By Turna Ray
As KRAS testing becomes part of standard clinical practice in guiding treatment for colorectal cancer patients, two competing drug companies are racing to look for gene markers beyond KRAS that will allow them to differentiate their EGFR-inhibiting monoclonal antibody therapies.
Last week, Merck KGaA, Northern Ireland-based Almac Diagnostics, and the UK-based Medical Research Council announced that they are working together to identify biomarkers of respond to Erbitux (cetuximab). The clinical trial will analyze samples from the MRC COIN trial to "identify whether biomarkers apart from KRAS status alone further define which patients may benefit most from the addition of cetuximab to chemotherapy."
The MRC COIN study is a phase III, multicenter, three-arm study to examine whether the combination of oxaliplatin and cetuximab with fluoropyrimidine chemotherapy as a first-line treatment of metastatic colorectal cancer improves patient outcomes.
This week, at the American Association for Cancer Research's annual meeting in Washington, DC, Amgen announced results from a trial that used next-generation sequencing to investigate whether mutations in nine genes "beyond KRAS" predict response to Vectibix (panitumumab). The company highlighted that this study — which analyzed tumor samples from a pivotal Phase III "408" trial comparing Vectibix plus best supportive care to BSC alone — is "the first time next-generation sequencing has been used to analyze tumor samples" from a late-stage study. In addition to the nine gene mutations, Amgen's study also looked at whether the previously unanalyzed exon 3 mutations of the KRAS gene were predictive of response to Vectibix.
European regulators in 2007 recommended the use of Vectibix only in patients with normal KRAS and then made the same recommendation for Erbitux a year later. Across the Atlantic, since the makers of Vectibix and Erbitux received a simultaneous class labeling update from the US Food and Drug Administration recommending KRAS testing ahead of drug administration, neither Amgen nor Bristol-Myers Squibb was able to be the first to market with a personalized product.
Since the labeling change in the US, industry observers have noted that it was very unusual for competing sponsors, such as Amgen and BMS/ImClone, to jointly submit retrospective clinical trial data to FDA's Oncologic Drugs Advisory Committee. In reviewing the data, the advisory committee felt that FDA should accept the companies' retrospective analysis to update labeling under certain conditions. However, the simultaneous labeling change didn't allow either sponsor a market advantage (PGx Reporter 12-17-08).
Historically, Erbitux — marketed by Merck KGaA in Europe and by BMS/ImClone in the US — has outsold Vectibix in the colorectal cancer treatment market. Financial analysts have opined that if Amgen had invested earlier in validating the association between KRAS mutations and Vectibix response and launched the drug as a personalized product from the start, then perhaps the company could have gotten a leg up on Erbitux in the marketplace.
Now, after missing the first-to-market opportunities with KRAS, both Merck KGaA and Amgen are strategically looking into additional gene markers, with the likely aim of being able to launch companion tests to their drugs that are more sensitive and accurate than the competition.
Merck KGaA refused to speak to PGx Reporter about its biomarker research for Erbitux, explaining that its contractual agreements restrict it from discussing any company activities in Europe with US media. Almac, Merck KGaA's research partner, also refused to comment on the specific response markers the collaborators will focus on when analyzing samples from the MRC COIN study, citing contractual obligations.
Amgen, however, discussed with PGx Reporter the results of its sequencing study involving Vectibix after the data was highlighted at AACR's opening press conference on April 18. The study, which sequenced samples collected as part of a previously conducted Phase III study, reported that in addition to KRAS, NRAS mutations may be predictive of Vectibix response.
Both BMS and Amgen are working with Qiagen subsidiary DxS to market companion diagnostics to gauge KRAS mutations. Stephen Little, Qiagen's vice president for personalized healthcare, told PGx Reporter this week that the company has every intention of expanding its panel of markers that gauge response to EGFR-inhibiting colorectal cancer drugs, but only when additional markers are sufficiently validated and found to be clinically useful.
"The problem with a lot of the genetic associations between mutations and drug response over the years is lack of reproducibility," Little said. "Absolutely, we want to make sure that we provide the best possible test to predict drug response, but at the same time we have to be cautious that there is real clinical utility to justify the incorporation of the markers.
"What I don't think we will be doing is introducing markers too early, until we are absolutely happy that it's the right thing to do," Little added.
Amgen's sequencing study using Vectibix is certainly a step toward validating additional markers for drug response.
Studies such as Amgen's '408 analysis, which used the Roche 454 sequencing platform, "can achieve unprecedented sensitivity and quantitative accuracy for tumor mutation detection" since the technology allows "for millions of DNA sequences to be read simultaneously in a single experiment," an Amgen spokesperson said.
In the study, researchers from Amgen and other organizations sequenced archival patient tumor samples from 288 patients in the '408 Phase III study for mutations in nine genes in the EGFR pathway: AKT1, BRAF, CTNNB1, EGFR, KRAS (exon 3), NRAS, PIK3CA, PTEN, and TP53. These samples had previously been analyzed for KRAS exon 2 mutations using DxS' allele-specific PCR.
"This analysis found that in addition to KRAS, mutations in NRAS, another member of the RAS gene family, were also associated with lack of response to Vectibix," the Amgen spokesperson said. "Patients with wild-type KRAS and wild-type NRAS tumors receiving Vectibix had improved progression-free survival compared with those receiving best supportive care, whereas those with wild-type KRAS and mutant NRAS tumors did not appear to benefit from Vectibix."
While the sequencing study showed NRAS mutations occur infrequently, the data do suggest that NRAS mutations could be a predictive biomarker for Vectibix response. This data is consistent with previously published genotyping studies. "Further investigation in larger studies is required to determine the predictive value of NRAS mutations," the Amgen spokesperson said.
The sequencing study yielded data for an average of 7.85 genes per patient and the data completeness for each gene ranged from between 84 percent to 99 percent. More than one mutant gene was identified in 109 tumors and 20 tumors had more than one mutation in a single gene.
Noting that drug makers may use next-generation sequencing to pinpoint response markers in Phase III drug development trials, the Amgen spokesperson acknowledged that the company "will be looking for ways to apply this technology to other programs in our therapeutic oncology development programs."