BERLIN—Several studies presented here at the joint Congress of the European CanCer Organisation and the European Society for Medical Oncology (ECCO15-ESMO34) confirm the need to perform KRAS-gene testing before prescribing epidermal growth factor receptor (EGFR) inhibitors to patients with colorectal cancer.
The studies also raise new questions as to which EGFR blocker should be used and whether the drugs have a role in the first-line setting.
In one Phase III study of patients with metastatic colorectal cancer with wild-type KRAS tumors, adding the anti-EGFR agent panitumumab to oxaliplatin-based chemotherapy as first-line treatment significantly improved progression-free survival time.
In patients with mutated KRAS, however, panitumumab proved detrimental, said principal investigator Jean-Yves Douillard, MD, Director of Clinical and Translational Research in the Medical Oncology Branch at Centre Rene Gauducheau in Nantes, France.
Dr. Douillard presented the results of the “203,” or PRIME (Panitumumab Randomized Trial in Combination with Chemotherapy for Metastatic Colorectal Cancer), study at a late-breaking session.
Initially, he explained, the study, for which Amgen provided funding, was designed to compare panitumumab plus FOLFOX4 (leucovorin, fluorouracil [5-FU], and oxaliplatin) with FOLFOX4 alone in the entire study population, but the trial was later amended to focus on testing in the wild-type KRAS patient subset, as it is now widely accepted that only these patients benefit from treatment with an anti-EGFR agent.
The main objective was to assess the effects of panitumumab on the primary endpoint of progression-free survival times, according to KRAS status. Secondary endpoints included overall survival times, objective response rate, time to response, duration of response, and safety.
In the study, 590 patients with metastatic adenocarcinoma of the colon or rectum, and no prior treatment for metastatic disease, were randomized to FOLFOX4, and 593 patients were randomized to FOLFOX4 plus panitumumab (6.0 mg/kg every two weeks).
Testing for KRAS status was prospectively gathered in the 93% of patients for whom samples were available, resulting in four final groups for analysis: 221 patients with wild-type KRAS who received FOLFOX4 alone, 219 with mutated KRAS who received FOLFOX4 alone, 325 with wild-type KRAS who received FOLFOX4 plus panitumumab, and 219 patients with mutated KRAS who received FOLFOX4 and panitumumab.
Of the total, 60% had wild-type KRAS.
KRAS Status Matters
Among patients with wild-type KRAS, the addition of panitumumab to FOLFOX4 significantly improved the median progression-free survival time to 9.6 months, versus 8.0 months for FOLFOX4 alone, corresponding to a 20% improvement in progression-free survival time in the panitumumab arm, Dr. Douillard reported.
In contrast, in patients with tumors harboring KRAS mutations, the progression-free survival time was significantly lower in the panitumumab arm: 7.3 vs 8.8 months for FOLFOX alone. This translates to a 29% greater risk of progression in the panitumumab arm.
The median overall survival time in patients with wild-type KRAS given panitumumab has not yet been reached, Dr. Douillard said. In patients with mutated KRAS, the addition of panitumumab again was associated with worse outcomes—a median overall-free survival time of 18.7 months vs 15.1 months for FOLFOX4 alone.
The addition of panitumumab to FOLFOX4 also improved the overall response rate in the KRAS wild-type patient population as measured by blinded central review: 55% vs 48% in the FOLFOX4-only arm.
In patients with mutated KRAS, the overall response rate was 40%, regardless of the treatment arm.
Adverse event rates were comparable across arms with the exception of toxicities known to be associated with anti-EGFR therapy such as rash, diarrhea, and hypomagnesemia, Dr. Douillard said.
Panitumumab-related Grade 3 infusion reactions were reported for two patients (less than 1%).
“Overall, panitumumab was well tolerated when administered with FOLFOX4,” he said.
“The trial confirmed that the only patients benefiting from panitumumab [first-line] are KRAS wild-type, but more importantly, in the KRAS-mutated patients there is a detrimental effect which is significant—a shorter progression-free survival time.”
A member of the audience noted that PRIME is the second study to suggest that EGFR inhibitors produce worse outcomes in patients with mutated KRAS.
In the Dutch CAIRO (Capecitabine, Irinotecan, and Oxaliplatin in Advanced Colorectal Cancer)-II study, the outcomes of patients with KRAS mutations were worse when the EGFR blocker cetuximab was added to a regimen of capecitabine, oxaliplatin, and bevacizumab for the first-line treatment of colorectal cancer (Tol et al: NEJM 2009;360:563-572).
