Context.—Mutation status of the KRAS gene identifies a distinct disease subtype of metastatic colorectal carcinoma that does not respond to antibody therapeutics targeting the epidermal growth factor receptor. This is currently the only validated marker in metastatic colorectal carcinoma with a clear implication in treatment selection. KRAS testing is widely accepted in clinical practice to guide metastatic colorectal carcinoma therapeutic decisions, and there are many commercially available platforms to perform the test.

Objective.—To evaluate the critical role of pathologists in the full implementation of KRAS testing by optimizing tumor tissue collection and fixation procedures and by choosing testing technologies and reliable Clinical Laboratory Improvement Amendments of 1988–certified laboratories to perform the tests.

Data Sources.—Prospective clinical trials, retrospective studies, and quality assessment and survey reports were identified in the following databases: PubMed, American Society of Clinical Oncology Proceedings (American Society of Clinical Oncology Annual Meeting and Gastrointestinal Cancer Symposium) and European Society for Medical Oncology Proceedings (Annals of Oncology European Society for Medical Oncology Congress and Annals of Oncology World Congress on Gastrointestinal Cancers).

Conclusions.—More bona fide standards are needed to address the variety of available test methods, which have different performance characteristics including speed, sensitivity to detect rare mutations, and technical requirements. Refined standards addressing timing of KRAS testing, laboratory performance and accuracy, quality assurance and control, proper tissue collection, and appropriate result reporting would also be greatly beneficial. Pathologists should be aware that the amount of information they need to manage will increase, because future trends and technological advances will enhance the predictive power of diagnostic tests or the scope of the biomarker panels tested routinely across tumor types.

In this exciting new era of targeted cancer therapy guided by molecular biomarkers, the work of the pathologist goes beyond initial tumor diagnosis and staging. The role of the pathologist in ensuring quality molecular diagnostic testing may shape important therapeutic choices for oncologists and patients. As gatekeepers of tissue collection and molecular testing, pathologists will be instrumental in optimizing the practical implementation of oncology biomarkers.

This need for continuing biomarker optimization is particularly apparent in the case of KRAS testing for metastatic colorectal carcinoma (mCRC). Strong phase 2 and 3 clinical data clearly indicate that mCRC patients having tumors with certain KRAS mutations (codon 12, 13) do not benefit from anti–epidermal growth factor receptor (EGFR) antibody therapy.15 These pivotal studies led to the inclusion (by the US Food and Drug Administration and the European Medicines Agency) of KRAS mutation analysis as a criterion to select patients with mCRC eligible for treatment with cetuximab or panitumumab.6,7 Continuing optimization and standardization of tissue collection, assay methods, and results reporting are critical for the overall quality of KRAS testing. The accuracy and timeliness of KRAS diagnostics are of paramount importance in directing mCRC patients to the most appropriate therapies.

In this article, we will review the clinical trial results leading to the validation of KRAS exon-2 mutations as a treatment-specific predictive biomarker. In addition, we will review current practice guidelines for KRAS testing that are important for the pathologist, including how tissues should be collected and what testing platforms are currently available as well as their distinguishing features and advantages/disadvantages. Finally, we will discuss practical issues that need to be addressed to drive quality improvements in KRAS testing going forward. These issues include timing of testing (after surgery versus recurrence diagnosis versus just prior to anti-EGFR therapy), tissue and KRAS testing quality control, clear and accurate reporting of results, and the inclusion of KRAS mutations as part of a panel of informative biomarkers. A greater appreciation of these key issues by the pathologist is important in realizing the full potential of KRAS diagnostic testing for metastatic colorectal cancer patients.

Pathologists are part of the multidisciplinary mCRC treatment team consulting with oncologists to help guide the appropriate use of the KRAS test and the subsequent interpretation of results.8 As such, pathologists should understand the clinical significance of KRAS mutation status and how the test influences treatment options for the targeted therapies cetuximab and panitumumab. They should also appreciate how KRAS testing has clearly met high validation standards for a clinically actionable biomarker for mCRC.

As described in the Perspectives on Emerging Technology report8 from the College of American Pathologists (CAP), levels of clinical evidence to use for decision making (eg, use of a molecular marker as a diagnostic test) are based on a classification system developed by the U.S. Preventive Services Task Force9 in 1996. Specifically, the highest level of evidence, level I, comes from randomized, controlled trials, whereas level II evidence comes from uncontrolled single-arm studies. Level III evidence refers to the opinions of respected authorities or reports of expert committees.

In the case of KRAS mutations, single-arm studies1014 that retrospectively examined anti-EGFR antibody treatment response in relation to marker status in various clinical settings provided strong level II evidence that patients with KRAS mutations in codon 12 or 13 do not benefit from anti-EGFR therapy with cetuximab or panitumumab. Importantly, even stronger level I evidence regarding KRAS mutations and treatment benefit with anti-EGFR antibodies has emerged from multiple phase 3 randomized, controlled trials1,3,1517 investigating both cetuximab and panitumumab in various clinical settings (first/second/third-line, single agent, combination therapy with different cytotoxic regimens; summarized in Table 1). It may be important to note that, in these trials, tissue collection was not always mandatory, and KRAS mutation testing in particular was not prespecified in their protocol; therefore, ascertainment rates were variable with the inherent risk for lack of representativeness of the evaluable populations (relative to the original intent-to-treat cohorts). Nonetheless, the results regarding the effect of KRAS mutations on treatment benefit have been consistent regardless of ascertainment rate, lending definitive support to this finding.

Table 1. 

Pivotal KRAS Data from Randomized Clinical Trials

Pivotal KRAS Data from Randomized Clinical Trials
Pivotal KRAS Data from Randomized Clinical Trials

The lack of benefit from anti-EGFR therapy in KRAS mutant patients was first evident in single-agent, randomized, controlled trials with cetuximab or panitumumab treatment of mCRC in second- or third-line settings. In a cetuximab versus best supportive care study3 in second- or third-line mCRC patients, statistically significant improvements in response rate (12.8% antibody arm versus 0% control arm), median progression-free survival (3.7 months antibody arm versus 1.9 months control arm), and median overall survival (9.5 months antibody arm versus 4.8 months control arm) were limited to patients with KRAS wild-type tumors (Table 1).

Similarly, in a panitumumab versus best supportive care study1 in chemotherapy-refractory mCRC, statistically significant improvements in response rate (17% antibody arm versus 0% control arm) and median progression-free survival (12.3 weeks antibody arm versus 7.3 weeks control arm) were limited to patients with KRAS wild-type tumors (Table 1). The availability of a control arm (not treated with the anti-EGFR agent) in these 2 trials allowed researchers to establish that the effect of KRAS mutations was specifically predictive, because the marker did not have any prognostic effect in patients assigned to the best supportive care arms. Following these studies of anti-EGFR agents as monotherapy, a series of phase 3 trials4,5,17 for both cetuximab and panitumumab has found the same predictive effect for KRAS mutations when these agents are combined with first-line chemotherapy, where again treatment benefit was limited to patients with KRAS wild-type tumors (Table 1).

