Trastuzumab prolongs overall survival in patients with brain metastases from Her2 positive breast cancer
Background: Brain metastases are frequently encountered in Her2 positive advanced breast cancer. It is still not clear, if trastuzumab treatment should be continued following their diagnosis. In this analysis we evaluated if trastuzumab was able to influence time to in-brain progression (TTP) and overall survival (OS). For this reason, we compared patients who continued on trastuzumab with a historical control group. Patients and Methods: Seventeen Her2 positive patients receiving whole brain radiotherapy for brain metastases and continuing on trastuzumab were identified. As historical control group, thirty-six patients treated before 2002 were identified from a breast cancer database. We performed a multivariate analysis (Cox regression) to explore which factors were potentially able to significantly influence TTP and OS. Results: Median TTP was 6 months, range 1–33+ months. Median OS was 7 months, range 1–38 months. Seventeen patients received trastuzumab after WBRT. Factors associated with prolonged TTP were KPS (p = 0.001), and intensified local treatment (p = 0.004). A trend towards longer TTP was observed in patients treated with trastuzumab (p = 0.068). OS was significantly influenced by KPS (p < 0.001), and continued antibody therapy (p = 0.001). Conclusion: Two parameters were significantly associated with prolonged OS: KPS and trastuzumab. While there was a trend towards prolonged TTP in patients with trastuzumab treatment after WBRT, this did not reach statistical significance. It appears therefore reasonable to suggest continuation of antibody therapy in patients with good performance status despite disease spreading to the brain. Concerning activity of trastuzumab in brain metastases themselves, no final conclusion is possible.
KeywordsAdvanced breast cancer Brain metastases Her2 positive disease Trastuzumab Whole brain radiotherapy
Breast cancer is still the main reason of cancer morbidity in women in most countries (Boyle and Ferlay , Smigal et al. ). While incidence is rising, mortality has declined over the last decade due to screening and improved adjuvant therapy options. Still, even in stage I and II disease, approximately 30% of patients are expected to experience a relapse (Mollick and Carlson ). Metastatic breast cancer remains an incurable disease. Today however, individualized, risk adapted palliative treatment options depending on tumor biology, symptoms, metastatic sites, time to disease recurrence, and patient wish, is available. In Her2 positive disease, which is associated with high recurrence rate and poor outcome (Slamon et al. , Boss et al. , Paik et al. , Kallioniemi et al. ), trastuzumab-based regimens, are well established (Slamon et al. , Pegram et al. ). Trastuzumab is a recombinant, humanized monoclonal antibody (rhMab4D5), targeting the extra cellular domain of Her2 (human EGFR related). Via different mechanisms, the substance inhibits Her2 receptor signaling and causes cell degradation. Mechanisms identified comprise inhibition of cleavage of the extra cellular domain, internalization and degradation of the Her2 receptor protein, decrease of cyclin dependent kinase 2 (CDK2) activity via p27 induction, inhibition of DNA repair, and antibody dependent cellular cytotoxicity (ADCC) (Clynes et al. , Nahta et al. ).
Since the introduction of trastuzumab, different groups reported a higher incidence of brain metastases in their patients. This is usually believed to stem from longer survival in patients with metastatic disease. Also, most anti-cancer drugs including monoclonal antibodies cannot pass an intact blood-brain barrier (Bendell et al. , Clayton et al. , Shmueli et al. , Herrero et al. ).
Local therapy options in brain metastases comprise neurosurgical resection, whole brain radiotherapy (WBRT), and boost irradiation (Herrero et al. , DiStefano et al. , Lagerwaard et al. , Mahmoud-Ahmed et al. ). Resection is usually utilized when there is a single lesion only. Further, neurosurgical interventions are necessary in case of severe neurological dysfunction, or occlusive hydrocephalus, or a lesion of unknown histology (Lee , Fokstuen et al. , Hendrickson et al. , Bindal et al. ). In case of multiple metastases, treatment usually consists of WBRT alone. Different total doses and fractions are utilized (Borgelt et al. , Lohr et al. ). Without major differences in efficacy, those regimens yield a symptomatic response in about 50% of patients, resulting in a median survival of up to 6 months (Sneed et al. , Broadbent et al. , Lutterbach et al. ). In patients with one to three intra-cerebral metastatic sites, boost irradiation is often applied. This greatly enhances progression free survival, while a possible influence on OS is still debated (Shaw et al. , Loeffler et al. , Lederman et al. , Noel et al. , Becker et al. ).
Recently published studies reported a higher OS in patients receiving immunotherapy with trastuzumab after WBRT (Lower et al. , Kirsch et al. ). Those analyses however are limited by two factors. First, patients with Her2 negative disease were included as control. This is potentially biasing results, as Her2 positive and negative breast cancers are distinct biological entities of disease. Further it was not described, why Her2 positive patients in the control group did not receive further trastuzumab (e.g. KPS < 70%).
