HER2 Support Group Forums - Good and Bad News

http://meetinglibrary.asco.org/recor...dbook#fulltext

Breast Cancer in the Central Nervous System: Multidisciplinary Considerations and Management
Presented Monday, June 5, 2017
Add to Collection
Article
Video
Slides
Back to 2017 ASCO Educational Book
Breast Cancer in the Central Nervous System: Multidisciplinary Considerations and Management
Authors:
Nancy U. Lin, MD,
Laurie E. Gaspar, MD, FASTRO, FACR, MBA, and
Riccardo Soffietti, MD
Author Disclosures
Author Affiliations
Corresponding author: Nancy U. Lin, MD, Department of Medical Oncology, Dana-Farber Cancer Institute, 450 Brookline Ave., Boston, MA 02215; email: nlin@partners.org.
Overview

Full Article

Key Points

References

Tables and Figures
Spread of cancer to the CNS, either in the form of parenchymal brain metastases or leptomeningeal disease continues to confer a poor prognosis and high symptom burden in many patients, though survival does appear to be improving over time in some patient subsets. Although the area of breast cancer brain metastases has historically been a relatively understudied area, several seminal clinical trials have altered the standard of care over the past few years, and other smaller studies have provided a variety of new treatment options for patients. Furthermore, multiple innovative investigational strategies are being tested in the clinic. For these reasons, more than ever, the management of brain metastases requires a thoughtful, multidisciplinary approach that integrates the anatomic and symptomatic burden of disease, the status of a patient’s extracranial disease and systemic therapy needs, prior therapies, and life expectancy.

RISK FACTORS FOR THE DEVELOPMENT OF CNS METASTASES

Breast cancer is the second most common cancer associated with brain metastases in the United States following lung cancer.1 As patients with advanced breast cancer live longer, the incidence of brain metastases appears to be increasing. In a subset of women, progression in the CNS has become a major life-limiting problem.

The incidence of brain metastases in patients presenting with stage I/II invasive breast cancer, according to subtype, is as follows2: luminal A, 0.1%; luminal B, 3.3%; luminal-HER2, 3.2%; HER2, 3.7%; and triple-negative, 7.4%.

Although these numbers are somewhat low, of those patients with distant metastases, approximately 30% to 50% will eventually develop brain metastases.2-5 Factors associated with an increased likelihood of brain metastases include young age, lymph node positivity, higher grade, hormone receptor negativity and HER2 positivity, and time from diagnosis to first metastasis.6 The time from the initial diagnosis of primary breast cancer to the development of brain metastases is also influenced by subtype, with the shortest interval observed for patients with triple-negative disease (27 months), and the longest interval observed for those with ER-positive, HER2-positive disease (54 months).7

PROGNOSTIC AND PREDICTIVE FACTORS OF SURVIVAL

The predictive factors and the prognosis of patients with brain metastases are now considered to be disease specific (Table 1). One tool that can be used is the Disease-Specific Graded Prognostic Assessment (DS-GPA).8 The prognostic factors within the breast-specific GPA are presented in Table 2. Note that the time from primary diagnosis to brain metastases was not an independent significant prognostic factor in the breast GPA and is therefore not a part of the index.7

TABLE 1.
Median Survival Time (Months) by DS-GPA8

Median Survival Time (Months) by DS-GPA8

Full Screen
TABLE 2.
Prognostic Factors and Assigned Score in Breast Cancer GPA8

Prognostic Factors and Assigned Score in Breast Cancer GPA8

Full Screen
The DS-GPA was based on the observed outcome of patients referred for a radiation therapy opinion, and patients underwent whole-brain radiotherapy (WBRT), stereotactic radiosurgery (SRS), surgery, or a combination of these treatments. Only 6% of patients had a GPA score of 1, with the remaining patients fairly equally distributed between GPA scores of 2, 3, and 4. This retrospective analysis cannot be used to predict the outcomes according to different treatments. Its utility lies in its use as a stratification tool for clinical trials and the comparison of results between clinical trials. It can also aid the oncologist in determining whether the patient might be best served by hospice.

