Second-line endocrine therapy for postmenopausal women with metastatic disease (Figure 29)

DR SCHWARTZBERG: This is a common case that we’re faced with every day — the patient who received chemotherapy, an adjuvant aromatase inhibitor and then relapses — and a bone-only relapse is positive disease.

We have few clinical trial data to suggest which therapy to use next in cases like this. The information we have on patients who relapse on an adjuvant aromatase inhibitor is recent, so we don’t have long-term results to examine and we have to extrapolate from studies in the metastatic setting where a front-line aromatase inhibitor was used.

Those data suggest we have three different options: (1) switch to a second aromatase inhibitor — either a steroidal or another nonsteroidal agent, (2) switch to fulvestrant, which is an estrogen receptor down-regulator or (3) use a SERM, and most commonly one would use tamoxifen in this case because the patient has not been exposed to it previously.

In head-to-head clinical trials in the second-line metastatic setting, the data suggest that fulvestrant is equivalent to an aromatase inhibitor and that fulvestrant might offer slight advantages in some of the endpoints, although not the primary endpoint.

Also, in the first-line setting, a trial comparing fulvestrant to tamoxifen demonstrated equivalence between these two agents for the patients with hormone receptor-positive disease.

I believe that hormonal therapy is absolutely the best option in this case. For a patient with bone-only, hormone-positive disease, even though she relapsed on adjuvant therapy, one might still get considerable mileage out of further endocrine therapy.

As for which endocrine therapy I would select, I prefer not to use another drug that works by the same mechanism. I would choose fulvestrant in this case, and I would use it alone.

If I were to use another aromatase inhibitor, the data don’t suggest that one agent is better than another. Typically, if a patient relapsed on a nonsteroidal agent, I would switch to a steroidal agent, and vice versa, because of the belief that some differences might exist in the mechanisms of resistance. So in this case I would select exemestane.

Fulvestrant loading dose

DR SCHWARTZBERG: I do use a loading dose with fulvestrant — that’s become the standard practice in my clinic. We start with an initial dose of 500 milligrams and then administer 250 milligrams two weeks and four weeks later, and then we begin the 28-day dosing schedule.

That regimen is based on preclinical data that suggest it takes a long time to reach a steady-state level of fulvestrant if you use the standard dosing that was initially approved, which is 250 milligrams every 28 days.

If you carefully examine the more recent studies, particularly those in which patients had previously been treated with an aromatase inhibitor and then received a second aromatase inhibitor, an aromatase inhibitor with another agent or fulvestrant, you see that progression-free survival decreases quickly in the first two to four months. Attaining a steady-state level of the drug so that it can work the way it’s supposed to as soon as possible and before the disease progresses is an important strategy.

In the EFECT study, which compared fulvestrant to exemestane, they used a loading dose. While that’s not the label dosing, I believe many people have come to adopt that strategy and, in my experience, no increase in toxicity occurs with the loading dose.

We don’t have the head-to-head studies comparing a loading dose to the monthly 250-mg schedule, but I see little disadvantage to using the higher dose.

Fulvestrant causes little toxicity. Some patients experience a local reaction from the intramuscular injection, and a small number experience hot flashes, gastrointestinal upset or headache, but it’s an extremely well-tolerated drug, and for patients who don’t mind coming in once a month to receive an injection, as opposed to taking a daily pill, it’s a nice alternative.

Rationale for trials combining fulvestrant with an aromatase inhibitor (Figure 31)

DR SCHWARTZBERG: Combining fulvestrant with an aromatase inhibitor is an interesting research strategy. A good amount of preclinical data supports the idea that we might observe greater efficacy if we not only shut off estrogen production, as an aromatase inhibitor would do, but also downregulate the estrogen receptor.

We know that in the preclinical animal models, a downregulator like fulvestrant can cause a feedback loop that increases the available estrogen. Thus, shutting off estrogen, which can then bind to the estrogen receptor as well as downregulating, theoretically leads to greater efficacy. We need to conduct a clinical trial to prove this.

I believe that if you had asked this question before the ATAC data were available, the majority would have predicted that the combination would show greater efficacy.

