ECOG-E3200: FOLFOX4 with bevacizumab versus FOLFOX4 versus bevacizumab in patients with previously treated advanced colorectal cancer

Safety

We did not observe any unexpected toxicities with the use of bevacizumab. One of the key things to keep in mind about ECOG-E3200 is that we used a higher dose of bevacizumab (10 mg/kg every two weeks) than has been reported in any of the colorectal cancer studies, except for that first trial by Dr Kabbinavar (Kabbinavar 2003). ECOG-E3200 demonstrated statistically significant increases in nausea, vomiting and neuropathy in the bevacizumab arm (Giantonio 2005b), which were probably due to these patients staying on treatment longer because of the addition of bevacizumab.

I don’t believe bevacizumab is enhancing these side effects; rather, it’s just allowing the patients to stay on treatment longer. We observed about a 15 percent incidence of Grade III neuropathy in the patients treated with FOLFOX4 and bevacizumab compared to nine percent in the patients treated with FOLFOX4 alone (Giantonio 2005b).

We observed bowel perforations in ECOG-E3200 similar to what was reported by Dr Hurwitz (Hurwitz 2004). We saw three cases in the patients treated with FOLFOX4 plus bevacizumab, three cases in the patients treated with bevacizumab alone and no cases in the patients treated with FOLFOX4 alone. Four of our six cases of perforation occurred within the first cycle of therapy, which is different from Dr Hurwitz’s experience. In his IFL plus bevacizumab study, there was no association with time on treatment and the development of the perforation.

Efficacy

In October 2004, an interim analysis was conducted and presented to the ECOG Data Monitoring Committee. Based on their review, they recommended the release of the study data. At the Data Monitoring Committee’s meeting, a statistically significant (p = 0.0024) improvement in median overall survival — the primary endpoint of the trial — was reported. The patients who received FOLFOX4 plus bevacizumab had a median overall survival of 12.5 months compared to 10.7 months for those treated with FOLFOX4 alone. The hazard ratio was 0.74; patients treated with bevacizumab in combination with FOLFOX had a 26 percent reduction in the risk of death (Giantonio 2005a; [4.1, 4.2]).*

It was recognized that these data were important because FOLFOX has evolved as front-line therapy for patients with metastatic colorectal cancer in the United States. Even though ECOG-E3200 is a study of second-line therapy, it makes many of us more comfortable in extrapolating these data to the front-line setting.

Clinical implications of ECOG-E3200

I think ECOG-E3200 adds very strongly to the existing data that bevacizumab in combination with FOLFOX, as first-line therapy, should further improve median overall survival. Given what we can accomplish with chemotherapy alone using FOLFOX and FOLFIRI, as demonstrated by Dr Tournigand (Tournigand 2004), conceivably, we can anticipate starting to move the median overall survival to two or more years. That is remarkable when just four years ago the median overall survival was about 12 months for the control arm of 5-FU/leucovorin in the IFL study (Saltz 2000).

In patients with metastatic disease, I generally start with FOLFOX plus bevacizumab. When they progress, I switch to FOLFIRI, and based on their insurance, I keep them on bevacizumab. When using FOLFIRI plus bevacizumab, I had been using a dose of 5 mg/kg; however, based on the results from ECOG-E3200, I’ll probably try to increase the dose to 10 mg/kg if I can obtain approval from the patient’s insurance company.

In terms of the management of some of the concerning side effects, particularly bowel perforation, I think it shouldn’t deter clinicians from using bevacizumab. All these perforations presented with abdominal pain. I think we just have to be a bit more vigilant in our evaluation. With prompt intervention, the bowel perforation can be effectively managed.

Regimens of bevacizumab in combination with oxaliplatin or irinotecan

The magnitude of benefit for FOLFOX4 with bevacizumab in ECOG-E3200 was about two months (Giantonio 2005a, 2005b), and the magnitude of benefit for IFL with bevacizumab, as reported by Dr Hurwitz, was a little more than four months (Hurwitz 2004). The important difference was that IFL was being used as first-line therapy.

A randomized Phase III study being conducted in Europe will ask a survival question for the addition of bevacizumab to FOLFOX in the first-line setting. If the magnitude of benefit is the same, I think that will add to the data set indicating that it is important for bevacizumab to be administered with chemotherapy.

One of the intriguing hypotheses, at least in colorectal cancer, is that the benefit seen with the addition of bevacizumab may be independent of the specific chemotherapy used. A number of mechanisms have been proposed to help explain the benefit associated with bevacizumab. One is that there may be improved chemotherapy delivery into the tumor, resulting in a higher tumor kill. There’s some merit to that, and emerging clinical data support it.

Dr Willett has looked at the changes in microvascular density and interstitial pressures in tumors from patients who have received bevacizumab. The patients underwent colonoscopy to have the measurements performed on the tumor, received a single dose of bevacizumab and 12 days later had a second colonoscopy to have those measurements repeated (Willett 2004a, 2004b).

In his study, we saw both a reduction in microvascular density and an improvement in interstitial pressures, which enhances the flow through the tumor and potentially allows greater drug delivery into the tumor.

Select publications

* Data discussed in the interview reflect data presented at ASCO GI 2005. Updated data presented at ASCO 2005 are shown in 4.1.

Dr Giantonio is an Assistant Professor of Medicine at the Abramson Cancer Center of the University of Pennsylvania in Philadelphia, Pennsylvania.