Consistent findings from all of these studies prompted an American Society of Clinical Oncology provisional clinical opinion18 that clearly stated (1) the correlation between KRAS mutations in either codon 12 or 13 and a lack of response to anti-EGFR therapy in patients with mCRC, and (2) the evidence of improved tumor response and survival (progression-free survival for panitumumab and/or overall survival for cetuximab) in response to anti-EGFR therapy in mCRC patients with wild-type KRAS (codon 12, 13) versus mutant KRAS. The provisional clinical opinion then recommended that all mCRC patients who are candidates for anti-EGFR therapies should have their tumors tested for KRAS mutations in a Clinical Laboratory Improvement Amendments of 1988–accredited laboratory, and if a KRAS codon 12 or 13 mutation is detected patients should not receive anti-EGFR antibody therapy.18 

Pathologists should clearly understand that the strong clinical evidence just described led to the regulatory consideration (Food and Drug Administration and European Medicines Agency) for KRAS testing in advanced mCRC. Prescribing information (or labeling) changes for panitumumab and cetuximab indicate that these drugs are not recommended for patients having colorectal carcinoma (CRC) tumors with KRAS mutations.6,7 ,KRAS testing has become part of standard practice guidelines as stated by the National Comprehensive Cancer Network (NCCN) and the American Society of Clinical Oncology. Studies19 using physician questionnaires have indicated KRAS mutation testing has become a widely adopted diagnostic procedure used to guide therapeutic decisions for mCRC patients. Given these developments, it is critical for the pathologist to understand current practice guidelines regarding tissue collection and assay technologies and results reporting to ensure accurate, high-quality KRAS testing.

Recommendations for Tissue Collection and Fixation

The pathologist plays a critical role in identifying appropriate sources of mCRC tumor cells for the analysis of KRAS mutation status. Although the tissue collection process has not been standardized, general recommendations have been made by the NCCN and CAP.

The NCCN guidelines for colon cancer20 recommend that KRAS testing can be performed on either primary tumor or metastatic tissue. In cases of metastatic disease in patients with recurrence, the general consensus is that the initial primary tumor specimen obtained at the time of surgical resection is appropriate to provide informative and conclusive results (even if a long period of time has elapsed). In the case of colorectal patients presenting with metastatic disease at first diagnosis, KRAS testing can be performed on a specimen isolated from metastatic tissue.21 These recommendations are based on evidence suggesting that KRAS mutations may occur early during the development of the subset of colorectal tumors harboring them; therefore, there is an expectation for marker status stability during the course of the disease and nearly total concordance between primary and metastatic tissues.22 For most cases of relapse to metastatic disease (>80%), pathologists can go back to archival primary tumor specimens for KRAS testing without the need for additional biopsy.

The College of American Pathologists has provided practical guidelines for tumor specimen collection in the Perspectives on Emerging Technology report.8 However, at the current time, there is no consensus recommendation for standardization of morphologic sample assessment and approval prior to DNA extraction. Genotyping laboratories continually urge pathologists to select CRC blocks that have no or minimal necrosis, which is achievable in the vast majority of cases. This area should also be enriched in intact tumor nuclei; more than 50% of all nuclei present is a typical benchmark (also quite feasible in most cases), although successful testing may be possible with less-enriched samples. Guidelines18 recommend that extracted samples be either provided fresh, stored in a preservation solution (eg, RNAlater [Qiagen, Valencia, California]); rapidly frozen and stored frozen; or fixed in neutral-buffered formalin (the preferred fixative) and embedded in paraffin. The generation of formalin-fixed paraffin-embedded (FFPE) blocks is perhaps the most common method used; as a result, FFPE sections are the most readily available clinical specimens for KRAS testing. The number of sections needed per patient case for standard pathology and molecular testing may depend on the area of the tumor (fewer sections may be needed for more compact tumors) and the complexity of the tumor histology.23 

Pathologists evaluate tumor tissue sections for the frequency of tumor cells using standard histologic techniques to ensure that a sufficient quantity of extracted tumor cell DNA is available for KRAS testing. Although this in itself is a routine step, accurate control of the basic histologic specimen specifications is becoming a critical aspect of tissue sampling as the sensitivity of available assay technologies increases. The frequency of tumor cells in the specimen and the general quality of the collected tissue and extracted DNA have important implications for the accuracy and the sensitivity of the KRAS testing method used. A true standardization of the testing procedure will require a critical appraisal of our sample quality control processes and reporting, an issue discussed later in this manuscript.

Established Technologies for Detecting KRAS Mutations

Despite the wide acceptance of KRAS mutation testing as a diagnostic tool for mCRC, a uniform consensus on optimal test methods is lacking. The NCCN guidelines20 mandate KRAS testing for CRC at diagnosis of metastatic disease, but do not recommend particular assay technologies. Comparability studies24 have shown that for certain testing methods there is often a lack of concordance with gold-standard sequencing results. To make a well-informed choice of a testing method, the pathologist needs a detailed understanding of how different tests compare in terms of sensitivity, accuracy, DNA amount and quality needed, length of time to get results, and cost.

The most commonly used methods for analyzing KRAS mutations can be categorized as direct sequencing, polymerase chain reaction (PCR)–based detection, and allele-specific hybridization. Key features and a comparison of these technologies are shown in Table 2. It is important to note that all technologies involve an initial PCR amplification of the KRAS target sequences. Direct sequencing of PCR products containing codon 12 and 13 regions has been traditionally considered the reference system against which other methods are compared. Traditional Sanger dideoxynucleotide cycle sequencing has the advantage of detecting all clinically relevant KRAS gene mutations, but it is relatively slow, costly, labor intensive, and technically challenging. Importantly, traditional cycle sequencing lacks the sensitivity achieved by other methods; the relative amount of KRAS mutant DNA needed for detection in a sample is 20% to 50% (Table 2).25 

Table 2. 