National Comprehensive Cancer Network guidelines as of January 2017 state that CNS imaging of patients with asymptomatic breast cancer is not indicated, based on lack of available evidence of benefit. However, prospective studies to evaluate the risks and benefits of CNS imaging are scant. Given the incidence and relatively short interval to presentation of brain metastases in patients with triple-negative disease, and the high incidence of CNS metastases in patients with HER2-positive breast cancer, further investigation of this issue is highly warranted.

LOCAL THERAPY

The Role of Surgery in Patients With Brain Metastases
Among local treatment options, surgery has a clear role in some subgroups of patients. Three phase III trials have compared surgical resection followed by WBRT with WBRT alone for a single brain metastasis (Fig. 1).9-11

FIGURE 1.
Management Algorithm for the Initial Treatment of Patients With Breast Cancer Brain Metastases
Management Algorithm for the Initial Treatment of Patients With Breast Cancer Brain Metastases
This figure provides a broad overview. For details and discussion of nuances of the recommendations, please refer to the text. Treatment recommendations will also depend on performance status, prior therapies, status of extracranial disease, comorbidities, and life expectancy. In most cases, outside of a clinical trial, surgery and/or radiation therapy will be given as initial therapy, and the systemic therapy will be determined by the status of a patient’s extracranial disease (i.e., continue prior systemic therapy if systemic disease is stable, and switch if systemic disease is progressive).

Expand
The first two studies have shown a survival benefit for patients receiving the combined treatment (median survival 10 vs. 4–6 months). In the Patchell study, patients who received surgery displayed a lower rate of local relapse (20% vs. 52%) and longer period of functional independence. The third study, which included more patients with active systemic disease (80% vs. 30%–40%) and a low Karnofsky performance status, did not show any benefit with the addition of surgery to WBRT. Therefore, class I evidence shows that the survival benefit of surgical resection in addition to WBRT is limited to the subgroup of patients with controlled systemic disease and good performance status.12 In properly selected patients with two or three brain metastases, who are in good neurologic condition and have controlled systemic disease, complete surgical resection yields results that are comparable to those obtained in single lesions.13 One caveat to these and much of the local therapy literature is that patients of multiple primary histologies were included in the trials, with a relatively small fraction of patients with breast cancer (typically 10%–20%), and thus, the recommendations are to some extent extrapolations based on a study population with primarily non–small cell lung cancer.

For the majority of patients, surgical resection allows an immediate relief of symptoms of intracranial hypertension, a reduction of focal neurologic deficits and seizures, and a rapid steroid taper. Gross total resection of a brain metastasis can be achieved with lower morbidity using contemporary image-guided systems, such as preoperative functional MRI, intraoperative neuronavigation, and cortical mapping.14 An early postoperative MRI can detect residual tumor in up to 20% of patients, and this is associated with an increased risk of local recurrence.15

The impact of surgical methodology on the complication rate and functional outcome, as well as on local relapse in patients with a single brain metastasis, has been recently analyzed. Overall, the study suggests that postoperative complication rates are not increased by en bloc resection, as compared with piecemeal resection, for lesions in eloquent brain regions or large tumors.16 Leptomeningeal dissemination can be a complication, especially for patients with posterior fossa metastases undergoing a piecemeal resection (13.8%) compared with en bloc resection (5%–6%).17

Last but not least, surgery is important in providing tissue for molecular analysis to define the molecular profile of the brain metastasis, which can be different from that of the primary cancer. This is critical in the near future for tailoring targeted therapies to the molecular profile of the brain metastases.

WBRT Compared With SRS Following Surgical Resection
Despite the randomized study by Patchell et al,18 in which it was found that patients with a single brain metastases who underwent surgical resection and postoperative WBRT had fewer recurrences of cancer in the brain and were less likely to die of neurologic causes as compared with patients treated with surgical resection alone, there has been controversy regarding the role of WBRT in this setting. This led to the N107C/CEC.3 cooperative group study randomly assigning patients with a resected brain metastasis to receive either WBRT or SRS to the cavity. SRS to unresected brain metastases was allowed in both groups. Patients were stratified between primary lung cancer, radio-resistant histologies, or other histologies. This study was presented at the Plenary Session during the 2016 American Society for Radiation Oncology Annual Meeting but is not yet published.19 Approximately 30% of enrolled patients fell into the other category, although breast cancer is not separated out otherwise. There was no reported difference in overall survival between the two treatment groups (11–12 months), with no difference seen according to age, extracranial disease status, number of brain metastases, histology, or size of resection cavity. However, there was a small but statistically significant difference in the cognitive deterioration–free survival favoring the SRS arm (2.8 months WBRT arm vs. 3.3 months SRS arm; p < .0001). Only 5.4% of patients in the WBRT arm were free of cognitive deterioration at 6 months as opposed to 22.9% in the SRS arm.

