Types of anthracyclines
Maximum lifetime cumulative doses mg/m2
Daunorubicin
550–800
Doxorubicin
400–550
Epirubicin
900–1000
Idarubicin
150–225
Mitoxantrone
100–140
Risk factors | Increased risk in case of |
|---|---|
Age at diagnosis | Young (<4 years) and old age (>65 years) |
Sex | Female |
Rate and schedule of anthracycline administration | Rapid infusion resulting in high peak dose |
Individual anthracycline dose | Higher daily dose |
Cumulative anthracycline dose | Increased cumulative dose |
Radiation therapy | Cumulative dose >30 Gy to the mediastinum or >5 Gy to the heart |
Concomitant therapy | Trastuzumab, cyclophosphamide, bleomycin, vincristine, amsacrine, and mitoxantrone |
Pre-existing cardiovascular disorders | Hypertension, coronary heart disease, valvular disorders, prior cardiotoxic treatment |
Medical comorbidities | Diabetes, obesity, renal dysfunction, pulmonary disease, endocrinopathies, hypocalcemia and hypomagnesemia, sepsis, infection, thyrotoxicosis, alcohol, and pregnancy |
Others | Trisomy 21 and African American ancestry |
Two recent meta-analyses examined if different dosing schedules and various anthracycline derivates reduce cardiotoxicity. The risk of clinical heart failure is significantly lower with an infusion duration of 6 h or longer of anthracyclines as compared to a shorter duration (relative risk (RR) = 0.27; 95 % confidence interval (CI) 0.09–0.81; 5 studies; 557 patients) (van Dalen et al. 2009). However, the same authors (van Dalen et al. 2009) reported no statistically significant difference in the occurrence of clinical heart failure in patients treated with a doxorubicin peak dose of <60 mg/m2 versus ≥60 mg/m2 (RR = 0.65; 95 % CI 0.23–1.88; 2 studies; 4,146 patients), a liposomal doxorubicin peak dose of 25 mg/m2 versus 50 mg/m2 (No patients in either treatment groups developed clinical heart failure; 1 study; 48 patients), and an epirubicin peak dose of 83 mg/m2 versus 110 mg/m2 (RR = 0.97; 95 % CI 0.06–15.48; 1 study; 1,086 patients). Regarding the risk of cardiotoxicity with various anthracycline derivatives, only liposomal-encapsulated doxorubicin is found to be associated with a significantly lower rate of clinical heart failure when compared to conventional doxorubicin (RR = 0.20; 95 % CI 0.05–0.75; 2 studies; 521 patients) (van Dalen et al. 2010). No evidence for a significant difference in the occurrence of clinical heart failure exists between epirubicin versus doxorubicin of the same dose (RR = 0.36; 95 % CI 0.12–1.11; 5 studies; 1,036 patients) (van Dalen et al. 2010).
Pathophysiology
The underlying mechanism of anthracycline-induced cardiotoxicity is complex and remains incompletely understood despite decades of research (Trachtenberg et al. 2011). Loss of myofibrils and cytoplasmic vacuolization caused by dilatation of the sarcoplasmic reticulum in cardiomyocytes are the most prominent histological features (Wouters et al. 2005). Oxidative stress caused by free radical formation is generally accepted as the main mechanism (Zuppinger and Suter 2010; Geiger et al. 2010; Yeh and Bickford 2009; Trachtenberg et al. 2011; Wouters et al. 2005). Reduction of the quinine moiety of anthracyclines to semiquinone generates reactive oxygen species that initiates a cascade of free radical formation (Trachtenberg et al. 2011), which causes damage to cells, cell membranes, and subcellular apparatuses (Trachtenberg et al. 2011; Wouters et al. 2005). Since cardiomyocytes have high oxidative metabolism as evidenced by the abundance of cardiac mitochondria (Trachtenberg et al. 2011) but with fewer natural antioxidants than other organs (Trachtenberg et al. 2011; Wouters et al. 2005), they are most susceptible to oxidative stress (Trachtenberg et al. 2011; Wouters et al. 2005). Furthermore, anthracyclines have a very strong affinity for cardiolipin, which is a phospholipid in the inner cell membrane of cardiac mitochondria that facilitates transport of anthracyclines (Trachtenberg et al. 2011; Wouters et al. 2005). This affinity leads to increased accumulation of anthracyclines inside cardiomyocytes (Trachtenberg et al. 2011; Wouters et al. 2005).
Other postulated mechanisms for anthracycline-induced cardiotoxicity include transcriptional changes in intracellular adenosine triphosphate (ATP) production in cardiomyocytes (Yeh and Bickford 2009); impaired formation of the myofilament protein known as titin in cardiac sarcomeres via calcium-dependent protease activation (Yeh and Bickford 2009; Trachtenberg et al. 2011); downregulation of transcription factors involved in sarcomere synthesis such as GATA4 (Trachtenberg et al. 2011); decrease in cardiac glutathione peroxidase activity (Yeh and Bickford 2009); depletion of cardiac stem cells (Trachtenberg et al. 2011); respiratory defects associated with mitochondrial DNA damage (Yeh and Bickford 2009; Trachtenberg et al. 2011), and impaired mitochondrial creatine kinase activity and function (Trachtenberg et al. 2011). Recently, Lyu et al. (2007) hypothesized that cardiotoxicity caused by doxorubicin may also be due to disruption of the activity of topoisomerase II beta.
2.1.1.2 Alkylating Agents
Drugs that contain reactive alkyl groups capable of forming covalent bonds with DNA are included in this group. Except cisplatin, these drugs were developed from nitrogen mustards, and their clinical use launched the era of cancer chemotherapy.
Incidence
Cyclophosphamide. No cardiotoxicity has been reported for low doses of cyclophosphamide (Floyd et al. 2005). However, acute cardiac toxicity has been described after high-dose cyclophosphamide (120–200 mg/kg) (Yeh and Bickford 2009; Floyd et al. 2005; Senkus and Jassem 2011), as commonly administered in high-dose conditioning regimens for bone marrow transplantation (Floyd et al. 2005). Clinical manifestations may include electrocardiogram changes (decreased amplitude of the QRS complex and nonspecific T wave of ST segment abnormalities), arrhythmias, conduction disorders, and hemorrhagic myopericarditis leading to pericardial effusion, tamponade, and death in some cases (Floyd et al. 2005; Senkus and Jassem 2011). While an asymptomatic transient decrease in ejection fraction has been reported that usually resolves over 3–4 weeks (Floyd et al. 2005), up to 28 % of patients may develop acute-onset fulminant heart failure after high-dose cyclophosphamide (Floyd et al. 2005). Risk of cardiotoxicity is increased with elderly patients, prior anthracycline or mitoxantrone therapy, and mediastinal radiation (Yeh and Bickford 2009; Floyd et al. 2005; Senkus and Jassem 2011).