The results of the ATAC trial probably explain why over half of the clinical investigators responded that no difference would be evident between the two arms, even though the preclinical data are compelling otherwise.

I was surprised that the combination in ATAC was no better than either agent alone and, to my mind, why that is remains to be explained. We don’t understand everything.

The models are not necessarily perfect, but they’re exciting and show that we may obtain greater efficacy from combination therapy that is rationally designed — and an aromatase inhibitor with fulvestrant is a rational design. This combination makes more sense from a preclinical model.

Breast Cancer Update 2007 (7)

DR IAN E SMITH: Fulvestrant seems to be as good as tamoxifen in up-front trials. It also seems to be as good as anastrozole, but it isn’t better.

One question is about the estrogen receptor becoming hypersensitized when it is reset. If the estrogen receptor is exposed to low doses of estrogen for a long time — as, for example, during prolonged aromatase inhibitor therapy — the receptor then seems to become hypersensitive to minute amounts of estrogen. So the question is whether fulvestrant would work better if you used an aromatase inhibitor concomitantly.

Two or three trials address this — one in the United Kingdom is called SoFEA. Patients who experience relapse on aromatase inhibitors are randomly assigned to fulvestrant or fulvestrant in combination with the aromatase inhibitor to test this question.

If another issue is that prolonged exposure to low estrogen doses hypersensitizes the receptor, then maybe we should be administering these therapies intermittently.

The latest idea being tested in clinical trials is intermittent aromatase inhibitor therapy — for example, three months on, three months off. In metastatic disease, the tumor marker CA15-3 may be useful in guiding therapy.

As soon as the levels go down, you stop and wait. Treatment can be restarted when the marker levels rise again to determine whether that approach is superior.

The Breast International Group trial 1-07 — the Study of Letrozole Extension (SOLE) — is like the MA17 trial, in which people who’ve been receiving endocrine therapy for five years are switched to either continuous or intermittent aromatase inhibitor therapy.

Nanoparticle albumin-bound (nab) paclitaxel versus standard taxane therapies (Figure 32)

DR SCHWARTZBERG: It’s always interesting to see the disparity among the clinical investigators and the community. Sometimes there’s no consensus within either group, but in this case, we clearly see that a large majority of the clinical investigators chose paclitaxel.

If you examine the data, which must be extrapolated a little, in my opinion it’s clear now that weekly is the best schedule for administering paclitaxel.

The CALGB data support paclitaxel as weekly therapy in the first-line metastatic setting, showing it to be superior in terms of response rate in addition to having less overall toxicity. Although a little more neuropathy occurs with the weekly versus the every three-week schedule, the hematologic toxicity is much less. The large adjuvant trial, ECOG-E1199, evaluated four schedules of taxane therapy: paclitaxel administered weekly and every three weeks and docetaxel administered weekly and every three weeks.

While no significant difference was evident in the main comparison, in the subgroup analyses, I believe weekly paclitaxel was clearly the winner in terms of less toxicity and greater efficacy, even in that large adjuvant setting in which it’s harder to tease out the effect in terms of disease-free survival.

As for docetaxel, I believe it’s been clearly shown that every three weeks is superior to the weekly schedule. Weekly docetaxel stirred a lot of interest around 2000, and some early data from Hainsworth and his group were provocative. Many people adopted that strategy, but I’ve always found fatigue to be a problem.

With docetaxel administered every three weeks, particularly at the full dose of 100 mg/m², quite a lot of neutropenia occurs.

However, growth factors can be used prophylactically, which is the standard according to the NCCN and ASCO guidelines, to avoid that toxicity, and a lower dose of 75 milligrams every three weeks is also effective.

I’m a little puzzled by the respondents who chose weekly docetaxel and paclitaxel every three weeks, given the abundance of data demonstrating that those are less effective regimens.

Personally, I voted for nab paclitaxel in this question. We have done a fair amount of research using weekly nab paclitaxel and have found that it’s not only an effective drug, either alone or in combination, but it’s also easy to administer and is not associated with much toxicity. In addition, nab paclitaxel delivers somewhat higher doses and does not require premedications.

