/antiplatelet vs 28% any grade, 4% grade 3–4 in those on ibrutinib alone).[7] Bleeding has also been reported in association with other drugs known to inhibit platelet function including vitamin E[10] and fish oil.[13]

Lastly, drug-induced thrombocytopenia does occur with ibrutinib and may also increase bleeding risk, although only to a mild degree. Grade III-IV thrombocytopenia has been reported at rates from 2% to 13% when the drug is used as a single agent.[7, 10]

The pathophysiology of ibrutinib-associated bleeding.

While Btk is important in platelet signaling, individuals with XLA do not have a bleeding phenotype, suggesting that both on- and off-target effects of ibrutinib contribute to aggravated bleeding.[15] Besides irreversibly inhibiting Btk, ibrutinib inhibits several other intracellular molecules important for platelet signaling including Tec (tyrosine kinase expressed in hepatocellular carcinoma), the first discovered kinase of the Tec family of protein-tyrosine kinases.[16] Both Btk and Tec are irreversibly inhibited by ibrutinib at clinically relevant concentrations with the IC₅₀ of 0.5 and 78 nM, respectively.[6] Btk and Tec function in the downstream signaling of several platelet transmembrane receptors, including the platelet collagen receptor glycoprotein VI (GPVI) and C-type lectin-like receptor 2 (CLEC-2). GPVI is responsible for platelet activation secondary to collagen and collagen-related peptide exposure.[17] (Figure 1) Platelets from patients with XLA have mildly diminished, but not abolished, collagen activation.[17, 18] It has been shown that in the absence of Btk, Tec signaling regulates platelet activation through GPVI resulting in only a mild deficit in collagen-mediated activation. Mouse platelets deficient in both Btk and Tec, however, are significantly inhibited and fail to undergo Ca²⁺ mobilization, aggregation, granular secretion, and spreading in response to collagen or the collagen analog, collagen related peptide (CRP), though they continue to aggregate normally in response to adenosine diphosphate (ADP).[18] This effect appears to be dose dependent, [19, 20] and has been confirmed clinically, as patients on ibrutinib have reductions in collagen-mediated platelet aggregation that correlates with the occurrence of clinical bleeding.[20–22] The effect was found to be partially reversed after holding the drug for 2.5 days and fully reversed after one week.[20, 21] Interestingly, there are several reports of individuals with congenital or acquired defects in GPVI, many of whom have a similar bleeding phenotype to patients on ibrutinib.[23].

Fig. 1.

The effects of ibrutinib on platelets. ADP, adenosine diphosphate; Btk, Bruton’s tyrosine kinase; Tec, tyrosine kinase expressed in hepatocellular carcinoma; TP, Thromboxane A2 receptor; VWF, von Willebrand factor.

Ibrutinib also inhibits downstream signaling of CLEC-2, a platelet transmembrane receptor reliant on Btk and Tec signaling.[24] CLEC-2 has been shown to mediate thrombus stability after platelet adhesion to damaged endothelium.[25] Mouse models show that ibrutinib abolishes platelet aggregation in the setting of CLEC-2 stimulation by the exogenous agonist rhodocytin.[26] The only known endogenous ligand of CLEC-2, podoplanin, is notably absent in vascular endothelial cells and platelets, yet is abundant in the lymphatic system.[27] Along these lines, podoplanin has been shown to be highly expressed in type I pneumocytes, renal podocytes and lymphatic endothelial cells, and podoplanin-platelet interactions have been shown to be required for the maintenance of high endothelial venule integrity.[28] Mice deficient in CLEC-2 signaling proteins developed disorganized lymphatics and petechia in utero.[27] It is unknown what effects ibrutinib-mediated inhibition of CLEC-2 signaling might have on pulmonary, renal or lymphatic systems in humans.