Ifosfamide. Arrhythmias, ST-T wave changes, and CHF associated with left ventricular dysfunction have been reported for ifosfamide in a dose response manner, usually observed at doses greater than 6.25–10 g/m2, and these toxicities are usually reversible with medical treatment (Floyd et al. 2005; Senkus and Jassem 2011). However, controversy exists regarding if concomitant administration of both ifosfamide and anthracyclines increases cardiotoxicity (Floyd et al. 2005).
Cisplatin. Acute cardiotoxicity of cisplatin includes supraventricular tachycardia, bradycardia, ST-T wave changes, left bundle branch block, acute ischemic events, myocardial infarction, and ischemic cardiomyopathy (Floyd et al. 2005; Senkus and Jassem 2011). Importantly, risk of cardiovascular diseases remains elevated many years after treatment with cisplatin (Senkus and Jassem 2011; Fung and Vaughn 2011). For instance, the risk of cardiovascular disease in long-term testicular cancer survivors who received cisplatin-based chemotherapy is approximately twofold greater than those treated with orchiectomy alone (Fung and Vaughn 2011). This increased risk of cardiovascular disease may be an indirect result of increased incidence of hypertension, hyperlipidemia, and metabolic syndrome observed in patients after cisplatin-based chemotherapy (Fung and Vaughn 2011).
Pathophysiology
The precise mechanism of cyclophosphamide-induced cardiotoxicity is unknown. Cyclophosphamide may cause endothelial capillary damage leading to extravasations of toxic metabolites resulting in damage to cardiomyocytes and hemorrhagic necrosis of the myocardium (Yeh and Bickford 2009; Floyd et al. 2005; Senkus and Jassem 2011). Ifosfamide may cause cardiotoxicity through a similar mechanism as cyclophosphamide due to their analogous structures (Yeh and Bickford 2009). Furthermore, ifosfamide can cause nephrotoxicity that results in delayed elimination of cardiotoxic metabolites along with disturbances of fluid, acid–base, and electrolyte homeostasis (Yeh and Bickford 2009; Floyd et al. 2005; Senkus and Jassem 2011).
Direct endothelial damage caused by cisplatin, as shown by elevations in von Willebrand factor, C-reactive protein, and soluble intercellular adhesion marker 1, may explain the increased risk of cardiotoxicity (Fung and Vaughn 2011). In addition, hypomagnesemia and hypokalemia resulting from cisplatin-induced nephrotoxicity may in turn lead to conduction abnormality and cardiac arrhythmias in the acute setting (Senkus and Jassem 2011).
2.1.1.3 Monoclonal Antibody-Based Tyrosine Kinase Inhibitors
Incidence
Trastuzumab. Trastuzumab, a recombinant DNA-derived humanized monoclonal antibody that selectively binds to the extracellular domain of the HER2 protein, is perhaps the best recognized targeted compound associated with a relatively high risk of cardiac complications (Floyd et al. 2005; Senkus and Jassem 2011). Various degrees of left ventricular systolic dysfunction, which occasionally leads to CHF, is the most common trastuzumab-related cardiac damage (Senkus and Jassem 2011). Improvement of symptoms usually occurs within 6 weeks after discontinuation of trastuzumab and its reinstitution is usually possible with resolution of symptoms (Senkus and Jassem 2011).
In the first landmark clinical trial that evaluated the efficacy and safety of trastuzumab by Slamon et al. (2001), New York Heart Association class III or IV cardiac dysfunction occurred in 27 % of the group given an anthracycline, cyclophosphamide, and trastuzumab; 8 % of the group given an anthracycline and cyclophosphamide alone; 13 % of the group given paclitaxel and trastuzumab; and 1 % of the group given paclitaxel alone. In more recent studies, however, the reported incidence of cardiac complications resulting from trastuzumab was lower with careful monitoring of cardiac function and avoidance of concomitant administration of anthracyclines (Senkus and Jassem 2011). The incidence of cardiac dysfunction ranges from 2 to 7 % when trastuzumab is used as monotherapy and 2–13 % when it is combined with paclitaxel (Yeh and Bickford 2009). More importantly, approximately 1 % of patient will ultimately develop symptomatic CHF (Hayes and Picard 2006).
A recent phase II trial by Rayson et al. (2011) examined the cardiac safety of adjuvant trastuzumab with liposomal doxorubicin in women with breast cancer. The incidence of cardiac toxicity or inability to administer trastuzumab due to cardiotoxicity was 18.6 % (n = 11) in the group with doxorubicin and cyclophosphamide followed by paclitaxel and trastuzumab compared to 4.2 % (n = 5) in the group that replaced doxorubicin with the liposomal formulation (Rayson et al. 2011). In a group of 30 patients with HER2-overexpressing metastatic breast cancer, Chia et al. (2006) reported that after treatment with liposomal doxorubicin and trastuzumab, no patient experienced symptomatic CHF; however, three patients experienced an asymptomatic absolute decline in LVEF of ≥15 % and all of them had previous exposure to anthracyclines.
Increased cumulative dose of trastuzumab does not appear to correlate with risk of cardiotoxicity. Risk factors for trastuzumab-induced cardiomyopathy are age >50 years, pre-existing cardiovascular disease and borderline LVEF, sequence of chemotherapy administration, mediastinal radiation and prior treatment with >300 mg/m2 cumulative dose of anthracyclines (Yeh and Bickford 2009).
