Abstract
Background:
Trastuzumab-associated cardiotoxicity remains an issue for patients with HER2-positive breast cancer. This pooled analysis of 3 adjuvant trials investigated the incidence, timing, impact on treatment completion and risk factors for trastuzumab-associated cardiotoxicity.
Methods:
This is an individual patient data level pooled analysis of HERA, NSBAP B-31, and NCCTG 9831 (Alliance Trials). Definitions of cardiac events were as per each individual study.
Results:
A total of 7,445 patients enrolled in the 3 trials were included in the analysis, of which 4,017 were in the trastuzumab and 3,428 in the control (observation) arms, respectively. Median follow-up exceeded 10 years (119.2 – 137.2 months). Nearly all patients (97.4%) in the trastuzumab arms received anthracycline-based chemotherapy. In total, 452 patients in the trastuzumab arms experienced a cardiac event (11.3%), with most being mildly symptomatic or asymptomatic left ventricular ejection fraction (LVEF) decrease (351 patients, 8.7%). Severe congestive heart failure was more common in the trastuzumab arm (2.3%) than in the control arm (0.8%). Most cardiac events occurred during trastuzumab treatment (78.1%) and cardiac events were the main cause of discontinuation across the sample (10.0%); nevertheless, a large majority of patients completed trastuzumab treatment (76.2%). Baseline risk factors that were significantly associated with development of cardiac events were baseline LVEF <60%, hypertension, body mass index >25, age ≥60 and non-Caucasian ethnicity.
Conclusion:
One year of trastuzumab increases the risk of cardiac events, though most consist of asymptomatic or mildly symptomatic LVEF drops. Adjuvant trastuzumab should be considered a safe treatment from a cardiac standpoint for most patients. Trastuzumab-associated cardiotoxicity is the main cause of discontinuation and further research is needed to individualize prevention and management.
Keywords: trastuzumab, LVEF, cardiotoxicity, breast cancer
Introduction
The treatment of patients with human epidermal growth factor receptor 2 (HER2) positive early breast cancer (BC) is based on combining adjuvant chemotherapy to 1 year of the monoclonal antibody trastuzumab [1]. Despite multiple attempts at changing this long-standing standard, either via dual HER2 blockade (lapatinib, pertuzumab), extended treatment (neratinib) or via reduction of the length of trastuzumab treatment (to 9 weeks, 3 months or 6 months), many patients continue to receive trastuzumab alone for 1 year [1]. The trastuzumab standard is based on a series of pivotal phase III trials of women with HER2-positive early breast cancer, including Herceptin Adjuvant (HERA; BIG1–01), National Surgical Adjuvant Breast and Bowel Project (NSABP) B-31, North Central Cancer Treatment Group (NCCTG) 9831 (Alliance Trials) and Breast Cancer International Research Group (BCIRG) 006 [2, 3].
These trials have shown that, similarly to what was seen in the advanced setting, trastuzumab use entails a significantly increased risk of cardiotoxicity, which most often manifests either as an asymptomatic decrease in left ventricular ejection fraction (LVEF) or, less frequently, as symptomatic congestive heart failure (CHF) [4–7]. Existing results, originated by cardiac-safety focused analysis of data from the individual trials, suggest that trastuzumab-associated cardiotoxicity is mostly clinically manageable, that it occurs typically during treatment and is frequently reversible with interruption of trastuzumab and adequate cardiac treatment and, moreover, that it can be predicted by a number of risk factors, which have varied across trials [8].
Nevertheless, the relatively low overall number of cardiac events in the individual trials has precluded a complete understanding of the occurrence and clinical course of trastuzumab-associated cardiotoxicity. The purpose of pooling the individual patient data from the three trials was to investigate potential risk factors for a cardiac event in patients treated with trastuzumab and the impact of risk factors being present based on a larger dataset with a larger number of patients with a cardiac event. These are exploratory analyses to be considered in addition to the published literature from individual trials.
Patients and Methods
Study Design
HERA, NSABP B-31 (NRG Oncology) and NCCTG N9831 (Alliance) were randomized phase III trials which assessed 1 year of trastuzumab vs observation in patients with HER2-positive early breast cancer, sequentially to or concomitantly with chemotherapy (anthracycline/taxane based) [2, 3, 9]. The 2-year trastuzumab arm of HERA was not included in this analysis considering that it’s not standard of care based on the lack of survival benefit shown with trastuzumab prolongation beyond 1 year [10]. The design, eligibility criteria and endpoints of each trial were described in detail in previous publications [2, 3]. Appendix 1 describes the essential aspects of the design of each of the 3 trials and Appendix 2 shows the consort diagram. Of note in NSABP B-31 and N9831, patients were randomized prior to the start of adjuvant therapy, with those experiencing cardiac events or LVEF reductions being prevented from starting trastuzumab use; on the contrary, in HERA, patients were randomized only after the completion of (neo)adjuvant chemotherapy plus radiotherapy (if indicated), and thus trastuzumab treatment was sequential. In HERA, furthermore, patients had to have a LVEF of ≥55% function after the completion of chemotherapy to be eligible for randomization. In NSABP B-31 and NCCTG N9831, LVEF on echocardiography (ECHO)/ Multigated Acquisition (MUGA) scan had to be above the lower limit of the institutional normal.
Definitions of Cardiac Events and Acute Recovery
Each of the 3 trials used different definitions for cardiac events (Appendix 3). Briefly, in HERA, cardiac events could be primary (cardiac death or New York Heart Association class III or IV CHF) or secondary (LVEF decrease that is either asymptomatic or mildly symptomatic). In NSABP B-31 cardiac events were defined as cardiac death or symptomatic CHF with an LVEF drop. In N9831, cardiac events were defined as symptomatic CHF or cardiac death. Acute cardiac recovery was likewise defined differently in each trial (Appendix 4).
