Navigating the Complexities of Brain Metastases Management

Abstract

The management of brain metastases, a potentially devastating complication of advanced cancers, has become increasingly complex with advancements in local and systemic therapies. Improved outcomes and extended survival for patients with metastatic solid tumors have led to a surge in the prevalence and possibly incidence of brain metastases, affecting up to 40% of individuals with solid tumors. Enhanced imaging technologies contribute to more accurate and early detection, shaping the understanding of the intricate landscape of this condition. Traditionally, surgery and radiation stood as the mainstays of treatment due to the limited efficacy of systemic therapies within the brain. However, emerging clinical data, particularly in melanoma, lung, and breast cancers, reveal promising results with novel systemic treatments such as immunotherapy and targeted therapies. Despite the historical exclusion of patients with active brain metastases from clinical trials, a shift is occurring towards a more inclusive approach. This chapter delves into the multifaceted challenges associated with managing brain metastases, with a focus on the evolving landscape of systemic approaches as well as the intricacies of shared decision-making, providing a comprehensive overview of the current state and future directions in navigating the complexities of brain metastases management.

Short Description

Management of brain metastases from solid tumors is an evolving field. This chapter explores the complexities in multidisciplinary management of brain metastases, offering insights into current strategies, future directions, and the importance of using high level communication skills to ensure patients receive goal-concordant care.

Brief Article Summary

This article reviews current practices and challenges in multidisciplinary management of brain metastases from solid tumors with a focus on advances in systemic therapies used to treat brain metastases, along with the communication skills required for optimal patient care.

INTRODUCTION

Central nervous system (CNS) metastasis is one of the most potentially devastating complications of advanced cancers. Advancements in local and systemic therapies for solid tumors have dramatically improved the outcomes and lifespan of patients with metastatic disease. This has resulted in an increase in the incidence of brain metastases. Up to one third of patients with solid tumors will develop brain metastases, resulting in an estimated annual incidence of 70,000–400,000 cases in the United States (1). Enhanced imaging modalities have also allowed for more accurate and earlier detection of brain metastases. For decades, local therapies such as surgery and radiation remained the mainstay of treatment of brain metastases, primarily due to lack of systemic therapies with meaningful intracranial activity. In addition, clinical trials historically excluded patients with active brain metastases, therefore limiting the extent of evidence on the intracranial efficacy of systemic therapies. While local treatment approaches with surgery and radiation continue to play a critical part in management of patients with brain metastasis, emerging clinical data from solid tumors, including melanoma, lung, and breast cancer, shows encouraging efficiency of novel systemic therapies such as immunotherapy and targeted therapies in controlling intracranial disease. In this chapter, we review the intricate landscape and multifaceted challenges associated with management of brain metastases originating from solid tumors such as melanoma, lung, and breast cancer. We focus on systemic therapeutic strategies for brain metastases, with a particular emphasis on the evolving landscape of immunotherapies and novel targeted agents (Table 1) as well as the intricacies of shared decision-making and communication skills required for optimal patient care.

Table 1.

Summary of selected clinical studies on systemic therapies for brain metastasis from different solid tumors

	Study type	Study population	Treatment	Intracranial Response rate	Median PFS (months)	Median OS (months)	Median Follow-up (months)
Melanoma
Margolin et al. 9	Phase II	Cohort A: asymptomatic	Ipilimumab	Cohort A: 24% Cohort B: 10%	Cohort A: 1.4 Cohort B: 1.2	Cohort A: 7.0 Cohort B: 3.7	Not available
		Cohort B: symptomatic
Kluger et al. 10	Phase II	Asymptomatic, no previous local treatment	Pembrolizumab	26%	2	17	24
Tawbi et al. 11	CheckMate 204 (Phase II)	Cohort A: asymptomatic Cohort B: symptomatic, on low dose corticosteroid	Ipilimumab plus Nivolumab	Cohort A: 53.5% Cohort B: 16.7%	Cohort A: NR Cohort B: 1.2	Cohort A: NR Cohort B: NR	Cohort A: 34.3 Cohort B: 7.5
Long et al. 12	ABC (Phase II)	Cohort A and B: asymptomatic, no previous local treatment Cohort C: symptomatic, disease failure after prior local therapy	Cohort A: Ipilimumab plus Nivolumab Cohort B: Nivolumab Cohort C: Nivolumab	Cohort A: 46% Cohort B: 20% Cohort C: 6%	Cohort A: NR Cohort B: 2.5 Cohort C: 2.3	Cohort A: NR Cohort B: 18.5 Cohort C: 5.1	17
Long et al. 14	BREAK-MB (Phase II)	Cohort A: no prior local treatment Cohort B: progression after prior local treatment	Dabrafenib	Cohort A: 39% Cohort B: 30%	Cohort A: at least 8 weeks Cohort B: at least 16 weeks	Cohort A: at least 16 weeks Cohort B: at least 21 weeks	Not available
Davies et al. 16	COMBI-MB (Phase III)	Cohort A: BRAF V600E, asymptomatic, no prior local treatment Cohort B: BRAF V600E, asymptomatic, with prior local treatment Cohort C: BRAF V600D/K/R, asymptomatic, with or without prior local treatment Cohort D: BRAF V600D/E/K/R, symptomatic, with or without prior local treatment	Dabrafenib plus Trametinib	Cohort A: 58% Cohort B: 56% Cohort C: 44% Cohort D: 59%	Cohort A: 5.6 Cohort B: 7.2 Cohort C: 4.2 Cohort D: 5.5	Cohort A: 10.8 Cohort B: 24.3 Cohort C: 10.1 Cohort D: 11.5	8.5
Non-small cell lung cancer
Reungwetwattana et al. 25	FLAURA (Phase III)	Asymptomatic or stable CNS metastases, EGFR mutated	Osimertinib	91%	NR	Not available	15
Gadgeel et al. 29	Pooled analysis of NP28761 (phase II) and NP28673 (phase II)	Asymptomatic treated or untreated CNS metastases, ALK mutated	Alectinib	64%	Not available	Not available	12.4
Dziadziuszko et al. 34	Pooled analysis of ALKA-372-001, STARTRK-1, and STARTRK-2 (phase I/II)	Asymptomatic or pretreated and controlled CNS metastases, ROS1 fusion–positive	Entrectinib	79%	12	Not available	15.8
Negrao et al. 37	KRYSTAL-1 (phase I/II)	Asymptomatic untreated CNS metastases, KRAS G12C mutated	Adagrasib	42%	5.4	11.4	13.7
Goldberg et al. 39	Phase II	Asymptomatic untreated or previously treated CNS metastases Cohort A: PDL1 >>1% Cohort B: PDL1 <1%	Pembrolizumab	Cohort A: 29.7% Cohort B: 0	Cohort A: 1.9	9.9	8.3
Nadal et al. 40	ATEZO-BRAIN (Phase II)	Asymptomatic CNS metastases	Atezolizumab plus carboplatin and pemetrexed	42.7%	6.9	Not available	31
Breast cancer
Krop et al. 62 Verma et al.66	EMILIA	Asymptomatic CNS metastases	T-DM1	Not available	5.9	26.8	19.1
Murthy et al. 46 Lin et al. 63, 64	HER2CLIMB (Phase II)	Asymptomatic CNS metastases or symptomatic after undergoing local therapy	Tucatinib in combination with trastuzumab and capecitabine.	47.3%	9.9	21.6	14
Cortes et al. 45 Modi et al. 65 André et al. 67	DESTINY-Breast01, −02, −03	Asymptomatic treated/stable or untreated/active CNS metastases,	Trastuzumab deruxtecan	Treated/stable brain metastases: 45.2% Untreated/active brain metastases: 45.5%	Treated/stable brain metastases: 12.3 Untreated/active brain metastases: 18.5	Not available	Not available

NR: not reached, OS: overall survival, PFS: progression free survival.

MELANOMA BRAIN METASTASES

Rationale for Using Immune Checkpoint Inhibition in Melanoma Brain Metastases

Melanoma has a high tendency to metastasize to the brain and is the third most common primary solid tumor associated with brain metastases after breast and lung cancer (2). Melanoma brain metastases (MBMs) can be associated with high morbidity and median survival was historically about 4-5 months (3). Development of novel immune checkpoint inhibitors (ICIs) such as anti–cytotoxic T-lymphocyte associated protein 4 (CTLA-4) and anti–programmed cell death 1 (PD-1) antibodies has led to a paradigm shift in the treatment of metastatic melanoma. However, significant challenges remain and approximately 45% of patients with metastatic melanoma develop acquired resistance to these therapies (4). ICIs have demonstrated promising clinical activity in brain metastases from solid tumors. For decades, the brain was considered an immune sanctuary, but emerging evidence has underscored the dynamic and finely regulated nature of the immune microenvironment within the brain (5). Studies in 2015 revealed the presence of functional meningeal lymphatic vessels in the brain, along with a high concentration of T cells and major histocompatibility complex (MHC) II-expressing immune cells in proximity to the dural sinuses (6). The immunoregulatory mechanisms at play in conjunction with blood brain barrier (BBB) limit the trafficking of immune cells into brain parenchyma, but the pathologic breakdown of BBB in brain metastasis could facilitate immune cell entry into the brain parenchyma. Comparative analysis of the immune profile of matched intracranial and extracranial melanoma metastasis samples revealed that compared to extracranial metastases MBMs have lower immune infiltrates with fewer T cells and myeloid cells but higher neutrophils, and less TCR diversity (7). However, MBM samples from patients treated with ICIs exhibit an increased number and diversity of infiltrating immune cells compared to treatment naïve MBM samples (8), indicating increased trafficking of immune cells into the brain in response to ICI therapy.

