You are here

Incidence of capecitabine-related cardiotoxicity in different treatment schedules of metastatic colorectal cancer: A retrospective analysis of the CAIRO studies of the Dutch Colorectal Cancer Group

European Journal of Cancer, May 2017, Volume 76, Pages 93-99



The frequency of capecitabine-related cardiotoxicity has been reported to be low but includes serious adverse events. We conducted a retrospective analysis of the incidence and severity of capecitabine-related cardiotoxicity in different regimens in the treatment of metastatic colorectal cancer in three randomised phase 3 studies.


We used data of cardiac events reported in the CAIRO, CAIRO2 and CAIRO3 studies of the Dutch Colorectal Cancer Group (DCCG) and analysed the incidence and severity of cardiac events in the different treatment regimens of the trials which all included the use of capecitabine. The following events were included: chest pain, newly diagnosed cardiac ischaemia/infarction, atrial fibrillation, other arrhythmias and heart failure, all graded according to National Cancer Institute Common Toxicity Criteria (NCI-CTC).


A total of 1973 patients were included, who received a total of 2461 capecitabine-based lines of treatment. Overall, 5.9% of patients ( n = 117) experienced at least one cardiac event, and 2.3% ( n = 46) experienced at least one grade ≥3 event. Three patients had two cardiac events. The most frequently observed cardiac event was ischaemia/infarction (2.9%, n = 57), followed by arrhythmias (2.0%, n = 40, including atrial fibrillation in 10 patients), chest pain (0.8%, n = 16) and heart failure (0.4%, n = 7). The highest incidence of cardiac events was observed in patients treated with capecitabine in combination with oxaliplatin and bevacizumab (12%, n = 43).


We observed capecitabine-related cardiotoxicity in 5.9% of patients, and severe cardiotoxicity in 2.3% of patients. Combination treatment with capecitabine, oxaliplatin and bevacizumab was associated with the highest risk of cardiotoxicity.



  • In this largest retrospective analysis to date, we observed cardiotoxicity in 5.9% of patients. The incidence of grade ≥3 cardiac events was 2.3%.

  • Cardiac ischemia/infarction and cardiac arrhythmias were the most frequently observed cardiac events.

  • Combination treatment with capecitabine, oxaliplatin and bevacizumab was associated with the highest risk of cardiotoxicity.

  • We support a thorough cardiac evaluation in every patient with cardiac symptoms during treatment with capecitabine.


Keywords: Capecitabine, Cardiotoxicity, Retrospective analysis.

1 Introduction

The most frequently observed toxicities of fluoropyrimidine therapy are diarrhoea, nausea, vomiting, stomatitis and hand-foot syndrome [1] . Cardiotoxicity-related to capecitabine includes serious but infrequently occurring adverse events. The most common symptom is angina-like chest pain, but a wide variety of effects and events may occur, including cardiac arrhythmias, hypertension, hypotension, heart failure, cardiac ischaemia, myocardial infarction, cardiogenic shock or sudden death [2 3 4] . The majority of events are observed during the first treatment cycle, which usually results in the discontinuation of treatment [3 5] . Proposed mechanisms that underlie fluoropyrimidine-induced cardiotoxicity are coronary artery vasospasms, vascular endothelial dysfunction, direct toxicity on the myocardium, and thrombogenicity due to altered rheological factors [6] . Coronary artery vasospasms are thought to be the main pathogenic factor [7] , and it has been suggested that NT-proBNP may act as a biomarker of cardiotoxicity during treatment with fluoropyrimidines [8] . The compound thought to be responsible for these effects is fluoroacetate. 5-Fluorouracil and capecitabine are largely degraded by dehydropyrimidine dehydrogenase (DPD) and in several steps catabolised into α-fluoro-β-alanine (FBAL). FBAL is further catabolised in fluoroacetate, which is known to be highly cardiotoxic and neurotoxic [9] . Dose, cardiac comorbidity and administration schedule of fluoropyrimidines may be predictive factors for cardiotoxicity, but without conclusive evidence [2 10 11] . The reported incidence of patients with cardiac symptoms and events varies—in a recent review of studies on cardiac effects of 5-FU or capecitabine, incidences of cardiotoxicity within the studies of more than 400 patients ranged from 1.2% to 4.3% of patients [10] . Serious cardiac events such as myocardial infarction, cardiogenic shock and cardiac arrest occurred in 0–2% [10] . No significant differences were observed in the incidence of cardiac events upon the use of capecitabine or intravenous 5-FU [12] . We performed a retrospective analysis of the incidence of cardiotoxic events upon the use of capecitabine in commonly used treatment regimens in three randomised phase 3 trials in patients with metastatic colorectal cancer (mCRC) conducted by the Dutch Colorectal Cancer Group (DCCG). To our knowledge, this is the largest retrospective analysis on this topic to date.

