Share this article on:

3 Randomized Phase III Trials Confirm: KRAS Status Is Predictive Biomarker

Carlson, Robert H.

doi: 10.1097/

ORLANDO, FL—Three randomized Phase III trials of epidermal growth factor receptor (EGFR) inhibitors have confirmed that tumor KRAS status is a predictive biomarker.

Two trials were in the first-line setting, and one was in second-line. In each trial the EGFR inhibitor was combined with a fluorouracil (5-FU)- based chemotherapy regimen and tested against the chemotherapy regimen alone.

In each case, as reported here at the Gastrointestinal Cancers Symposium, KRAS status appeared to predict the outcome. Patients with wild-type KRAS tumors had better outcomes with the chemotherapy-EGFR-inhibitor regimen than with chemotherapy alone; but patients with mutant KRAS tumors responded approximately the same to either the chemotherapy-EGFR-inhibitor arm or chemotherapy-alone arm.

Back to Top | Article Outline

Panitumumab ‘181' Trial

A combination regimen of panitumumab plus FOLFIRI (5-FU, irinotecan, and leucovorin) was tested against FOLFIRI alone as second-line treatment for metastatic colorectal cancer in the open-label, international, randomized Phase III “181” trial.

The data presented showed that panitumumab-FOLFIRI improved progression-free survival time and response rates, but only in patients with wild-type KRAS.

Median progression-free survival in patients with wild-type KRAS was 5.9 months for those in the panitumumab-FOLFIRI arm, compared with 3.9 months for the FOLFIRI-alone arm, a statistically significant difference.

But in patients with mutant KRAS there was virtually no difference between the two study arms. Median progression-free survival was 5.0 months for panitumumab-FOLFIRI and 4.9 months for FOLFIRI alone, said principal investigator Marc Peeters, MD, PhD, Coordinator of the Department of Hepatogastroenterology of the Digestive Oncology Unit at University Hospital Ghent in Belgium.

The differences in objective response by KRAS status were just as striking. Partial response rates by central review for KRAS wild-type tumor patients receiving panitumumab-FOLFIRI were 35% of 297, compared with 10% of 285 receiving FOLFIRI alone.

Partial response rates in patients with mutant KRAS were also very similar, 13% of 232 receiving panitumumab plus FOLFIRI had partial responses, versus 14% of 237 on FOLFIRI alone.

There were no complete responses in either arm.

There was a trend toward improved median overall survival in patients with wild-type KRAS receiving panitumumab-FOLFIRI, 14.5 vs 12.5 month with mutant KRAS, but not a statistically significant difference, Dr. Peeters said.

But basically, “the addition of panitumumab had no effect on patients with KRAS gene mutations.”

Patients were accrued throughout Europe (including Russia), the United States, Japan, and Australia. One prior chemotherapy regimen for metastatic colorectal cancer was allowed, but not prior EGFR-inhibitor therapy or irinotecan.

In Arm 1 of the trial, patients received panitumumab at 6.0 mg/kg every two weeks plus FOLFIRI every two weeks. In Arm 2, patients received FOLFIRI alone every two weeks.

A total of 91% of patients had KRAS analysis results: 55% of those were wild-type KRAS and 45% the mutant version.

Among 597 patients with wild-type KRAS, 303 were randomly assigned to receive panitumumab plus FOLFIRI, and 294 to FOLFIRI alone.

Among 486 with mutated KRAS, 238 received panitumumab-FOLFIRI, and 248 FOLFIRI alone.

Dr. Peeters said panitumumab was well tolerated, with an adverse-event profile as expected for an EGFR antibody combined with FOLFIRI.

Back to Top | Article Outline

CRYSTAL in First-Line

In an update and final analysis of the international Phase III CRYSTAL trial, Eric Van Cutsem, MD, PhD, Professor of Internal Medicine at University Hospital Gasthuisberg in Leuven, Belgium, confirmed that KRAS tumor mutation status is predictive across all efficacy endpoints in treating advanced-colorectal cancer patients with a combination of cetuximab and FOLFIRI.

The final analysis also showed that BRAF is a marker of poor prognosis.

“This is the first time in a randomized study that the addition of cetuximab to FOLFIRI in first-line treatment of patients with advanced colorectal cancer with wild-type KRAS tumors significantly improved overall survival compared with FOLFIRI alone,” Dr. Van Cutsem said.

