Roxanne Nelson

October 09, 2014

MADRID — In the management of metastatic colorectal cancer, it is time to go beyond KRAS testing and do expanded RAS testing, according to experts.

Up until recently, KRAS status was the only validated predictive biomarker for patients with metastatic colorectal cancer, but it has become evident that expanded RAS testing is needed, experts said during a special press briefing here at the European Society for Medical Oncology Congress 2014.

Europe has already made the move, and expanded RAS testing is now a prerequisite for patients with metastatic colorectal cancer, said Dick Arnold, MD, director of the Department of Medical Oncology at Klinik für Tumorbiologie in Freiburg, Germany, and moderator of the panel. "We now have to identify all patients who bear a RAS mutation, which predicts those who are very unlikely to benefit from anti-epidermal growth factor receptor [EGFR] treatment or may potentially be harmed by it."

However, in the United States, expanded RAS testing is not a prerequisite for treatment.

Dr Alan Venook

There are no drugs currently on the market that require extended RAS testing, and the US Food and Drug Administration (FDA) has not included that as an indication, said Alan Venook, MD, Madden Family Distinguished Professor of Medical Oncology and Translational Research at the University of California, San Francisco. "Most academic centers are doing it, although in the community they are not."

The European Medicines Agency has already modified its authorization for anti-EGFR agents (such as cetuximab [Erbitux]) to exclude patients with any RAS mutations, but the FDA still labels these agents as being for patients without KRAS exon 2 mutations.

"The NCCN guidelines now include all RAS testing as mandatory, and the FDA is lagging behind," he said. "But part of that is identifying an assay that can work well."

Beyond KRAS

Previous guidance focused on KRAS testing. Only patients with metastatic colorectal cancer who had wild-type KRAS tumors were treated with an anti-EGFR agent because others were unlikely to benefit.

However, even though wild-type KRAS patients are likely to respond to anti-EGFR agents, response is not universal. This has led to a quest to find reasons for this lack of response, explained Eric Van Cutsem, MD, PhD, professor of internal medicine at the University of Leuven in Belgium.

About 60% of patients have wild-type KRAS tumors, he noted; the remaining 40% have a mutation in exon 2.

However, extended RAS testing finds further mutations in about 15% of these wild-type KRAS patients, Dr Van Cutsem reported.

The implication is that of the 60% of patients who appear on KRAS testing to have wild-type tumors, and are therefore suitable for treatment with cetuximab, 15% have other mutations that are found on expanded RAS testing. It is these patients who might not benefit from cetuximab.

For these patients, there are other therapeutic options, such as the angiogenesis inhibitor bevacizumab (Avastin).

However, it has not been established when to use bevacizumab with chemotherapy and when to use an agent such as cetuximab in patients with colorectal cancer.

Cetuximab or Bevacizumab?

A large trial recently addressed this question, and found no difference between the two agents.

The phase 3 CALGB/SWOG 80405 trial evaluated different first-line treatments in patients with KRAS codon 12 and 13 wild-type metastatic colorectal cancer.

Patients received chemotherapy with irinotecan, fluorouracil, and leucovorin (FOLFIRI) or with oxaliplatin, fluorouracil, and leucovorin (mFOLFOX6), plus either bevacizumab or cetuximab.

The study results, presented earlier this year at the annual meeting of the American Society of Clinical Oncology, showed that outcomes and overall survival were similar with either of the chemotherapy regimens and either of the targeted agents. The researchers concluded that either regimen was appropriate in first-line treatment.

However, the researchers have gone back and reanalyzed the data, taking into account results from expanded RAS testing on samples from all patients with wild-type KRAS tumors.

Results of the expanded RAS analysis were presented during the press briefing by Heinz-Josef Lenz, MD, from the division of medical oncology at the University of Southern California Norris Comprehensive Cancer Center in Los Angeles.

The original study involved unselected patients, but after 1420 patients were enrolled, the study was amended to include only patients with wild-type KRAS tumors. The final cohort of 1137 was comprised of 333 preamendment and 804 postamendment patients.

"We know from other clinical trials that mutations outside exon 2 of KRAS show that they do not benefit from VEGF receptor inhibitors," said Dr Lenz. "We wanted to test if the expanded RAS analysis would change the clinical outcome."

Expanded RAS testing included an analysis of KRAS exon 3 and 4 and NRAS exon 2, 3, and 4. The sensitivity for detection was 0.01%. "We used the most sensitive and advanced technology available for testing," he explained.

The primary end point of the study was overall survival.

In the wild-type RAS subgroup, median overall survival was pushed beyond 30 months. However, the researchers found that the difference between cetuximab and bevacizumab in combination with chemotherapy was not significant (32.0 vs 31.2 months). There was also no difference in progression-free survival.

There were some differences between the two groups, but it did not reach statistical significance, Dr Lenz reported.

However, in the wild-type RAS subgroup, response was better with cetuximab than with bevacizumab (68.6% vs 53.6%; P < .01).

"I think it is very important that newly diagnosed patients with metastatic colorectal cancer be tested for RAS," concluded Dr Lenz. "Overall survival of 30 months in both arms sets a new benchmark for patients with metastatic disease."

First-line therapy should reflect treatment goals and take potential adverse effects into consideration, he added, as well as dose intensity, treatment duration, tumor shrinkage, second-line therapies, and additional biomarkers for anti-EGFR and anti-VEGF therapies.

Possible Cure?

Dr Venook presented a subset analysis from the study in which wild-type KRAS patients underwent surgery as part of the treatment strategy. The goal was to determine the characteristics and long-term outcome of patients enrolled in the trial.

He reported that there are patients who might be cured, even in the presence of advanced metastatic disease, but emphasized that this is still a "work in progress."

In the cohort, 179 patients underwent surgery at some point after chemotherapy, and 130 reached a stage of nonevidence of disease (NED). Currently, 96 patients are alive and 50 remain NED, he reported. Median overall survival in these patients was 60 months.

"There were 130 patients who at some point had no evidence of cancer, so they are potentially curable," said Dr Venook.

About 60% of these patients received cetuximab before going to surgery and about 40% received bevacizumab. "We don't know yet which is better," he noted.

Further RAS testing might indicate who is cured and who is not, but there is "still a lot of work to do," said Dr Venook. Expanded RAS testing might help distinguish prognosis in this already select group of patients, he added.

Each of the examined strategies resulted in an overall survival of more than 30 months in all treatment groups in both trials, noted Dr Arnold. "This is the longest overall survival reported, and beyond the promising activity in first-line treatment, it is related to the fact that most patients also had active second- and/or further-line regimens with the complementary strategy."

"This sets the standard, and the 'winners' in these trials are the patients," he said. "The findings underline that treatment with any monoclonal antibody should be regarded as the standard."

Dr Arnold added that "RAS testing is the beginning, not the end, of the biomarker story in metastatic colorectal cancer."

"In terms of molecular biology, the current situation of using only RAS testing is not satisfactory. In the future, we need more precise biomarker definitions, and possibly smaller biomarker-defined subgroups of patients," he said in a statement. "This is not a time for simple answers. Choosing the most appropriate first-line therapy in colorectal cancer remains a complex and multifactorial decision."

European Society for Medical Oncology (ESMO) Congress 2014: Abstracts 501O and LBA10. Presented September 29, 2014.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: