Acute coronary syndrome in patients undergoing anticancer therapies: A single-center, controlled case study

,


Introduction
Anticancer therapies can be accompanied by cardiovas cular complications, including acute coronary syndrome (ACS). A recently published registry has shown that cancer survivors, compared to the general population, are at a high er risk of cardiovascular morbidity and mortality and that they represent a large group of patients undergoing percu taneous coronary intervention (PCI): 1 in every 13 patients. 1 Cancer at various stages and treated with various anticancer therapies is reported in about 15% of patients with ACS. 2 The association between cancer and ACS is complex and multifactorial. Many cancers have risk factors in common with coronary artery disease: older age, male sex, smoking and obesity. 3 Cancer itself leads to a prothrombotic state, oxidative stress and the progression of atherosclerosis. 4 Addi tionally, anticancer treatment may increase thrombotic risk and lead to cardiotoxic effects, since chemotherapy and ra diotherapy exert proinflammatory and vasospastic effects. 5,6 Optimal ACS treatment in cancer patients can be dif ficult, as these patients are at risk of both stent thrombosis and the bleeding that is often increased due to thrombo cytopenia. In clinical practice, cancer reported in anam nesis can change the treatment plan due to the unknown prognosis of life length and the higher risk of bleeding, as well as to thrombotic events which may accompany the treatment.
The aim of our study was to analyze the treatment of ACS administered to patients with cancer during or soon after the end of their anticancer therapy.

Methods
Based on the hospital database, 3 investigators indepen dently reviewed the discharge cards, medical history re ports and all available medical documents of patients hos pitalized in our Cardiology Departments because of ACS from January 2012 to December 2017. The keywords for the search were ACS, STelevation myocardial infarction (STEMI), nonSTsegment elevation myocardial infarction (NSTEMI), unstable angina (UA), cancer, and neoplasm, using ICD codes C0C97, D37D48 and I20I22.
Next, we searched for patients in whom ACS presented during or soon after (up to 6 months after the end of anti cancer therapy or treatment). The medical records of these patients were carefully screened to analyze the ACS treat ment. The control group consisted of consecutive patients admitted for ACS during the same period, but without a diagnosis of cancer. They were matched with the study group in terms of age, gender and clinical type of ACS. The statistical analyses of continuous data were performed using a ttest in the case of normal distributions, and nonparametric tests (Mann-Whitney U test) in the case of nonnormally distributed or ordinal data. Pvalues <0.05 were considered statistically significant.

Results
Overall, 8,327 records with ACS reported were retrieved from the hospital database, of which 441 records were of patients with a diagnosis of cancer. Finally, 32 records based on the inclusion criteria and deemed adequate for the purpose of our study were screened in detail. The con trol group consisted of 32 patients with a similar age and sex distribution and the same frequency of ACS types.
We analyzed the data obtained from 32 consecutive cancer patients at a mean age of 70 ±9 years (58-88 years) 17 of whom (53.1%) were men admitted from the emergen cy department due to ACS occurring a median of 7 months after their cancer diagnosis. In the cancer group, the most common disease was lung cancer, diagnosed in 9 patients (28.1%). From the remaining patients, breast cancer was in 6 (18.8%), prostate cancer in 4 (12.5%), colon cancer in 4 (12.5%), gastric cancer in 2, ovarian cancer in 2, head and neck cancer in 2, endometrial cancer in 1, nonHodgkin lymphoma in 1, and chronic myeloid leukemia in 1. For 29 of the patients, it was a newly diagnosed tumor, while in the remaining 3 it was a recurrence of cancer. In 22 pa tients (69%), ACS occurred during anticancer treatment: chemotherapy in 15 patients, hormonotherapy in 4, com bined radiation and chemotherapy in 2, and during immu notherapy in 1. In 10 patients (31%), ACS developed within 6 months of the end of anticancer treatment: in 5 patients who had previously undergone chemotherapy and after thorax radiation in the other 3 patients. Twentytwo pa tients (68.7%) had received chemotherapy, and the most commonly used anticancer drugs which could poten tially affect coronary arteries or lead to arterial throm bosis were platinum compounds and fluoropyrimidines, administered in 31.2% and 18.6% of patients, respectively. Data on the chemotherapy agents which were adminis tered is shown in Table 1. Thirteen patients (40.6%) had recently undergone radiotherapy. Of the 13, 7 (21.9%) had undergone thorax irradiation and in 2 patients the can cer recurred more than 10 years after radiotherapy. Seven patients (21.9%) received combined chemotherapy and radiotherapy.