Also at the meeting, data were presented from the “181” trial showing that panitumumab administered in combination with FOLFIRI (leucovorin, 5-FU, and irinotecan) prolonged the median progression-free survival time by two months in patients with KRAS wild-type metastatic colorectal cancer, compared with treatment with FOLFIRI alone.
In the Phase III study, also sponsored by Amgen, 597 patients were randomized to receive panitumumab plus FOLFIRI or FOLFIRI alone as a second-line treatment. Patients had metastatic adenocarcinoma of the colon or rectum and only one previous chemotherapy regimen for their metastatic disease.
In patients with wild-type KRAS tumors, the median progression-free survival time was 5.9 months in the panitumumab arm, compared with 3.9 months in the FOLFIRI-alone arm, a significant difference.
Overall survival time was also better in patients with wild-type tumors receiving panitumumab plus FOLFIRI: a median of 14.5 months, compared with 12.5 months for FOLFIRI alone, but the difference was not significant.
“There was no evidence of benefit in patients with mutated KRAS tumors, but unlike in first-line therapy in PRIME, these patients did not do worse on panitumumab,” said one of the investigators, Michel Ducreux, MD, PhD, Professor of Gastroenterology at Institut Gustave Roussy in Villejuif, France.
The lead researcher, Marc Peeters, MD, PhD, Professor in the Department of Hepatogastroenterology at the University Hospital Ghent in Belgium, said that while not statistically significant, the two-month improvement in overall survival time in the panitumumab arm is clinically significant: “If you look at the curves, there is a real difference between both arms, which means there is a risk reduction of 27% with panitumumab, and this is clinically significant. In this population which has already received chemotherapy, and some who already received bevacizumab, this is an important gain, and in general, clinically significant.”
But Dr. Peeters also cautioned that panitumumab works only in wild-type KRAS, not mutant KRAS, “There is no benefit in the mutant population, but no negative effect in the mutants. If you look to other trials—e.g., PRIME—you see a negative effect on the outcome when you add an EGFR inhibitor. You need to test first. If you want to treat with panitumumab, you need to know [the KRAS status],” he said.
“So, KRAS is a major biomarker to decide on a treatment in colorectal cancer. In second-line therapy, panitumumab is not harmful, but in first-line it is.”
Dr. Douillard was asked for this article why he thought it was that panitumumab was detrimental as first-line, but not second-line, therapy. “Second-line patients are naturally selected,” he explained. “Also, the detrimental effect was seen with Erbitux [cetuximab] and FOLFOX but not reported with FOLFIRI, so it could be an oxaliplatin issue. That is a hypothesis only, but we have to look at that in detail.”
Also reported at the ECCO-ESMO meeting, two studies showed conflicting results regarding the addition of cetuximab to chemotherapy in the first-line treatment of metastatic colorectal cancer.
New data from the COIN (COntinuous chemotherapy plus cetuximab or INtermittent chemotherapy) trial showed no benefit to adding cetuximab to chemotherapy in patients with wild-type KRAS or in patients overall.
In contrast, a second study conducted by Merck Serono called CRYSTAL (Cetuximab Combined with Irinotecan in First-Line Therapy for Metastatic Colorectal Cancer), found that patients treated with cetuximab and chemotherapy lived significantly longer than patients on chemotherapy alone.
The randomized COIN trial, sponsored by the Medical Research Council, included 815 patients randomized to chemotherapy alone and 815 patients randomized to chemotherapy plus cetuximab. For chemotherapy, patients could get oxaliplatin/5-FU/leucovorin every two weeks or oxaliplatin/capecitabine every three weeks at the option of the physician or patient, but the choice had to be made before randomization.
As in the PRIME trial, the analysis was changed prospectively to focus on patients with KRAS wild-type. KRAS data were available for 84% of patients.
Results showed that among patients with wild-type KRAS, the median overall survival times were 17.9 months in the chemotherapy-only arm and 17.0 months in the cetuximab arm, a nonsignficant difference.
The only patients to be helped were a small group with a wild-type KRAS gene whose inoperable cancer had metastasized only to the liver, reported the primary investigator, Timothy Maughan, MD, Professor of Cancer Studies at the School of Medicine at Cardiff University.
In these patients, the addition of cetuximab reduced the risk of dying by about 23%, although the group was too small for the result to be statistically significant, he said.
