Select Excerpts from the Interview

Track 2

DR LOVE: Can you describe the background of the HERA study?

DR LEYLAND-JONES: The design of the HERA study came about in a very pragmatic way. A group of us met on a Saturday in Frankfurt and spent a number of hours trying to come up with a common regimen. It became clear that it was impossible to agree upon one combined regimen.

The following morning, we simply drew the current trial design, in which we decided to include any prior chemotherapy regimen within reason, and mandated that eligible patients must have completed at least four cycles (Piccart-Gebhart 2005). We also decided to adopt the three-weekly trastuzumab regimen, which had been tested extensively worldwide.

Another factor that drove the trial design was the fact that it takes 18 weeks, either on the weekly or the three-weekly schedule, to reach steady-state levels of trastuzumab. So basically, with one year of therapy, you’ve devoted over a third of the treatment time in achieving steady state. We decided to include a two-year arm also. That was supportive of the data that came from Rich Pietras and Dennis Slamon’s group, which indicated that you need continuous attenuation of HER2 signaling to derive the most benefit (Pietras 1998).

So the four characteristics of the initial trial design were: Patients had to receive adjuvant chemotherapy; trastuzumab was administered following the completion of all chemotherapy, radiation and surgery; the three-weekly trastuzumab regimen was adopted; and a three-arm design of observation versus one year of trastuzumab versus two years of trastuzumab was developed.

DR LOVE: What were the determinants of patient eligibility?

DR LEYLAND-JONES: There were many discussions about how patients with node-negative disease fared if they had poor pathology. So we decided to include anyone with a tumor size of greater than one centimeter, including patients with node-negative disease. We did not include patients with inflammatory breast cancer. One third of patients enrolled in the trial had nodenegative disease.

Track 3

DR LOVE: Can you talk about the results that were observed?

DR LEYLAND-JONES: The findings were striking. The three-arm trial design mandated the accrual of more than 4,500 patients. In practice, accrual reached 5,070 women. The first interim analysis was planned at 475 events. That analysis showed a hazard ratio of 0.54 (p < 0.0001) for disease-free survival, which was remarkable after only one year of median follow-up (Piccart-Gebhart 2005).

DR LOVE: What about the survival data?

DR LEYLAND-JONES: Survival was not yet statistically significant after one year of median follow-up. That will be pursued, and additional data may be presented at the 2005 San Antonio Breast Cancer Symposium.

DR LOVE: When you look at different patient subsets, particularly according to status, age and other factors, what did you see in terms of relative risk reduction?

DR LEYLAND-JONES: As has been common across most of the trials, the data have not shown any huge differences between the subsets in general (4.1). However, the HERA study is the only trial in which one third of the patients had node-negative disease, and in this respect, the data are utterly striking.

Everything falls the same way, with exactly the same hazard ratio, whether it’s a subgroup with negative nodal status, one to three positive nodes or four or more positive nodes. This gives us enormous confidence to treat node-negative disease with the same regimens that we utilize to treat node-positive disease.

This is all a credit to the work of Dennis Slamon (Slamon 1987), who recognized that the expression of HER2 was critical to tumor biology and that this is a beautifully targeted therapy. In Melody Cobleigh’s study of trastuzumab as a second- or third-line single agent in patients with HER2-positive disease, there was an objective response rate of approximately 15 percent (Cobleigh 1999). That objective response increased to almost 35 percent in Chuck Vogel’s study, when it went into use as front-line therapy (Vogel 2002).

We’ve seen huge incremental gains with docetaxel/trastuzumab combinations and paclitaxel/carboplatin/trastuzumab combinations as front-line treatments. In many ways, we expected trastuzumab to be associated with increases in efficacy during the adjuvant studies; however, I don’t believe any of us thought it would be this dramatic.

DR LOVE: What did you see when you looked at the different chemotherapeutic regimens that patients received in the HERA trial? Were you able to tease anything out of that analysis?

DR LEYLAND-JONES: Around two thirds of the patients were treated with a straight anthracycline regimen: FAC, FEC or AC. Approximately six percent were treated with a nonanthracycline/nontaxane-containing regimen such as CMF. Around 25 percent were treated with an anthracycline/taxane.

So the anthracycline/taxane group appeared to do worse, although the confidence intervals are fairly wide because it’s a smaller number (4.1). Patients treated with an anthracycline/taxane regimen tend to be a group with a worse prognosis. They tend to have large numbers of positive nodes, larger tumors and worse pathologies.

After essentially any adjuvant chemotherapy regimen with trastuzumab given sequentially, the hazard ratio is 0.54, and there was a significantly decreased incidence of distant metastases. So it shows the power of administering trastuzumab sequentially.

Track 4

DR LOVE: Can you discuss cardiac toxicity in the HERA trial?

DR LEYLAND-JONES: That is one of the key features of this trial. The entry criteria included completion of chemotherapy, surgery and radiation therapy prior to trastuzumab. Therefore, the randomization took place with trastuzumab starting either six weeks after completion of the radiation or surgery or seven weeks after the last cycle of chemotherapy.

This is a very clear separation, and what this resulted in was a difference in Grade III/IV cardiac toxicity, which was 0.54 percent in the trastuzumab group versus zero percent in the observation group (4.2).

