Refining Therapy in Patients with HER2-Positive Breast Cancer with Central Nervous System Metastasis

Abstract

Background

Brain metastasis (BM) is a major clinical problem in metastatic breast cancer (MBC), occurring in 50% of patients with human epidermal growth factor receptor 2-positive (HER2+) breast cancer. Historically omitted from clinical trials, recent studies of novel HER2-targeted agents have focused on HER2+ BM patients, addressing stable but also progressing BM and leptomeningeal carcinomatosis (LMC).

Summary

This review aimed to summarize the most relevant data on treating patients with HER2+ BM and LMC.

Key Messages

The treatment paradigm for patients with HER2+ MBC has changed. Local therapies play an important role, but accumulating evidence on the intracranial activity and clinical benefit of anti-HER2 targeting therapies might lead to a shift in the paradigm on treating BM in the next few years towards more widespread use of systemic therapy.

Introduction

Brain metastasis (BM) is a major clinical problem in breast cancer (BC) and significantly shortens patient survival. Approximately 50% of patients with metastatic human epidermal growth factor receptor 2-positive (HER2+) BC will develop BM [1]. Rates of BM across all metastatic BC (MBC) are increasing [2]. An increase in incidence was noted in patients first treated with trastuzumab for HER2+ MBC, proposed to be unmasking the biological predilection for BM, in the face of controlled extracranial disease [3]. Following the introduction of adjuvant trastuzumab, a meta-analysis reported that the rate of brain relapse was 2.6%, equating to a 76% increase compared with patients who had not received adjuvant trastuzumab [4].

When addressing BM, two clinical phenotypes should be distinguished:

• Treated/stable BM − patients have received prior central nervous system (CNS)-directed therapy for their BM and CNS is stable

• Active BM − new or progressive BM not treated with CNS-directed therapy since documented progression and/or associated with oedema and/or symptoms [5].

Patients with active BM have been historically excluded from clinical trials. However, this paradigm has been changing, with recent trials including and/or being specifically designed to evaluate treatment approaches for these patients.

Current Standards

At diagnosis of MBC, screening for BM is not standard of care. However, if detecting CNS metastases would alter the choice of systemic therapy, neuroimaging should be performed [6]. When BM is detected, treatment decisions should be made by a multidisciplinary team, taking into account the number, size, and location of BM; disease extension; extra-CNS disease; performance status; expected survival; and neurological symptoms [7, 8, 9]. Prognostic models can aid treatment decisions [10]. Patients' management should be individualized, and palliative and supportive care should be offered. For raised intracranial pressure and peri-tumoral oedema, corticosteroids are recommended. Seizures can be managed with anticonvulsants [7]. For HER2+ BC, current guidelines discuss the role of systemic targeted therapy, particularly for asymptomatic patients with favourable prognostic factors [6, 7]. Systemic regimens with evidence of CNS disease activity should also be considered when a patient develops intracranial disease progression after local therapy [9]. In Figure 1, we suggest an algorithm for managing patients with BM from HER2+ BC.

Fig. 1.

Suggested approach for the managing of patients with brain metastases from HER2-positive breast cancer. HER2, human epidermal growth factor receptor-type 2; SRS, stereotactic radiosurgery; WBRT, whole brain radiotherapy.

Role of Local Therapies

If local therapies are considered, surgery, stereotactic radiosurgery (SRS), and whole brain radiotherapy (WBRT) are the options [8]:

Surgery

Surgery is generally preferred in patients with absent or controlled extracranial disease and 3 or less BM [11]. Also, there are specific scenarios where it should be considered for the immediate therapeutic effect: large BM (>3 cm diameter) causing raised intracranial pressure or neurological impairment; metastasis in the posterior fossa (due to the risk of hydrocephalus); and cystic or necrotic BM, which may respond less well to SRS. Surgery with diagnostic intention is required, for example, when neuroimaging is equivocal; primary tumour diagnosis is not established or when changes in molecular profile may impact clinical decision-making [7]. Post-operative radiation may be considered to improve local control [8]. Two randomized clinical trials (RCTs) compared surgery plus WBRT with WBRT alone in patients with single BM, and both showed a survival benefit with the addition of surgery: one trial with 48 patients found fewer local recurrences (20% vs. 52%; p < 0.02), improved survival (40 vs. 15 weeks; p < 0.01), and better quality of life [12]. The other, with 63 patients, also found a significant overall survival (OS) benefit (10 vs. 6 months; p = 0.04). The benefit from surgery was mainly seen in patients with stable extracranial disease (median OS 12 months) [13].