Established Technologies for KRAS Mutation Analysis

Established Technologies for KRAS Mutation Analysis
Established Technologies for KRAS Mutation Analysis

Pyrosequencing has led to substantial improvements in sensitivity and more rapid generation of results. This technology uses the release of pyrophosphate during the sequencing reaction as a detection method leading to greater sensitivity, higher throughput, and faster results than traditional sequencing, although it is more expensive. Unlike traditional Sanger sequencing, the short sequencing reads associated with pyrosequencing allow for successful KRAS testing of poorer-quality, partially degraded DNA samples (Table 2).26 

But perhaps most of the KRAS testing methods used in clinical practice are based on direct analytic use of PCR technology. KRAS mutations can be detected during the PCR amplification process itself or afterwards using different analytic techniques such as high-resolution melting analysis, high-performance liquid chromatography, or restriction fragment length polymorphism. High-performance liquid chromatography and restriction fragment length polymorphism are technically challenging, are less common in clinical practice, and will not be discussed in any detail. The underlying principle of high-resolution melting analysis is that the presence of mutations in KRAS impacts the melting temperature of double-stranded DNA. Melting point curves can be used to distinguish wild-type from KRAS mutant alleles in tumor DNA samples using relatively rapid and inexpensive methods. However, the need for sequencing to confirm that abnormal melting curves are due to KRAS mutations (and not another alteration) may limit the use of this test in clinical practice.26 

Several robust technologies have been developed to incorporate allele-specific probes and/or allele-specific primers during the PCR amplification of the KRAS gene. The amplification-refractory mutation system (ARMS) detects mutations using primers with single-base alterations to selectively amplify mutant alleles, even if they are at a low concentration in a tumor DNA sample. Thus, ARMS reaches a higher level of sensitivity than sequencing, detecting as low as 1% relative presence of KRAS mutant DNA. ARMS results can be generated rapidly (<2 days) from paraffin sections. However, unlike sequencing, ARMS provides only a categoric result (positive/negative) for a single mutation per reaction. Recently, ARMS has been further optimized as a quantitative PCR platform.26 

The ARMS technology has been combined with bifunctional fluorescent primer/probe molecules (Scorpions technology; DxS, Manchester, United Kingdom) in a quantitative PCR platform. This combination forms the basis for the TheraScreen KRAS testing kit, initially developed by DxS and now commercialized by QIAGEN Benelux BV (Venlo, the Netherlands). TheraScreen is a commercially available test; it is an approved in vitro diagnostic test in Europe and is reaching widespread acceptance in the United States. Disadvantages of TheraScreen include its cost, a requirement for more tissue than other methods, and the limited testing of only 7 of the most common mutations.26 Advantages include high sensitivity, the ability to generate results with poor quality/partially degraded DNA samples, and a closed-tube 1-step process.26 In contrast, direct sequencing uses a multiple step process that involves direct handling of PCR products, a potential source of contamination. A basic level of quantitative PCR expertise is required, as opposed to higher levels of expertise needed for direct sequencing.23 

A third unique group of KRAS tests uses the principle of allele-specific hybridization followed by a detection step. In one method, oligonucleotides complementary to wild-type KRAS or one of the mutant alleles are allowed to hybridize to KRAS gene PCR products, and mutations are detected by changes in melting temperature (Table 2). Although rapid, this method is not very sensitive for rare mutant alleles, and it is expensive, requiring specialized equipment and analysis software.26 In another variation of allele-specific hybridization called PCR clamping and melting curve technique, KRAS mutations are detected by melting curve analysis of PCR products after hybridization of fluorescence-labeled mutation-specific sensor probes.17 

Another hybridization method with improved performance is the commercially available and European Medicines Agency–approved in vitro diagnostic–certified ViennaLab assay (ViennaLab Diagnostics GmbH, Vienna, Austria). KRAS mutations are detected by hybridization of biotinylated PCR products to an array of allele-specific oligonucleotides immobilized on membrane strips. Mutation-specific hybridization events are detected by an enzymatic color reaction similar to that used in ELISA methods. With speed and sensitivity comparable to TheraScreen (1% KRAS mutant frequency), the ViennaLab assay has advantages of being less expensive and inclusive of more mutations (13 rather than 7 for TheraScreen).25 In contrast to TheraScreen, which uses a standard real-time PCR machine, ViennaLab requires a thermocycler incubator.25 

Comparability studies have shown that many of the most common commercially available technologies are accurate as measured by concordance with direct-sequencing results from a reference laboratory.24,27 Therefore, the choice of KRAS testing methodology may be driven more by the availability of particular assays, relative costs, physician familiarity, and the turnaround time needed for results. DNA sample quality and tumor DNA content are important factors for the pathologist to consider when choosing a technology for KRAS testing. These factors may lead to discrepancies between sequencing and quantitative PCR results.23 

With the increasing sensitivity of KRAS testing methods, the detection of mutations present at a low relative level in mCRC tumors is becoming optimized. In a recent retrospective study28 analyzing 111 mCRC patient samples, the use of KRAS analysis methods with sensitivities less than 10% (10% mutation frequency) led to the additional identification of 13 KRAS mutations that were not detected by direct sequencing. All 13 newly identified mutations occurred in cetuximab or panitumumab nonresponders.28 This study highlights the need for future KRAS testing sensitivity standards to optimize the identification of all anti-EGFR nonresponders.

Next-Generation Sequencing Technologies

The future of personalized medicine for mCRC will likely include the use of high-throughput next-generation sequencing technologies to examine hundreds of potentially clinically actionable alterations in cancer-related genes, including KRAS. A recent next-generation sequencing study29 retrospectively analyzed 75 cancer tissue samples (from various tumor types including colon) using a next-generation diagnostic test currently being developed by Foundation Medicine, Inc (Cambridge, Massachusetts). This test sequences 2574 exons representing 176 genes using the Illumina HiSeq 2000 (Illumina Inc, San Diego, California) platform. Results were 100% concordant with conventional, single-gene analyses reported previously by reference laboratories for KRAS, BRAF, and EGFR. The test identified additional novel and known mutations, many of which could be clinically relevant.29 In order for next-generation sequencing to reach utility in clinical practice, technical challenges will have to be met. Those challenges include extensive infrastructure requirements (including expensive equipment) and the need for high-level computational expertise. Next-generation sequencing provides the potential to fully characterize the genome-wide mutational profile of a tumor in one single assay, and the challenge will be to identify clinically relevant gene alterations that could help drive optimal therapeutic decisions for mCRC patients.

“Liquid Biopsies:” Measuring KRAS Mutations in Patient Blood or Plasma

A future trend in KRAS diagnostics may be the measurement of mutations using tumor DNA from patient plasma or circulating tumor cells from whole blood. Earlier studies investigated molecular testing of plasma (or stool) as a means for disease screening or early detection; those studies, although promising, revealed issues with the sensitivity and, particularly, with the specificity of those tests carried out in nontumor samples, yielding unacceptable rates of false-positive results (ie, KRAS mutations found in the serum of healthy donors).3033 If optimized, however, this liquid-biopsy approach could be used to rapidly generate real-time KRAS data, with results generated as close as possible to the time of anti-EGFR treatment initiation. In a recent study, 78% of KRAS mutations detected in mCRC primary tumor samples were also detected in plasma. Quantitative levels of plasma KRAS mutations (pKRAS) also correlated with disease control in mCRC patients receiving third-line cetuximab plus irinotecan therapy.34 Preliminary studies35 have indicated that circulating tumor cells captured from blood samples (using the CellSearch [Veridex LLC, Raritan, New Jersey] or circulating tumor cell chip platforms) might be amenable to the analysis of molecular markers including KRAS. Because KRAS mutations were measured using an artificial system of tumor cells spiked into human blood to mimic real circulating tumor cells, more studies are needed to demonstrate the potential for clinical utility. In total, although the sensitivity and specificity levels are suboptimal for current clinical practice, blood sampling for KRAS mutations represents a promising technology that could simplify considerably the sample-collection process.

