Cancer and CAD share multiple pathophysiological pathways (Figure 1). Several classic risk factors, such as smoking or diabetes, play a relevant role for both. Inflammation and chronic oxidative stress are also of relevance. An interesting example is the CANTOS trial, where the Interleukin-1β inhibitor canakinumab lead to both a reduction in cardiovascular events and fatal cancer.12 Cancer therapies can also often have unwanted coronary toxicity. Another important aspect to bear in mind is that the manifestations of CAD in cancer patients are often due not only to classical atherosclerotic disease, but also due to vasospasm and microvascular disease, and perhaps to a greater extent than in the general population.10

Figure 1.

Mechanisms involved in the association of cancer with coronary artery disease.

(0,24MB).

Figure partly generated using Servier Medical Art, provided by Servier, licensed under a Creative Commons Attribution 3.0 unported license.

The general principles of cardiovascular risk assessment should, of course, also apply to cancer patients. Classic risk factors should be sought and detailed as in the general population. This is especially important as cancer patients often have a higher incidence of classic cardiovascular risk factors compared to the general population.4,13,14

Whether cancer (or specific primary tumors) is an independent risk factor for CAD, is still subject to debate, However, some studies have suggested that gastric, pancreatic, lung, urogenital and lymphoproliferative cancers may be at particular risk.15,16

It is clear is that several cancer therapies pose a clear risk of CAD toxicity. Table 1 outlines the major therapies and their known mechanisms. The incidence of clinical manifestations, either in CCS or ACS presentation, while poorly defined, is usually reported <5% for most agents. However, for some drugs, especially antimetabolites, the incidence of coronary toxicity exceeds 10% in some case series, varying according to posology, timing and route of administration.10,15,17–19 5-Fluorouracyl (5-FU), a drug commonly used for the treatment of many gastro-intestinal cancers, as well as breast cancer, is particularly concerning, with no consensus having been reached regarding an approach. Some advocate routine anti-anginal prophylaxis for such patients, while others recommend cessation of chemotherapy with this agent and initiation of anti-anginal drugs after toxicity. Rechallenge is possible, under supervision, but should generally be avoided if possible.20 Current ESC guidelines suggest that severe CAD should first be excluded by CT or coronary angiography and once prophylactic multidrug anti-anginal therapy has been started.5

It is also worth noting that angiogenesis inhibitors typically induce or increase hypertension, which indirectly contributes to CAD. Lastly, while CAD toxicity manifests principally during therapy, some drugs may produce late-onset disease, after cessation, such as ponatinib, nilotinib and cisplatin.19

Radiotherapy is also a very important cancer specific risk factor for CAD development. Three main clinical characteristics should be remembered: (1) the location of lesions is often specific, including ostial, posing special challenges both regarding risk and revascularization; (2) mediastinal irradiation is especially important, such as (left-sided) breast cancer21 and Hodgkin's lymphoma, and may be mitigated by the use of techniques such as breath holding; (3) CAD often presents very late, sometimes more than a decade after cessation of therapy and is especially concerning for patients who have received treatment at a young age.18,22 As a result, current ESC guidelines suggest screening for high-risk patients who previously underwent chest radiotherapy including the heart, from five years after completion of radiotherapy, ideally with imaging modalities.5

Some strategies (listed in Figure 2) can be considered to reduce CAD burden. The patient specific cardiovascular risk profile should first be assessed as usual in any other setting. Then, cancer specific risk factors, including chemotherapy agents and radiotherapy should be evaluated. Table 1 or a similar example should be routinely available and updated for those dealing with these patients.

Figure 2.

Recommended strategies to reduce the risk of coronary toxicity in cancer patients. CV: cardiovascular; CRVF: cardiovascular risk factors; FUP: follow-up; RT: radiotherapy.

(0,24MB).

Considering the baseline classic cardiovascular risk profile of the patient, if therapy with a potentially coronary toxic cancer agent is planned, the patient should be regarded as high risk (patients undergoing paclitaxel with no classic risk factors) or very high risk (patients undergoing therapy with 5-FU, or a less relevant agent but in combination with cardiovascular risk factors) in terms of CAD management. If so, aggressive risk factor modification should be employed and prophylactic anti-anginal drugs considered. Antiplatelet therapy may also be considered, taking into account the patients’ bleeding risk, as well as the presence of anemia or thrombocytopenia, which is an ever-present issue in these patients. Lastly, information regarding the timing of cancer therapy should be available, with shorter follow-up intervals during administration of these drugs.