SRS With or Without WBRT
Several randomized studies have examined the outcome of SRS with or without WBRT.20-23 One of these was a randomized controlled trial published in 2006 by Aoyama et al20 (JROSG 99-1), which randomly assigned 132 patients with up to four brain metastases amenable to SRS. The primary endpoint was overall survival, but secondary outcomes included local recurrence, rate of salvage brain treatment, functional preservation, toxic effects, and cause of death. The study was closed earlier than the planned accrual when an interim analysis determined that more than 800 patients would be required to detect a significant difference in the primary endpoint. Breast cancer made up only 7% of enrolled patients, the majority being non–small cell lung cancer. In the SRS-only group, the median survival time and the 1-year actuarial survival rate were not significantly different between the two groups. However, the group receiving SRS and WBRT had a lower intracranial recurrence rate at 1 year (47% vs. 77%; p < .001), and required less frequent salvage treatment as opposed to the SRS-only group.

Another larger study of 359 patients, of which 12% had breast cancer, randomly assigned patients with up to four brain metastases to receive local therapy (surgery or SRS) with or without WBRT.23 Overall survival was similar between the two treatment arms (p = .89), although the local control (surgery vs. surgery WBRT: 59% to 27%, p < .001; SRS vs. SRS + WBRT: 31% vs. 19%, p = .04) and need for further salvage therapy (51% vs. 16%, p value not reported) were improved in the WBRT arm.

Lastly, a 2015 meta-analysis by Sahgal et al24 of these randomized studies found that patients age 50 or younger had a significant survival benefit (p = .04) when SRS alone was used. This analysis found that these results were similar between patients with lung cancer and those with breast cancer, although the authors acknowledged the problems with small sample sizes. The authors concluded that SRS alone is the recommended initial therapy of patients age 50 or younger with one to four brain metastases.

The above findings showing that WBRT is not associated with improved survival, when combined with the data regarding the neurocognitive effects of WBRT, have led to many guidelines recommending SRS only for patients with one to four brain metastases (American Association of Neurological Surgeons, unpublished data, 2017).22,25

Quality of Life and Cognitive Dysfunction Following WBRT
Cognitive dysfunction following WBRT represents a topic of increasing importance. Historically, radiation-induced dementia with ataxia and urinary incontinence was described in up to 30% of patients by year 1 who were receiving unconventional, large-size fractions of WBRT (6–8.5 Gy), which are no longer used.26 The picture on CT/MRI was that of a leukoencephalopathy (diffuse hyperintensity of the periventricular white matter on T2-weighted and fluid attenuation inversion recovery images) with associated hydrocephalus, for which placement of a ventriculoperitoneal shunt could be of some clinical value. When using more conventional size fractions (up to 3 or 4 Gy per fraction), the risk is that of mild cognitive dysfunction, consisting mainly in learning and memory impairment with a variable degree of damage of the white matter and cortical atrophy on MRI.

In recent years, several randomized trials have shed light on the short- and long-term effects of WBRT on neurocognitive function and quality of life. Aoyama et al compared the neurocognitive function of patients who underwent SRS alone or SRS plus WBRT.27 Similar proportions of patients in both arms (p = .85) achieved a three point or more improvement in their Mini Mental State Examination score shortly after therapy (2–3 months). However, subsequent deterioration of neurocognitive function in long-term survivors (up to 36 months) after WBRT was observed. In a small randomized trial, Chang et al have shown that patients treated with SRS plus WBRT were at greater risk of a decline in learning and memory function at 4 months after treatment compared with those receiving SRS alone.21 A randomized phase III trial (Alliance trial) has compared SRS alone with SRS plus WBRT in patients with one to three brain metastases using a primary neurocognitive endpoint, defined as decline from baseline in any seven cognitive tests at three months.28 Neurocognitive decline was significantly more frequent after SRS plus WBRT compared with SRS alone (91.7% vs. 63.5%, p < .001). On individual tests, there was more cognitive deterioration in immediate memory (30.4% vs. 8.8.2%, p = .004), delayed memory (51.1% vs. 19.7%, p < .001), and verbal fluency (18.6% vs. 1.9%, p = .01) in the SRS plusWBRT arm. Finally, a quality-of-life analysis of the EORTC 22952-26001 trial has shown over 1 year of follow-up no significant differences in the global health-related quality of life, but patients undergoing adjuvant WBRT instead of observation had lower transient cognitive functioning, physical functioning, and more fatigue.22