2 Methods

2.1 Patients and regimens

In the CAIRO study, 820 patients with mCRC were randomised to receive either first-line treatment with capecitabine monotherapy, second-line treatment with irinotecan and third-line treatment with capecitabine plus oxaliplatin (CAPOX) or first-line treatment with capecitabine plus irinotecan (CAPIRI) and second-line treatment with CAPOX [13] . In the CAIRO2 study, 755 patients with previously untreated mCRC were randomly assigned to capecitabine, oxaliplatin and bevacizumab (CAPOX-B) or the same regimen plus cetuximab (CAPOX-BC) [14] . The CAIRO3 study included 558 patients with mCRC who had stable disease or better after six induction cycles of CAPOX-B and were randomised between maintenance treatment with capecitabine and bevacizumab (CAP-B) or observation [15] . These six induction cycles were not a prospective part of the study and therefore no toxicity data on these cycles are available. After first progression, patients in both study arms were to receive CAPOX-B until second progression.

Treatment cycles in all studies were administered at intervals of 3 weeks. Capecitabine monotherapy was administered on day 1–14 at a dose of 1250 mg/m 2 bid, capecitabine in combination with irinotecan, oxaliplatin or oxaliplatin plus bevacizumab (with or without cetuximab) on day 1–14 at a dose of 1000 mg/m 2 bid and capecitabine with bevacizumab as maintenance treatment was administered continuously at a dose of 625 mg/m 2 bid.

The CAIRO study excluded patients with serious concomitant diseases that may have prevented the safe administration of chemotherapy, and did not explicitly exclude patients with a history of a cardiovascular disease. Patients were excluded from CAIRO2 if they had previous intolerance of adjuvant chemotherapy or clinically significant cardiovascular disease, including unstable angina pectoris, recent myocardial infarction within 12 months, uncontrolled hypertension or previous cerebrovascular disease. Patients were excluded from CAIRO3 if they had experienced any toxic effect from capecitabine, oxaliplatin or bevacizumab during pre-randomisation induction treatment that would prevent its safe continuation or reintroduction or any significant cardiovascular disease within 12 months before randomisation, including symptomatic congestive heart failure, myocardial ischaemia or infarction, unstable angina pectoris, serious uncontrolled cardiac arrhythmia, arterial thrombosis, cerebrovascular event or pulmonary embolism.

2.2 Procedures

For this analysis, we included all patients treated with capecitabine who received at least one dose of study treatment. We categorised patients according to the different lines of treatment: first-, second- and third-line of treatment, respectively. Maintenance treatment with capecitabine plus bevacizumab after 6 induction cycles in CAIRO3 was considered a second-line treatment for the purpose of this analysis. Adverse events were graded according to NCI common toxicity criteria version 2.0 (CAIRO) and 3.0 (CAIRO2 and CAIRO3). Version 2.0 and 3.0 did not majorly differ in grading the seriousness of events. Although grade 5 adverse events were not included in version 2.0, reported treatment-related cardiac deaths in CAIRO were considered as grade 5 event in this analysis. A serious cardiac event was defined as a grade ≥3 event. In our analysis, we defined cardiotoxicity as one of the following events: chest pain which was suspected to be of cardiac origin, newly diagnosed cardiac ischaemia/infarction, heart failure, atrial fibrillation and other arrhythmias, all as assessed by the local investigator.

