Novel agents are showing impressive ability to cross the blood–brain barrier and impair tumor development in several malignancies, raising the prospect of achieving a long-elusive goal of anticancer therapy.
Eudocia Q. Lee, MD
Novel agents are showing impressive ability to cross the blood—brain barrier (BBB) and impair tumor development in several malignancies, raising the prospect of achieving a long-elusive goal of anticancer therapy. With advancements in lung cancer leading the way, investigators are increasingly seeking to incor-porate into clinical trials patients who develop brain metastases from primary tumors else-where in the body.
The BBB, with its limited permeability to systemic treatment, has traditionally interfered with cancer treatments. Some chemotherapy agents can pass through the BBB to attack cancer cells, including temozolomide (Temodar), lomustine (Gleostine), and carmustine (BiCNU). However, chemotherapy is generally not as effective as surgery and radiotherapy (RT), which remain standard treatments for brain tumors. Recent advances with targeted therapies and immunotherapies have fueled optimism that systemic therapies can play an important role in brain tumor management. Major advances have been made in the central nervous system (CNS) penetrance of targeted therapies for ALK-rearranged and EGFR-mutant non—small cell lung cancer (NSCLC), HER2-positive breast cancer, and melanoma.
For further progress, the enrollment of patients with brain metastases in clinical trials is essential. Statistics show that a huge unmet need exists for effective therapies that can address these cancers. Each year, more than 200,000 patients receive a diagnosis of metastatic brain cancer. This cancer occurs most often in patients with advanced primary lung cancer but also is common in breast cancer and melanoma.1 Up to 50% of patients with lung cancer, 10% with melanoma, and 16% with breast cancer will develop secondary lesions in the brain. For many of these patients, survival may be measured in months rather than years (Table 1).2
“Historically, most trials of systemic ther-apies have excluded patients with brain metastases on the assumption that patients with brain metastases made poor clinical trial candidates,” said Eudocia Q. Lee, MD, senior physician and assistant professor of neurology at Dana-Farber Cancer Institute in Boston. “But there’s data that properly selected patients with brain metastases can participate in phase I studies and provide important information.”Multiple trials, many of which are detailed below, are providing important results to guide clinicians in the effective use of targeted therapies and immunotherapies, setting the stage for additional breakthroughs in the treatment of what remain highly therapy-resistant cancers.ALK-Rearranged NSCLC
Next-generation agents targeting ALK-rearranged NSCLC are among the greatest success stories for achieving intracranial disease control with systemic agents, according to Ayal A. Aizer, MD, assistant professor of radiation oncology at Brigham and Women’s Hospital in Boston, Massachusetts. Among these drugs, there has been great improvement in a short amount of time.
“If you look at ALK-rearranged NSCLC, crizotinib [Xalkori] had some potential but lagged in terms of intracranial efficacy relative to its systemic efficacy,” Aizer said. “The newer agents, like alectinib [Alecensa], ceritinib [Zykadia], and brigatinib [Alunbrig], have shown a lot of promise.”
In a retrospective analysis of the PROFILE 1005 and 1007 trials, crizotinib versus chemotherapy was associated with systemic and intracranial disease control in patients with ALK-rearranged NSCLC and brain metastases. However, progression of intracra-nial lesions following initiation of crizotinib was observed in 20% of patients (51/253).3
Additionally, an analysis of single-patient findings from PROFILE 10014 illustrated crizotinib’s limited CNS activity. A low concentration of crizotinib was noted in the patient’s cerebrospinal fluid. “This was one of the first instances in the reported literature where we had a drug that has excellent systemic control in the body but, because of the blood—brain barrier, was not actually getting to the brain,” said Tejas Patil, MD, a hematology/oncology fellow at University of Colorado School of Medicine in Aurora. PROFILE 1001 began enrolling in 2008.
Thus, the development of second- and third-generation ALK tyrosine kinase inhibitors (TKIs) focused on increasing CNS penetrance and activity. The results of a randomized phase III trial5 showed that alectinib led to a greater rate of 12-month event-free survival (68.4% vs 48.7%) and fewer CNS progression events (12% vs 45%) than crizotinib in previously untreated patients with advanced ALK-rearranged NSCLC. Additionally, the results of a phase III trial6 of patients with metastatic ALK-rearranged NSCLC showed that 35% of the patients with brain metastases had an intracranial response with ceritinib (vs 5% who received chemotherapy).
Brigatinib, which was granted accelerated approval by the FDA in April 2017 for patients with ALK-rearranged NSCLC who had progressed on or were intolerant to crizotinib, has also demonstrated robust CNS activity that may make it an attractive option for firstline therapy. The phase III ALTA-1L trial7 findings showed that brigatinib versus crizotinib led to improvements in 12-month progression-free survival (67% vs 43%) and intracranial response (78% vs 29%) in patients with advanced ALK-positive NSCLC who were naïve to ALK-targeted therapy.
Additionally, preliminary results of a phase I/II trial8 showed that lorlatinib (Lorbrena), an ALK/ROS1 TKI, induced an intracranial response in the target lesions of 19 of 35 patients with ALK-rearranged or ROS1-positive NSCLC that had metastasized to the brain. Patil also recommended that the reporting of intracranial and extracranial overall response rates, as done in this trial and the ALTA-1L trial, should be continued in future investigations.
“Subgroup analyses show there was a progression-free survival benefit with brigatinib that was most pronounced in patients with brain metastases at baseline,” Patil said. “What that tells you is patients who had brain metastases at baseline were getting a drug that slowed progression in the brain.”In April 2015, osimertinib (Tagrisso), a third-generation EGFR TKI, was approved for patients with metastatic EGFR T790M-positive NSCLC who have disease progression during or after EGFR-TKI therapy. Pooled data from 2 phase II trials showed that patients with T790M-positive advanced NSCLC and CNS metastases had overall response and disease control rates in the CNS of 54% and 92%, respectively, in response to osimertinib.9 The median best percentage change from baseline in CNS target lesion size was 53%, and CNS response was observed regardless of prior RT to the brain.Although the brain is not a common site for first metastatic recurrence when considering all types of breast cancer, over half of patients with HER2-positive disease will develop brain metastases, according to Nancy U. Lin, MD, associate chief, Division of Breast Oncology, Susan F. Smith Center for Women’s Cancers and director of the Metastatic Breast Cancer Program at Dana-Farber Cancer Institute in Boston.
Tucatinib and neratinib (Nerlynx) are 2 HER2targeting TKIs that have shown potential CNS activity, Lin noted. The results of a phase IB study showed that tucatinib elicited a 42% brain-specific objective response in patients with heavily pretreated metastatic HER2-positive breast cancer.10 No systemic therapies are specifically approved by the FDA to treat brain metastases in breast cancer, but that could change. The ongoing phase II HER2CLIMB trial (NCT02614794) is investigating the addition of tucatinib to capecitabine (Xeloda) and trastuzumab (Herceptin) in patients with advanced HER2-positive breast cancer. This trial is the first registration trial in the breast cancer field to include patients with active brain metastases, according to Lin.
Additionally, data from a phase II trial presented at the 2017 American Society of Clinical Oncology (ASCO) Annual Meeting showed that neratinib induced a volumetric brain tumor response in 49% (18/37) of patients with HER2-positive breast cancer and brain metastases.11 Lin noted that the data were sufficiently promising for the National Comprehensive Cancer Network to include neratinib with capecitabine as a regimen to consider for patients with HER2-positive breast cancer and brain metastases. “That’s an unusual situation, because neratinib at the moment does not have a label for treatment of metastatic HER2-positive disease,” Lin said.
The ongoing phase III NALA trial (NCT01808573), which randomizes patients with metastatic HER2-positive breast cancer to receive neratinib plus capecitabine or lapatinib (Tykerb) plus capecitabine, is expected to further clarify the role of neratinib in treatment of brain metastases, although Lin pointed out that trial enrollment was restricted to patients with stable (not active) brain metastases only.
Partial trial findings released in mid-December demonstrate that the neratinib-containing combination led to an improvement in the time to intervention for brain metastases over the lapatinib-containing regimen (P = .043).12Up to 56% of patients with melanoma and brain metastases have a mutation in the BRAF gene,13 and presence of a BRAF mutation and disease in the CNS are independent predictors of shorter survival.14 The results of a phase I dose-escalation study15 showed that the BRAF V600E inhibitor dabrafenib (Tafinlar) reduced brain lesion size in 9 of 10 patients with metastatic melanoma and untreated brain metastases, and findings from the phase II COMBI-MB trial showed that dabrafenib plus trametinib (Mekinist), a MEK1/MEK2 inhibitor, elicited an intracranial response in 58% of patients with BRAF V600E—positive asymptomatic brain metastases with no prior local brain therapy.16 Vemurafenib (Zelboraf), another BRAF inhibitor, also produced intracranial responses of 33% and 23% in patients with untreated and treated brain metastases, respectively, in melanoma.17Although the intracranial responses from targeted agents for BRAF-mutated melanoma initially appeared promising, a pooled analysis of individual patient data from randomized trials showed that most patients present with disease recurrence within 12 months of starting dabrafenib plus trametinib, highlighting the need for more durable treatment options.18
“We know that targeted therapy…can induce up to a 58% response rate in the brain, which was a dramatic improvement over previous therapies,” said Hussein Tawbi, MD, director of melanoma clinical research and early drug development, Department of Melanoma Medical Oncology, The University of Texas MD Anderson Cancer Center in Houston. “However, what we found out from targeted therapy was that [those agents] work for a shorter duration than they do outside of the brain.” Tawbi led an open-label phase II study19 of ipilimumab (Yervoy) plus nivolumab (Opdivo) followed by nivolumab until disease progression or unacceptable toxicity in patients with metastatic melanoma and at least 1 measurable nonirradiated brain metastasis. At the median follow-up of 14 months, the results showed an intracranial clinical benefit rate of 57% with ipilimumab plus nivolumab, but the most remarkable finding was the durability of response, according to Tawbi. “More than 90% [of patients] who had a response continued to respond, which was really fascinating,” he said. “We were inducing these responses in the brain at a high rate, and patients didn’t need further radiation or surgery.”
Although the combination regimen led to treatment-related grade 3 or 4 toxicities in 55% of patients, those who experienced toxicity from the combination were able to safely receive single-agent immunotherapy with nivolumab. Investigating less-toxic immunotherapy regimens that have similar or better efficacy will be a top priority moving forward, Tawbi noted.
Although pembrolizumab (Keytruda) recently became part of the standard of care for first-line metastatic NSCLC, the early data on intracranial activity in NSCLC appear less promising than the data in melanoma. An open-label phase II trial20 results showed that pembrolizumab led to intracranial responses in 4 of 18 patients with melanoma and 6 of 18 patients with PD-L1—positive NSCLC and untreated brain metastases, and Patil stated that the role of immunotherapy in reduction of brain metastases remains an open question. “The trial we need is chemotherapy plus immunotherapy versus immunotherapy in patients with brain metastases, and I don’t think that trial exists,” said Patil. “That would help answer the [question of] the role that immunotherapy is having in reduction of brain metastases.”
Similarly, Lin stated that the CNS activity of immunotherapy in brain metastases associated with breast cancer is unknown largely due to the exclusion of patients with active brain metastases from the majority of clinical trials. She and her colleagues have ongoing trials investigating atezolizumab (Tecentriq) plus stereotactic radiosurgery in patients with triple-negative breast cancer that has metastasized to the brain and HER2-directed therapy plus immunotherapy in patients with HER2-positive breast cancer with brain metastases. “The jury is still out as to whether checkpoint inhibitors will have activity in [brain metastases in] breast cancer and what the risk/benefit profile will look like,” Lin said.
Furthermore, how immune checkpoint inhibitors elicit an intracranial response is still an ongoing area of investigation, especially in light of the traditional thought that the CNS is an immunoprivileged zone and CNS responses to immune checkpoint inhibitors vary across tumor types. CNS response rates to checkpoint inhibitor therapy are higher for intracranial metastases from melanoma than for NSCLC. Ongoing investigations are studying possible mechanisms of action, such as whether the PD-1/PD-L1 and CTLA-4 antibodies can cross the BBB despite their relatively large size or if immune modulation induced by these agents allows cytotoxic T cells into the brain tumor microenvironment.21
“We don’t know if T cells are activated outside the brain and then go in and attack the tumor or if the drugs actually get in there and activate the T cells that are sitting in the tumor,” Tawbi said. “We are in the process of answering that question of where the T cells get activated and how they exert their effect. But it seems to be an immune response because it has all of the characteristics of an immune response.”The increasing trial inclusion of patients with brain metastases has been a key factor in improving patient care and the discovery of CNS-penetrating agents, and inclusion criteria should be progressively expanded to include patients with active and symptomatic brain metastasis when safe and appropriate, Aizer said.
“The more we can be flexible with allowing patients with modest intracranial disease burdens to be on these studies and allow them to receive salvage therapy while continuing the study agent, the better we’re going to do for patients and the better designed our trials will be,” Aizer said. In response to a request from oncologists and patient advocates, the National Cancer Institute recently revised its protocol for trial accrual to include some patients with brain metastases.
Lee agreed that agents with CNS activity still represent a major unmet need in the cancer community, but she hopes that the recent success seen in ALK-targeted therapies will encourage inclusion of patients with brain metastases in clinical trials for cancers in which brain metastases are prevalent. She and Lin were part of the Response Assessment of Neuro-Oncology working group that submitted a proposal published in Lancet Oncology that described strategies for developing inclusion criteria for patients with brain metastases in clinical trials.22 Lin was also a member of the ASCO and Friends of Cancer Research eligibility working group that drafted guidance documents that outlined approaches for the inclusion of patients with brain metastases in all phases of clinical trials.23
“Our hope from these 2 initiatives is that [they] will spur on more active inclusion of patients with brain metastases in clinical trials where the patient population is expected to have a high prevalence of CNS involvement,” said Lin. “These efforts have been going on over the past year, and I think we’ll have to see over time what sort of impact the recommendations have.”In summary, impressive results have been achieved so far with targeted and immunotherapies (Table 25-7, 9-11, 15-18, 20-21). Moving forward, making the treatment of brain metastases a top priority in cancers with a high incidence of brain metastases will be important for developing agents with CNS penetration, according to Lee. “Looking at ALK-positive lung cancer, we’re seeing a lot of success in that field with respect to [treatment of] brain metastases.” The reason why is that investigators have designed trials that included a focus on brain metastases, Lee said.
Tawbi added that basic research that uses brain metastases models for different cancers also will help optimize development of agents with CNS and tumor penetrance. “One of the big [regrets] that we have right now is that there aren’t very good brain metastases models that will help us study the interaction between the brain and tumor microenvironment,” Tawbi said. “We have great basic science researchers [at The University of Texas MD Anderson Cancer Center] who have published some great stuff in this setting, and we really need to develop [models] to be able to understand what the next target should be.”