Despite the biologic premise of ICIs in brain metastases, landmark trials of ICIs in melanoma excluded patients with active brain metastases, largely due to concerns about CNS penetration of antibodies and potential for exacerbation of neurological symptoms. However, with the emergence of durable responses with ICIs in controlling extracranial disease, studies shifted their focus to specifically investigate the intracranial activity of ICIs and began actively enrolling patients with brain metastasis in clinical trials. Available data shows that ICIs have a meaningful intracranial activity with improvement in outcomes in both CNS disease control and overall survival in metastatic melanoma. Intracranial activity of ipilimumab, an anti-CTLA-4 antibody, was first described in a phase II trial of patients with metastatic melanoma in 2012 (9). Intracranial disease control was achieved in 24% of patients with asymptomatic MBMs and 10% of patients with symptomatic MBMs who required corticosteroids for symptom control. Median overall survival was 7 months in asymptomatic patients and only 3.7 months in the symptomatic cohort. Later, the intracranial activity of the anti-PD-1 antibody, pembrolizumab, was shown in a phase II study in patients with asymptomatic MBMs. In this study, intracranial responses were seen in 26% of patients with a median overall survival of 17 months (10). The combination of ipilimumab and nivolumab proved to be more effective in controlling intracranial disease compared to anti-PD1 monotherapy, as shown in CheckMate-204 and ABC studies (11, 12). In the CheckMate-204 study the combination of ipilimumab and nivolumab was associated with response rates of up to 57% in patients with asymptomatic MBMs, with 33% of patients achieving an intracranial complete response. Although response rates were lower in patients with symptomatic MBMs, the efficacy was durable in both populations with 3-year overall survival rates of 71.9% and 36.6% in asymptomatic and symptomatic cohorts, respectively (11). The RELATIVITY-047 trial is a phase II-III trial which evaluated the efficacy of the novel anti-lymphocyte activation gene-3 (LAG3) agent, relatlimab, in combination with nivolumab in patients with metastatic melanoma (13). Although patients with untreated brain metastases were excluded, the 2-year subgroup analysis showed promising results with a longer time to development of CNS metastasis in patients treated with relatlimab plus nivolumab. An ongoing phase II study (BLUEBONNET, NCT05704647) is investigating the intracranial efficacy of relatlimab plus nivolumab in patients with active MBMs.

Targeted Therapies in Melanoma Brain Metastases

The early clinical trials leading to the approval of targeted therapies for BRAF-mutated metastatic melanoma often excluded patients with brain metastasis. The phase II BREAK-MB trial investigated the intracranial activity of the BRAF inhibitor, dabrafenib, in patients with BRAF V600 mutant melanoma and asymptomatic brain metastasis, yielding an intracranial response rate of 39% in patients with untreated MBMs and 30.8% in those with prior brain directed local therapy (14). Similar results were observed with vemurafenib in a pilot study of patients with BRAF V600 mutated melanoma and symptomatic MBMs (15). The COMBI-MB phase II trial explored the combination of BRAF and MEK inhibitors, dabrafenib and trametinib in 4 different cohort of patients with BRAF V600E mutated melanoma and either asymptomatic or symptomatic MBMs with or without prior brain directed local therapy (16). After a median follow up of 8.5 months, intracranial responses were seen in 58% of patients with untreated and asymptomatic MBMs, and 56% of patients with previously asymptomatic MBMs who had received prior local therapy. 59% of patients with symptomatic MBMs with or without prior local therapy achieved an intracranial response. However, despite favorable responses, the median duration of response was only 6.5 months, which is shorter than the median duration of response with extracranial disease (12.0 months) (16, 17). The SECOMBIT phase II trial demonstrated superior clinical benefits with the combination of ipilimumab and nivolumab as the first-line treatment (18). The study allowed for inclusion of patients with MBMs provided they had stable disease for >4 weeks and were on steroid doses less than 10mg prednisone equivalent. The "sandwich" approach, involving initial targeted therapy followed by immunotherapy maintenance, showed clinical benefit in rapidly progressing cases (18). Updated 4-year OS data from SECOMBIT indicated a superior brain-metastasis survival benefit for first-line combination immunotherapy, with or without initial targeted therapy induction, potentially underlying its overall survival benefit (19).

LUNG CANCER BRAIN METASTASES

Using precision medicine to optimize the management of patients with brain metastases from lung cancer

Approximately 40% of patients with non-squamous non-small cell lung cancer (NSCLC) in the United States will have a molecular alteration that provides the opportunity to use a targeted therapy (20). Notably, the frequency of targetable alterations varies in different populations across the world based on ethnicity and smoking patterns. As therapeutics have advanced over the last two decades since the discovery of the EGFR mutation in 2004, the number of patients with lung cancer who are eligible for a targeted therapy has increased over time and likely will increase further in the coming years with the development of additional therapeutics. Targeted therapy has revolutionized the treatment paradigm for managing patients with advanced NSCLC. No longer are all patients with advanced lung cancer receiving chemotherapy as their initial treatment, which results in a median OS of approximately 1 year (21). Instead, an oral, molecularly targeted drug is used for patients whenever possible, with median survival rates increasing beyond 5 years as the therapeutic options have advanced (22).

In addition to the benefit of the targeted therapies for lung cancer becoming more potent and selective, in many cases the drugs have also become more brain-penetrant. For example, the first-generation of EGFR tyrosine kinase inhibitors (TKIs) including erlotinib and gefitinib have some CNS activity, however the third-generation EGFR inhibitor osimertinib has superior brain penetration and likely in part accounts for its superiority over the first-generation agents (23). With highly active, CNS-penetrant targeted therapies proving to be effective for a sizable portion of patients with advanced NSCLC, the management of brain metastases has shifted over the last several years. Whereas the standard algorithm for treating brain metastases previously included local therapy for all patients prior to initiating systemic therapy, there is increasing recognition that targeted therapies may be effective at controlling CNS disease.

Current management of patients with advanced EGFR-mutant lung cancer was informed by the phase 3 trial FLAURA that compared osimertinib to first-generation EGFR TKIs and demonstrated an improvement in OS, making osimertinib the standard first-line therapy for patients with EGFR-mutant lung cancer (24). A subset analysis from this trial focusing on brain metastases showed that the risk of CNS progression was reduced with osimertinib, with the median CNS progression free survival (PFS) not reached with osimertinib versus 13.9 months with standard EGFR TKIs (hazard ratio [HR] 0.48, 95% confidence interval [CI] 0.26-0.86; p=0.14) (25). Moreover, the CNS response rate was higher with osimertinib compared to erlotinib or gefitinib in patients with baseline brain metastases (91% vs 68%) and CNS PFS was prolonged as well (not reached vs 13.9 months, HR 0.48, 95% CI 0.26-0.86). Based on these encouraging results, common practice has evolved to include upfront EGFR TKI therapy rather than standard radiation therapy for patients with brain metastases. Accordingly, several retrospective studies have attempted to determine whether upfront EGFR TKI is a beneficial strategy compared to upfront radiation. Although an older retrospective study found an improvement in OS with upfront radiation compared to upfront TKI (26), most patients included received a first-generation TKI which limits extrapolation to current practice using a third-generation agents. A more recent retrospective study focusing on third generation TKIs did demonstrate similar outcomes for patients who received EGFR TKIs alone compared with those who received radiation followed by TKI, with a trend towards an improved PFS in those who received TKI alone (27). Although there has not yet been a randomized trial of osimertinib to treat brain metastases from EGFR-mutant NSCLC, oncologists frequently recommend deferring radiation therapy and instead use upfront osimertinib to avoid the potential toxicity of radiation for this subset of patients who may have long-term survival.

ALK-rearranged lung cancer also represents a paradigm shift for management of brain metastases given the development of several brain-penetrant ALK TKIs. The initial ALK inhibitor crizotinib that was used for this subset of patients had some activity in the CNS, but the newer generation of ALK TKIs are more brain penetrant and have superior outcomes. Notably, alectinib, brigatinib, and lorlatinib all have demonstrated high rates of durable CNS responses (28-32), and can be considered for upfront therapy of brain metastases or in patients who have CNS progression on an earlier generation ALK TKI. Specifically, the CNS response rate is around 80% for all 3 agents, and each has demonstrated prolonged duration of CNS response compared to crizotinib. Similarly, next-generation TKIs against ROS-1 including entrectinib and repotrectinib are highly effective in the CNS (33, 34), and should be considered in patients with brain metastases. In a small subset of 24 patients treated with entrectinib who had baseline brain metastases, the intracranial response rate was 79% and median CNS PFS was 12 months with entrectinib. There is less data available for other more rare subsets of patients with brain metastases, with small series showing effective CNS control for capmatinib in NSCLC with a MET exon 14 skipping mutation (35) and selpercatinib in RET-rearranged NSCLC (36). In contrast, the KRAS G12C inhibitors appear to have less CNS activity, in line with their lower overall response rates compared to targeted therapies for other subsets of NSCLC. Although adagrasib in particular did show an intracranial response of 42%, responses were typically not durable, and the median CNS PFS was only 5.4 months (37).

Clearly the revolution in precision medicine has reached patients with NSCLC, including those with brain metastases. As the therapies become more effective and brain-penetrant, we are seeing patients with CNS disease have prolonged survival, making it imperative that we carefully consider the impact of CNS-directed therapies that could have long-term side effects, particularly whole brain radiation therapy. We now have many examples of targeted therapies that can result in durable benefit in the CNS, potentially allowing radiation to be deferred in favor of upfront systemic therapy. Even so, it remains critical to develop better CNS-penetrating targeted therapies as CNS progression remains a risk in patients with oncogene-driven lung cancer.

Immunotherapy for NSCLC Brain Metastases

With incorporation of PD-1 and PD-L1 inhibitors into first-line therapy for patients with advanced NSCLC, we are witnessing more durable responses than ever seen before in patients with advanced lung cancer. The benefit of immunotherapy in patients with brain metastases was demonstrated in subset analyses from the large phase 3 randomized trials, though notably most trials required local therapy prior to initiating systemic therapy. For example, the KEYNOTE-189 trial which randomized patients with advanced non-squamous NSCLC to either chemotherapy plus pembrolizumab or chemotherapy plus placebo demonstrated an overall survival benefit with the addition of pembrolizumab in patients who had baseline brain metastases, with a median OS of 19.2 months vs 7.2 months with chemotherapy alone (HR 0.42 [95% CI 0.24-0.67]), which was a similar benefit as seen in patients without brain metastases (38). Importantly, the activity of immunotherapy has also been demonstrated in patients with untreated brain metastases, albeit in relatively small prospective trials. A phase II trial of pembrolizumab enrolled 37 patients with PD-L1-positive NSCLC who had untreated or progressing brain metastases that were 5-20 mm and were not causing neurologic symptoms. The trial demonstrated a CNS response rate of 29.7%, which was similar to the systemic response rate. Moreover, several patients on the trial were able to avoid radiation therapy to the brain and more than one-third were alive for at least 2 years after starting treatment (39). A more recent trial of chemotherapy plus atezolizumab in patients with untreated brain metastases also demonstrated CNS activity. Among the 40 enrolled patients, the intracranial response rate was 45% and the overall PFS rate at 12 weeks was 62.2% (40). Despite the clear evidence that immunotherapy has activity in the CNS in patients with NSCLC, additional studies are needed to determine whether certain patients may be able to avoid radiation and receive upfront systemic therapy with an immunotherapy-based regimen.