Treatment Interruption
In order to safeguard patients from cardiotoxicity, each trial had recommendations for interruption of trastuzumab treatment in case of cardiac events. Briefly, in HERA patients with severe CHF or confirmed LVEF decreases had to interrupt treatment. In NSABP B-31 patients with confirmed symptomatic CHF discontinued treatment and patients with asymptomatic LVEF decrease followed guidelines in the protocol for treatment interruption.. In N9831 patients with substantial LVEF decrease lasting beyond four weeks had to interrupt treatment. Appendix 4 details these rules further.
Cardiac Monitoring
Each study followed a different schedule for cardiac monitoring, particularly for long-term cardiac follow-up after the end of study treatment (Appendix 5). HERA required cardiac monitoring with an ECHO or MUGA to be performed annually between years 5 through 10 [4]. NSABP B-31 required cardiac history forms to be submitted yearly after 5 years [7]. Any indications of possible late cardiac events were followed up with specific queries. In NCCTG N9831, a subset (approximately 33.4%) of patients who went on to trastuzumab (arms B and C) or observation (arm A) had an LVEF evaluation at 6 years [5, 6].
Identifying cardiac events
For NSABP B-31 and N9831, the definition of a cardiac event in the protocol could be considered as cardiac death or severe CHF (NYHA class III/IV). For the pooled analysis, patients in the trastuzumab arm of NSABP B-31 who discontinued trastuzumab as a result of asymptomatic declines in LVEF were considered as having asymptomatic CHF. Patients in trastuzumab arms of N9831 who discontinued trastuzumab as a result of asymptomatic declines in LVEF were considered as having asymptomatic CHF during the trastuzumab treatment period. Further details are described in Appendix 6.
Statistical Analysis
The NSABP B-31 and N9831 datasets consider the intention to treat (ITT) population. It is important to note that for NSABP B-31, the ITT population differs from the primary cardiac safety analysis population, which excluded those who did not initiate trastuzumab based on an unacceptable post-AC cardiac evaluation. The trastuzumab arm from the HERA trial considers patients who received at least one dose of trastuzumab. The HERA patients who were randomized to observation and selectively crossed over to trastuzumab after the release of the trial results were censored at the time of selective crossover for the purpose of this analysis.
Baseline patient and tumor characteristics were described by trial and in the combined arms. Reasons for trastuzumab discontinuation and the incidence of cardiac events were summarized by trial and for the combined trastuzumab arm. The time period of the occurrence of cardiac events in the combined trastuzumab arm was summarized. Cardiac events described as occurring during trastuzumab treatment included those taking place up to 28 days after last administration of trastuzumab. The incidence of any cardiac event was summarized in the combined trastuzumab arm by potential risk factor. The 95% confidence interval (CI) for the incidence of cardiac events was calculated as an exact 95% CI for one sample binomial using the Pearson-Clopper method. The 95% CI for the difference in incidence of cardiac events was calculated using the Hauck-Anderson correction. Baseline factors selected for analysis were chosen based on previously published literature on the subject as well as the current understanding of the underlying physiopathology of cardiotoxicity. The impact of having multiple concurrent risk factors was investigated in the combined trastuzumab arm by calculating the cumulative incidence of any cardiac event. The mean LVEF value was displayed for the combined trastuzumab arm at visits up to month 18 and for the trastuzumab arms by trial up to month 36. The statistical analysis was performed in SAS 9.4, and further details on programming algorithms are in Appendix 6.
Results
Patient Characteristics
The 3 trials randomized a total of 7,445 patients, with 4,017 in the pooled 1-year trastuzumab arm and 3,428 in the pooled observation arm. The median follow-up in the combined trastuzumab arm was 132.0 months (IQR 119.9–142.6 months) (Appendix 7). Table 1 summarizes patient characteristics for both the trastuzumab and observation arms for all patients included in this analysis, as well as for the individual trials (with further detail given in Appendix 8). While in NSABP B-31 all patients received concomitant trastuzumab with taxanes, in N9831 some patients received trastuzumab only after the end of therapy – sequential therapy (67.3%); in HERA all patients were treated sequentially.
