Conference Coverage

Liquid biopsy falls short for isolated brain lesions in lung cancer



– Liquid biopsy appears inadequate to detect molecular aberrations in patients with non–small cell lung cancer (NSCLC) who have isolated central nervous system (CNS) progression, according to investigators.

Dr. Mihaela Aldea of Gustave Roussy Institute in Villejuif, France Will Pass/MDedge News

Dr. Mihaela Aldea

Plasma circulating tumor DNA (ctDNA) analysis detected molecular abnormalities in almost all patients with systemic disease progression, compared with just two out of five patients with isolated brain lesions, reported lead author Mihaela Aldea, MD, who presented findings at the European Lung Cancer Conference.

Dr. Aldea, of Gustave Roussy Institute in Villejuif, France, said that “central nervous system progression is an example of hard-to-biopsy disease and is common in oncogene addicted non–small cell lung cancer, making it a potential setting to employ ctDNA analysis.” However, Dr. Aldea noted that the blood-brain barrier limits passage of molecules such as ctDNA into systemic circulation, leading to hypothetical skepticism within the medical community, despite “very limited” data.

“Currently, the actual performance of ctDNA in patients with lung cancer and isolated CNS progression remains largely unknown,” Dr. Aldea said, “so this is the question that we put in our study.”

Dr. Aldea and her colleagues screened 959 patients with NSCLC who were involved in prospective trials at Gustave Roussy between 2016 and 2018. Study inclusion required that patients have a molecular alteration detected via tissue sample and at least 1 ctDNA sample available from the time of CNS progression. Molecular alterations included ALK, EGFR, KRAS, ROS1, HER2, BRAF, TP53, and MET. Through these criteria, the study population was narrowed to 58 patients and 66 ctDNA samples, of which 21 were from patients with isolated CNS (I-CNS) progression and 45 were from patients with systemic disease progression (S-CNS). CtDNA was conducted with next generation sequencing and compared with imaging, molecular, and clinical patient data.

Most patients in the I-CNS group were female (94%), compared with about half of the S-CNS group (59%). Rates of adenocarcinoma and smoking history were relatively similar between I-CNS and S-CNS patients; in contrast, S-CNS patients had a median of two metastatic sites, compared with one in the I-CNS group. Rates of ALK, KRAS, and EGFR aberrations were slightly higher in the I-CNS group, whereas HER2, TP53, MET, and BRAF abnormalities were found only in the S-CNS group. Relating to the central hypothesis, 98% of S-CNS patients tested positive for at least one actionable driver via ctDNA analysis, compared with just 38% of I-CNS patients (P less than .0001). Resistance mutations were detected more commonly in the S-CNS group, although not significantly, which Dr. Aldea attributed to small population size.

“Plasma liquid biopsy is not a reliable marker for analyzing the molecular landscape of CNS progression,” Dr. Aldea concluded, adding that patients with isolated brain lesions may need to be treated with “more potent drugs” even when resistance mutations are not detected.

The investigators disclosed financial relationships with Celgene, Daiichi Sankyo, Eli Lilly, and others.

SOURCE: Aldea et al. ELCC 2019. Abstract 110O.

Next Article: