NGS-Based Cancer Tests Reveal Limitations of Manual Variant Analysis

Junaid Shabbeer, Clinical Science Director QIAGEN Silicon Valley

Junaid Shabbeer Clinical Science Director QIAGEN Silicon Valley

As we work with our early access partners to put the finishing touches on QIAGEN’s Ingenuity Clinical web application for DNA sequence variant interpretation in clinical labs, we sat down with Junaid Shabbeer, the new clinical science director here at QIAGEN Silicon Valley. Junaid has been directing clinical molecular diagnostic labs for years, and his experience in the field has helped shape Ingenuity Clinical to ensure that it will be just what this community needs.

Junaid, who joined our team earlier this year, earned his PhD in molecular biology and received ABMG certification in clinical molecular genetics. He has spent 20 years in the field, working at Mount Sinai in New York City, Myriad Genetics, Genomic Health, and Ariosa Diagnostics, among others. He has made extensive use of next-gen sequencing for gene panel and other tests.

As a lab director at many of those organizations, Junaid has seen firsthand the challenges of filtering and interpreting variants generated from clinical genetic tests. One of the biggest: classifying variants as benign or pathogenic, rather than just giving them the “unknown significance” label. Junaid remembers situations where a variant seemed intuitively to be benign, but because it had only been seen in a couple of pedigrees with limited clinical information, “we just didn’t have the confidence to call it benign, and we had to downgrade it to a variant of unknown significance,” he says. The same held when variants were described in published research studies using cell lines or mouse models, and the information suggested a variant was pathogenic, but its source meant it wasn’t directly applicable to human cases. “What’s frustrating at times is that you’re trying your best to help physicians and patients,” Junaid says. “Something in my gut would say that a variant was deleterious or benign, but we just didn’t have enough experience with the variant to classify it as anything other than unknown.” It was always a triumph when enough information coalesced to more definitively classify a variant.

Another significant factor in analysis is time. “When I first arrived at Myriad Genetics, I was involved in building the variant analysis team,” he says. “We had our own internal database, experience with pedigrees, and great statistical tools — but we still relied on Google Scholar as one of our primary methods for retrieving papers.” Taking that time to track down relevant papers, read them, and consider their findings in light of a particular variant was a rate-limiting step. “As test volume increased, that step became much more time-consuming,” Junaid recalls. “It became very taxing.”

Much of Junaid’s experience is in cancer-related testing. For hereditary cancer, he says, the most important task was building out the bibliography of sources to help with variant analysis. In addition to existing commercial options, he says, lab directors and variant specialists want to be absolutely certain that every relevant paper and finding has been considered when weighing how each variant should be classified. “You want to be absolutely certain,” he says. “And the director might have his or her own experience with these variants from looking at other families, so any analytical approach needs to support that data as well.”

When it comes to somatic cancer, Junaid notes that the major hurdle is the sheer number of genes involved. “The volume of clinical information used to interpret variants in these genes is a real challenge,” he says. Also, while hereditary cancer screening has some well-established rules for variant classification — such as ACMG guidelines — this is not the case for somatic cancer. If a confirmed variant has not been reported before, the information for interpreting and determining the significance of that variant is harder to come by.

Naturally, we are thrilled that Junaid is deploying his considerable expertise to help with development of the Ingenuity Clinical product. As he says, “The questions that we’re asking are the same that any lab director would ask in weighing a variant.” The vast amount of content in Knowledge Base, the brain powering Ingenuity Clinical, will help ensure that clinical lab teams are confident that all relevant information has been considered in variant interpretation. It will eliminate the need for those time-consuming literature searches, and could even bring enough information to bear so that clinical analysts can upgrade the classification of variants out of the “unknown significance” black hole, whether those variants are from somatic or hereditary cancer.

We are currently working with our early access partners on the new product, and we look forward to making the tool available to a broader audience in the near future.