Antiplatelet and Wound Healing Implications of Immunotherapy and Targeted Cancer Therapies in the Perioperative Period

The objective of this narrative review is to identify data regarding the hematologic and wound healing effects of targeted cancer therapy drug classes and the most frequently used individual agents, including guidelines for cessation before and after procedural intervention. Relevant literature was obtained by searching the PubMed database. The prescribing information (“package inserts”) of targeted cancer therapy agents was also reviewed.^2–19 

Bruton’s tyrosine kinase plays an integral role in the B-cell receptor signaling pathway, which is critical to the maturation of B-cell lymphocytes. Once the B-cell receptor is engaged by an antigen at the cell surface, Bruton’s tyrosine kinase is recruited to the cell membrane and activates other kinases, leading to the activation of nuclear factor–kB and transcription of genes involved in B-cell migration, proliferation, and survival.²⁰  In certain B-cell malignancies, this pathway can become aberrantly active, resulting in unchecked B-cell proliferation, which makes it an attractive therapeutic target. Bruton’s tyrosine kinase inhibitors have been associated with a dramatic change in the treatment of chronic lymphocytic leukemia and other B-cell malignancies. Ibrutinib (Imbruvica, Pharmacyclics, USA), a first-generation Bruton’s tyrosine kinase inhibitor, covalently binds to Bruton’s tyrosine kinase, irreversibly inhibiting the downstream signaling cascade. In in vitro models of chronic lymphocytic leukemia, Bruton’s tyrosine kinase inhibition has been shown to reduce cell migration, proliferation, and survival, as well as induce apoptosis.²¹ 

After its success in clinical trials, ibrutinib received Food and Drug Administration (Silver Spring, Maryland) approval for treatment of mantle cell lymphoma in 2013, with subsequent approval for chronic lymphocytic leukemia, marginal zone lymphoma, and Waldenström’s macroglobulinemia. However, ibrutinib is associated with a toxicity profile including hemorrhage, hypertension, and cardiac arrhythmias that can lead to drug discontinuation in up to 41% of patients.²²  Second-generation Bruton’s tyrosine kinase inhibitors were developed to bind more specifically to Bruton’s tyrosine kinase and reduce adverse effects. Acalabrutinib (Calquence, AstraZeneca Pharmaceuticals, USA) and zanubrutinib (Brukinsa, BeiGene, USA), the two Food and Drug Administration–approved second-generation Bruton’s tyrosine kinase inhibitors currently available, have a reduced incidence of bleeding, atrial fibrillation, and hypertension.^23,24  In early 2023, a third-generation Bruton’s tyrosine kinase inhibitor, pirtobrutinib (Jaypirca, Eli Lilly and Company, USA), was approved for relapsed and refractory mantle cell lymphoma, although it carries similar risks of bleeding, atrial fibrillation, and hypertension as second-generation inhibitors.²⁵ 

Hemorrhage is one of the most serious and pertinent perioperative side effects of Bruton’s tyrosine kinase inhibition. Bleeding events attributed to Bruton’s tyrosine kinase inhibitors can range in severity from minor mucocutaneous bleeding to life-threatening central nervous system hemorrhage. Bleeding events are graded in severity using the Common Terminology Criteria for Adverse Events scale with major bleeding defined as Grade 3 (severe or medically significant but not immediately life-threatening; hospitalization or prolongation of hospitalization indicated), Grade 4 (life-threatening consequences; urgent intervention indicated), or Grade 5 (death related to adverse event).²⁶  In trials, ibrutinib was associated with bleeding events in 51% of patients, with major bleeding events (common terminology criteria for adverse events 3 or greater) or any central nervous system bleeding occurring in 5%.²⁴  Despite improvements, a significant bleeding risk still exists with the use of second- and third-generation Bruton’s tyrosine kinase inhibitors, including major bleeding events in 4.5% of patients on acalabrutinib,²⁵  6% on zanubrutinib, and 3% on pirtobritinib.^23–25 

