COMMENT FROM THE TREATING PHYSICIAN

ABRAHAM B SCHWARZBERG, MD: I treated her for approximately six months with paclitaxel/trastuzumab and she had a complete clinical response, but she developed severe nail problems, pain and fatigue. At that point, we decided to take a break to address local control of her primary tumor, and she underwent a lumpectomy with axillary dissection. She was premenopausal, so we also removed her ovaries at the same time, as a hormonal therapeutic maneuver, and we started tamoxifen.

The question then became, what do we do about the liver metastasis? In the treatment of metastatic disease there is no clear indication of the correct course of action. The decision was made to rebiopsy the liver nodule and perform radiofrequency ablation (RFA). No disease was evident during the biopsy or RFA. Necrotic tissue was found but nothing was viable, and she tolerated the procedure fine. She continues to take tamoxifen.

Select Commentary from Interviews Conducted at SABCS 2008

DR HAYES: A number of philosophical approaches to this patient are available. One is to acknowledge that she has metastases, and the treatment is palliation. Therefore, one wants to choose a treatment most likely to work with the fewest side effects to keep her feeling normal as long as possible.

I tell my patients, “The bad news is you have metastatic breast cancer, and we will probably not cure you. The good news is you have metastatic breast cancer, and we have many treatment options now.”

Many of those options are interchangeable, and the key is to work with the patient and her family to find the best treatments for that patient. There’s no right or wrong there.

If she is totally asymptomatic, has a single liver metastasis and has no evidence of end-organ dysfunction, we have lots of “wiggle room.” I would probably start her with endocrine therapy, with the specific choice dependent on what we wish to accomplish.

It would be reasonable to start her on tamoxifen. One could also make a case for tamoxifen and trastuzumab, but then she is tethered to your clinic because she will have to come in every three weeks for IV administration. Others would use ovarian ablation and tamoxifen. Data suggest that the response rate is higher with the combination, but no data indicate that survival is improved with that approach.

One could administer chemotherapy/trastuzumab, but what is the objective of that approach? This woman is in no danger of dying from her disease in the next two months. In fact, she’s in no danger of having substantial end-organ dysfunction that would preclude her from receiving chemotherapy and bevacizumab down the road.

Evidence does not indicate that she will live longer by receiving up-front chemotherapy/trastuzumab. You will be subjecting her to toxicities that she can probably avoid — although, frankly, you will be administering this treatment down the road.

If the treating oncologist consulted with me and expressed a desire to treat her with chemotherapy/trastuzumab, I wouldn’t argue strongly against that approach — it’s simply not what I would do.

DR MILLER: From the long-term follow-up series at MD Anderson, we know that a small group of patients are potentially cured of their metastatic disease, or, at least, long-term survivors have been followed for up to 10 years without evidence of recurrence or more cancer.

We don’t know a lot about who those women are, but we do know who they are not. They are not women who are symptomatic from their metastatic disease or have multiple, bulky sites of metastatic disease. They are not women who don’t have a complete response to their systemic therapy. These women were predominantly treated with anthracycline-based regimens and hormone therapy, if that would have been appropriate for their disease. None of them were assessed for HER2 or treated with HER2-targeted therapies.

I tell these patients that in general metastatic disease is not curable, but some exceptions occur. In a younger patient who has limited, asymptomatic metastatic disease, it would be reasonable to begin with the assumption that a few women are long-term survivors, and a key component would be a complete response to initial systemic therapy.

Therefore, we would start with an aggressive plan that would offer her the best chance of achieving a complete response, and then the rest of our decisions would follow from how well she responded. For this woman with a HER2-positive tumor, I would start with something like the TCH regimen. She would be eligible for ECOG-E1105, which randomly assigns women to paclitaxel with trastuzumab and optional carboplatin and then randomly assigns them to bevacizumab or not.

I have a patient on ECOG-E1105 who is exactly like this woman, and if she has a complete response to chemotherapy, we will proceed with local therapy for her primary tumor — whether that is lumpectomy or radiation therapy or axillary dissection — continue trastuzumab and then add in hormone therapy.