BREAST CANCER BRAIN METASTASES

In the latest decades, therapeutic advances for patients with early-stage, locally advanced, and metastatic breast cancers are significantly increasing rates of recurrence-free, progression-free and overall survival (41-51). Nonetheless, the prolonged survival with systemic therapies, better control of extracranial disease, and improvement of imaging techniques appear to result in a rise in the incidence of brain metastases from breast cancer.(52) Brain metastases occur in approximately 10-30% of patients with metastatic breast cancer, making it one of the most common sites of distant metastasis (52). Certain subtypes of breast cancer are associated with a higher risk of brain metastases, which are seen in approximately one-third of patients with human epidermal growth factor receptor 2 (HER2)-positive, in one third with triple-negative and in 15% of hormone receptor (HR)-positive breast cancers.(52) In addition to breast cancer subtype, advanced stage at diagnosis, inflammatory breast cancer, younger age, presence of germline breast cancer susceptibility gene 1 or 2 (BRCA1 or BRCA2) mutations, pulmonary metastases, and African American ancestry are also described as associated with an increased risk, although some of these may be confounding factors as also associated with diagnosis of triple-negative breast cancers (TNBC) (52-58).The prognosis for patients with brain metastases from breast cancer is still generally poor, with median survival ranging from a few months to about a year with less common cases of longer survival, especially in those with HER2-positive breast cancers due to improvements in systemic therapy since the advent of trastuzumab (59-61). Local treatment options often include surgery, SRS and WBRT, while systemic therapy choices vary from chemotherapies to targeted therapies and to novel agents such ICIs and antibody-drug conjugates (ADCs). The choice of treatment depends on factors such as the number and size of brain lesions, extent and control of extracranial disease, performance status, and molecular subtype. For instance, in HER2-positive breast cancer, targeted therapies such as trastuzumab, tucatinib and capecitabine, or ADCs such as trastuzumab emtansine (T-DM1) and trastuzumab deruxtecan (T-DXd) have demonstrated efficacy in controlling brain metastases and can sometimes be offered prior to local therapy to avoid neurological adverse events (62-67). For instance, in patients with intracranial measurable disease, the addition of tucatinib to trastuzumab and capecitabine significantly increased intracranial response rates from 20% to 47% while T-DXd increased intracranial response rates from around 30% with treatment of physician’s choice including T-DM1 to 45%. The use of ICI has shown intracranial disease control in select patients with brain metastases from breast cancers of all subtypes, but lack of predictive biomarkers and non-negligible rates of immune-related adverse events limit broader ICI use for this indication (68). For example, a phase 2 study assessing the efficacy of pembrolizumab in patients with brain metastases from tumors of all subtypes included 35 breast cancer patients and showed intracranial activity in 37% of the patients (90% CI: 24–52%) (68).

Despite advances in treatment, challenges remain, including the development of resistance to targeted therapies, neurocognitive deficits associated with radiation therapy, and limited therapeutic options with intracranial activity. Future research efforts are focused on identifying novel therapeutic targets and optimizing the sequencing and combination of treatments for patients with brain metastases from breast cancer. Additionally, upcoming clinical trial investigators are urged to consider not automatically excluding patients with active brain metastases from clinical trials if no biological concerns are predicted given importance of randomized controlled data for these patients (69). Continued research efforts are needed to address the unmet needs in this patient population and improve treatment strategies.

Role of Immunotherapy in Renal Cell Carcinoma Brain Metastasis

ICIs have also exhibited promising activity in brain metastases from other solid tumors such as renal cell carcinoma (RCC). Among genitourinary malignancies, RCC has the highest tendency to metastasize to the brain. ICIs have significantly improved the outcomes for patients with metastatic RCC (mRCC) and provide a more durable response compared to TKIs. However, the optimal systemic therapy for patients with mRCC and brain metastases is not defined. The first prospective study to investigate the intracranial activity of ICIs in mRCC was the GETUG-AFU 26 NIVOREN phase II trial, in which patients with asymptomatic RCC brain metastases (either untreated or previously treated) who had progressed on TKI were treated with nivolumab. Nivolumab monotherapy was associated with an intracranial response rate of approximately 12%, with median intracranial PFS of 2.7 months in patients with untreated brain metastases and 4.8 months in those with previously treated brain metastases (70). The phase III Checkmate 920 trial investigated the efficacy of the combination of nivolumab plus ipilimumab in patients with mRCC and clinical features excluded in the pivotal Checkmate-214 study including the presence of brain metastases. The investigator assessed overall response rates in patients with brain metastases was 32% with a median duration of response of 24 months, and median PFS of 9 months (71).