Quality Control: Tissue Collection and Testing Laboratories

There are several important details that need to be considered by the pathologist to ensure the high-quality tissue collection and laboratory selection required for accurate KRAS testing. Although the output in KRAS mutation testing is a simple classification, uniformity in reporting is lacking, and details such as the percentage of tumor cells in a specimen, the use of compatible fixation methods, and the choice of a qualified and well-validated laboratory are often overlooked in final reports.

Formalin-fixed paraffin-embedded tumor blocks of primary CRC tumors are reviewed by the pathologist for tumor cell content using hematoxylin and eosin staining, a critical initial step in KRAS molecular testing. High variability has been observed in the estimation of percentage of tumor cells in hematoxylin and eosin–stained paraffin sections.36 Histologic analyses performed in a large European quality assessment study36 revealed that for some samples laboratories reported 10% to 20% tumor cell frequency whereas others reported 90% to 100%. Interestingly, there was no correlation between the percentage of tumor cells measured and the level of the section in the FFPE block, suggesting that observer variability was the major determinant of these discrepancies.36 These observations illustrate how molecular testing practices are revealing persisting shortcomings in the standardization of pathologic assessments; although centralized training schemas can be proposed as one possible approach to increase uniformity and reliability, it is clear that this is still a pending issue.

Although a 50% benchmark is common, pathologists have used neoplastic cell frequency cutoff levels of up to 70% for choosing tumor FFPE sections for further immunohistochemistry, cytogenetic, and molecular analyses.28 In cases where samples do not meet this criteria, microdissection or macrodissection methods can be performed to reduce the presence of nonneoplastic tissues.28 Enrichment of tumor cells can limit “competition” of DNA from other surrounding cell types such as fibroblasts or endothelial cells in PCR amplification reactions, reducing the potential for KRAS testing errors.26 Pathologists can also address the issue of low tumor cell content by ordering a test with a high degree of sensitivity to detect rare KRAS mutations. Standards are needed based on a better understanding of how the frequency of neoplastic cells translates to KRAS mutation detection sensitivity. Standards could also lead to more consistent ascertainment of the initial tumor cell frequency.

Pathologists should consider additional tissue collection standards, including how the FFPE block should be selected and what tumor cell frequency is sufficient before enrichment methods need to be considered. Pathologists could also benefit from standards dictating how tissues should be submitted to laboratories, including the number and size of sections (eg, two 20-µm sections or one 40-µm section) and whether samples should be submitted on glass slides or stored in Eppendorf tubes.

Recent technical advances have led to an improved ability to use DNA from fixed FFPE tissue samples for molecular testing (eg, KRAS). However, formalin fixation can damage DNA and introduce artificial mutations,26 particularly in samples that are overfixated, which may result when exposing small tissue samples to formalin overnight.37 Sampling of sufficient cellular material is therefore critical when using formalin fixation. Nonformalin solutions may also be problematic: for example, acidic solutions (such as Bouin) and decalcifying agents (such as ethylene diamine tetraacetic acid and ethyleneglycoltetraacetic acid) are not well suited to the recovery and preservation of nucleic acids.37 Heavy metal fixatives, including B5 and Zenker, also contribute to DNA degradation and are thus not useful for samples that will be subsequently analyzed by PCR. Recommendations from the Perspectives on Emerging Technology report8 and the NCCN20 suggest neutral-buffered formalin fixation for KRAS testing. In multiple quality assessment studies,26,36 the use of nonbuffered formalin or nonformalin fixation methods by certain laboratories led to poor DNA yield/quality and failures to generate definitive results.

The pathologist's choice of a KRAS testing laboratory is a critical one because the accuracy and timeliness of KRAS genotyping have important treatment implications for mCRC patients. The NCCN guidelines20 recommend that KRAS testing only be performed in laboratories that are certified under the Clinical Laboratory Improvement Amendments of 198838 as qualified to perform highly complex clinical laboratory tests. Pathologists can also identify CAP-accredited laboratories where regular inspections evaluate key performance areas such as the techniques being used, specimen handling, test reporting, and laboratory equipment.39 

The Perspectives on Emerging Technology report from CAP indicates that pathologists “should be able to carefully evaluate the KRAS testing technology utilized and processes employed to ensure confidence in the results.” 8 Results from recent quality assessment studies provide some important perspectives for pathologists to consider in their evaluation of testing methods and quality. By comparing KRAS test results of known standard mCRC samples sent to different regional and reference testing laboratories, the European quality assessment study32 assessed overall laboratory quality and performance. Accuracy was clearly an issue; just 70% of the European laboratories correctly identified all KRAS mutations.36 At least 1 in 10 samples were incorrectly genotyped in more than 30% of laboratories, and there were additional technical failures where no results were generated.36 In another quality assessment study, discrepancies across 6 laboratories were observed. Some issues and potential areas of improvement identified were the quality of samples (eg, appropriate fixation methods) and the sensitivity of methods used to detect samples containing a low frequency of KRAS mutations.26 Specific standards for fixation and test sensitivity would help address these issues.

Pathologists should look for a stringent level of internal and external quality assessment and quality control in the laboratories that they choose for KRAS testing. The aforementioned studies underscore the need for more quality assessment studies across regional and reference laboratories (particularly in the United States, where reports are lacking), and regular laboratory participation in proficiency strategies aimed at improving KRAS testing quality. To this end, pathologists should be aware of the KRAS proficiency test from CAP, which provides a standardized, external measure of testing accuracy.39 New and established laboratories should consider routine participation in external proficiency tests by measuring samples of known KRAS mutation status and comparing results with a reference laboratory. Accuracy is particularly critical when rapid treatment decisions depend on KRAS testing results.

Reporting of KRAS Testing Results: Need for Standardization?

Although KRAS test reporting has not been standardized, accepted guidelines for reporting results of molecular tests are based on International Organization for Standardization 15189:2007 requirements for medical laboratories and a CAP document40 providing additional guidance. Some of the key items that should be included on a test report are sample collection details (eg, FFPE specimen), percentage of tumor cells, specific reason for testing, genotype, correct nomenclature, an interpretation of the result, list of the KRAS mutations tested, and the testing methodology used. Based on the recent external quality assessment36 involving reference and community laboratories in Europe, deficiencies in KRAS test reporting are still a concern.

