Colorectal Cancer Trials Support Gene Testing for Two Drugs
Adapted from the NCI Cancer Bulletin.
A trio of new studies adds to the growing evidence that patients with colorectal cancer should have their tumors tested for genetic mutations prior to starting therapy with cetuximab (Erbitux®) or panitumumab (Vectibix®). Tumors with certain mutations are unlikely to respond to the drugs, and these patients should be spared the expense and side effects of the medications, the findings suggest.
The new findings confirm and extend the results of prior studies, including a retrospective analysis of a major clinical trial, the CRYSTAL trial, that found cetuximab to be ineffective in patients with mutated forms of a gene called KRAS. Since that study was presented, the National Cancer Institute (NCI) has been modifying cetuximab trials to include testing for mutations in the KRAS gene. In Europe, treatment with panitumumab is restricted to patients with normal KRAS genes.
The first new study was a retrospective analysis of the only cetuximab trial to show a survival benefit for patients with advanced colorectal cancer. In the now-closed CO.17 trial (see the protocol summary), the benefits of cetuximab were limited to patients whose tumors carried normal, or unmutated, forms of the KRAS gene.
As the researchers reported last year, the cetuximab group lived, on average, six weeks longer than those who received supportive care. But only some patients benefited, and to understand why, the researchers went back and analyzed tumor samples from 394 of the 572 participants.
Some clear patterns emerged: Among patients with normal KRAS genes, survival was nearly double for the cetuximab group compared to the supportive care group (9.5 months versus 4.8 months). The progression-free interval was also superior in the group of patients treated with cetuximab, 3.7 months versus 1.9 months.
Among patients with mutated KRAS genes, however, there was essentially no difference in survival between the two treatment groups, the researchers reported in the October 23, 2008, New England Journal of Medicine (NEJM). KRAS mutations were present in 42 percent of the tumors.
"These results represent an exciting transition in the treatment of cancer," said co-author Dr. Derek Jonker of the University of Ottawa. "Whereas in the past - and specifically in the CO.17 trial - we treated a large number of people and had a small effect, we now begin to have the ability to identify the patients who are most likely to benefit and then tailor treatment based on the unique genetic makeup of each person's cancer."
The research to date leads to the "reasonable conclusion" that all patients with advanced colorectal cancer who are being considered for anti-EGFR therapy should undergo KRAS testing, according to an accompanying editorial.
The clinical results are consistent with preclinical research on EGFR in colorectal tumors. The protein sits on the cell surface and controls a number of signaling pathways involved in cell growth. KRAS mutations can constitutively activate signaling pathways (for example, MAPK) normally controlled by EGFR. The researchers believe that because this activation occurs "downstream" of EGFR, drugs that inhibit the receptor will not affect the abnormal signaling.
Mutations in other genes, such as BRAF, may also impair the effectiveness of anti-EGFR therapy, the editorial notes. About 15 percent of colorectal cancers have disease-related mutations in the BRAF gene, which is part of the MAPK pathway.
A day after the NEJM study appeared, Italian researchers shared the results of another trial that showed that metastatic colorectal tumors with BRAF mutations did not respond to cetuximab or panitumumab. None of the patients with the mutations responded to these drugs, whereas all of the responders had normal BRAF genes.
The study included 113 patients, and KRAS mutations accounted for 30 percent of the non-responsive cases. BRAF mutations explained another 14 percent, leaving more than half of the nonresponsive cases unexplained.
Additional molecular markers will be needed to better define patients who are unlikely to benefit from EGFR-targeted treatment, said Dr. Federica Di Nicolantonio of the University of Turin. She presented the results at the 20th EORTC-NCI-AACR Symposium on Molecular Targets and Cancer Therapeutics in Geneva, Switzerland. (The results were subsequently published December 10, 2008, in the Journal of Clinical Oncology; see the journal abstract.)
A third study looked at the KRAS status of 715 patients in four panitumumab trials. None of the patients whose tumors had mutated KRAS genes responded to panitumumab (defined as tumor shrinkage or no tumor growth), compared with almost 14 percent of those with normal KRAS genes who did respond. Dr. Daniel Freeman of Amgen, Inc., the manufacturer of panitumumab, presented the results at the second EORTC-NCI-ASCO Annual Meeting on Molecular Markers in Cancer in Hollywood, FL.
During a press briefing, Dr. Bruce Johnson, director of thoracic oncology at the Dana-Farber Cancer Institute, said that the results were timely, important, and "quite consistent" with the cetuximab study in the NEJM.
Editor's note: A more recent retrospective analysis of data from 579 patients with advanced colorectal cancer showed that patients with a particular mutation in KRAS did appear to benefit from cetuximab. In this analysis of data from patients who had received cetuximab either in a clinical trial or outside of a trial, patients with the mutation (called p.G13D) survived for longer than patients with other KRAS mutations who received cetuximab (medians of 7.6 months versus 5.7 months). In addition, overall survival was similar between patients with the p.G13D mutation and patients with a normal KRAS gene. The results were published October 27, 2010, in JAMA. Conversely, an updated analysis of the CRYSTAL trial, with longer follow-up time and data on KRAS status available for more patients, found that the benefit of cetuximab continued to be limited to patients with normal KRAS. These findings were published May 20, 2011, in the Journal of Clinical Oncology. Results from a third study of 1,630 patients, the COIN trial, further complicated the issue by showing that even in patients with normal KRAS overall survival was worse when treated with cetuximab and chemotherapy rather than chemotherapy alone (17.0 months versus 17.9 months). Regardless of treatment, the scientists found that overall survival varied depending on the mutations present in a patient’s tumor. The data were published June 18, 2011 in The Lancet.