Brain Metastases: Evolving Concepts at ASCO 2016

Recent research on brain metastasis (BM) from two disparate primary malignancies (non-small cell lung cancer [NSCLC] and gastroesophageal cancer) may impact patient care in each case. Abstracts presented by Cleveland Clinic researchers at the 2016 American Society of Clinical Oncology (ASCO) Annual Meeting in Chicago communicate related findings.

“Brain metastases is a significant and serious complication of systemic cancers ‒ 10 times more common than primary malignant brain tumors, so this is an important clinical problem we see in clinic every day,” says Manmeet Ahluwalia, MD, Director of Cleveland Clinic’s Brain Metastasis Research Program and a study author.

Non-small-cell lung cancer BM

“Targeted therapies have really changed the outcome of a select group of patients with lung cancer, and this research evaluated the impact of epidermal growth factor receptor (EGFR) and anaplastic lymphoma kinase (ALK) mutations in outcomes of patients with brain metastases,” says Dr. Ahluwalia. Activating mutations in EGFR and ALK constitute 10 to 15 percent and 5 percent of NSCLC, respectively.

Historically, the number of BMs has been considered a key factor. And, in wild-type NSCLC, patients who had one BM generally did better than those who had two-to-three, or four or more. However, emerging evidence suggests that, in some ways, BM in patients with mutations in EGFR and ALK is different.

“We found in our cohort that the number of brain metastases did not impact the overall survival rate for NSCLC patients with activating mutations in the EGFR gene and the ALK gene,” says Dr. Ahluwalia, describing the Cleveland Clinic study. This is contrary to the earlier knowledge where the number of brain metastases significantly impacts survival.

With IRB approval, Cleveland Clinic’s Rose Ella Burkhardt Brain Tumor and Neuro-Oncology Center database was used to identify patients treated between 2000 and 2014 for NSCLC BM with mutational status (EGFR/ALK/wild).

In the updated analysis for the meeting, a total of 1,078 NSCLC BM patients treated in that time period were identified. Among 348 with mutational analysis, 91 were positive for mutation (ALK = 23; EGFR = 68), while 257 were wild type. Median age at diagnosis was 60.0 (29.8, 82.6) and the Karnofsky performance scale (KPS) level was >80 in 50 percent (44 patients). Median number of BM was two (range 1, 99).

Highlights from the results include:

• Median overall survival (OS) for ALK and EGFR + NSCLC BM was 19.93 months vs. 9.87 months for wild type (p .001).
• The number of BMs in the mutation-positive group didn’t significantly impact OS (BM 1 with 1 months vs. 2-3 with 19.1 months and >3 with 23.7 months, p .74), whereas fewer metastases in wild type had significantly better OS on multivariate analysis excluding extracranial metastases (BM=1 with 13.8 months, 2-3 with 11.0 months and >3 with 8.1 months, p .006).

“This analysis showed that in patients with EGFR- and ALK-positive lung cancer, the number of brain metastases did not impact survival,” Dr. Ahluwalia says. “The analysis also highlighted that the outcomes for patients with brain metastases with EGFR-positive and ALK-positive mutational status are significantly better than for patients who do not have these mutations.”

Recent NCCN guidelines on NSCLC have been updated to reflect advancing knowledge with respect to EGFR-positive and ALK-positive disease.

Gastroesophageal cancer with brain metastases (GECBM)

Dr. Ahluwalia and his co-authors also presented a second abstract on prognostic factors and outcomes in patients with BMs from gastroesophageal cancer (GECBM). Although BM occurs in <4 percent of all tumors of the GI tract, the incidence of BM from gastrointestinal cancers may be rising, in part due to more effective systemic treatments and prolonged survival.

Not much is known about the outcomes of these patients, according to Dr. Ahluwalia. The Disease Specific Graded Prognostic Assessment (DS-GPA) ‒ which is based solely on the KPS ‒ is a commonly used prognostic index in patients with BM.

Cleveland Clinic’s Burkhardt Brain Tumor Center database was used to identify 63 GECBM patients who were treated between 2002 and 2014.

The main highlight from the study was that, on multivariate analysis, the number of BM and the presence of symptoms were independent predictors of OS, in addition to KPS.

Proposal for revising DS-GPA:

A revised DS-GPA for GECBM based on KPS, number of BMs and symptoms is proposed. The revised DS-GPA assigns an increasing number of points to categories representing fewer metastases, a higher KPS and the absence of symptoms.

Combining the factors and summing the points, three groups are defined:

• Unfavorable (total points <10); median OS 2.5 months
• Intermediate (total points 11-13); median OS 5.0 months
• Favorable (total points >13); median OS 8.8 months

“This analysis of this research shows that there are a number of other factors which impact the outcomes of these patients that include a number of brain metastases, presence of lung or liver metastases and performance status,” says Dr. Ahluwalia.

The abstracts detailing both of these studies were featured at the 2016 American Society of Clinical Oncology (ASCO) Annual Meeting June 3-7.

Dr. Ahluwalia is an Associate Professor in the Department of Medicine, Clinic Lerner College of Medicine of Case Western Reserve University (CCLCM) where he subspecializes in treatment of patients with brain tumors and brain metastases. He is the Director, Brain Metastasis Research Program and the Associate Director, Clinical Trials, Operations in the Rose Ella Burkhardt Brain Tumor and Neuro-Oncology Center of the Neurological Institute of Cleveland Clinic. He is the Section Head of Neuro-Oncology Outcomes and is a staff in Cleveland Clinic Cancer Center and has joint appointment in the Developmental Therapeutics Program, Case Comprehensive Cancer Center.