Abstract
Background/Aim: There is controversy around the use of high-sensitive troponin T (hs-TnT) as an early biomarker of cardiac toxicity in patients with breast cancer on trastuzumab (T). Patients and Methods: Patients receiving adjuvant or neo-adjuvant T for early HER2-positive breast cancer were prospectively included. Transthoracic echocardiograms and matched hs-TnT before T and at 3, 6, and 9 months were performed on all patients. Congestive heart failure, cardiac death, a decline in left ventricular ejection fraction (LVEF) of more than 10% from baseline even if it is still within the normal range, or a drop in LVEF below 55% were all considered signs of cardiac toxicity. Results: In total, 24 patients (median age: 57; range=39-79 years) were enrolled. Anthracyclines were administered to all patients but three as part of neo/adjuvant treatment before T. Cardiovascular toxicity was observed in 3 out of 24 (12.5%) patients: two non-symptomatic LVEF declines (8.3%) and one heart failure episode (4.2%). In the entire population, the mean baseline hs-TnT level was 10.1±8.8 pg/ml, and after 3, 6, and 12 months, no appreciable change was observed. Patients with cardiac toxicity had mean hs-TnT levels higher than those without (18.3±12.3 vs. 8.2±7.2 pg/ml; p=0.049). A definite trend was evident in the chi-square test (chi2=3.52; p=0.06). Conclusion: In anthracycline-exposed patients with early breast cancer, hs-TnT may be able to identify those at risk of developing cardiac toxicity during neo/adjuvant T treatment.
Elimination or at least reduction of unnecessary anticancer treatment interruptions is now the aim of cardio-oncology. To describe this approach the term of “Permissive Cardio-oncology” has been used (1). To this goal an early recognition of cardiac toxicity is of utmost importance.
Anthracyclines and anti-HER2 agents are still two of the cornerstones in HER2-positive breast cancer treatment with a dramatic impact on survival (2-4). However, these drugs are associated with a significant burden of cardiac toxicity somehow limiting their efficacy (5, 6) even if this effect seems to be less pronounced with modern dosing regimens (7). Paradoxically, in very early stage breast cancer survivors the risk of death from cardiac disease exceeds that of cancer recurrence (8). The role of pre-treatment classic cardiovascular risk factors (CVRF) in predicting early and late cardiac toxicity from anthracyclines and trastuzumab has been clearly demonstrated (9-11), as stated in recently published focused European Society of Cardiology guidelines and previously in an article by the American Society of Clinical Oncology (12, 13). In spite of the role of baseline classic CVRF, the search for newer (other than reduction of left ventricle function) and more accurate factors able to early recognize cardiac toxicity has resulted into the clinical interest in biomarkers testing (14).
Myocardial cell injury, changes in myocardial deformation, and asymptomatic left ventricular dysfunction are likely the first signs of cardiotoxicity. If ignored and untreated, these conditions may eventually result in overt heart failure after months or years. Cardiovascular troponins and natriuretic peptides, such as B-type natriuretic peptide (BNP) and NT-terminal-pro-hormone BNP (NT-proBNP), work in concert to help identify patients who are at risk of cardiotoxicity and could help to choose the best imaging technique (15). The demonstration of the predictive role of cardiac biomarker testing in cardiac toxicity from high dose anthracyclines in advanced cancers dates back to the pivotal works of Cardinale et al. (11, 16) but a similar role in patients treated with lower dose anthracyclines or trastuzumab treatment is still to be proven. As proof of this, an Italian survey in cardio-oncology centers pointed out that biomarkers are poorly used in clinical practice (17). On these grounds we conducted this prospective single arm trial to test the hypothesis that pretreatment high-sensitive troponin T level could predict cardiac toxicity in HER2-positive early breast cancer patients receiving adjuvant or neo-adjuvant trastuzumab.
Patients and Methods
Study population. The 3T trial is an Italian single-center prospective interventional trial on the role of pre-treatment troponin-T levels in predicting cardiac toxicity of neo-adjuvant or adjuvant trastuzumab in early breast cancer patients aged 18 years or more. Patients with histologically confirmed invasive early ductal or lobular stage I-III breast cancer (according to UICC-TNM 7th edition) were prospectively enrolled. Other histologies were excluded. HER-2/neu analysis was performed using immunohistochemical (IHC) analysis and confirmatory FISH in ICH unclear 2+ cases only as part of the routine work-up according to local pathology laboratory practice. Only HER2-positive cases were enrolled.