If she has a complete response to her initial therapy, hormone therapy will offer her the best chance of long-term survival. With that goal in mind, I would make certain that her ovaries were fully suppressed or removed and then I would administer tamoxifen to maximize the benefit of her hormone therapy. The difficult question is how long to continue trastuzumab if she has a complete response. Treatment until progression reverts to the mind-set that the disease is not curable. I have an equal number of women who pick an arbitrary time to discontinue trastuzumab and others who continue trastuzumab.

If she doesn’t have a complete response, then the best data we have indicate she does not have the possibility of long-term survival and we need to shift our goals to preserving the quality of her life while still trying to extend it. So we would perhaps make different decisions about local therapy, about more aggressive hormone therapy and about how long we might continue the trastuzumab.

DR HAYES: I don’t believe that resecting a liver metastasis does any good in breast cancer. The data simply do not support that approach.

Many years ago, Don Morton performed metastatectomies on a variety of sites and suggested that those patients who had a metastatectomy lived longer than those who did not, but of course he chose patients who were most likely to be able to undergo the resection.

If you can perform the resection of the liver metastasis laparoscopically and do it safely, then it may be okay. However, any time you touch a patient with a procedure, a risk is incurred, and I don’t see any evidence that this will improve her outcome.

I have sent patients for isolated hepatic treatment, but more commonly at our institution it’s a radiation therapy approach. Our group is not performing RFA, but I believe it’s reasonable.

Frankly, most of us would only do this for patients who have an isolated liver metastasis and a long disease-free interval, which are the patients who will live a long time anyway. I might consider local treatment for the hepatic lesion if the patient wanted to avoid systemic therapy for the next couple of years, but I don’t believe you will cure the patient.

DR MILLER: I can’t say that resection of the liver metastasis is an unreasonable consideration for a patient who, by your best imaging, has one lesion, but I’m not a fan of this approach and, in general, I haven’t done that. The bulk of the data that we have suggests that one lesion is the one you can see but many more exist that you cannot see.

RFA is another approach that’s often brought up as a way of administering local therapy but perhaps with less toxicity and less recovery time than surgery. However, I don’t believe it will meaningfully change her likelihood of being a long-term survivor or the quality of her life. So it’s not something that I’ve recommended to patients in this situation.

COMMENT FROM THE TREATING PHYSICIAN

KENNETH R HOFFMAN, MD, MPH: The patient accepted anastrozole and trastuzumab and had a dramatic response. The tumor in the breast is now resectable and the lung masses have essentially disappeared.

The question now is whether to perform a proactive mastectomy for local tumor control, which she’s willing to do, or do we simply continue systemic therapy indefinitely?

Select Commentary from Interviews Conducted at SABCS 2008

DR HAYES: All of us have patients like this, and I find them both challenging and rewarding. As in society in general, if all of our patients thought exactly alike it would be a boring place.

These patients enrich our practice, and I enjoy taking care of them. It’s always sad when a patient refuses therapy — out of...I don’t want to say ignorance because that sounds pejorative, but out of a lack of understanding of what we can do for them with few side effects.

The challenge is to help this patient understand that we can do her a lot of good without a lot of toxicity. This is when a thoughtful and compassionate physician and, more importantly, nurse, can be of great value. My nurse practitioner routinely saves me because she loves these folks and gets involved.

I would recommend antiestrogen therapy for this patient and would probably add trastuzumab. My guess is that she’ll fare reasonably well with that. I have a patient like this whom I have treated for a number of years. She and I have become close, but she refuses to admit that the benefit she’s experienced is because of my treatment. She always attributes it to complementary treatments that she engages in.

I would probably recommend removal of the breast lesion for palliation, depending on how much response she’s had to systemic therapy. Then the question is whether to resect the lesion and add radiation therapy or simply irradiate the lesion if she’s had a nice response.

These treatments will not prolong her survival, but the goal is to keep the lesion from breaking down in the future. In my practice, we probably see six to 10 cases like this each year. Granted, we serve several million people, and we find it’s not only the uneducated person who delays seeking medical attention.

Indeed, the patient may be well educated. My impression is that it’s a mixture of people from all walks of life.

I believe, although I’ve never seen this documented, that these cases involve a fair amount of spirituality. Some people believe their faith will lead them through this problem, so they keep ignoring the issue. It behooves us to work with that. We cannot swim upstream and fight that theory, so I try to make patients understand that I want to work with their beliefs, and frequently we can do that. It’s a matter of gaining trust and doing the right thing.