Patients with arterial hypertension, diabetes, or other vascular diseases are at a higher risk of developing cognitive dysfunction. The pathogenesis of this radiation damage could consist of an injury of the endothelium of small vessels that leads to an accelerated atherosclerosis and ultimately to a chronic ischemia, resulting in a picture similar to that of the small vessel disease of vascular dementia. For this reason, there is interest in investigating vascular dementia treatments to prevent or reduce radiation-induced cognitive decline. One of these approaches is using memantine in combination with WBRT. Memantine is a noncompetitive, low affinity antagonist of the N-methyl-D-aspartate (NMDA) receptor, which is one of the receptors activated by glutamate, the principal excitatory neurotransmitter. Memantine has the potential to block the excessive NMDA stimulation following ischemia, which ordinarily could lead to excitotoxic damage of the normal brain. In a recently published randomized, double-blind, placebo-controlled phase II trial (RTOG 0614), the use of memantine during and after WBRT resulted in a mild improvement of cognitive function over time, specifically delaying time to cognitive decline and reducing the rates of decline in memory, executive function, and processing speed.29 The use of another neurotransmitter regulator, such as donepezil, has shown only modest improvements in cognitive function in a controlled trial, especially among patients with greater pretreatment impairment.30

Radiation-induced cognitive deficits may result, at least in part, from a radiation injury to the neuronal stem cells in the subgranular zone of the hippocampus.31 These stem cells are responsible for maintaining neurogenesis, which is critical for preserving memory function, especially in terms of encoding new episodic memories. Low-dose irradiation in rodents results in a blockade of hippocampal neurogenesis and damage of the neurogenic microenvironment, leading to significant short-term memory impairment. Thereby, it has been hypothesized that sparing the hippocampus during WBRT (hippocampal-avoidance WBRT, HA-WBRT) could prevent the damage of the neuronal progenitor cells and better preserve memory functions.32 The recent single-arm, phase II RTOG 0933 trial has suggested that the conformal avoidance of the hippocampus during WBRT is associated with some sparing of memory and quality of life; specifically, performance on standardized memory tests declined 7% from baseline to 4 months in patients treated with hippocampal-avoidance WBRT, as compared with 30% in an historical control group.33 Importantly, 4.5% of patients developing intracranial progression had involvement of the hippocampal-avoidance area by metastatic disease. In this regard, building on results of RTOG 0933 and RTOG 0614, NRG-CC001 is a phase III trial evaluating the potential combined neuroprotective effects of hippocampal avoidance in addition to memantine during WBRT for brain metastases.34

Clinical Challenges: Tumor Progression, Radionecrosis, and Pseudoprogression
A critical issue is the distinction between post-treatment effects and true tumor progression in some particular scenarios. Following SRS, changes such as an increase in contrast enhancement, necrosis, edema, and mass effect on MRI are difficult to interpret: in this regard, PET with 18F-fluorodeoxyglucose, amino acids or 18F-fluorodeoxythymidine, MRI perfusion, and magnetic resonance spectroscopy may provide additional information, though are rarely diagnostic.35-38 In general, careful monitoring with MRI, sometimes for many months, is needed. Radiation necrosis is commonly treated with steroids. Hyperbaric oxygen and/or the anti-VEGF agent bevacizumab, which may allow stabilization/normalization of the vascular permeability, can be useful in patients not responding to steroids,39 while surgical resection is needed in some patients.