Table 1:
Patient Baseline Characteristics
| Trials Combined | HERA | NSABP B-31 | N9831 | |||||
|---|---|---|---|---|---|---|---|---|
| Trastuzumab N=4017 | Control N=3428 | Trastuzumab N=1682 | Control N=1717 | Trastuzumab N=1055 | Control N=1047 | Trastuzumab N=1280 | Control N=664 | |
| NODAL STATUS | ||||||||
| Not assessed* | 191 (4.8%) | 181 (5.3%) | 191 (11.4%) | 181 (10.5%) | 0 | 0 | 0 | 0 |
| 0 positive nodes | 610 (15.2%) | 601 (17.5%) | 538 (32.0%) | 560 (32.6%) | 0 | 0 | 72 (5.6%) | 41 (6.2%) |
| 1–3 positive nodes | 1737 (43.2%) | 1434 (41.8%) | 477 (28.4%) | 499 (29.1%) | 611 (57.9%) | 600 (57.3%) | 649 (50.7%) | 335 (50.5%) |
| ≥4 positive nodes | 1479 (36.8%) | 1211 (35.3%) | 476 (28.3%) | 476 (27.7%) | 444 (42.1%) | 447 (42.7%) | 559 (43.7%) | 288 (43.4%) |
| Unknown | 0 | 1 (0.0%) | 0 | 1 ( 0.1%) | 0 | 0 | 0 | 0 |
| ADJUVANT CHEMOTHERAPY REGIMEN | ||||||||
| No anthracyclines | 101 (2.5%) | 102 (3.0%) | 99 (5.9%) | 101 (5.9%) | 0 | 0 | 2 (0.2%) | 1 (0.2%) |
| Anthracycline but no taxane | 1167 (29.1%) | 1216 (35.5%) | 1138 (67.7%) | 1173 (68.3%) | 29 (2.7%) | 43 (4.1%) | 0 | 0 |
| Anthracycline and taxane | 2747 (68.4%) | 2094 (61.1%) | 445 (26.5%) | 443 (25.8%) | 1024 (97.1%) | 988 (94.4%) | 1278 (99.8%) | 663 (99.8%) |
| None/unknown | 2 (0.0%) | 16 (0.5%) | 0 | 0 | 2 (0.2%) | 16 (1.5%) | 0 | 0 |
| AGE (years) | ||||||||
| < 35 | 285 (7.1%) | 239 (7.0%) | 124 (7.4%) | 130 (7.6%) | 69 (6.5%) | 62 (5.9%) | 92 (7.2%) | 47 (7.1%) |
| 35–39 | 417 (10.4%) | 361 (10.5%) | 183 (10.9%) | 176 (10.3%) | 103 (9.8%) | 108 (10.3%) | 131 (10.2%) | 77 (11.6%) |
| 40–49 | 1381 (34.4%) | 1165 (34.0%) | 566 (33.7%) | 583 (34.0%) | 367 (34.8%) | 352 (33.6%) | 448 (35.0%) | 230 (34.6%) |
| 50–64 | 1634 (40.7%) | 1437 (41.9%) | 688 (40.9%) | 713 (41.5%) | 436 (41.3%) | 451 (43.1%) | 510 (39.8%) | 273 (41.1%) |
| ≥65 | 300 (7.5%) | 226 (6.6%) | 121 (7.2%) | 115 (6.7%) | 80 (7.6%) | 74 (7.1%) | 99 (7.7%) | 37 (5.6%) |
| PATHOLOGICAL TUMOR SIZE (cm) | ||||||||
| Not assessed* | 191 (4.8%) | 181 (5.3%) | 191 (11.4%) | 181 (10.5%) | 0 | 0 | 0 | 0 |
| 0- <2 | 1273 (31.7%) | 1134 (33.1%) | 536 (31.9%) | 536 (31.2%) | 319 (30.2%) | 358 (34.2%) | 418 (32.7%) | 240 (36.1%) |
| ≥2-<5 | 2150 (53.5%) | 1772 (51.7%) | 843 (50.1%) | 845 (49.2%) | 595 (56.4%) | 574 (54.8%) | 712 (55.6%) | 353 (53.2%) |
| ≥5 | 391 (9.7%) | 321 (9.4%) | 103 (6.1%) | 139 (8.1%) | 138 (13.1%) | 111 (10.6%) | 150 (11.7%) | 71 (10.7%) |
| Unknown | 12 (0.3%) | 20 (0.6%) | 9 (0.5%) | 16 (0.9%) | 3 (0.3%) | 4 (0.4%) | 0 | 0 |
| HISTOLOGICAL GRADE OF TUMOR | ||||||||
| 3 | 1750 (43.6%) | 1718 (50.1%) | 1020 (60.6%) | 1025 (59.7%) | 730 (69.2%) | 693 (66.2%) | Not available | Not available |
| 2 | 835 (20.8%) | 884 (25.8%) | 543 (32.3%) | 568 (33.1%) | 292 (27.7%) | 316 (30.2%) | Not available | Not available |
| 1 | 60 (1.5%) | 67 (2.0%) | 37 (2.2%) | 38 (2.2%) | 23 (2.2%) | 29 (2.8%) | Not available | Not available |
| Not assessed | 71 (1.8%) | 78 (2.3%) | 71 (4.2%) | 78 (4.5%) | 0 | 0 | Not available | Not available |
| Unknown | 21 (0.5%) | 17 (0.5%) | 11 (0.7%) | 8 (0.5%) | 10 (0.9%) | 9 (0.9%) | Not available | Not available |
Neo-adjuvant chemotherapy
Discontinuation of Trastuzumab Treatment
As per all study protocols, trastuzumab dose reductions were not allowed during treatment. Table 2 summarizes the occurrence, as well as the causes of trastuzumab treatment discontinuation in total and by study. Cardiac safety was the most common reason for treatment discontinuation in the combined study population (10.0%). In HERA, disease recurrence was the most common cause of trastuzumab discontinuations (5.4%) followed by cardiac safety (5.2%).
