Efficacy of bevacizumab in combination with chemotherapy in first- and second-line clinical trials

The ECOG E3200 trial, which randomly assigned patients with previously treated advanced colorectal cancer to receive FOLFOX4 with or without bevacizumab, demonstrated a positive survival advantage with the addition of bevacizumab (Giantonio 2005).

These data could have an immediate impact on the clinical use of bevacizumab, particularly in patients with refractory disease or in second-line therapy.

Previously, FOLFOX had not shown an independent survival advantage in the second-line setting. In Rothenberg’s study, FOLFOX showed improvement in time to progression, but not survival (Rothenberg 2003). In E3200, patients who received FOLFOX alone clearly did as well as patients who received FOLFOX in Rothenberg’s study. In E3200, the addition of bevacizumab increased survival by almost a couple of months.

In the IFL plus bevacizumab front-line trial, a dramatic improvement occurred in survival and time to progression in patients who received the combination (Hurwitz 2004). Trials such as these support our argument that adding bevacizumab to chemotherapy regimens in the first- or second-line setting can result in a positive outcome for patients in terms of survival and progression-free survival.

The magnitude of benefit was greater in the front-line study, and we don’t know whether that is because of some front-line phenomenon that wouldn’t be seen in second-line metastatic colon cancer or if something more “additive” occurs with bevacizumab and irinotecan compared to oxaliplatin.

Synergy between bevacizumab and chemotherapy

The front-line study examining 5-FU/leucovorin with or without bevacizumab in patients with a poor performance status demonstrated an advantage from the combination (Kabbinavar 2004). Clearly, synergy exists between bevacizumab and 5-FU, and it appears that when you start combining it with other chemotherapeutic agents, that benefit carries over or adds up even further (1.1).

What may be happening with bevacizumab in a number of settings is that it controls the growth of the cancer, prevents progression and, therefore, adds to progression-free survival, which is now translating into overall survival.

I believe bevacizumab works by changing the dynamics of the interstitial pressure within the tumor, which facilitates the delivery of chemotherapy. If indeed that is its mechanism of action, then the particular chemotherapy utilized shouldn’t matter, and bevacizumab should improve its efficacy.

Continuation of bevacizumab after progression

One question the E3200 trial does not answer is whether a patient who progresses on a front-line combination of chemotherapy plus bevacizumab should be maintained on bevacizumab in the second- and third-line settings.

The suspected mechanism of action of bevacizumab suggests that patients should be maintained on this agent, but other factors — including cost and side effects — must be considered. If a patient is tolerating bevacizumab, I believe it would be difficult to justify discontinuing it when switching from a front-line to a second-line therapy. Even though E3200 doesn’t answer that question directly, I believe continuing bevacizumab is a reasonable strategy in the nonprotocol setting, given the survival data in the front- and second-line settings.

Nonprotocol use of first-line FOLFOX with bevacizumab

I was surprised that after the data presented at the 2003 ASCO meeting on bevacizumab/IFL, clinicians shifted towards using FOLFOX plus bevacizumab (Hurwitz 2004). I believe this occurred because the FDA gave free reign to use bevacizumab with any intravenous 5-FU-containing regimen. The pendulum was already swinging toward using FOLFOX as the standard front-line approach for patients with metastatic colon cancer, so oncologists went ahead and combined it with bevacizumab. That requires two leaps of faith — first, infusional 5-FU in FOLFOX rather than bolus 5-FU in the IFL regimen, and, second, oxaliplatin rather than irinotecan — without a lot of supportive data. The data from the E3200 trial support both leaps, showing bevacizumab to be safe and effective with infusion 5-FU and oxaliplatin.

Rationale for adjuvant clinical trials combining FOLFOX and bevacizumab

To date, drugs that are effective in metastatic disease are also effective in the adjuvant setting, so the obvious next step after E3200 is to test FOLFOX plus bevacizumab as adjuvant therapy. Based on the MOSAIC trial data, FOLFOX offers clear benefit in the adjuvant setting (de Gramont 2005), whereas the IFL data were negative in the adjuvant setting.

