Early cardiotoxicity after adjuvant concomitant treatment with radiotherapy and trastuzumab in patients with breast cancer

In a prospective observational monocentric population study, we included 175 consecutive patients with HER2-positive breast cancer (stage I-III) without disease recurrence, who received adjuvant treatment with trastuzumab and RT to the breast/chest wall between June 2005 and December 2010 at the Institute of Oncology in Ljubljana.

All patients were treated according to the clinical guidelines, namely with surgery, CT, endocrine therapy in case of hormone receptor positive disease, trastuzumab, and RT. Trastuzumab treatment started before RT or on the first day of RT at the latest. Altogether, 203 consecutive patients with HER2-positive breast cancer were invited to participate in the study, among them 28 patients refused to be involved in the study. Informed consent was obtained from all individual participants included in the study. The study was internationally registered at ClinicalTrial.gov (identifier NCT 01572883), and it was approved by the Republic of Slovenia National Medical Ethics Committee.

In the study framework, between December 2011 and July 2012, after the treatment with trastuzumab had been completed, we performed clinical examinations, echocardiographic measurements of LVEF, and measurements of NT-proBNP levels in all patients. Baseline LVEF was determined either by means of echocardiography or by radionuclide ventriculography. Before the clinical examination took place patients had fulfilled the questionnaires about smoking, concomitant diseases, and problems related to cardio-vascular diseases. All other data were collected from the patients’ records. Patients were classified in New York Heart Association (NYHA) classes, according to the NYHA classification, as well as in World Health Organisation (WHO) performance classes, according to the WHO classification.

The criteria for the adjuvant treatment with trastuzumab regarding tumour, nodal stage and cardiac function were the same as in pivotal adjuvant trials: tumours larger than 2 cm if node negative disease, any tumour size if node positive disease, WHO performance status zero or one, no serious concomitant cardiac disease, and treatment with adjuvant CT.4

According to clinical guidelines patients were operated with either breast conservation surgery or mastectomy and either sentinel node biopsy or axillary dissection. After the operation and CT they were irradiated on the Cobalt machine or on the linear accelerator. Two-dimensional RT (2D RT) or three-dimensional conformal RT (3D CRT) were mostly used. Some of the patients received electron-beam chest wall irradiation, sometimes in combination with concomitant photon-beam irradiation of the periclavicular region (regional RT). Whole breast RT was required in all patients who underwent breast cancer surgery. In addition to the irradiation of the breast/thoracic wall all patients with 4 or more positive axillary lymph nodes also received regional RT. Parasternal lymph nodes were not specifically included in the irradiated area.

Patients were irradiated with a total dose (TD) = 25 x 2 Gy, 5 fractions per week. A minority received RT with TD = 17 or 18 x 2.5 Gy, 5 fractions per week. RT was performed 3 or more weeks after CT had been completed and concurrently with trastuzumab treatment as well as hormonal therapy in case of hormone receptor positive breast cancer.

At the beginning of the primary systemic therapy baseline echocardiography was performed at different clinical institutions according to the shortest waiting time for the examination. Contrary to this, all control echocardiographies were performed in one institution (Department of Cardiology, University Medical Centre Ljubljana) by three cardiologists, and they were carried out on the same device (Aloka SSD-α10, Tokyo, Japan). Conventional and tissue-Doppler echocardiography was performed on each patient.20,21 Normal range for LVEF was 50% or more.

All radionuclide ventriculographs were performed at the Institute of Oncology in Ljubljana (Gamma Cam Siemens; erythrocytes were labelled with in vivo method, activity 740-952 MBq) Normal range for LVEF was 50% or more. LVEF was calculated with the programme Intermedical/Medicview.

NT-proBNP was determined with the analyser Cobas e 411 (Roche). According to the instructions of the manufacturer, the values of the NT-proBNP below 125 ng/l exclude heart dysfunction.

HER2 immuno histochemistry (IHC) expression was scored as follows: 0, no staining or faint membrane staining in ≤ 10% of tumour cells; 1+, incomplete membrane staining that is faint perceptible in >10% of tumour cells,; 2+, incomplete and/or weak to moderate membrane staining in >10% of tumour cells or complete and intense staging in ≤ 10% of tumour cells; and 3+, complete, intense circumferential membrane staining in >10% of tumour cells. HER2 scores of 0 and 1+ were considered negative. All IHC2+ tumours were tested for gene amplification by fluorescent in situ hybridization (FISH). HER2 IHC 3+ and FISH-amplified tumours were considered positive.

The study was designed as an equivalence study. For the purpose of the analysis the patients were divided into two groups: (1) left breast cancer group: patients irradiated on the left side of the chest (breast cancer of the left breast) and (2) right breast cancer group: patients irradiated on the right side of the chest (breast cancer of the right breast).

