Skip to main content
Premium Trial:

Request an Annual Quote

Gene Panel Sequencing Study Reveals Distinct Drivers in African American Lung Cancer


CHICAGO (GenomeWeb News) – Researchers from Ohio State University, GenomOncology, and the Vanderbilt-Ingram Cancer Center have come up with a gene sequencing panel aimed at understanding and more effectively treating non-small cell lung cancer in individuals with African American ancestry.

At an American Society of Clinical Oncology annual meeting poster session this weekend, Luiz Araujo, a post-doctoral researcher in the Ohio State University Comprehensive Cancer Center's thoracic oncology program, described results from a retrospective study applying this gene sequencing panel to samples from almost 100 African American individuals with NSCLC.

That analysis suggested that 'classic' NSCLC genes — frequently mutated genes such as EGFR identified in NSCLC tumors in the past — appear to be mutated at much lower rates in tumors from African American individuals with this lung cancer tumor type. Instead, the newly tested tumors tended to contain more frequent mutations in other cancer-related genes, particularly TP53.

"We found a very low frequency in 'classic' mutations among African Americans [with NSCLC]," Araujo told Clinical Sequencing News.

"Only about 24 percent of the cases had a classic, druggable driver mutation," he said, noting that in past studies of unselected patients, many of them Caucasian, more than half of individuals with NSCLC typically have targetable tumor mutations involving the best known NSLC genes.

Such findings have important treatment implications, Araujo explained, since frequently mutated genes found in the past are sometimes considered for targeted treatment. They also hint that there are still to be detected driver mutations behind many NSCLC cases — including some that could be targeted by new therapies or by drugs developed for other cancer types.

Prior studies suggest that EGFR mutations may be found at lower frequencies in NSCLC tumors from African American patients than in those from Caucasian patients. That pattern has been debated, though, since studies in different populations have included variable proportions of lung cancer patients with or without a smoking history.

Nevertheless, differences in African American NSCLC patient responses to EGFR-targeted therapies — including findings reported in the Journal of Clinical Oncology in 2009 — are consistent with the notion that targetable mutations may differ in NSCLC tumors from African American individuals.

By scouring the lung cancer literature, the researchers narrowed in on a set of 81 lung cancer-specific genes with known or suspected therapeutic relevance, which they used to develop a gene sequencing panel built around a custom Agilent Haloplex kit.

After validating the panel on more than a dozen cell lines and formalin-fixed, paraffin-embedded NSCLC samples with known mutation patterns, the team used it to nab the selected sequences from retrospectively collected NSCLC samples from 99 African American NSCLC patients between the ages of 41 and 76 years.

More than 60 percent of the samples came from patients with stage I disease, though stage II, III, and IV tumors were included in the analysis as well. Fifty of the NSCLCs had been sub-classified as adenocarcinomas, while 31 were squamous cell carcinoma. The remaining tumors represented adenosquamous and other forms of NSCLC.

In the validation stage of the study, they sequenced the gene panel set using Illumina's HiSeq 2500, unearthing known mutations in EGFR, KRAS, and other genes in the previously characterized cell line and FFPE samples.

Within the patient tumor samples, meanwhile, the team identified some 3,500 non-synonymous variants, which were subsequently whittled down through comparisons with the dbSNP database, SIFT and PolyPhen analyses, and manual curation.

Each of the patient tumors contained between zero and 11 potential driver mutations — three each, on average, Araujo noted. But the location and frequency of these mutations did not match that described in past studies of NSCLC. And more than three-quarters of the tumors lacked mutations involving the most recurrently mutated genes described previously.

For example, the EGFR gene was altered in only around 5 percent of the samples and the KRAS gene was altered in 16 percent of the African American tumors. Other classic NSCLC drivers such as PIK3CA and NRAS were each mutated in 1 percent of samples, Araujo reported.

"We demonstrated a relatively low frequency of classic driver mutations in NSCLC among [African Americans]," he and his co-authors noted in their ASCO abstract. "This group may benefit from the identification of novel drivers through comprehensive genomic approaches."

On the other hand, the team tracked down recurrent mutations in many genes that act as drivers in other cancer types and lung cancer sub-types. These included TP53 (mutated in almost half of the tumors), as well as LRP1B, KEAP1, SMARCA4, and MLL2.

At the pathway level, the researchers saw potential roles for processes ranging from cell survival, proliferation, and stem cell maintenance to epigenetic regulation and Wnt-signaling.

In parallel to that mutational analysis, the researchers tested germline samples from the African American patients at a few dozen ancestry-informative SNPs to examine the proportion of African and European ancestry in each patient's genome.

By bringing together genetic ancestry and tumor somatic mutation patterns, Araujo explained, they were able to explore the possibility that particular mutations might be more apt to appear in tumors from those with more or less African or European ancestry.

On average, the patients had around 70 percent African and 30 percent European ancestry, Araujo noted. But he and his colleagues did not see a relationship between any of the genetic ancestry markers assessed and specific somatic driver mutations.

"It leads to the conclusion that genetic [ancestry] by itself might not be a predictor of which kind of biology this lung cancer will have," Araujo said.

In its current form, the 81-gene panel led to driver mutations in tumors from 82 of the 99 individuals tested, with at least half of those appearing in potentially targetable genes.

As more targetable driver mutations are detected, the team plans to add them to the gene panel used to search for actionable mutations in tumor samples from African Americans with NSCLC.

"We have to add more genes because we still don't have a lot of drivers, so that's something to be evaluated," Araujo said.

The researchers are interested in using the existing panel to test samples from additional African American NSCLC cases, Araujo said. As they work to untangle lung cancer biology — and potential vulnerabilities — in different populations, they also hope to apply it to lung cancer samples from Latino American individuals, who typically have African, European, and Native American ancestry.

Araujo emphasized the need to standardize the gene sequencing panel so it can be applied to lung cancer samples in both research and clinical settings, but said the team has no plans to commercialize it.