Almost all cancer patients (30 (94%)) presented with at least 1 cardiovascular risk factor or comorbidity. Coro nary artery disease was reported in 13 (40%) patients, and previous myocardial infarction in 3 patients. In the control group, all patients had at least 1 cardiovascular risk factor or comorbidity, and coronary artery disease was diagnosed in 14 (44%) patients, while previous myocardial infarc tion was found in 11 patients. The clinical characteristics of both groups are shown in Table 2.
Upon admission to the hospital, 19 patients (59%) from the control group presented with dyspnea, and 7 (22%) with typical angina. In the control group, the main ACS symptom was typical angina in 28 patients (87%), while dyspnea was reported by 4 patients (13%). The most com mon clinical ACS manifestation was NSTEMI, diagnosed in 16 cancer patients (50%). Unstable angina (UA) had occurred in 10 patients (31.3%) and STEMI in 5 (15.6%). In 1 patient with severe anemia, a type 2 myocardial in farction (MI) had occurred. The incidence of the various clinical ACS types was the same as in the control group.
In the cancer group, coronary angiography was done in 25 patients (78%). Seventeen patients (53%) under went percutaneous coronary angioplasty (PCA), with the implantation of a drugeluting stent (DES) in 12 pa tients and a bare metal stent (BMS) in the remaining 5 patients. None of the patients were treated with balloon angioplasty (POBA) or referred for coronary artery bypass grafting (CABG). In 6 patients, a thirdgeneration DES was implanted. In the control group, all patients were referred for coronary angiography. Twentythree (72%) were treated with PCA: DES implantation in 21 patients and BMS im plantation in 1, while 1 patient was treated with POBA. In 4 patients, a thirdgeneration DES was implanted. As in the cancer group, none of the patients were referred for CABG. In the cancer group, PCA with stent implanta tion was performed in 3 patients (60%) with STEMI, 9 with NSTEMI and 5 with UA, while in the control group it was in 5 (100%), 12 and 6 patients, respectively.
The median time from hospital admission to PCA among the cancer patients was 10 h; it was 105 min in the patients with STEMI (ranging from 30 min to 10 h), 14.5 h in patients with NSTEMI (from 30 min to 10 days) and 13.5 h in those with UA (from 30 min to 4 days).
In the control group, the median time between admis sion and PCA was 7.5 h (p = 0.6 vs the cancer group); it was 35 min in the patients with STEMI (ranging from 30 min to 8 h; p = 0.6), 11.5 h in the patients with NSTEMI (from 30 min to 6 days; p = 0.8) and 5 h in those with UA (from 60 min to 42 h; p = 0.1).
Coronary catheterization was performed using the ra dial approach in all but 1 cancer patient, in whom femoral access was used. There were no PCArelated complica tions or serious bleeding. In 2 patients, bleeding in the ar terial puncture area was noted (1 patient with femoral ac cess), which was selflimiting and did not require medical intervention. In the control group, coronary angiography was performed using the radial approach in 26 patients (81%), via a brachial artery in 2 and via a femoral artery in 4 (12%). No PCArelated complications or bleeding were reported in the control group. Data on the ACS treatment administered to individual patients are shown in Table 3.
Among the cancer patients, aspirin was administered to 29 (94%), clopidogrel to 23 (74%) and 22 (71%) patients obtained double antiplatelet therapy (DAPT). Nine pa tients did not receive antiplatelet treatment because they reported anemia or severe bleeding in anamnesis, and 1 was allergic to aspirin. Enoxaparine was administered to 16 patients (51.6%). A loading dose of 300 mg of aspirin was administered to 20 patients and 600 mg of clopido grel to 23. Ticagrelor and prasugrel were not used in our patients. Triple antithrombotic therapy, involving aspirin, clopidogrel and enoxaparine, was used in 13 patients. None of the patients received a vitamin K antagonist (VKA) or a novel oral anticoagulant (NOAC). In 1 patient, data about antiplatelet and antithrombotic treatment was not obtained. Gastrointestinal bleeding occurred in 1 patient and nasal bleeding in another 1, but medical intervention or prolongation of hospitalization were not required. Statin  was administered to 25 patients (80.6%), βblockers in 28 (87.1%) and an angiotensin converting enzyme inhibitor (ACEI) or angiotensin receptor blocker (ARB) was admin istered to 23 (74.2%).