The main takeaway from currently available evidence is that cancer patients can still derive important benefits from revascularization, and thus a generally conservative approach is inappropriate. However, given the much larger risk of complications as well as mortality in these patients, a case-by-case evaluation should be undertaken. It is also important to bear in mind that performing a coronary angiography does not mean proceeding with revascularization – as a result, the threshold for undertaking both should be quite different.

Current ESC Guidelines recommend an invasive approach for STEMI or high risk NSTE-ACS, provided life expectancy is ≥6 months, whereas with low-risk NSTE-ACS a conservative approach may be attempted.5 While we generally agree with these recommendations, we believe more factors should be taken into consideration, as outlined below. Additionally, in the STEMI or very high risk NSTE-ACS context, information regarding life expectancy often cannot be obtained in a timely manner, given its emergency setting. As such, the application of such of the life expectancy principle may not be feasible in a considerable number of cases. For chronic coronary syndromes, current ESC guidelines recommend an initial conservative approach,5 with which we agree.

In light of the above, we believe the following principles should apply, as outlined in Figure 3:

• -

  In patients with acute coronary syndrome, invasive risk stratification should generally apply according to current guidelines. Importantly, a thorough history regarding cancer status and vital prognosis should be sought before proceeding with catheterization in the setting of NSTE-ACS, favoring a conservative approach for low-risk cases or short life expectancy (<6 months). If a specific cancer therapy is clearly related to the occurrence of an ACS, temporary or permanent cessation of such therapies should be considered

• -

  In patients with chronic coronary syndromes, initial conservative management should be the default, except in the presence of severe symptoms (CCS class ≥3), critical anatomy (left main) or impaired left ventricle systolic function

• -

  Revascularization should preferably be undertaken with PCI, given the very scarce data available for CABG in this population, as well as the general clinical impression that surgery is often poorly tolerated in these patients; CABG should not, however, be routinely withheld from these patients, but rather evaluated individually; importantly, hybrid strategies, especially if minimally invasive approaches are feasible, should also be considered. Current ESC guidelines suggest CABG for patients with extensive CAD for whom PCI is a poor technical option, with Heart Team discussion and if prognosis exceeds 12 months.5 Furthermore, given that patients with radiation-induced CAD have in increased risk of complications after mediastinal RT,32 guidelines also suggest PCI with DES may be favored over CABG with multivessel disease and a high SYNTAX score (>22), if feasible5

• -

  Low-risk primary tumor location (especially breast cancer), local and non-metastatic disease are favorable scenarios for revascularization, given their lower risk of complications

• -

  High-risk primary tumor location (especially lung cancer), advanced and metastatic disease are unfavorable scenarios for revascularization. Proceeding in these patients requires judicious use of every strategy to minimize bleeding, close follow-up and realistic management of expectations

• -

  Non-solid tumors and active disease are mixed scenarios regarding revascularization – while the risk of intervention is higher, the potential benefit may also be greater.

Figure 3.

Factors favorable and unfavorable for percutaneous coronary intervention in clinical decision making. ACS: acute coronary syndromes; CCS: chronic coronary syndromes.

(0,16MB).

Specific evidence for exact red blood cell (RBC) and platelet levels’ thresholds regarding angiography and PCI in cancer patients is lacking.

Anemia is an almost ubiquitous problem in cancer patients. The etiology is often multifactorial and should ideally be clarified prior to the procedure. Corrective factors, such as vitamin B12, folate, iron or erythropoietin should be provided prior to the procedure whenever possible, thereby minimizing anemia, seeking help from other specialties regularly. If necessary, RBC transfusion may be administered. The general threshold is 7 g/dL in most guidelines, including those of ESMO.33 Notwithstanding, given the increased risk of bleeding and complications in cancer patients, as well as the unwanted effects of blood products administration during the procedure, higher Hb levels (closer to 9–10 g/dL) may be a more sensible option, especially if radial access is not available, complex PCI is planned or there is a specific reason for more aggressive antithrombotic therapy.