Stereotactic Radiosurgery

SRS is the delivery of high doses of radiation with high accuracy to intracranial targets via stereotactic or image guidance. It is commonly used as a single fraction (15–24 Gy) but can also be fractionated [14]. It is particularly suitable for single, small, or inaccessible tumours and can be used in brain lesions with volumes of 15–25 cm³. Neurotoxicity and local failure after SRS increase with lesion size; therefore, SRS should generally be limited to lesions with a diameter of 3 cm or less [8]. A prospective study of 1,194 patients who received SRS alone showed a similar OS (10.8 months; HR = 0.974) in patients with 2–4 versus 5–10 BM [15]. Due to high local recurrence rates after surgery alone (50–60% risk of local recurrence at the surgical site within 6–12 months) [16, 17], two RCTs evaluated SRS to resection cavity after surgery of BM: one study which randomized 194 patients to post-operative WBRT or SRS found no difference in OS (12.2 vs. 11.6 months, HR = 1.07, p = 0.7). Neurocognitive dysfunction at 6 months was more frequent after WBRT (52 vs. 85 per cent; p < 0.00031) [18]. Another study randomized 132 patients to either postoperative SRS or observation and found higher local control rates for SRS (72 vs. 43% at 12 months; HR = 0.46, p = 0.015), with similar median OS (17 vs. 18 months; HR = 1.29, p = 0.24). The most important risk factor for local recurrence in patients treated with SRS was preoperative tumour diameter larger than 2.5 cm [17].

Whole Brain Radiotherapy

WBRT may be indicated in patients with multiple or large BM not amenable to surgery or SRS. The most commonly used regimen is 20–30 Gy in 5–10 fractions [8]. WBRT after surgery or SRS improved local and distant brain control, but not OS, and was associated with an unfavourable effect on neurocognitive function [16]. The main purpose of WBRT in patients with a good performance status not suitable for SRS or surgery is to improve neurological deficits caused by BM and to prevent further deterioration of neurological function. A meta-analysis found that WBRT alone was associated with a median OS of 3–6 months, with 10–15% of patients alive at 1 year [19]. Modulated intensity techniques and hippocampal sparing radiation can minimize cognitive deterioration [20]. Furthermore, an RCT showed a trend towards neurocognitive protection when combined with WBRT [21]. A trial comparing WBRT plus memantine with hippocampal sparing WBRT plus memantine reported significantly preserved patient-reported quality of life and prevention of cognitive decline in the latter group [22].

Other Local Therapies

Laser interstitial thermal therapy (LITT) is an innovative intervention currently under evaluation for recurrent brain tumours following other local therapies but also in the management of radiation necrosis. Local control data are promising; however, the role of LITT managing BM requires further studies [23].

Systemic Therapy Targeting HER2 Disease

A systematic review and meta-analysis showed that HER2+ MBC patients with CNS disease have a better OS when treated with HER2-targeted therapy versus non-targeted therapy [24]. Therefore, most relevant data on treatment of CNS metastases with HER2-targeted therapy will be reviewed. Table 1 summarizes the most relevant completed and ongoing studies.

Table 1.

Evidence on anti-HER2−targeted therapy and brain metastasis − more relevant completed and ongoing studies