Common reporting deficiencies include no list of KRAS mutations tested, lack of referral to anti-EGFR therapy if the result is positive, and incorrect nomenclature describing the mutation.36 Some of these reporting issues are critical for the interpretation of results and could be clinically impactful. For example, if no mutations are found, the pathologist must know which mutations were screened because different technologies assess different mutations. A negative result from a test that covers only a subset of clinically relevant KRAS mutations does not fully inform treatment decisions.

Conversely, a positive test result must indicate the specific mutation sites, because current standards for treatment eligibility/ineligibility refer only to mutations found on the KRAS gene codons 12 and 13. However, these standards may be evolving: recent evidence41,42 suggests that the G13D transition, currently included in the panel of mutations excluding a patient from cetuximab treatment, may not be predictive of a lack of benefit. Moreover, a paucity of data regarding the biomarker potential of the rare codon 61 mutation excludes this marker from any practical consideration at this point, although available evidence43 suggests that these mutations are indicative of cetuximab resistance. Thus, clinicians unfamiliar with the various mutations and the latest clinical data related to them may find these reporting details somewhat confusing, if provided as a mere list without context. A clear note should be made of the clinically actionable items: simply, whether the tumor sample has a wild-type or mutated KRAS that warrants specific treatment choices. There is a compelling need to establish standards so that KRAS results, together with their clinical context and therapeutic significance, can be clearly and accurately communicated from the laboratory to the pathologist and subsequently to medical oncologists and patients.

The Timing of KRAS Testing in mCRC: At Diagnosis or Relapse?

There are currently no standards for the timing of KRAS testing in mCRC. The results of the test are required in order to assign a patient to anti-EGFR therapy. Because current data support the use of these agents as first or later lines of therapy, the choice of when to perform the test is left at the discretion of the treating oncologist. As a contrasting example, the wide acceptance of trastuzumab as adjuvant therapy for patients with HER2+ breast cancer has led to standards requiring immediate tumor molecular testing of the estrogen and progesterone receptors and HER2 in all newly diagnosed patients to plan for the immediate therapy.44 In colon cancer, where adjuvant therapy is not a universal option, and where, when used, it is not dependent on molecular markers, biomarker testing is not always seen as an immediate priority.

KRAS testing is currently not part of the standard pathology workup for colon cancer. Often, medical oncologists request KRAS testing late in the treatment course, to determine whether their patients with mCRC are eligible for third-line therapy with the anti-EGFR agents panitumumab and cetuximab. The timing of KRAS testing in clinical practices was investigated in a questionnaire study19 of KRAS testing trends in Asia, Europe, and Latin America. Investigators found that patients receiving second- or third-line therapy were more likely to have a KRAS mutation test (70% and 75%, respectively), perhaps reflecting the approved use of anti-EGFR antibodies for third-line use.19 The fact that 64% of first-line mCRC patients were tested for KRAS mutations even though anti-EGFR therapy may not have been the choice at that time suggests that oncologists were planning for subsequent treatments after the possibility of first-line treatment failure.

Routine, standardized KRAS testing at initial workup for mCRC would minimize potential delays (waiting period for the results) at the time of progression or treatment failure. From a principle standpoint, medical oncologists and pathologists could consider performing KRAS testing on all colon cancer patients at the time of surgery, so that in the event of relapse, KRAS genotype information would be universally and immediately available to guide treatment decisions. However, the cost-effectiveness of this approach in practice could be questioned. A middle-of-the-road approach would be to follow a standard practical rule, stratifying patients undergoing surgery according to their relapse risk (based on T and N staging), and setting a fixed relapse risk level that would warrant proactive KRAS testing, given the likelihood of those patients presenting with metastatic disease later on. Higher-risk patients with T2, stage III, or high-risk stage II mCRC tumors with transmural invasion could be tested for KRAS mutations after diagnosis and staging because there is a very significant chance the KRAS data will be needed.

Another approach that could reach that balance is to implement testing for all patients at the time of diagnosis with metastatic disease, which is the recommendation issued by the NCCN Guidelines Panel.20 Both these approaches ensure that the information about KRAS mutation is part of any mCRC clinical file from the initiation of therapy for metastatic disease, which may optimize the treatment planning process as well as streamline the logistics of patient management. In any case, standards are needed to strike the right balance between unnecessary testing and securing the timely availability of KRAS data.

Additional Biomarkers for mCRC: The Panel Approach

This review focuses on the practical implementation of KRAS testing in mCRC, but, as the number of clinically important molecular markers grows across tumor types, KRAS mutation becomes one of many markers relevant in oncology. One emerging trend in CRC reflected in several recent reports is the comprehensive evaluation of several biomarkers to enhance patient selection, given the still significant proportion of mCRC patients with wild-type KRAS tumors that do not respond to anti-EGFR therapies. Alterations in several genes downstream from KRAS in the EGFR signaling pathway (including BRAF, NRAS, PIK3CA, and PTEN) have been reported4547 that might be associated with lack of benefit from anti-EGFR agents in patients with KRAS wild-type tumors. Multiple testing could therefore narrow even further the population of patients eligible for anti-EGFR therapy. Some preliminary retrospective analyses45,48 suggest that nonresponders can be identified by alterations in BRAF, PIK3CA, and PTEN and that patients who are “quadruple negative” for these mutations (including KRAS) may have the highest probability of response to anti-EGFR therapies. Recent data47 suggest that the simultaneous measurement of all 4 markers could identify up to 70% of anti-EGFR nonresponding patients with mCRC. Unfortunately, although these reports explore the value of nested testing to enrich the patient population in responders, there is a lack of well-controlled data (similar to that available for KRAS mutations) confirming that the effect observed for some of those individual markers is truly predictive and not prognostic. Additional data from larger controlled prospective studies are needed before the use of this multi-marker panel can be recommended for clinical practice.

One such marker, for example, is BRAF V600E mutation status, which, in a retrospective analysis49 of mCRC patients treated with either panitumumab or cetuximab, correlated with a lack of treatment response and significantly shorter progression-free and overall survival compared with wild-type BRAF. However, clinical evidence from randomized, controlled trials2 indicates that BRAF mutations have negative prognostic value for mCRC and define a subset of CRC patients who respond poorly to both chemotherapeutic and targeted agents.

