Real-World Testing Practice Leave Room For Improvement in Colon Cancer

December 26, 2019

Recent study results suggest that guideline recommendations for biomarker testing in metastatic colon cancer are not consistent with real-world care. A retrospective review of the COTA Real World Data database showed suboptimal adherence to testing guidelines for RAS, BRAF, and microsatellite instability (MSI)/mismatch repair deficiency in this patient population.

Stuart Goldberg, MD

Guideline recommendations for biomarker testing in metastatic colon cancer (mCC) are not consistent with real-world care, recent study results suggest. A retrospective review of the COTA Real World Data database showed suboptimal adherence to testing guidelines forRAS,BRAF, and microsatellite instability (MSI)/mismatch repair deficiency (dMMR) in this patient population.1

“Current medical guidelines recommended multigene mutation testing for [patients with mCC] to ensure patients are receiving the optimal treatment available, but it seems like a majority of these patients—even those who are receiving targeted treatments—are not being comprehensively tested,” said Stuart Goldberg, MD, a lead study investigator from Hackensack University Medical Center in New Jersey. He served as COTA’s chief scientific officer during the project. “Recently, we’ve been able to see the promise of personalized medicine for achieving better outcomes, but proper genomic testing is needed in order to realize these benefits and to improve overall patient outcomes,” he said.

Certain mutations inRASandBRAF,as well as overexpression of HER2, have been correlated with poor responses to anti-EGFR monoclonal antibodies, such as cetuximab (Erbitux) and panitumumab (Vectibix), and have been included in relevant biomarker testing guidelines as negative prognostic factors for these therapies. Additionally, tumors with HER2 overexpression and mutations inBRAFV600E have demonstrated durable responses and disease control with available targeted therapies in these specific populations.1

MSI is a predictive biomarker for response to immune checkpoint blockade in solid tumors, for which the PD-1 inhibitor pembrolizumab (Keytruda) was approved in patients with MSI-high or dMMR disease.2The National Comprehensive Cancer Network (NCCN) now recommends dMMR testing by MSI or immunohistochemistry for all patients with mCC.3

Even though organizations such as NCCN and the American Society of Clinical Oncology advocate for these tests, published reports of testing rates remain low.

For the current analysis, 1497 patients with mCC were identified for observational study. Patients were found to be positive for testing if they had biomarker testing at any time during their disease course, including before dates of study or metastatic progression.RAStesting was identified as being eitherextended, defined as a laboratory or pathology report indicating an analysis of exon 2, 3, and 4 for bothKRASandNRAS, orlimited, defined as analysis that did not include all 3 exons in both genes. When the data were available, testing methodologies including next-generation sequencing (NGS), polymerase chain reaction, and immunohistochemistry were collected and reported.

Patients were consideredguideline alignedif they received testing that correlated with the latest NCCN guidelines recommendations for that biomarker for each respective year. Aligned testing from years 2013 to 2014 would have includedKRAStesting only by any methodology. In 2015, extendedNRASandKRAStesting was added, followed by extendedBRAFtesting in 2016 and 2017. MSI/dMMR was not included in the aligned analysis because at the time, it was recommended for testing only in patients <50 years.

Baseline characteristics of the patient set included a median age of 64 years, de novo disease presentation in 89% of patients, and adenocarcinoma histology in 94%. Half of patients were women, and slightly more than three-quarters of patients were treated at academic centers.

Testing Rates forKRASandNRAS

KRASandNRAStesting rates by any methodology were 52% and 38%, respectively, between 2013 and 2017. Among patients tested, 62% harbored an alteration inKRASand 7% inNRAS.

Guideline-aligned testing was completed in 601 patients (40%). Receiving treatment at an academic center versus a community center was a positive predictive factor for completing guideline-aligned testing (44% vs 29%;P<.001), as were age <65 years versus ≥65 years at diagnosis (44% vs 35%;P<.001) and female versus male gender (44% vs 36%;P<.01).

Forty-one percent of patients completed guidelines-alignedRAStesting, which was more likely to occur in the academic setting (44% vs 29%;P< .001), in patients <65 years (44% vs 35%;P<.001), and in women (45% vs 37%;P<.01). Those who presented with de novo metastases as opposed to disease that progressed from an earlier stage were more likely to have completeRAStesting (43% vs 25%;P<.001).


Of patients who could have receivedBRAFtesting between 2016 and 2017 (n = 546), 43% fell into the guideline-aligned group. Factors that predisposed patients to higher testing rates included receiving care at academic centers (47% vs 25%;P<.001), de novo metastatic disease (46% vs 23%;P<.001), and female gender (48% vs 38%;P<.05).

Over the entire study period,BRAFtesting was performed in 41% of patients, with 17% of those harboring a mutation.

MSI/dMMR Prevalence

Between 2016 and 2017, guideline-aligned testing for dMMR was completed in 51% of patients (276 of 546); during the whole study period, testing was done in 45%. Seven percent were MSI high or harbored ≥1 MMR deficiency, with none of the evaluable factors correlated with a higher likelihood of testing.

Effects of Testing on Therapy

In this patient cohort, 11.8% received anti-EGFR systemic therapy; of those, 28% had adequate testing forRASandBRAF. Additionally, 7% of those receiving cetuximab or panitumumab were positive forKRASorNRAS, although it is not known whether testing occurred before or after initiation of therapy.

The investigators found that although rates for individual biomarker testing rose steadily, expanding guidelines meant that real-life practices did not necessarily keep up with current recommendations, and the number of patients who were considered guideline aligned was lower in 2017 than at the start of the study. However, increased use of NGS between 2015 and 2017 showed positive trends toward adequate testing.

It was estimated that if every patient in the study who received any testing had instead received NGS multigene panel testing, there would have been a nearly 50% increase in the percentage of patients who had guideline-aligned testing. Despite this, more than 40% would have received no biomarker testing at all, suggesting that additional barriers exist.

Barriers to testing in mCC were not studied, but the investigators hypothesized that tissue availability, turnaround time, knowledge of the treating physician, insurance coverage, and patient preference or eligibility might be factors affecting adequate testing. Sequential testing, in which negative results in 1 set of genes initiate testing in the next set, rather than testing genes together, may also present challenges.

Implementation of liquid biopsies, or plasma-based NGS assays, may help overcome some barriers to testing. Analysis of circulating tumor DNA has demonstrated high sensitivity, specificity, and concordance with tissue genotyping and may be a valid option for patients in whom a tissue biopsy is not possible.

The study investigators concluded that better adherence to testing guidelines may result in improved patient outcomes by reducing patient exposure to expensive and ineffective systemic therapies and instead helping guide physicians to better approaches earlier in the treatment cycle.


  1. Gutierrez ME, Price KS, Lanman RB, et al. Genomic profiling for KRAS, NRAS, BRAF, microsatellite instability, and mismatch repair deficiency among patients with metastatic colon cancer [published online December 6, 2019]. JCO Precis Oncol. doi: 10.1200/PO.19.00274.
  2. FDA grants accelerated approval to pembrolizumab for first tissue/site agnostic indication. FDA website. Published May 23, 2019. Accessed December 22, 2019.
  3. NCCN Clinical Practice Guidelines in Oncology (NCCN Guidelines): Colon Cancer, Version 1.2020. National Comprehensive Cancer Network website. Published December 19, 2019. Accessed December 22, 2019.