Besides the GPVI and CLEC-2-mediated pathways, ibrutinib also interferes with platelet GPIb-mediated platelet functions. Btk has been found to be essential in Von Willebrand factor (vWF)-induced signaling and GPIb-dependent thrombus formation in vivo.[29] GPIb binds to vWF, effectively tethering platelets to injured vessel walls, and stimulates multiple platelet pathways and platelet cytoskeletal reorganization.[30] In vitro data found that pre-incubation of blood from healthy donors with ibrutinib decreased the firm platelet adhesion with vWF under high shear stress while sparing platelet rolling and expression of GPIb.[20] This effect has been correlated clinically, as platelets from patients on ibrutinib with a bleeding phenotype minimally adhered to vWF under flow compared with patients with no bleeding symptoms.[20] It has been suggested that the inhibitory effect of ibrutinib on vWF-GPIb interactions may partially explain the clinical phenotype of bleeding in the microvasculature where shear stress is elevated.[20]

Lastly, in vitro experiments have suggested that ibrutinib also inhibits platelet adhesion to fibrinogen. Ordinarily, binding of fibrinogen to integrin αIIbβ3 promotes platelet adhesion, spreading and clot retraction by evoking outside-in signaling as positive feedback for platelet activation.[31]. In vitro data suggests ibrutinib inhibits the αIIbβ3 outside-in signaling pathway, which has been shown to involve Btk.[32] It has recently been shown that irreversible inhibition of Btk with two ibrutinib analogs in vitro decreased human platelet activation, phosphorylation of Btk, P-selectin exposure, spreading on fibrinogen, and aggregation under shear flow conditions.[33] Moreover, short-term studies of ibrutinib analogs administered to non-human primates also showed abrogation of platelet aggregation in vitro.[33]

Important questions include why some patients do not bleed on ibrutinib and whether we can predict who is at risk for bleeding. Several factors may explain the variability of reported bleeding events, including differences in dosing, disease type, use of other antiplatelet and anticoagulant agents, severity of thrombocytopenia, and across-trial variability in the reporting of bleeding events. Also plausible are individual differences in platelet responsiveness to ibrutinib. Small studies have found that the degree of inhibition of either collagen- or ristocetin-mediated platelet aggregation in patients on ibrutinib correlates with bleeding.[20, 21, 34] Variability in disease-associated platelet defects may also play a role. Patients with CLL, for instance, often have varying degrees of mild platelet dysfunction which is likely exacerbated by ibrutinib. Some authors have suggested that interactions between potential baseline platelet abnormalities and ibrutinib-related effects may determine the risk of bleeding.[22]

One study compared markers of coagulation in patients with CLL at baseline and directly after starting ibrutinib, including platelet function assays (PFA-110™), vWF activity and antigen, factor VIII (FVIII) levels, and platelet aggregation studies.[22] Increased baseline PFA-100 Epinephrine closure time was predictive of bleeding. Baseline levels of FVIII and vWF activity were generally high in CLL patients, and subsequently decreased on ibrutinib. Baseline levels of FVIII and vWF activity in the lower range of normal were also predictive of bleeding. The authors suggested that a pro-inflammatory state associated with CLL may drive increased levels of FVIII and vWF, with baseline elevations conferring some protection against bleeding.[22]

Does ibrutinib have potential cardiovascular protective effects?

While ibrutinib is not an ideal agent for cardiovascular protection given its off-target and side effects, the question of what cardiovascular benefit it offers is relevant to patients already on antiplatelet agents planning to start ibrutinib. The possibility to safely discontinue antiplatelet agents in these patients without increasing their risk of cardiovascular events is appealing and would likely result in less bleeding. Current use of antiplatelet agents in patients not on ibrutinib include aspirin or P2Y12-antagonist (clopidogrel, prasugrel, ticagrelor) monotherapy for primary and secondary prevention of cardiovascular disease, and dual antiplatelet therapy (DAPT) in the settings of acute stroke, myocardial infarction, or coronary artery stenting. [35] The true risks and benefits of stopping antiplatelet agents in patients taking ibrutinib at high risk of cardiovascular events is unknown, though suggestions of its potential cardioprotective efficacy can be drawn from studies targeting the specific platelet receptor pathways affected by ibrutinib.