Bevacizumab. Bevacizumab is a recombinant humanized monoclonal antibody that binds to and inhibits the biologic activity of human vascular endothelial growth factor (VEGF) (Floyd et al. 2005). Hypertension and thromboembolic events are both well-recognized vascular toxicities of bevacizumab (Senkus and Jassem 2011). Hypertension occurs in 22–36 % of patients (Senkus and Jassem 2011) and this risk is increased 3 times with low dose and 7.5 times with high dose of bevacizumab as shown in a meta-analysis of 7 trials (n = 1,850) (Zhu et al. 2007). Incidence of arterial thromboembolic events is approximately 4–5 % (Senkus and Jassem 2011). When used alone, grade 2–4 left ventricular dysfunction developed in 2 % of patients (Floyd et al. 2005). Among patients who received concurrent anthracyclines, CHF occurred in 14 % of them and in 4 % of those who had prior exposure to anthracycline only (Floyd et al. 2005). Risk factors for bevacizumab-induced cardiovascular toxicity are age greater than 65 years and prior arterial thromboembolic events (Floyd et al. 2005).
Pathophysiology
HER2 protein is critical in the embryonic cardiogenesis and pathogenesis of cardiac hypertrophy (Yeh and Bickford 2009; Senkus and Jassem 2011) and it activates transcription factor AP-1 and nuclear kappa B factor, which are involved in the pathogenesis of cardiac hypertrophy and cellular response to stress respectively (Senkus and Jassem 2011). Inhibition of this pathway by trastuzumab leads to abnormal growth, repair, and survival of cardiomyocytes (Yeh and Bickford 2009) and it may also cause ATP depletion and contractile dysfunction of cardiomyocytes by disrupting the mitochondrial integrity through dysregulation of the BCL-X proteins (Yeh and Bickford 2009). Other proposed mechanisms include drug–drug interaction with anthracyclines; induction of immune-mediated destruction of cardiomyocytes; and an indirect consequence of trastuzumab-related effects outside the heart (Floyd et al. 2005).
The underlying mechanism for development of bevacizumab-induced CHF involves hypertension and inhibition of angiogenesis that causes reduction of myocardial capillary density, cardiac fibrosis, and global contractile dysfunction (Senkus and Jassem 2011). Inhibition of angiogenesis may also explain the increased risk of arterial thromboembolic events with bevacizumab (Yeh and Bickford 2009). Since VEGF is responsible for endothelial cell proliferation and survival, inhibition of VEGF decreases their capability to generate in response to trauma and causes defects in their lining that exposes subendothelial collagen, which subsequently activates tissue factor and increases the risk of thrombotic events (Yeh and Bickford 2009). Furthermore, inhibition of VEGF also leads to decreased concentrations of nitric oxide and prostacyclin and overproduction of erythropoietin that result in increased hematocrit and blood viscosity, all of which may predispose patients to risks of thromboembolism (Yeh and Bickford 2009).
2.1.2 Cardioprotective Interventions
Dexrazoxane, which is an EDTA-like chelator of iron (Geiger et al. 2010), is the most widely investigated agent for reducing anthracycline-induced heart failure (van Dalen et al. 2011). By removing iron from the anthracycline-iron complex or by binding to free iron, it prevents the formation of oxygen radicals, which are thought to be the main mechanism of anthracycline-induced cardiomyopathy (Wouters et al. 2005). A recent meta-analysis by van Dalen et al. (2011) reported that dexrazoxane is associated with lower risks of clinical heart failure (RR = 0.29; 95 % CI 0.20–0.41; 10 studies; 1,619 patients). Furthermore, no evidence was found for a difference in response rate or survival between the dexrazoxane and control groups in this meta-analysis (van Dalen et al. 2011).
Dexrazoxane is currently approved in the United States and the European Union (Geiger et al. 2010) and is usually administered after a cumulative dose of 300 mg/m2 or at the beginning of an anthracycline-based chemotherapy (Geiger et al. 2010). The American Society of Clinical Oncology has published detailed guidelines regarding adjunctive use of dexrazoxane (Hensley et al. 2009) as summarized in Table 3. Aside from dexrazoxane, seven other agents, including N-acetylcysteine, phenethylamine, coenzyme Q10, a combination of vitamins E and C and N-acetylcysteine, L-carnitine, carvedilol, and amifostine, have been studied and demonstrated no cardioprotective effects (Hensley et al. 2009).
Table 3
American society of clinical oncology guidelines for use of dexrazoxane in 2008 (Geiger et al. 2010)
Category | Recommendation |
|---|---|
Breast cancer | |
Initial use in patients with metastatic breast cancer | It is recommended that dexrazoxane not routinely be used for patients with metastatic breast cancer receiving initial doxorubicin-based chemotherapy |
Delayed use in patients with metastatic breast cancer who have received more than 300 mg/m2 of doxorubicin | It is suggested that the use of dexrazoxane be considered for patients with metastatic breast cancer who have received more than 300 mg/m2 of doxorubicin in the metastatic setting and who may benefit from continued doxorubicin-containing therapy; treatment of patients who received more than 300 mg/m2 in the adjuvant setting and are now initiating doxorubicin-based chemotherapy in the metastatic setting should be individualized, with consideration given to the potential for dexrazoxane to decrease response rates as well as decreasing the risk of cardiac toxicity; these patients were not included in the clinical trials of dexrazoxane |
Use in patients receiving adjuvant chemotherapy for breast cancer | The use of dexrazoxane in the adjuvant setting is not suggested outside of a clinical trial |
Other malignancies | |
Use in adult patients with other malignancies | The use of dexrazoxane can be considered in adult patients who have received more than 300 mg/m2 of doxorubicin-based therapy; caution should be exercised in the use of dexrazoxane in setting in which doxorubicin-based therapy has been shown to improve survival |
Use in pediatric malignancies | There is insufficient evidence to make a recommendation for the use of dexrazoxane in the treatment of pediatric malignancies |
Other anthracycline doses and schedules | |
Use in patients receiving other anthracyclines or other anthracycline dose schedules | On the basis of the available data and extrapolations from the experience with doxorubicin plus dexrazoxane, the use of dexrazoxane may be considered for patients responding to anthracycline-based chemotherapy for advanced breast cancer and for whom continued epirubicin therapy is clinically indicated; data for using dexrazoxane with epirubicin for treatment of other cancers are limited; data are insufficient to make a recommendation regarding the use of dexrazoxane with other potentially cardiotoxic agents |