The mechanism underlying the increased risk of bleeding with Bruton’s tyrosine kinase inhibition is drug-induced platelet dysfunction. More specifically, ibrutinib has been shown to inhibit collagen-mediated platelet aggregation, leading to an unstable thrombus formation.²⁷  This is believed to be related to on- and off-target inhibition of other protein kinases, including Tec and Src family kinase. Bruton’s tyrosine kinase, Tec, and Src family kinase function in the downstream signaling pathway of platelet collagen receptor glycoprotein VI, leading to platelet activation induced by collagen exposure. Ibrutinib inhibition of these protein kinases leads to decreased platelet aggregation as depicted in figure 1.^28,29  Although acalabrutinib has also been found to inhibit Tec in in vitro studies, thrombus formation was unaffected. An additional study found that acalabrutinib did not inhibit Src family kinase. Thus preserved Src family kinase function may be the key distinction between ibrutinib and acalabrutinib in maintaining stable thrombus formation and the basis behind the relative decrease in bleeding risk with second-generation Bruton’s tyrosine kinase inhibitors³⁰  (fig. 2). Pirtobrutinib has a unique binding site that makes it highly selective for Bruton’s tyrosine kinase. It does not inhibit Tec or Src family kinase, which decreases the off-target bleeding risk to an even greater extent as compared to other Bruton’s tyrosine kinase inhibitors.²⁵ 

The National Comprehensive Cancer Network (Plymouth Meeting, Pennsylvania) guidelines recommend holding Bruton’s tyrosine kinase inhibitors perioperatively due to their bleeding risk.³¹  For major procedures, these agents should be held for 7 days before and 7 days after the procedure, as the antiplatelet effects are felt to be fully reversed within 1 week of cessation. For minor procedures, it is recommended to hold these medications for 3 days before and 3 days after the procedure. For emergent or unplanned procedures, bleeding risk can be mitigated with platelet transfusion.³² Ex vivo data also show that von Willebrand factor and Factor VIII transfusion may enhance platelet function in patients treated with Bruton’s tyrosine kinase inhibitors.³⁰ 

For patients with chronic lymphocytic leukemia who have significant oncologic disease burden or aggressive disease, there is risk of inciting a disease flare if the Bruton’s tyrosine kinase inhibitor is held shortly after initiating therapy.³³  This risk is mitigated with a longer duration of Bruton’s tyrosine kinase inhibitor therapy, a reflection of better disease control. Consequently, elective surgical procedures should be scheduled before initiating Bruton’s tyrosine kinase inhibitors or after the patient has been on therapy for more than 6 to 12 months.³⁴ 

Angiogenesis, the sprouting of new blood vessels from existing vasculature, is one of the hallmarks of cancer. This constitutively active process supports the rapid growth of tumors by supporting delivery of nutrients vital for cell growth.³⁵  The vascular endothelial growth factor signaling pathway drives tumor angiogenesis, and targeting this pathway is a key therapeutic strategy. Through its inhibition, the tumor can be effectively starved of nutrients, thereby retarding its growth. Vascular endothelial growth factor A (commonly referred to as vascular endothelial growth factor) is a growth factor that activates the angiogenesis pathway and is overexpressed in most solid tumors, making it a primary antineoplastic target. The most well-known vascular endothelial growth factor inhibitor is the monoclonal anti–vascular endothelial growth factor antibody bevacizumab (Avastin, Genentech Inc., USA), which binds to vascular endothelial growth factor and blocks its binding to vascular endothelial growth factor receptors on endothelial cells. Bevacizumab also decreases vascular permeability and permits more optimal delivery of cytotoxic chemotherapy to cancer cells.³⁶  Ramucirumab (Cyramza, Eli Lilly, USA) also works on this pathway, binding to vascular endothelial growth factor receptor 2, blocking its activation. Additionally, there are other multitargeted tyrosine kinase inhibitors that target vascular endothelial growth factor receptors and other antineoplastic tyrosine kinases, including sorafenib (Nexavar, Bayer Healthcare Pharmaceuticals, USA), sunitinib (Sutent, Pfizer Labs, USA), regorafenib (Stivarga, Bayer Healthcare Pharmaceuticals), axitinib (Inlyta, Pfizer Labs), lenvatinib (Lenvima, Eisai, USA), pazopanib (Votrient, Novartis, Switzerland), ponatinib (Iclusig, ARIAD Pharmaceuticals, USA), tivozanib (Fotivda, AVEO pharmaceuticals, USA), vandetanib (Caprelsa, Sanofi, France), ziv-aflibercept (Zaltrap, Sanofi, France), and cabozantinib (Cabometyx, Exelixis, USA; Cometriq, Exelixis).³⁷ 

