Skip to main content
Premium Trial:

Request an Annual Quote

Genomics in the Journals: Arrays 2014.08.07

NEW YORK (GenomeWeb) – A Maryland-based team used real-time, quantitative PCR to screen for human papillomavirus in blood and saliva in order to detect HPV-positive head and neck cancers, track their recurrence, and predict survival patterns.

As they reported in the Journal of the American Medical Association Otolaryngology-Head & Neck Surgery journal, researchers from Johns Hopkins Medical Institutions and Baltimore's Milton J. Dance Head and Neck Center performed RT-qPCR tests focused on the E6, E7, and beta-actin proteins in HPV-16, a high-risk HPV subtype associated with enhanced risk of oropharyngeal squamous cell carcinoma and other head and neck cancers.

Using this screening method, they tested blood and saliva samples from 93 individuals with squamous cell carcinoma originating in the oropharyngeal region or unknown primary site, both before and after treatment.

For the 81 HPV-positive cases in that cohort, the team determined that HPV-16 status in blood and saliva samples taken after treatment appeared to predict disease recurrence with almost 70 percent sensitivity and more than 90 percent specificity. The results of DNA tests done prior to treatment, meanwhile, seemed to offer a reliable means of identifying tumors involving HPV-16 infections.

"Until now, there has been no reliable biological way to identify which patients are at higher risk for recurrence, so these tests should greatly help do so," the study's senior author Joseph Califano, an otolaryngology researcher at Johns Hopkins and medical director at the Milton J. Dance Head and Neck Center, said in a statement.


In Nature Genetics, an international team led by investigators in the US and Germany described a handful of new genomic loci linked to pancreatic cancer risk. As part of a study dubbed PanScan III, the team turned to a multistage genome-wide association study approach to look for genetic variants associated with pancreatic cancer, starting with genotyping profiles for 1,582 individuals with pancreatic cancer and 5,203 individual without.

Following the first phase of the study, the researchers considered variant patterns in another 3,525 cases and 3,642 controls before performing a meta-analysis that included data from both stages of the study. In the process, they verified pancreatic cancer associations at four sites in the genome for individuals with European ancestry.

Two more SNPs, not implicated in earlier pancreatic cancers studies, had genome-wide significant ties to the condition in that meta-analysis, while 11 more variants had more tenuous associations with pancreatic cancer.

From there, the researchers tested that set of 13 variants in 2,576 more individuals with pancreatic cancer and more than 5,500 unaffected controls before doing another meta-analysis of cases and controls from all three study stages. All told, they turned up five new loci with genome-wide significant pancreatic cancer associations — involving sites on chromosome 5, 7, 13, 16, and 22 — and one suggestive locus on chromosome 8.


A GWAS published in the Journal of the American Medical Association pointed to a set of 16 SNPs in and around the CYP2C gene that were associated with serious adverse skin reactions in individuals taking the antiepileptic drug phenytoin.

Researchers from Taiwan's Chang Gung Memorial Hospital, the National Yang-Ming University, and elsewhere did array-based genotyping at more than 900,000 SNPs in 130 phenytoin-tolerant controls and 168 individuals who'd experienced severe reactions to the drug, including severe cutaneous adverse reactions and less severe reactions known as maculopapular exanthemas. They also considered genotypes for almost 3,700 population controls from Taiwan, Japan, and Malaysia.

For the first phase of the study, the team focused on Taiwanese samples, identifying 16 suspicious SNPs on chromosome 10 in genotyping data for 60 individuals with severe adverse skin reactions linked to phenytoin use and more than 400 population controls from Taiwan. Half of those appeared in CYP2C genes, which code for cytochrome P450 oxidase enzymes known for their roles in drug response.

In follow up experiments, the researchers not only assessed such variants by targeted sequencing in the Taiwanese population, but also replicated their findings in phenytoin-tolerant and –susceptible individuals and population controls from Japan and Malaysia.

A missense variant in the CYP2C9*3 gene showed ties to especially enhanced risk of severe skin reactions in phenytoin takers. The study's authors determined that the variant delayed clearance of phenytoin from the body, leading to prolonged presence of the drug in patients' bloodstreams.

"These findings may have potential to improve the safety profile of phenytoin in clinical practice and offer the possibility of prospective testing for preventing phenytoin-related severe cutaneous adverse reactions," they concluded.


A team from the US, Canada, and Spain defined 11 main subtypes for cancers originating in a dozen tissue types. As they reported in Cell, the researchers considered integrated genomic and proteomic profiles generated by the Cancer Genome Atlas for more than 3,500 cancer samples from 12 cancer types.

Using array-based copy number, methylation, transcript expression, and microRNA levels for the tumors — together with proteomic profiles and mutation patterns detected by whole exome sequencing — the researchers were able to cluster samples into groups that reflected the tumor's tissue of origin in some cases and encompassed multiple tissues of origin in others.

For instance, five of the 11 subtypes they found coincided well with the tissue in which the primary tumor was detected. But other subtypes included samples from multiple tissues of origin, including a subtype comprised of lung squamous samples, head and neck tumors, and some bladder cancers.

Other bladder cancers belonged to distinct pan-cancer subtypes, the study's authors noted, with a cluster containing lung adenocarcinomas. Similarly, the analysis highlighted the dramatic molecular differences that can occur between breast cancer subtypes, pointing to shared features between basal or "triple-negative" breast cancer and cancers found in the ovary.

"This genomic study not only challenges our existing system of classifying cancers based on tissue type, but also provides a massive new data resource for further exploration, as well as a comprehensive list of the molecular features distinguishing each of the newly described cancer classes," the Buck Institute for Research on Aging's Christopher Benz, a co-corresponding author on the study, said in a statement.