In the control group, 31 (97%) patients received aspirin, 25 (78%) clopidogrel, 1 (3%) received ticagrelor, and 26 (81%) were given DAPT. A loading dose of 300 mg of aspirin was administered to 18 patients, and 600 mg of clopidogrel to another 18 patients. One patient with type 2 myocardial infarction due to tachyarrhythmia was treated conserva tively and received only anticoagulant therapy (enoxapa rine). In 3 other patients, antithrombotic therapy was ad ministered: warfarine in 2 patients and dabigatran in 1. One patient had nasal bleeding which was treated conservati vely, with no significant drop in hemoglobin concentration. Thirty patients (94%) received statin, 25 (78%) βblockers and 29 (91%) patients were administered ACEI or ARB. None of the patients from either study group suffered from severe thrombocytopenia, and hemoglobin concen tration in the 2 groups varied from 8.7 mg/dL to 14 mg/dL and from 8.2 mg/dL to 16.0 mg/dL, respectively. In echo cardiographic examination performed before hospital discharge, the mean left ventricular ejection fraction was 55 ±12% (33-66%) in the cancer group and 50 ±10% (25-74%) in the control group (p = 0.16).
Inhospital mortality among all patients admitted to our cardiology departments due to ACS in the study period was 6.16%. In the cancer group, during a median of 5 days (3-31 days) of hospitalization 2 patients died (6.25%) due to sudden cardiac arrest and pulseless electrical activity. There were no more deaths from ACS during the following 30 days, though data regarding outcome were not collected on 2 patients. In the control group, the median duration of hospitalization was 6 days (2-22 days) and there were no hospital deaths or deaths from ACS during the follow ing 30 days.

Discussion
The results of the study indicate that the majority of pa tients with ACS presenting during or soon after anticancer treatment can be treated according to the current ACS guidelines, without adversely affecting inhospital prog nosis or the duration of hospitalization. The vast majority of cancer patients received antiplatelet therapy, and even if DAPT or enoxaparine was used, it was not accompanied by an increased risk of bleeding during hospitalization. Most patients were given a βblocker, ACEI or ARB and statin. However, coronary angiography was performed too rarely (in only 78% of the cancer patients) and only 53% underwent PCA with stent implantation.
The pathogenesis of ACS among cancer patients in cludes the impact of classic cardiovascular risk factors, as well as the influence of a prothrombotic state, oxida tive stress and tumorinduced atherosclerosis. 4 The effects of anticancer treatment (i.e., chemotherapy, radiotherapy or surgery) must also be considered. The majority of our cancer patients presented with cardiovascular risk factors or comorbidities, with no significant differences in com parison with the controls (apart from dyslipidemia) which was more frequent among the control group. In addition, the median time from cancer diagnosis was 7 months, which may affect the occurrence of ACS. It has been reported that the incidence of ACS in patients with newly diagnosed cancer increases in the first 6 months from di agnosis and then decreases after a year to increase again in more advanced cancer stages. 6,7 In our patients, we must also consider the influence of recent anticancer therapy on the development of their ACS, as chemotherapy and radiotherapy can exert prothrombotic, proinflammatory and vasospastic effects. 5,6,8 We focused on an early car diac manifestation of anticancer therapy complication, although 2 patients from our group presented with tumor recurrence years after thorax radiotherapy. In this case, the mechanism of ACS may differ, and it could be a result of fibrosis or calcification within the coronary arteries.