2.3 Statistical analysis

The incidence of cardiotoxicity was expressed as the percentage of patients that experienced one or more cardiac event. Analyses were performed to compare the incidence of cardiac events between the different lines of treatment administered, particularly between the first-line regimens. Comparisons between second or third regimens were considered biased inasmuch as these patient groups were treated differently in their previous regimens. Considering the differences in capecitabine starting dose and median number of cycles between treatment groups, an exploratory analysis that corrected for the cumulative target dose was performed on the incidence of cardiac events. Comparisons between incidences of cardiac events were made using the chi-squared test. Values of p ≤ 0.05 were considered significant.

3 Results

A total of 1973 out of 2133 patients received at least one dose of capecitabine and were included in this analysis, receiving a total of 2461 lines of capecitabine-based treatment. In first-line, capecitabine monotherapy was administered to 397 patients, CAPIRI to 398 patients, CAPOX-B to 366 patients, and CAPOX-BC to 366 patients. In second-line, CAPOX was administered to 356 patients, CAP-B to 278 patients, and CAPOX-B to 168 patients. CAPOX-B as third-line treatment was administered to 132 patients. A total of 111 (14%) patients in CAIRO, 107 (15%) patients in CAIRO2 and 87 (16%) patients in CAIRO3 had received previous adjuvant therapy. In first-line, the median number of cycles in the capecitabine monotherapy group was 6 (range 1–45), in the CAPIRI group 7 (1–42), in the CAPOX-B group 10 (1–44) and in the CAPOX-BC group 9 (1–44). In second-line, the median number of cycles in the CAPOX group was 5 (1–23), in the CAP-B group 9 (interquartile range (IQR) 5–18) and in the CAPOX-B group 6 (IQR 5–10). The median number of cycles in the third-line CAPOX-B group was 6 (IQR 3–7).

Overall, 117 (5.9%) patients experienced at least one cardiac event, and 46 (2.3%) experienced at least one grade ≥3 cardiac event. Three patients suffered from two events. The most frequently observed cardiac event was cardiac ischaemia/infarction (2.9%, 57 events). Arrhythmia (including atrial fibrillation in 10 patients) occurred in 2.0% (40 events), chest pain in 0.8% (16 events) and heart failure in 0.4% (7 events). Four (0.16%) cardiac deaths were reported. Causes for death were acute myocardial infarction (1), heart failure (1), complete AV-block (1) and ventricular fibrillation (1). The distribution of all grade and grade ≥3 cardiac events is shown in Table 1 . The distribution of cardiac events between the different regimens is shown in Tables 2 and 3 .

  All grade n (%) Grade ≥3 n (%)
Cardiac ischemia/infarction 57 (2.9) 28 (1.4)
Atrial fibrillation 10 (0.5) 1 (0.05)
Arrhythmias, other 30 (1.5) 13 (0.7)
Heart failure 7 (0.4) 5 (0.3)
Chest pain a 16 (0.8)
Total 120 (6.1) 47 (2.4)

a Suspicious for cardiac origin as assessed by treating physician.

Table 1Distribution of cardiac events.

Capecitabine (n = 397) (%) CAPIRI (n = 398) (%) CAPOX-B (n = 366) (%) CAPOX-BC (n = 366) (%)
All grade Grade ≥3 All grade Grade ≥3 All grade Grade ≥3 All grade Grade ≥3
Cardiac ischemia/infarction 11 (3) 4 (1) 13 (3) 10 (2) 13 (4) 5 (1) 8 (2) 6 (2)
Atrial fibrillation 5 (1) 3 (1) 1
Arrhythmias, other 4 (1) 5 (2) 2 (1) 14 (4) 6 (2) 7 (2) 4 (1)
Heart failure 4 (1) 3 (1) 2 (1) 1
Chest pain a 1 7 (2) 4 (1) 0
Total 15 (4) 4 (1) 19 (5) 12 (3) 43 (12) 14 (4) 24 (7) 12 (3)

a Suspicious for cardiac origin as assessed by treating physician.

CAPIRI, capecitabine plus irinotecan; CAPOX-B, capecitabine, oxaliplatin and bevacizumab; CAPOX-BC, capecitabine, oxaliplatin and bevacizumab plus cetuximab.

Table 2Distribution of cardiac events between first-line regimens.

