Skip to main content
Premium Trial:

Request an Annual Quote

At AACR, Scientists Detail Efforts to Combat Resistance to Targeted Cancer Drugs

WASHINGTON (GenomeWeb Daily News) — While cancer therapies targeted to the genetic makeup of patients' tumors have allowed for more precise treatment of disease, tumors bombarded with such drugs may develop resistance, even to such directed drugs.

Resistance may be either innate, as part of a subpopulation of the tumor, or acquired as new mutations arise. As a slate of researchers recounted during a plenary session yesterday at the American Association for Cancer Research annual meeting, they are developing approaches to identify how such resistance arises and to find ways to combat it.

"Targeted agents work very well on average," said René Bernards, a group leader in the molecular carcinogenesis division at The Netherlands Cancer Institute and chief scientific officer at Agendia. However, he added, "resistance ultimately develops."

There are two general ways in which resistance to a drug may arise: A mutation arises in the drug target or a separate pathway, which Jeff Engleman from Massachusetts General Hospital dubbed a 'bypass track' during his talk, providing an alternate way for the downstream processes to be activated. Combinations of targeted drugs could be used to hit both the original mutation and the new mutation or the bypass track, Engleman said.

The typical way to identify drugs aimed at these secondary changes, moving from cell lines to mice to patients, can take a long time, he said. Instead, Engleman and his colleagues turned to screens to find combinations of drugs that overcome such resistance in cell lines and in biopsies isolated from patients.

To date, Engelman's group has screened about 30 lung cancer cell lines, 16 with EGFR mutations and 14 with ALK changes, to identify effective therapies and link those therapies to a biomarker.

They have also conducted repeat biopsies of about 165 tumors to examine the genetic makeup of the resistant tumor. "In each case, the original EGFR mutation or ALK translocation was present in the resistant lesion," Engleman said.

Further, he and his team demonstrated that tumors tend to contain a variety of resistance clones that may be present in the tumor population at different abundance levels.

Indeed, Todd Golub's team at the Broad Institute performed whole-exome sequencing on tumor and normal samples from people with multiple myeloma, assessing their heterogeneity, and found that there was a median of four to five different subclones per tumor. They saw similar rates, he added, in samples from ovarian cancer patients.

"I think there are some significant clinical implications of tumor heterogeneity," Golub said.

Serial biopsies further indicated to Engleman and his team that, due to selective pressures imposed by drugs, those population levels of resistant subclones could vary dynamically.

Still, the most lethal combination of drugs to affect those changing populations needs to be uncovered. Bernards and his colleagues at NKI noted that, even with current targeted therapies, different tumors that harbor the same mutations respond to targeted drugs in different ways. For example, both melanoma and colorectal cancer may contain BRAF V600E mutations, but BRAF inhibitors are only effective in treating melanomas, not colorectal cancers. Colorectal cancer, Bernards said, must have a kind of intrinsic resistance.

Using a short hairpin RNA-based screen, Bernards and his colleagues searched for a gene whose down-regulation would interact with a known cancer gene to cause cell death of the tumor. From this, they identified three hairpin factors to EGFR, which is found upstream of BRAF. They confirmed, however, in in vitro and mouse studies that inhibition of EGFR interacted strongly with BRAF inhibition to be effective in colorectal cancer.

Through a separate pathway, EGFR can also increase AKT, leading colorectal tumor cells to not respond to BRAF inhibitors.

A clinical trial, Bernards added, is underway to test the combination of BRAF and EGFR mutations for treating colorectal cancer. He noted that eight months lapsed between the publication of the study detailing this work, in Nature in January 2012, to the start of the trial.

BRAF inhibitors are effective by themselves in melanoma, as melanomas do not typically express EGFR. Among those tumor cells that do develop a mutation that allows them to express EGFR, that subclone then becomes resistant to BRAF inhibitors.

While many resistance characteristics may be due to new mutations that arise in tumor cells, the Broad's Golub noted that the tumor microenvironment could also help tumor cells evade the reach of cancer therapies.

Golub added that the majority of tumor cells with BRAF V600E mutations respond to the protein kinase inhibtor vemurafenib, but some do not, as there appears to be a partial innate resistance to the drug, he said.

On a 384-well plate, Golub and his colleagues examined how 45 cancer cell lines fared in a variety of microenvironments, containing a selection of 23 different stromal cell types in response to 35 different cancer therapeutics. "A number of cancer types were rendered resistant," he said. Certain fibroblasts, he added, appeared to rescue melanoma from vemurafenib.

Using an antibody assay, they found that the fibroblast cells were secreting hepatocyte growth factor, or HGF, which is enough to confer resistance as it activates its receptor, MET, which, in turn, leads to the activation of the AKT signaling pathway. The addition of an antibody blocking HGF further indicated HGF is necessary and sufficient for resistance.

Golub and his team also found that some dermal fibroblasts appeared to enable resistance to gemcitabine, but their antibody array was unable to find any hits. Through an HPLC assay, they saw that gemcitabine and its metabolite dFfU change in proportion in response to the fibroblast. The conversion between gemcitabine and dFdU is mediated by CDA, but RNAi knockdown of CDA still leads to resistance. The mechanism behind how this resistance phenotype works isn't yet understood. "We're still tracking this effect," Golub said.

"The microenvironment may be a common source of innate drug resistance," he added.