Larger trials needed to confirm benefit of targeted therapies, immunotherapy for brain metastases
Brain metastases are a common complication of a variety of cancer types and a major cause of morbidity and mortality.
As patients with cancer are living longer, and as detection methods improve, the incidence of brain metastases is increasing. Ten percent to 20% of adults with cancer will develop brain metastases, which most frequently originate from lung or breast cancer, or melanoma. Up to 65% of patients with lung cancer will develop brain metastases.
Brain metastases can differ molecularly from the primary tumor, making them especially challenging to treat.
“Sometimes when a patient has brain metastases and a previous history of lung cancer, they can acquire new genetic alterations in the brain than what was in the lungs,” Manmeet S. Ahluwalia, MD, director of the Brain Metastases Research Program at Cleveland Clinic and a HemOnc Today Editorial Board Member, said in an interview. “It’s a highly complex disease.”
Due to these complexities, brain metastases historically have been associated with an extremely poor prognosis.
Surgical and radiation techniques are used to subdue tumor growth but have been linked to long-term side effects and are often ineffective against multiple lesions. Many chemotherapy agents cannot cross the blood-brain barrier, limiting their effectiveness.
However, studies presented last year at the ASCO Annual Meeting showed systemic treatment with immunotherapy and targeted agents may redefine the treatment of brain metastases.
HemOnc Today spoke with radiation oncologists and neuro-oncologists about the changing standard of care for brain metastases; how new data impact the role of radiotherapy; and how immunotherapy and targeted therapies may change the prognosis for these patients.
Standard of care
Any type of cancer can metastasize to the brain, causing additional complications and treatment challenges.
Surgery is the preferred treatment for a single brain metastasis and for those larger than 4 cm. However, surgery is invasive, has no impact on other lesions and is usually followed by radiotherapy.
“Surgery is generally reserved for diagnostic purposes and relief of lesions that can’t be otherwise controlled, such as for bleeding or massive pressure, as well as to treat tumor regrowth or stereotactic radiation therapy radionecrosis,” Kim A. Margolin, MD, clinical professor in the department of medical oncology and therapeutics research at City of Hope, told HemOnc Today.
The National Comprehensive Cancer Network recommends stereotactic radiosurgery (SRS) — a nonsurgical technique that spares healthy tissue — for patients with one to three or four metastases, and whole-brain radiotherapy (WBRT) — which delivers radiation to the entire brain, even healthy tissue — for patients with advanced disease and more metastases.
Many oncologists try to limit the use of WBRT because of the risk for severe long-term cognitive decline. Radiation therapy also can lead to radionecrosis, or the death of soft tissues.
In other instances — particularly melanoma — WBRT has minimal antitumor activity and is only considered when other approaches fail.
In an Alliance clinical trial, Brown and colleagues found the addition of adjuvant WBRT to SRS improved intracranial tumor control at 6 months (88.3% vs. 66.1%) and 12 months (84.9% vs. 50.5%; P < .001 for both) among 213 patients with one to three brain metastases. However, this did not translate to improved median OS (7.5 months vs. 10.7 months).
Further, cognitive decline at 3 months appeared more frequent after the combination (88% vs. 61.9%; P = .002).
Despite its limitations, radiation therapy remains the mainstay for patients with brain metastases who are unable to undergo surgery.
“There definitely is a role and benefit for radiation therapy to the brain,” Mark M. Awad, MD, PhD, assistant professor of medicine at Harvard Medical School, said in an interview. “It can help many patients, particularly in situations where you can focus radiation therapy in a smaller number of spots.”
Both SRS and stereotactic radiation therapy (SRT) — another option for precise radiotherapy — are used for localized tumors. SRS delivers a large dose of radiation on a single day, whereas SRT spans multiple days and may be more appropriate for difficult-to-reach lesions.
Jimenez and colleagues evaluated SRT among 156 patients with brain metastases from solid tumors. Results showed a 36% rate of local progression at the index lesion at 24 months and modest incidence of adverse events (23.7%), the most common of which were fatigue and headache. Only three patients experienced grade 1 cognitive impairment.
Focal therapy alone leaves the rest of the brain uncontrolled, thus still requiring systemic treatment.
“With new agents, which work better systemically, the question now is whether they can work in the brain, as well,” Stuart H. Burri, MD, radiation oncologist at Levine Cancer Institute at Carolinas HealthCare System, said in an interview. “We don’t want to stop using the proven effective methods for brain control just because some patients do OK [on new therapy] without stronger evidence from large prospective controlled trials and evidence that nonresponders — currently around half or more of patients — are not harmed by withholding standard-of-care radiation or surgery. The goal is to find effective systemic therapy to lower the need for radiation, specifically WBRT.”
Durable response from immunotherapy
The effectiveness of immunotherapy among patients with brain metastases had been largely unknown due to their exclusion from clinical trials.
Further, even when a metastatic cancer shows benefit from systemic treatment, patients can still experience progressive disease in the brain or new metastases — known as escape metastases — highlighting the need for more therapeutic options.
The ASCO Annual Meeting showcased multiple abstracts that suggested immunotherapies could be an option.
In the Anti-PD-1 Brain Collaboration (ABC) clinical study, Long and colleagues showed nivolumab (Opdivo, Bristol-Myers Squibb) alone or with ipilimumab (Yervoy, Bristol-Myers Squibb) induced considerable antitumor activity and objective responses among patients with melanoma brain metastases. Treatment appeared safe and without central nervous system-specific adverse events.
Rate of intracranial response appeared highest among patients treated with the combination (42%), followed by patients who received nivolumab alone (previously untreated, 20%; treated, 6%).
Treatment-related grade 3 and grade 4 toxicities occurred more frequently among treatment-naive patients who received combination therapy (96%) or nivolumab alone (68%) than among previously treated patients (56%).
“The responses seen with immunotherapy are durable,” Ahluwalia said. “Benefit can last many months to years, something we do not usually see with targeted therapy.”
In the CheckMate 204 study, Tawbi and colleagues also evaluated nivolumab and ipilimumab among patients with brain metastases from melanoma. The intracranial overall response rate was 56%.
Treatment-related grade 3 and grade 4 toxicities occurred among 48% of patients, 8% of which were neurologic.
“The nivolumab and ipilimumab combination produces better response rates than the rates we’ve seen with either drug used as single agents,” which are around 20% with nivolumab and 10% with ipilimumab, Ahluwalia said. “With the combination approach, we see the benefits are synergistic.”
Awad noted that, although not every patient in the CheckMate 204 and ABC trials had responses in the brain, the responses were durable.
“In some cases, you can use immunotherapy and avoid radiation therapy and still see responses in the brain,” he said.
Despite their promise, immunotherapies do not benefit all cancers. For example, immunotherapy has not demonstrated good responses among patients with HER-2-positive breast cancer.
Clinical trials have excluded patients who require steroids to control neurological symptoms because they are unlikely to respond to immunotherapies.
“Immunotherapies generally consist of large molecules that don’t themselves get into the brain, but stimulate T lymphocytes that are capable of entering the brain and brain metastases,” Margolin said. “However, T cells are suppressed by steroids, which are often used to control edema from brain metastases.”
To further improve outcomes, researchers also are assessing a combination of radiotherapy and immunotherapy.
Ahmed and colleagues showed SRT was well tolerated among 36 patients with advanced melanoma irradiated within 6 months of receiving nivolumab.
Local control following radiation reached 91% at 6 months and 85% at 12 months. Median OS was 12 months from nivolumab initiation.
No treatment-related neurologic toxicities or scalp reactions occurred, with the exception of grade 2 headaches.
However, 11% of patients experienced local brain metastasis failures, with at least a 20% increase in volume at previously irradiated areas.