Table 2:
Causes of trastuzumab treatment discontinuation
| Trastuzumab Arms Combined N=4017 | HERA N=1682 | NSABP B-31 N=1055 | N9831 N=1280 | |
|---|---|---|---|---|
| Completed trastuzumab | 3059 (76.2%) | 1416 (84.2%) | 720 (68.2%) | 923 (72.1%) |
| Discontinued trastuzumab | 954 (23.7%) | 266 (15.8%) | 331 (31.4%) | 357 (27.9%) |
| Safety reasons | 524 (13.0%) | 123 (7.3%) | 218 (20.7%) | 183 (14.3%) |
| Cardiac safety* | 401 (10.0%) | 88 (5.2%) | 200 (19.0%) | 113 (8.8%) |
| Other safety | 123 (3.1%) | 35 (2.1%) | 18 (1.7%) | 70 (5.5%) |
| Recurrence of disease | 144 (3.6%) | 90 (5.4%) | 23 (2.2%) | 31 (2.4%) |
| Other reason | 286 (7.1%) | 53 (3.2%) | 90 (8.5%) | 143 (11.2%) |
NOTE: The NSABP B-31 dataset and N9831 dataset considers an ITT population
As the NSABP B-31 dataset considers an ITT population, this number includes patients who did not initiate trastuzumab based on an unacceptable post-chemotherapy cardiac evaluation
Incidence of Cardiac Events
Out of 4,017 patients in the trastuzumab arms, 452 (11.3%) experienced a cardiac event (Table 3). Seven cardiac deaths occurred in the combined trastuzumab arms of the trials (0.2%) as compared to 5 in the combined control arms (0.1%). Severe CHF (NYHA III/IV) was more common in the trastuzumab arms with 94 (2.3%) patients with a reported event as compared to 28 (0.8%) in the control arms (Appendix 9). Three hundred and fifty one (8.7%) of the reported events in the trastuzumab arms was asymptomatic or mildly symptomatic (mild heart failure {NYHA II} or asymptomatic drops in LVEF {NYHA I}). In terms of timing of occurrence in relation to trastuzumab, 78.1% of reported cardiac events occurred during trastuzumab use and only 10.6% after anti-HER2 treatment completion. The remaining 11.3% of cardiac events occurred in patients who did not start trastuzumab (i.e during chemotherapy). Appendix 10 summarizes the timing of events in relation to trastuzumab use. The incidence of cardiac events was numerically higher in the concurrent trastuzumab arms as compared to the sequential arms (Appendix 11).
Table 3:
Incidence of cardiac events in the trastuzumab arms
| Trastuzumab Arms Combined N=4017 | HERA N=1682 | NSABP B-31 N=1055 | N9831 N=1280 | |
|---|---|---|---|---|
| Any cardiac event* | 452 (11.3%) | 92 (5.5%) | 205 (19.4%) | 155 (12.1%) |
| Did not start trastuzumab | 51 (1.3%) | - | 49 (4.6%) | 2 (0.2%) |
| During trastuzumab treatment** | 353 (8.8%) | 67 (4.0%) | 144 (13.6%) | 142 (11.1%) |
| After trastuzumab treatment | 48 (1.2%) | 25 (1.5%) | 12 (1.1%) | 11 (0.9%) |
| Cardiac death | 7 (0.2%) | 3 (0.2%) | 2 (0.2%) | 2 (0.2%) |
| Severe CHF (NYHA class III/IV) | 94 (2.3%) | 15 (0.9%) | 41 (3.9%) | 38 (3.0%) |
| Did not start trastuzumab | 3 (0.1%) | - | 2 (0.2%) | 1 (0.1%) |
| During trastuzumab treatment** | 71 (1.8%) | 12 (0.7%) | 28 (2.7%) | 31 (2.4%) |
| After trastuzumab treatment | 20 (0.5%) | 3 (0.2%) | 11 (1.0%) | 6 (0.5%) |
| Asymptomatic or mildly symptomatic (NYHA class II) | 351 (8.7%) | 74 (4.4%) | 162 (15.4%) | 115 (9.0%) |
| Did not start trastuzumab | 46 (1.1%) | - | 46 (4.4%) | 0 |
| During trastuzumab treatment** | 281 (7.0%) | 55 (3.3%) | 116 (11.0%) | 110 (8.6%) |
| After trastuzumab treatment | 24 (0.6%) | 19 (1.1%) | 0 | 5 (0.4%) |
The precise definition of the types of cardiac event was not consistent across the trials
During trastuzumab treatment includes up to 28 days after last administration of trastuzumab
In NSABP B31, the timing of the asymptomatic or mildly symptomatic CHF from randomization is set as the timing of last administration of trastuzumab or, for patients who did not start trastuzumab, as the last administration of chemotherapy.
The precise date of asymptomatic or mildly symptomatic CHF in N9831 could not be determined. The timing can be approximated from trastuzumab treatment discontinuation.
Acute Recovery from Cardiac Events
Unfortunately, the heterogeneity existing among the definitions of acute recovery among trials did not allow for a pooled analysis of this particular study endpoint.
Risk Factors for Cardiac Events
Nearly all patients in this study received anthracycline-based chemotherapy (97.5%), precluding analysis of the influence of anthracycline as a risk factor for cardiac events. Identified risk factors for the development of cardiac events in the combined sample were LVEF at baseline <60%; history of hypertension, high body mass index (BMI) (>25), age ≥60 and being ethnically non-Caucasian (Table 4). There were too few patients with a history of diabetes to draw conclusions, despite the numerical risk increase. Having 2 or more risk factors increases the risk of experiencing a cardiac event over time, as compared to 0 or 1 risk factor (Figure 1). By 12 months, the cumulative incidence of having a cardiac event in patients with two or more risk factors is approximately twice the cumulative incidence in patients with zero risk factors (Figure 1). Further details on programming algorithms for Figure 1 are described in Appendix 6.