Thrombocytopenia is another challenge commonly encountered, either due to the cancer itself or chemotherapy drugs. Current SCAI recommendations3 and ESC Guidelines5 suggest a threshold of 30000 cells/μL for PCI (although 50000 cells/μL is preferable) and 50000 cells/μL for CABG, which is generally agreed upon and which we endorse. SCAI recommendations consider, however, that there is no lower limit for diagnostic angiography.3 Considering the risk of spontaneous bleeding for levels <10000 cells/μL, coronary angiogram is probably best avoided in such extreme levels, unless deemed absolutely necessary and provided radial access is easily obtainable. Platelet transfusion should be undertaken if necessary and feasible for achieving the desirable threshold. Table 2 summarizes these considerations.

Radial access provides a safer choice for PCI with well-established evidence in multiple clinical trials34,35 and should be the default strategy for cancer patients, as in the general population according to current guidelines.10 If femoral access is inexorable, ultrasound guided puncture may decrease the risk of bleeding36,37 and should be sought, provided operators are familiar with the technique. Blind puncture should be avoided and if closure devices are used, especially plug-based devices, pre-closure angiography/ultrasound may be considered, also bearing in mind that endothelization may be slower in patients on chemotherapy, whose risk of infection is also greater.3 Suture devices may be preferred, if operators have the experience. The use of hydrophilic sheaths should also be considered.

Drug-eluting stents (DES) should be the default strategy while performing PCI. SCAI guidelines3 suggest the use of bare metal stents (BMS) when a minimum of 1 month of dual antiplatelet therapy (DAPT) is necessary, but ample evidence has arisen since their publication in 2016 to support DES as standard, as current ESC recommendations do.6 The rates of stent thrombosis are generally very low38,39 and even <1% in more recent trials across multiple current DES platforms.40,41 Even in patients on radiotherapy, where endothelization may be altered, there is observational data to support the use of DES.30 The improved survival of cancer patients further supports DES, as mid to long-term results should be increasingly a concern.42 Bioresorbable scaffolds should be avoided given the disappointing results of past trials43 and the still somewhat experimental nature of the platforms. Plain old balloon angioplasty (POBA) should be reserved for patients who cannot tolerate even a minimum of one month DAPT, although this technique is no more than an a priori intention, as stents may always be necessary for bailing out complications such as dissection or abrupt vessel closure.3,44

It is well known that physiology-guided PCI is generally superior to angiography guided PCI.45 The indexes with the highest level of evidence are FFR46 and iFR.47,48 The cut-off for proceeding with revascularization should follow the standard practice of use,6 although at times a lower threshold (such as the FFR 0.75 cut-off of the DEFER trial49) may be used for selected cancer patients, especially when delays in treatment caused by DAPT are especially nefarious for the oncologic outcome.3

Intravascular ultrasound (IVUS)-guided PCI has also repeatedly been shown to produce better results, with reduced risk of target vessel failure and stent thrombosis.50,51 The same has also been reported with optical coherence tomography, although evidence is less extensive.52,53

Because of the increased risk of bleeding complications in cancer patients and also target lesion revascularization29 and the improved results adjuvant technology can provide43 – i.e. physiology and intra-coronary imaging – these should always be considered, in order to achieve a technically “perfect” result, while also taking into account the added complexity of intervention as well. Table 3 summarizes these considerations.

As bleeding is the single most common complication in cancer patients, every effort should be sought to maximize safety, remembering that commonly used risk stratification scores (such as DAPT,54 PRECISE-DAPT,55 or CRUSADE56) do not take into account cancer as part of their stratification variables, although this has more recently been recognized by other authors.57 Although exact quantification by use of a specific score is lacking, bleeding risk varies according to primary tumor nature, location, stage and metastasis. Bleeding risk should therefore be considered taking these variables into account, which are unique to cancer patients. It is especially important to recall the severity of bleeding due to volume (such as GI) or critical location (such as intra-cranial). Additionally, cancer patients often require surgery on a regular basis, with minimal delays. On the order hand, chemotherapy may theoretically delay stent endothelization, perhaps contributing to the observation that target lesion revascularization is more common in cancer patients.29

Considering the above, shortening DAPT periods is by far the most common necessity in cancer patients. Fortunately, a growing number of clinical trials during the last decade have consistently shown that with current stent technology, 1 month DAPT is a safe and effective strategy, with similarly low rates of ischemic events and significant reduction of bleeding events.38–41 Importantly, the largest of these trials included malignancy other than skin (including GI or lung) as one of the high bleeding risk criteria.40,41 Longer periods of DAPT (3 months) have also been studied with good results.58 As a result, shortening DAPT to a minimum of 1 month can be generally considered whenever bleeding risk is high (such as colon, lung, metastatic cancers, or those with significant anemia or thrombocytopenia) or short-term surgery is warranted.