In this retrospective analysis, we investigated factors potentially influencing local time to progression (TTP) and overall survival (OS) in a population of patients with brain metastases from Her2 positive breast cancer. We investigated, if patients derived a benefit in terms of TTP and OS from continuation of trastuzumab after WBRT. For this reason, we compared patients receiving further antibody therapy with a historical control group of Her2 positive individuals.
Patients and Methods
All data were collected at the Department of Medicine 1 and Cancer Centre, Clinical Division of Oncology, and the Department of Radiotherapy, Medical University of Vienna, Vienna, Austria. The retrospective analysis was performed in accordance with the ethical regulations of the Medical University of Vienna.
Data from all patients treated from 2003 until 2006 with WBRT for brain metastases from Her2 positive breast cancer, who received further trastuzumab were retrieved from patient charts and a breast cancer database. Patients without ongoing antibody therapy were not included, because they were either deemed resistant to trastuzumab, or had a KPS too low for further systemic therapy. A total number of seventeen patients were identified. In a second step, we retrieved data from all patients with Her2 positive disease and brain metastasis, who were treated until 2002. This cut-off was chosen, as in 2002 routine application of trastuzumab after WBRT was initiated. A total number of fifty-five patients were identified. Fourteen patients were excluded, as they presented with KPS < 70%, or because not all necessary data were available. Another five patients had to be excluded because Her2 status was analyzed using methods other than HercepTest® or fluorescence in situ hybridization (FISH). Therefore, a total of 36 patients were available as control group for this analysis.
WBRT was applied at a 6 MV LINAC (linear accelerator) or a cobalt machine by lateral opposed fields. Total dose was 30 Gy, applied in 10 fractions within two weeks. In case of one to three metastases ≤2 cm, an additional stereotactic boost was applied at a Gamma knife with a dose of 16 to 20 Gy on the 50% isodose, or at a 6 MV LINAC with 20 Gy applied on the 80% isodose. In case of tumor size ≥2.5 cm two times 10 Gy were applied at a 6 MV LINAC. In some cases, pre-radiotherapy tumor resection has been performed.
Her2 status was assessed using the HercepTest® (Dako A/S, Glostrup, Denmark) or dual color fluorescent in situ hybridization (FISH; PathVision® HER2 DNA probe kit, Vysis Inc., Downers Grove, IL, USA). Tumors were classed as Her2-positive if they had a staining intensity of +++ on the HercepTest®; if a score of ++ was gained, the tumors were reanalyzed using FISH. Tumors with Her2 gene amplification again were deemed eligible for trastuzumab. Trastuzumab was administered at a dose of 6 mg/kg body weight every three weeks following a loading dose of 8 mg/kg body weight on day one (Leyland-Jones et al. ).
Cranial CT or MRI scans were performed one and three months after WBRT, and every three months thereafter. From 2000 onwards, patients routinely had MRI scans for staging evaluation.
TTP was defined as time from end of WBRT until documented disease progression by MRI scans of the brain (when available) or cranial CT scans. OS after WBRT was defined as interval from WBRT until death of any cause. TTP and OS were estimated using the Kaplan-Meier product limit method; results were compared with the log-rank test. To analyze factors significantly influencing potentially TTP and OS, a multivariate analysis (Cox regression model) was performed. For comparison between the subgroups, the independent t-test was used. p values < 0.05 were considered to indicate statistic significance. The following factors were included in the Cox regression model as independent variables: age (≥65 vs. <65 years), grading, estrogen and progesterone receptor status, invasive ductal or lobular carcinoma, initial tumor stage (metastatic versus localized), metastatic sites (visceral versus non-visceral), KPS (=70% vs. >70%), intensified local treatment (boost irradiation, neuro-surgical resection vs. WBRT only), ongoing systemic treatment after WBRT (no further systemic therapy, chemotherapy only, trastuzumab). Data were analyzed as of December 2006. All statistics were calculated using statistical package for the social sciences (SPSS®) 12.0 software (SPSS Inc., Chicago, IL, USA).
A total of fifty-three patients (pts) are evaluable for this retrospective analysis. Median age was 49 years (range 27–79 years), and median KPS was 80%, with 14 patients having a KPS of <80%. Minimum KPS of all pts included was 70%. Twelve patients had metastases only to the bones and/or the soft tissue, while thirty-six had also visceral involvement. The remaining five patients had cerebral metastases as only site of tumor spreading. Thirty-eight pts had more than one metastatic site. Trastuzumab before WBRT was administered in 36 pts (67.9%), and median duration was 12 months, range 3–43 months. A boost irradiation or neurosurgical resection was utilized in 25 individuals (47.1%). After local therapy, seventeen (31.1%) pts continued on antibody therapy, with a median duration of 8 months, range 1–37 months.
