College of American Pathologists
Printable Version



CAP Today




February 2012

Cobas BRAF V600 mutation test and Zelboraf Cobas BRAF V600 mutation test and Zelboraf

We appreciate the timely cover story about the joint FDA approval of vemurafenib (Zelboraf) and its companion diagnostic test, the Cobas 4800 BRAF V600 Mutation Test (“Cobas BRAF test”), and would like to address some of the issues raised in the article (December 2011, page 1).

The article says that this joint FDA approval “leaves an estimated 50 to 100 laboratories, which have been using LDTs to detect BRAF mutations for years without FDA approval, in a state of uncertainty.” However, this joint approval does not restrict laboratories from offering BRAF tests for clinical applications other than metastatic melanoma, where no approved tests exist.

A key issue is patient safety. The FDA ensures that the diagnostic tests it approves meet clinical and analytical validation requirements that assure clinicians and patients of their safety and effectiveness. Given this high standard, CAP-accredited laboratories should consider the statement in the FDA-approved label that an FDA-approved test be used for selecting vemurafenib patients as setting a standard that should apply to any tests in this context. Thus, it is not unreasonable to expect laboratories to modify techniques or testing where required by a drug’s package insert.

The article suggests that most labs were not ready for the rapid approval of vemurafenib and the Cobas BRAF test. The article noted that other indications have provided the bulk of test volume in prior years. However, a clear-cut indication for Cobas BRAF testing in melanoma did not exist prior to development of a highly effective targeted therapy. It seems reasonable to expect modification or initiation of new testing where it involves a companion diagnostic such as this, tied to the efficacy of a new therapy as demonstrated in a clinical trial. No other BRAF test has been shown to have clinical utility in selecting patients for treatment with vemurafenib. In addition, no reasonable barrier exists for labs to provide this test in the United States. The test is available now from many U.S. reference laboratories with turnaround times that are shorter than those of most other methods, inclusive of the overnight shipping needed.

The article also notes that some labs are using multiplexed assays that screen for several mutations at once, including BRAF mutations other than the predominant V600E (1799T>A) mutation. However, the clinical utility of these other mutations, such as V600K, has not been clearly demonstrated, given the small number of such patients treated to date with vemurafenib, and additional prospective clinical trials are required to establish the efficacy of the drug in tumors harboring these variant BRAF mutations. Thus, treating patients with these other mutations must be considered investigational at this time.1

The sample cohort was not selected based on specimen size, as suggested in the article, but consisted of consecutive samples received for screening in the BRIM-2 and BRIM-3 clinical trials. All specimens with less than 50 percent tumor content were macrodissected prior to BRAF mutation testing by both Cobas testing and Sanger sequencing. Thus, the relatively poor performance of Sanger in the studies cannot be ascribed to the testing of specimens with low tumor content. In fact, if the results of Sanger sequencing analysis had been used to select patients for the clinical trials, as many as 20 percent of patients who could have been treated with vemurafenib would have been precluded from treatment due to either false-negative or invalid test results.

Finally, the cost of the Cobas test was challenged in the article. The cost of the Cobas test, including DNA isolation, is comparable to that of other methods and comes with clinical and analytical validations to assure clinicians and patients of the test’s efficacy and safety, as established by randomized clinical trials.

In summary, the companion diagnostic test for vemurafenib meets FDA requirements for its particular intended use, as shown in prospective clinical trials (BRIM-2 and BRIM-3). Despite concerns raised in the article, the Cobas BRAF test demonstrated better performance than Sanger sequencing in selecting patients within those trials. Other characteristics of the test (analytical performance and clinical performance) have been described fully in two manuscripts now in press.2,3 The FDA’s linkage of vemurafenib to a well-validated companion diagnostic provides a safe and effective path to personalized medicine for patients and their treating physicians.


  1. Zelboraf U.S. Prescribing Information.
  2. Halait H, DeMartin K, Shah S, et al. Analytical performance of a real-time PCR-based assay for V600 mutations in the BRAF gene, used as the companion diagnostic test for the novel BRAF inhibitor vemurafenib in metastatic melanoma. Diagn Mol Pathol. In press.
  3. Anderson S, Bloom K, Vallera DU, et al. Multi-site analytic performance studies of a real-time PCR assay for the detection of BRAF V600E mutations in formalin-fixed paraffin-embedded tissue specimens of malignant melanoma. Arch Pathol Lab Med. In press.

Michael C. Dugan, MD
Chief Medical Officer

H. Jeffrey Lawrence, MD
Senior Director, Clinical Research Genomics and Oncology

Roche Molecular Systems Inc.
Pleasanton, Calif.

Doctors’ dining room Doctors’ dining room

John Plotz, professor of English at Brandeis University, in a recent essay discussed acedia in writers, monks, and laboratory scientists (Their noonday demons, and ours. New York Times Book Review. Dec. 25, 2011). The surgical pathologist, working in isolation, faced with trays and trays of slides, most with rather pedestrian and unexciting lesions, is also susceptible to acedia (defined as spiritual torpor, languor, or ennui). The mind easily wanders; the Internet is tempting. What is the market doing? Did I just doze off? Did I already dictate this case? What day is this?

The Benedictines addressed this problem by collective singing and shared meals. The new transformed/involved pathologist also needs some sort of “enforced sociability.” The seclusion of the diagnostic process can be tempered by a mid-morning and noontime trip to the doctors’ dining room. Caffeine and calories can be restorative, but the interaction with clinicians, even if contentious but most often collegial, is always stimulative and invigorating.

I recommend that pathologists take breaks and have their meals with the clinicians in the doctors’ dining room, not sequestered in their offices. Pathologists cannot be significant or involved if they are invisible. The doctors’ dining room or the surgeons’ lounge are where we learn what the clinicians require or deem important. It is where we get requests for consultations and where early warnings of problems in the laboratory may surface. It is where pathologists can assert their legitimate membership on the patient care team.

Richard E. Horowitz, MD
Clinical Professor
Department of Pathology
University of Southern California
Keck School of Medicine
Los Angeles