Table 4:
Risk Factors for cardiac event - Trastuzumab Arms Combined
| Variable | n | Incidence of Cardiac Event* | 95% CI for Incidence** | Difference in Incidence (%) | 95% CI for the Difference*** |
|---|---|---|---|---|---|
| SCREENING LVEF**** | |||||
| Lower limit ≤ LVEF < 60% | 1026 | 157 (15.30) | (13.15, 17.65) | 5.41 | (2.91, 7.91) |
| ≥ 60% | 2973 | 294 (9.89) | (8.84, 11.02) | ||
| 60% ≤ LVEF < 65% | 1181 | 128 (10.84) | (9.12, 12.75) | 1.57 | (−0.69, 3.84) |
| ≥65% | 1792 | 166 (9.26) | (7.96, 10.70) | ||
| PREVIOUS RADIOTHERAPY | |||||
| Left sided | 1537 | 163 (10.61) | (9.11, 12.25) | −0.74 | (−3.35, 1.86) |
| None | 934 | 106 (11.35) | (9.39, 13.56) | ||
| HYPERTENSION | |||||
| Yes | 729 | 112 (15.36) | (12.82, 18.19) | 5.03 | (2.15, 7.92) |
| No | 3282 | 339 (10.33) | (9.31, 11.42) | ||
| HIGH BODY MASS INDEX | |||||
| >25 | 2338 | 311 (13.30) | (11.95, 14.75) | 4.87 | (2.84, 6.91) |
| 20 ≤ BMI ≤25 | 1412 | 119 (8.43) | (7.03, 10.00) | ||
| SMOKING STATUS | |||||
| Yes | 349 | 40 (11.46) | (8.32, 15.28) | 0.71 | (−3.01, 4.42) |
| No | 2380 | 256 (10.76) | (9.54, 12.07) | ||
| DIABETES | |||||
| Yes | 80 | 17 (21.25) | (12.89, 31.83) | 10.73 | (1.00, 20.45) |
| No | 2651 | 279 (10.52) | (9.38, 11.76) | ||
| AGE (years) | |||||
| ≥ 50 to ≤ 59 | 1280 | 144 (11.25) | (9.57, 13.11) | 1.50 | (−0.68, 3.69) |
| < 50 | 2083 | 203 (9.75) | (8.51, 11.10) | ||
| ≥60 | 654 | 105 (16.06) | (13.32, 19.10) | 5.74 | (2.66, 8.81) |
| < 60 | 3363 | 347 (10.32) | (9.31, 11.40) | ||
| RACE | |||||
| Caucasian | 3374 | 360 (10.67) | (9.65, 11.76) | −3.57 | (−6.55, −0.59) |
| Non-Caucasian | 639 | 91 (14.24) | (11.62, 17.19) |
The precise definition of the types of cardiac event was not consistent across the trials
Exact 95% CI for one sample binomial using Pearson-Clopper method
Approximate 95% CI for difference of two rates using Hauck-Anderson correction
The LVEF eligibility criteria was not consistent across the trials
LVEF = left ventricular ejection fraction
Figure 1:
cumulative incidence plot of any cardiac events in the Trastuzumab combined arm based on competing risks
LVEF Value Over time
Figure 2A depicts mean LVEF value over time for patients in the trastuzumab arms. Figure 2B depicts mean LVEF value over time by trial. The graph shows that mean LVEF falls progressively during trastuzumab treatment but recovers to near pre-treatment levels approximately 6 months after the end of trastuzumab treatment. Further details on programming algorithm for Figure 2 are in Appendix 6.
Figure 2A:
plot of mean LVEF value over time up to month 18
Figure 2B:
plot of mean LVEF value over time up to month 36
Discussion
To our knowledge, this study, including individual data originating from 3 well conducted randomized clinical trials with long term follow-up is the largest study investigating trastuzumab-associated cardiotoxicity in the adjuvant setting. Patients in the combined trastuzumab arms had a numerically increased incidence of cardiac events, as compared to patients in the combined observation arms, though these cardiac events in most cases were asymptomatic or mildly symptomatic and occurred during trastuzumab treatment.
The pathophysiology of anti-HER2 treatment associated cardiac dysfunction is not yet completely understood [8]. HER2 seems to play an important role in cardiac homeostasis, particularly when the myocardium suffers physiological stress [11]. This explains why anti-HER2 treatment is particularly cardiotoxic in situations in which myocardial oxidative stress is heightened, i.e. during or immediately after the use of anthracyclines, in patients with arterial hypertension or pre-existing cardiac diseases [12–14]. In contrast to anthracyclines, which have the potential to induce myocyte cell death, inhibition of HER2 signaling leads to cardiac dysfunction either by affecting the cardiomyocyte contraction or by interfering with mitochondrial metabolism [15, 16].
This framework for comprehending trastuzumab-associated cardiotoxicity is further strengthened by currently available data on cardiac damage biomarkers, which suggest that troponins are elevated early in trastuzumab treatment and almost exclusively in patients who are pre-treated with anthracycline-based chemotherapy [17, 18]. In our study, this is well reflected in the progressive yet slight (mean still above 60%) reduction in mean LVEF for the combined population, followed by the recovery approximately 6 months after the end of treatment.
Clinically, trastuzumab-associated cardiotoxicity manifests most often as asymptomatic LVEF drops and mildly symptomatic CHF, with cardiac death being an exceedingly rare occurrence [8]. Additionally, most cardiac events occur during the trastuzumab treatment phase and are reversible. Heterogeneity in the definition of acute recovery among trials, as well as the differences in long-term cardiac follow-up has precluded combined analysis of the recovery data. Results from the individual trials suggest a high chance of recovery with interruption of treatment and adequate cardiac treatment [4, 5, 7]. For example, 81.2% of patients in HERA who experienced a cardiac event eventually recovered (median time to recovery was 7.2 months).
Ten percent of patients in the combined trastuzumab arms discontinued trastuzumab treatment due to cardiac events. Discontinuation, particularly if it occurs early in treatment (before completing 6 months), can compromise outcomes. Therefore, in our view, discontinuation of anti-HER2 treatment due to cardiotoxicity is not always necessary. For many patients, especially those with asymptomatic LVEF drops, maintaining trastuzumab use while conducting closer follow-up with a cardiologist and adequate cardiac treatment may be the appropriate first management option. Short interruptions with resumption after intervention (with ACE inhibitors and/or beta-blockers) and LVEF recovery is also a valid option [19]. The results of the PERSEPHONE trial, which suggest the non-inferiority of 6 months of adjuvant trastuzumab vs 1 year with a substantial reduction in cardiac events (12% to 9%) and non-inferior survival outcomes also points towards the possibility of halting treatment after 6 months, though this remains a hotly debated topic [20, 21].