Experimental models of thrombosis and atherosclerosis have suggested that GPVI blockade may reduce thrombotic risk. GPVI-deficient animal models suggest protection against platelet adhesion, thrombus propagation, leukocyte recruitment and neointimal hyperplasia after mechanical arterial injury.[36, 37] [38] Subsequent animal models have shown that GPVI deficiency significantly impairs thrombus formation caused by ultrasound or needle-induced plaque rupture [39, 40] and reduces neointimal proliferation after carotid artery injury.[41, 42] Lastly, antibody blockade of GPVI inhibits plaque-induced thrombus formation.[43, 44] Several drugs including direct blocking molecules, antibodies, and signal transduction blockers against GPVI have been explored, though none have been evaluated for safety or efficacy in humans to date.[45] As Btk and Tec inhibition affects multiple platelet signaling pathways, ibrutinib cannot be directly compared to agents that inhibit the GPVI receptor alone, particularly in terms of safety. The impact on the GPVI pathway of newer inhibitors more specific to Btk is also unclear. Alternative BTK inhibitors are discussed separately below.

Inhibiting the vWF-GPIb axis is less well-studied, but has been suggested as a potential antithrombotic target in vascular disease. [46]. Animal models show that arterial thrombus formation is inhibited by antibodies targeting the GPIb. [47, 48] Human studies have evaluated inhibition of the vWF-GPIb interaction as a potential therapy for thrombotic thrombocytopenic purpura, though no studies have examined GPIb inhibition as a potential target for cardiovascular risk reduction. [49]

Despite both these pathophysiologic suggestions and animal data, there is insufficient data addressing ibrutinib’s efficacy or safety for primary or secondary cardiovascular risk reduction. With the collection of more clinical data, it may be possible to better clarify the role of discontinuing other cardiovascular protective agents in patients taking ibrutinib.

Clinical management of ibrutinib in patients at risk for bleeding, and the management of ibrutinib-associated bleeding, based on the current body of evidence.

Ibrutinib is generally used in an older patient population with inherently higher cardiovascular and bleeding risks. These risks can be mitigated with forethought and an understanding of the pathophysiology of ibrutinib-associated bleeding. In the following sections, we outline risk reduction strategies to prevent bleeding in patients with indications for antiplatelet agents, anticoagulation, pain management needs, and atrial fibrillation, based on the current understandings of the pathophysiology and risks of ibrutinib-associated bleeding. The evidence behind these recommendations is limited by the currently available data and emphasizes the need for further investigation.

Management with antiplatelet agents.

Many patients treated with ibrutinib have a history of, or risk factors for, cardiovascular disease, warranting antiplatelet therapy for primary or secondary prevention. Likewise, patients may have cardiovascular events during therapy requiring coronary artery stenting and DAPT for extended periods. In making decisions about antiplatelet therapy concurrently with ibrutinib, particularly for primary or secondary prevention of cardiovascular events, it is important to consider patients’ overall life expectancy along with their bleeding and cardiovascular risks. While some patients may have very indolent disease, patients with a short life expectancy are unlikely to benefit from primary risk reduction, and may already be at a sufficiently decreased risk on ibrutinib.

Reported clinical trial data suggest that the addition of antiplatelet agents to ibrutinib increases the risk of bleeding.[7] DAPT with aspirin and a P2Y₁₂-antagonist raises the risk of major bleeding by 40–50% compared to single antiplatelet therapy;[50] adding ibrutinib to DAPT is therefore likely to further increase this risk. Indeed, in vitro data has suggested ibrutinib combined with P2Y₁₂ antagonists has an additive antiplatelet effect.[31] For these reasons, we are cautious about concurrent use of ibrutinib with other antiplatelet agents. Based on the available data, we recommend the following in patients taking ibrutinib:

1. Patients should be cautioned against the use of NSAIDs, fish oils, vitamin E, and inadvertent use of aspirin-containing products.