Many chemotherapeutic agents predispose one to ACS, as they may provoke coronary vasospasm, endothelial damage or arterial thrombosis and they may aggravate atherosclerosis. 9 A number of chemotherapeutic agents may lead to acute coronary events, mainly cisplatin, 5fluo rouracil, vincristine, rituximab, and BCRABLdirected ty rosine kinase inhibitors. Likewise, paclitaxel, capecitabine, VEGF inhibitors, erlotinib, nilotinib, and ponatinib have been reported to exert vascular toxicity, especially in coro nary arteries. The time of ischemia onset varies widely. It may occur within hours of infusion or several days af terwards. 10 Cisplatinrelated risk can persist even after the end of chemotherapy. 11 Modern radiotherapy aims to focus the radiation beam on the invaded tissue; complications of such treatment are rarer than with the previously used methods. The mecha nism of radiationrelated damage to the coronary arter ies is similar to chemotherapy, and it could be an effect of endothelial injury, coronary vasospasm, atherosclerotic plaque rupture or thrombosis. 9,12 Such injuries are usually located in the ostia and proximal segments of the coronary arteries. 13 The risk of radiationrelated coronary artery disease depends on the radiation dosage and the volume of the irradiated heart. 9 It can manifest early, during or soon after the end of radiation, or with a delay, even after 10-15 years.
As opposed to patients from the control group, who re ported angina as the main ACS symptom, many of the can cer patients presented with dyspnea at hospital admis sion, which is consistent with the observations of other authors. 14-16 Radio and chemotherapyrelated neurotoxic ity can affect the ability to feel pain, so in effect patients after anticancer treatment complain of angina less often. The occurrence of ACS manifests as dyspnea in 44.3% of cancer patients, chest pain in 30.3% and hypotension in 22.7%. 16 As a result of either the higher prevalence of si lent ischemia or the altered perception of angina after anticancer treatment, cancer patients seek emergency care after some delay. Emphasis should be placed on cardiac checkup before anticancer treatment with any cardio toxicity potential and regularly after the end of the treat ment in order to reveal complications of anticancer therapy at an early stage.
The most common clinical manifestation of ACS in our patients was NSTEMI. This finding is in accordance with other studies, which reported that in 85% of can cer patients the ACS manifested as NSTEMI and in 15% as STEMI. 17 Conservative treatment of ACS in cancer patients leads to poor survival rate. 18 An analysis of treat ment in patients with metastatic cancer who developed STEMI or NSTEMI revealed that invasive treatment with PCI resulted in a 2 to 3fold reduction in inhospital mor tality. 19 However, the results of one study indicated that cancer patients undergoing PCI due to STEMI had poorer survival after 1 year (10.7% vs 5.4%) and higher cardiac mortality, which especially pertained to those diagnosed up to 6 months before the onset of ACS. 20 The optimal treatment in this group remains indefinite, because there is no data available on cancer patients in the PCI regis tries. Formerly, cancer patients were excluded from most major randomized, controlled ACS trials. According to the available data, PCI offers a better prognosis in this group of patients, but the need for antiplatelet therapy after stenting should be taken into account in treatment planning. Double antiplatelet therapy can cause hemor rhagic complications and anemia, especially gastrointes tinal and urinary bleeding.
It has been proven that early PCI improves outcomes in ACS independently from the patients' group. 21 In the general population, the frequency of PCI has increased during the last 2 decades, from 11.9% to 60.8% of patients admitted with STEMI. This corresponds with significantly lower 30day mortality and overall mortality. 22 In Poland, according to a recent registry, PCI was performed in 96.2% of patients with STEMI, and in 76.3% of patients with NSTEMI or UA. 23 In our groups, 78% of cancer patients were referred for coronary angiography, and 53% were treated with PCA and stent implantation, while in the con trol group it was 100% and 72% of patients, respectively. Our data indicates that invasive treatment of ACS is less common in patients with cancer, despite current guide lines. Moreover, in STEMI patients with cancer, the me dian time from admission to PCI was 105 min, while it was 35 min in the control group. Guidelines recommend that the interval between arrival at the hospital and intracoro nary balloon inflation (doortoballoon time) during pri mary PCI should be 90 min or less. In the STEMI registry, the median doortoballoon time was 83 min 24 ; thus, it was too late in our patients.