Anti-HER2−targeted therapy	Study	Type of study	Treatment arms	Number of patients with baseline BM	Type of BM included at baseline	CNS radiological evaluation	Neuroimaging assessment	Outcome evaluated for patients with BM
Pertuzumab	CLEOPATRA [27]	Phase III RCT	Pertuzumab, trastuzumab, docetaxel versus placebo, trastuzumab, docetaxel	None	None	Only if clinical suspicion of BM	RECIST criteria	Incidence of CNS metastasis as the first site of disease progression (similar: 13.7% with pertuzumab vs. 12.6% with placebo) Time do BM development (15 with pertuzumab vs. 11.9 months with placebo (H R = 0.58; p = 0.0049)
	PATRICIA [28]	Phase II	Pertuzumab and high-dose trastuzumab	39	CNS progression despite prior RT	MRI	RANO criteria	CNS-ORR: 11% (95% CI: 3–25) CBR at 4 months: 68%; at 6 months: 51%
T-DM1	KAMILLA [30]	Phase IIIb single-arm	T-DM1	398 (126 with measurable disease)	Asymptomatic BM	Not mandatory	RECIST criteria	ORR 21.4% (95% CI: 14.6–29.6) CBR 42.9% (95% CI: 34.1–52.0) Median PFS 5.5 months (95% CI: 5.3–5.6) Median OS 18.9 months (95% CI: 17.1–21.3)
	EMILIA [31]	Phase III	T-DM1 versus capecitabine plus lapatinib	95	Treated and asymptomatic BM	MRI or CT-scan	RECIST criteria	CNS progression: 22.2% versus 16.0% Median PFS: 5.9 versus 5.7 months (HR = 1.00; 95% CI: 0.54–1.84; p = 1.000)
T-DXd	DESTINY-Breast 01 Phase II [34]	T-DXd	24	Treated and asymptomatic BM	Not mandatory	RECIST criteria	ORR: 58.3% Median PFS: 18.1 months CNS response rate: 50%
	DESTINY-Breast 03Phase III RT T-DXd versus T-DM1 [35]	82	Treated and asymptomatic BM	Not mandatory	RECIST criteria	ORR: 67.4% versus 20.5%, CNS RR: 63.9% versus 33.4%, Median PFS: 15 versus 3 months
	TUXEDO [36]	Phase II	T-DXd	15	Untreated or progressing after local treatment	MRI	RANO criteria	CNS RR: 73.3% Median PFS: 14 months
	DEBBRAH [37]	Phase II	T-DXd	21	Stable or progressing BM, including LMC	MRI	RANO criteria	CNS RR: 46.2% (95% CI: 19.2–74.9) ORR: 66.7% (cohort 1), 80.0% (cohort 2), 66.7% (cohort 3)
Lapatinib	[60]	Phase 1	Lapatinib	11	New or progressive CNS metastasis or LMC	MRI	RECIST criteria	CNS RR: 27.2% (2 with LM)
	LANDSCAPE [40]	Phase II	Capecitabine-lapatinib	45	Untreated BM	MRI	RECIST criteria	CNS RR: 66% CNS-PFS: 5.5 months
	EGF100151 [41]	Phase III RCT	Capecitabine-lapatinib versus lapatinib	NE	Asymptomatic BM	Not mandatory	RECIST criteria	CNS as the first prog ression site: 2% versus 7%
	CEREBEL [42]	Phase III RCT	Capecitabine-lapatinib versus trastuzumablapatinib	NA	None	MRI	RECIST criteria	CNS as first progression site: 3% versus 5% PFS: 6.6 versus 8.1 months (HR, 1.30; 95% CI: 1.04–1.64), OS: 22.7 versus 27.3 months (HR, 1.34; 95% CI: 0.95–1.90)
Neratinib	NEfERT-T[45]	Phase III RCT	Neratinib-paclitaxel versus trastuzumab-paclitaxel	18	Asymptomatic BM	Not mandatory	RECIST criteria	CNS recurrence: 8.3% versus 17.3%, RR 0.48; (95% CI: 0.29–0.79) p = 0.002)
	NALA [46]	Phase III RCT	Capecitabine-neratinib versus capecitabinelapatinib	101	Asymptomatic and stable BM	Not mandatory	RECIST criteria	CNS intervention: 23% versus 29% (p = 0.043)
Tucatinib	HER2CLIMB [48]	Phase III RCT	Capecitabine-trastuzumabtucatinib versus capecitabine-trastuzumab	291	Stable or active BM	MRI	RECIST criteria	Median PFS: 7.6 versus 5.4 months (HR 0.48; 95% CI: 0.34–0.69; p < 0.001) 1 −year CNS-PFS: 40.2% versus 0% (HR 0.32; 95% CI: 0.22–0.48; p < 0.0001) 2-years OS: 51 % versus 40%
Pyrotinib	PHENIX [51]	Phase III RCT	Capecitabine-pyrotinib versus capecitabine	31	Asymptomatic BM	NA	RECIST criteria	PFS: 6.9 vs. 4.2 months (HR = 0.32, p = 0.011)
	PERMEATE [52]	Phase II	Capecitabine-pyrotinib	78	Untreated BM (cohort 1) progressing BM after local treatment (cohort 2)	MRI	RECIST criteria	CNS RR: 74.6% (cohort 1) and 42.1% (cohort 2)

BM, brain metastasis; CBR, clinical benefit rate; CI, confidence interval; CNS, central nervous system; CT, computed tomography; HR, hazard ratio; LMC, leptomeningeal carcinomatosis; MRI, magnetic resonance; ORR, overall response rate; PFS, progression-free survival; RR, response rate; OS, overall survival; RANO, Response Assessment in Neuro-Oncology; RECIST, Response Evaluation Criteria in Solid Tumors; RCT, randomized controlled trial.