2. Consider stopping aspirin in patients on ibrutinib who have low or moderate cardiovascular risk.

3. For patients at high cardiovascular risk that may compromise their survival, including those with recent MI or stroke, consider ibrutinib plus 81 mg of aspirin. We recommend against higher doses of aspirin in light of data suggesting increased bleeding with no benefit.[51]

4. For patients with recent bare metal cardiac stent placement, consider delaying or holding ibrutinib while on DAPT. After the required DAPT period, consider ibrutinib plus 81 mg of aspirin.

5. For patients with recent drug eluting stent placement, consider replacing ibrutinib with an alternative treatment strategy given the extended duration of recommended DAPT.

6. Some authors initially trial ibrutinib at a lower dose (280 mg/day) in patients on other antiplatelet agents or anticoagulants and slowly increase to treatment dose if bleeding does not occur. This dose escalation strategy is based on in vitro data suggesting that the antiplatelet effects of ibrutinib are dose-dependent.[19, 20] It is important to note, however, that there is no clinical data to endorse this practical strategy, and that studies have shown significant bleeding rates at both lower (420mg/day)[7, 10–12] and higher (560mg/day)[8, 9] doses of ibrutinib.

Management with anticoagulants:

Ibrutinib has been associated with unacceptable bleeding rates in combination with vitamin K antagonists and thus should not be given to patients on warfarin.[7, 11] Other anticoagulants were allowed in clinical trials and are used concurrently with ibrutinib in some studies.[14] DOACs have been shown to cause fewer bleeding events than warfarin in multiple phase III trials, and are likely a safer class of anticoagulant to combine with ibrutinib.[52] However, current data regarding the combination of DOACs and ibrutinib is insufficient to draw strong conclusions on the associated bleeding risk, which is presumed higher than either agent alone. The risks and benefits of anticoagulation should be considered on a case by case basis and relayed to the patient. It is important to use the appropriate duration of anticoagulation and to avoid anticoagulation in combination with ibrutinib for extended periods without an appropriate indication. Contemporary guidelines recommend three months of anticoagulation for patients with venous thromboembolism provoked by surgery or an alternative transient risk factor, and consideration of indefinite anticoagulation for patients with unprovoked thrombosis who are not at a high risk of bleeding.[53]

When considering the combination of ibrutinib and DOACs, it is also important to note that ibrutinib primarily undergoes CYP3A4-mediated metabolism, and CYP3A4 inhibitors will increase ibrutinib exposure.[54] Rivaroxaban and apixaban in particular undergo CYP3A4-mediated metabolism (33% and 25%, respectively), while dabigatran and edoxaban do not.[55] It is not known how the presence of both substrates (ibrutinib and a CYP3A4-metabolized DOAC) affects their respective metabolism, particularly as DOAC levels are not routinely monitored. Clinicians should be conscious of medications which impact the CYP3A4 pathway in patients taking ibrutinib, especially in those taking ibrutinib and DOACs concurrently. Importantly, concurrent administration of additional CYP3A4 inhibitors (e.g. fluconazole) will likely increase plasma concentrations of both drugs, potentially leading to a significantly increased risk of bleeding.

For patients potentially requiring ibrutinib concurrent with anticoagulation we propose the following:

1. For patients who require short courses of anticoagulation for a provoked VTE, consider postponing ibrutinib therapy if feasible and using an alternative agent in the interim. If overlapping with ibrutinib, we recommend using a DOAC such as apixaban given the smaller bleeding rates as compared to warfarin.[56] Avoid concurrent antiplatelet therapy unless there is a strong indication.

2. For patients requiring extended anticoagulation for unprovoked VTE, consider using another agent or using a DOAC such as apixaban, again avoiding other antiplatelet agents unless strongly indicated. Half-dose apixaban can be safely used after 6 months of therapy, likely decreasing bleeding risk.[57]

3. Patients should be cautioned against the use of NSAIDs, fish oils, vitamin E, and inadvertent use of aspirin-containing products.