Also, some data from the clinical trials suggest the neuropathy, which is the dose-limiting side effect of paclitaxel in general, resolves faster with nab paclitaxel than with Cremophor-based paclitaxel.

Use of premedications with nab paclitaxel (Figure 33)

DR SCHWARTZBERG: I am surprised by these responses and find the reasons for using premedications interesting. The emetic potential of paclitaxel is considered low, so there is no standard reason to use steroids for that purpose with nab paclitaxel as half of the oncologists suggested.

Most of us are concerned about the use of steroids, particularly when patients are receiving therapy on a weekly basis. Women frequently have problems with insomnia and nervousness, and older patients are frequently diabetic or have a prediabetic condition. Although steroids are wonderful drugs that we use every day, we should use them only when it’s appropriate.

The fact that a third of the practicing oncologists have a treatment algorithm that includes premedications when administering nab paclitaxel must be changed immediately.

We’re moving into an era of evidence-based therapy, of quality medicine and pay for performance. The insurers are increasingly scrutinizing the way we’re treating patients and the supportive measures we use, and this practice would be difficult to justify.

In an era in which electronic medical records are used increasingly, it seems it would be simple to alter the algorithm and take out the premedications for nab paclitaxel while retaining them for Cremophor-based paclitaxel.

As for hypersensitivity reactions, they are rare with nab paclitaxel in my experience, nor do I recall seeing any in our Phase II study of nab paclitaxel combined with capecitabine.

In this trial, approximately 50 patients received nab paclitaxel, 125 mg/m² on days one and eight, and capecitabine, 1,000 mg/m² BID on days one through 14 every 21 days.

We recorded a high response, approximately 60 percent overall, and the mean time to progression was nine months, which compares favorably to other combination regimens.

We are encouraged by this regimen and are hoping to repeat the study with a similar schedule and the addition of bevacizumab, which we believe we can add because the reported toxicity was reasonable.

A few patients experienced neutropenia and required dose delays or reductions, but in general no unusual toxicity occurred and we found full doses could be combined easily.

Click on the image to enlarge

Efficacy and tolerability of nab paclitaxel (Figure 34)

DR SCHWARTZBERG: At the San Antonio Breast Cancer Symposium in 2006, Bill Gradishar presented data from a randomized Phase II trial comparing weekly or every three-week nab paclitaxel to q3wk docetaxel as first-line therapy for metastatic breast cancer.

In that study, the weekly nab paclitaxel appeared to be superior to the every three-week schedule and also to docetaxel. They also evaluated weekly nab paclitaxel at 100 mg/m² and 150 mg/m², and while the higher dose appeared to work a little better, it brought more toxicity.

Still, the toxicity of weekly nab paclitaxel was substantially less than that of docetaxel. I believe these data are the reason why the majority of both doctors in practice and the clinical investigators responded that they believe nab paclitaxel is safer and more tolerable.

Assessing the efficacy depends on how you interpret the data because it is a randomized Phase II study. I would vote for weekly nab paclitaxel as somewhat more efficacious than the best dosing of docetaxel, which is the every three-week schedule that was used in Gradishar’s trial.

I hope we have the opportunity to study nab paclitaxel in the adjuvant setting as a treatment option. Currently, the use of docetaxel is resurgent in the adjuvant setting, particularly with the nonanthracycline regimens — TCH for HER2-positive tumors and TC for HER2-negative disease.

Breast Cancer Update 2007 (6)

DR WILLIAM J GRADISHAR: Nab paclitaxel was developed to take advantage of the significant antitumor activity of the taxanes but also to avoid some of their side effects. Solvents typically used with drugs such as docetaxel or Cremophor-based paclitaxel are absent, and instead the paclitaxel is administered in an albumin delivery system to increase the amount of drug that reaches the tumor tissue. That’s the underlying rationale.

What’s been shown to date, both through some of the early Phase I and Phase II trials and ultimately the Phase III trial, is that when administered every three weeks, nab paclitaxel was superior to solvent-based paclitaxel administered every three weeks.

Despite more of the paclitaxel being administered in the nab preparation than with the every three-week solvent-based paclitaxel, less neutropenia occurred. A different kind of neuropathy appeared to be present that resolved more quickly. A greater antitumor effect was also observed in terms of response rate and improved progression-free survival.