Use in patients receiving high-dose anthracycline therapy | There are no new data addressing the use of dexrazoxane, and there are no new data regarding the clinical use of high-dose anthracyclines; thus, the panel has elected to delete this particular guideline statement, since its clinical relevance appears limited |
Use in patients with cardiac risk factors | There is insufficient evidence on which to base a recommendation for the use of dexrazoxane in patients with cardiac risk factors or underlying cardiac cause |
Monitoring therapy | |
Termination of anthracycline therapy for patients receiving dexrazoxane | Patients receiving dexrazoxane should continue to undergo cardiac monitoring; after cumulative doxorubicin doses of 400 mg/m2, cardiac monitoring should be frequent; the panel suggests repeating the monitoring study after 500 mg/m2 and subsequently after every 50 mg/m2 of doxorubicin; the panel suggests that the termination of dexrazoxane/doxorubicin therapy be strongly considered in patients who develop a decline in LVEP to below institutional normal limits or who develop clinical congestive heart failure |
Dose of dexrazoxane | It is suggested that patients who are being treated with dexrazoxane receive dexrazoxane at a ratio of 10:1 with the doxorubicin dose, given by slow IV push or short IV infusion, 15–30 min before doxorubicin or epirubicin administration; a ratio of 10:1 with the epirubicin dose may be reasonable; however, it should be noted that the optimal dose ratio has not been determined |
2.1.3 Cardiac Monitoring
Regular monitoring of heart function is important during chemotherapy with anthracyclines and trastuzumab (Yeh and Bickford 2009). Echocardiography and multi-gated acquisition (MUGA) scan are the most common non-invasive procedures performed to evaluate LVEF (Geiger et al. 2010; Yeh and Bickford 2009) in order to monitor and diagnose chemotherapy-induced cardiomyopathy (Yeh and Bickford 2009). While echocardiography can identify valvular, pericardial disease, and both systolic and diastolic dysfunction, MUGA scans primary detect decline in left ventricular dysfunction only (Yeh and Bickford 2009). Although endometrial biopsy remains the gold standard for diagnosis of cardiac dysfunction, its invasive nature has limited its use (Yeh and Bickford 2009). In addition, stress testing and dobutamine stress echocardiogram have also been studied extensively but with mixed reported value in their utility to enhance the diagnostic sensitivity for left ventricular dysfunction (Carver et al. 2007). Tables 4 and 5 summarize proposed respective guidelines for cardiac surveillance during doxorubicin (Schwartz et al. 1987) and trastuzumab (Keefe 2002) therapies.
Normal baseline LVEF ≥50 % | Abnormal baseline LVEF <50 % |
|---|---|
After cumulative dose of 250–300 mg/m2: second examination of LVEF | Baseline LVEF ≤30 %: doxorubicin therapy should not be initiated |
After cumulative dose of 400 mg/m2 in patients with known cardiac risk factors and 450 mg/m2 in absence of risk factors: third examination of LVEF with sequential monitoring of cardiac function before each subsequent dose thereafter | Baseline LVEF between 30–50 %: LVEF should be monitored before each dose of doxorubicin |
Functional signs of cardiotoxicity and/or absolute decrease in LVEF ≥10 % associated with a decline to a level of overall LVEF ≤50 %: discontinue doxorubicin therapy | Absolute decrease in LVEF ≥10 % and/or overall LVEF ≤30 %: discontinue doxorubicin therapy |
Table 5
Proposed guidelines for the management of patients treated with trastuzumab (Smith et al. 2010)
Action | ||||
|---|---|---|---|---|
Physical status | LVEF | Trastuzumab | LVEF monitoring | Management |
Asymptomatica | ↓ but normal | Continue | Repeat in 4 weeks | |
↓ >10 points but normal | Continue | Repeat in 4 weeks | Consider β-blockers | |
↓ 10–20 points and LVEF >40 % | Continue | Repeat in 2–4 weeks *Improved: monitor *Not improved: stop trastuzumab | Treat for CHF | |
↓ >20 points to <40 % or LVEF <30 % | Hold | Repeat in 2 weeks *Improved to >45 %: restart trastuzumab *Not improved: stop trastuzumab | Treat for CHF | |
Symptomatica | ↓ <10 points | Continue | Search for noncardiac pathology (e.g., anemia) | |
↓ >10 points and LVEF >50 % | Continue | Repeat in 2–4 weeks *Stable or improved: continue trastuzumab *Worsened: stop trastuzumab | Treat for CHF | |
↓ <30 points | Stop | Treat for CHF | ||
There is evidence that several biochemical markers, including elevations in troponin I (Cardinale et al. 2002, 2004) and B-type natriuretic peptide (BNP) (Nousiainen et al. 2002), may indicate early myocardial injury before development of left ventricular dysfunction. Therefore, there has been interest in serial measurements of BNP to detect changes in left ventricular function (Carver et al. 2007). However, no study has thus far validated this as a routine measurement or screening tool in this population (Carver et al. 2007).
2.2 Pulmonary Toxicity
A wide variety of chemotherapeutic agents has pulmonary toxicities (Carver et al. 2007; Limper 2004) as shown in Table 6. Onset of these adverse effects may be acute or insidious (Limper 2004), with some of them causing permanent lung damage in long-term cancer survivors (Limper 2004; Huang et al. 2011). Clinical manifestations include dyspnea, nonproductive cough, and fever, which may develop weeks to years after chemotherapy (Limper 2004). Aside from hilar lymphadenopathy that is commonly associated with methotrexate-induced lung injury, there is no characteristic radiographic pattern that is specific for other chemotherapy agents. Bleomycin and busulfan are the most relevant in terms of late effects and will be the main focus of this section.
Class of chemotherapy | Agents |
|---|---|
Antibiotics | Bleomycin |
Mitomycin C | |
Alkylating agents | Busulfan |
Cyclophosphamide | |
Chlorambucil | |
Procarbazine | |
Antimetabolites | Methotrexate |
Cytosine arabinoside | |
Fludarabine | |
Gemcitabine | |
Antimicrotubules agents | Docetaxel |
Paclitaxel | |
Vinca alkaloids | |
Nitrosamines | Carmustine |
Topoisomerase inhibitors | Etoposide |
2.2.1 Types of Chemotherapeutic Agents
2.2.1.1 Bleomycin
Bleomycin is an antibiotic mixture of two copper chelating peptides; fermentation products of Streptococcus verticillus, used in the curative treatment of lymphomas and testicular tumors.