The NSABP is undertaking a study of FOLFOX plus bevacizumab versus FOLFOX alone. As it stands now, patients will receive 12 months of bevacizumab — six months concurrent with chemotherapy, followed by an additional six months of bevacizumab alone — hoping for an antiangiogenic effect on microscopic disease and the prevention of relapses.

The Intergroup is initiating a key trial for patients with Stage III disease that will have three arms. One arm will utilize FOLFOX as the standard of care, a second arm will utilize 12 cycles of fluorouracil/leucovorin with or without irinotecan (FOLFIRI), and the third arm will utilize six cycles of FOLFOX and six cycles FOLFIRI. Adding a biologic agent has also been discussed. It’s an interesting study evaluating whether we can double up on chemotherapy and further improve outcome in the adjuvant setting; however, this trial is contingent on a positive result in the PETACC-3 study, which is evaluating adjuvant FOLFIRI.

Adjuvant therapy for patients with Stage II disease

It is interesting that we tend to back away from adjuvant therapy in patients who have a lower risk, when it may be more appropriate to do exactly the opposite. Those are the patients with whom we should be the most aggressive. In the Stage II subset analysis of the MOSAIC study, the patients who received FOLFOX had a three-year disease-free survival of 87 percent. To my knowledge, that’s the highest number ever reported for Stage II patients, and it’s a compelling number in the clinic.

The real issue with using 12 cycles of FOLFOX in the adjuvant setting is the high rate of neurotoxicity seen at the end of that six-month treatment period. It’s not life threatening, but it’s a nuisance for patients.

In breast cancer we are accustomed to utilizing adjuvant chemotherapy for relatively small gains, meaning two to four percent absolute gain. I believe we should be equally aggressive when treating patients with colon cancer, and we should incorporate these adjuvant therapies as often as possible. By adopting these new therapies, we’re going to cure more patients of this disease.

Safety and efficacy data from the X-ACT adjuvant trial

The X-ACT trial was conducted in Europe with approximately 2,000 patients, half of whom received full-dose capecitabine — 1,250 mg/m² bid, two weeks on, one week off — and the other half received the Mayo Clinic 5-FU/leucovorin regimen (Cassidy 2004b). Both of those recipes are considered “too spicy” in the United States, so we were surprised when the safety data revealed that approximately 60 percent of patients did not require a dose reduction in either arm.

Although the remaining approximately 40 percent in both arms needed dose reduction — either due to hand-foot syndrome or the more toxic side effects of the Mayo Clinic regimen, including neutropenia, sepsis and diarrhea. However, the capecitabine arm was significantly less toxic.

The trial was designed to be an equivalence study, but the analysis showed capecitabine to be superior by a couple of percentage points in both disease-free and overall survival (Cassidy 2004a). Therefore, based on the lower toxicity and slightly higher efficacy data, I believe capecitabine is a better option than the Mayo Clinic 5-FU/leucovorin regimen.

Capecitabine in the adjuvant setting for Stage II colon cancer

Often when discussing treatment with young, healthy patients with lower-risk Stage II disease, I consider using infusion 5-FU or sometimes even FOLFOX, but some patients don’t want to approach it that aggressively, and some oncologists even debate whether to offer chemotherapy to those patients. I believe capecitabine, although controversial, is a good option. I recognize that we don’t have firm randomized trial data with capecitabine in Stage II disease, but in the X-ACT study, capecitabine was equivalent to bolus 5-FU, so I believe it’s an option for patients with low- to moderate-risk Stage II disease.

Adjuvant chemotherapy for Stage III disease

In patients with Stage III disease, I generally use FOLFOX6 as adjuvant therapy because I believe the two-day infusion schedule is the optimal way to administer 5-FU intravenously. I discuss the MOSAIC and X-ACT data with the patients, and some patients ask about using CAPOX. I usually steer them toward one of the other options. I have treated patients with CAPOX in the adjuvant setting, and my intuition tells me that regimen is adequate so I would rather give that than nothing; however, we need the trial data to be certain.

Select publications

Dr Marshall is an Associate Professor and Director of Developmental Therapeutics and GI Oncology at Lombardi Comprehensive Cancer Center at Georgetown University in Washington, DC.