In patients receiving immunotherapy-based treatments, an initial increase in the number and size of metastases can be followed by radiographic stabilization or regression. This pattern might be related to the mechanism of action of immunotherapy, including immune infiltrates and the time to mount an effective immune response. If immune response–related radiographic changes are suspected, the advice is to not interrupt immunotherapy treatment until a short interval scan is obtained.40,41

SYSTEMIC THERAPY

Evidence for Efficacy of Available Endocrine Therapies and Cytotoxic Chemotherapies
To date, no systemic agents have gained regulatory approval for the treatment of breast cancer brain metastases. Nevertheless, as summarized in Table 3, CNS activity in case series or in small, prospective studies has been reported across a range of cytotoxic drugs. For example, Rivera and colleagues reported their experience in a phase I trial testing the combination of capecitabine and temozolomide in 24 patients with breast cancer brain metastases (14 newly diagnosed; 10 with progressive brain metastases after prior local therapy). The observed CNS objective response rate was 18%, with median time to progression of 12 weeks.42 Given the general lack of activity of temozolomide in breast cancer, it would be reasonable to assume that the majority of the activity in this trial can be attributed to the capecitabine. Data from a small series of seven patients treated at Memorial Sloan Kettering Cancer Center corroborate the observation of capecitabine activity in the CNS.43

TABLE 3.
Summary of Case Reports, Case Series, and Prospective Studies Testing Cytotoxic Chemotherapy in Patients With Breast Cancer Brain Metastases

Summary of Case Reports, Case Series, and Prospective Studies Testing Cytotoxic Chemotherapy in Patients With Breast Cancer Brain Metastases

Expand
Anthracyclines are also associated with CNS responses in breast cancer, with response rates ranging widely from 17% to 62%, in small experiences. Single-agent temozolomide appears to have minimal activity in breast cancer and does not clearly add to other agents when given in combination.51,53

CNS activity for platinum salts has been reported in older case series, including response rates of 38% to 55%, albeit in a patient population less heavily pretreated than a typical patient seen today.47,48 In this context, recent efforts (not limited to the brain metastasis space) to identify predictive markers of platinum benefit are relevant. The TNT trial evaluated the efficacy of taxanes versus platinums in the first-line treatment of metastatic triple-negative breast cancer. Of note, patients with active brain metastases were excluded. The study demonstrated differential activity according to BRCA1/2 germline status, with a substantially higher rate of response in extracranial sites to carboplatin compared with docetaxel in BRCA1/2 carriers.56 This hypothesis has not been formally tested in the CNS; however, anecdotally, Jennifer Ligibel, MD, and Judy Garber, MD, MPH, have observed durable CNS responses (including one in excess of three years) in some BRCA1/2 carriers treated with platinum salts (personal communication, January 2017).

Case reports of CNS responses to a variety of endocrine agents, including tamoxifen and aromatase inhibitors, are also present in the literature; however, patients with estrogen receptor–positive tumors typically present with brain metastases late in their disease course, when their disease has become hormone refractory.57-60

In general, if considering off-label use of systemic therapy for the treatment of breast cancer brain metastases, the choice of therapy should be in accordance with a patient’s tumor subtype, prior therapies, performance status, and comorbidities, in keeping with national and international guidelines for management of metastatic breast cancer (Fig. 2). As with the practice guidelines for use of cytotoxic chemotherapy in patients with extracranial metastases, sequential single-agent chemotherapy is generally preferable to combination chemotherapy.

FIGURE 2.
Options for Systemic Treatment of Breast Cancer Brain Metastases
Options for Systemic Treatment of Breast Cancer Brain Metastases
This figure provides a broad overview. For details and discussion of nuances of the recommendations, please refer to the text. Note that very few prospective trials have been conducted to evaluate the role of systemic therapy for breast cancer brain metastases, and, in many cases, the recommendations above are therefore based on case reports or case series, or extrapolation from systemic therapy trials. See text for details. Of note, there have not been randomized trials directly comparing a local (i.e., surgery, radiation therapy) versus systemic approach for the treatment of breast cancer brain metastases. Treatment recommendations will depend on performance status, prior therapies, status of extracranial disease, comorbidities, and life expectancy.

