The following article features coverage from the 2021 American Society of Clinical Oncology (ASCO) Annual Meeting. Click here to read more of Oncology Nurse Advisor‘s conference coverage.

A model for predicting the risk for brain metastases developing in patients with non-small cell lung cancer (NSCLC) was described in a poster presentation at the 2021 American Society of Clinical Oncology (ASCO) Annual Meeting.

Thirty percent to 50% of patients with NSCLC will develop brain metastases, reported Protiva Rahman, PhD, of Vanderbilt University Medical Center in Nashville, Tennessee, and colleagues. Median survival for these patients is less than 1 year following the diagnosis of brain metastasis. The researchers aimed to devise a method to predict patients’ probability of developing brain metastasis.

Patients with NSCLC Stage II-IV were identified from the American Association for Cancer Research Project GENIE Biopharma Consortium dataset. The researchers used demographic and anticancer treatment data, as well as features from next-generation sequencing data for these patients, and tested 5 types of machine learning models for predicting brain metastasis. Model development was performed with a training dataset, and a test dataset was used for analyzing model performance.

Continue Reading

A total of 956 patients were included in the full dataset, with data from 20% of these patients analyzed in the test dataset. Univariate analysis identified several features that appeared linked to brain metastasis, including etoposide therapy, mutations in EGFR and TP53 genes, amplifications of EGFR and ERBB2 genes, and deletions involving CDKN2A, CDKN2B, and RB1 genes. In this analysis, decreased risk of brain metastasis was reportedly associated with mutations in KRAS and NOTCH1 genes and treatment with nivolumab, alectinib, atezolizumab, pembrolizumab, gemcitabine, or vinorelbine.

The statistical modeling was able to predict brain metastasis with an area-under-the-curve of 0.71 (±0.07). Performance across the models was similar.

“Our models have reasonable prediction performance,” stated Dr Rahman in a presentation accompanying the research team’s poster. She also noted that competing risk of death was not included in these models, but the team is updating models to include this.

Disclosures: This research was supported by American Association for Cancer Research GENIE BioPharma Collaborative. Some study authors declared affiliations with biotech, pharmaceutical, and/or device companies. Please see the original reference for a full list of disclosures.

Read more of Oncology Nurse Advisor’s coverage of the 2021 ASCO Annual Meeting by visiting the conference page.


Rahman P, LeNoue-Newton M, Chaugai S, et al. Clinical and genomic predictors of brain metastases (BM) in non-small cell lung cancer (NSCLC): an AACR Project GENIE analysis. J Clin Oncol. 2021;39(suppl 15; abstr 2032). doi:10.1200/JCO.2021.39.15_suppl.2032