“There is evidence that the combination of immunotherapy and radiation may have a greater effect,” Burri said. “There should be immunotherapy trials done that include radiation arms to evaluate as another comparator.”
Overall, more data are needed, “preferably from trials such as the melanoma ABC and CheckMate trials, in which patients are randomly assigned to receive a backbone of immunotherapy for systemic brain metastases with or without SRS — using the latter as salvage for escape metastases,” Margolin said.
The impact of immunotherapy on radionecrosis incidence remains unclear.
“These are very preliminary data in abstract form only showing modest response rates to immunotherapy, which are far superior to historic levels but appear lower than standard-of-care radiation,” Burri said. “We need more data before we start treating patients with immunotherapies and other systemic therapies.”
Because the agents are promising, patients with asymptomatic brain metastases should be treated on prospective controlled randomized trials with a standard-of-care control arm, Burri added.
“We are focusing on the potential advantages in the responders, but we also need to be aware of causing harm to nonresponders by delaying current standard-of-care therapies,” he said.
Targeted therapies have shown impressive brain responses, Awad said, although many agents do not cross an intact blood-brain barrier.
For instance, in the COMBI-MB trial, Michael A. Davies, MD, PhD, associate professor and deputy chairman of the department of melanoma medical oncology at The University of Texas MD Anderson Cancer Center, and colleagues evaluated whether dabrafenib (Tafinlar, Novartis) plus trametinib (Mekinist, Novartis) — FDA approved for patients with BRAF V600-mutated melanoma — also benefited disease spread to the brain.
The analysis included 125 patients with brain metastases who received no more than two prior metastatic melanoma systemic treatments.
Results, presented at ASCO, showed intracranial response rates ranged from 44% to 59% across study cohorts, nearly as high as ORRs from these agents for patients without brain metastases.
Median duration of response in cohort A — the largest group (n = 76), all of whom had no prior brain-directed therapy — was 6.5 months (95% CI, 4.9-10.3), which was generally shorter than that observed with dabrafenib and trametinib among patients without brain metastases.
“There was good initial activity; however, as we followed them, we saw the majority of responses were relatively short-lived,” Davies said. “The results highlight the need to develop new strategies to improve durability of responses we achieve with these patients.”
Thirty-five percent of patients experienced a serious adverse event. Common serious treatment-related adverse events included pyrexia for dabrafenib (6%) and decreased cardiac ejection fraction (4%).
“Results showed treating patients with brain metastases [with this combination] was safe,” Davies said. “We didn’t see any unexpected toxicities or those different from other trials in patients without brain metastases.”
“On a practical note, MAPK inhibitor therapy has a rapid onset of action and can often be started with minimal delay following the diagnosis of brain metastases, [whereas] treatment planning for SRS can sometimes delay the initiation of therapy by a week or more,” Margolin said.
However, the lack of a control arm limits the general applicability of the results.
“The trial was well designed and well run, but there was no conventional control arm, only an experimental arm, which is a problem,” Burri said. “Further, the response rates are, again, not great.”
In the AURA3 phase 3 trial, Mok and colleagues reported osimertinib (Tagrisso, AstraZeneca), an EGFR inhibitor, appeared superior to chemotherapy among 419 patients with advanced NSCLC, asymptomatic brain metastases and an EGFR T790M mutation.
Among evaluable patients with at least one measurable CNS metastasis, the CNS ORR reached 70% among patients treated with osimertinib compared with 31% among patients treated with chemotherapy (OR = 5.13; 95% CI, 1.44-20.64).
“Targeted therapies are working very well in the brain, as well as in the rest of the body, so you can give the same drug to treat the brain metastases as the systemic disease,” Ahluwalia said.
Targeted therapies also are showing activity for metastatic breast cancer.
“Treatment choices are tailor-made based on the genetic mutations breast cancer patients’ brain metastases may harbor, as we are living in the era of precision medicine,” Ahluwalia said.