Monoclonal Antibodies

Trastuzumab

Due to its high molecular weight, trastuzumab, a monoclonal antibody against HER2, has been considered not to cross the BBB. However, BM may increase the permeability of the BBB. Strong evidence of the benefit of the use of trastuzumab in patients with BM is lacking, but observational studies have reported clinical benefit. In RegistHER, 377 (37.3%) patients developed CNS metastases. The use of trastuzumab was associated with an improvement in the median OS following diagnosis of CNS disease (17.5 vs. 3.8 months) [25]. Retrospective studies also found an OS benefit with trastuzumab after BM development [26, 27].

Pertuzumab

The CLEOPATRA trial, which assessed the addition of the monoclonal antibody anti-HER2 pertuzumab to trastuzumab plus chemotherapy in the first-line treatment of HER2+ MBC, excluded patients with baseline BM. However, an exploratory analysis showed that while the incidence of BM as the first site of disease progression was similar, the median time to its development was 11.9 months in placebo arm versus 15 months in the pertuzumab arm (HR = 0.58; p = 0.0049). Further, a trend in OS favouring pertuzumab may be partly attributed to delaying the onset of BM [28]. Focussing on the treatment of BM, the combination of pertuzumab with high-dose trastuzumab was investigated in a phase II trial − PATRICIA − that included 39 patients with HER2+ MBC with BM and CNS progression despite prior RT. CNS overall response rate (ORR) per Response Assessment in Neuro-Oncology (RANO) BM criteria was low at 11%; however, CNS clinical benefit rate (CBR) was 68% at 4 months and 51% at 6 months, indicating some intracranial activity of this combination [29].

Margetuximab

Margetuximab, a chimeric anti-HER2 immunoglobulin G1 monoclonal antibody, in association with chemotherapy, was evaluated for previously treated HER2+ advanced BC (ABC) − SOPHIA trial[30]. BM was allowed if stable. Overall, 13.9% of patients in the margetuximab arm and 12.6% in the trastuzumab arm had BM, but no subgroup analysis for these patients is available.

Antibody-Drug Conjugates

Trastuzumab Emtansine

The KAMILLA trial tested ado-trastuzumab emtansine (T-DM1), a HER2-direct antibody-drug conjugate, in previously treated HER2+ MBC. Three hundred ninety-eight patients had baseline BM, of which 126 with measurable BM. The best ORR and CBR were 21.4% (95% CI: 14.6–29.6) and 42.9% (95% CI: 34.1–52.0), respectively. Median progression-free survival (PFS) and OS were 5.5 (95% CI: 5.3–5.6) months and 18.9 (95% CI: 17.1–21.3) months, respectively [31].

In the pivotal trial of T-DM1 in MBC-EMILIA, 95 patients had been treated for asymptomatic BM at baseline. In an exploratory retrospective subgroup analysis, an improvement in OS was observed with T-DM1 compared with the control arm (HR = 0.38; p = 0.008; median: 26.8 vs.12. 9 months). However, no difference in PFS was observed: median PFS was 5.9 months for T-DM1 and 5.7 months for lapatinib plus capecitabine (HR = 1.00 [95% CI: 0.54–1.84]; p = 1.00) [32]. Interesting data also came from the KATHERINE trial, which assessed T-DM1 in an adjuvant setting in case of residual disease post-neoadjuvant treatment in HER2+ BC. The CNS was one of the sites of first recurrence in approximately 5% of patients in both treatment groups [33]. Also, CNS incidence at any time was similar between arms (6.1% vs. 5.4%, respectively), as was median OS after CNS recurrence (12.5 months with T-DM1 vs. 14.3 months with trastuzumab) [34].

Trastuzumab Deruxtecan

Trastuzumab deruxtecan (T-DXd), a novel HER2-direct antibody-drug conjugate, was tested in advanced HER2+ BC following failure of T-DM1 (DESTINY-Breast 01). Enrolment of patients with stable BM was allowed. A subgroup analysis of 24 patients with BM was published showing that ORR, median PFS, and median duration of response were comparable to the total population (58.3% vs. 60.9%; 18.1 months vs. 16.4 months, and 16.9 months vs. 14.8 months). Also, in patients with available baseline BM diameter data, the CNS RR was 50% (7 out of 14) [35].