While the development of vascular endothelial growth factor inhibitors has improved outcomes in many solid tumors, disruption of the vascular endothelial growth factor pathway may also lead to serious toxicities, including in the perioperative setting. Vascular endothelial growth factor signaling has been found to promote both prothrombotic and antithrombotic pathways; thus, the use of vascular endothelial growth factor inhibitors have been associated with both increased risk of bleeding and increased risk of thrombosis.

A large meta-analysis reviewed bleeding events for bevacizumab and ramucirumab in clinical trials and found that 3.5% and 1% of patients, respectively, experienced severe bleeding outside the perioperative and periprocedural settings.³⁸  Among tyrosine kinase inhibitors, bleeding is also of concern, with rates of mild to moderate bleeding reported in up to 20% of patients.³⁶  Despite these data, there are limited guidelines on tyrosine kinase inhibitor cessation to mitigate bleeding risk during the perioperative or periprocedural period, and recommendations for cessation are centered on the wound healing issues, as discussed in the next section.

Vascular endothelial growth factor is involved in both pro- and antithrombotic signaling pathways, and therefore, vascular endothelial growth factor inhibitors can exert seemingly contradictory effects on hemostasis (fig. 3). The mechanism by which vascular endothelial growth factor inhibitors promote hemorrhage is likely multifaceted and remains relatively unelucidated. In vitro studies have shown that vascular endothelial growth factor increases the activity of the prothrombotic plasminogen activator inhibitor 1, tissue factor, and von Willebrand factor.^39,40  As all these proteins are involved in promoting coagulation, vascular endothelial growth factor inhibition may be associated with increased risk of hemorrhage.

However, the prothrombotic effects of vascular endothelial growth factor inhibitors have also been associated with clinically significant thromboembolic disease including venous thromboembolism in approximately 12% of patients on bevacizumab and 2 to 6% of patients treated with tyrosine kinase inhibitors.^41,42  Rates of arterial thromboembolism are lower, with the incidence ranging from 1 to 5%.⁴¹  Arterial thromboembolism is primarily associated with bevacizumab use and only rarely reported with tyrosine kinase inhibitors.³⁶  The mechanism of thrombosis associated with vascular endothelial growth factor inhibitors is complex, including mediation through antiplatelet effects. Under normal physiologic conditions, vascular endothelial growth factor prevents platelet aggregation by upregulating downstream pathway mediators, NO, and prostacyclin. In platelets, NO decreases platelet activation and aggregation by acting upon the thromboxane A2 receptor. Prostacyclin is also thought to have an antagonistic effect directly on thromboxane A2, thereby decreasing platelet activation. Furthermore, vascular endothelial growth factor leads to increased expression of other antithrombotic mediators including tissue plasminogen activator and urokinase and increases urokinase receptor activity.⁴³  Vascular endothelial growth factor inhibitors, therefore, promote platelet aggregation and thrombosis by reversing the constitutional processes of vascular endothelial growth factor. Additionally, bevacizumab immune complexes may directly bind the Fc gamma receptors on platelets and promote platelet activation⁴⁴  (fig. 3).

Inhibition of angiogenesis by vascular endothelial growth factor inhibitors has been associated with impaired wound healing, and the preoperative use of tyrosine kinase inhibitors has been associated with impaired bowel anastomotic healing.^45,46  In the formation of granulation tissue, angiogenesis and capillary growth into the wound provide a critical conduit for nutrients to promote healing.⁴⁷  Vascular endothelial growth factor plays a role in multiple steps of wound healing angiogenesis, including vasodilation via the production of NO, basement membrane degradation, endothelial cell migration, and proliferation⁴⁵  (fig. 3).