Results from the phase 2 TBCRC 022 clinical trial — also presented at ASCO — showed a combination of neratinib (Nerlynx, Puma Biotechnology) and capecitabine induced intracranial activity among 39 patients with HER-2-amplified disease and brain metastases.
Eighteen patients (49%; 95% CI, 32-66) had a volumetric ORR, indicating at least a 50% reduction in all target CNS lesions. Overall, 12-month survival was 63% (95% CI, 43-77).
Eighteen patients (49%) experienced grade 3 toxicity, whereas no patients experienced grade 4 toxicity.
The study is expected to complete in December.
“These are the best response rates we’ve seen in HER-2-positive breast cancer patients with brain metastases,” Ahluwalia said.
Consequences of targeted therapy
The efficacy of targeted therapy has revealed additional research questions.
For instance, researchers are assessing whether a targeted approach can prevent the development of brain metastases.
In the ALEX study — presented at ASCO — Shaw and colleagues randomly assigned 303 patients with stage IIIB or stage IV ALK-positive NSCLC to 600 mg alectinib (Alecensa, Genentech; n = 152) — an oral, small molecule, ATP-competitive tyrosine kinase inhibitor of ALK — or 250 mg crizotinib (Xalkori, Pfizer; n = 151) twice daily.
Alectinib appeared associated with an 84% reduction in the risk for CNS progression (HR = 0.16; 95% CI, 0.1-0.28). At 1 year, brain metastases occurred among 9% of the alectinib-treated group compared with 41% of the crizotinib-treated group.
These improvements translated into a 53% reduced risk for progression and mortality (HR = 0.47; 95% CI, 0.34-0.65).
Still, targeted therapies will only benefit patients whose tumors depend on that target and do not rapidly mutate to lose this dependence, Margolin said.
The blood-brain barrier poses another challenge.
“Targeted therapies generally do not cross the blood-brain barrier but, at least in melanoma, have been associated with major responses when the tumor is not already resistant to systemic treatment,” Margolin said. “This is presumably due to partial permeability of the blood-brain barrier in the presence of macroscopic tumor metastases.”
Because targeted therapies can effectively control cancer in the body and shrink tumors in the brain, the eventual hope is to avoid or delay radiation.
“There is a need to develop improved therapies targeting cancer everywhere in the body, but particularly the brain,” Awad said.
Still, the data as they stand may not be enough to change the role of radiation.
“My only concern is the interpretation of these preliminary clinical trials,” Burri said. “Are oncologists going to start thinking we no longer have to do other treatments, such as radiation? It is far too premature to think these limited trials — almost all with less than 100 patients and mostly presented only in abstract form — have changed the standard of care.”
Participation in clinical trials
Although these studies show promise, the data are limited by the small cohorts.
More clinical trials with larger numbers of patients with brain metastases, as well as study designs that allow comparisons among best-available regimens, are needed to enhance existing data.
Historically, patients with brain metastases have been excluded from clinical trials due to concerns for worse outcomes or different potential side effects than patients without brain metastases.
“There has been essentially no regulation of how to design, conduct and report trials for patients with brain metastases on systemic therapy,” Margolin said. “But, trials for these patients are becoming more important, with newer agents, methods of imaging, combinations with local therapies and overall better outcomes.”
Davies said it is “very frustrating” to tell a patient they can’t participate in a clinical trial because they have brain metastases. However, as more clinical trials of immunotherapies and targeted therapies demonstrate positive results, randomized trials are more likely.
“It is now an open question for the field with these multiple treatment options — each has strengths and weaknesses — about whether we are at the point to have randomized trials for patients with brain metastases,” Davies said. “Hopefully, in addition to moving in a direction of more trials for these patients, there will be a change in the paradigm, so these patients aren’t excluded from large clinical studies.”