In response to the accumulation of molecular markers important in oncology, pathology practice protocols may opt to operate with single requisitions that include a panel of key markers applicable across tumor types (Table 3). Arguments for this approach are the procedural streamlining and the inclusive concept of generating a patient-specific tumor biomarker data set that would be available to guide current and future treatment decisions. An argument against the panel approach is that it would generate irrelevant information because many markers define treatment decisions for one tumor type or therapeutic approach but not others, and the indiscriminate listing of information for tumors for which the information is not relevant could foster confusion. KRAS, EGFR, and BRAF are excellent examples (Table 3): indeed, although the importance of KRAS mutation status in predicting patient response to anti-EGFR antibodies in mCRC is unambiguous, no apparent correlation between KRAS mutation and cetuximab benefit has been observed in non–small cell lung cancer.50,51 Conversely, KRAS mutation in non–small cell lung cancer has been investigated5255 as a marker for resistance to another anti-EGFR therapeutic class, the tyrosine kinase inhibitors, with conflicting results. Because tyrosine kinase inhibitors and EGFR-targeting antibodies differ in their molecular action, it is perhaps not surprising that the biomarkers that define their activity are divergent; further exemplifying this point, mutations in the EGFR gene clearly mark non–small cell lung cancer tumors for which tyrosine kinase inhibitors are the preferred option, but data have not shown an obvious relationship between EGFR mutation and patient response to anti-EGFR antibodies in non–small cell lung cancer or in any other tumor type.50,51 

Table 3. 

Key Solid Tumor Oncology Biomarkers With Implications in Clinical Practice

Key Solid Tumor Oncology Biomarkers With Implications in Clinical Practice
Key Solid Tumor Oncology Biomarkers With Implications in Clinical Practice

Another example of a tumor-specific biomarker is mutated BRAF, which for the moment appears to specifically inform treatment decisions in melanoma,56 but not in CRC (as described above) or in other tumor types. The situation becomes more complex if agents against the identified aberration are available or under investigation, but no/limited data are available in the tumor type where the mutation has been identified. Despite the fact that BRAF mutations are present in up to 10% of mCRC tumors, a small study57 of vemurafenib in 20 mCRC patients selected for BRAF mutations resulted in just a 5% response rate. Small sample size and questions about drug exposure (pharmacokinetic values were 20% lower than in melanoma studies) should prompt additional studies.57 Therefore, it is possible that future clinical studies could point to a role for BRAF mutations in predicting mCRC patient response to vemurafenib, but current data do not warrant any action.

A comprehensive panel approach that could be applied specifically to mCRC patients, or more broadly across all cancer types, could lead to cost and work efficiencies compared with 1-marker-at-a-time tests. In particular, efficiencies could arise from the application of a single technology (measuring many molecular markers) to identical DNA samples from the same tumor sections. In the future, next-generation sequencing technologies methods could be used to rapidly screen a large panel of markers from a single tumor DNA sample. Reporting and filing processes could become more streamlined when multiple markers are measured simultaneously using the same technology. Another potential advantage is that data for a particular marker, which may not be currently informative for a particular tumor type or clinical setting, will be available for a patient if new supporting data emerges in this rapidly evolving field. It will be up to oncologists and pathologists to decide whether the all-inclusive panel approach will provide excessive data (much of which would not be currently actionable) or whether it is worth having as much data as possible to guide future treatment decisions. Technological advances allowing the rapid concomitant evaluation of many markers, leading to cost and technical efficiencies, might make this decision easier.

KRAS mutation testing is now widely accepted in clinical practice to guide therapeutic decisions regarding the use of anti-EGFR antibodies in mCRC. Many commercially available platforms and testing laboratories are now available, but areas in which improvements can be made have been identified. Pathologists can help develop standard practices that maximize the number of patients for whom timely diagnostic information becomes available in a cost-effective manner. Pathologists can also drive improvements in the overall quality of KRAS testing by developing standards for tumor tissue collection, for laboratory quality assurance programs, and to improve reporting of results. Pathologists must be familiar with the pros and cons of the latest assay technologies to make informed decisions about which KRAS test to use. Future trends include the potential for panel testing of multiple biomarkers using new technologies such as next-generation sequencing and the possibility of real-time measurements of KRAS mutations in patient blood or plasma. Pathologists should appreciate and embrace these developments during the rapid evolution of KRAS testing.