All study patients received trastuzumab (8 mg/kg i.v. loading dose followed by 6 mg/kg i.v. maintenance every 21 days) according to oncologist prescription based on international guidelines as part of neo-adjuvant or adjuvant treatment. No patient received double trastuzumab plus pertuzumab treatment. Study participants received standard chemotherapy before trastuzumab as part of neo/adjuvant treatment. In case of paclitaxel treatment, trastuzumab was administered concomitantly. Chemotherapy regimens used were: EC x4 (epirubicin i.v. bolus 90 mg/m2 day 1+cycloposphamide i.v. bolus 600 mg/m2 day 1 q21) followed by weekly i.v. paclitaxel ×12 (80 mg/m2 day 1 weekly); FEC ×6 (5-fluorouracil i.v. bolus 600 mg/m2 day 1+epirubicin i.v. bolus 60 mg/m2 day 1+cyclophosphamide i.v. bolus 600 mg/m2 day 1 every 21 days); TC ×4 (docetaxel i.v. 75 mg/m2 day 1+cyclophosphamide i.v. bolus 600 mg/m2 day 1 every 21 days). Patients received any other concomitant supportive care and/or hormone therapy (if appropriate).
Exclusion criteria included: concomitant cardiac disease with reduced left ventricle ejection factor (LVEF) contraindicating trastuzumab treatment; inadequate cardiac window; inability to complain with study procedures.
Patients’ names were coded and not revealed. All relevant clinical, echocardiographic and pathological information were recorded in an appropriate database. Clinical records were extracted and collected locally by investigators and survival data were collected from the registry office if necessary. The study received approval by the local ethical committee (Approved on 24 March 2011 by Comitato Etico USL 12 Viareggio, Regional ref. number 342, Delibera di autorizzazione n 134 del 27/04/2011). Informed consent was signed by all patients before any study-related procedure.
Study procedures. hs-TnT analysis was performed locally at the central laboratory of Versilia Hospital according to laboratory policy using commercially available tests. The upper limit of normal range was 14 pg/ml and all values above this threshold were considered as positive. hs-TnT was measured in each patient at four time points: before trastuzumab and 3, 6, and 12 months after treatment start. The biomarker was added to the routine pre-therapy blood tests, avoiding the need for extra blood samples. Results of hs-TnT test were not available to clinicians.
Participants underwent transthoracic echocardiography (TTE) (Aplio 400, Toshiba, Tokyo, Japan) by the same operator (MLC) at the cardiology department of Versilia hospital to reduce interobserver variability. The study protocol consisted of TTE assessment before starting trastuzumab (baseline), at 3, 6, and 12 months after trastuzumab start paired with troponin dosing. All echocardiographic measurements were performed using deidentified images. All available conventional echocardiographic modalities were used to analyze structural and functional cardiac changes. All measurements and the echocardiographic evaluation of the chambers followed the recommendations available at that time (18). Diastolic measures including early (E) and late diastolic (A) mitral inflow velocities, tissue Doppler imaging–based early diastolic mitral septal and lateral annular velocities (e’), left atrial volume, and tricuspid regurgitation velocity were measured. The E/e’ ratio was calculated using the average of the Doppler-defined septal and lateral annular velocities (e’) for left-sided filling pressures. The left ventricular ejection fraction (LVEF) was calculated using the two-plane Simpson method. Left atrial volume was calculated using the biplane method. Diastolic dysfunction grade was determined according to the EACVI guidelines.
Endpoints and statistical analysis. Primary end-point of the study was the ability of hs-TnT values to predict cardiac toxicity of trastuzumab. Secondary end-point was the correlation between hs-TnT values and decreased LVEF. Cardiac toxicity was defined as the occurrence of one or more of the following: signs or symptoms of congestive heart failure, cardiac death, drop >10% of LVEF from baseline even within the normal range or drop of LVEF below the upper limit of normality (55%).
Given the observational aim of the study no formal statistical design was set up but, taking into consideration a predicted percentage of cardiac toxicity of approximately 10%, the trial was originally designed to enroll a minimum of 50 patients. Study accrual stopped early after 24 patients were enrolled due to administrative issues. Descriptive statistics were calculated for all the variables, including continuous variables (reported as mean values and standard deviations or median and range) and categorical variables (reported as numbers and percentages). Serum hs-TnT values are expressed as mean±standard deviation and their differences were tested for significance using Student’s t-test. Statistical significance of categorical variables distribution was evaluated using the chi-square test. All calculations were performed using the STATA statistical software package (Stata, College Station, TX, USA) and the results were considered statistically significant when the p-value was ≤0.05.