In an era when we’re increasingly using weekly therapy and when many perceive docetaxel to be the most active single-agent anticancer therapy for breast cancer, what most people want to know is, how does nab paclitaxel compare to a weekly taxane schedule? How does it compare to docetaxel?

We conducted a randomized Phase II trial, which we reported at the 2006 San Antonio meeting and updated at ASCO 2007. Patients with metastatic breast cancer were randomly assigned to first-line treatment with a dose of 300 mg/m² of nab paclitaxel every three weeks, 100 mg/m² of docetaxel every three weeks or nab paclitaxel administered weekly three out of four weeks at a dose of either 100 or 150 mg/m².

In December 2006, we reported that the weekly nab paclitaxel schedules were more active from the standpoint of antitumor activity than either every three-week docetaxel or every three-week nab paclitaxel. The weekly treatment arms were not only active but were also well tolerated, particularly the 100-mg/m² dose, which appeared at the time to be the optimal schedule.

The weekly schedule with 150 mg/m² had a slightly higher response rate, but it also is associated with slightly more toxicity. We did not see much of a difference in terms of progression-free survival between these two arms.

The weekly treatment arms were associated with response rates in the 60-plus percent range, markedly higher than the every three-week treatment arms of either nab paclitaxel or docetaxel.

Part of the more recent ASCO presentation was the response rate findings from the independent radiology review. As expected, there was a drop-off in response rates among all four treatment arms.

However, consistent with the original investigator-reported findings, response rates for both weekly nab schedules remained numerically superior to every three-week docetaxel or every three-week nab paclitaxel.

In the December 2006 analysis, we would have said that the progression-free survival is not different across the nab paclitaxel treatment arms, but all are superior to docetaxel administered every three weeks.

What’s emerging now is that both the every three-week nab paclitaxel and the weekly schedule of 150-mg/m² nab paclitaxel arms appear to be the superior treatments.

However, from the standpoint of efficacy and tolerability, the 150-mg/m² schedule appears to be the treatment arm to be pursued in a pivotal Phase III trial.

I believe one of the things that will come out of the upcoming randomized trial is whether the added antitumor efficacy that’s presumed to be associated with the weekly schedule will offset what might be slightly more toxicity than we see with lower-dose weekly schedules of nab paclitaxel.

One of the interesting observations made across all the reported nab paclitaxel trials is the notion that the neuropathy might be different.

One of the first things people would have considered is that with this agent, when you eliminate the Cremophor, no neuropathy should occur.

But what has been observed in every trial — even in the Phase I trials — is that with high doses, you see neuropathy even in the absence of Cremophor. This might be attributable to the chemotherapy drug itself. So neuropathy occurs with nab paclitaxel — that seems to be a consistent finding.

The numbers are not huge, but there appears to be resolution of the neuropathy to the point at which you can readminister the chemotherapy drug within approximately three weeks.

In other words, you see a decrease in the severity of the neuropathy to the point at which you feel comfortable readministering the drug.

That’s in contrast to what we typically see when patients develop Grade III neuropathy with solvent-based paclitaxel, with which the duration of the neuropathy is much longer.

In terms of other side effects, the degree and frequency of significant neutropenia are decreased with the nab paclitaxel every three-week and weekly schedules, relative to the three-weekly docetaxel, and minimal febrile neutropenia is associated with nab paclitaxel at the doses evaluated.

Additionally, in contrast to docetaxel, in our study the incidence of stomatitis is clearly less frequent whether you’re using every three-week or weekly schedules of nab paclitaxel.

Chemotherapy and bevacizumab for hormone-resistant metastatic disease (Figure 35)

DR SCHWARTZBERG: I find these responses provocative.

The investigators are apparently driven by clinical trials, whereas the practicing oncologists are more creative in terms of different options. Based on these responses, there’s clearly no consensus among practitioners about the right approach.

When I answered this question, I selected capecitabine alone based on the description of the patient as having minimal tumor symptoms.