1.
Amado
RG
,
Wolf
M
,
Peeters
M
,
et al.
Wild-type KRAS is required for panitumumab efficacy in patients with metastatic colorectal cancer
.
J Clin Oncol
.
2008
;
26
(
10
):
1626
1634
.
2.
Van Cutsem
E
,
Bajetta
E
,
Valle
J
,
et al.
Cetuximab plus irinotecan, fluorouracil, and leucovorin as first-line treatment for metastatic colorectal cancer: updated analysis of overall survival according to tumor KRAS and BRAF mutation status
.
J Clin Oncol
.
2011
;
29
(
15
):
2011
2019
.
3.
Karapetis
CS
,
Khambata-Ford
S
,
Jonker
DJ
,
et al.
K-ras mutations and benefit from cetuximab in advanced colorectal cancer
.
N Engl J Med
.
2008
;
359
(
17
):
1757
1765
.
4.
Douillard
JY
,
Siena
S
,
Cassidy
J
,
et al.
Randomized, phase III trial of panitumumab with infusional fluorouracil, leucovorin, and oxaliplatin (FOLFOX4) versus FOLFOX4 alone as first-line treatment in patients with previously untreated metastatic colorectal cancer: the PRIME study
.
J Clin Oncol
.
2010
;
28
(
31
):
4697
4705
.
5.
Peeters
M
,
Price
TJ
,
Cervantes
A
,
et al.
Randomized phase III study of panitumumab with fluorouracil, leucovorin, and irinotecan (FOLFIRI) compared with FOLFIRI alone as second-line treatment in patients with metastatic colorectal cancer
.
J Clin Oncol
.
2010
;
28
(
31
):
4706
4713
.
6.
Erbitux [package insert]
.
Princeton, NJ
:
Bristol-Myers Squibb Co
;
2011
.
7.
Vectibix [package insert]
.
Thousand Oaks, CA
:
Amgen Inc
;
2011
.
8.
College of American Pathologists
.
POET Report: perspectives on emerging technology: KRAS testing for colorectal cancer
.
http://www.cap.org/apps/docs/committees/technology/KRAS.pdf. Published September 2009. Accessed November 22, 2011
.
9.
US Preventive Services Task Force
.
Guide to Clinical Preventive Services. 2nd ed
.
Baltimore, MD
:
Lippincott, Williams & Wilkins
;
1996
.
10.
Lièvre
A
,
Bachet
J
,
Boige
V
,
et al.
KRAS mutations as an independent prognostic factor in patients with advanced colorectal cancer treated with cetuximab
.
J Clin Oncol
.
2008
;
26
(
3
):
374
379
.
11.
De Roock
W
,
Piessevaux
H
,
De Schutter
J
,
et al.
KRAS wild-type state predicts survival and is associated to early radiological response in metastatic colorectal cancer treated with cetuximab
.
Ann Oncol
.
2008
;
19
(
3
):
508
515
.
12.
Khambata-Ford
S
,
Garrett
C
,
Meropol
NJ
,
et al.
Expression of epiregulin and amphiregulin and K-ras mutation status predict disease control in metastatic colorectal cancer patients treated with cetuximab
.
J Clin Oncol
.
2007
;
25
(
22
):
3230
3237
.
13.
Di Fiore
F
,
Blanchard
F
,
Charbonnier
F
,
et al.
Clinical relevance of KRAS mutation detection in metastatic colorectal cancer treated by cetuximab plus chemotherapy
.
Br J Cancer
.
2007
;
96
(
8
):
1166
1169
.
14.
Benvenuti
S
,
Sartore-Bianchi
A
,
Di Nicolantonio
F
,
et al.
Oncogenic activation of the RAS/RAF signaling pathway impairs the response of metastatic colorectal cancers to anti-epidermal growth factor receptor antibody therapies
.
Cancer Res
.
2007
;
67
(
6
):
2643
2648
.
15.
Van Cutsem
E
,
Köhne
CH
,
Hitre
E
,
et al.
Cetuximab and chemotherapy as initial treatment for metastatic colorectal cancer
.
N Engl J Med
.
2009
;
360
(
14
):
1408
1417
.
16.
Punt
C
,
Tol
J
,
Rodenburg
CJ
,
et al.
Randomized phase III study of capecitabine, oxaliplatin, and bevacizumab with or without cetuximab in advanced colorectal cancer (ACC), the CAIRO2 study of the Dutch Colorectal Cancer Group (DCCG) [abstract]
.
J Clin Oncol
.
2008
;
26
(
15 suppl
):
LBA4011
.
17.
Bokemeyer
C
,
Bondarenko
I
,
Makhson
A
,
et al.
Fluorouracil, leucovorin, and oxaliplatin with and without cetuximab in the first-line treatment of metastatic colorectal cancer
.
J Clin Oncol
.
2009
;
27
(
5
):
663
671
.
18.
Allegra
C
,
Jessup
M
,
Somerfield
MR
,
et al.
American Society of Clinical Oncology provisional clinical opinion: testing for KRAS gene mutations in patients with metastatic colorectal carcinoma to predict response to anti–epidermal growth factor receptor monoclonal antibody therapy
.
J Clin Oncol
.
2009
;
27
(
12
):
2091
2096
.
19.
Ciardiello
F
,
Tejpar
S
,
Normanno
N
,
et al.
2011. Uptake of KRAS mutation testing in patients with metastatic colorectal cancer in Europe, Latin America and Asia
.
Target Oncol
.
2011
;
6
(
3
):
133
145
.
20.
National Comprehensive Cancer Network
.
NCCN clinical practice guidelines in oncology: colon cancer V2.2012
. .
21.
van Krieken
J
,
Jung
A
,
Kirchner
T
,
et al.
KRAS mutation testing for predicting response to anti-EGFR therapy for colorectal carcinoma: proposal for an European quality assurance program
.
Virchows Arch
.
2008
;
453
(
5
):
417
431
.
22.
Artale
S
,
Sartore-Bianchi
A
,
Veronese
SM
,
et al.
Mutations of KRAS and BRAF in primary and matched metastatic sites of colorectal cancer
.
J Clin Oncol
.
2008
;
26
(
25
):
4217
4219
.
23.
Kotoula
V
,
Charalambous
E
,
Biesmans
B
,
et al.
Targeted KRAS mutation assessment on patient tumor histologic material in real time diagnostics
.
PLoS One
.
2009
;
4
(
11
):
e7746
.
24.
Oliner
K
,
Juan
T
,
Suggs
S
,
et al.
A comparability study of 5 commercial KRAS tests
.
Diagn Pathol
.
2010
;
5
:
23
.
25.
Herreros-Villanueva
M
,
Aggarwal
G
.
KRAS assay selection: sensitivity and accuracy in clinical application [published online ahead of print June 9, 2011]
.
Mol Biol Rep
.
doi: 10.1007/s11033-011-0997-6
.
26.
Soulieres
D
,
Greer
W
,
Magliocco
AM
,
et al.
2010. KRAS mutation testing in the treatment of metastatic colorectal cancer with anti-EGFR therapies
.
Curr Oncol
.
2010
;
17
(
suppl 1
):
S31
S40
.
27.
Whitehall
V
,
Tran
K
,
Umapathy
A
,
et al.
A multicenter blinded study to evaluate KRAS mutation testing methodologies in the clinical setting
.
J Mol Diagn
.
2009
;
11
(
6
):
543
552
.
28.
Molinari
F
,
Felicioni
L
,
Buscarino
M
,
et al.
Increased detection sensitivity for KRAS mutations enhances the prediction of anti-EGFR monoclonal antibody resistance in metastatic colorectal cancer
.
Clin Cancer Res
.
2011
;
17
(
14
):
4901
4914
.
29.
Ross
J
,
Lipson
R
,
Yelensky
R
,
et al.
Comprehensive next-generation sequencing for clinically actionable mutations from formalin-fixed cancer tissues [abstract]
.
J Clin Oncol
.
2011
;
29
(
15 suppl
):
10564
.
30.
Sorenson
GD
.
Detection of mutated KRAS2 sequences as tumor markers in plasma/serum of patients with gastrointestinal cancer