4. Some authors initially trial ibrutinib at a lower dose (280 mg/day) in patients on other antiplatelet or anticoagulants and slowly increase to treatment dose if bleeding does not occur. This dose escalation recommendation is based on in vitro data which suggest that the antiplatelet effects of ibrutinib are dose dependent.[19, 20] It is important to note however that there is no clinical data to confirm the usefulness of this practice, and that studies have shown significant bleeding rates regardless of lower (420mg/day)[9–12] or higher (560mg/day)[7, 8] treatment doses.

Management of bleeding

Low-grade bleeding can be managed with supportive care or by holding ibrutinib for a short period of time. For high grade (III-IV) bleeding, we recommend the following:

Non-CNS bleeding:

For patients admitted to the hospital for non-CNS bleeding, or those requiring transfusions, we recommend stopping ibrutinib and transfusing platelets. In vitro data suggests that transfusion of platelets to achieve 50% fresh platelets should correct hemostasis.[20] The time since the last dose of ibrutinib should be considered if transfusing platelets, as the drug may still inhibit platelets within the initial 3–4 hour half-life.[21] If holding ibrutinib alone, the antiplatelet effects should be partially reversed after 2.5 days.[20, 21]

CNS Bleeding:

Although there have been reports of ibrutinib-associated CNS bleeding effectively treated with platelet transfusion, [58] recent data suggest that platelet transfusion may have some detrimental effects which must be considered. In the PATCH trial, investigators compared platelet transfusion verses supportive care in patients who presented with spontaneous antiplatelet agent-associated intraparenchyma bleeding. The platelet transfusion group, which was mostly composed of patients taking aspirin alone, was significantly more likely to have adverse advents, and had a significantly higher mortality (adjusted common odds ratio 2·05, 95% CI 1·18–3·56; p=0·0114).[59] The authors concluded that platelet transfusion cannot be recommended in clinical practice as a hemostatic measure in patients with intraparenchyma hemorrhage taking antiplatelet agents.[59] Although no explanation is readily available, the data is compelling, and since the publication of this study it has been our practice to not offer platelet transfusion to patients with antiplatelet agent-associated CNS bleeding. There is however, no data to determine if this finding holds true in patients with ibrutinib-associated CNS bleeding. Ibrutinib has distinctly different antiplatelet effects and bleeding rates compared to other antiplatelet agents, and as such, the risk/benefit of transfusion in the setting of ibrutinib-associated CNS bleeding is uncertain. We recommend the decision to transfuse platelets in this setting be individualized on a case by case basis. No strong recommendations can be made due to a lack of specific data.

Management of elective and unplanned procedures.

Ibrutinib should be held for 7 days prior to major procedures, as the antiplatelet effects of ibrutinib have been shown to be fully reversed after a week off therapy.[20, 21] Ibrutinib can be restarted 1–3 days postoperatively.[11, 13] The risk of bleeding during urgent, unplanned procedures can be mitigated with platelet transfusion to achieve 50% fresh platelets.[20] The time since the last dose of ibrutinib should be considered if transfusing platelets, as outlined above.[21]

Management of atrial fibrillation

Though ibrutinib may have cardioprotective effects via platelet inhibition, an additional, less-desirable side effect is an increased risk of atrial fibrillation. This was shown in a systematic review and meta-analysis which found a pooled relative risk of 3.5 (95% CI 1.8–6.9, p<0.0001) for developing atrial fibrillation in those treated with ibrutinib versus alternative therapies.[2] Experimental mouse models have suggested a potential culprit for this association: the phosphoinositide 3-kinase (PI3K)-Akt pathway, critical for protecting cardiac myocytes under stress and itself closely tied to Btk and Tec pathways.[60] This risk appears to be ongoing for as long as patients take the drug, with rates up to 16% reported after 2 years of therapy with ibrutinib.[61]