Results
Study population. We enrolled 24 patients that met all inclusion criteria and none of the exclusion criteria. Study population characteristics are summarized in Table I upper panel. The majority of patients were post-menopausal with a median age of 57 years (range=39-79 years) affected by invasive ductal carcinoma. Most study patients presented with hormone-sensitive disease (79%) and received hormone therapy according to menopausal status. Trastuzumab was delivered as part of the adjuvant treatment in 80% (19 out of 24 cases) and as part of neo-adjuvant therapy in 5 out of 24 women. All study patients received taxanes (paclitaxel or docetaxel) previous to T and all patients but three (88%, 21 out of 24) received anthracyclines before T as part of neo/adjuvant therapy.
Baseline study population characteristics (n=24).
Eleven out of 24 (46%) patients presented with at least one classic cardiovascular risk factor (CVRF) with hypertension and smoking being the most frequent. Only two out of 24 cases (8%) reported two or more classic CVRF. Median body mass index fell into the normal range. Overall, the CV risk profile of study the population could be considered as moderate-to-low.
Cardiac toxicity and hs-TnT levels. Overall, three out of 24 (12.5%) patients experience cardiac toxicity: two non-symptomatic LVEF drop (8.3%) and one heart failure episode (4.2%) requiring hospitalization (Table I, bottom panel). The two cases with non-symptomatic LVEF drop presented a mean reduction of EF of 13%. No hospitalization was required but T treatment was temporarily stopped and medical treatment with beta-blockers (BB) and angiotensin-converting enzyme inhibitors (ACE-I) was started. A rapid recovery of cardiac function in both cases was observed and T resumed without any new toxicity. ACE-I and BB were continued all along T treatment. The case needing hospitalization for heart failure presented with serious impairment of cardiac systolic function (LVEF 35%) with dyspnea; she received i.v. diuretics, ACE-I and BB, and T was permanently discontinued. Cardiac MR confirmed the reduced EF and showed a pattern of T1 and T2 mapping and late gadolinium enhancement according to cardiac toxicity (non-ischemic). Coronary angiography showed no coronary atherosclerosis. The patient was discharged after four days with partial recovery of cardiac function (LVEF 45%) increasing over time and reaching 50% after three months. All episodes occurred early within the first six months of treatment with a median time of onset of 3 months (range=2-6 months). Interestingly, the patient developing heart failure did not present with baseline CVRF whereas the two cases with non-symptomatic LVEF drop reported hypertension and smoking history.
Mean baseline hs-TnT level was 10.1±8.8 pg/ml in the whole population and we did not observe any significant change at during treatment at 3, 6, 9 and 12 months. Five out of 24 (21%) patients showed pre-T higher hs-TnT levels. Mean baseline hs-TnT levels in patients with cardiac toxicity were significantly higher than those in patients without toxicity (18.3±12.3 vs. 8.2±7.2 pg/ml; p=0.049) (Figure 1). The chi-square test showed a clear trend (chi2=3.52; p=0.06). Only the two cases with non-symptomatic LVEF drop presented with baseline hs-TnT levels out of normal range (31.5 pg/ml and 16 pg/ml) while the case with acute heart failure showed normal pre-trastuzumab hs-TnT levels (7.3 pg/ml).
Comparison of baseline hs-TnT levels in early breast cancer patients without (dark grey) or with (light grey) cardiac toxicity during neo/adjuvant trastuzumab.
No correlation was found between hs-TnT levels and the magnitude of the EF decrease or any other echocardiography parameters.