Incidence
As many as 20 % of patients develop pulmonary disease after bleomycin (Limper 2004) and the mortality rate from bleomycin-induced pulmonary diseases reaches as high as 1 % (Limper 2004). Bronchiolitis obliterans, eosinophilic hypersensitivity, and interstitial pneumonitis are common bleomycin-related pulmonary disorders (Yousem et al. 1985). Unexplained nonproductive cough and dyspnea on exertion are frequently first signs of bleomycin-induced pneumonitis (BIP) (Comis 1990; Sleijfer 2001) followed by onset of fever, tachypnea, cyanosis, and dyspnea with progressive lung injury (Comis 1990; Sleijfer 2001).
There is a linear relationship between the cumulative dose of bleomycin and the incidence of pulmonary toxicity in animal models (Sleijfer 2001). Studies have shown that the incidence of BIP increases from 3 to 5 % with cumulative bleomycin dose <300 mg to 20 % with doses >500 mg (Fung and Vaughn 2011). Aside from cumulative dose of bleomycin, other risk factors that predispose patients to bleomycin-induced lung injury include mediastinal radiation, renal dysfunction, increased age, smoking, exposure to high-inspired oxygen concentration, and pre-existing pulmonary comorbidity (Carver et al. 2007; Fung and Vaughn 2011; Limper 2004). Concomitant administration of cyclophosphamide, vincristine, doxorubicin, and methotrexate with bleomycin has also been reported to increase risk of pulmonary fibrosis (Huang et al. 2011).
Pathophysiology
The primary mechanism for BIP is direct endothelial damage from bleomycin (Huang et al. 2011; Sleijfer 2001; Cooper et al. 1988), most likely caused by induction of cytokines and free radicals (Sleijfer 2001). These cytokines activate lymphocytes and upregulate adhesion molecules of the endothelial cells, which facilitates adhesion and influx of inflammatory cells, including macrophages, neutrophils, and lymphocytes, into the lung interstitium via the endothelium (Sleijfer 2001). Damage of the endothelial cells, along with infiltration of inflammatory cells into the interstitium, subsequently activates fibroblasts to deposit collagen, which causes pulmonary fibrosis (Sleijfer 2001).
2.2.1.2 Busulfan
Incidence
Busulfan is an alkylating agent with myelosuppressive properties, and is commonly used in the myeloablative conditioning regimen of bone marrow transplantation for hematologic malignancies (Limper 2004; Bilgrami et al. 2001) and idiopathic pneumonia syndrome (IPS) has been described in this population (Bilgrami et al. 2001). The incidence of busulfan pulmonary toxicity is approximately 6 %, with a range of 2.5–43 % (Limper 2004). The average time from initiation of therapy to onset of pulmonary symptoms is approximately 3.5 years, with some patients developing symptoms up to 10 years after treatment (Limper 2004). Dyspnea, cough, and fever are common clinical manifestations and occur in a more insidious nature than with other chemotherapy-related lung disease (Limper 2004). While some patients improve after discontinuation of busulfan and administration of corticosteroids, many develop progressive lung impairment that may ultimately lead to death (Limper 2004). Unlike bleomycin, it is uncertain if busulfan-related lung injury occurs in a dose-dependent fashion (Limper 2004). However, concomitant radiation and cytotoxic agents, such as cyclophosphamide, thiotepa, and melphalan, appear to increase pulmonary toxicities (Limper 2004; Bilgrami et al. 2001).
Pathophysiology
The exact mechanism for busulfan-induced lung injury is unknown, but cell-mediated immune reactions and release of cytokines are thought to be involved (Bilgrami et al. 2001). Atypical oval or elongated cells, intra-alveolar rather than interstitial fibrosis, and multinucleated giant cell containing eosinophilic nuclear inclusions are common histological features with busulfan-induced lung toxicity (Bilgrami et al. 2001). Furthermore, the number of type I pneumocytes decreases while type II pneumocytes proliferate, delamellate, and migrate into alveolar sacs (Bilgrami et al. 2001). Consequently, this may lead to extensive accumulation of alveolar debris that yields a pattern similar to alveolar proteinosis (Limper 2004).
2.2.2 Pulmonary Function Monitoring
Historically, pulmonary function test (PFT), which includes measurement of lung volumes and oxygen diffusion capacity, has been used to monitor for BIP (Fung and Vaughn 2011). Bleomycin was traditionally withheld if the diffusion capacity of carbon monoxide (DLCO) fell below 40–60 % of the pretreatment value (Comis 1990). Sleijfer et al. (1995) examined the validity of using DLCO to detect onset of BIP in testicular cancer patients and showed that DLCO declined in both groups of patients receiving bleomycin, etoposide, and cisplatin (BEP) and etoposide and cisplatin (EP). However, the decrease in DLCO of the BEP group became significant only at 12 weeks after completion of treatment (P < 0.01) when compared to the EP group. Therefore, the authors concluded that a decline in DLCO during active treatment should not be used as the basis for discontinuation of bleomycin therapy. In long-term survivors of Hodgkin’s disease, PFT monitoring has been reported in several clinical trials and they showed that early decline in PFT is followed by subsequent improvement over time in most patients (Carver et al. 2007).
Aside from DLCO, Haugnes et al. (2009) have studied the validity of using other PFT measures to monitor for pulmonary damage in 1,049 long-term testicular survivors after chemotherapy. They reported a decline in both the predicted forced vital capacity (FVC) and forced expiratory volume in one second (FEV1) in this cohort of patients. However, the onset of the decrease in these parameters is unclear and therefore, routine use of FVC and FEV1 to monitor for chemotherapy-induced pulmonary damage is not currently recommended during chemotherapy for testicular cancer patients.
2.2.3 Management of Chemotherapy-Induced Pulmonary Toxicities
Withholding bleomycin at the earliest clinical signs or symptoms of pulmonary toxicities is the most effective treatment for BIP (Fung and Vaughn 2011). Although data regarding role of corticosteroids in management of BIP from prospective randomized trials is lacking, corticosteroids (for example, prednisone 60–100 mg/day) are the mainstay of treatment currently (Sleijfer 2001). For BIP that are refractory to standard-dose corticosteroids, azathioprine (150 mg/day) and high-dose corticosteroids have been used successfully (Fung and Vaughn 2011; Maher and Daly 1993). Similarly, early treatment with corticosteroids is common for busulfan-induced pulmonary toxicities (Bilgrami et al. 2001).
