Skip to main content
Premium Trial:

Request an Annual Quote

Wash U Team Looks to Expand on Initial Success of Sequencing in Breast Cancer Clinical Trial


By Bernadette Toner

ORLANDO — Researchers from the Washington University School of Medicine in St. Louis this week reported initial results of sequencing 50 tumor/normal pairs as part of an ongoing breast cancer clinical trial and discussed their plans to extend the same approach into future studies.

At the annual American Association of Cancer Research meeting here this week, Matthew Ellis, a professor of medicine at the Wash U School of Medicine and a lead investigator on the project, discussed the use of whole-genome sequencing in 50 pretreatment tumor core biopsy specimens and blood normals as part of the American College of Surgeons Oncology Group's Z1031 study.

Ellis said that the team adopted sequencing as part of the trial in order to gain insight into the mechanisms that underlie resistance to aromatase inhibitor therapy.

In a separate presentation at AACR, Elaine Mardis, co-director of the Washington University Genome Center, also discussed the Z1031 study and noted that the researchers are applying a similar approach to another ACOSOG trial, Z1041, which is studying trastuzumab plus chemotherapy in HER2-positive breast cancer. Mardis said that trial will also use sequencing to identify signatures for resistance to therapy.

In his discussion of the Z1031 study, Ellis said that about half of the tumor samples, 24, came from patients who were resistant to aromatase treatment, and the remaining 26 samples came from patients who responded to the treatment. All the samples were estrogen receptor-positive.

After using a paired-end strategy to sequence to 30X coverage on the Illumina platform, the Wash U team identified more than 1,700 mutations that differed between tumors and normals.

Ellis said that individual tumors exhibited a wide range of variants. "Some had a handful, while others had more than a hundred," he said. In addition, resistant tumors tended to harbor more variants and generally had "more complicated genomes" than responsive tumors.

An analysis of the most frequently mutated genes revealed two that had been identified in previous studies — PIK3CA, which was mutated in about 40 percent of the cases, and TP53, which was mutated in about 20 percent of the cases.

They also identified a novel gene, MAP3K1, that appeared to play a role in ER-positive breast cancer. It was mutated in around 10 percent of cases in the initial 50 tumor/normal pairs, and the team later verified its occurrence in an additional 171 breast cancer tumors.

The majority of the MAP3K1 mutations are frame-shifts, so they essentially "knock the kinase out, which was a bit of a surprise," Ellis said. In addition, the Wash U researchers saw large deletions as well as structural variation in this gene. "This is an advantage of whole-genome sequencing," he noted, since some deletions were located outside of exons, "so this wouldn't have been picked up with exome sequencing."

Ellis said that aside from those three top genes and two others, ATR and MYT3, that occurred in at least 10 percent of the cases, most other genes occurred at a frequency of five percent or less. "It's a somewhat disappointing result in many ways because it says that breast cancer is extraordinarily complicated and not explained by a short list of commonly recurring genes, but by a large list of relatively rarely mutated genes," he said.

In order to use this information to gain insight into improved treatments for non-responders, Ellis and colleagues mapped the complete mutation list to what is referred to as the "druggable genome" — targets for which there are currently drugs on the market or in clinical trials. Of the 1,700 total variants, Ellis said that 307 were in the druggable genome, though he noted that not all of the drugs in that list are designed for cancer treatment.

The team did find a handful of mutations that would be targets for several cancer drugs, however, including imatinib, sorafenib, lapatinib, tretinoin, and clonazepam.

Ellis stressed that much further work will be necessary to validate whether these drugs would be effective in patients harboring these mutations.

In her talk, Mardis proposed a possible study design for a follow-on trial, in which patients with ER-positive disease would have their tumors sequenced, and, depending on their response signature, would be stratified into different trial arms: those predicted to respond to the aromatase inhibitor would receive that therapy only; those with mutations in therapeutic targets would receive the aromatase inhibitor plus a targeted therapy; and those deemed resistant to aromatase therapy would receive the targeted therapy alone.

Furthermore, Mardis said that her team has been considering a "paradigm for clinical sequencing" that would take advantage of the increased output of the Illumina HiSeq, which she noted, now exceeds 300 gigabase pairs per flow cell in a 10-day run.

Mardis said that it would be possible to generate, on a single flow cell, 30x coverage of a normal genome, 60x coverage of a tumor genome, the exomes for both the tumor and the normal sample, as well as a single lane for RNA-seq, thus providing full characterization of a cancer patient's genome.

"This is just an idea," Mardis noted. "We're investigating it, but we think it might hold water."

Ellis concluded that "a genome-forward approach to clinical trials and clinical care appears compelling," but stressed that "further work is required to validate therapeutic opportunities presented by mutations in the druggable genome."

In a discussion of the study following Ellis's talk, Matthew Meyerson of the Dana-Farber Cancer Institute said that the work is significant because it is "one of the first major large-scale genomic analyses using next-generation sequencing in the context of a cancer clinical trial."

He added that the MAP3K1 inactivation was independently observed in a Dana-Farber exome sequencing study of 116 breast cancer samples.

In terms of clinical implications, Meyerson said that the druggable genome analysis "adds to the range of pathways that can be targeted for breast cancer treatment."

More broadly, he said, it "demonstrates the feasibility of breast cancer genome analysis by next-generation sequencing in the clinical trial and clinical setting."

Have topics you'd like to see covered by Clinical Sequencing News? Contact the editor at btoner [at] genomeweb [.] com.

The Scan

Germline-Targeting HIV Vaccine Shows Promise in Phase I Trial

A National Institutes of Health-led team reports in Science that a broadly neutralizing antibody HIV vaccine induced bnAb precursors in 97 percent of those given the vaccine.

Study Uncovers Genetic Mutation in Childhood Glaucoma

A study in the Journal of Clinical Investigation ties a heterozygous missense variant in thrombospondin 1 to childhood glaucoma.

Gene Co-Expression Database for Humans, Model Organisms Gets Update

GeneFriends has been updated to include gene and transcript co-expression networks based on RNA-seq data from 46,475 human and 34,322 mouse samples, a new paper in Nucleic Acids Research says.

New Study Investigates Genomics of Fanconi Anemia Repair Pathway in Cancer

A Rockefeller University team reports in Nature that FA repair deficiency leads to structural variants that can contribute to genomic instability.