N = 53
Median age (years) (range)
49 years (27–79)
Patients ≥65 years
Karnofsky Performance Score median
Patients KPS > 70%
Estrogen receptor/progesterone receptor positive
Her2 positive IHC/FISH*
Invasive ductal / lobular carcinoma
Bones/soft tissue only
Metastatic sites median (range)
2 sites (0–5)
>1 metastatic site
Adjuvant endocrine therapy
Palliative endocrine therapy
Trastuzumab before WBRT
Duration (months) (range)
12 months (3–43)
Time to diagnosis of brain metastases
Median (months) (range)
35 months (1–232)
Time to progression (brain)
Median (months) (range)
23.5 months (1–166)
Intensified local treatment
Systemic treatment after WBRT
Duration (months) (range)
8 months (1–37)
Patient characteristics (trastuzumab vs. control)
KPS < 80%
Initial tumor stage IV
Invasiv ductal carcinoma
Intensified local treatment
Median OS after detection of brain metastases was 7 months, range 1–38 months, 95% CI 2.98–11.02. Corresponding numbers for median in-brain time to progression were 6 months, range 1–33+, 95% CI 5.01–6.98.
In the subgroup analysis, OS for patients receiving further trastuzumab was 21 months, range 3–38 months, 95% CI 11.28–30.72, as compared to nine months (range 1–23; 95% CI 2.06–14.99) in patients receiving further chemotherapy without trastuzumab, and 3 months (range 1–14; 95% CI 2.18–3.82) in patients without any further systemic therapy. Log-rank test revealed a statistically significant difference (p < 0.001).
TTP in patients treated with trastuzumab after WBRT was median 9 months (range 2–33+; 95% CI 4.35–13.65), as compared to 6 months (range 1–9+; 95% CI 4.46–7.54) in patients with continuation of systemic therapy without trastuzumab, and 2 (range 1–9+; 95% CI 1.43–2.57) in individuals without further systemic treatment. Again, log-rank test revealed a significant difference (p = 0.001).
In patients with KPS > 70%, OS was median 10 months (range 2–38; 95% CI 4.10 to 15–90), as compared to 3 months (range 1–19; 95% CI 1.77–4.23) in patients with KPS 70% (p = 0.0001). Corresponding numbers for TTP are 7 months (1–33+; 95% CI 4.12–9.88), and 2 months (1–9; 95% CI 1.16–2.84), respectively (p < 0.0001).
OS in patients undergoing intensified local therapy was 16 months (3–38; 95% CI 10.56–21.44), and median 3 months (1–14; 95% CI 2.19–3.81) in patients treated with WBRT only (p < 0.0001). Numbers for TTP are 9 months (2–33+; 95% CI 4.00–14.00) in patients with intensified local therapy as compared to 2 months (1–8+; 95% CI 1.19–2.81) (p < 0.0001).
Results (Cox regression)
Invasiv ductal/lobular carcinoma
Initial tumor stage
Intensified local treatment
Trastuzumab after WBRT
Different groups reported a higher incidence of brain metastases in patients with Her2 positive breast cancer since the introduction of trastuzumab (Clayton et al. , Shmueli et al. , Herrero et al. ). In the analysis presented here, we tried to establish if those individuals would derive any benefit from ongoing application of the antibody.
In a Cox regression model, trastuzumab and Karnofsky Performance Score >70% were significantly associated with longer survival after whole brain radiotherapy. A trend towards prolonged survival in patients with intensified local therapy (i.e. neurosurgical resection or boost irradiation) was observed, which however did not reach statistical significance. On the other hand, longer time to in-brain disease progression was predicted by intensified local therapy and KPS, while only a trend towards longer TTP in patients on trastuzumab treatment was found. Patient age did influence neither OS nor TTP significantly. While the significant influence of KPS was expected, data concerning age are somewhat contradicting evidences from different other trials. It is, for example, well known from individuals suffering from primary brain tumors that young age and a high performance score are associated with better outcome (Fazeny-Dorner et al. ). Other groups reported a significant benefit for patients with a KPS of at least 80% (as confirmed in our data) and age below 65 years with regard to survival in brain metastases treated with stereotactic radiosurgery (Fernandez-Vicioso et al. , Serizawa et al. ). An analysis of three RTOG studies formulated a prognostic index derived from over 1000 patients with a variety of solid tumors (Gaspar et al. ). Young patients (below 65 years) with controlled extracranial disease and a KPS of higher than 70% had the best outcome in terms of survival, with a median survival of 7.1 months after WBRT. Those data however derive from trials including patients with different primary tumors, therefore reducing comparability.