A study comparing PCI outcomes in patients with and without cancer history proved that those reporting cancer in anamnesis received stents less often. Moreover, a delay in invasive treatment, assessed by the time between di agnosis and balloon inflation, was evident in the cancer group. In this study, higher early cardiac mortality was linked to anemia and cardiogenic shock during PCA, which occurred more frequently in cancer patients. 20 In the majority of study patients, PCI was performed by the radial approach. One of 2 bleeding events at the puncture site in our group occurred in a cancer patient on whom PCI was performed via the femoral artery. Ac cording to the literature, femoral artery access is associated with a higher risk of bleeding, even with the use of vascular closure devices after coronary angiography. 25 The femoral approach should be used in patients with abnormal Allen's test results in both hands, with arterial lines, those who have had bilateral mastectomy or multiple radial proce dures and in those on hemodialysis. Radial artery access is preferred for others. 26 The small number of bleeding events in our patients, which were selfterminated and clinically insignificant, may be due to the fact that none of the patients had severe thrombocytopenia during the treatment of ACS. This fact allowed aspirin to be safely administered in 94% of cancer patients and DAPT to be used in 71%, similarly to the con trol group (97% and 78% of patients, respectively). There is no platelet count which limits coronary catheteriza tion, 26 and the use of aspirin in ACS treatment among cancer patients with a platelet count below 100,000/µL was associated with a higher 7day survival rate compared to those who did not receive aspirin (90% vs 6%). 27 Double antiplatelet therapy with clopidogrel can be used in pa tients with a platelet count of 30,000-50,000/µL. Ticagre lor and prasugrel should not be used in cancer patients due to the high risk of bleeding in this group. 17 In 70% of cancer patients who underwent PCA, a drug eluting stent was implanted, which is contrary to other studies reporting that BMSs are used more often in cancer patients. 13,20 The antiproliferative effects of chemotherapy may delay the normal endothelization process observed among noncancer patients after stent implantation, 26 which may favor the use of DES in cancer patients with sufficient prognosis. Drugeluting stent has lower rates of stent thrombosis 18 and with thirdgeneration DES the duration of DAPT can be shortened to 3-6 months in ACS patients. However, in cancer patients with a plate let count from 10,000/µL to 30,000/µL or if DAPT cannot be used or in those demanding surgery or chemotherapy within the next 4 weeks, balloon angioplasty should be considered. 26 With balloon angioplasty, DAPT is required for at least 2 weeks. 26 The vast majority of our patients were treated with a βblocker, ACEI or ARB and statin. It was reported that not only aspirin use, but also βblocker use, in cancer pa tients as in the general population, improves survival rates in ACS. 16 For unknown reasons, βblockers are less likely to be administered to cancer patients. 20 Each cancer pa tient with ACS should be considered for optimal therapy with an antiplatelet drug or drugs, statin, ACEI or ARB and a βblocker.
The most important limitation of our study is the small size of the cancer group and the lack of longterm obser vation after hospital discharge. Our data does not include the stage of cancer or planned further anticancer treat ment, though this was not the focus of our study. The short observation period may have resulted in the low frequency of bleeding complications in the time of recommended DAPT therapy after PCI. The data was collected from our hospital database based on the medical recognition on the information cards at hospital discharge. If a diag nosis of cancer was missing on this card, the patient may not have been included in the analysis. This may to some extent explain the small number of patients from our hos pital database who met the inclusion criteria for the study. As our sample is small, conclusions regarding the popula tion of cancer patients with ACS should only be drawn with special caution. We included data on inhospital mortal ity among ACS patients without cancer who were treated in our cardiology departments, but no direct comparison was made with the cancer group.

Conclusions
Our data suggests that cancer patients with ACS should be treated according to the current guidelines for ACS in the general population, taking into consideration ad ditional factors related to cancer. Data regarding ACS management in cancer patients is still lacking, as the cur rent information is most often based on small population studies and expert consensus. According to our results, patients with ACS onset during or shortly after anticancer therapy are too rarely treated invasively. Moreover, those with STEMI are referred for coronary angiography too late after hospital admission. The presence of cancer and active anticancer treatment should not limit the effective and safe treatment of ACS.