Select Excerpts from the Interview

Track 7 - 10

DR LOVE: Can you describe the CONFIRM trials, which studied the use of PTK/ZK (vatalanib) in patients with untreated metastatic colon cancer?

DR HECHT: CONFIRM-1 enrolled previously untreated patients with metastatic colorectal cancer (Hecht 2005), and CONFIRM-2 evaluated second-line treatment of patients with metastatic colorectal cancer. All patients in the CONFIRM-1 study received 5-fluorouracil and oxaliplatin, administered as FOLFOX-4.

Patients were randomly assigned to receive either 1,250 mg of PTK/ZK daily or placebo. The identical regimen was used in CONFIRM-2 for patients who had failed treatment with 5-fluorouracil and irinotecan.

DR LOVE: Can you discuss the trial results?

DR HECHT: In both CONFIRM trials, progression-free survival was measured rigorously using a central assessment, which was completely blinded as to how the patient was doing clinically.

In CONFIRM-1, the final progression-free survival analysis showed a modest improvement, which was not significant (HR = 0.88), but investigator-assessed progression-free survival, a more common measurement, did reach statistical significance (HR = 0.83; p = 0.026).

Progression-free survival was one of two primary endpoints; the other was overall survival. At this time, we only have data for progression-free survival, and that’s what was presented at ASCO 2005 (Hecht 2005).

One of the things seen in CONFIRM trials that I think was very interesting was the correlation of progression-free survival and LDH, which has been evaluated as both a prognostic indicator and a predictive marker for chemotherapy. Historically, LDH has been used as a stratification criterion to make certain both arms of a study are balanced. In these trials, LDH and performance status were used as stratification criteria.

When you look at the patients who had high LDH – the worst-prognosis group – they derived the greatest benefit from PTK/ZK. In fact, with the addition of PTK/ZK, the hazard ratio of 0.88 decreased to 0.68 for centrally assessed progression-free survival.

Track 12

DR LOVE: I am curious about your thoughts on the antitumor mechanism of action of bevacizumab.

DR HECHT: The original thinking was that bevacizumab inhibited the growth of new blood vessels. In order for tumors to grow beyond a certain size, the angiogenic switch is f lipped, and a mutation causes the secretion of growth factors, which leads to the supply of new blood vessels to the tumor.

However, another name for vascular endothelial growth factor is vascular permeability factor. Tumor blood vessels are leaky, and tumors tend to have a high intratumoral pressure. One of the thoughts — this is sometimes called the Jain hypothesis (Jain 2001) — is that the vasculature that’s associated with tumors is abnormal. It’s tortuous. The vessels are dilated, inefficient, and also very leaky. By giving anti-angiogenic therapy, you might be normalizing the vasculature.

This process would have several effects (2.1), one of which could be to allow a more efficient delivery of chemotherapy by providing more efficient blood vessels and by reducing intratumoral pressure. It’s harder to get chemotherapy in against a pressure gradient.

A study by Chris Willett (Willett 2004), published in Nature Medicine, had a huge impact in terms of this thinking, although it had a small number of patients. The study showed that following treatment of rectal cancer with bevacizumab, blanching and shrinkage of the tumor occurred and less blood (ie, perfusion and volume) appeared to be going to the tumor (2.2).

In addition, interstitial pressure of the tumor fell and significant decreases in microvascular density occurred. This observation was made using bevacizumab alone, which offers some validation that at least a portion of the Jain hypothesis appears to be correct.

Bevacizumab increases response rates, and no one expected anti-angiogenic therapies to increase response rates. All we ever expected was to cause stabilization of disease by keeping new blood vessels from growing. Instead, virtually all the trials with bevacizumab have shown an increase in response rate.

In fact, the colon cancer trials — the original 5-FU trial that Kabbinavar published (Kabbinavar 2003), the IFL study by Herb Hurwitz (Hurwitz 2004), Kabbinavar’s other 5-FU trial (Kabbinavar 2005), the TREE-2 trial (Hochster 2005, 2006) — all show approximately a 10 percent improvement in response rate with the addition of bevacizumab.

Track 15

DR LOVE: Do you continue using bevacizumab in a patient who has disease progression?

DR HECHT: Nobody knows the best approach in this situation, and there are studies in development right now to answer that question. One of the problems has been that the standard of care in colon cancer has changed so rapidly that it’s been difficult to catch up and fill in all the gaps.

The ECOG-E3200 trial, which Bruce Giantonio presented at ASCO this year (Giantonio 2005), showed that the combination of FOLFOX and bevacizumab was better than FOLFOX alone in patients who were bevacizumab-naïve and who had failed an irinotecan-containing regimen. However, that group of patients no longer exists because there are very few patients who don’t receive bevacizumab front line.

The answer depends on how bevacizumab works. If bevacizumab works only by blocking the growth of new blood vessels, maybe giving it after disease progression doesn’t make any sense.

However, if bevacizumab works by facilitating the delivery of chemotherapy, perhaps it does make some sense.

DR LOVE: What do you do in your own practice?

DR HECHT: I have a discussion with the patient. Theoretically, it could make sense to continue treatment; however, more toxicity might occur. We try to enroll the patient in a clinical trial, since we are in an academic medical center, but I do present that option to my patients.

Track 17 - 18

DR LOVE: The AVANT trial, like the NSABP-C-08 trial, is considering what might be the most interesting question in adjuvant therapy right now: the use of bevacizumab. However, it is also comparing capecitabine to infusional 5-FU. What are your thoughts on this?

DR HECHT: I believe that this will provide another therapy option. Personally, I have patients who are very happy with infusional 5-FU, even when given the choice between the two agents.

Other patients may prefer not to have a central line or may not want to wear a pump, but remember that oxaliplatin requires a central line anyway.

The issue that will remain unanswered following those trials is the optimal duration of bevacizumab. One thought was that if you were preventing the growth of new blood vessels, maybe bevacizumab should be given to patients for the rest of their lives.

Obviously, that’s not a practical option from either a toxicity standpoint or a resource standpoint.

How long do you treat? Remember, we still don’t know how long to treat with cytotoxic agents. Some data — British data, for example — suggest shorter courses of chemotherapy may be as good as six-month regimens.

Not long ago, we were treating patients for a year. At this time, we’re continuing the fluoropyrimidine/oxaliplatin and bevacizumab for six months.

DR LOVE: Do you use capecitabine with oxaliplatin in the clinical adjuvant setting?

DR HECHT: I have done so in special circumstances. The data from an efficacy standpoint showing functional equivalence between capecitabine-containing regimens and infusional fluorouracil-containing regimens are good (Twelves 2005a, b).

Randomized Phase III trials of capecitabine alone versus 5-fluorouracil have been conducted (Twelves 2005a, b), so the use of capecitabine in this setting is not a huge extrapolation.

Select publications