Expand
Investigational Cytotoxic Approaches
In terms of cytotoxic chemotherapy, the trend for the development of compounds focused on the treatment of brain metastases has been to engineer compounds that more effectively penetrate the blood-brain barrier (Table 4). Two of the compounds furthest in development are etirinotecan pegol (NKTR-102) and ANG1005. NKTR-102 is a long-acting topoisomerase-I inhibitor that prolongs exposure to SN38, the active metabolite of irinotecan. CNS activity in breast cancer has been previously observed in the clinic with the parent compound irinotecan.50 In a mouse model of breast cancer brain metastases (MDA-MB-231Br), NKTR-102 prolonged survival compared with conventional irinotecan.61 A phase III trial for patients with heavily pretreated breast cancer (either without brain metastases or with stable brain metastases on study entry) was recently reported.62 Though a negative study overall, a potential signal was observed in the stable brain metastasis subset and a confirmatory study is currently under way. ANG1005 is a novel taxane derivative that is able to penetrate the blood-brain barrier via the low-density lipoprotein receptor–related protein.63 CNS responses have been observed in early-phase studies, and additional studies are ongoing.64,65

TABLE 4.
Ongoing Trials of Systemic Therapy for Breast Cancer Brain Metastases

Ongoing Trials of Systemic Therapy for Breast Cancer Brain Metastases

Full Screen
HER2-Targeted Therapies
The development of HER2-targeted therapies has dramatically improved overall outcomes for patients with HER2-positive breast cancer, both in the early and advanced stages. Despite these improvements, up to half of patients with advanced HER2-positive breast cancer will relapse in the CNS.

The small molecule tyrosine kinase inhibitor (TKI) lapatinib has been studied as a single agent and in combination with chemotherapy in multiple prospective clinical trials. As a single agent in pretreated patients, its activity is modest at best, with CNS responses observed in only 6% of patients.66 Greater activity has been observed in combination with capecitabine, with CNS response rates ranging from 18% to 38% in pretreated patients, and a CNS response rate of 66% in the newly diagnosed setting.66-70 Responses have been durable in many cases, and overall, the results support the concept of evaluating HER2-targeted TKIs for the treatment of brain metastases.

HER2-targeted TKIs in clinical development include neratinib, afatinib, and tucatinib, among others (Table 4). Neratinib is an irreversible inhibitor of EGFR and HER2 currently in late-stage clinical testing (NALA, neratinib plus capecitabine versus lapatinib plus capecitabine in patients with HER2+ metastatic breast cancer who have received two or more prior HER2-directed regimens in the metastatic setting, NCT01808573). However, the ongoing phase III study excludes patients with active brain metastases. The Translational Breast Cancer Research Consortium is conducting a phase II study evaluating neratinib in patients with progressive brain metastases. Results of the monotherapy neratinib cohort have been published, with an observed CNS objective response rate of 8%.71 Results of the neratinib/capecitabine combination cohort are anticipated in mid- to late 2017. Like neratinib, afatinib inhibits both EGFR and HER. Although it has gained regulatory approval in lung cancer, clinical development in breast cancer has been terminated based on negative results of the LUX-Breast 1 and LUX-Breast 3 randomized trials.72,73 In particular, in the LUX-Breast 3 trial, the combination of vinorelbine and afatinib did not afford any additional benefits to patients with HER2-positive breast cancer brain metastases compared with treatment of provider choice.72

In contrast to either neratinib or afatinib, tucatinib (ONT-380; ARRY-380) selectively targets HER2 and has minimal activity against EGFR, leading to a more favorable toxicity profile, with less diarrhea and rash. The active metabolite appears to cross the blood-brain barrier, and improvements in survival have been reported in preclinical models of breast cancer brain metastases. In a phase I study of tucatinib with trastuzumab, CNS responses were observed in 7% of patients; approximately one-third of patients achieved stable disease of 16 weeks or longer.74 The triplet of trastuzumab-capecitabine-tucatinib has been studied in a phase IB study among patients with highly refractory, HER2-positive metastatic breast cancer. Objective responses were observed in 61% of patients, including in 42% of patients who had measurable CNS disease at baseline.75 The approach is now being tested in an ongoing randomized trial that includes patients with and without brain metastases.

