Molecular Testing for Colon Cancer Falls Far Short of the Guidelines

Stuart L. Goldberg, MD

Between 2013 and 2017, just 40% of patients with metastatic colon cancer (mCC) received guideline-recommended biomarker testing that could have improved treatment decisions, according to findings from a retrospective study. Results also showed that adherence to evidence-based recommendations for genotyping was poor in academic and community settings alike.¹

“Our study demonstrates that many patients who would be eligible for genomic profiling that would direct their care are not getting the appropriate testing,” study coauthor Stuart L. Goldberg, MD, said in an interview.

Although National Comprehensive Cancer Network (NCCN) guidelines recommend RAS, BRAF, and microsatellite instability (MSI)/mismatch repair deficiency (dMMR) testing for all patients with metastatic colorectal cancer (mCRC), among the study’s 1497 patients with mCC, a subset of mCRC, the testing rates were 41%, 43%, and 51%, respectively. Of note, patients were more likely to have testing performed if they were treated at an academic center versus a community center (44% vs 29%; P <.001). However, testing rates for all biomarkers were suboptimal at both. Most patients (1152; 77%) were evaluated at an academic center.¹

Testing aligned with NCCN guidelines was more prevalent in certain patient groups: those who presented with de novo metastatic disease compared with those who progressed from an earlier stage (43% vs 25%; P <.001); those diagnosed when <65 years versus ≥65 years (44% vs 35%; P <.001); and women versus men (45% vs 37%; P <.05).¹

At the study’s outset, the authors hypothesized that their findings would show that genomic testing rates for guideline-recommended biomarkers had improved over time, but this was not the case (Figure 1). “Guidelines significantly expanded between 2013 and 2017, [but] the number of patients receiving full guideline- aligned genotyping in 2017 was lower than it was at the start of the study,” they wrote. The American Society of Clinical Oncology also recommends biomarker testing for all patients with mCRC.²

Failure to receive guidelinerecommended biomarker testing may put patients at risk for not receiving appropriate immunotherapy options or for receiving ineffective anti-EGFR monoclonal antibody (mAb) therapy, the authors noted.

“These are the guideline-compliant tests that need to be conducted in mCC,” said Goldberg, chief of the division of Outcomes and Value Research at the John Theurer Cancer Center in Hackensack New Jersey, part of the Hackensack Meridian Health network. “Genomic profiling is not research anymore; it’s available and ‘prime time,’ [but] we’re not using the technology appropriately.”

Benefits of Biomarker Testing

Figure 1. (Click to Enlarge)

Biomarker testing can also reveal the interventions that are not likely to elicit a response, potentially reducing patients’ exposure to ineffective, toxic therapies. “If every patient with mCC had molecular testing, many patients would not be subject to expensive treatments that do not work. But if you don’t order the tests, you won’t give the right therapies,” Goldberg said.

For example, in mCRC, KRAS exon 2 mutations correlate with a lack of benefit from anti-EGFR mAbs, including cetuximab (Erbitux) and panitumumab (Vectibix). If molecular testing uncovers such a mutation, treatment regimens that include anti-EGFR mAbs can be avoided. Proper testing would spare approximately 40% of patients with mCRC who might otherwise receive mAb-based treatment.¹ “Testing in KRAS exons 3 or 4 [and in] NRAS exons 2, 3, or 4 finds 20% more RAS mutations than exon 2 testing alone, adding another 8% of patients with mCRC who will have poor response to anti-EGFR [therapy],” the authors wrote.

BRAF mutations, specifically V600E, add another 9% to the total of those who would have similarly poor response, and HER2 amplification is also a predictor of negative anti-EGFR mAb response. HER2 amplification presents in about 2% of mCRC cases, increasing to almost 60% the percentage of patients with mCRC in whom ineffective anti-EGFR agents could be avoided.

mAb therapies are expensive relative to the costs associated with available molecular tests.¹ The authors calculated that the cost of providing mAb therapy to just 4.6% of the undergenotyped patients receiving inappropriate therapy for 1 year would pay for testing all 1497 patients in the study. Genomic profiling, Goldberg concluded, can lead to a reduction in treatment costs.

Table. (Click to Enlarge)

Observed real-world mutation rates for KRAS, NRAS, and BRAF are 62%, 7%, and 17%, respectively. Investigators used these data to estimate the number of patients who would have been considered candidates for anti-EGFR agents had all participants completed molecular testing. In total, only 14% (210 patients) of patients with mCC would receive a normal wildtype readout for KRAS, NRAS, and BRAF alike, and consequently be qualified for anti-EGFR therapy.¹

When evaluating the actual testing rates seen in the study, investigators found that 895 patients (60%) had no documented mutational testing, indicating that 685 patients might have unnecessarily been offered anti-EGFR treatment.¹

Applying Real-World Data

Goldberg and colleagues used COTA’s Real World Data database to identify patients with histologically confirmed mCC who were diagnosed between January 1, 2013, and December 1, 2017. Based in Boston, COTA Healthcare applies analytics to fragmented real-world data to interpret patient information that is accumulated nationwide from physician notes; radiology, pathology, and surgery reports; and genomic testing results.