The Response Assessment in Neuro-Oncology-Brain Metastases working group developed guidelines to improve clinical trial design to include patients with brain metastases. The guidelines detailed frequent limitations patients with brain metastases may experience in clinical trials or for enrollment and offered recommendations for overcoming these limitations.
The group recommended three strategies for enrollment based on a drug’s activity in the brain:
- For trials of systemic therapy with limited brain activity, patients with stable CNS disease — including those who have had SRS and are not steroid dependent for control of symptoms or edema — should be permitted and patients with active CNS disease should be excluded;
- For trials of drugs with initial evidence of brain activity, patients with stable and active CNS disease should be permitted to compare the drug’s activity in the brain compared with the rest of the body; and
- Early trials of drugs with unclear activity in the brain should include a cohort of patients with brain metastases to determine whether the drug has benefit and to better select one of the other two strategies for later trials.
“The careful management of brain metastases with newer approaches will allow patients to participate in more treatment options on or off trial than ever before,” said Margolin, who was on the panel and helped write the guidelines. “This can only improve the outcomes and our knowledge of the most accurate and current data. Patients with brain metastases won’t be disregarded as being beyond all hope.”
A greater understanding of brain metastasis is needed before immunotherapy and targeted therapies become standard of care.
“Brain metastases continue to be a major problem limiting the lifespan and the quality of life of many patients,” Margolin said. “Essential to their control is an understanding of how they originate and differ from other metastases, as well as the unique anatomic and biologic requirements for their control.” – by Melinda Stevens
Click here to read the , “Does WBRT still have a role for limited brain metastases?”
Ahmed KA, et al. Ann Oncol. 2016;doi:10.1093/annonc/mdv622.
Brown PD, et al. JAMA. 2016;doi:10.1001/jama.2016.9839.
Camidge DR, et al. Lancet Oncol. 2017;doi:10.1016/S1470-2045(17)30693-9.
Davies MA, et al. Lancet Oncol. 2017;doi:10.1016/S1470-2045(17)30429-1.
Goldberg SB, et al. Lancet Oncol. 2016;doi: 10.1016/S1470-2045(16)30053-5.
Jimenez RB, et al. Adv Radiat Oncol. 2018;doi:10.1016/j.adro.2017.05.008.
Kim, JM, et al. J Neurooncol. 2017;doi:10.1007/s11060-017-2442-8.
Mok T, et al. J Clin Oncol. 2017;doi:10.1200/JCO.2017.35.15_suppl.9005.
Zindler JD, et al. BMC Cancer. 2017;doi:10.1186/s12885-017-3494-z.
The following were presented at ASCO Annual Meeting; June 2-6, 2017; Chicago:
Davies MA, et al. Abstract 9506.
Freedman RA, et al. Abstract 1005.
Long GV, et al. Abstract 9508.
Shaw AT, et al. Abstract LBA9008.
Tawbi HA-H, et al. Abstract 9507.
For more information:
Manmeet S. Ahluwalia, MD, can be reached at firstname.lastname@example.org.
Mark M. Awad, MD, PhD, can be reached at email@example.com.
Stuart H. Burri, MD, can be reached at firstname.lastname@example.org.
Michael A. Davies, MD, PhD, can be reached at email@example.com.
Kim A. Margolin, MD, can be reached at firstname.lastname@example.org.
Disclosures: Ahluwalia reports honoraria from Elsevier, a consultant or advisory role with Monteris Medical and research funding from Novartis. Davies reports consultant or advisory roles with Genentech/Roche, GlaxoSmithKline, Novartis, Sanofi and Vaccinex; and research funding to his institution from AstraZeneca, Genentech/Roche, GlaxoSmithKline, Merck, Myriad Genetics, Oncothyreon and Sanofi. Margolin reports consultant or advisory roles with ImaginAb, Lion Biotechnologies and Pfizer; and research funding from Altor Bioscience and Merck. Burri reports no relevant financial disclosures.