NEW YORK (GenomeWeb News) – A gene fusion identified in DNA and RNA sequences from a never-smoker lung cancer patient may act as a driver in a subset of lung cancer cases, a new study suggests.
In a paper appearing online yesterday in Genome Research, a South Korean team described how they used genome and transcriptome sequencing to find a gene fusion suspected of contributing to lung cancer formation in a 33-year-old never-smoker with lung adenocarcinoma. The fusion involved two chromosome 10 genes: KIF5B and RET. Researchers found the same fusion in two of 20 lung adenocarcinoma samples screened in their replication experiments.
"Our data demonstrate that a subset of [non-small cell lung cancers] could be caused by a fusion of KIF5B and RET, and suggest the chimeric oncogene as a promising molecular target for the personalized diagnosis and treatment of lung cancer," corresponding author Jeong-Sun Seo, director of Seoul National University's Genomic Medicine Institute and chairman of the Korean genomics firm Macrogen, and colleagues wrote.
Although smoking is a well-known risk factor for lung cancer and is believed to be a very common cause of the disease in the Western world, Seo and his co-authors explained, roughly one-quarter of lung cancers diagnosed globally occur in individuals who have never smoked. In some Asian countries the proportion is even higher, they noted, especially for lung tumors classified as non-small cell lung cancers.
In the hopes of learning more about the genetic features that might predispose non-smokers to lung cancer, the team used the Illumina HiSeq 2000 and GAIIx to do massively parallel sequencing on genomic DNA from matched tumor, normal, and metastasis samples from a 33-year-old Korean man with lung adenocarcinoma.
Using a similar strategy they also sequenced the transcriptome of a liver metastasis sample from the patient.
For the genome-sequencing arm of the project, the team generated sequence covering the liver metastasis genome to an average depth of almost 48 times. The matched normal blood genome was sequenced to an average depth of 28 times.
The primary lung cancer genome, generated using DNA from a fresh-frozen paraffin-embedded sample, was sequenced to an average depth of around 20 times. Sequence from the primary lung tumor genome was not used for the initial analysis owing to the low DNA quality in the FFPE sample, researchers explained, though the primary tumor sequence was used for validation stages of the study.
When they sifted through sequence for the liver metastasis and blood genomes, researchers uncovered eight non-synonymous single nucleotide changes and two small insertions or deletions in the tumor that were not present in the matched normal sample.
Nevertheless, they could not find any mutations affecting in a set of genes previously tied to lung and other cancers, such as EGFR, BRAF, PIK3CA, or KRAS. And based on the genes that were affected by the single nucleotide variations in the tumor, the team argued that these SNVs were not likely to have contributed to lung cancer development.
"The somatic mutations may be present in the primary lung cancer or may have occurred during metastasis," the study authors noted. "In any case, we suggest they are unlikely to be driver mutations."
Likewise, in their analyses of genome and transcriptome sequences the researchers did not find the fusion between EML4 and ALK genes that has been reported in some past studies of lung adenocarcinomas from relatively young patients.
Instead, their search uncovered a fusion involving the kinesin family gene KIF5B and the receptor tyrosine kinase gene RET, both from chromosome 10. The fusion appears to have developed as the result of a chromosome 10 inversion, they reported, and leads to much higher expression of the known proto-oncogene RET.
Through their PCR amplification and Sanger sequencing experiments, the team demonstrated that the KIF5B-RET fusion was present not only in the patient's liver metastasis but also in his primary lung cancer and in a bone metastasis.
The KIF5B-RET fusion turned up in lung cancers from other individuals as well: analyses of five lung adenocarcinoma transcriptomes sequenced previously and of data from targeted sequencing experiments on 15 more lung adenocarcinoma samples uncovered two cases involving the fusion.
"These results clearly show that KIF5B-RET fusion is not rare and that the fusion transcript exists in the primary lung adenocarcinomas," Seo and colleagues noted. "In addition, because it would be very unlikely to find identical non-functional fusion genes in different cancer tissues, these results also provide indirect evidence that expression of the KIF5B-RET fusion gene has an important functional impact in lung cancer."