I N T E R V I E W  Harry D Bear, MD, PhD Dr Bear is Chairman of the Division of Surgical Oncology, Professor of Surgery and Microbiology and Immunology and Walter Lawrence Jr Distinguished Professor in Oncology at Virginia Commonwealth University School of Medicine’s Massey Cancer Center in Richmond, Virginia.

Select Excerpts from the Interview*
* Conducted on December 10, 2004

 Track 2

DR LOVE: Can you summarize the updated NSABP trial B-27 data?

DR BEAR: Although the numbers are low, the addition of docetaxel, preoperatively, in NSABP-B-27 doubled the pathologic complete response rate from 13 percent to 26 percent. In 2004, we presented the overall and disease-free survival data at San Antonio (Bear 2004).

The overall survival did not show any significant difference among the groups or between each of the docetaxel groups and AC alone. There was a trend towards improved disease-free survival with the addition of docetaxel, particularly when given preoperatively. When we examined relapse-free survival, which was similar to the definition of disease-free survival in the CALGB trial 9344 (Henderson 2003), we saw a significant improvement with the addition of preoperative docetaxel compared to AC alone.

The difference between the relapse-free survival result and the disease-free survival result was probably caused by a chance event, which was an increase in the number of second malignancies at other sites and in the contralateral breast in some of the patients who received docetaxel. Since that’s included in disease-free survival, it sort of wipes out the beneficial effect. Most of the relapse-free survival benefit was caused by a reduction in local recurrences. We have not yet seen any difference in distant disease-free survival, which, in terms of patient survival, is probably the most important outcome.

DR LOVE: What’s your interpretation of these data?

DR BEAR: I believe it relates to a number of factors, some of which are biological. The Skipper hypothesis is that metastatic clones of tumor cells behave differently from the primary tumor. We thought we had disproven that in the NSABP-B-18 trial (Fisher 1998), but it may be true in some patients, so that a patient who has a good response in the breast may not necessarily have a good response in the metastatic disease sites.

Probably the most important reason is a statistical issue.

One of the other major factors is that this group of patients, as compared to the CALGB-9344 (Henderson 2003) or NSABP-B-28 (Mamounas 2005) trials, is probably diluted by some better-risk patients. In those trials, all the patients had known positive nodes, whereas in B-27, the patients’ nodal status is unknown and certainly included a number of ER-positive, node-negative patients whose benefit from any chemotherapy is probably fairly small. If you look at the actual magnitude of the effect in terms of hazard ratios, it’s really quite similar to CALGB-9344 or NSABP-B-28 with the addition of paclitaxel, so I think it’s largely a sample size issue, as well as some potential biostatistical parameters that neutralized any effect on survival.

 Track 7

DR LOVE: What’s going on with NSABP-B-35, the trial in DCIS?

DR BEAR: That is a very interesting trial, and it’s continuing to accrue very well. That study is examining what kind of an anti-estrogen should be given to patients with ER-positive DCIS. All the patients have been treated for DCIS with a lumpectomy and radiation therapy, and they are then randomly assigned to tamoxifen or anastrozole.

DR LOVE: What’s been your experience in terms of tolerance of the aromatase inhibitors and anastrozole, as compared to tamoxifen?

DR BEAR: For the most part, they’re very well tolerated. Patients seem to do quite well with anastrozole, and I think it’s somewhat less of a problem with hot f lashes and other side effects, compared to tamoxifen (3.1). However, aromatase inhibitors are certainly not trouble free.

DR LOVE: What’s your take on the ATAC data (Howell 2005) and some of the other data that have been coming out in terms of switching aromatase inhibitors (Boccardo 2005; Coombes 2004; Goss 2003; Jakesz 2005a, b)?

DR BEAR: I believe it’ll probably precipitate a fairly wholesale change in the primary treatment of ER-positive postmenopausal breast cancer patients. It clearly is the end of five years of tamoxifen as a standard treatment; I don’t think we’ll see that anymore.

 Track 8

DR LOVE: What’s the status of the NSABP sentinel lymph node trial — B-32?

DR BEAR: Dr Julian presented the initial results from that trial at San Antonio in 2004, which consisted of the pathologic data and accuracy data for the half of the trial that received a sentinel node biopsy plus an axillary node dissection ( Julian 2004). This was a trial of 5,600 patients randomly assigned to sentinel node biopsy with or without an axillary node dissection.

The false-negative rate in that trial was 9.5 percent, which some people thought was higher than expected, but it’s right in line with other multicenter validation trials. There are individual institutional trials with much lower rates done by a few surgeons. Immunohistochemistry examination of the sentinel nodes was done as part of the trial, but it was done centrally, and it was blinded, so we don’t know the results of those.

DR LOVE: What is the current role of sentinel node biopsy in the clinical setting?

DR BEAR: I think it probably should be the standard of care for most patients who don’t have a specific contraindication to it. It has been in my practice since the completion of the trial. That’s what I offer most patients with invasive breast cancer.

DR LOVE: Does the false-negative rate of 9.5 percent concern you?

DR BEAR: It doesn’t really concern me. I feel pretty confident in the technique. We’ve been doing it a long time, and I’ve seen very few, if any, axillary recurrences that were unexpected, and that’s probably the most significant thing that could happen. Obviously, the main answer to that question really is going to have to come from the long-term follow-up of the B-32 trial.

 Track 9

DR LOVE: Can you discuss the development and validation of the Oncotype DX assay?

DR BEAR: This is an exciting era in the management of breast cancer. Soonmyung Paik and the NSABP have developed Oncotype DX, a genomic assay that reliably evaluates gene expression in paraffin-fixed tumor tissue. They obtained tumor tissue from a large number of the patients participating in the NSABP-B-14 trial and, using the assay, established a panel of 21 genes that were accurate at predicting the risk of recurrence among patients who were randomly assigned to either tamoxifen or placebo (Paik 2004a).

The panel of 21 genes as a predictor of recurrence — called the risk score — was validated in the NSABP-B-20 trial. In B-20, patients with node-negative, ER-positive disease were randomly assigned to tamoxifen alone or tamoxifen plus chemotherapy. In that group of patients, not only did the risk score estimate the risk of recurrence, it also significantly predicted whether or not patients would receive any benefit from adding chemotherapy to tamoxifen. Only the patients in the high-risk group — not the low- or intermediate-risk group — showed significant benefit from adding chemotherapy to tamoxifen (Paik 2004b).

A lot of unanswered questions remain, including patients with node-positive disease who are already at higher risk, a priori. Additionally, does this assay apply to other types of patients? That is one of the things we’re going to assess in the neoadjuvant setting in the NSABP-B-40 study.

DR LOVE: Are the patients in the NSABP-B-40 neoadjuvant trial going to be evaluated with a genomic assay?

DR BEAR: In the B-40 trial, core biopsies will be taken and retrieved for the purpose of gene expression analysis. We will be able to look at that panel of 21 genes as one of the ways to assess gene expression, which is a critical part of the study.

SELECT PUBLICATIONS