Two groups reported beneficial results in patients with ongoing trastuzumab therapy after detection of brain metastases. An analysis conducted by Kirsch et al. (Kirsch et al. ) compared outcome in terms of OS between three different populations. A Her2 positive group, with ongoing trastuzumab (n = 36), was compared to results from a Her2 positive group without further antibody therapy (n = 11), and a Her2 negative population (n = 48). Patients in the subgroup with continued antibody therapy had the best outcome, with an OS of 22.4 months, similar to the results reported in our study (21 months median OS), as compared to 9 months in the control groups. There is however a potential bias in this trial: the majority of patients in the control group had Her2 negative disease. Therefore, two biologically distinct tumor entities were compared. Also a second, conducted by Lower et al. (Lower et al. ), compared outcome between Her2 positive and negative patients. Further, in the study conducted by Kirsch et al., no data concerning KPS and extracranial metastatic sites are provided. This is of potential importance, as a recently published retrospective analysis, which included 174 patients with brain metastases from advanced breast cancer (43 with Her2 positive disease), reported that ongoing systemic therapy, KPS and metastatic sites were significantly influencing OS (Bartsch et al. ). In difference to the studies mentioned above, we therefore tried to rule out an inclusion bias by comparing two Her2 positive subgroups. Further, in the patients treated from 2003 on, we excluded all individuals who did not receive further trastuzumab due to low performance score or other reasons.
Contradicting the data from Kirsch et al. and Lower et al., both of which suggested an improved survival in Her2 positive disease with cerebral involvement, Tham et al. found a shorter OS in this population as compared to Her2 negative individuals (Tham et al. ). An important difference here is that Tham’s study spanned the years from 1970 to 1999, before the trastuzumab era. This again would emphasis the importance of ongoing trastuzumab therapy in patients with cerebral lesions, most probably due to its beneficial role on systemic disease.
An open question regards the possible activity of systemic treatment in brain metastases. There is some evidence that systemic chemotherapy is active in brain metastases from breast cancer, with a response rate of 50% (Rosner et al. , Boogerd et al. ). It is hypothesized that the blood-brain barrier is dysfunctional within brain metastases. Tumor vasculature tends to be relatively permeable, as evidenced by enhancement of lesions with water-soluble contrast agents (Lin et al. ). This effect is enhanced by WBRT. Another study presented a significant influence of palliative systemic therapy after WBRT on cerebral TTP, which remained significant also in a multivariate analysis (Bartsch et al. ). In the univariate analysis of our study, we also observed a significantly longer TTP in patients with ongoing trastuzumab. This significance however was lost in the multivariate model, although a strong trend was maintained. In the light of these facts, it might be assumed that there is some activity from trastuzumab on brain lesions themselves. This is further strengthened by a recent trial clearly showing that trastuzumab levels in cerebrospinal fluid are increased under conditions of an impaired blood-brain-barrier (e.g. after WBRT) (Stemmler et al. ). In general, however, the possible activity of trastuzumab in brain metastases remains a matter of debate.
It is commonly believed that patients with brain metastases from advanced breast cancer do not die from cerebral progression, but from progression of systemic disease. This was obviously the major limiting factor in the survival of women before the trastuzumab era, and since the introduction of the antibody, the proportion of patients with controlled systemic disease dying from brain lesions has grown (Lin and Winer ). This in term is underlined by the study conducted by Kirsch et al. (Kirsch et al. ). The authors were able to show that the difference in survival between the subgroups (i.e. patients who received trastuzumab and who did not), did not correlate with control of brain metastases. Further, most studies analyzing the beneficial role of intensified local therapy reported a prolonged time to in-brain disease progression, but no influence on OS (Patchell et al. , Noordijk et al. ). As the influence of trastuzumab on in-brain TTP in our trial did not reach significance in the multivariate analysis, it must be assumed that the impact on OS stems from control of systemic disease, rather than control of brain lesions.
In difference to other trials evaluating the role of trastuzumab after detection of cerebral spreading, this is the first study comparing two homogenous subgroups, without the inclusion of Her2 negative patients as control. Yet, a different assessment method of cerebral disease status (CT scans before 2000, and MRI scans thereafter) might potentially bias our results. Therefore, as this is a retrospective analysis with all subsequent limitations, our results must be interpreted with caution. Still, as it is not likely that the question of continued antibody therapy after WBRT is going to be addressed in randomized trials due to ethical reasons, we believe that our data are of value.
- 10.Clynes RA, Towers TL, Presta LG et al (2006) Inhibitory Fc receptors modulate in vivo cytotoxicity against tumor targets. Nat Med 6:443Google Scholar
- 15.Herrero A, Grandez R, Puertolas T et al (2004) High incidence of brain metastases at the time of death in women with metastatic breast cancer treated with trastuzumab. J Clin Oncol 22(suppl 14):765sGoogle Scholar
- 27.Lutterbach J, Bartelt S, Stancu E et al (2002) Patients with brain metastases: hope for recursive partitioning analysis (RPA) class 3. Radiother Oncol 263:399–345Google Scholar