Atrial fibrillation in patients taking ibrutinib poses a unique challenge in balancing thrombosis and bleeding risks. A major cause of morbidity and mortality in patients with atrial fibrillation is cardioembolic stroke, the risk of which can be estimated by models such as the widely-used CHA2DS2VASc system.[62] Contemporary guidelines recommend therapeutic anticoagulation for individuals at sufficiently elevated stroke risk (CHA2DS2VASc score of 2 or greater).[62] Individuals with a CHA2DS2VASc score of zero can safely omit antithrombotic therapy, while those with a CHA2DS2VASc score of 1 may be managed with no antithrombotic therapy, anticoagulation, or aspirin, based on physician and patient preference.[62]

In patients with atrial fibrillation receiving ibrutinib, however, the cumulative bleeding risk from both antiplatelet and anticoagulant effects can be dangerously high. As noted above, warfarin has been associated with serious bleeding prompting exclusion of patients taking this drug from subsequent trials.[7, 11, 22] Clinical data examining the safety of ibrutinib taken in conjunction with DOACs is lacking, though small studies have not reported excess bleeding when using these agents together; as previously noted, the number of patients on DOACs in these trials was limited.[14] In addition, it is currently unknown whether the multiple antiplatelet effects of ibrutinib itself confer any meaningful protection against cardioembolic stroke in atrial fibrillation. Statistically significant reductions in stroke have been found by adding clopidogrel to aspirin, speaking to the additive effects of inhibiting multiple platelet signaling pathways;[63] however, this combination is still inferior to anticoagulation with warfarin alone.[64] Additional studies are necessary to clarify how ibrutinib might fit into atrial fibrillation stroke prophylaxis strategies.

We recommend the following management of patient on ibrutinib who develop atrial fibrillation.

1. If patents have a high bleeding risk, including mandated use of DAPT, or a history of significant bleeding, consider using an alternative agent to ibrutinib and managing their atrial fibrillation per conventional guidelines.

2. In patients who are to continue ibrutinib and have a sufficiently elevated stroke risk per standard models (CHA₂DS₂VASc score of 2 or greater), we recommend a DOAC be used as opposed to a vitamin K antagonist, without additional antiplatelet agents. We at our institution, as well as our colleagues across the United States, have successfully combined ibrutinib with DOACs in patients who develop atrial fibrillation. As discussed above, in this setting we use a dose escalation strategy for ibrutinib (often starting at 280 mg/day), slowly increasing to standard treatment doses if bleeding does not occur.

3. For patients at low risk for cardioembolic stroke (CHA₂DS₂VASc score of zero), we recommend continuing ibrutinib alone without the addition of anticoagulation or alternative antiplatelet agents.

4. For patients at intermediate risk for cardioembolic stroke (CHA2DS2VASc score of one), we recommend continuing ibrutinib alone, or ibrutinib plus aspirin, based on patient-physician preferences and consideration of individualized risks.

5. In patients requiring DAPT (e.g. those with recent coronary artery stenting) who develop atrial fibrillation and have a high stroke risk, we recommend discontinuing ibrutinib due to the significant bleeding risk and switching to alternative therapy for lymphoproliferative disorders.

6. In patients with an indication for aspirin for cardioprotective benefit who develop atrial fibrillation with elevated stroke risk (CHA₂DS₂VASc score of 2 or greater), we recommend treating with a DOAC plus ibrutinib alone and stopping other antiplatelet agents. This is based on prospective data showing that single antiplatelet therapy with anticoagulation had equal efficacy and superior safety to dual antiplatelet therapy with anticoagulation.[65]