Heterogeneity in the definition for cardiac events (Appendix 3), in the point of patient inclusion and in the follow up schedule used in the different studies are among the challenges facing the field of cardiac-oncology and the authors of guidelines today (Appendix 4) and are therefore major limitations of our study. Although the European Society of Cardiology (ESC) has proposed definitions for cardiac events to be used in clinical trials, including those testing anti-HER2 agents, these have thus far not been extensively used [19].
Several unanswered research questions remain, including: (1) what is the long term risk of cardiac dysfunction for patients receiving trastuzumab (and/or newer anti-HER agents); notably when patients suffer from pre-existing cardiac illness; (2) what is the long term clinical course of patients who experienced dysfunction during treatment, notably when they eventually develop other cardiac diseases (such as coronary insufficiency). It is hoped that trials such as APHINITY, in which pertuzumab plus trastuzumab vs trastuzumab was tested, will provide further cardiac data in the future [22]. Additionally, registry studies such as SystHERs () should help evaluating the cardiotoxicity associated with anti-HER2 agents in “real life” populations. Currently available “real life” retrospective cohorts suggest that trastuzumab-associated cardiotoxicity is more common phenomena than data coming from clinical trials results suggest, notably in older patients [8]. A recent study, including over 16.000 patients with early breast cancer from the Truven health Marketscan database detected 692 patients who developed CHF during or after cancer treatment (4.2% of the entire sample). For patients who received trastuzumab, the incidence was 8.3% (vs 2.7% for those who did not use trastuzumab, p<0.001), with nearly all events being identified within the first year [23].
The possibility of trastuzumab-associated cardiotoxicity should be taken into consideration when making decisions regarding adjuvant trastuzumab use despite the low overall incidence of cardiac events in our study. Our study identified a few risk factors, including LVEF at baseline <60%, history of hypertension, high BMI (>25), and age ≥60. Though the eligibility criteria of HERA precluded a combined analysis of patients with LVEF between 50% and 54%, it is relevant to mention that these lower LVEF levels were significantly associated with cardiac events in a cardiac-specific analysis of NSABP B-31 [7]. Considering the outstanding long-term benefit of adjuvant trastuzumab, excluding patients from treatment based solely on one of these risk factors should not be a standard approach. As our results show, multiple risk factors substantially increase the risk of a cardiac event. Hence, when multiple factors (3 or more) are present a non-anthracycline regimen should be preferred, such as the taxane, carboplatin and trastuzumab regimen tested in BCIRG-006 (with the possibility of adding pertuzumab or taxane and trastuzumab tested in the APT trial (for lower risk patients) as well as closer follow-up and pro-active measures to control risk factors [24, 25]. In addition, cardioprotective pharmacological interventions, as per the results of the PRADA, MANTICORE or the carvedilol and lisinopril trials, cannot be considered as standard at this point, but should be the focus of further studies [26–28]. Additionally, based on the results of PERSEPHONE, a shorter (6 months) adjuvant trastuzumab regimen may be considered for a certain number of patients in some countries where trastuzumab is not broadly available.
In conclusion, the results of our study, which is the largest individual patient level combined analysis concerning trastuzumab-associated cardiotoxicity performed to date shows an increase in the incidence of cardiac events, though most cases are asymptomatic or mildly symptomatic nature and occurring during trastuzumab treatment. One-year of adjuvant trastuzumab should be considered as a safe treatment from a cardiac stand-point for most patients. Collaboration between cardiologists and oncologists is necessary to advance the field and reduce the risk of cardiotoxicity during cancer treatments.
Supplementary Material
Appendices
Appendix 1 (Figure): Individual trial designs
Appendix 2 (Figure): Consort Diagram
Appendix 3 (Table): Individual trial cardiac event definitions
Appendix 4 (Table): Definitions for Acute Recovery and Recommendations for Interruption of Trastuzumab Treatment due to Cardiotoxicity
Appendix 5 (Table): Individual trial cardiac follow up
Appendix 6 (Table): Definitions in programming algorithms
Appendix 7 (Table): Summary of Median Follow-Up Time
Appendix 8 (Table): Patient and Tumor Characteristics (Expanded)
Appendix 9 (Table): Incidence of Severe CHF and Cardiac Death (Trials Combined)
Appendix 10 (Table): Period of occurrence of cardiac events in relationship to trastuzumab use
Appendix 11 (Table): Incidence of Cardiac Event (Trastuzumab Sequential and Concurrent)
Acknowledgements
The authors would like to acknowledge the teams of HERA, NSABP B −31, NCCTCG N9831 and PACS004 for their help in making this project possible.
Financial Disclosures
The NSABP B-31 Trial was supported by NCI grants: U10CA180868, −180822, UG1–189867, and U24–196067.
Footnotes
Publisher's Disclaimer: This Author Accepted Manuscript is a PDF file of an unedited peer-reviewed manuscript that has been accepted for publication but has not been copyedited or corrected. The official version of record that is published in the journal is kept up to date and so may therefore differ from this version.
Compliance with Ethical Standards
Conflict of Interest
Evandro de Azambuja has received honoraria and advisory board from Roche/GNE and Seattle Genetics; travel grants from Roche/GNE and GSK/Novartis and research grants to his Institute: Roche/GNE; Astra-Zeneca, GSK/Novartis, Servier
Noam Pondé has received honoraria from AZ and Eli Lilly and travel grants Eli Lilly and Novartis.
Marion Procter has declared that her institution received funding from Roche in respect to the APHINITY trial.