Discussion
Myocardial cell injury, changes in myocardial deformation, and asymptomatic left ventricular dysfunction are likely the first signs of cardiotoxicity. If ignored and untreated, these conditions may eventually result in overt heart failure after months or years (19). Therefore, the timing of cardiac toxicity diagnosis is of primary importance. Delayed or misdiagnosis could potentially interfere with active oncologic treatments and/or result in permanent cardiac damage. The overall incidence of cardiac side effects varies significantly among studies. A recent meta-analysis pointed out that in the included randomized-controlled studies, the average post-anthracycline decline in main LVEF in the placebo (without cardio-protective drugs) arm was 4.5% but it peaked at 17% in patients receiving higher than 300 mg/m2 total dose of anthracyclines (20). This lower than-previously-reported incidence of cardiac toxicity from anthracyclines has been confirmed; authors reported that with an anthracycline median dosage of 385 mg/m2, the pooled mean drop in LVEF in the placebo arms was 5.4% (21). The definition used for cardiac toxicity greatly impacts on the overall reported incidence with broader definitions resulting in higher incidence of cardiac toxicity up to 37.5% and this could be misleading for general audience (22). To define a common path to manage cardiac side effects of anticancer agents and to reduce confusion a common definition of major cardiovascular toxicities has been proposed (23).
The real impact of anti-HER2 drugs in terms of cardiac side effects is difficult to determine while the wide majority of patients received trastuzumab after an anthracyclines-based regimen. Battisti et al., reported an overall cardiac toxicity rate of 16.6% but serious toxicity was rare (heart failure with NYHA class II in 4.5%, and class III-IV in 0.5% of patients). Early interruptions of treatment for cardiac reasons were reported in 3.8% of the cases with no fatalities noted (24). Trastuzumab-based regimens without anthracyclines are somehow increasingly used, especially for patients with low-risk diseases or those who have several cardiovascular risk factors. In this setting, the pure incidence of cardiac toxicity of trastuzumab seems to be lower (1.2%), but we should take into consideration that women with multiple CV risk factors were excluded (25).
Recently, retrospective data on trastuzumab cardiotoxicity in metastatic breast cancer patients have been reported. Despite the low percentage of patients previously treated with anthracyclines (37%), the reported overall rate of cardiac toxicity was 14%. The majority of patients received chest radiotherapy and this could explain, at least in part, this observation (26).
Our data showed an overall rate of cardiac toxicity of 12,5% during trastuzumab treatment with two cases of non-symptomatic LVEF drop and one heart failure episode requiring hospitalization. This percentage is in accordance with the previously mentioned study of Battisti focusing of breast cancer patients only, even if the percentage of women with anthracyclines exposure was lower (65.3%). In fact, the percentage of anthracyclines-exposed patients in our study was really high (88%) but this did not result in increased cardiac toxicity. However, the CV risk profile differed between the two studies with only 8% of cases presenting with two or more CVRF in our experience compared to 60% of the previous study. Taken together, our observations confirm previous data showing anthracyclines as a major risk factor for trastuzumab cardiotoxicity (27-30).
Pre-treatment with anthracyclines did not cause a widespread hs-TnT increase in our experience. Mean baseline hs-TnT levels in the study population was within the normal range before trastuzumab despite a clear majority of women had previously received an anthracyclines-based chemotherapy regimen. Even if the absolute number of cardiac toxicity events was low, the two cases with non-symptomatic LVEF drop presented with post-anthracyclines and pre-trastuzumab hs-TnT levels out of normal range indicating a baseline sub-clinical cardiac damage. In this scenario trastuzumab served as triggering factor acting on a pre-existing myocardial damage.
The relation between anthracyclines treatment and troponins has been the subject of research for a long time. A pivotal study involving 703 cancer patients found that patients with normal troponin I levels before and after anthracycline-based chemotherapy had a low incidence of cardiac events (1% during the >3-year follow-up), whereas patients with elevated troponin I during the course of ChT had a higher incidence of major adverse cardiac events; of note, patients received very different chemotherapy regimens with high-dose the use of which is today limited in clinical practice (31). Although this result has to be further validated, a more recent investigation showed that absolute changes in hs-Tn levels were particularly predictive of future cardiotoxicity in individuals treated with anthracyclines (32). Increased hs-Tn levels at baseline may also be a sign of a higher risk of cardiac events, according to other research (33).
The use of troponins to predict trastuzumab cardiotoxicity is ambiguous, but it seems to be more effective in patients who have previously been exposed to anthracyclines in relation to the well-known myocardial damage (34, 35). The CARDIOTOX registry reported a really high percentage of serum biomarker abnormalities (78.4%) in cancer patients (mainly breast and lymphoma patients) receiving a potentially cardiotoxic drug but severe and clinically significant toxicities were rare (22).