Pain management on ibrutinib

Ibrutinib’s antiplatelet effects also demand careful consideration when addressing patient’s pain symptoms. NSAIDs inhibit platelet function through non-selective inhibition of the COX-1 enzyme constitutive in platelets, leading to reduced synthesis of the platelet-stimulatory molecule thromboxane A2.[66] Use of non-selective NSAIDs in conjunction with ibrutinib has been shown to increase the risk of both minor and major bleeding and should therefore be avoided.[14, 22] Celecoxib, the only available COX-2 selective NSAID, has been shown to have no effect on platelet function, and therefore should not further increase the risk of bleeding associated with ibrutinib.[67] Likewise, meloxicam has minimal to no effects on platelets.[68] Selective serotonin reuptake inhibitors (SSRI) do have some documented antiplatelet effects and increased risk for bleeding;[69] via similar mechanisms, serotonin-norepinephrine reuptake inhibitors (SNRIs) used for neuropathic pain may be just as likely to cause bleeding, although current clinical data is conflicting.[70] Acceptable analgesics with no major adverse platelet effects include acetaminophen and opioids, as well as anti-epileptic drugs such as gabapentin, pregabalin, and tricyclic antidepressants.

Are providers conscious of bleeding risk?

There is limited data describing physician and patient awareness of the bleeding risks associated with ibrutinib. One group performed a medication review of 96 CLL patients beginning ibrutinib and found that 64% of the patients were concurrently taking medications that increased risk of ibrutinib toxicity, including CYP3A4 inhibitors carbamazepine, rifampin, and rifabutin.[71] When beginning ibrutinib, 9% of patients were taking an anticoagulant and 30% were on aspirin.[71] The data showed that 10% of the patients had clinically significant bleeding, with 4% (4 patients: 1 on enoxaparin, 2 on a SSRI, 1 on a NSAID) requiring hospitalization.[71]

Future Btk inhibitors

Newer Btk inhibitors are more selective for Btk with less off-target kinase inhibition and may reduce bleeding as compared to ibrutinib.[72] Phase 1 data for ONO/GS-4059, a “second generation” Btk inhibitor, shows clinical activity in CLL and non-Hodgkin lymphoma.[73] This study did not exclude patients based on anticoagulant use, and the use of anticoagulant therapy was not associated with increased risk of bleeding for the 31.1% of patients receiving anticoagulation.[73] Of these 28 patients who were receiving anticoagulation, 18 were treated with only prophylactic doses, and 5 were being anticoagulated for atrial fibrillation (4 of 5 cases were pre-existing before initiation of the study drug). Grade III-IV hematologic toxicities recovered spontaneously and included neutropenia in 10%, anemia in 13.3%, and thrombocytopenia in 13.3%.[73] One episode of grade III hemorrhage was reported (psoas hematoma) in a patient who did not receive anticoagulants, resulting in withdrawal from the trial. [73]

Acalabrutinib is another novel second-generation Btk inhibitor with increased selectivity for Btk. Whereas ibrutinib completely suppressed Tec activity, an equivalent dose of acalbrutinib had minimal effect in vitro on Tec.[74] Btk suppression alone is not likely to lead to a strong bleeding phenotype, just as individuals with XLA generally do not bleed. Assessment of in vivo platelet function in samples from human subjects treated with acalabrutinib showed no reduction in platelet-vessel wall interactions, in contrast to control patients treated with ibrutinib.[74] In a phase I-II study of acalabrutinib in 61 relapsed CLL patients that excluded patients requiring warfarin therapy but allowed other anticoagulants, no major hemorrhage was noted. However, Grade 1–2 petechia were noted in 16% of patients, and contusion in 18%, which in summation represents low grade bleeding not dissimilar to that reported with ibrutinib. [74]

More data is needed for the second-generation Btk inhibitors (acalabrutinib, ONO/GS-4059, BGB-3111) to understand their long-term safety and efficacy. A phase III trial comparing acalabrutinib with ibrutinib is underway for high-risk CLL patients.[75]

Conclusion

Ibrutinib is an effective agent for the treatment of several lymphoproliferative disorders but results in increased rates of low and high-grade bleeding through several platelet inhibitory mechanisms. With careful planning and recognition, these risks can be mitigated. While data is lacking to accurately define the risks of combining ibrutinib with other antiplatelet agents and anticoagulants, educated risk assessment can be made with the current body of knowledge. Future studies are needed to determine the appropriate role of other antiplatelet agents in combination with ibrutinib and the safest means of combining the drug with anticoagulants.