.
Clin Cancer Res
.
2000
;
6
(
6
):
2129
2137
.
31.
Chien
CC
,
Chen
SH
,
Liu
CC
,
et al.
Correlation of K-RAS codon 12 mutations in human feces and ages of patients with colorectal cancer (CRC)
.
Transl Res
.
2007
;
149
(
2
):
96
102
.
32.
Haug
U
,
Hillebrand
T
,
Bendzko
P
,
et al.
Mutant-enriched PCR and allele-specific hybridization reaction to detect K-ras mutations in stool DNA: high prevalence in a large sample of older adults
.
Clin Chem
.
2007
;
53
(
4
):
787
790
.
33.
Dong
SM
,
Traverso
G
,
Johnson
C
,
et al.
Detecting colorectal cancer in stool with the use of multiple genetic targets
.
J Natl Cancer Inst
.
2001
;
93
(
11
):
858
865
.
34.
Pallisgaard
N
,
Spindler
K
,
Vogelius
IS
,
Jakobsen
AKM
.
Cell-free DNA, KRAS, and BRAF mutations in plasma from patients with metastatic colorectal cancer treated with third-line cetuximab and irinotecan [abstract]
.
J Clin Oncol
.
2011
;
29
(
15 suppl
):
3599
.
35.
Punnoose
E
,
Atwal
S
,
Spoerke
JM
,
et al.
Molecular biomarker analyses using circulating tumor cells
.
PLoS One
.
2010
;
5
(
9
):
e12517
.
36.
Bellon
E
,
Ligtenberg
M
,
Tejpar
S
,
et al.
External quality assessment for KRAS testing is needed: setup of a European program and report of the first joined regional quality assessment rounds
.
Oncologist
.
2011
;
16
(
4
):
467
478
.
37.
Hewitt
SM
,
Lewis
FA
,
Cao
Y
.
Tissue handling and specimen preparation in surgical pathology: issues concerning the recovery of nucleic acids from formalin-fixed, paraffin-embedded tissue
.
Arch Pathol Lab Med
.
2008
;
132
(
12
):
1929
1935
.
38.
US Department of Health and Human Services
.
Centers for Medicare and Medicaid Services
.
Clinical Laboratory Improvement Amendments (CLIA). http://www.cms.gov/clia. Accessed November 22, 2011
.
39.
College of American Pathologists
.
Accreditation and laboratory improvement
. .
40.
Gulley
M
,
Braziel
R
,
Halling
KC
,
et al.
Clinical laboratory reports in molecular pathology
.
Arch Pathol Lab Med
.
2007
;
131
(
6
):
852
863
.
41.
De Roock
W
,
Jonker
DJ
,
Di Nicolantonio
F
,
et al.
Association of KRAS p.G13D mutation with outcome in patients with chemotherapy-refractory metastatic colorectal cancer treated with cetuximab
.
JAMA
.
2010
;
304
(
16
):
1812
1820
.
42.
Tejpar
S
,
Bokemeyer
C
,
Celik
I
,
et al.
Influence of KRAS G13D mutations on outcome in patients with metastatic colorectal cancer (mCRC) treated with first-line chemotherapy with or without cetuximab [abstract]
.
J Clin Oncol
.
2011
;
29
(
15 suppl
):
3511
.
43.
Loupakis
F
,
Ruzzo
A
,
Cremolini
C
,
et al.
KRAS codon 61, 146, and BRAF mutations predict resistance to cetuximab plus irinotecan in KRAS codon 12 and 13 wild-type metastatic colorectal cancer
.
Br J Cancer
.
2009
;
101
(
4
):
715
721
.
44.
National Comprehensive Cancer Network
.
NCCN clinical practice guidelines in oncology: breast cancer V2.2011
. .
45.
Bardelli
A
,
Siena
A
.
Molecular mechanisms of resistance to cetuximab and panitumumab in colorectal cancer
.
J Clin Oncol
.
2010
;
28
(
7
):
1254
1261
.
46.
Laurent-Puig
P
,
Cayre
A
,
Manceau
G
,
et al.
Analysis of PTEN, BRAF, and EGFR status in determining benefit from cetuximab therapy in wild-type KRAS metastatic colon cancer
.
J Clin Oncol
.
2009
;
27
(
35
):
5924
5930
.
47.
Sartore-Bianchi
A
,
Di Nicolantonio
F
,
Nichelatti
M
,
et al.
Multi-determinants analysis of molecular alterations for predicting clinical benefit to EGFR targeted monoclonal antibodies in colorectal cancer
.
PLoS One
.
2009
;
4
(
10
):
e7287
.
48.
De Roock
W
,
Claes
B
,
Bernasconi
D
,
et al.
Effects of KRAS, BRAF, NRAS, and PIK3CA mutations on the efficacy of cetuximab plus chemotherapy in chemotherapy-refractory metastatic colorectal cancer: a retrospective consortium analysis
.
Lancet Oncol
.
2010
;
11
(
8
):
753
762
.
49.
Di Nicolantonio
F
,
Martini
M
,
Molinari
F
,
et al.
Wild-type BRAF is required for response to panitumumab or cetuximab in metastatic colorectal cancer
.
J Clin Oncol
.
2008
;
26
(
35
):
5705
5712
.
50.
Khambata-Ford
S
,
Harbison
C
,
Hart
L
,
et al.
Analysis of potential predictive markers of cetuximab benefit in BMS099, a phase III study of cetuximab and first-line taxane/carboplatin in advanced non–small-cell lung cancer
.
J Clin Oncol
.
2010
;
28
(
6
):
918
927
.
51.
O'Byrne
KJ
,
Gatzemeier
U
,
Bondarenko
I
,
et al.
Molecular biomarkers in non–small-cell lung cancer: a retrospective analysis of data from the phase 3 FLEX study
.
Lancet Oncol
.
2011
;
12
(
8
):
795
805
.
52.
Linardou
H
,
Dahabreh
IJ
,
Kanaloupiti
D
.
Assessment of somatic k-RAS mutations as a mechanism associated with resistance to EGFR-targeted agents: a systematic review and meta-analysis of studies in advanced non–small-cell lung cancer and metastatic colorectal cancer
.
Lancet Oncol
.
2008
;
9
(
10
):
962
972
.
53.
Zhu
C
,
da Cunha Santos
G
,
Ding
K
,
et al.
Role of KRAS and EGFR as biomarkers of response to erlotinib in National Cancer Institute of Canada Clinical Trials Group Study BR.21
.
J Clin Oncol
.
2008
;
26
(
26
):
4268
4275
.
54.
Brugger
W
,
Triller
N
,
Blasinska-Morawiec
M
,
et al.
Prospective molecular marker analyses of EGFR and KRAS from a randomized, placebo-controlled study of erlotinib maintenance therapy in advanced non–small-cell lung cancer
.
J Clin Oncol
.
2011
;
29
(
31
):
4113
4120
.
55.
Douillard
JY
,
Shepherd
FA
,
Hirsh
V
,
et al.
Molecular predictors of outcome with gefitinib and docetaxel in previously treated non–small-cell lung cancer: data from the randomized phase III INTEREST trial
.
J Clin Oncol
.
2010
;
28
(
5
):
744
752
.
56.
Chapman
P
,
Hauschild
A
,
Robert
C
,
et al.
Improved survival with vemurafenib in melanoma with BRAF V600E mutation
.
N Engl J Med
.
2011
;
364
(
26
):
2507
2516
.
57.
Kopetz
S
,
Desai
J
,
Chan
E
,
et al.
PLX4032 in metastatic colorectal cancer patients with mutant BRAF tumors [abstract]
.
J Clin Oncol
.
2010
;
28
(
15 suppl
):
3534
.
58.
Bokemeyer
C
,
Bondarenko
I
,
Hartmann
JT
,
et al.
Efficacy according to biomarker status of cetuximab plus FOLFOX-4 as first-line treatment for metastatic colorectal cancer: the OPUS study
.
Ann Oncol
.
2011
;
22
(
7
):
1535
1546
.

Author notes

Dr Ross is a consultant and a Scientific Advisory Board member with Bristol Myers Squibb, Inc.