  Second regimen Third regimen
CAP-B (n = 278) (%) CAPOX (n = 356) (%) CAPOX-B (n = 168) (%) CAPOX-B (n = 132) (%)
All grade Grade ≥3 All grade Grade ≥3 All grade Grade ≥3 All grade Grade ≥3
Cardiac ischemia/infarction 5 (2) 3 (1) 3 (1) 4 (2) 1 (1)
Atrial fibrillation 1 1 (1)
Arrhythmias, other 1 (1) 1 (1)
Heart failure 1 1
Chest pain a 2 (1) 1 (1) 1 (1)
Total 9 (3) 4 (1) 3 (1) 7 (4) 1 (1) 2 (2)

a Suspicious for cardiac origin as assessed by treating physician.

CAP-B, capecitabine and bevacizumab; CAPOX, capecitabine plus oxaliplatin; CAPOX-B, capecitabine, oxaliplatin and bevacizumab.

Table 3Distribution of cardiac events between second- and third-line capecitabine-containing regimens.


First-line treatment with CAPOX-B was associated with a significantly higher incidence of all grade (12% versus 4%; p < 0.001) and grade ≥3 (4% versus 1%; p = 0.01) cardiac events compared to capecitabine monotherapy. First-line treatment with CAPIRI showed no significant difference in the incidence of any grade cardiac events compared to capecitabine monotherapy (5% versus 4%; p = 0.48), but the incidence of grade ≥3 events was significantly higher upon treatment with CAPIRI than capecitabine monotherapy (3% versus 1%; p = 0.04). Comparable findings were observed for CAPOX-BC compared to capecitabine monotherapy (7% versus 4%; p = 0.08 for all grade; 3% versus 1%; p = 0.03 for grade ≥3). Compared to first-line treatment, we observed a lower incidence of any grade (2% versus 6%; p < 0.001) and grade ≥3 (0.5% versus 3%; p < 0.001) cardiac events upon treatment with capecitabine-containing regimens in the second- and third-line. Exploratory analyses correcting for the cumulative target dose in the different treatment groups showed that all significant differences were maintained.

Of 46 patients that experienced a serious cardiac event a total of 15 patients (33%) used one or more of the following drugs: anticoagulants, antiplatelet agents, cholesterol-lowering drugs, nitrates, beta blockers, calcium channel blockers and (other) antihypertensive drugs. The use of one or more of the above-mentioned drugs among patients who experienced a serious cardiac event was not significantly higher compared to the use of medication in those who did not experience serious cardiac events (33% versus 29%; p = 0.64).

4 Discussion

In this retrospective analysis of 1973 patients from three studies using capecitabine for metastatic colorectal cancer we observed all grade and grade ≥3 cardiac events in 5.9% and 2.3% of patients, respectively.

This is higher than the overall incidence of 1.2–4.3% that has been reported in previous retrospective studies that reviewed more than 400 patients 10 11 12 16 17 18 19 . Previous studies predominantly reported patients with ischaemic symptoms or events. Next to patients with chest pain with suspected cardiac origin and cardiac ischaemia or infarction, we included heart failure, atrial fibrillation and other arrhythmias, which may account for this difference. Earlier studies considered cardiac adverse events as hypotension, tachycardia, palpitations and dyspnoea also as possibly 5-FU-induced [2] , but we chose to exclude these adverse events because of the non-specificity of the symptoms for a cardiac cause.

Chest pain is the most frequently reported symptom in literature and should be evaluated when present [2 10] . Evaluation by electrocardiography (ECG), or by serum cardiac enzyme analysis or even by coronary angiography is often not performed after episodes of capecitabine-induced chest pain, especially if short-lasting [2 6] . We therefore included all patients with a suspected cardiac-related chest pain in the analysis. By doing so, we may have slightly overestimated the incidence of all grade cardiac toxicity. On the other hand, the retrospective design of our study may have resulted in an underestimation of the incidence of capecitabine-related cardiotoxicity. This is supported by data from prospective trials in which symptomatic and asymptomatic cardiac events upon the use of capecitabine have been reported in up to 34.6% of patients (5.5–34.6%) [3 11 20] . In patients receiving infused 5-FU, silent characteristic ischaemic changes and asymptomatic arrhythmias on ECG recordings have been observed in 64% and 38% of patients, respectively [21 22] , indicating that many cardiotoxic events remain undiagnosed. Notwithstandingly, it supports our decision to include arrhythmias as possible capecitabine-induced cardiac events.