2.3 Nephrotoxicity
Nephrotoxicity is an inherent adverse effect of several chemotherapeutic agents (Kintzel 2001) as shown in Table 7. Mechanism of chemotherapy-induced renal dysfunction usually involves vasculature or tubular damage of the kidneys along with inadequate renal perfusion (Kintzel 2001; Skinner 2011). Hypertension, proteinuria, and varying degrees of renal insufficiency are common clinical manifestations of chemotherapy-induced nephrotoxicity (Jones et al. 2008). While these symptoms may be transient, chronic renal insufficiency may develop insidiously in a minority of patients (Skinner 2011). Long-term renal dysfunction associated with cisplatin and ifosfamide (Fung and Vaughn 2011; Jones et al. 2008) have been described and will be the main focus of this section.
Class of chemotherapy | Agents |
|---|---|
Antibiotics | Mitomycin |
Alkylating agents | Ifosfamide |
Antimetabolites | Azacitidine |
Gemcitabine | |
Methotrexate | |
Pentostatin | |
Nitrosamines | Carmustine |
Semustine | |
Platinum agents | Carboplatin |
Cisplatin | |
Vascular endothelial growth inhibitors | Bevacizumab |
Sorafenib | |
Sunitinib | |
Others | Interferon-α |
2.3.1 Types of Chemotherapeutic Agents
2.3.1.1 Cisplatin
Cisplatin has broad antineoplastic activity, and plays an indispensible role in the curative treatment regimens for testicular cancer. It has activity against ovarian cancer as well as cancers of the head and neck, lung, bladder, and esophagus, and is used as adjuvant therapy in many of these cancers.
Incidence
Clinical manifestations of cisplatin nephrotoxicity include acute and/or chronic renal insufficiency, hypokalemia, and hypomagnesemia, which many cause paresthesia, tremor, tetany, and convulsions (Fung and Vaughn 2011; Skinner 2011). Among long-term testicular cancer survivors treated with cisplatin, approximately 20–30 % of them developed a long-term deterioration in renal function, with an average decline of 20–30 % in glomerular filtration rate (GFR) (Fung and Vaughn 2011). A recent study by Skinner et al. (2009) examined the incidence of long-term nephrotoxicity in 27 children who received cisplatin. At 10-year follow-up, they reported that there was no recovery of cisplatin-induced renal toxicity, as measured by GFR and serum magnesium. GFR was <60 ml/min/1.73 m2 in 11 % of them at both completion and 10 years after chemotherapy.
Risk factors for development of cisplatin nephrotoxicity include high cumulative cisplatin dose (≥500 mg/m2), increased rate of infusion, older age, pre-existing renal disease, renal radiation (≥15 Gy), and concomitant treatment with other nephrotoxins, such as aminoglycosides, amphotericin, immunosuppressants, and methotrexate (Fung and Vaughn 2011; Skinner 2011; Jones et al. 2008). In adults, renal toxicity is less common with low-dose cisplatin (20 mg/m2/day) than high-dose (40 mg/m2/day) (Skinner 2011) while children who received >40 mg/m2/day developed significantly more nephrotoxicity than those at a lower dose (Skinner 2011).
Pathophysiology
Cisplatin exerts direct toxic effects on both the renal tubules and collecting ducts (Fung and Vaughn 2011). Filtration of unbound cisplatin at the glomerulus leads to its uptake into the renal tubular cells (Yao et al. 2007) where it is partially metabolized to nephrotoxic molecules, which subsequently cause cell injury (Yao et al. 2007). Cisplatin also has other intracellular effects that cause tubular damage and dysfunction which may explain sodium, potassium, and magnesium wasting in patients (Yao et al. 2007). These mechanisms include direct cytotoxicity with reactive oxygen species, activation of mitogen-activated protein kinases, induction of apoptosis, and stimulation of inflammation and fibrinogenesis (Yao et al. 2007). Furthermore, cisplatin may also induce renal vasculature damage, resulting in decreased blood flow and ischemic insult to the kidneys (Pabla and Dong 2008).
2.3.1.2 Ifosfamide
Incidence
Ifosfamide has a critical role in the treatment of many solid pediatric malignancies, including bone and soft-tissue sarcoma, Wilm’s tumor, neuroblastoma, and germ cell tumors (Lawson et al. 2008). Proximal tubular dysfunction and decline in GFR, are the most common clinical manifestation of ifosfamide-induced nephrotoxicity (Jones et al. 2008), causing rise in serum creatinine and blood urea nitrogen levels, oliguria, and proximal tubular wasting of electrolytes, glucose, and amino acids (Kintzel 2001). Although most acute renal tubular dysfunction resolves, a minority of kidney damage is permanent and potentially progressive (Jones et al. 2008).
In a study of 16 pediatric osteosarcoma patients after high-dose ifosfamide (14 g/m2 per course), 25 % of them experienced a reduction in GFR, as defined by ≥3 times increase in normal serum creatinine level (Berrak et al. 2005). Skinner et al. (2000) also reported that 17–50 % of pediatric sarcoma patients treated with conventional doses of ifosfamide subsequently developed progressive renal insufficiency. Furthermore, Suarez et al. (1991) reported that among 74 children with malignant mesenchymal tumors who received ifosfamide, 5 % of them developed persistent renal tubulopathy that resulted in Fanconi syndrome.
Risk factors that may predispose patients to ifosfamide-induced nephrotoxicity include younger age at treatment (<5 years at time of treatment), higher cumulative dose (≥60 g/m2), renal radiation (≥15 Gy), and concurrent administration of other nephrotoxins, including cisplatin, amphotericin, immunosuppressants, and methotrexate (Skinner 2011; Jones et al. 2008; Skinner et al. 2000). While most published reports of ifosfamide-induced renal damage were in infants and young children, the significance of age as a risk factor remains unclear (Skinner 2011). Although several studies suggest that young age is a significant risk factor (Skinner 2011), this does not appear to predict long-term nephrotoxicity at 10 years (Skinner et al. 2010). Instead, Oberlin and colleagues recently reported that older age at treatment is associated with decreased GFR at 10 years among 183 pediatric patients who received a median dose of 54 g/m2 of ifosfamide (RR = 1.08; 95 % CI 1.01–1.15). Concurrent cisplatin predisposes patients to higher risk of renal toxicity (Fung and Vaughn 2011). Loebstein et al. (1999) reported that the incidence of nephrotoxicity increased from 33.5 to 41.4 % when cisplatin was included in the chemotherapy of pediatric testicular cancer patients after a cumulative ifosfamide dose of 45 mg/m2.