Current ESC guidelines state that high potency P2Y12 inhibitors (ticagrelor, prasugrel) should be avoided in patients with a platelet count of <50000/μL, but may be considered in selected cases.6 These agents carry a higher risk of bleeding than clopidogrel as observed in both clinical trials and registries,59–61 with very limited evidence for cancer patients. As a result, while we generally agree with current guidelines, we believe clopidogrel should be the standard, with ticagrelor or prasugrel reserved for specific cancer patients with lower risk of bleeding, such as non-advanced, non-metastatic breast cancer without significant anemia or thrombocytopenia. Importantly, interactions may be of greater relevance with more potent P2Y12 inhibitors, whereas for clopidogrel it should be remembered that paclitaxel increases its concentration and therefore bleeding risks.3 Lastly, current ESC guidelines state that with a platelet count <30000 cells/μL, clopidogrel should be avoided and aspirin withheld with <10000 cells/μL.6 This is a strategy we endorse.

For patients requiring oral anticoagulation, multiple clinical trials have shown that dual therapy with OAC (warfarin or DOACs) and clopidogrel (avoiding aspirin except for a very brief period peri-procedurally) is feasible and safe following PCI.62–66 While cancer patients are mostly on anticoagulation therapy for venous thrombosis/embolism rather than atrial fibrillation, multiple clinical trials have also shown that edoxaban, rivaroxaban and apixaban can be effectively used in cancer patients.67–69 As a result, considering the very high bleeding risk of cancer patients, a combination of dual therapy with a DOAC and clopidogrel following PCI should be the default strategy. As there is limited evidence on dabigatran in cancer patients, together with the slightly increased rates of stent thrombosis observed in the RE-DUAL trial62 with the 110 mg dosage of dabigatran (which might have been preferred for cancer patients given their bleeding risk), an anti-Xa DOACs should be preferred.

Lastly, regarding intra-procedural anticoagulation, current guidelines should be followed, bearing in mind that every effort should be made to minimize bleeding. Routine use of fondaparinux for UA/NSTEMI, using heparin dosages and ACT values closer to the lower limits of target intervals (70 U/kg or 50–70 U/kg and 250 s, respectively) and avoidance of regimen crossovers and GpIIbIIIa inhibitors should be sought.3,6 A summary can be found in Table 4.

Aortic stenosis (AS) is a common issue in cancer patients, as they share several risk factors.70 While chemotherapy, especially anthracyclines a are associated with increased risk of valvular disease,71 radiation is perhaps the single greatest cancer specific risk factor, accelerating the emergence and progression of the disease, with general screening for cardiac complications from five years after treatment cessation.70 Current available evidence suggests that surgical aortic valve replacement (SAVR) outcomes are worse for patients with past mediastinal radiotherapy72 and that transcatheter aortic valve implantation (TAVI) results in better in-hospital outcomes and survival than SAVR in patients at intermediate or high risk for heart surgery.73

Current ESC guidelines recommend TAVI for patients at intermediate surgical risk or greater, while also highlighting that STS or EuroSCORE stratification tools may not fully appreciate the surgery-related risk of these patients.5 Hence, patients with past radiotherapy should always undergo a thorough evaluation, as radiation may often induce pulmonary or ascending aortic fibrosis (favoring TAVI), pericardial fibrosis (potentially requiring heart surgery and thereby favoring SAVR) and CAD (which may warrant PCI or CABG, thereby influencing AS treatment modality).74