Cardiac ischaemia or infarction was the most commonly reported serious adverse event (2.8%, n = 28). Unfortunately, ischaemic events were not always thoroughly evaluated and a distinction between a thromboembolic event or coronary vasospasms was often not reported by the local investigators. As for grade ≥3 arrhythmias, events included ventricular fibrillation, complete AV-block, atrial flutter, atrial fibrillation and ventricular tachycardia.

In our analysis, first-line treatment with CAPOX-B was associated with a higher incidence of cardiac events compared to capecitabine monotherapy (12% versus 4%, respectively). Although rare cases of oxaliplatin-induced arrhythmias have been described [23] , bevacizumab may be more likely the contributing factor to this finding since bevacizumab is associated with an increased risk of hypertension and a small, but significant risk of arterial thrombotic events [24] . Indeed, any grade hypertension was reported in 43% of patients treated with first-line CAPOX-B or CAPOX-BC (data not shown). The events that were added in patients treated with bevacizumab in the first line were events related to arrhythmia and heart failure, which suggests that bevacizumab may have been responsible for the extra events. An exploratory comparison of our data between CAPOX and CAPOX-B as second-line regimens showed a similar trend with increased number of events, but should be interpreted with caution since these patients were previously treated with different regimens.

A high incidence of fatal cardiovascular events upon treatment with CAPIRI has been reported in the EORTC 40015 trial, resulting in its early discontinuation [25] . An interim analysis of the CAIRO study on safety of the first 400 randomised patients did not show a difference between treatment arms in cardiac ischaemia/infarction events [26] . However, after this interim analysis five myocardial infarctions were observed in the CAPIRI group, compared to one myocardial infarction in the capecitabine group. This difference turned out to be significant, but the number of events is relatively low.

Results of previous studies on intravenous 5-fluorouracil and capecitabine have suggested that the vast majority of cardiac events occur during the first cycle of administration [3 4 17 27] . Indeed we observed a significantly higher incidence of cardiac events during the first-line capecitabine-containing regimens compared to second- and third-line regimens, and 47% of all events occurred during the first three palliative cycles. However, all types of cardiac events were observed during all lines of treatment. Hence, physicians should be aware of the potential of cardiotoxicity at any time during treatment.

The use of medication for symptomatic heart disease or prophylactic treatment was not significantly higher among patients who experienced serious cardiac events. Pre-existing heart disease, the use of calcium channel blockers or nitrates were shown to be predisposing factors by Meyer et al. [17] . A history of ischaemic heart disease was suggested to be a risk factor in more studies [16 21] , but the number of patients presenting with cardiotoxicity were few. Therefore, a history of ischaemic heart disease should not be an absolute contraindication for treatment with fluoropyrimidines.

Currently, the treatment of capecitabine-induced cardiotoxicity is symptomatic. Re-challenging capecitabine is considered unattractive in view of the high rates of recurrent cardiac toxicity that have been reported [2 4] , although re-challenging therapy at a reduced dose in combination with calcium channel blockers and nitrates was suggested to be feasible on the basis of one retrospective series [19] . A better understanding in the pathophysiological mechanisms may lead to better and possibly even preventive treatment options. Drugs with a lower incidence of cardiotoxicity but with comparable efficacy results are warranted. The oral fluoropyrimidine S-1 may be promising in this respect. S-1 has shown non-inferiority in terms of efficacy in different treatment regimens in mCRC and advanced gastroesophageal cancer compared to capecitabine and intravenous 5-FU in Asian populations, and has no reported cases of cardiotoxicity in clinical trials [29 30 31 32] . Preliminary results of the SALTO trial—a randomised phase 3 trial comparing S-1 with capecitabine (with the addition of bevacizumab in both arms as an option) in the first-line treatment of mCRC in a Western population—have not shown any serious cardiac events in either treatment arm [33] . S-1 consists of the 5-FU prodrug tegafur and two modulators, gimeracil and oteracil potassium. Gimeracil inhibits the degradation of 5-FU by DPD inhibition, thereby significantly decreasing the serum level of FBAL and consequently the serum level of the cardiotoxic compound fluoroacetate. In fact, the area under the curve (AUC) of FBAL upon the administration of 1255 mg/m 2 capecitabine is 18 fold greater than the AUC of FBAL during treatment with S-1 [34 35] . Evidence that S-1 can be safely administered after cardiotoxicity-related discontinuation of capecitabine or intravenous 5-FU is limited to case-reports [36] .