Given the risks of bleeding, thrombosis, and impaired wound healing with vascular endothelial growth factor inhibitors perioperatively, surgical procedures need to be carefully planned. This is of clinical relevance because the oncologic treatment plan for many solid tumors in which bevacizumab and other vascular endothelial growth factor inhibitors are used includes surgery. For example, the resection of synchronous hepatic disease in oligometastatic colorectal cancer may include neoadjuvant and/or adjuvant treatment with chemotherapy plus bevacizumab.⁴⁸  In a study pooling data from two randomized clinical trials evaluating metastatic colorectal patients treated with bevacizumab, Scappaticci et al. assessed postoperative wound healing complications in patients before or during treatment with bevacizumab. They found wound healing complications were increased in patients who had major surgery during bevacizumab therapy compared to a control group (13% vs. 3.4%). This risk was mitigated if bevacizumab was administered 28 to 60 days after surgery, with wound healing complications occurring in 1.3% of bevacizumab-treated patients and 0.5% of the control group.⁴⁹  With a half-life of 20 days, a 4- to 8-week interval between cessation of bevacizumab and surgery is generally recommended, although this interval varies by site of primary tumor resection. Postoperatively, 4 to 8 weeks is recommended before reinitiation of bevacizumab.^48,50 

Although tyrosine kinase inhibitors have much shorter half-lives than bevacizumab, perioperative complications are still pertinent. Rates of impaired wound healing and thromboembolism are comparatively less than those associated with bevacizumab.⁴⁷  In a large multicenter retrospective study of stage IV renal cell carcinoma patients treated with a tyrosine kinase inhibitor who underwent cytoreductive nephrectomy, perioperative tyrosine kinase inhibitor use was independently associated with a higher risk for postoperative complications within 30 and 90 days of nephrectomy. These complications were broadly categorized; however, they did include vascular and wound healing complications.⁵¹  Currently, there are insufficient data to conclude the optimal delay between tyrosine kinase inhibitor treatment and surgery, but recommendations are provided in table 1 based on package inserts and empiric practice.⁴⁸ 

Immune checkpoint inhibitors are increasingly prevalent as a treatment option for various cancers. Inhibitors of programmed cell death, programmed cell death ligand-1, and cytotoxic T-lymphocyte antigen-4,work by activating T-lymphocytes and stimulating the immune system to target tumor cells.⁵²  The more commonly used Food and Drug Administration–approved immune checkpoint inhibitors include pembrolizumab (Keytruda, Merck, USA), ipilimumab (Yervoy, Bristol Myers Squibb, USA), nivolumab (Opdivo, Bristol Myers Squibb), atezolizumab (Tecentriq, Genentech, USA), cemiplimab (Libtayo, Regeneron, USA), avelumab (Bavencio, Pfizer, USA), and durvalumab (Imfinzi, AstraZeneca, USA).

A systematic review of neoadjuvant immune checkpoint inhibitors in resectable non–small cell lung cancer did not report any postoperative bleeding or wound healing complications.⁵³  Several small, retrospective studies have looked at tumor resection within the time period of immune checkpoint inhibitor administration and did not show impaired wound healing or severe postoperative bleeding. The timeframe of immune checkpoint inhibitor administration was variable and ranged from 1 day to 123 weeks perioperatively and 1 to 74 days postoperatively.^54–58 

There are currently no recommendations to hold or delay immune checkpoint inhibitor administration perioperatively due to hematologic or wound healing concerns. While data regarding these effects are limited, immune checkpoint inhibitors are associated with a broad range of immunotherapy-related adverse effects. In the perioperative period, particular attention should be paid to potential endocrine, cardiac, pulmonary, and gastrointestinal effects, which are beyond the scope of this review. Furthermore, consideration should be given to patients currently experiencing immune-related adverse effects secondary to immune checkpoint inhibitors as these are typically treated with high-dose steroids that carry their own attendant perioperative risk.⁵⁸ 

There are many tyrosine kinase inhibitors that inhibit tyrosine kinases aside from Bruton’s tyrosine kinase and vascular endothelial growth factor. Other targets include BRAF, MEK, FGFR, EGFR, MET, and HER². The perioperative risks of these tyrosine kinase inhibitors are less known with no clear guidance about management perioperatively.