Regarding patients without past chest radiation, TAVI is currently recommended as the default choice for the treatment of symptomatic severe aortic stenosis in older or more frail patients, despite the fact that age cut-offs vary between American75 and European76 guidelines. Considering the good outcomes of TAVI vs. SAVR across patients with high,77,78 intermediate79,80 and low surgical risk81,82 and the increased frailty of cancer patients, TAVI seems the logical choice for most of these patients. However, cancer patients have generally been excluded from the major trials. Currently available evidence is entirely observational and focuses mostly on TAVI. Comparative studies of TAVI vs. SAVR suggest that the while outcomes (namely mortality) may be generally similar between SAVR and TAVI, the latter enables earlier administration or resumption of cancer therapies. Effective comparisons are limited, as groups are generally rather heterogeneous, with TAVI patients being older and with more comorbidities. No differences across cancer type, staging or active vs. past cancer were found.83,84 With regards to TAVI specifically, one meta-analysis focused on TAVI outcomes in patients with active cancer: short-term outcomes were generally similar between patients with cancer vs. those without, with a trend toward increased pacemaker and bleeding rates in cancer patients. One-year mortality, however, was significantly higher in cancer patients, suggesting that the bulk of mortality was derived from the oncologic rather than the CV disease.70 Another larger meta-analysis,85 partially overlapping with the former,70 found no differences of outcomes after TAVI at any time point between cancer vs. no cancer patients, but also included patients with past rather than active cancer, suggesting that active disease has a relevant contribution for prognosis, as would be expected.

Another important issue is the incidental detection of cancer in CT taken for TAVI planning. The diagnosis of cancer in these patients is estimated in 2.8% according to a recent study and may have implications in the treatment of AS with delay or even cancellation of the TAVI procedure. However, these cases should be discussed with Oncology to establish the prognosis of the disease and decide the best approach to both diseases, in a similar way as other patients with already diagnosed with active cancer.86

The aggregate of currently available evidence therefore suggests that:

• (1)

  Cancer patients can derive benefit from aortic valve intervention

• (2)

  Active cancer patients may have a worse outcome and may be at increased risk of complications

• (3)

  Evidence supporting intervention is greater and more rapidly expanding for TAVI

As a result, we believe that TAVI should be the default strategy for cancer patients, if feasible, especially those who previously underwent chest radiotherapy. It should neither be routinely denied nor offered, but rather decided on a case-by-case basis. Currently available evidence suggests that most patients probably benefit from intervention, provided expected survival is likely to exceed >1 year, as suggested in current guidelines.75,76Table 5 summarizes these considerations.

Cancer patients are at increased risk of mitral regurgitation, as radiotherapy and myocardial cardiotoxicity induced by anthracyclines, trastuzumab or other agents may contribute to development of primary (organic) or secondary (functional) regurgitation.3,18,71 Percutaneous treatment of mitral valve disease is rapidly expanding, with fewer durable results in primary regurgitation87 and mixed results for secondary regurgitation, although available evidence suggests a relevant prognostic benefit in patients where the severity of mitral regurgitation has not progressed too far.88,89 Little evidence concerning cancer patients is available, although a history of cancer seems to adversely affect prognosis in patients undergoing percutaneous edge-to-edge mitral repair.90 Given the frailty of cancer patients, percutaneous repair may be of use as definitive therapy or bridging therapy for surgery for improving functional status enabling cancer therapies. For surgical intervention, minimally invasive techniques should be considered. Table 6 summarizes these considerations.

The added bleeding risk of cancer patients, together with anemia and thrombocytopenia, render OAC sometimes challenging, especially for cases with GI cancer, where bleeding is especially common, including with DOACs.67–69,91 As a result, LAAO would theoretically be a good option for such cases. While still there is still no agreement, initial evidence comparing LAAO with warfarin, as well as LAAO outcomes in a large real-world registry92,93 has recently been complemented by the PRAGUE-17 trial, where LAAO has shown non-inferiority to treatment with DOACs, with significantly reduced bleeding.94 Notwithstanding the scarcity of evidence in cancer patients, with the exclusion of cancer patients from LAAO clinical trials, observational studies have suggested increased risk of in-hospital mortality (regardless of cancer status)95 and in-hospital stroke (for active cancer).96 The authors speculate on a possible relation between the hypercoagulable state in cancer and additional challenges of antithrombotic therapies peri-procedurally, as patients undergoing LAAO usually require at the very least a short period of mono antiplatelet therapy (often with an initial period of DAPT or DOAC). Thus, while LAAO may be considered for cancer patients intolerant to OAC on a case-by-case decision, extra-caution should be exercised peri-procedurally. Table 7 summarizes these considerations.