Local Therapies and Sequencing with Systemic Therapy

In patients with larger (>2 cm) or symptomatic brain lesions, upfront surgery remains the mainstay of treatment. In addition to controlling neurological symptoms, upfront tumor debulking prior to initiation of ICIs have been associated with improved patient survival (72). In cases with multiple metastatic brain lesions in which surgical resection is not feasible, local therapies such as stereotactic radiosurgery (SRS) or whole brain radiation (WBRT) are often utilized to control symptoms, allow for tapering of corticosteroids, and eventually serve as a bridge to systemic therapies such as ICIs which provide a more durable intracranial response. Limited prospective data exists on comparing SRS with WBRT. The QUARTZ study in patients with non-small cell lung cancer and poor prognosis did not show a survival or quality of life benefit from WBRT (though it did allow decreased corticosteroid use at end of life) (73), and two small prospective studies in patients with up to 4 metastatic brain lesions demonstrated no survival benefit to addition of WBRT to SRS and showed less cognitive decline in patients receiving SRS alone (74, 75). In practice, SRS is the preferred initial modality for patients with brain metastases from solid tumors, and WBRT is usually reserved for patients with a large number of brain metastases or leptomeningeal disease. American Society for Radiation Oncology (ASTRO) brain metastases guideline provides a practical approach to the best RT method based on number and size of brain metastases and patient performance status (76). The optimal timing for initiating ICIs, either before or after SRS, is unclear. The synergistic effect of radiation with immunotherapy is known as the abscopal effect and refers to the phenomenon in which radiation induces a distant anti-tumor response in non-irradiated sites (77), via increasing tumor antigen presentation and release of inflammatory cytokines (78). This synergistic effect was shown in a number of retrospective studies in which combination of ipilimumab with SRS led to increased median overall survival (79). Although there is a strong rationale for combining radiation with ICI therapy, there is conflicting information on additive treatment related toxicities such as radiation necrosis. In a retrospective analysis of 480 patients with brain metastases from melanoma, renal cell carcinoma or non-small cell lung cancer, treatment with single agent anti-PD1 therapy was associated with an increased risk of symptomatic radiation necrosis (HR, 2.56; 95% CI, 1.35-4.86; P = .004) (80). However, the timing of SRS relative to ICI treatment was not reported. Conversely, in a retrospective analysis of 54 patients with MBMs who received ipilimumab within 4 months of SRS, no difference in symptomatic radiation necrosis was seen compared to patients treated with SRS alone (81). Additional prospective studies are needed to determine the incidence of radiation necrosis in relation to timing of ICI treatment.

MULTIDISCIPLINARY MANAGEMENT OF BRAIN METASTASES

The many questions and challenges that arise when managing patients with brain metastases have led to the development of brain metastasis tumor boards at many institutions, and dedicated multi-disciplinary clinics dedicated to patients with brain metastases. These tumor boards provide a forum to review patients cases with multiple sub-specialties that are critical to the care of patients with CNS disease, including medical oncology, radiation oncology, neuro-oncology, and neurosurgery. Additionally, the presence of expertise in CNS radiology is critical given the challenging nature of interpreting brain imaging in many cases, as is the inclusion of a pathologist with proficiency in brain pathology. Knowledge of the unique supportive care necessary for patients with brain metastases is also extremely helpful given the potential physical and psychological challenges that many of our patients confront. Common discussions in brain metastasis tumor boards include decision-making regarding upfront management of patients with CNS metastases (particularly considerations around local therapy versus upfront systemic therapy), treatment options for patients with progression in the CNS, and questions of possible radiation necrosis versus tumor progression in previously radiated brain lesions. Given the importance of continuing to advance the field, a brain metastasis tumor board also provides an excellent opportunity for research teams to identify patients who may be eligible for a clinical trial or a non-therapeutic study of brain metastases.

COMMUNICATION AND SHARED DECISION-MAKING IN THE FACE OF UNCERTAINTY

When a patient comes into the office stating, “I saw the report in the portal that says I have brain metastases”, multiple questions should come to mind. What does the patient assume? What have they read online? What have they heard from family members? The misconceptions can be tremendous, particularly because the management of brain metastases is continuing to undergo a radical transformation, allowing patients with solid tumors metastatic to the brain to live longer and with better quality of life than ever before. Advances in systemic therapy, radiotherapy, surgery, prognostic tools and even communication science allow for improved tailoring of treatment; how do we best talk with patients and their families about the treatment options to ensure goal concordant care in the context of all of these options? This section will review prep work (prognostic tools, treatment algorithms) and the importance of optimal communication regarding five fundamental conversations to ensure patients are getting care that matches their individual wishes. The review is based on the paucity of literature on how to talk to patients with brain metastases and the 10+ years one author (JJ) has spent learning and teaching communication skills using SICP (https://www.ariadnelabs.org/serious-illness-care/, accessed 3/2/24) and VitalTalk (vitaltalk.org, accessed 3/2/24) methodology to teach hundreds of learners (from medical students to attending physicians) different approaches critical to caring for patients with advanced cancer, including those with brain metastases. The five conversations vital to talking with patients about brain metastases are similar to broader serious illness conversations and include delivering serious news, discussing prognosis, assessing values/goals, discussing treatment options and advance care planning (82).