In the long run, this patient will be exposed to multiple agents, so I want to ease her into chemotherapy. Obviously, in a case like this the physician should speak frankly with the patient about the need for chemotherapy and describe the different options. It in fact becomes the patient’s decision.

Another option I would consider would be paclitaxel and bevacizumab because that combination seems to have the best time-to-progression data in randomized trials, although this issue hasn’t been evaluated head to head.

This regimen is complicated, however, involving weekly infusions and a second drug. It changes the lifestyle of the patient, whereas capecitabine alone is an oral therapy and it’s similar to what she has already been on, so that is frequently the choice of patients as they ease back into chemotherapy.

Click on the image to enlarge

Click on the image to enlarge

TAnDEM trial data and treatment of hormone receptor-positive, HER2-positive, metastatic disease (Figures 36-37)

DR SCHWARTZBERG: This case is realistic because three years of tamoxifen therapy is approximately the point patients had reached before the adjuvant trastuzumab data were released, so we have a fair number of these patients.

The TAnDEM data confirmed that the subgroup of patients who have ER-positive, PR-positive breast cancer that is also HER2-positive are the least likely to respond to hormonal therapy alone.

Their median time to progression was approximately two and a half months, which is rather short. The addition of trastuzumab increased that by 60 or 70 percent, to four months or more.

While trastuzumab increased the disease-free survival and time to progression, these results are still somewhat modest. One might ask what would have been the efficacy of single-agent trastuzumab in these patients? We might have seen the same effect, but we don’t know because that wasn’t studied, which is one criticism of this study.

To me, the most important feature of the TAnDEM trial is that investigators saw a survival advantage, and that’s why I would vote for endocrine therapy with trastuzumab in this case.

The trial data suggest that patients with HER2-positive metastatic disease should begin an anti-HER2 therapy right away, and that might impact their overall survival. Specifically, I would use a nonsteroidal aromatase inhibitor with trastuzumab for this patient.

On the other hand, if a patient presented with a similar history, relapsing on endocrine therapy, but was symptomatic with visceral disease, I would move on to chemotherapy with trastuzumab. I would do that because the response rates are high, as is symptom control, with that combination.

In addition, it has shown a survival advantage, and I don’t believe we have time to wait for progression when a patient is symptomatic.

If half the patients will experience disease progression at four months anyway, they may be very sick, and then you may not have an opportunity to salvage the situation.

Clinical trial evaluating fulvestrant with capecitabine for disease progression after adjuvant endocrine therapy

DR SCHWARTZBERG: I found it fascinating to see in these responses that clinicians are starting to consider combining endocrine therapy and chemotherapy again, which was anathema in the field for many years.

I have a particular interest in that strategy and have recently launched a trial for the ACORN Network evaluating fulvestrant and capecitabine for patients who are failing at or within 12 months of completing an adjuvant aromatase inhibitor.

The goal is to evaluate the idea that the patients who fail an aromatase inhibitor have relatively endocrine-resistant disease and that many of them experience disease progression within the first few months of therapy.

That may be due to the pharmaco-dynamics and pharmacokinetics of the drug and a matter of getting the drug in fast enough, or maybe it’s due to the fact that this disease is more intrinsically resistant, if patients fail an aromatase inhibitor, and they may not respond as well to a second-line hormonal therapy.

We simply don’t know the data on that yet, although some suggestion has emerged that the second-line therapy won’t be as effective for a patient who has already received an aromatase inhibitor and maybe not as good as after having failed front-line tamoxifen, as in a patient who is naïve to therapy.

In this trial we’re using a metronomic approach to capecitabine, administering a low dose daily, which we believe will be well tolerated. It’s a Phase II trial, but the goal is to improve time to progression for these patients with hormone receptor-positive disease who fail an aromatase inhibitor and are still receiving endocrine therapy.

We also believe that the data on the antagonism between hormonal therapy and chemotherapy are scant. No work has been done since the 1970s, and that work was done with tamoxifen, which is an estrogen agonist, so those data may not relate at all to the interaction between chemotherapy and an ER downregulator like fulvestrant or, for that matter, an aromatase inhibitor.