Although current guidelines support molecular testing in patients with mCRC, the study focused solely on patients with mCC and did not include rectal cancer cases. Evaluation for actionable mutations was considered “guideline-aligned” if NCCN guidelines recommended testing for the biomarker for the full year analyzed. Guideline-aligned testing in 2013 and 2014 included testing for KRAS by any methodology; 2015, extended testing (defined as multiple exon testing) of NRAS and KRAS; and 2016 and 2017, extended testing of KRAS and NRAS, BRAF testing by any methodology, and MSI/dMMR analysis by any methodology (Table).¹

At the time of data extraction, COTA’s database included deidentified demographic, diagnostic, treatment, and quality-of-care information for patients with colon cancer that was collected from the electronic health records of 23 oncology practices throughout the United States and from 258 oncologists in Arkansas, Maryland, Michigan, New Jersey, New York, and Tennessee.¹

KRAS and NRAS

Findings showed that guideline-aligned RAS testing was performed in 41% of patients (610/1497; Figure 2). Just more than half (52%; 777/1497) had KRAS testing by any methodology and 38% (566/1497) had NRAS testing by any methodology. Among those evaluated for a KRAS mutation, 62% had an alteration. Meanwhile, 7% of patients tested for NRAS harbored the genetic abnormality.

Investigators said polymerase chain reaction (PCR) was the most prominent testing methodology for KRAS in 2013. Next-generation sequencing (NGS) supplanted PCR as the preferred modality from 2014 on. For NRAS, NGS was the dominant testing approach for the duration of the study.¹

BRAF

BRAF testing was not guideline-recommended until 2016. During this 1-year period, 43% of patients (235/546) had guideline-aligned testing. However, BRAF testing was performed prior to its inclusion in mCC guidelines: between 2013 and 2015, 41% of patients (613/1497) were tested by any methodology for BRAF alterations; 17% of these patients tested positive for the mutation.

Figure 2. (Click to Enlarge)

In their analysis of the different biomarker testing methodologies used to check for each mutation, Goldberg and colleagues noted that the insights provided by different tests vary. For example, NGS multigene panel testing, which affords comprehensive coverage of actionable mutations such as BRAF, RAS, and MSI, can yield more data, and thus contribute more to reversing undergenotyping rates, than can a single-gene test.¹

In this study, use of a tissue- or plasma- based NGS test in each patient would have meant 59%, rather than just 40%, of patients would have had guideline-aligned biomarker screening, investigators said. Still, they noted, more than 40% of patients still would not have received molecular testing, suggesting that other barriers obstruct broad adoption of tissue-based testing options in mCC.¹

Potential hindrances to biomarker assessment in this setting have not been well studied, the authors wrote, but might include tissue availability, turnaround time, physician education/knowledge, cost/insurance coverage, patient preference, and eligibility for therapy based on performance status and comorbidities.¹

However, Goldberg argues that lack of access to molecular testing is no longer a valid excuse to not pursue guideline- aligned genotyping. “Commercial laboratories perform genomic testing, so every physician should have the ability to order the test. And for colon cancers specifically, the testing protocols that we evaluated are guidelinerecommended procedures,” he said.

MSI/dMMR

Like BRAF testing, MSI/dMMR testing was not guideline-recommended until 2016, when dMMR evaluation was recommended for all patients with mCC. In this final year of the study, 51% of patients (276/546) had guideline-aligned testing for an alteration.

Over the whole study period, including 2013, 2014, and 2015, when MSI/dMMR testing was recommended only for a subset of patients with mCC, testing rates were low: 667 (45%) of the study’s 1497 patients were screened for dMMR, and 46 (7%) were found to be MSI-H or to harbor ≥1 MMR deficiency.¹

The authors’ discovery that just 40% of patients with mCC observed between 2013 and 2017 received guideline-aligned biomarker testing highlights the need for broader adoption of the guideline-backed testing protocols, according to Goldberg.

“We’re [progressively] moving toward the catchphrase of ‘precision oncology,’ where you match the right patient with the right therapy, and this is becoming increasingly possible as we develop broader genomic testing, but you’ve got to order the tests,” Goldberg said.

References

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. Study shows only 40 percent of patients with metastatic colon cancer receive guideline-recommended biomarker testing [news release]. Redwood City, CA: Guardant Health; December 11, 2019. investors.guardanhealth.com/ news-releases/news-release-details/study-shows-only- 40-percent-patients-metastatic-colon-cancer. Accessed December 17, 2019.

NGS was the most common evaluative modality for BRAF mutations between 2013 and 2017. Importantly, the number of patients who were tested for KRAS, NRAS, and BRAF by NGS increased each year of the study.

“There’s a cost—value aspect: Insurance companies will argue that [biomarker tests] are expensive, but these tests are going to point to the right treatment,” Goldberg said. “The money that is spent on therapies that do not work would have paid for the tests. You’re going to spend more money on diagnostics, but you’re going to save a lot of money on ineffective treatments.”

Because biomarker testing can improve treatment selection, more patients must be tested for actionable mutations, Goldberg said. Genotyping can identify somatic mutations that predict patient response to specific therapies, offering patients an opportunity to receive effective personalized treatments. Notably, BRAF mutations and HER2 overexpression are positive predictors of response to targeted therapy; MSI status is a positive predictor of response to immune checkpoint inhibitors.¹