Richard Adams, MD, Senior Lecturer in Oncology at Cardiff University, a COIN investigator, said the findings suggest that patients with unresectable liver metastases might benefit from cetuximab as first-line therapy, but that most patients will not.
The researchers also said there was a nonsignificant trend indicating a “suggestion of benefit” for progression-free survival times in patients receiving cetuximab plus oxaliplatin/5-FU/leucovorin and a “suggestion of detriment” in patients getting chemotherapy including capecitabine. As a result, the team urged caution in the selection of the chemotherapy to be used with cetuximab in clinical practice.
Updated results from the CRYSTAL study showed that patients with wild-type KRAS tumors who took cetuximab and chemotherapy had a median survival time of 23.5 months, compared with 20.0 months in patients who received chemotherapy alone. That translates into a 20% significantly reduced risk of dying in the cetuximab arm, said Eric Van Cutsem, MD, Professor of Medicine and Digestive Oncology at University Hospital Gasthuisberg in Belgium.
In addition, there was a 30% lower risk of disease progression with cetuximab than with chemotherapy alone, and the likelihood of achieving a tumor response doubled to 57.3% from 39.7%, he reported.
Continuous vs Intermittent Therapy
A second part of the COIN trial was designed to answer the question of whether it is best to give continuous chemotherapy or treat for 12 weeks, stop treatment until progression, and then restart the same chemotherapy.
There was a small survival advantage to a continuous strategy, although that comes at a cost of 10 weeks more of chemotherapy, which means not only the extra time but also more side effects and toxicity, Dr. Adams noted.
“We are left with [the need for] a discussion with patients of the pros and cons, saying to them, ‘At the end of 12 weeks, you have a choice.’”
Putting It Into Practice
When OT asked both the researchers themselves and experts not involved with the studies to sum up the new information on EGFR inhibitors, all agreed that the findings reinforced the importance of KRAS testing.
A US researcher in the audience noted that the FDA recently announced that cetuximab and panitumumab have undergone labeling changes to reflect the fact that they show no benefit in people whose colorectal tumors have mutations of the KRAS gene.
Dr. Adams said he wondered if there are biomarkers other than KRAS that may further define who will benefit from EGFR inhibitors: “Does COIN say to us that there are biomarkers we are missing, particularly when EGFR inhibitors are given in combination with chemotherapy, that KRAS is not enough? I think going back to this [COIN] data and trying to find biomarkers that identify subgroups will be very useful, and that requires a prospective trial to assess that.”
Asked how they choose between panitumumab and cetuximab, those commenting pointed out that there are no head-to-head data, but they agreed that data on efficacy, skin toxicity, and diarrhea appear comparable.
Dr. Douillard said that panitumumab is better tolerated than cetuximab and that no premedication with steroids or antihistamines is needed. He pointed to the lack of allergic reactions and the lack of need for renal monitoring as additional reasons to choose panitumumab over cetuximab.
Dr. Peeters said, “You can say the efficacy is in the same line, but there are other points that favor panitumumab: It is given every two weeks and it is very convenient for the patient because there is no premedication. But a head-to-head trial has never been done.”
Dr. Adams said he believes that first-line therapy should be oxaliplatin or irinotecan and second-line therapy should be whichever of these was not used as first-line therapy. Either of the EGFR inhibitors can then be used as third-line therapy, he said.
“It depends on what you [are used to], the cost, and convenience to patients. What I personally use is cetuximab, but I don't have a hang-up between the two, and I think they've shown very similar efficacy and toxicity.
“The majority of patients are getting a benefit from chemotherapy first-line,” Dr. Adams continued. “I don't think [using an EGFR inhibitor as first-line therapy] would be to the benefit of all of my patients with KRAS wild-type tumors, given the current data. It may cause them to have toxicity without benefits. What is important to patients is quality of life, [minimizing] toxicity, and how long they are going to live.”
A member of the audience, who identified himself as being from France, said, “[The results presented] will change what we do. We'll use panitumumab earlier; it is less allergic, and there are fewer acute side effects. I think panitumumab is easier to use, and efficacy is comparable. Currently we are using almost all cetuximab, but in a year we may be doing 50% panitumumab.”
Richard Sullivan, MD, PhD, Head of Clinical Programmes at Cancer Research UK, said that further research is needed to identify which patients would most benefit from panitumumab versus cetuximab.
“It is not clear where panitumumab should be used yet,” he said. “The 1.6 month improvement in progression-free survival is really thin, and clinical trials are the best results. When it is used in the real world, that number nearly always drops.”