In conclusion, our retrospective analysis of three phase 3 studies indicates that a significant number of patients experience cardiotoxicity upon the use of capecitabine, with the highest incidence of cardiac events being observed in patients treated with the combination of capecitabine, oxaliplatin and bevacizumab. Cardiac ischaemia and arrhythmias are the most common capecitabine-related cardiac events. We support a thorough cardiac evaluation in every patient who presents with cardiac symptoms during treatment with capecitabine. A better understanding of the underlying pathophysiological mechanisms is warranted. Preliminary data on fluoropyrimidines with a lower incidence of cardiotoxicity should be confirmed.

Conflict of interest statement

None declared.


This research did not receive any specific grant from funding agencies in the public, commercial or not-for-profit sectors.


  • [1] J. Cassidy ,C. Twelves ,E. Van Cutsem ,P. Hoff ,E. Bajetta ,M. Boyer ,et al. First-line oral capecitabine therapy in metastatic colorectal cancer: a favorable safety profile compared with intravenous 5-fluorouracil/leucovorin. Ann Oncol. 2002;13(4):566-575 Crossref
  • [2] K. Becker ,J.F. Erckenbrecht ,D. Haussinger ,T. Frieling. Cardiotoxicity of the antiproliferative compound fluorouracil. Drugs. 1999;57(4):475-484 Crossref
  • [3] M. Ng ,D. Cunningham ,A.R. Norman. The frequency and pattern of cardiotoxicity observed with capecitabine used in conjunction with oxaliplatin in patients treated for advanced colorectal cancer (CRC). Eur J Cancer. 2005;41(11):1542-1546 Crossref
  • [4] M.W. Saif ,M.M. Shah ,A.R. Shah. Fluoropyrimidine-associated cardiotoxicity: revisited. Expert Opin Drug Saf. 2009;8(2):191-202 Crossref
  • [5] M. de Forni ,M.C. Malet-Martino ,P. Jaillais ,R.E. Shubinski ,J.M. Bachaud ,L. Lemaire ,et al. Cardiotoxicity of high-dose continuous infusion fluorouracil: a prospective clinical study. J Clin Oncol. 1992;10(11):1795-1801 Crossref
  • [6] A. Polk ,K. Vistisen ,M. Vaage-Nilsen ,D.L. Nielsen. A systematic review of the pathophysiology of 5-fluorouracil-induced cardiotoxicity. BMC Pharmacol Toxicol. 2014;15 :47 Crossref
  • [7] T. Südhoff ,M.D. Enderle ,M. Pahlke ,C. Petz ,C. Teschendorf ,U. Graeven ,et al. 5-Fluorouracil induces arterial vasocontractions. Ann Oncol. 2004;15(4):661-664 Crossref
  • [8] S.A. Jensen ,P. Hasbak ,J. Mortensen ,J.B. Sorensen. Fluorouracil induces myocardial ischemia with increases of plasma brain natriuretic peptide and lactic acid but without dysfunction of left ventricle. J Clin Oncol. 2010;28(36):5280-5286 Crossref
  • [9] M. Arellano ,M. Malet-Martino ,R. Martino ,P. Gires. The anti-cancer drug 5-fluorouracil is metabolized by the isolated perfused rat liver and in rats into highly toxic fluoroacetate. Br J Cancer. 1998;77(1):79-86
  • [10] A. Polk ,M. Vaage-Nilsen ,K. Vistisen ,D.L. Nielsen. Cardiotoxicity in cancer patients treated with 5-fluorouracil or capecitabine: a systematic review of incidence, manifestations and predisposing factors. Cancer Treat Rev. 2013;39 :974-984 Crossref