With the addition of patients since the preliminary CRYSTAL report, overall survival in the update is now statistically significant, he said. Median overall survival among the 666 patients with wild-type KRAS tumors was 23.5 months for the 316 patients in the cetuximab-FOLFIRI arm, compared with 20.0 months for 350 patients receiving FOLFIRI alone.

Progression-free survival with wild-type KRAS tumors was 9.9 months in the cetuximab-FOLFIRI arm, vs 8.4 months for FOLFIRI alone.

Back to Top | Article Outline

BRAF Marker of Poor Prognosis

A retrospective subgroup analysis was conducted in CRYSTAL to investigate associations between KRAS/BRAF tumor mutation status and progression-free time and overall survival time. This update showed that BRAF tumor mutations are a marker of poor prognosis in the first-line setting, Dr. Van Cutsem said.

Among 566 patients with KRAS wild-type/BRAF wild-type tumors, median overall survival was 25.1 months for the 277 patients receiving cetuximab-FOLFIRI, compared with 21.6 months for the 289 receiving FOLFIRI alone.

Progression-free survival time for patients with KRAS wild-type/BRAF wild-type tumors was 10.9 months for the combination regimen versus 8.8 months for FOLFIRI alone.

For the 56 patients with KRAS wild-type/BRAF mutant tumors, overall survival was 14.1 months for 26 patients receiving cetuximab-FOLFIRI, and 10.3 months for the 33 receiving FOLFIRI alone.

Progression-free survival for patients with KRAS wild-type/BRAF mutant tumors was 8.0 months for the combination arm and 5.6 months for FOLFIRI alone.

Back to Top | Article Outline

Panitumumab in First-Line Setting

The international Phase III PRIME (Panitumumab Randomized Trial in Combination with Chemotherapy for Metastatic Colorectal Cancer to Determine Efficacy, aka “203” trial) added panitumumab to FOLFOX4 for first-line treatment of metastatic colorectal cancer patients.

Patients were randomly selected to receive either 6.0 mg/kg of panitumumab and FOLFOX4 every two weeks, or FOLFOX4 alone every two weeks.

For patients with wild-type KRAS tumors, median progression-free survival in the combination regimen arm was 9.6 months for the panitumumab-FOLFOX4 arm, vs 8.0 months with FOLFOX4 alone, a statistically significant difference, said first author Salvatore Siena, MD, Professor in the Division of Oncology at Ospedale Niguarda Ca'Granda in Milan.

“I call this a positive trial because of improvement in progression-free survival with the combination regimen,” he said.

The objective response rate for wild-type KRAS also favored the combination arm—55% of the 317 patients in that arm vs 48% of 323 patients receiving FOLFOX4 alone.

There was only one complete response in the trial, in the FOLFOX4-alone arm.

In the mutant KRAS arm, the response rate was 40% of 215 patients in the combination arm and also 40% of 211 patients in the FOLFOX4-alone arm.

But median progression-free survival with FOLFOX4 alone was superior in patients who had KRAS-mutant tumors: 8.8 vs 7.3 months for patients who received panitumumab-plus-FOLFOX4 therapies.

“The mechanism for this is unknown,” Dr. Siena said.

There was a trend toward improved overall survival in favor of panitumumab-FOLFOX4—23.9 months vs 19.7 months for FOLFOX4 alone.

“The difference was not significantly significant, but in my opinion, 4.2 months longer survival is clinically meaningful,” Dr. Sienna said.

Back to Top | Article Outline

Prime Toxicities

The EGFR inhibitor was associated with skin toxicities in both the KRAS wild-type and mutant tumors, 36% and 30%, respectively, Dr. Siena said.

In the FOLFOX4-alone arm the rate was 2% in wild-type KRAS and 1% in the mutant KRAS tumors.

The rate of neutropenia was similar in all four groups—approximately 42%.

Dr. Siena concluded that the overall survival curves by KRAS status indicate a biological subgroup effect, because some patients with wild-type KRAS did benefit from panitumumab. “The fact that the mutant-KRAS group did worse with panitumumab throughout the survival curve suggests resistance to panitumumab in that group,” he explained.

Back to Top | Article Outline

Are Improved Outcomes Worth the Toxicities?