Historically, monoclonal antibodies such as trastuzumab, pertuzumab, or trastuzumab-emtansine were thought to be too large and bulky to penetrate the blood-brain barrier. However, studies utilizing 89Zr-labeled trastuzumab as a PET tracer have suggested there is some penetration through a disrupted blood-tumor barrier.76 Penetration across the blood-tumor barrier is supported by emerging case reports and case series of the CNS activity of trastuzumab-emtansine (TDM1), with response rates qualitatively similar to that reported against extracranial disease.77-79 A prospective U.S. study is being planned. Another approach under active investigation in the ongoing PATRICIA study is the use of high-dose trastuzumab (6 mg/kg IV weekly) in combination with standard pertuzumab every 3 weeks to drive up concentrations of trastuzumab in brain metastases. Accrual to the study is ongoing.

Clinical Challenges: Is There a Role for Chemoprevention of Brain Metastases?
There is great interest in preventing the emergence of brain metastases (primary prevention) and prolonging the time to subsequent CNS progression in patients who receive initial local therapy (secondary prevention). A common clinical question is whether systemic therapy should be modified following SRS to include a “CNS-active” regimen. At present, there is no direct evidence to support this approach, though the existing data have been sparse and do not perfectly address this question.

In the EMILIA trial, a subset analysis was performed among patients who entered the study with stable brain metastases to determine whether the benefit of trastuzumab-emtansine seen in the overall study (compared with lapatinib-capecitabine) held up in the brain metastasis subset.80 Patients in the brain metastasis subset appeared to derive similar relative benefits in terms of overall survival prolongation with trastuzumab-emtansine, and there was no obvious signal in favor of lapatinib-capecitabine with respect to CNS progression. In addition, there were no obvious differences in the incidence of new CNS metastases among patients who entered the study without brain metastases at baseline (2% with trastuzumab-emtansine; 0.7% with lapatinib-capecitabine; p = not significant). The analysis had several limitations including that CNS scans were not mandated per protocol and emerging reports supporting CNS activity of trastuzumab-emtansine, making this potentially a comparison between two “CNS-active” regimens.

In general, the consensus approach for treatment of such patients is to continue the prior systemic therapy after SRS, if the systemic disease remains well controlled, but this is an area ripe for clinical trials.81

Other Targeted Approaches Under Clinical Investigation
A number of other targets are being explored in the treatment of breast cancer brain metastases, including CDK4/6 inhibitors, PARP inhibitors, and immunomodulatory therapies (Table 4).

When combined with endocrine therapy, the CDK4/6 inhibitors palbociclib and ribociclib prolong progression-free survival compared with endocrine therapy alone.82-84 Among the CDK4/6 inhibitors, abemaciclib appears to have the best CNS penetration in preclinical models and has been demonstrated to reach therapeutic levels in human “window of opportunity” studies when given prior to planned resection.85,86 Phase II studies of both palbociclib and abemaciclib for brain metastases are ongoing.

Given the frequency of brain metastases in patients with triple-negative breast cancer, there is interest in developing novel targeted approaches in this patient population. PARP inhibitors have clear activity against extracranial metastases in BRCA1/2 carriers; however, single-agent activity in sporadic triple-negative breast cancer has been disappointing.87 Three large randomized phase III trials comparing PARP inhibitors with standard chemotherapy in BRCA1/2 carriers with metastatic breast cancer are ongoing, but all three studies exclude patients with active brain metastases. There is also ongoing interest in combination approaches to sensitize BRCA1/2 wild-type breast cancer to PARP inhibitors, including combinations with platinum salts. The U.S. cooperative groups are collaborating on a planned randomized study (S1416) that will examine cisplatin with or without the PARP inhibitor veliparib (ABT-888). Given that veliparib crosses the blood-brain barrier in preclinical models, a CNS-specific cohort will be concurrently enrolled, with a primary endpoint of progression-free survival.88

There are accumulating preclinical and clinical evidence suggesting that the immune system is critical for disease outcome in breast cancer, particularly in the triple-negative and HER2-positive subtypes.89,90 Moreover, PD-L1 expression appears to be common in breast cancer brain metastases.91 Data in patients with melanoma and lung cancer support potential efficacy of immune checkpoint inhibitors in the CNS.92 Beyond CTLA-4 and PD-1/PD-L1 inhibitors, there is a wealth of novel immunomodulatory compounds, including STING and GITR agonists, and inhibitors of IDO, TIM3, and LAG3. Unfortunately, the vast majority of ongoing trials of immunotherapy in breast cancer specifically exclude patients with active brain metastases, though there are some studies in this population that have recently opened or are in development (Table 4).

