Next-generation sequencing has identified additional breast cancer patients who may benefit from HER2-targeted therapies but are missed by current tests for HER2 gene amplification such as fluorescence in situ hybridization and immunohistochemistry.
Researchers from Washington University have re-analyzed data from eight breast cancer genome-sequencing projects totaling more than 1,500 patients. They found activating mutations in the HER2 gene in 25 patients who did not have HER2 gene amplifications and would therefore not be deemed eligible under current guidelines to receive targeted therapy.
While the overall incidence of HER2 somatic mutations was low, at around 1.6 percent, this could still translate into 4,000 women per year in the US alone.
Additionally, most of the patients had mutations in druggable regions of the gene. Nearly 70 percent of patients, 17 out of 25, had mutations in the tyrosine kinase domain of HER2, while around 20 percent of patients, 5 out of 25, had mutations in the extracellular domain.
The results, which were published this week in Cancer Discovery, have spurred the launch of a clinical trial to test HER2-targeted drugs in women with these mutations.
Currently, women with breast cancer are tested to see if they have additional copies of the HER2 gene, which would make them eligible for approved targeted therapies such as Genentech's trastuzumab (Herceptin) or pertuzumab (Perjeta), GlaxoSmithKline's lapatinib (Tykerb), or for clinical trials of Puma Biotechnology's neratinib and Genentech's trastuzumab-DM1.
However, recent sequencing studies have found that somatic mutations in the gene may also increase gene activation and drive tumorigenesis, suggesting that these women may also benefit from such targeted therapies, even though they do not have elevated copies of the gene.
"We're identifying patients who would benefit from HER2-targeted drugs that are being missed by current screening tests," Ron Bose, lead author of the paper and a medical oncologist at Wash U's Siteman Cancer Center, told Clinical Sequencing News.
Bose and the Wash U team originally discovered HER2 somatic mutations in two patients that had been sequenced as part of the American College of Surgeons Oncology Group's clinical trial Z1031, which was conducted to identify genomic pathways that may predict response and resistance to aromatase inhibitor therapy in women with estrogen-positive breast cancer (CSN 6/13/2012).
The two cases were originally classified as HER2 negative by the FISH and IHC tests used to measure HER2 gene amplification.
They then scoured whole-genome sequence data from breast cancer sequencing studies that were part of the Cancer Genome Atlas as well as studies that were done at Wash U's Genome Institute. Out of around 1,500 patients included in the eight studies they identified 25 with somatic mutations in HER2 genes that had been classified as HER2 negative from standard testing because these patients did not have extra copies of the HER2 gene.
To determine the functional consequences of the somatic mutations, the researchers first looked at available expression data from RNA-seq studies.
The RNA-seq data was helpful on two different levels, said Bose. First, it helped to confirm that the mutations were in fact real, he said. And second, "if we're going to give a drug treatment based on the presence of a mutation, we need to know that it's being expressed at the RNA and protein level."
Looking at RNA-seq data from six patients that were part of the TCGA study, they confirmed that in all six cases, the HER2 mutations were being expressed. In the two cases from the Wash U aromatase inhibitor trial, the team used PCR and Sanger sequencing to sequence tumor RNA and detected the mutant allele in both of those cases as well.
Next, they set out to characterize the somatic HER2 mutations. One mutation had been previously characterized in non-small cell lung cancer. They also identified a mutation that is known to confer resistance to lapatinib, an approved drug that targets HER2. To determine the effect of the other mutations, the team designed in vitro assays, tested the effect of the mutations in xenograft models, and also tested the effects of different HER2 drugs on the cell assays.
In total, the team characterized 13 somatic mutations from the 25 patients. Seven mutations found in 13 patients were found to activate HER2. The clinical effect of one mutation, a deletion, found in three patients could not be determined; however, the deletion is homologous to EGFR exon 19 deletions, which are found in non-small cell lung cancer patients and are sensitive to the AstraZeneca and Teva's gefitinib (Iressa). Three mutations found in two patients showed no functional effect. Two mutations have not been functionally tested yet.
When testing drugs on the assays, the team found that neratinib was a potent inhibitor of all of the HER2 mutations. Lapatinib was less potent and, as expected, had no effect on the cells with the mutation that predicted resistance. Trastuzumab had a more complex effect because its action includes activation of the immune system, and the researchers determined that they could not adequately model the drug's effects on the assays.
Bose said the next step is to begin a clinical trial that will enroll women with metastatic breast cancer who have tested negative for HER2 gene amplification. Those women will have the HER2 gene sequenced and, if somatic mutations are found, will receive neratinib.
The trial will be run at four sites — Washington University School of Medicine, the Dana-Farber Cancer Institute, Memorial Sloan Kettering, and the University of North Carolina at Chapel Hill.
Bose anticipates that the trial will run about two years. In the first step, around 500 patients will be screened and at least 10 treated. In total, he said, the goal will be to treat around 30 patients.
If the trial in women with metastatic cancer is successful, the researchers would then expand it to look at women who are newly diagnosed or who have just had surgery to remove the primary tumor, he added.
The hope is that the trial will ultimately lead to better outcomes, by identifying more women who will benefit from HER2-targeted therapy, Bose said.
This could work in two ways, he explained. The additional sequence information could be added to existing HER2 testing, so that women would first receive a standard HER2 test via IHC or FISH and also have the HER2 gene sequenced, or it could lead to the development of a completely new next-gen sequencing-based test that would identify mutations as well as amplifications.
"It would be very exciting to have a panel of genes to test in breast cancer, and for all of them, we'd know their implications," Bose said.
Already, a number of companies and academic groups have developed sequencing-based cancer panels, many of which include HER2. Currently, the gene is included because mutations in it have been demonstrated to play a role in lung cancer, said Bose. However, if the clinical trial for HER2 mutations in breast cancer proves to be successful, then such panels could also be used to guide treatment for breast cancer patients, he said.