DESTINY-Breast 03 trial studied T-DXd versus T-DM1 in HER2+ MBC. In the 82 patients with stable BM included, median PFS was significantly better at 15 months with T-DXd versus 3 months with T-DM1, ORR was 67.4% versus 20.5%, and intracranial RR (iRR) 63.9% versus 33.4%, respectively [36]. In the context of this signal of T-DXd intracranial benefit, there are two ongoing phase II trials specifically addressing the efficacy of T-DXd in patients with BM: TUXEDO-1 and DEBBRAH. TUXEDO-1 recruited 15 patients with previously treated HER2+ MBC and newly diagnosed BM, or BM progressing after local therapy. According to RANO criteria, the most recent update showed an iRR of 73.3%, by RANO criteria. At 11 months of median follow-up, PFS was 14 months [37]. DEBBRAH recruited 41 pre-treated patients with HER2+ or HER2-low-expressing ABC with CNS disease, divided into 5 cohorts: (1) HER2+ ABC with non-progressing BM after RT and/or surgery; (2) HER2+ or HER2-low-expressing ABC with asymptomatic untreated BM; (3) HER2+ ABC with progressing BM after local treatment; (4) HER2-low-expressing ABC with progressing BM after local treatment; (5) HER2+ or HER2-low-expressing ABC with leptomeningeal carcinomatosis (LMC) [38]. In cohort 1, 16-week PFS rate was 87.5% (95% CI: 47.3–99.7; p < 0.001). The iRR was 50.0% (95% CI: 6.7–93.2) in cohort 2 and 44.4% (95% CI: 13.7–78.8; p < 0.001) in cohort 3. Overall, the iRR in patients with active BMs was 46.2% (95% CI: 19.2–74.9) [39]. A report on all cohorts is awaited. DESTINY-Breast 12 is also ongoing, with the primary objective of cohort 2 being PFS in patients with BM at baseline: untreated and not needing immediate local therapy or previously treated stable or progressing (NCT04739761).

Tyrosine Kinase Inhibitors

Lapatinib. In the phase II LANDSCAPE trial, lapatinib, a dual EGFR and HER2 tyrosine kinase inhibitor (TKI), in combination with capecitabine was evaluated in HER2+ MBC with untreated BM. CNS-ORR was 66% and CNS-PFS was 5.5 months. Median time to RT was 8.3 months [40]. However, in a meta-analysis of eight studies in patients with HER2+ MBC with BM, pooled CNS-ORR with lapatinib plus capecitabine was much lower: 29% [41]. EGF100151 trial permitted baseline stable BM; 2% of patients had progressive BM while receiving lapatinib plus capecitabine compared with 7% of patients receiving capecitabine alone (p = 0.10) [42]. The CEREBEL study enrolled HER2+ MBC without BM. The incidence of CNS metastases as the first site of relapse was comparable with lapatinib plus capecitabine (3%) and trastuzumab plus capecitabine (5%; treatment difference, −1.6% [95% CI: −2 to 5%]; p = 0.360) [43].

Neratinib. Neratinib (EGFR and HER2 TKI) has shown minor activity as monotherapy in previously treated HER2+ MBC with progressive BM [44]. However, for patients with progressive HER2+ CNS metastases receiving neratinib plus capecitabine, CNS-ORR was 49% among TKI-naïve patients and 33% among lapatinib previously treated patients [45].

NEfERT-T trial allowed inclusion of patients with stable BM. CNS recurrence was lower with neratinib plus paclitaxel than with trastuzumab plus paclitaxel (8.3% vs. 17.3%; RR 0.48 [95% CI: 0.29–0.79]; p = 0.002) [46].

The NALA trial randomized MBC to neratinib plus capecitabine versus lapatinib plus capecitabine after two lines of anti-HER2-targeted therapies. BM was allowed if asymptomatic or stable, although baseline brain screening was not mandatory. Overall, there was a significantly lower incidence of intervention for BM in neratinib arm versus lapatinib arm (23% vs. 29%; p = 0.043) [47].

Afatinib. Afatinib (an irreversible EGFR and HER2 TKI) was investigated in previously treated and progressive BM of HER2+ BC-LUX-Breast 3 trial. This was a negative trial, since afatinib arms were not better than the investigator's choice of treatment [48].

Tucatinib

HER2CLIMB was the first large RCT in patients with HER2+ MBC to include patients with active BM, in addition to stable BM. Brain imaging at baseline was mandatory. Almost half of patients (47.5%) had BM: 40% treated and stable, 37% treated and progressing and 23% untreated. Among patients with BM, PFS was significantly longer with the addition of tucatinib, a potent reversible HER2 TKI: median PFS was 7.6 months versus 5.4 months in the placebo arm (HR = 0.48; 95% CI: 0.34–0.69; p < 0.001). Additionally, 1-year CNS-PFS was 40.2% and 0%, respectively (HR = 0.32; 95% CI: 0.22–0.48; p < 0.0001) [49]. This significant improvement was consistent in both patients with active and stable BM. In the overall population, an estimated 2-years OS was 51% in the tucatinib and 40% in the placebo arms. This OS benefit was consistent across all subgroups, including patients with CNS metastases (HR = 0.60 [0.44–0.81]) [50]. These results questioned the relevance of TKI in the adjuvant setting for HER2+ BC to prevent BM development [51]. COMPASS-HER2 is ongoing and evaluating tucatinib plus T-DM1 in case of residual disease following neoadjuvant chemotherapy. CNS disease-free survival is a secondary objective (NCT04457596).

Pyrotinib. PHENIX trial evaluated pyrotinib (a potent and irreversible EGFR and HER2 TKI) plus capecitabine versus placebo plus capecitabine. Baseline asymptomatic BM was allowed. In this subgroup, PFS was higher (6.9 months vs. 4.2 months, HR = 0.32, p = 0.011) with pyrotinib combination [52]. Pyrotinib was also specifically investigated in association with capecitabine in HER2+ MBC with BM − PERMEATE phase II trial. Patients were enrolled in cohort A, if RT-naive or cohort B in case of disease progression or new lesions after WBRT or SRS. The iRR (by Response Evaluation Criteria in Solid Tumors (RECIST) criteria) was 74.6% in cohort A (95% CI: 61.6–85.0; 44 of 59 patients) and 42.1% (20.3–66.5; 8 of 19 patients) in cohort B [53].

Epertinib. Epertinib (a potent reversible EGFR, HER2 and HER4 TKI) is being studied for previously treated HER2+ MBC. To our knowledge, human data are only available from a phase I/II study of epertinib with trastuzumab with or without chemotherapy. Across all cohorts, a reduction on the longest diameter was observed for 4 of the 5 patients with CNS target lesions, by RECIST criteria [54].

Combination of Systemic Therapy with Radiotherapy

Targeted therapies and radiotherapy can have a synergistic effect, but potential safety concerns should be considered [55]. Data on safety and efficacy of anti-HER2 therapies plus RT are limited. Experts consider that for patients with stable systemic disease and limited CNS progression amenable to SRS, anti-HER2 monoclonal antibodies can be continued throughout RT. In case of repetitive CNS progression over a short period of time, deferring SRS and switching of systemic therapy might be an option [8].

Leptomeningeal Carcinomatosis

LMC has a dismal prognosis, barely reaching 8 months with treatment. Although HER2+ BC shows CNS tropism, frequency of this intrinsic subtype among LMC patients is 14–29% [56]. LMC can cause a plethora of neurological signs and symptoms that may be subtle and unspecific. Cerebrospinal fluid (CSF) cytology has impaired sensitivity, which can lead to a diagnosis delay. Treatment options include RT, intrathecal therapy (IT), systemic treatment, and symptomatic care. Evidence to support treatment is limited [57]. Brain imaging and CSF cytology are not quantitative, making it challenging to assess therapy response. Advances in CSF liquid biopsy are helping to address this unmet need and should be incorporated into clinical trials [58].

RT can be used for symptomatic sites, restore CSF flow, enhance efficacy from IT therapy, or when BM co-exists. IT chemotherapy can also be used [57], with methotrexate being the most frequently used agent. For systemic treatment, newer agents are being investigated: A peptide-drug conjugate − paclitaxel trevatide − showed CNS and systemic response and better survival rate in 16 heavily pre-treated HER2+ BC patients with LMC [59]. Pembrolizumab was studied in patients with LMC from solid tumours, including 6 patients with HER2+ BC. The study found 4 of these 6 patients alive at 3 months [60].

