Skip to main content
Premium Trial:

Request an Annual Quote

Lack of FDA Guidance on KRAS Testing Creates Confusion; Pharmas See Lower Sales


The lack of clear guidance from the US Food and Drug Administration regarding KRAS testing to target treatment with anti-EGFR monoclonal antibodies for colorectal cancer is affecting sales of such drugs and putting drugmakers in a difficult spot with regard to what they can and cannot say to doctors.

During an earnings call Last week, BMS said confusion over KRAS testing caused fourth-quarter US sales of its colorectal cancer drug Erbitux to slide 2 percent to $179 million from the year-ago period.

BMS CEO Jim Cornelius was also quoted by Reuters saying: "We have to admit that Erbitux sales suffered in the fourth quarter and it has to do with KRAS testing, this genetic phenomenon, which is a predictor of whether Erbitux will be efficacious."

According to Cornelius, "it caused a lot of confusion among oncologists, and there are probably some [patients] that deserved to get Erbitux that didn't."

Also last week, Amgen, which markets the anti-EGFR colorectal cancer drug Vectibix, said fourth-quarter US sales of the drug dropped 24 percent to $25 million year over year.

"We're in a tough position right now with Vectibix," an Amgen spokesperson told Pharmacogenomics Reporter. "We made it clear during [an] FDA advisory committee meeting in December that current literature strongly suggests that Vectibix is not effective in a certain population. But without an [FDA] recommendation for testing in the label we can't legally discuss off-label KRAS testing with doctors."

In December, FDA's Oncologic Drugs Advisory Committee met to discuss the type and amount of data needed to support product labeling using biomarkers when prospective studies have not been conducted.

The committee considered Vectibix and Erbitux in discussing the issues surrounding retrospective analyses, including assay validation, appropriate sample requisition, and biomarker discovery. Amgen and BMS have each submitted retrospective analyses in an effort to update the labels of their drugs to show they are more efficacious in patients with metastatic colorectal cancer who carry a wild-type version of the KRAS gene [see PGx Reporter 12-17-2008].

[ pagebreak ]

While European regulators have narrowed the indication for the two colorectal cancer drugs to include only patients whose tumor carries the wildtype KRAS gene, the FDA has refused to make such a recommendation absent prospective clinical trial data.

An FDA spokesperson told Pharmacogenomics Reporter this week that the agency "is currently working with Amgen and Imclone/Lilly to obtain adequate information to update the label of these two agents." Erbitux developer ImClone was acquired by Lilly last year.

Although "the advisory committee recommended inclusion of KRAS information in the package inserts of the two agents," the panel "did not comment on the specific label language or the placement of the KRAS information in the specific section of the package insert," the FDA spokesperson said. "This is negotiated and finalized between the FDA and the sponsor based on the available scientific and clinical data."

One reason the FDA is taking time to mull the clinical data on KRAS testing has to do with the false-positive rate of assays, according to Sudeep Basu, senior analyst in the healthcare division at the consultancy Frost & Sullivan.

"While KRAS testing does stratify patients as responders and non-responders, the responder population does not do so in totality. There are still some patients who are KRAS wild-type but do not respond to therapy as well," said Basu, who is currently leading projects focused on stratified medicine. "This can be one reason for some delay in obtaining clarity in the regulatory process."

The spokesperson from the FDA cited an article published in the Oct. 23 issue of the New England Journal of Medicine, in which only 12.8 percent of patients with wild-type KRAS status had a complete or partial tumor response to Erbitux. Conversely, 87.2 percent of the patients positively selected by KRAS did not respond to the therapy.

However, this study also showed that 28.3 percent of patients with KRAS wild-type alleles treated with Erbitux survived for one year while 20.1 percent of those with a mutated gene did not, the agency spokesperson pointed out.

Although the FDA has delayed issuing this recommendation, the scientific community has begun to accept KRAS as a predictive marker of response to EGFR-inhibiting treatments in metastatic colorectal cancer.

[ pagebreak ]

The American Society of Clinical Oncology last month issued its preliminary opinion on the matter advising doctors to test colorectal cancer patients for KRAS status prior to administering Vectibix and Erbitux. Based on the fact that as many as 40 percent of colon cancer patients have the KRAS mutation, ASCO said routine testing for KRAS mutations in patients with metastatic colorectal cancer could save the US health system up to $604 million per year from the cost of the drug cetuximab alone.

Despite the confusion regarding testing, BMS CFO Jean-Marc Huet told investors during its earnings call last week that the company "expects this market to be a growth driver long term as oncologists are able to identify those patients most likely to benefit based on KRAS testing." He did not say what role the test will play in this growth. BMS was not available to comment in time for publication.

However, without definitive guidance from the FDA on KRAS testing, drugmakers may need to consider alternative education and marketing approaches for their drugs.

"Given that ASCO has now put its weight behind the issue by releasing a provisional clinical opinion in favor of KRAS testing, it may make the environment more conducive for industry to reach out to physicians and reimbursement agencies more actively," Frost & Sullivan's Basu noted.

He suggested drug firms consider a two-pronged approach by increasing patient awareness for KRAS testing and actively engaging in discussions with physicians.

Without specific labeling language recommending or requiring KRAS testing, Amgen and BMS cannot directly recommend doctors to conduct KRAS testing before administering the drugs.

An Amgen spokesperson said if a doctor calls its information hotline asking about KRAS testing, the company can point to published studies in peer-reviewed journals.

According to one doctor who runs a personalized-medicine practice in New York and Connecticut, it might benefit the drug companies to train their sales reps to become more comfortable speaking to doctors in terms of subpopulations.

"Pharma likes to be able to say that everyone should take their drugs. That's what the [blockbuster] model is set up for," Steven Murphy, founder of HelixHealth and a doctor trained in medical genetics, told Pharmacogenomics Reporter. "Pharma reps need to get more nuanced."