  • [11] C. Kosmas ,M.S. Kallistratos ,P. Kopterides ,J. Syrios ,H. Skopelitis ,N. Mylonakis ,et al. Cardiotoxicity of fluoropyrimidines in different schedules of administration: a prospective study. J Cancer Res Clin Oncol. 2008;134(1):75-82 Crossref
  • [12] E. Van Cutsem ,P.M. Hoff ,J.L. Blum ,M. Abt ,B. Osterwalder. Incidence of cardiotoxicity with the oral fluoropyrimidine capecitabine is typical of that reported with 5-fluorouracil. Ann Oncol. 2002;13(3):484-485 Crossref
  • [13] M. Koopman ,N.F. Antonini ,J. Douma ,J. Wals ,A.H. Honkoop ,F.L. Erdkamp ,et al. Sequential versus combination chemotherapy with capecitabine, irinotecan, and oxaliplatin in advanced colorectal cancer (CAIRO): a phase III randomised controlled trial. Lancet. 2007;370 :135-142 Crossref
  • [14] J. Tol ,M. Koopman ,A. Cats ,C.J. Rodenburg ,G.J. Creemers ,J.G. Schrama ,et al. Chemotherapy, bevacizumab, and cetuximab in metastatic colorectal cancer. N Engl J Med. 2009;360(6):563-567 Crossref
  • [15] L. Simkens ,H. Van Tinteren ,A. May ,A.J. ten Tije ,G.J. Creemers ,O.J. Loosveld ,et al. Maintenance treatment with capecitabine and bevacizumab in metastatic colorectal cancer (CAIRO3): a phase 3 randomised controlled trial of the Dutch Colorectal Cancer Group. Lancet. 2015;385(9980):1843-1852 Crossref
  • [16] R. Labianca ,G. Beretta ,M. Clerici ,P. Fraschini ,G. Luporini. Cardiac toxicity of 5-fluorouracil: a study on 1083 patients. Tumori. 1982;68(6):505-510
  • [17] C.C. Meyer ,K.A. Calis ,L.B. Burke ,C.A. Walawander ,T.H. Grasela. Symptomatic cardiotoxicity associated with 5-fluorouracil. Pharmacotherapy. 1997;17(4):729-736
  • [18] P. Tsibiribi ,J. Descotes ,Lombard-Bohas ,C. Barel ,B. Bui-Xuan ,M. Belkhiria ,et al. Cardiotoxicity of 5-fluorouracil in 1350 patients with no prior history of heart disease. Bull Cancer. 2006;93(3):E27-E30
  • [19] S.A. Jensen ,J.B. Sorensen. Risk factors and prevention of cardiotoxicity induced by 5-fluorouracil or capecitabine. Cancer Chemother Pharmacol. 2006;58(4):487-493 Crossref
  • [20] D. Koca ,T. Salman ,I.T. Unek ,I. Oztop ,H. Ellidokuz ,M. Eren ,et al. Clinical and electrocardiography changes in patients treated with capecitabine. Chemotherapy. 2011;57(5):381-387 Crossref
  • [21] S. Rezkalla ,R.A. Kloner ,J. Ensley ,M. al-Sarraf ,S. Revels ,A. Olivenstein ,et al. Continuous ambulatory ECG monitoring during fluorouracil therapy: a prospective study. J Clin Oncol. 1989;7(4):509-514 Crossref
  • [22] U. Yilmaz ,I. Oztop ,A. Ciloglu ,T. Okan ,U. Tekin ,A. Yaren ,et al. 5-fluorouracil increases the number and complexity of premature complexes in the heart: a prospective study using ambulatory ECG monitoring. Int J Clin Pract. 2007;61(5):795-801 Crossref
  • [23] H.J. Kim ,S.H. An ,Y.H. Cho ,S.Y. Kim ,H.G. Lee ,S.Y. Yoon. Oxaliplatin-induced Torsades de pointes and long QT syndrome in a patient with gastric cancer. Acta Oncol. 2013;52(6):1223-1224 Crossref
  • [24] H. Hurwitz ,N.C. Tebbutt ,F.F. Kabbinavar ,B.J. Giantonio ,Z.Z. Guan ,L. Mitchell ,et al. Efficacy and safety of bevacizumab in metastatic colorectal cancer: pooled analysis from seven randomized controlled trials. Oncologist. 2013;18(9):1004-1012 Crossref