Anti-HER2 therapy for disease progression status-post adjuvant trastuzumab (Figures 38-39, 41)

DR SCHWARTZBERG: I agree with the general approach of deciding which anti-HER2 therapy to use based on the length of the disease-free interval.

Extrapolating from what we know about chemotherapy, the longer the interval between adjuvant therapy and relapse, the more likely the patient is to respond to similar agents or even the same agent. In previous years, that was tested and shown to be true. How long is long enough to go back to trastuzumab is an arbitrary decision.

We’re fortunate now to have a second anti-HER2 agent, lapatinib. However, when you’re considering these patients who unfortunately relapse with HER2-positive disease, you have to consider the long haul.

Undoubtedly, these patients will be exposed to both lapatinib and trastuzumab again, so it comes down to a sequencing question rather than simply picking one agent over the other.

I have patients who are living with extensive metastatic disease three, four or five years with a variety of anti-HER2-directed therapies. You have to simply make the decision as to which agent to use initially.

In years to come, we may learn that if a patient has failed trastuzumab, even at three years, lapatinib is the best drug to use, but our data set on that agent is much more limited than our data set on trastuzumab.

For patients who you believe may respond again, trastuzumab-based therapy makes sense. It is well known and has minimal toxicity. For patients who are refractory — that is, they relapse while on trastuzumab or within a few months or even a year — I’ll use lapatinib.

The cutoff I use is that if they finished trastuzumab less than a year or 18 months ago, I use lapatinib. We have data now that show the combination of lapatinib and capecitabine is effective in those patients.

Breast Cancer Update 2007 (3)

DR BRIAN LEYLAND-JONES: A big controversy existed about continuing trastuzumab on progression. It seemed people would say, “Well, if the patient had a good, prolonged response to firstline therapy, I might be more likely to continue the trastuzumab.” Will that now go totally out the window, and will people simply go to second-line lapatinib?

DR MARK D PEGRAM: A strong sentiment will probably emerge to change classes of inhibitors.

The lessons learned from other targeted therapy approaches — the estrogen receptor — are that by changing the strategy of therapeutic targeting, you might capture additional responses, albeit with perhaps somewhat lower frequency and not as long a duration, but nevertheless resulting in tangible clinical benefit.

This issue of trastuzumab duration in the metastatic setting has never been put to rest in a randomized clinical trial, which is unfortunate because once the tyrosine kinase inhibitors are available in the community, that question will probably become impossible to address.

Use of bevacizumab for metastatic breast cancer in clinical practice (Figure 42)

DR SCHWARTZBERG: I was surprised that 24 percent of the practicing oncologists indicated that they have not used bevacizumab for metastatic breast cancer.

It’s clearly related to reimbursement. I can tell you that in my own practice, which is generally a relatively liberal reimbursement environment, I undergo tremendous scrutiny about this.

Many carriers do not pay for bevacizumab at all because it’s not yet FDA approved, and from those that do pay for it, I am receiving approval only when it’s combined with paclitaxel, based on the data submitted to the FDA. I’ve tried using it with nab paclitaxel, too, and sometimes it’s reimbursed and sometimes it’s not.

An amazing aspect of being an oncologist that is endlessly fascinating is the different way people respond to receiving chemotherapy.

Some patients when faced with chemotherapy will say, “This is the time I want to get aggressive. I want bevacizumab with chemotherapy because I’ve read that it’s the best option.”

Patients with metastatic disease know they’re receiving therapy to keep their cancer under control.

Frequently they’ve received oral therapy, or sometimes a monthly injection, for years, but coming in for intravenous chemotherapy and all that goes with it — blood counts, anti-nausea agents, steroids and other supportive care medicines — changes the whole dynamic.

They see themselves in a different way, and treatment decisions are a collaboration between the patient and the physician.

Bevacizumab combined with endocrine therapy (Figure 43)

DR SCHWARTZBERG: I found it interesting that a quarter of the practicing oncologists have used bevacizumab with endocrine therapy for metastatic disease in practice. I have not used that.

We have a paucity of data, although at least one Phase II trial of letrozole and bevacizumab has been reported, and that showed a higher response rate with the combination than one would expect with an aromatase inhibitor alone.

The combination of bevacizumab and endocrine therapy does make preclinical sense, because some evidence exists for a feedback loop as you downregulate the estrogen receptor.

In this case, if you block estrogen production and ER activity decreases, then a compensatory increase in VEGF occurs, so some kind of feedback loop may be working in this situation through the hormonal system.

However, while it makes sense to do, I personally wouldn’t do it in the absence of more data. The regimen requires the introduction of IV therapy and a great deal of expense.

The increase in toxicity associated with the combination is probably modest, but in my opinion, it still needs more data support before it should be used off study.

Efficacy of bevacizumab combined with chemotherapy in the metastatic setting (Figure 44)

DR SCHWARTZBERG: In deciding whether to use bevacizumab in a case like this, I believe you have to go back to the clinical trial data.

If you examine ECOG-E2100, you see that it suggests that this type of patient treated with paclitaxel alone on the best schedule — weekly — will surprisingly have only about a 20 percent response rate.

However, with the addition of bevacizumab the response rate increases and, perhaps more importantly, the progression-free survival doubles, albeit with some increase in toxicity.

Generally, I believe it’s a good idea to add bevacizumab to paclitaxel in these cases, particularly for a patient like this who is taxane naïve.

Nab paclitaxel in combination with bevacizumab (Figure 44)

DR SCHWARTZBERG: From a scientific perspective, I don’t know of any reason whatsoever why nab paclitaxel should work any differently with bevacizumab than paclitaxel does. It’s the same drug, simply a different delivery system.

In fact, some evidence suggests that you may actually deliver more drug to the tumor with this agent because of the way the particles are distributed on the nanoparticle albumin. That would account for its milligram-per-milligram increased efficacy or, at least, the fact that we can deliver more drug with less toxicity.

In the end, it’s still paclitaxel, and I would agree that it’s generally a good idea to add bevacizumab to nab paclitaxel.

I believe that’s true in the adjuvant setting also, and if reimbursement were not an issue, I would use it in that setting because I believe it brings less toxicity and it’s easier to deliver.

Use of bevacizumab in elderly patients (Figure 45)

DR SCHWARTZBERG: For the majority of 85-year-old patients, I would probably not advise adding bevacizumab to paclitaxel, but I would be comfortable with that being done. I wouldn’t want to stress an 85-year-old heart with bevacizumab unless the patient had no comorbidities and was symptomatic from her disease, in which case I would consider it.

I believe it’s fairly clear, particularly from the work of Hy Muss in CALGB, that among older patients, response rates to standard therapies are similar but toxicity is increased, so we have to balance toxicity and efficacy.

Adding bevacizumab to paclitaxel does increase toxicity, particularly cardiovascular toxicity and hypertension.

In the ECOG-E2100 trial, 15 percent of patients had Grade III hypertension and required therapy.

XCaliBr data: Response to capecitabine with bevacizumab based on estrogen receptor status (Figure 45)

DR SCHWARTZBERG: At ASCO in 2007, George Sledge presented data from the Phase II XCaliBr trial that evaluated capecitabine with bevacizumab as firstline treatment in the metastatic setting.

He presented some unexpected data, but that’s why clinical trials are so endlessly fascinating — because they sometimes produce answers you don’t expect. They send you in new directions, and I would describe the XCaliBr trial as one of those that is hypothesis generating.

It was not a comparative study, but it seemed to show only a modest benefit from capecitabine with bevacizumab for the whole patient group.

That mirrors the original trial that Kathy Miller reported of capecitabine and bevacizumab in heavily pretreated patients, which did not meet its primary endpoint.

However, when George went back and analyzed the ER status, this study demonstrated a dramatically better response to the combination in patients with ER-positive versus ER-negative disease. The time to progression was more than doubled for the patients with ER-positive disease, and it was rather short for those with ER-negative tumors.

This is probably not what most people would have anticipated. Rather, one would have expected that the patients with ER-negative disease might actually respond better to chemotherapy, or at least as well as those with ER-positive breast cancer, but that’s not what we saw.

Still, this was a small trial, and this issue should be explored prospectively. However, considering these data, perhaps in the future we’ll confirm that it’s the patients with ER-positive tumors that benefit the most from capecitabine and bevacizumab.

Breast Cancer Update 2007 (2)

DR ANDREW D SEIDMAN: Currently, outside of a clinical trial I generally follow the ECOG-E2100 paradigm.

For patients who are not participating in our AC/nab paclitaxel/bevacizumab pilot trial but for whom taxanes are appropriate, I use paclitaxel and bevacizumab.

Occasionally, I will have patients who have received an adjuvant taxane within the past year and have relapsed, and my inclination at that point is to use capecitabine and bevacizumab, based on Kathy Miller’s reported Phase III trial. Those are probably the two most common scenarios.

Despite the doubling of the response rate, it does concern me that Kathy Miller’s trial did not show a significant increase in the time to progression with capecitabine. Certainly a difference is evident between that population and that of the E2100 trial with regard to the extent of prior therapy.

I don’t see any reason to suspect that the addition of bevacizumab to one particular cytotoxic agent in breast cancer versus another will make a big difference in terms of efficacy.

The RIBBON 1 trial, which allows a repertoire of commonly used chemotherapy regimens in the first-line setting, should inform us as to whether we need to worry about which agent we combine with bevacizumab.

My practice follows the highest level of evidence-based medicine. So when I use bevacizumab, for the majority of patients, I use it with paclitaxel.

There are certain unique circumstances in which paclitaxel is not appropriate, so I find the occasion to combine bevacizumab with other agents. Capecitabine would be the next most common agent followed probably by vinorelbine and gemcitabine.

I don’t think I’ve ever used an anthracycline with bevacizumab for metastatic disease. Primarily, I use taxane-based therapy.

Controversies regarding the continuation of bevacizumab upon disease progression (Figure 47)

DR SCHWARTZBERG: In responding to this question, more than half of the practicing oncologists indicated that they would present patients with the option of continuing bevacizumab and switching to another chemotherapy at the time of disease progression, yet only about 20 percent of investigators said they would do so.

I believe this reflects the fact that the practitioners tend to see not only breast cancer but all types of cancer and they draw inferences — as I do, myself — from other diseases when appropriate.

In colon cancer, we have the BRiTE registry data that show a significant benefit to continuing bevacizumab beyond disease progression, although it’s biased by the fact that they are longitudinal registry data. We participated in the registry, and it’s possible that the majority of the clinicians you polled participated also because it was a large registry.

I believe clinicians are extrapolating those data from colon to breast cancer, but I suggest a couple of notes of caution. First, we have the negative capecitabine with bevacizumab trial in breast cancer, although that wasn’t second line. It was in the third-line setting and beyond, so that’s perhaps different from the colon data.

In addition, as the data are maturing in colon cancer, a suggestion is emerging that bevacizumab might have a ceiling effect in the sense that if you administer better first-line chemotherapy in colorectal cancer, you obtain less incremental benefit from the addition of bevacizumab. In other words, it worked well with IFL, which wasn’t as good as FOLFOX or XELOX, and yet when you add bevacizumab to those regimens, one interpretation is that it confers some benefit, but it’s a modest benefit.

If you extrapolate that to breast cancer, then we probably do need to see the results of the RIBBON 1 and other ongoing trials to make sure that bevacizumab added to chemotherapy is the same across different types of chemotherapy. We don’t know the answer to that yet.

Cardiovascular toxicities associated with bevacizumab (Figure 48)

DR SCHWARTZBERG: The clinical trials show that 15 to 20 percent of patients on bevacizumab will require antihypertensive therapy.

As for thromboembolic events, that might vary somewhat from disease to disease and is also probably age and comorbidity dependent. I believe that in the breast cancer trial, it was around four percent.

It’s becoming clear now that most of the thromboembolic events with bevacizumab are arterial and probably associated with a minimal increased risk, if any at all, of venous thromboembolism.

You have to screen your patients carefully, and those who have arterial problems or significant coronary artery disease need to be evaluated carefully before considering bevacizumab.

Those patients who have preexisting renal disease, particularly proteinuria of any degree, should probably not receive the drug. Nor should patients who have hard-to-control hypertension, but all other patients are probably good candidates.

Click on the image to enlarge

Select Publications