Pathophysiology
The underlying pathogenesis of ifosfamide-induced nephrotoxicity is currently poorly understood but direct cytotoxicity by ifosfamide or its metabolites is most likely implicated (Skinner 2011), which may involve production of reactive oxygen species that causes disruption in tubular cell energy pathways and membrane function (Skinner 2011). Chloroacetaldehyde (CAA), a metabolite of ifosfamide, is thought to be the primary causative agent for ifosfamide-induced renal damage (Skinner 2011). In a renal tubular cell culture model, CAA causes an experimental Fanconi syndrome possibly by blocking active transport and increasing the permeability of tubular cell membranes (Skinner 2011).
2.3.2 Renal Protective Interventions
Aggressive hydration is the best strategy to reduce the risk of cisplatin nephrotoxicity (Fung and Vaughn 2011). However, evidence for forced diuresis using mannitol or furosemide is controversial (Launay-Vacher et al. 2008). While several experimental reports (Cvitkovic et al. 1977; Pera et al. 1979) have suggested that these diuretics decreased cisplatin nephrotoxicity in animals, two randomized studies in humans (Al-Sarraf et al. 1982; Santoso et al. 2003) showed no such effect. At the same time, concomitant administration of nephrotoxic agents, such as aminoglycosides, non-steroidal anti-inflammatory drugs, and iodinated contrast media, should be avoided in patients receiving cisplatin (Fung and Vaughn 2011). In 2008, the European Society of Clinical Pharmacy Special Interest Group on Cancer Care published their recommendations on the prevention of cisplatin nephrotoxicity (Launay-Vacher et al. 2008) as shown in Table 8.
Table 8
ESCP SIG Cancer Care recommendations on the prevention of cisplatin nephrotoxicity (van Dalen et al. 2009)
Before administration |
1. Estimate GFR or CrCL using MDRD or Cockcroft-Gault formula, respectively |
2. Ensure euvolemia is present |
Dosage |
Adjust cisplatin dosage according to the patient’s renal function |
Administration |
Administer the platinum slowly |
Hydration |
1. Use a saline solution infusion that produces a brisk diuresis |
2. Urine flow should be maintained at 3–4 1/24 h the preceding day and for the next 2–3 days |
3. Do not use diuretics, neither mannitol nor furosemide |
4. There are no data for patients who already are on diuretics, for another concomitant disease such as hypertension |
After administration |
1. When feasible, determine serum creatinine 3–5 days after completion of the course |
2. Monitor magnesium levels routinely and supplement when necessary |
3. Avoid co-adminstration of nephrotoxic drugs (aminoglycosides, non-steroidal anti-inflammatory drugs, iodinated contrast media, zoledronate, etc.) |
4. Re-evaluate renal function before the next course |
The most important measure to reduce the risk of ifosfamide nephrotoxicity is hydration (Fung and Vaughn 2011). While sodium 2-mercaptoethanesulfonate (Mesna) is effective in preventing hemorrhagic cystitis (Lawson et al. 2008), its renal protective effects are uncertain (Lawson et al. 2008). Electrolyte supplementation for patients with persistent electrolyte wasting is also critical after ifosfamide administration (Jones et al. 2008).
2.3.3 Surveillance of Nephrotoxicity
Due to risks of long-term nephrotoxicity from chemotherapy, patients should be monitored regularly using a defined surveillance protocol (Skinner 2011). According to the Children’s Oncology Group of the United States (2008), pediatric oncology patients should have medical check-up at least yearly that includes blood pressure monitoring and urinalysis. At the first long-term follow-up visit (≥2 years after completion of chemotherapy), a basic metabolic panel should be obtained to monitor electrolytes, blood urea nitrogen, and creatinine levels. In addition, these patients should use over-the-counter non-steroidal anti-inflammatory drugs with caution.
2.4 Peripheral Neuropathy
Chemotherapy-induced peripheral neuropathy (CIPN) is a common, potentially severe side effect that can affect the quality of life in long-term cancer survivors (Cavaletti and Marmiroli 2010; Manji 2011). Sensory neuropathies, including paresthesias and pain, are the most common symptoms of CIPN (Pachman et al. 2011), and they usually begin from the fingers and toes and spread proximally in a “glove and stocking” distribution (Pachman et al. 2011). Onset of CIPN usually begins weeks to months after initial treatment with complete resolution of symptoms (Cavaletti and Marmiroli 2010; Manji 2011; Pachman et al. 2011). However, in some cases, CIPN is only partially reversible and can become permanent (Pachman et al. 2011). Table 9 lists the cytotoxic agents commonly associated with CIPN. In this section, we will focus on antimicrotubule agents, platinum analogs, and proteasome inhibitors.
Class of chemotherapy | Agents |
|---|---|
Antimicrotubules agents | Docetaxel |
Ixabepilone | |
Paclitaxel | |
Vincristine | |
Platinum agents | Carboplatin |
Cisplatin | |
Oxaliplatin | |
Proteasome inhibitor | Bortezomib |
Others | Thalidomide |
2.4.1 Type of Chemotherapeutic Agents
2.4.1.1 Antimicrotubule Agents
Paclitaxel, ixabepilone, and vincristine are microtubule inhibitors commonly associated with CIPN (Cavaletti and Marmiroli 2010; Manji 2011). While paclitaxel and ixabepilone both induce polymerization and stabilization of tubulin dimers, vincristine prevents tubulin polymerization from soluble dimers into microtubules (Cavaletti and Marmiroli 2010).
Incidence
The incidence of paclitaxel-induced grade 3 and 4 sensory neuropathy ranges from 2 to 32 % depending on the dose and schedule of its administration (Carlson and Ocean 2011), with its clinical manifestations including paresthesias, numbness, and/or pain in a stocking-glove distribution; decreased proprioception and vibration; sensory ataxia; gait disturbance; and impaired deep tendon reflexes (Cavaletti and Marmiroli 2010; Carlson and Ocean 2011). In some patients, myopathy accompanied by muscle weakness in foot, finger, and ankle extensor muscles also develop (Cavaletti and Marmiroli 2010; Carlson and Ocean 2011). Onset of peripheral neuropathy usually occurs after cumulative doses of 100–300 mg/m2 of paclitaxel (Carlson and Ocean 2011). Dose-dense administration of paclitaxel, albumin-bound paclitaxel, and concomitant administration of platinum agents are other risk factors for paclitaxel-induced CIPN (Manji 2011; Carlson and Ocean 2011).