Imatinib, commonly used to treat chronic myeloid leukemia and gastrointestinal stromal tumors, primarily inhibits Bcr-Abl but also inhibits PDGF, stem cell factor, and c-kit. Preoperative imatinib use in gastrointestinal stromal tumor patients has been associated with impaired bowel anastomosis healing and increased risk for gastrointestinal hemorrhage.^50,59  Despite this, the National Comprehensive Cancer Network guidelines state imatinib can be stopped just before surgery and restarted when the patient is able to tolerate oral medications.⁶⁰ 

The neoadjuvant use of the tyrosine kinase inhibitors sunitinib and sorafenib was studied in renal cell cancer surgical cases and was not associated with bleeding or wound healing complications, but increased rates of intraoperative adhesions were observed.⁶¹  There are numerous other case reports and case series of tyrosine kinase inhibitor–associated perioperative morbidity including delayed tracheoesophageal puncture complications in thyroid cancer.⁶²  As there are no guidelines regarding holding other tyrosine kinase inhibitors around the time of surgery, a discussion of risk versus benefit depending on the surgical procedure, the patient’s oncologic history, and assessment of the available literature on the potential complications of the agent in question is warranted.

Additional targeted cancer therapy classes include monoclonal antibodies with varied targets and mechanisms of action. These include cetuximab (Erbitux, Lilly, USA), panitumumab (Vectibix, Amgen, USA), pertuzumab (Perjeta, Genentech, USA), rituximab (Rituxan, Genentech), trastuzumab (Herceptin, Genentech), enfortumab (Padcev, Astellas Pharma, Japan), and daratumumab (Darzalex, Janssen Pharmaceuticals, Belgium). Monoclonal antibodies can also be conjugated to other cytotoxic drugs, such as brentuximab vedotin (Adcetris, Seagen, USA), polatuzumab vedotin (Polivy, Genentech), and inotuzumab (Besponsa, Pfizer).⁶³  An emerging class is Bispecific T-Cell Engager antibodies, which have two binding sites that bring tumor cells into contact with a patient’s own T-cells, allowing the T-cell to exert cytotoxic activity. Blinatumomab (Blincyto, Amgen) is an example of a Bispecific T-Cell Engager used in treatment of acute lymphoblastic leukemia.⁶⁴  There are no reported wound healing or hematologic complications noted with these drugs, and they are generally continued in the perioperative period.

Please see table 1 for a summary of the perioperative and hematologic implications of targeted cancer therapies.

The antiplatelet effects of vascular endothelial growth factor and Bruton’s tyrosine kinase inhibitors are potentially the most important perioperative issues to recognize. Patients are likely to be at risk of increased blood loss during surgery, including severe hemorrhage. Greater attention to vascular access and blood product availability may be warranted when therapy has not been discontinued. Increased bleeding risk with neuraxial anesthesia, deep plexus, or peripheral nerve blocks has not been well-characterized but should be assumed. Notably, current guidelines from the American Society for Regional Anesthesia and Pain Medicine (Pittsburg, Pennsylvania) regarding regional anesthesia in the setting of antithrombotic or thrombolytic therapy do not address targeted cancer therapy agents.⁶⁵  Some targeted cancer therapy agents have recommendations for cessation before surgery, but these recommendations may be related to both hematologic and wound healing effects, without specific consideration of increased risk with neuraxial or regional anesthesia techniques. Guidelines from oncologic societies and package inserts primarily pertain to the risk of hemorrhage and intraoperative blood loss, and the risk of iatrogenic epidural hematoma is not specifically addressed. Although regional anesthesia-related epidural or soft tissue hematoma has not yet been reported in the setting of targeted cancer therapy use, it is certainly a plausible theoretical concern. For example, Bruton’s tyrosine kinase inhibitor surgical recommendations are often arbitrarily divided into “minor” or “major” surgeries (table 1), but it is conceivable that “minor” surgery could be performed under neuraxial anesthesia with no consideration for increased risk.