In both group we compared the difference between LVEF, measured at the beginning of the adjuvant systemic therapy (LVEF₀), and LVEF measured after the treatment with trastuzumab (LVEF₁) had been completed. The difference between the two measurements was marked as Δ LVEF = LVEF₀ - LVEF₁ and we made comparisons between the two groups. A 95% confidence interval for the difference of means (CI) was estimated. Groups were labelled as equivalent if 95 % CI did not include Δ LVEF of 10 percentage points, but it included the value 0.

In calculating the necessary sample size we assumed that there were no differences in the population between the investigated groups, and that the standard deviation of the difference of Δ LVEF would be 10 percentage points. The level of significance was set to 5% and the desired power to at least 80%. We considered the difference Δ LVEF of 10 percentage points or less as clinically irrelevant. The calculation using PASS (version 12) showed that under these assumptions the equivalence study enrolling 20 patients per group would have 86% power. In addition to the primary objective we, namely, wanted to analyse also echocardiographic parameters at the control examination after the completed treatment with trastuzumab and RT. The data is presented as mean (standard deviation, SD) or median (interquartile range Q1–Q3) for continuous variables as appropriate, and number (%) for categorical variables.

The difference between the groups (the left breast cancer group and the right breast cancer group) for continuous variables was tested with t-test, Welsch t-test, or Mann-Whitney test as appropriate. The assumption of normality was verified with Shapiro-Wilks test, and Bartlett test was used to test the assumption of variance equality.

The association between the two groups (left breast cancer and right breast cancer group) and categorical variables was tested with χ² test, as well as with Yates continuity correction of Fischer exact test as appropriate.

A p-value of less than 0.05 was considered as statistically significant. The analysis was performed with R language for statistical computing (R version 3.0.1.).22

There were 84 patients (48%) in the left breast cancer group, and 91 patients (52%) in the right breast cancer group. Median age was 59 (52–67) years in the left breast cancer group, and 55 (46–63) years in the right breast cancer group (p = 0.009). Median observation time was 57 (37–71) months. Patients and tumour characteristics as well as the associated diseases are described in Table 1. Among all the patients, 35 of them (20%) had arterial hypertension, but there were no significant differences between the two groups.

Table 1

Patients and tumour characteristics, concomitant diseases

All patients received CT. Among all of them 95 patients (54.3% of all the patients) received one of the following CT schemes: doxorubicin, cyclophosphamide (AC) / epirubicin, cyclophosphamide (EC) / 5-FU, doxorubicin, cyclophosphamide (FAC) / 5-FU, epirubicin, and cyclophosphamide (FEC) with a sequence of taxanes, and 69 patients (39.4% of all the patients) received one of the following schemes: AC / EC / FAC / FEC without taxanes. Only 11 patients (6.3% of all the patients) did not receive any anthracyclines (6 patients in the left breast cancer group and 5 patients in the right breast cancer group). None of the patients received concomitant anthracyclines and trastuzumab. There were no significant differences found in the CT schemes used, hormone therapies, and in the mean received cumulative doses of anthracyclines, taxanes, cyclophosphamide, and trastuzumab between the two groups. Systemic treatment is described in detail in Table 2.

Table 2

Systemic treatment and surgical characteristics

Among all the patients, 39 patients (79.4% of all the patients) were treated with 2D RT technique, and 25 patients (14.3%) with 3D CRT technique. RT treatment features are presented in Table 3.

Table 3

Radiotherapy treatment features

The analysis showed no statistically significant differences between initial and control LVEF in the observed groups. Data is presented in detail in Table 4.

Table 4

Analysis of the difference in left ventricular ejection fractions (Δ LVEF)

The time between the introduction of trastuzumab and the beginning of RT, the time between LVEF₀ measurement and the time beetween LVEF₀ and LVEF₁ did not differ between the observed groups (p = 0.596, 0.506 and 0.089, respectively).

Overall, we found an important reduction of the LVEF (a decrease of LVEF for 10 percent points or more or a final value of LVEF < 50) in 9 patients (6%) (4 patients in the left breast cancer and 5 in the right breast cancer group), of which 8 patients were classified as NYHA class 1.

Median NT-proBNP, measured after the completed treatment with trastuzumab, was 111.0 (56.7–182) ng/l in the left breast cancer group, and 90.0 (58-170) ng/l in the right breast cancer group (p = 0.545).

Echocardiographic parameters are presented in Table 5. A comparison of echocardiographic parameters showed that the patients who received RT to the left breast/thoracic wall did not have significantly worse systolic or diastolic left ventricular function.

Table 5

Echocardiografic parameters (n = 175)

Table 6

Pericardial effusion

We found significantly more pericardial effusions (9 [11%]) in the left breast cancer group than in the right breast cancer group (1 [1%]) (p = 0.007). The thickness of pericardial effusion was > 1cm in 1 patient in the left breast cancer group, all others were < 1 cm.