  • [25] C.H. Köhne ,J. De Greve ,J.T. Hartmann ,I. Lang ,P. Vergauwe ,K. Becker ,et al. Irinotecan combined with infusional 5-fluorouracil/folinic acid or capecitabine plus celecoxib or placebo in the first-line treatment of patients with metastatic colorectal cancer. EORTC study 40015. Ann Oncol. 2008;19(5):920-926 Crossref
  • [26] M. Koopman ,N.F. Antonini ,J. Douma ,J. Wals ,A.H. Honkoop ,F.L. Erdkamp ,et al. Randomised study of sequential versus combination chemotherapy with capecitabine, irinotecan and oxaliplatin in advanced colorectal cancer, an interim safety analysis. A Dutch Colorectal Cancer Group (DCCG) phase III study. Ann Oncol. 2006 Oct;17(10):1523-1528 Crossref
  • [27] N.C. Robben ,A.W. Pippas ,J.O. Moore. The syndrome of 5-fluorouracil cardiotoxicity. An elusive cardiopathy. Cancer. 1993;71(2):493-509 Crossref
  • [29] K. Muro ,N. Boku ,K. Sugihara ,A. Tsuji ,S. Sameshima ,H. Baba ,et al. Irinotecan plus S-1 (IRIS) versus fluorouracil and folinic acid plus irinotecan (FOLFIRI) as second-line chemotherapy for metastatic colorectal cancer a randomized phase 2/3 non-inferiority study (FIRIS study). Lancet Oncol. 2010;11 :853-860 Crossref
  • [30] Y.S. Hong ,Y.S. Park ,H.Y. Lim ,J. Lee ,T.W. Kim ,K.P. Kim ,et al. S-1 plus oxaliplatin versus capecitabine plus oxaliplatin for first-line treatment of patients with metastatic colorectal cancer: a randomised, non-inferiority phase 3 trial. Lancet Oncol. 2012;13 :1125-1132 Crossref
  • [31] Y. Yamada ,D. Takahari ,H. Matsumoto ,H. Baba ,M. Nakamura ,K. Yoshida ,et al. Leucovorin, fluorouracil, and oxaliplatin plus bevacizumab versus S-1 and oxaliplatin plus bevacizumab in patients with metastatic colorectal cancer (SOFT): an open-label, non-inferiority, randomised phase 3 trial. Lancet Oncol. 2013;14(13):1278-1286 Crossref
  • [32] J.A. Ajani ,M. Buyse ,M. Lichinitser ,V. Gorbunova ,G. Bodoky ,J.Y. Douillard ,et al. Combination of cisplatin/S-1 in the treatment of patients with advanced gastric or gastroesophageal adenocarcinoma: results of noninferiority and safety analyses compared with cisplatin/5-fluorouracil in the First-Line Advanced Gastric Cancer Study. Eur J Cancer. 2013;49(17):3616-3624 Crossref
  • [33] C.J. Punt ,L.H. Simkens ,J.M. Van Rooijen ,A.J. Van de Wouw ,O.J. Loosveld ,G.J. Creemers ,et al. Randomized phase 3 study of S-1 versus capecitabine in the first-line treatment of metastatic colorectal cancer (mCRC): the SALTO study of the Dutch Colorectal Cancer Group. J Clin Oncol. 2016;34 suppl; abstr 3640
  • [34] B. Reigner ,J. Verweij ,L. Dirix ,J. Cassidy ,C. Twelves ,D. Allman ,et al. Effect of food on the pharmacokinetics of capecitabine and its metabolites following oral administration in cancer patients. Clin Cancer Res. 1998;4(4):941-948
  • [35] Y. Yamada ,T. Hamaguchi ,M. Goto ,K. Muro ,Y. Matsumura ,Y. Shimada ,et al. Plasma concentrations of 5-fluorouracil and F-beta-alanine following oral administration of S-1, a dihydropyrimidine dehydrogenase inhibitory fluoropyrimidine, as compared with protracted venous infusion of 5-fluorouracil. Br J Cancer. 2003;89(5):816-820 Crossref
  • [36] K. Muneoka ,Y. Shirai ,N. Yokoyama ,T. Wakai ,K. Hatakeyama. 5-Fluorouracil cardiotoxicity induced by alpha-fluoro-beta-alanine. Int J Clin Oncol. 2005 Dec;10(6):441-443 Crossref