We have to remember that the denominator is approximately 1,700 patients. In hard numbers, Grade III/IV CHF occurred in nine patients versus zero. In contrast, the Intergroup data reported CHF in 31 patients who received trastuzumab and in four patients from the control group (Romond 2005). So there was considerably less cardiac toxicity in the HERA trial if you’re looking at cross-trial data.

The Intergroup study had a median follow-up of two years, whereas follow-up in HERA was one year. There are differences between the trials, of course. The LVEF cutoff criterion in the HERA study was 55 percent as opposed to 50 percent in the Intergroup trial. Having said that, at least looking at the HERA data, per se, there was a very low risk of cardiac toxicity.

DR LOVE: What is your current adjuvant treatment approach for patients with HER2-positive breast cancer?

DR LEYLAND-JONES: For our younger, fit patients, we use the current Intergroup/ NSABP method: ACTH, exactly as in NSABP-B-31. The NSABP and Intergroup trials showed that older patients or those who started post-AC with an LVEF in the 50 to 54 range were at a much higher risk for cardiac toxicity. Therefore, we tend to use a HERA type of regimen in the older patients or in patients who have a lower LVEF.

Track 5

DR LEYLAND-JONES: It’s an excellent question. All I can say is, the results are early, and the median follow-up is relatively short. If you look at the HERA data, the hazard ratio is 0.54 for disease-free survival. The cardiac toxicity rate is low, and the efficacy and safety results are based on more than 1,600 patients per arm.

In the NCCTG study, the number of events in the concurrent versus sequential analysis is small — about 130 events (4.3). The complete number of events expected are 530, so they are only about a quarter of the way through the data. The idea that synergy between agents exists may well pan out in the end, and giving these drugs concurrently is perhaps better than giving them sequentially.

Track 8

DR LOVE: What was seen in the HERA trial in patients with ER-positive and HER2-positive tumors?

DR LEYLAND-JONES: The ER-positive population was half the population. Equal benefit from trastuzumab was seen in the ER-positive population. Again, I believe this is another indicator of the importance of HER2 biology.

DR LOVE: In the clinical setting, how do you approach the use of hormonal therapy with these patients? In the trials, patients received hormonal therapy together with trastuzumab after chemotherapy was completed. Is that your general approach?

DR LEYLAND-JONES: We utilize chemotherapy and trastuzumab followed by an aromatase inhibitor in patients who have HER2-positive and ER-positive disease. The original protocol for HERA included only tamoxifen. It was subsequently modified to include the aromatase inhibitors. As shown in the 2004 San Antonio presentation, many patients with ER-positive, HER2- positive disease have a downregulation of the PR, and the aromatase inhibitors are better in this treatment group (Dowsett 2005).

DR LOVE: So you start the aromatase inhibitor after the chemotherapy?

DR LEYLAND-JONES: Yes.

DR LOVE: In the metastatic setting, do you utilize a combination of hormonal therapy and trastuzumab?

DR LEYLAND-JONES: We are participating in a trial of anastrozole with or without trastuzumab (4.4). We already have first-hand experience with that combination; however, the data are not yet available for this trial. The natural biology drives the use of this combination, so we utilize trastuzumab and aromatase inhibitors in select patients in the clinical setting.

Track 9

DR LOVE: The patient with multiple positive nodes — five to 10 positive nodes — who is a year or two out from chemotherapy is still at substantial risk for subsequent recurrence. Do you think trastuzumab should be discussed with these patients?

DR LEYLAND-JONES: This is an agonizing situation. The evidence is the evidence, and national advisory committees everywhere use it. The general advice is to restrict the use of adjuvant trastuzumab to within six months of completing chemotherapy. One or two guidelines are taking it out to a year. At our center in Canada, however, we are restricted by the data, which support utilizing trastuzumab within six months of completing chemotherapy.

Track 10

DR LOVE: What about dose-dense chemotherapy and trastuzumab?

DR LEYLAND-JONES: We’ve treated patients with that regimen. A number of people question the best way of incorporating the trastuzumab. We do not have the appropriate pharmacokinetic data on this, but a number of doctors are administering a 4 mg/kg two-weekly regimen.

There is some evidence that using loading doses can help. Instead of giving the 2 mg/kg weekly dose, it makes sense to combine it with the dose-dense chemotherapy regimen using a 4 mg/kg two-weekly regimen, which makes it easier for the patient during the time of transition from AC to trastuzumab. In a clinical setting, it is natural that physicians would adopt trastuzumab into a dose-dense kind of regimen.

Track 13

Neoadjuvant trastuzumab

DR LOVE: What were your thoughts about the data from Aman Buzdar evaluating neoadjuvant trastuzumab/chemotherapy?

DR LEYLAND-JONES: This study had one of the most dramatic complete response rates ever seen in the neoadjuvant setting. The trial was discontinued after accruing 34 patients because it showed a huge difference between the two arms (Buzdar 2005). This causes a dilemma, because you can show patients a 67 percent pCR rate with the combined neoadjuvant regimen; however, the regulatory authorities look at this and say, “Well, you have 18 patients in the one arm and 16 in the other. Show us something that is more significant clinically.”

Unfortunately, because the data are so compelling, I’m not sure whether we are going to have a large neoadjuvant trial. At the moment, the vast majority of practicing oncologists are using trastuzumab in the neoadjuvant setting. The appropriate regimen is not well demonstrated, and we do not have a large trial confirming the benefit of trastuzumab.

Select publications