Matteo Lambertini has received honoraria Therame and consulting fee from TEVA
Pryia Rastogi has received travel and accommodations supported by Lilly, AZ, and GNE/Roche
Martine Piccart has declared board Member (Scientific Board) : Oncolytics, Radius; consultant (honoraria) : AstraZeneca, Camel-IDS, Crescendo Biologics, Debiopharm, G1 Therapeutics, Genentech, Huya, Immunomedics, Lilly, Menarini, MSD, Novartis, Odonate, Periphagen, Pfizer, Roche, Seattle Genetics; research grants to her Institute: AstraZeneca, Lilly, MSD, Novartis, Pfizer, Radius, Roche-Genentech, Servier, Synthon
Thomas Suter has declared research Support from Novartis and Amgen; consultant from Pfizer and advisory Board from Alnylam
Richard D Gelber declares that his institution receives support for his salary from Roche, AstraZeneca, Merck, Novartis, Ipsen, Celgene, Pfizer, and Ferring.
Dimitrios Zardavas has declared employment from BMS
Karla V. Ballman has declared consulting or Advisory Role from ARIAD; Medtronic; Takeda; Agenus; Patents, Royalties, Other Intellectual Property: Prostate cancer signature patent (Inst); Expert Testimony - Janssen Oncology; Lilly
Lise Roca, Alvaro Moreno Aspitia and Reena Cecchini have no conflict of interest to declare
Ethical approval
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.
Informed consent: Informed consent was obtained from all individual participants included in each of the studies (HERA, NSABP B-31 (NRG Oncology) and NCCTG N9831 (Alliance)).
References
- 1.Pondé NF, Zardavas D, Piccart M (2018) Progress in adjuvant systemic therapy for breast cancer. Nature Reviews Clinical Oncology. 10.1038/s41571-018-0089-9 [DOI] [PubMed] [Google Scholar]
- 2.Piccart-Gebhart MJ, Procter M, Leyland-Jones B, et al. (2005) Trastuzumab after adjuvant chemotherapy in HER2-positive breast cancer. N Engl J Med 353:1659–1672. 10.1056/NEJMoa052306 [DOI] [PubMed] [Google Scholar]
- 3.Perez EA, Romond EH, Suman VJ, et al. (2014) Trastuzumab Plus Adjuvant Chemotherapy for Human Epidermal Growth Factor Receptor 2-Positive Breast Cancer: Planned Joint Analysis of Overall Survival From NSABP B-31 and NCCTG N9831. Journal of Clinical Oncology 32:3744–3752. 10.1200/JCO.2014.55.5730 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.de Azambuja E, Procter MJ, van Veldhuisen DJ, et al. (2014) Trastuzumab-associated cardiac events at 8 years of median follow-up in the Herceptin Adjuvant trial (BIG 1–01). J Clin Oncol 32:2159–2165. 10.1200/JCO.2013.53.9288 [DOI] [PubMed] [Google Scholar]
- 5.Advani PP, Ballman KV, Dockter TJ, et al. (2015) Long-Term Cardiac Safety Analysis of NCCTG N9831 (Alliance) Adjuvant Trastuzumab Trial. J Clin Oncol 10.1200/JCO.2015.61.8413 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Advani PP, Ballman KV, Dockter TJ, et al. (2016) Long-Term Cardiac Safety Analysis of NCCTG N9831 (Alliance) Adjuvant Trastuzumab Trial. Journal of Clinical Oncology 34:581–587. 10.1200/JCO.2015.61.8413 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Romond EH, Jeong J-H, Rastogi P, et al. (2012) Seven-Year Follow-Up Assessment of Cardiac Function in NSABP B-31, a Randomized Trial Comparing Doxorubicin and Cyclophosphamide Followed by Paclitaxel (ACP) With ACP Plus Trastuzumab As Adjuvant Therapy for Patients With Node-Positive, Human Epidermal Growth Factor Receptor 2-Positive Breast Cancer. Journal of Clinical Oncology 30:3792–3799. 10.1200/JCO.2011.40.0010 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Pondé NF, Lambertini M, de Azambuja E (2016) Twenty years of anti-HER2 therapy-associated cardiotoxicity. ESMO Open 1:e000073 10.1136/esmoopen-2016-000073 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Romond EH, Perez EA, Bryant J, et al. (2005) Trastuzumab plus Adjuvant Chemotherapy for Operable HER2-Positive Breast Cancer. New England Journal of Medicine 353:1673–1684. 10.1056/NEJMoa052122 [DOI] [PubMed] [Google Scholar]
- 10.Goldhirsch A, Gelber RD, Piccart-Gebhart MJ, et al. (2013) 2 years versus 1 year of adjuvant trastuzumab for HER2-positive breast cancer (HERA): an open-label, randomised controlled trial. Lancet 382:1021–1028. 10.1016/S0140-6736(13)61094-6 [DOI] [PubMed] [Google Scholar]
- 11.Hervent A-S, De Keulenaer GW (2012) Molecular Mechanisms of Cardiotoxicity Induced by ErbB Receptor Inhibitor Cancer Therapeutics. International Journal of Molecular Sciences 13:12268–12286. 10.3390/ijms131012268 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Crone SA, Zhao Y-Y, Fan L, et al. (2002) ErbB2 is essential in the prevention of dilated cardiomyopathy. Nat Med 8:459–465. 10.1038/nm0502-459 [DOI] [PubMed] [Google Scholar]
- 13.Grazette LP, Boecker W, Matsui T, et al. (2004) Inhibition of ErbB2 causes mitochondrial dysfunction in cardiomyocytes: implications for herceptin-induced cardiomyopathy. J Am Coll Cardiol 44:2231–2238. 10.1016/j.jacc.2004.08.066 [DOI] [PubMed] [Google Scholar]