DR WINER: Although we all wonder how a patient can let this go on for so long, I’d remind people that when you get into a car accident and dent your fender, at first it looks scratched up but after a while you stop noticing it. I believe that the longer you ignore something, the easier it is to ignore it.

Often when people finally do come in, I find they’re open to receiving treatment. Of course, patients delay for a range of reasons, but it may not be a pathologic process.

For treatment of the primary tumor, we have a number of retrospective studies evaluating whether to remove it in the face of metastatic disease, and most, if not all, suggest a benefit, even a survival benefit. The problem is that they are all deeply flawed. They’ve all attempted to control for all the various covariates, but they can’t.

What we need is a prospective trial. Seema Khan will be conducting a study through ECOG evaluating surgery performed once a patient responds to systemic therapy versus delaying surgery until the time of progression. I believe that will be an important trial.

For now, whether to remove the primary tumor is patient dependent, but the cases in which I consider it involve low-volume metastatic disease, typically when a response to therapy has occurred and, in particular, when a patient, upon hearing that it’s not clear what to do, feels strongly about wanting to do something in the way of local therapy.

DR BURRIS: I’ve seen two patients like this in the past six months, and both had extenuating life circumstances that contributed to their delay in seeking treatment.

One lost her home in a tornado, and the other had a spouse who was ill. So some women neglect to come in not because they don’t want medical care but because they put their family first and end up with a late diagnosis.

We have to assess the situation and find out what all the factors are that caused them to delay seeking care. If it really is a situation of significant denial associated with depression, then we get them involved with a psychologist.

For some patients, their initial inclination is that they don’t want chemotherapy, but that may change after they speak with you. It’s a day-to-day decision, and we must be open minded about how we treat these patients.

Although this woman says she won’t take chemotherapy, she still has many choices, and it’s quite a decision. She could take any one of three aromatase inhibitors or tamoxifen or fulvestrant. She also could choose biologics — lapatinib, trastuzumab or bevacizumab — and they can also be combined.

We have data on all these agents. In addition, she could at some point consider entering a clinical trial of trastuzumab-DM1.

DR GRALOW: We sometimes see patients who refuse any kind of therapy from a conventional provider, but we do the best we can for these patients and we try to be there if they change their mind.

I’ve had patients whom I referred, at their request, to a naturopath, and I’ve continued to follow them jointly.

Eventually most of these patients come back to us for some form of systemic therapy, sometimes when they become symptomatic or frightened enough that they want to be more aggressive.

I’m certain a few have never come back for treatment, but regardless, I believe our job is to be there for them, listen to their concerns and work with them.

COMMENT FROM THE TREATING PHYSICIAN

STEVEN P KANNER, MD: This patient wanted a treatment that wouldn’t require much effort on her part and would be covered by insurance. We put her on fulvestrant, and she’s shown a good response to therapy. We have seen improvement in her scans, and her tumor markers are normal. However, her markers were normal before, so I’m not certain why a bone scan was originally ordered.

Select Commentary from Interviews Conducted at SABCS 2008

DR HAYES: The first thing I would do is put this patient on a bisphosphonate, which, if nothing else, would substantially prolong how long she remained asymptomatic.

I would want to be certain that the bone scan was truly positive. I always tell my fellows to review the plain films or CT to be certain they are lytic lesions.

Also, this is a case in which I believe circulating markers can be helpful. If the patient has circulating tumor cells or an elevated CA15-3 in the 50 to 100 units/L range, that indicates metastatic breast cancer, and it could be almost nothing else.

If the bone metastasis consisted of an isolated, single focus and the plain films and markers were negative, I would either do nothing and repeat the scan later, or I would biopsy the lesion, depending on what the patient wanted to do.

The alternative is probably one of the best therapies we have, and that’s tamoxifen. By switching her from an aromatase inhibitor to tamoxifen, we are administering a different antiestrogen therapy, and even if it doesn’t benefit her, I don’t believe it will cause any harm.

WILLIAM J GRADISHAR, MD: I wouldn’t have ordered imaging studies for this patient. However, once I knew the disease was present, I would start endocrine therapy in addition to a bisphosphonate. I believe that exemestane and fulvestrant are equivalent choices, and a case could even be made for tamoxifen.