MANAGEMENT OF LEPTOMENINGEAL DISEASE

Estimates of the incidence of leptomeningeal metastases vary widely, ranging from 2% to 40%, either alone or associated with parenchymal brain metastases. In a case series of patients with leptomeningeal disease (1998 to 2013) from Memorial Sloan Kettering Cancer Center, both HER2-positive (26% of cases) and triple-negative (25% of cases) breast cancer subtypes were overrepresented, suggesting they are associated with a propensity toward dissemination in the leptomeninges.93 Invasive lobular histology also appears to be associated with leptomeningeal spread.94 The prognosis is poor with median survival of 3.5 to 6 months and 20% survival rate at 1 year.93,95 Since patients can experience very poor survival, it is critical to consider prognostic factors early in weighing management options, including consideration of a more palliative/hospice-oriented course, as appropriate. Favorable prognostic factors include HER2-positive subtype, preserved performance status, and CNS-only involvement. Unfavorable prognostic factors include poor performance status, progressive/treatment-refractory extracranial disease, and major neurological deficits.

The most typical management approach is radiation to sites of bulk disease followed by consideration of intra–cerebrospinal fluid (CSF) and/or systemic therapy. Radiation-based approaches, including WBRT, have the potential to provide rapid relief of symptoms, and should be strongly considered, particularly for patients presenting with neurological deficits. Intra-CSF chemotherapy has a role for palliation of neurologic symptoms and should be considered for patients with a large tumor cell load in the CSF.96 Methotrexate, liposomal cytarabine, and thiotepa are the drugs of choice. At the time of the placement of an Ommaya catheter, and prior to injecting drugs into the CSF, flow studies are recommended to rule out the existence of subarachnoid blocks, as these could preclude optimal distribution of drug and increase the risk of leukoencephalopathy.97-100 Systemic chemotherapy has been used off-label to treat patients with leptomeningeal disease based on observed efficacy in case reports and small case series. Regimens with reported efficacy (with caveats given the very limited data) include tamoxifen, aromatase inhibitors, high-dose intravenous methotrexate, capecitabine, lapatinib/capecitabine, and platinum salts.

From an investigational standpoint, leptomeningeal disease has frequently been excluded from clinical trials. However, a number of ongoing trials are exploring new therapeutic options (Table 4). Of note, a phase I/II study of intrathecal trastuzumab has recently completed accrual. In this study, trastuzumab was reconstituted in preservative-free sterile water, USP or preservative-free 0.9% sodium chloride, with an induction phase of more frequent administration, followed by tapering of the frequency of administration. The recommended phase II dose has been identified, and efficacy results from the phase II portion are expected later this year.101 In general, we have not incorporated use of intrathecal trastuzumab into routine clinical practice, pending efficacy results of this study.

CONCLUSION

Increasingly, the management of breast cancer with brain metastases (parenchymal or leptomeningeal disease) requires close multidisciplinary collaboration, balancing the patient’s disease burden in the CNS and extracranially, prior therapies, performance status, comorbidities, life expectancy, and preferences, with available treatment options. Surgical resection should be strongly considered in patients presenting with a single brain metastasis, or a large, symptomatic mass, particularly if they have good performance status and controlled extracranial disease. For patients with expected longer survival, the use of up-front SRS and avoidance of WBRT in the setting of a limited number of brain metastases is preferred. Although there are still no systemic therapies approved for the treatment of breast cancer brain metastases, a number of regimens have demonstrated clear activity in prospective experiences and can be considered in the clinic. At present, systemic therapy is an option for patients whose CNS disease has progressed through standard local therapy, and it can even be considered in patients with newly diagnosed disease in lieu of local approaches in some circumstances (e.g., asymptomatic or minimally symptomatic patients). Moving forward, many novel, promising approaches are being tested in the clinic, and results are eagerly awaited.