- 14.ElZarrad MK, Mukhopadhyay P, Mohan N, et al. (2013) Trastuzumab Alters the Expression of Genes Essential for Cardiac Function and Induces Ultrastructural Changes of Cardiomyocytes in Mice. PLoS ONE 8:e79543 10.1371/journal.pone.0079543 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Ewer MS, Lippman SM (2005) Type II Chemotherapy-Related Cardiac Dysfunction: Time to Recognize a New Entity. Journal of Clinical Oncology 23:2900–2902. 10.1200/JCO.2005.05.827 [DOI] [PubMed] [Google Scholar]
- 16.Ewer MS, Ewer SM (2015) Cardiotoxicity of anticancer treatments. Nature Reviews Cardiology 12:547–558. 10.1038/nrcardio.2015.65 [DOI] [PubMed] [Google Scholar]
- 17.Zardavas D, Suter TM, Van Veldhuisen DJ, et al. (2017) Role of Troponins I and T and N-Terminal Prohormone of Brain Natriuretic Peptide in Monitoring Cardiac Safety of Patients With Early-Stage Human Epidermal Growth Factor Receptor 2-Positive Breast Cancer Receiving Trastuzumab: A Herceptin Adjuvant Study Cardiac Marker Substudy. J Clin Oncol 35:878–884. 10.1200/JCO.2015.65.7916 [DOI] [PubMed] [Google Scholar]
- 18.Ponde N, Bradbury I, Lambertini M, et al. (2017) Cardiac biomarkers for early detection and prediction of trastuzumab and/or lapatinib-induced cardiotoxicity in patients with HER2-positive early-stage breast cancer: a NeoALTTO sub-study (BIG 1–06). Breast Cancer Research and Treatment. 10.1007/s10549-017-4628-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Zamorano JL, Lancellotti P, Rodriguez Muñoz D, et al. (2017) 2016 ESC Position Paper on cancer treatments and cardiovascular toxicity developed under the auspices of the ESC Committee for Practice Guidelines: The Task Force for cancer treatments and cardiovascular toxicity of the European Society of Cardiology (ESC). European Journal of Heart Failure 19:9–42. 10.1002/ejhf.654 [DOI] [PubMed] [Google Scholar]
- 20.Earl HM, Hiller L, Vallier A-L, et al. (2019) 6 versus 12 months of adjuvant trastuzumab for HER2-positive early breast cancer (PERSEPHONE): 4-year disease-free survival results of a randomised phase 3 non-inferiority trial. Lancet 10.1016/S0140-6736(19)30650-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Pivot X, Romieu G, Debled M, et al. (2019) 6 months versus 12 months of adjuvant trastuzumab in early breast cancer (PHARE): final analysis of a multicentre, open-label, phase 3 randomised trial. Lancet 10.1016/S0140-6736(19)30653-1 [DOI] [PubMed] [Google Scholar]
- 22.von Minckwitz G, Procter M, de Azambuja E, et al. (2017) Adjuvant Pertuzumab and Trastuzumab in Early HER2-Positive Breast Cancer. New England Journal of Medicine 377:122–131. 10.1056/NEJMoa1703643 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Henry ML, Niu J, Zhang N, et al. (2018) Cardiotoxicity and Cardiac Monitoring Among Chemotherapy-Treated Breast Cancer Patients. JACC: Cardiovascular Imaging 11:1084–1093. 10.1016/j.jcmg.2018.06.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Slamon D, Eiermann W, Robert N, et al. (2011) Adjuvant Trastuzumab in HER2-Positive Breast Cancer. New England Journal of Medicine 365:1273–1283. 10.1056/NEJMoa0910383 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Tolaney SM, Guo H, Pernas S, et al. (2019) Seven-Year Follow-Up Analysis of Adjuvant Paclitaxel and Trastuzumab Trial for Node-Negative, Human Epidermal Growth Factor Receptor 2–Positive Breast Cancer. JCO 37:1868–1875. 10.1200/JCO.19.00066 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Gulati G, Heck SL, Ree AH, et al. (2016) Prevention of cardiac dysfunction during adjuvant breast cancer therapy (PRADA): a 2 × 2 factorial, randomized, placebo-controlled, double-blind clinical trial of candesartan and metoprolol. European Heart Journal ehw022 10.1093/eurheartj/ehw022 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Pituskin E, Mackey JR, Koshman S, et al. (2017) Multidisciplinary Approach to Novel Therapies in Cardio-Oncology Research (MANTICORE 101–Breast): A Randomized Trial for the Prevention of Trastuzumab-Associated Cardiotoxicity. Journal of Clinical Oncology 35:870–877. 10.1200/JCO.2016.68.7830 [DOI] [PubMed] [Google Scholar]
- 28.Guglin M, Krischer J, Tamura R, et al. (2019) Randomized Trial of Lisinopril Versus Carvedilol to Prevent Trastuzumab Cardiotoxicity in Patients With Breast Cancer. Journal of the American College of Cardiology 73:2859–2868. 10.1016/j.jacc.2019.03.495 [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Appendices
Appendix 1 (Figure): Individual trial designs
Appendix 2 (Figure): Consort Diagram
Appendix 3 (Table): Individual trial cardiac event definitions
Appendix 4 (Table): Definitions for Acute Recovery and Recommendations for Interruption of Trastuzumab Treatment due to Cardiotoxicity
Appendix 5 (Table): Individual trial cardiac follow up
Appendix 6 (Table): Definitions in programming algorithms
Appendix 7 (Table): Summary of Median Follow-Up Time
Appendix 8 (Table): Patient and Tumor Characteristics (Expanded)
Appendix 9 (Table): Incidence of Severe CHF and Cardiac Death (Trials Combined)
Appendix 10 (Table): Period of occurrence of cardiac events in relationship to trastuzumab use
Appendix 11 (Table): Incidence of Cardiac Event (Trastuzumab Sequential and Concurrent)