Researchers from Washington University in St. Louis have published a case study on a patient who was successfully treated with a targeted therapy based on results of the university's clinical cancer sequencing.
The case report — of a woman with a metastatic thymic carcinoma who had failed standard therapy and was found, by Wash U's sequencing test, to have an activating KIT mutation that guided successful treatment with a tyrosine kinase inhibitor — appeared online in the Journal of Thoracic Oncology this month.
The authors hope the paper will serve as a single-subject clinical trial — evidence of the utility of Wash U's comprehensive targeted cancer sequencing service, which the university's Genomics and Pathology Services laboratory has been offering on the Ilumina HiSeq an MiSeq since 2011.
Ian Hagemann, the study's first author and an associate medical director of the GPS lab, told Clinical Sequencing News this week that the patient described in the group's new report was referred to Wash U in 2012 — relatively early in the university's clinical cancer sequencing program — after experiencing disease progression during treatment with standard chemotherapy strategies.
The Wash U team sequenced the patient's tumor DNA using what was at the time a 25-gene targeted sequencing panel, and identified a three-nucleotide deletion in the KIT codon 579. According to the group, though not previously associated with thymic tumors, this alteration had been reported to be associated with response to the Novartis' TKI Gleevec (imatinib) in other tumor types.
Based on the sequencing results, the patient's physician began treating her with Gleevec. After three months of therapy, the patient showed stable disease and reduction in tumor size, which persisted to her latest follow-up 23 months after initiating therapy, the study authors reported.
According to Hagemann, the sequencing metrics for this reported case — mean coverage of more than 3,000-fold, and more than 50-fold for about 99 percent of targeted positions — were on par with other cases the group has analyzed subsequently.
"This was one of our earlier cases, but subsequent cases have given similar results," he said.
Catherine Cottrell, the study's corresponding author, and the medical director of Wash U's GPS lab, told CSN that the lab has now sequenced more than 1,200 patients.
For about 40 percent of these clients, Hagemann added, the group has been able to identify a potentially targetable mutation.
"We say 'potentially' because there are so many considerations that go into the treatment decision above and beyond what we are able to ascertain, [such as] comorbid conditions that prevent someone from getting a drug, or insufficient access, or maybe data we think looks strong on paper that has been refuted in a recent abstract we haven't seen," he said.
For this and other reasons, only some of this 40 percent of cancer patients for whom sequencing identifies a potential drug target have gone on to be treated based on the results.
The group has not collected hard numbers for how this plays out as a percentage of patients who have been sequenced. However, "the number of patients that have been treated based on the results [of this testing] is something we are actively investigating," Hagemann said.
For the group's case-report patient, Hagemann said several factors influenced the successful implementation of NGS-directed therapy.
For one, he said, the patient's doctor, "a sophisticated geneticist in his own right," with significant experience using targeted therapies and interpreting genomic data, was prepared and confident in implementing the results of sequencing into his treatment strategy.
"This testing is most powerful in the hands of a knowledgeable ordering physician," Hageman said. "We've found that ordering physicians who have less background in genetics are a little more unsure of what to do with the results of this kind of genomic testing.
"One of our missions as clinical genomicists is to provide education to ordering clinicians," he added. "We'd like to be a resource [for] questions like 'what do you mean that a variant is present at a low allele frequency?' or 'what does it mean that a variant may be actionable?' These are all questions people have and we want to be able available to answer them."
Additionally, he said, the patient's cancer type and mutation status made interpretation of the sequencing results easier than it could have been.
"What was a little special about this [case] is that it’s a perfect example of what you might call the 'sweet spot' of clinical next-gen sequencing: a tumor type that is rare enough that there is a small number of standard therapies available, but [one] that has been studied before with some published [genomic] literature," he said.
"A tumor for which nothing is known on the molecular level is also a good candidate for this type of testing, but it's harder to interpret sequence results when there is a vacuum of knowledge surrounding the underlying molecular events," he added.
On the technical side, the patient also had sufficient tumor available for successful sequencing. According to Hagemann, the Wash U lab currently sets its minimum input cutoff at 200 nanograms of DNA, or 150 nanograms with a disclaimer that the results may not be reliable.
As Wash U has continued to offer clinical targeted cancer sequencing, it has increased its panel from 25 to 40 genes as of late last year with plans for a third version in 2014.
For the subject of the team's recent case report, the NGS results were confirmed by Sanger sequencing, but Cottrell said that the lab no longer bothers with this verification step.
"For the first fifty cases, variants detected by next-generation sequencing were Sanger verified, and to avoid discrepant analysis we verified the mutation-negative regions within our targeted genes as well," she said. "All results were concordant and due to the high sensitivity and specificity of our comprehensive cancer sequencing assay we subsequently felt confident to report out solely on the basis of our NGS results."
Cottrell and Hagemann said that the team hopes to publish additional case studies from the cohort of patients who have received its clinical cancer sequencing service, despite difficulties of publishing such reports.
Case reports are currently "having a moment" in the precision medicine field," Hagemann said.
"But while case reports conveying this type of story are becoming easier to publish, journals have a limited amount of space available and there are a limited number of venues," he added.
"The whole concept of [personalized medicine] is that n equals one in almost all of these cases," he explained. "It's not possible to do a prospective study of 200 thymic carcinomas all of whom have an activating KIT mutation. That study could never be done… So we hope, as early adopters of this technology, that the [US Food and Drug Administration] and payors will recognize that the case report is no longer the lowest level of clinical evidence and, for acting on genomic data, it can be the highest."