Presented results highlight the role of baseline hs-TnT level in predicting cardiac toxicity in early breast cancer patients receiving trastuzumab. Mean baseline pre-trastuzumab hs-TnT values were significantly higher in patients experiencing cardiac toxicity. Two different clinical situations could be defined in our population. Approximately 20% (5 out of 24) of women exposed to chemotherapy (with 22 exposed to anthracyclines) presented with pre-trastuzumab levels of hs-TnT over the normal range. Two out five (40%) experienced cardiac toxicity compared with 1 out of 19 (5%) with post-anthracycline/pre-trastuzumab normal hs-TnT levels. In this clinical scenario, baseline high hs-TnT levels identified a subgroup of patients at higher risk for cardiac side effects. This subgroup may benefit from a closer cardiology evaluation schedule or from a cardiac protective strategy given the subclinical myocardial damage already caused by anthracycline treatment. The ICOS-ONE trial randomly assigned 273 adult patients with solid cancer (mainly breast cancer) treated with anthracyclines to a standard cardio-protective arm with enalapril or a troponin-modulated cardio-protective arm with enalapril given only to those patients showing a rise in troponin levels. Interestingly, the percentage of patients with a troponin value over the normal range during anthracycline treatment ranged from 23% to 26% with no difference between the two arms. The percentage of increased hs-TnT (21%) we reported is in accordance with that reported in the ICOS-ONE trial. Authors concluded that in light of enalapril’s effectiveness in preventing LV dysfunction, a troponin-triggered approach would be more practical (36).
A second clinical scenario surfaced. As stated before, one case with normal baseline hs-TnT values experienced a serious cardiac toxicity requiring hospitalization for heart failure and early trastuzumab discontinuation. In this case troponin levels couldn’t help to predict future cardiac toxicity and probably pharmacological preventive strategies will be not useful.
Study limitations need to be considered when analyzing the presented results. Basically, the reduced sample size (due to an early stop of accrual) does not allow general conclusions of the efficacy of hs-TnT in defining a subgroup of women with breast cancer at high risk for cardiac toxicity of trastuzumab. Moreover, the lack of a pre-anthracyclines (or pre-chemotherapy) hs-TnT levels do not allow the determination of the time course variation of troponin during treatment. Only cases with baseline normal values and a rise during anthracycline treatment could be clearly defined as at high risk for toxicity.
Conclusion
The results of the 3T trial confirm the clinical utility of hs-TnT in risk stratification for cardiac toxicity of trastuzumab so possibly reducing the need for planned cardiac imaging in the low risk group while providing a rational basis for a closer cardiology evaluation schedule or for a cardiac protective strategy in women with post-anthracycline/pre-trastuzumab rise in troponin. This information could be of primary importance during the COVID-19 pandemic or in any situation requiring to reduce patient transfer to hospital. The majority of patients received anthracyclines just before trastuzumab, so our results confirm the value of hs-TnT as early biomarker of cardiac damage in anthracycline and trastuzumab-treated patients. Study observations need to be confirmed on a larger sample size.
Acknowledgements
The Authors thank the study team of Versilia Hospital, nurses, patients, and their families for participation in the study.
Footnotes
Authors’ Contributions
Conceptualization: Maria Laura Canale, Giancarlo Casolo, Domenico Amoroso, Fabio Maria Turazza and Andrea Camerini; Data curation: Cheti Puccetti; Formal analysis: Maria Laura Canale, Cheti Puccetti and Andrea Camerini; Investigation: Maria Laura Canale, Sara Donati, Elio Venturini and Andrea Camerini; Methodology: Sara Donati, Irma Bisceglia, Domenico Amoroso, Elio Venturini and Nicola Maurea; Project administration: Giancarlo Casolo; Resources: Fabio Maria Turazza; Supervision: Giancarlo Casolo, Irma Bisceglia and Domenico Amoroso; Visualization: Domenico Amoroso and Nicola Maurea; Writing – original draft: Maria Laura Canale and Andrea Camerini; Writing – review & editing: Maria Laura Canale, Giancarlo Casolo, Sara Donati, Irma Bisceglia, Cheti Puccetti, Elio Venturini, Nicola Maurea, Fabio Maria Turazza and Andrea Camerini. All Authors contributed to manuscript revision and approved the submitted version.
Conflicts of Interest
The Authors declare that there are no conflicts of interest in relation to this study.
- Received May 18, 2023.
- Revision received June 15, 2023.
- Accepted June 19, 2023.
- Copyright © 2023, International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.