Why does talking to patients about brain metastases continue to be a challenge in today’s era of modern oncology care? The care of patients with brain metastases has changed dramatically over the past 25 years. Validated prognostic tools demonstrate differences in survival seen in that time period: The recursive partitioning analysis (RPA) for brain metastases, first published in 1997, showed differences in median survival of 2.3 months for patients with the worst prognosis (those with Karnofsky Performance Status (KPS) <70) compared with 7.1 months for patients with the best prognosis (age <65, KPS>70, controlled primary and brain metastases as the only site of disease) (83). In comparison, the diagnosis-specific graded prognostic assessment (DS-GPA) incorporates histology specific characteristics, including molecular markers, to further predict survival for patients with brain metastases, ranging from median survivals of 3 months for the worst versus 47 months for the best prognosis categories of patients with NSCLC, breast cancer, melanoma, GI cancers or renal cell carcinoma (84). With this wide variability in predicted survival for patients with brain metastases, how do clinicians “meet patients where they are at” and ensure patients are getting the care that they want? How should clinicians respond when patients with a good performance status and prognosis are immediately ready for hospice, but may have a chance at prolonged survival? When patients with poor anticipated survival insist that “Another clinician told me the brain metastases are nothing to worry about”? When patients lose their capacity to participate meaningfully in conversations about care?

Prior to engaging in any of the five conversations described, there is “pre-work” required. Creating a team that can care for patients with brain metastases is of critical importance, including neurosurgeons, neurologists, neuro-radiologists, radiation oncologists, medical oncologists, and palliative medicine clinicians. In an optimal environment, this team will together consider prognosis, treatment options, anticipated outcomes, side effects of therapy and elicitation of patient preferences as critical to patient management, whether in the setting of a dedicated clinic or in separate visits. A pilot brain metastases clinic at the Cross Cancer Institute at University of Alberta showed rapid multi-disciplinary evaluation and high patient satisfaction scores (85). Other multi-disciplinary clinics focus on integration with palliative care (86) or allow histology specific interdisciplinary management of patients with breast cancer brain metastases, showing improved clinical trial enrollment and encouraging overall survival (87). The value of multi-disciplinary care stems from the ability to have multiple providers assess patients and discuss treatment options, be they local approaches including surgery or radiotherapy, systemic therapies or hospice care.

Even without a dedicated brain metastasis clinic, validated tools become a fundamental part of the pre-work. The DS-GPA and other prognostic models can help clinicians, patients and families understand how much time might be ahead (84). The ASTRO/ASCO/SNO guidelines for the management of brain metastases provide detailed algorithms about what treatment options may be appropriate, including making recommendations on the role of systemic therapy without local therapy (as previously described in this paper), and options for local therapies including surgical resection, stereotactic radiotherapy, whether single fraction or fractionated, whole brain radiotherapy with or without hippocampal avoidance and memantine and the role of supportive care/hospice without radiotherapy (88). ASTRO further released specific guidelines with algorithms that guide radiotherapeutic care of patients with brain metastases (76). The QUARTZ study suggests there is no difference in quality adjusted life years or survival for patients with NSCLC treated with palliative whole brain radiotherapy to 20 Gy in 5 fractions compared with optimal supportive care. Thus, for patients with poor prognosis, the guidelines suggest no radiotherapy. Even so, a recent “moment-in-time” analysis shows that the median survival of a global, unselected group of 462 patients treated with whole brain radiotherapy was only 84 days with wide variation in dose-fractionation, suggesting that there are opportunities for improved outcomes for patients with brain metastases and short survival (89). While the guidelines are invaluable in the care of patients with brain metastases and understanding potential outcomes, there is nuance to the discussions required to help patients make decisions that are right for them.

The five fundamental conversations that are critical to helping patients make clinical decisions about brain metastasis management (serious news, prognosis, eliciting values, discussing treatment options and advance care planning) (Figure 1) are sometimes accomplished in one visit, but often require multiple visits with different team members to ensure patients are being treated with goal concordant care. What steps are important in managing these conversations?

Figure 1. Communication guide for patients with brain metastases.

NURSE: Name, Understand, Respect, Support or Explore the emotion

A critical opening is to set a mutual agenda with the patient, letting them know what the clinician hopes to cover and asking what the patient/family would like to discuss. Using an Ask/Tell/Ask methodology can help with each step - asking what patients know, what they want to know and if it is okay to continue the conversation before delivering a concise headline, responding to emotion using NURSE statements and checking for understanding before continuing (CITATION: https://www.vitaltalk.org/guides/responding-to-emotion-respecting/; accessed 3/2/24).

1. Delivering serious news. Oncology clinicians in the United States recognize that this step has changed fundamentally since the passage of the 21st Century Cures Act required automatic and immediate release of medical information through an electronic portal (including imaging and pathology results that could contain “serious news”) (90, 91). Multiple guides for discussing serious news exist, including SPIKES, GUIDE and other acronyms that suggest finding an appropriate setting, assessing patient understanding, delivering a simple headline, responding to emotions, and then coming up with a plan for moving forward. With the Cures Act, the first ask of “tell me what you have heard about the scans” becomes increasingly important: patients may have already seen information about brain metastases (and have asked friends/family or Google about the diagnosis) and may not understand the medical jargon nor the implications of the diagnosis. The delivery of serious news, then, may actually be delivering new information, clarifying what has already been seen or confirming (or gently correcting) what patients have heard. Regardless, after elucidating patients’ understanding, it is critical to utilize a single “headline” statement to deliver the most important news about without jargon followed by a pause for processing. Expressions of emotion may indicate that a patient has understood and is processing a threat. Questions such as “How could this have happened?” or “Will I ever be the same?” likely indicate strong emotion rather than factual questions and, as such, should be responded to with empathic statements as described in the NURSE mnemonic to Name, Understand, Respect, Support or Explore the emotion (92). Silence is also critical as a tool. For many patients, further discussion may be appropriate at the initial visit, but for some, simply letting the patient know that the clinician will be there for support may be all that is possible at the initial discussion (“I will walk this path with you wherever it may lead”).