For HER2+ BC with LMC, IT administration of trastuzumab has been described in case reports. Maximum tolerated dose of IT weekly trastuzumab was 150 mg [61].

In another study, high-dose lapatinib was given intermittently and sequentially with capecitabine for patients with HER2+ BC with isolated CNS progression (BM and LMC). Two of 11 patients had LMC and were on study for at least 6 months before progression [62].

The role of newer anti-HER2 systemic drugs is being assessed in patients with LMC: tucatinib with trastuzumab and capecitabine is being investigated for HER2+ BC (NCT03501979). In addition, a cohort of the DEBBRAH trial is also investigating the role of T-DXd in 7 patients with HER2+ or HER2-low-expressing MBC with LMC.

Conclusion

For specific subtypes of lung cancer and melanoma, first-line systemic therapy is an established treatment paradigm for asymptomatic BM, delaying local therapy until intracranial progression. For BC, specifically in HER2+ disease, there is an emerging shift towards this paradigm. Nevertheless, treating BM from HER2+ MBC remains a clinical challenge and requires multidisciplinary management. Local therapies play an important role, particularly in patients with good prognostic factors. However, accumulating evidence on the intracranial activity and clinical benefit of anti-HER2-targeting therapies is likely to lead to more widespread use of systemic therapy in the management of HER2+ BM.

Nowadays, BM is no longer an absolute exclusion criteria. Most ongoing anti-HER2 clinical trials accept stable BM, and some also allow active BM. This shift is also accompanied by an important change in trial design: inclusion of end points to evaluate intracranial activity. Standardization in evaluating these outcomes is essential, for example, using RANO criteria. In addition, there are an increasing number of studies that directly address BM as their primary end point.

With emerging knowledge on the effectiveness of drugs on CNS metastasis, refining therapy in patients with HER2+ BC CNS metastases may even start prior to the development of symptomatic BM. The role of screening patients with MBC is being addressed in 3 prospective studies, using magnetic resonance imaging screening for asymptomatic high risk MBC patients (NCT03881605, NCT03617341, and NCT04030507).

Furthermore, recognizing the brain as a frequent site of first relapse in HER2+ disease, adjuvant trials also need to be designed with CNS outcomes as secondary end points. These strategies might improve the historically poor prognosis of HER2+ MBC patients with BM.

Conflict of Interest Statement

Marta Filipa Freire Vaz Batista: consulting/advisory board: AstraZeneca; speaker/conferences: Daiichi-Sankyo and Nutricia; travel fees: AstraZeneca, Daiichi-Sankyo, Pfizer, and GSK. Inês Eiriz: speaker/sonferences: Novartis; travel fees: Gilead, Novartis, and Pierre Fabre.

Amanda Fitzpatrick: no conflict of interest to declare. Fanny Le Du: consulting/advisory board: Daiichi-Sankyo, Lilly, Seagen Inc., Novartis, Pfizer, Roche, and Sandoz; speaker/conferences: Amgen, Lilly, Novartis, and Pierre Fabre; travel fees: Daiichi-Sankyo, Lilly, Novartis, Pierre Fabre, and Pfizer.

Sofia Braga: consulting/advisory board: Daiichi-Sankyo, AstraZeneca, Novartis, and Roche; speaker/conferences: Daiichi-Sankyo, AstraZeneca, Novartis, and Roche; travel fees: Daiichi-Sankyo, AstraZeneca, Novartis, and Roche. Diogo Alpuim Costa has received honoraria from the Portuguese Navy, CUF Oncologia, and NTT DATA and has served as a speaker, advisory board member, or has received research or education funding from AstraZeneca, CUF Oncologia, Daiichi-Sankyo, Gilead, Hoffmann-La Roche, Merck KGaA, Merck Sharp & Dohme, Nestlé, Novartis, Pfizer, Nanobiotix, Puma Biotechnology Inc., Sanofi, Seagen Inc., and Uriage.

Funding Sources

No funding sources.

Author Contributions

Marta Filipa Freire Vaz Batista was responsible for the concept and design. Marta Filipa Freire Vaz Batista, Inês Eiriz, Amanda Fitzpatrick, Fanny Le Du, Sofia Braga e Diogo, and Alpuim Costa were responsible for writing, reviewing, and revision of the manuscript. All the authors read and approved the final version of this paper.

Funding Statement

No funding sources.