Furthermore, the combination of targeted cancer therapy agents with other antiplatelet agents, such as aspirin or nonsteroidal anti-inflammatory drugs, given their dual mechanisms of decreasing platelet aggregation, may have an additive effect and further increase bleeding risk. Currently, there is neither literature nor expert consensus to guide management in these situations, but a conservative approach is likely appropriate.

The effects of poor wound healing have far-reaching consequences, and knowledge of this potential complication with targeted cancer therapy use is important to many medical specialties. Wound infection and anastomotic breakdown are associated with significant costs and morbidity related to the need for debridement, infection control, and/or surgical reconstruction.⁶⁶ 

Surgeons and anesthesiologists may easily overlook the importance of appropriate timing of cessation of targeted cancer therapies in the perioperative period, especially considering that the length of cessation for some of these medications is quite prolonged (table 1). Additionally, while simple discontinuation of therapy for the recommended period may be reasonable in some cases, the medical oncologist/hematologist may indicate that prolonged cessation would be associated with an unacceptably high risk of cancer progression. As such, informed discussion and coordination between the surgical team and the treating oncologist is important to optimize perioperative outcomes while mitigating the risks of withholding targeted cancer therapies. Anesthesia preoperative clinics may be uniquely situated to identify both the agents and surgical procedures that place patients at highest risk for compromised wound healing and advocate for careful perioperative planning. Furthermore, given the complexity, varied profiles, and ongoing evolution of these drugs, coordination with a pharmacist with expertise in this field should be considered.

The use of targeted cancer therapies is expanding to encompass more patients and additional cancer types, with extended courses of treatment that may include lifelong administration. Furthermore, new agents are rapidly being developed and will be increasingly encountered in the perioperative period. Many of these agents are unfamiliar to surgical and anesthesia professionals, and while often colloquially referred to as “immunotherapy,” there are important distinctions between the mechanistic classes of these agents that may be overlooked. In particular, vascular endothelial growth factor and Bruton’s tyrosine kinase inhibitors require careful identification and consideration around the time of surgery due to their hematologic (vascular endothelial growth factor and Bruton’s tyrosine kinase inhibitors) and wound healing (vascular endothelial growth factor inhibitors) implications.

Our review has limitations largely related to the rapidly evolving nature of the field and limited data on studying perioperative risks. The bleeding risks of targeted cancer therapies reported in clinical trials include spontaneous bleeding events as well as trauma-related events, but these agents have not been studied specifically in the perioperative or periprocedural environment. As most targeted cancer therapies have been used for less than 10 yr, they have no or limited postmarketing data on procedural risk. Many of the recommended cessation and reinitiation times are obtained from the package inserts, rather than being based on real-world experience, or even expert consensus opinion, and none are specific to neuraxial or regional anesthesia. Wound healing data are similarly limited and nonspecific for both the drug and surgical procedure. Consequently, only generalized recommendations can be made for perioperative cessation times, recognizing that the length of cessation is impractical for many surgical procedures with greater acuity. Additionally, while great effort was made to include a comprehensive review of the most impactful targeted cancer therapy agents, the scope of this review did not allow for inclusion of all agents and targets. This presents an opportunity for future work in this field. Finally, this review did not comprehensively address other potential perioperative concerns with targeted cancer therapies, including cardiovascular complications, pulmonary toxicity, venous thromboembolism, and the potential for myelosuppression.

In the rapidly evolving field of targeted cancer therapies with their remarkable impact on cancer survival, it is imperative that perioperative physicians recognize these agents and understand their surgical and procedural implications. Until more literature is available, the anesthesiologist has the difficult task of interpreting disparate or limited guidelines and recommendations. Having a basic understanding of the hematologic and wound healing implications of targeted cancer therapy agents will hopefully promote the development of an appropriate perioperative plan. It is hoped that this document helps facilitate shared decision-making with the patient and relevant medical disciplines, including surgeons, oncologists, and intensivists, as well as providing a framework for discussion and further investigation in this area moving forward.

The authors would like to thank Ms. Roisin Khor-Brogan (Salt Lake City, Utah) for her assistance with figure design and preparation.

Support was provided solely from institutional and/or departmental sources.

Dr. Brogan has served as a speaker and consultant for Medtronic (Ireland). The other authors declare no competing interests.