2. Discussing Prognosis. Not all patients want to hear about how much time they might have. Equipping them with information that will be helpful with shared decision-making is critical, but forcing information on them (such as a median survival) that may not be useful (and may even be detrimental) may harm the physician patient relationship (93). So how should prognosis be discussed? Multiple theoretical models exist, but simply asking patients and families what type of information might be helpful to them can help clinicians provide information in meaningful formats, including ranges of time, descriptions of function or independence (CITATION: Serious Illness Care Program, https://www.ariadnelabs.org/serious-illness-care/for-clinicians/, visited 3/2/24), best case/worst case scenarios (PMID 28062349, 35246401) or communicating the uncertainty in the clinical scenario. A straightforward description of what the future might hold, delivered in terms the patient can understand, should be followed by further opportunities to respond to patient emotions. Sometimes, this discussion can be reserved for the future or this may be all that a patient can handle in one visit.

3. Eliciting patient values/priorities. Discussing what is important to a patient/family as they look to the future is predicated on the patient having an understanding of what might lie ahead. This discussion of values/priorities may change based on the clinical scenario. If, for example, a patient believes that their life expectancy is very short because a close friend died two weeks after learning they had brain metastases, their values and priorities may be different than if they learn they have a favorable prognosis and may live for years with good quality of life with appropriate treatment. Thus, the “goals of care” conversation must be based in patient understanding of their clinical scenario in which the clinician is the expert in medicine knowledge and the patient and family are the experts in what is important to them, what they are frightened about in facing the future and how those gaps can be bridged as a team. Guidelines for discussing goals exist in multiple frameworks, including VitalTalk (vitaltalk.org, accessed 3/2/24), Serious Illness Conversation Guide (https://www.ariadnelabs.org/serious-illness-care/, accessed 3/2/24) and others (82). Whatever framework is utilized, open ended questions inquiring about hopes and fears can frame ongoing discussions about the care best suited for the individual patient.

4. Reviewing treatment options. The various treatment options for patients with brain metastases are expanding exponentially. However, even with the dramatic progress in the care of patients with brain metastases, most patients with solid malignancies with brain metastases will ultimately die from their cancer. The ASCO/ASTRO/SNO guidelines provide detailed algorithms for thinking about integrating neurosurgical, radiotherapeutic, systemic therapy and supportive care interventions to optimize care for patients with brain metastases. These reviews, including enrolling patients in clinical trials of each modality, should continue to form the basis for caring for patients with brain metastases.

5. Advance care planning. Even if patients have a good prognosis, planning for the future, including thinking about what care is appropriate if patients have clinical decline, is critical to the care of patients with brain metastases. Inquiring about healthcare proxy, Medical Orders for Life Sustaining Treatment (MOLST), wishes regarding hospitalization, ICU stays and what is important to the patient with end of life care is imperative to ensure goal concordant care. If the patient has already completed an advanced care plan, the diagnosis of brain metastases may be a good impetus to review that plan. If no advance care has been discussed, the topic should be added to the agenda for a future visit to ensure the patient will receive the care they want.

CONCLUSION

Brain metastases account for significant morbidity and mortality in patients with advanced cancer. Over the past decade, advancements in cancer therapeutics have markedly improved patient outcomes, resulting in longer survival and subsequently a potential rise in the incidence of brain metastasis. Despite this increasing incidence, treatment options for brain metastases remain limited. While novel therapies such as immunotherapies and targeted agents have shown promising results in controlling extracranial disease, development of effective systemic therapies for intracranial metastases have lagged behind. This is partly attributed to the historical exclusion of patients with active brain metastases from clinical trials, limiting the availability of data on clinical efficacy and safety of novel therapies. As our understanding of the distinctive microenvironment of brain metastasis advances, it becomes imperative to actively include patients with brain metastasis in prospective studies. With recent advances in prognostication, surgery, radiotherapy, systemic therapies and supportive care, the science of communication is more important than ever before. Regardless of the care that patients ultimately want, meeting them where they are is vital to supporting them on their cancer journey. Using the “ask/tell/ask” model, clinicians can provide information in ways that are meaningful, can respond to emotions empathically, and can enhance the clinician patient relationship to provide the highest level of care regardless of outcome. Managing patients with brain metastasis poses distinct challenges and complexities, emphasizing the critical role of multidisciplinary care.

Practical Applications Bullet Points.

• Management of brain metastases requires a multidisciplinary approach focusing on the choice and sequence of local and systemic therapies and clear patient-clinician communication.

• Immune checkpoint inhibitors and targeted therapies have shown promising intracranial activity with acceptable safety profile in brain metastases from melanoma, non-small cell lung cancer, breast and renal cell carcinoma.

• Patients with active brain metastases should be actively included in future prospective studies to gain better insight into intracranial activity of novel therapeutics.