The number of brain metastases does not have an impact on overall survival (OS) in patients with non-small cell lung cancer (NSCLC) with an EGFR mutation or ALK rearrangement, Cleveland Clinic investigators discovered.
Suresh Balasubramanian, MD
The number of brain metastases does not have an impact on overall survival (OS) in patients with nonsmall cell lung cancer (NSCLC) with an EGFR mutation or ALK rearrangement, Cleveland Clinic investigators discovered through examination of their neuro-oncology database.
The finding contrasts to wild-type NSCLC, in which fewer brain metastases was associated with significantly better OS, said Suresh Balasubramanian, MD, at the 2016 ASCO Annual Meeting.
Activating EGFR mutations in NSCLC are present in 10% of US patients and 35% of patients in East Asia, and approximately 5% of NSCLC cases harbor ALK rearrangements.
Multiple brain metastases confer worse outcomes than single brain metastasis in NSCLC patients treated with stereotactic radiation therapy without whole brain radiation. A larger number of brain metastases can now be treated with radiosurgery alone without resorting to whole brain radiation, said Dr. Balasubramanian, neuro-oncology fellow at Cleveland Clinic.
In addition, newer chemotherapies and targeted treatments can achieve higher concentrations in the central nervous system than older agents, potentially improving outcomes in mutated NSCLC with brain metastases.
The Cleveland Clinic Neuro-oncology Center’s database was used to analyze variables that could affect the prognosis of NSCLC with brain metastases. The investigators identified 348 NSCLC patients with known mutational status who were treated between 2000 to 2014, with OS from the diagnosis of bone metastasis being the primary outcome of interest. Of the 348 patients, 257 had wild-type disease and 91 were mutation positive. There was no significant difference between the groups in the number of brain metastases, but those with mutation-positive NSCLC more often had extracranial manifestations.
A similar proportion of those with mutation positive and wild-type NSCLC received stereotactic radiation with or without surgery (33.3% vs. 39%) as first-line treatment. More patients in the mutation-positive group compared with wild type received a combination of whole brain radiation and stereotactic radiation (16% vs. 6.4%) as first line. About half (49.5%) of the mutation-positive group received first-line targeted therapy compared with none of the wild-type patients. In the mutation-positive patients, 65.2% with ALK rearrangement and 72.7% with EGFR mutation received targeted therapy after a diagnosis of bone metastasis, either as adjuvant therapy or after progression of disease.
Univariate analysis showed that the number of brain metastases had no impact on OS or progression-free survival (PFS) in the mutation-positive group, but in the wild-type group, multiple brain metastases predicted worse OS and PFS. The presence of extracranial metastases in the wild-type group had an unfavorable impact on OS in the wild-type group, leading the investigators to design 2 multivariate models: excluding extracranial metastases and a separate model including extracranial metastases.
On multivariate analysis in patients without extracranial metastases, “we have similar results as what we saw in the univariate analysis; that is, the number of brain metastases did not impact the mutation-positive cohort whereas it did in the wild-type group,” said Dr. Balasubramanian.
Multivariate analysis without extracranial metastases showed no significant impact of brain metastases on OS in the mutation-positive group (HR 0.95, P=0.74) but a significant impact in the wild-type group (HR 1.31, P=0.006). In the mutation-positive group, there was no OS difference between patients with 1, 2 to 3, and >3 brain metastases. In the wild-type group, however, OS was worse in patients with ≥2 brain metastases compared with 1 brain metastasis. Estimated median OS in the wild-type group was 13.8 months with 1 brain metastasis, which was significantly better (P= 0.007) than the 11-month median OS with 2 to 3 brain metastases and 8.1-month median OS with >3 brain metastases.
There was no statistically significant difference in OS in the wild-type group with 2 to 3 brain metastases compared with the wild-type group with >3 brain metastases.
Multivariate analysis including extracranial metastases again showed an impact of brain metastases on OS only in the wild-type group. However, the impact of the number of brain metastases on OS when including extracranial metastases was less than significant (P= 0.063) than in multivariate analysis without extracranial metastases.
The modest impact of brain metastases on outcomes in mutated NSCLC, “is in contrast to the radiation scenario, where the number of brain metastases is a key determinant of what therapy may be best. In patients managed with systemic therapy, number of brain metastases may be less important,” said invited discussant Ayal Aizer, MD, radiation oncologist, Brigham and Women’s Hospital, Boston.
The blood-brain barrier limits the penetration of conventional chemotherapy but modern targeted therapies offer more potential for intracranial control in mutated/rearranged lung cancer, breast cancer, melanoma, and others, he said.
Systemic therapy for mutated/rearranged NSCLC is a viable alternative to radiation if the patient is asymptomatic, close magnetic resonance imaging surveillance is employed, “and if the radiation option would otherwise be whole brain radiation,” said Dr. Aizer.
Balasubramanian SK, Venur VA, Chao ST, et al. Impact of EGFR and ALK mutation on the outcomes of non-small cell lung cancer (NSCLC) patients with brain metastases. Presented at: ASCO 2016; June 3-7, 2016; Chicago, IL. Abstract 2005.