Approximately 1–24 % of patients experienced grade 3 and 4 sensory neuropathy after ixabepilone, depending of its dose and administration schedule (Carlson and Ocean 2011). Onset of CIPN usually occurs after cumulative doses of 40–120 mg/m2 of ixabepilone (Carlson and Ocean 2011). Compared to peripheral neuropathy associated with paclitaxel, peripheral neuropathy induced by ixabepilone is very similar in nature but its recovery appears to be considerably faster (Cavaletti and Marmiroli 2010; Carlson and Ocean 2011). Examining a dataset of 1,540 patients who received ixabepilone as a monotherapy or in combination with capecitabine, Vahdat et al. (2012) reported pre-existing neuropathy as the only risk factor for ixabepilone-associated peripheral neuropathy.
Vinca alkaloids, including vincristine, vinorelbine, and vinflunine, are all associated with CIPN (Cavaletti and Marmiroli 2010; Carlson and Ocean 2011). After discontinuation of vincristine, up to 30 % of patients may continue to experience peripheral neuropathy (Carlson and Ocean 2011). Aside from sensory neuropathy, moderate to severe autonomic dysfunction is also common after vincrisitine (Cavaletti and Marmiroli 2010; Carlson and Ocean 2011), with colicky abdominal pain and constipation being its symptoms (Cavaletti and Marmiroli 2010; Carlson and Ocean 2011). Furthermore, few cases of paralytic ileus and megacolon have also been reported (Carlson and Ocean 2011). With the newer generation of vinca alkaloids, such as vinorelbine and vinflunine, patients usually develop a primarily sensory neuropathy that is reversible after discontinuation of agents (Carlson and Ocean 2011).
Pathophysiology
The mechanisms for microtubules inhibitors-induced peripheral neuropathy have not been elucidated (Cavaletti and Marmiroli 2010; Carlson and Ocean 2011). Disruption of the active transport of proteins and other compounds within the neuron has been proposed as the main mechanism since intact function of the microtubules is critical for anterograde and retrograde axonal transport and the survival of neurons (Carlson and Ocean 2011). There are evidence that taxanes, such as paclitaxel and docetaxel, target both the axons and the soma of the sensory neurons (Cavaletti et al. 1995, 1997; Persohn et al. 2005) with dorsal root ganglions as the initial site of injury (Carlson and Ocean 2011; Cavaletti et al. 2000). Through signal transduction-mediated pathways, microglial activation within the spinal cord and macrophage activation within the dorsal root ganglion and peripheral nerves appear to be the key pathways in the pathogenesis of taxane-induced peripheral neuropathy (Cavaletti and Marmiroli 2010). The pathophysiology for ixabepilone-induced peripheral neuropathy is hypothesized to be similar to that of the taxanes due to their similar mechanism of action (Carlson and Ocean 2011). Unlike ixabepilone and the taxanes, however, vinca alkaloids induce depolymerization of microtubules (Cavaletti and Marmiroli 2010). This effect on tubulin dimers causes alteration in the length, arrangement, and orientation of axonal microtubules (Cavaletti and Marmiroli 2010), which may explain axonal transport dysfunction and Wallerian-like axonal degeneration (Cavaletti and Marmiroli 2010) observed in patients after treatment with vinca alkaloids.
2.4.1.2 Platinum Agents
Platinum compounds form DNA intrastrand adducts and interstrand crosslinks that disrupt the structure and function of DNA (Cavaletti and Marmiroli 2010) with peripheral neuropathy being a significant side effect of both cisplatin and oxaliplatin (Fung and Vaughn 2011; Cavaletti and Marmiroli 2010; Manji 2011).
Incidence
Approximately 7–31 % of testicular cancer patients undergoing cisplatin-based chemotherapy develop acute peripheral neuropathy (Bajorin et al. 1993; de Wit et al. 2001; Nichols et al. 1998). After a median follow-up of 10 years, persistent symptomatic peripheral neuropathy occurred in up to 30 % of testicular cancer patients after 3–4 cycles of cisplatin-based chemotherapy (Brydoy et al. 2009; Glendenning et al. 2010). Numbness, tingling, decreased proprioception, and impaired vibratory and temperature sensation are common clinical manifestations of cisplatin-induced peripheral neuropathy (Fung and Vaughn 2011; Cavaletti and Marmiroli 2010; Amptoulach and Tsavaris 2011). Onset of peripheral neuropathy usually occurs after a cumulative cisplatin dose of >300 mg/m2 with 50–90 % of patients experiencing neuropathy after a cumulative dose of >500 mg/m2 (Fung and Vaughn 2011; Amptoulach and Tsavaris 2011). Cumulative dose of cisplatin is the most significant risk factor for cisplatin-induced neuropathy (Fung and Vaughn 2011; Brydoy et al. 2009; Amptoulach and Tsavaris 2011).
The incidence of acute oxaliplatin-induced peripheral neuropathy ranges from 65 to 85 % depending on the dose and schedule of its administration (Argyriou et al. 2008), with its clinical symptoms including paresthesias and dysesthesias in the extremities and perioral region that are exacerbated by cold exposure (Amptoulach and Tsavaris 2011). In 1–2 % of patients, transient laryngopharyngeal dysesthesia that manifests as shortness of breath and swallowing difficulty may develop (Amptoulach and Tsavaris 2011; Argyriou et al. 2008). Although these acute toxicities are reversible in approximately 80 % of patients and usually resolve completely in about 40 % of patients 6–8 months after discontinuation of oxaliplatin (Amptoulach and Tsavaris 2011), chronic neuropathy has been reported (Amptoulach and Tsavaris 2011; Argyriou et al. 2008). Its incidence is related to various risk factors, including single dose per course, cumulative dose, administration schedule, and pre-existing peripheral neuropathy (Amptoulach and Tsavaris 2011; Argyriou et al. 2008). Table 10 compares the clinical characteristics of cisplatin and oxaliplatin-induced neurotoxicity.
Cisplatin | Oxaliplatin | |
|---|---|---|
Dose-limiting toxicity | Peripheral neuropathy | Peripheral neuropathy |
Symptoms | Paresthesia | Paresthesia, sensory ataxia, and dysesthesia |
Location | Extremities | Extremities, perioral area |
Time-course onset |