Biopsies of bone-only metastases can be problematic because a negative result can be caused by a sampling error. So one can make a case for treating empirically with endocrine therapy. However, if the patient developed any visceral or soft tissue disease that could be biopsied easily, I would do so.

In cases such as this one, treatment is palliative, not curative, so the goals of selecting an endocrine therapy include delaying the time until chemotherapy is needed and maintaining quality of life. In managing metastatic disease, we don’t have to see the lesions disappear. Stable disease is an acceptable endpoint, and it has an outcome that’s equivalent to objective response. This concept is now used with chemotherapy, but it was initially identified with endocrine therapy.

Certainly with bone-only disease, evidence supports continuing endocrine therapy as long as you don’t see disease progression. It would be great if all evidence of metastases disappeared, but if the bone scan remains stable, that’s acceptable. I would use endocrine therapy to the point of exhaustion, as long as the patient doesn’t develop a visceral crisis or another problem that motivates us to use chemotherapy. A fraction of patients will respond to successive maneuvers with different endocrine agents.

We’ve spent 20 years trying to figure out which endocrine agent or sequence is best in these cases. The bottom line is that it probably doesn’t make a big difference. You can switch between classes of endocrine therapy, or within the class of aromatase inhibitors you can switch between a steroidal and a nonsteroidal aromatase inhibitor. In EFECT, fulvestrant was compared to exemestane among patients whose disease progressed on a nonsteroidal aromatase inhibitor, and the outcomes were absolutely identical regarding side effects, clinical benefit rate, objective response rate and time to disease progression.

One of the questions this case brings up is whether we should order general screening tests for asymptomatic patients such as this one. I believe that the simple answer is no. This is a difficult issue to explain to patients. To illustrate that point, only a few days ago, a group of clinicians, including myself, Cliff Hudis and Hal Burstein, were meeting with a group of patient advocates. We were to update what was new in the NCCN guidelines.

The first point was whether we should conduct PET scanning, tumor markers, et cetera, for these patients. Cliff succinctly answered, “Absolutely not.” After 45 minutes, we were still trying to explain to the advocates why it isn’t a good thing for a patient who is otherwise well. The fact is that we have little evidence that identifying asymptomatic metastatic disease sooner, whether it’s two weeks or two months, will make a significant difference in overall outcomes for patients, but this can be a difficult concept for some patients to understand.

COMMENT FROM THE TREATING PHYSICIAN

LOWELL L HART, MD: I treated this woman with single-agent, weekly nanoparticle albumin-bound (nab) paclitaxel, which has been well tolerated. This led to a rapid tumor response. Based on the most recent PET/CT results, she has experienced a near-complete response.

The treatment has not caused any side effects other than having to come to my office briefly each week. She has had no difficulty with blood counts and absolutely no neuropathy. The primary issue here is should bevacizumab be used whenever a taxane is administered as first-line therapy?

Select Commentary from Interviews Conducted at SABCS 2008

DR HAYES: This patient’s disease may or may not be endocrine refractory. She’s never received tamoxifen, which is a good drug. However, she is developing end-organ dysfunction, and this patient doesn’t have a lot of wiggle room. She already has shortness of breath.

If you administer tamoxifen and two or three months later you discover that was the wrong choice, then you’re in trouble. Her symptoms will be much worse, which makes it even more difficult to administer chemotherapy.

So I would probably proceed with chemotherapy and use paclitaxel/bevacizumab, which seems like the perfect first-line therapy.

Because this patient had a great response to nab paclitaxel, Dr Hart questioned whether he should administer bevacizumab or whether he could administer it with another, second-line chemotherapy when the disease progresses.

I don’t believe we’re obligated to use bevacizumab with a taxane up front. I would probably administer bevacizumab as first-line therapy because we have a prospective, randomized trial suggesting that time to disease progression is prolonged.

If you are in a practice in which bevacizumab is not a viable option, I can’t argue that it’s the wrong decision. Let’s recall that the taxanes are effective and our goal is palliation, so a taxane by itself is appropriate.