“But do these trials change the way we practice?” asked the Discussant for the three papers, J. Randolph Hecht, MD, Professor of Clinical Medicine, Hematology-Oncology, and Director of the Gastrointestinal Oncology Program at the David Geffen School of Medicine at UCLA.

“Clearly only a subset of patients with metastatic colorectal cancer derived any benefit from anti EGFR therapy, but then, clearly some patients do,” he said. “The problem is that almost all patients get significant toxicities that can affect their quality of life, even if these were usually not Grade 3 or Grade 4.”

He said various groups have investigated various biomarkers to try to predict which patients would benefit from these antibodies, but studies to date have been in small series and in a retrospective fashion.

“With all these [EGFR antibody] drugs the toxicity is skin, diarrhea, and mucositis, and I'm troubled by this. It's been seen in other biologics trials, and we have to look after these patients very carefully.”

Commenting on Dr. Siena's Prime Trial of panitumumab, Dr. Hecht said it was well done, and that it hit its primary endpoint, although not the secondary endpoint of overall survival.

“Should panitumumab be approved for first-line therapy? That's a regulatory issue but I believe [the manufacturers] are going to look for regulatory approval.”

He said the main question is not cetuximab versus panitumumab, but EGFR antibodies versus bevacizumab as front-line therapy for KRAS-wild-type patients.

“With EGFR antibodies you get an increased response rate, but also toxicity”—mainly skin and GI, but that must be compared with rare catastrophic toxicities with bevacizumab, he said.

Two ongoing randomized trials do address that question, he noted: The Phase II PEAK Trial, of FOLFOX6-panitumumab versus FOLFOX6-bevacizumab; and the long-delayed CALGB 80405 Trial, of FOLFOX/FOLFIRI-cetuximab vs FOLFOX/FOLFIRI-bevacizumab.

“The CALGB Trial has dropped the biological-combination arm, FOLFOX/ FOLFIRI-cetuximab-bevacizumab, which will help answer this question,” he said.

Moving to Dr. Peeters's 181 panitumumab trial, which analyzed KRAS prospectively, Dr. Hecht noted a statistically significant improvement in progression-free survival with the antibody. But the difference in overall survival between the two arms, just two months, did not reach statistical significant.

As seen in other trials, the 181 trial had a robust improvement in overall survival with an advantage in response rate, which might be important in getting patients to liver resection, even with the increased toxicity, Dr. Hecht said.

In the updated results from the retrospective CRYSTAL trial, with more patients analyzed, there is now a statistically significant overall survival advantage for the cetuximab regimen, he continued.

“I'm not sure it changes an awful lot, but as Dr. Van Cutsem said, there is clearly no decrement, even though the numbers are very small.”

Dr. Hecht asked investigators in these types of trials to also do a meta-analysis, “because no one trial is going to have enough patients” to answer all the important questions.

Regarding the CRYSTAL update also showing BRAF as a marker of poor prognosis, Dr. Hecht said, “For your practice, the NCCN guidelines committees have already started talking about using BRAF as a marker. And unfortunately that's often being done already in patients [with commercially available assays], but I'm not sure we should do it so quickly.”

Back to Top | Article Outline

Summarizing the 3 Studies

Standard therapy for metastatic colorectal cancer has gotten somewhat more complicated, he said, summarizing the three EGFR antibody studies: “5-FU with oxaliplatin or irinotecan plus bevacizumab is used for the majority of patients in first-line treatment of advanced colorectal cancer with wild-type KRAS. A regimen of 5-FU with oxaliplatin or irinotecan plus cetuximab, or a regimen of 5-FU with oxaliplatin plus panitumumab, would both be reasonable options.”

For second line, 5-FU with oxaliplatin or irinotecan plus bevacizumab can be used, but irinotecan-cetuximab is reasonable if the patient has not already received an EGFR antibody. And FOLFIRI- panitumumab could also be used in second line, but again, only if there no prior therapy with an EGFR antibody has been given.

“But in patients with mutant-KRAS tumors, there is clearly an unmet need for new drugs and for creative thinking,” he said.

© 2010 Lippincott Williams & Wilkins, Inc.
Home  Clinical Resource Center
Current Issue       Search OT
Archives Get OT Enews
Blogs Email us!