In terms of administering bevacizumab after she progresses on the taxane, we are getting outside of evidence-based data. We don’t have evidence that the next line of therapy with or without bevacizumab is better. We know that after progression on a couple of prior therapies, capecitabine in combination with bevacizumab is not effective.

I have used bevacizumab in that setting, and a couple of patients have fared well. Was that because of the bevacizumab or because the next chemotherapy was effective? I don’t know, so I don’t believe there’s a right or wrong answer. I have spoken at great length with Kathy Miller, who is probably the leading investigator of bevacizumab in breast cancer, and she doesn’t believe the benefits in the ECOG-E2100 trial of paclitaxel/bevacizumab are specific to bevacizumab.

She believes that the specific chemotherapeutic agent doesn’t matter. Rather, it’s the earlier use that likely makes it more effective. Biology is consistent with that belief. Once angiogenesis is turned on, it is difficult to get rid of it compared to preventing new angiogenesis and, therefore, new metastases.

When I have a patient with metastatic disease for whom I’m considering first-line therapy, I ask the patient, “Do you want to take a pill or do you want IV therapy?”

This woman has bone metastases, so you will use bisphosphonates anyway, and it probably won’t matter to her if she receives IV chemotherapy. If she had lung-only metastases and you didn’t want to administer bisphosphonates, it would be reasonable to administer capecitabine by itself.

If nab paclitaxel had been developed as the first taxane and was off patent now and inexpensive, it would be our drug of choice. Unfortunately, nab paclitaxel was our third taxane, and the benefits compared to standard paclitaxel are not great.

The occasional patient who has terrible reactions to standard paclitaxel probably doesn’t experience the same reactions with nab paclitaxel. You may also be able to administer nab paclitaxel a little longer because it’s more likely to be effective.

DR GRALOW: I probably would have opted for paclitaxel and bevacizumab. Also, it would be reasonable to enroll this patient on the CALGB-40502 trial, evaluating paclitaxel/bevacizumab versus nab paclitaxel/bevacizumab versus ixabepilone/bevacizumab.

I’m impressed with nab paclitaxel and find it easier to administer than paclitaxel. It doesn’t require premedications, we see less allergic reactions and patients prefer it. In many cases we seek insurance preauthorization and, if we obtain it, then that’s my choice. If not, I choose paclitaxel.

DR BURRIS: If a patient has triple-negative breast cancer, then I typically decide to use weekly paclitaxel and bevacizumab. I am certainly influenced by the ER/PR positivity, and if a patient is asymptomatic, then I may start out with weekly paclitaxel or capecitabine. However, the data are much weaker for bevacizumab in the second- or third-line setting, so I find fewer and fewer patients for whom I don’t offer a taxane with bevacizumab up front.

COMMENT FROM THE TREATING PHYSICIAN

MICHAEL A SCHWARTZ, MD: This is the only patient I have treated with bevacizumab who had symptomatic nephrotic syndrome. This patient responded to bevacizumab treatment and then progressed off of it. The nephrotic syndrome was reversible. She experienced proteinuria but did not have a significant increase in her creatinine. So is it worth the risk of rechallenging with bevacizumab in this type of situation?

Select Commentary from Interviews Conducted at SABCS 2008

DR HAYES: The development of nephrotic syndrome with bevacizumab has been reported, but it is uncommon. I agree with George Sledge and Kathy Miller, who commented that, in general, bevacizumab is a well-tolerated drug with occasional serious adverse effects. These include hemorrhage, hypertension and renal dysfunction. In this patient, I would not have pushed the bevacizumab. I probably would have used it as this physician did, but at the first sign of nephrotic syndrome, I would have stopped it because I don’t believe you will improve her survival much with it, and now you’re making her sicker rather than better.

DR MILLER: I have seen nephrotic-range proteinuria, but without the other symptoms of nephrotic syndrome, in two patients. Grade III/IV proteinuria occurs in only about three percent of patients treated with bevacizumab.

I would stop the bevacizumab for this patient. She has developed two major bevacizumab- associated toxicities. She had difficult-to-control hypertension, which is also a rare situation. She received a modest reexposure to bevacizumab with capecitabine and had more trouble with hypertension and nephrotic-range proteinuria. Her disease is progressing, and we don’t have any data that continuing bevacizumab beyond disease progression is helpful.

Select publications