Professor of Surgery and Consultant Surgeon at the University of Edinburgh; and Clinical Director of the Edinburgh Breast Unit at the Western General Hospital in Edinburgh, Scotland.
J. Michael Dixon, MBChB, MD
Professor of Surgery, Consultant Surgeon, University of Edinburgh Clinical Director of the Edinburgh Breast Unit, Western General Hospital, Edinburgh, Scotland.
Dr Dixon: I think there are a number of advances. Number one is that we’re diagnosing cancer at an earlier stage. That’s partly due to screening, but it’s also due to better patient education, women coming forward with small lumps, better investigative tools, better radiology, better image-guided core biopsies, etc.
The second big advance is in treatment. More patients are now receiving breast-conserving surgery, and the surgery now produces much better cosmetic outcomes. We now understand that big surgeons don’t need to make big incisions—that is, surgeons don’t need to remove the cancer with a lot of surrounding tissue. There has been a dramatic movement away from routinely removing all the axillary nodes with the advent of sentinel node biopsy. That’s been a major improvement. It has improved the speed with which patients recover after surgery. Because most of the patients are node-negative, they can have a sentinel node biopsy as their only axillary node procedure. This reduces the number of women who get significant arm problems and lymphedema after surgery. More recently, we’ve learned that perhaps some women with positive nodes after sentinel lymph node biopsy don’t need an axillary lymph node dissection even if they do have one or two positive sentinel lymph nodes.
The other advance in treatment is in systemic therapy. For hormonal therapy, we have the aromatase inhibitors that have demonstrated superiority over tamoxifen in postmenopausal women. In chemotherapy, there has been wider use of taxanes and anthracyclines compared with standard cyclophosphamide, methotrexate, and 5-fluorouracil regimens. Then, of course, there is the discovery of anti-HER2 drugs with Herceptin or, recently, with lapatinib, and pertuzumab.
The first lesson we’ve learned is that neoadjuvant endocrine therapy is very effective therapy in older women with larger cancers. For women who have a lot of other morbidities, you can treat these patients with a drug like letrozole or anastrozole alone, and that’s probably the only treatment they’re going to require. It at least controls their breast cancer until the time of their death. So, the first thing is that these new drugs alone are sufficient treatment.
The second thing we’ve learned is that endocrine therapy works differently than chemotherapy in how it shrinks the cancer. It’s a lot slower. The treatment duration that’s required to shrink the cancer down to a size where you can perform breast-conserving surgery can be years—rather than a few weeks, which is the case for neoadjuvant chemotherapy. In other words, you might give 4 to 8 cycles of neoadjuvant chemotherapy over a period of 12 to 24 weeks, whereas neoadjuvant endocrine therapy is given over a period of 9 to 10 months.
We also find that in neoadjuvant endocrine therapy the cancer implodes and shrinks from the center. It produces a central scar and pulls the cancer inward. That’s great news for the surgeon because if you’ve got a 4-cm cancer and it implodes, you end up with a 1-cm cancer by the time you’re finished with endocrine therapy.
The patterns of response to neoadjuvant chemotherapy aren’t the same. Neoadjuvant chemotherapy is sort of like a shotgun approach. You pull the trigger and 90% of the cancer disappears, but the same area of the breast is affected. So, when the surgeon operates and removes the cancer that may be left, he or she often finds that the area of breast involvement is still very extensive.
The other important point about endocrine therapy is that if you have a patient with estrogen receptor [ER]-positive disease, she can take medication for several weeks and then have her operation when she wants to. If the patient is going away on holiday or needs to work a few weeks before her operation, you can put her on an aromatase inhibitor and schedule the operation when it’s convenient for the patient. Breast cancer isn’t the same critical event anymore; it’s now a survivable disease. So, patients can actually plan surgery around their life.
Researchers are now investigating giving endocrine therapy for a couple of weeks before and after surgery to determine the long-term effects. Because endocrine therapy switches off proliferation in most cancers, giving a short period of endocrine treatment and seeing whether the cancer responds with a switch-off proliferation, researchers may be able to predict long-term outcomes better than even some of the standard genomic profiles.
Another important aspect of neoadjuvant therapy is that it gives you tremendous insight into how these drugs work and why they don’t work. Preoperative drug administration is a tremendous way of getting insight into how we move forward with this disease and understanding the biology, which is the other biggest change I’ve seen over the past 5 years. Doctors are doing studies where the endpoints are biological rather than survival-based because patients are doing that well and few relapse anymore.
Exclusive interview with
If you take a group of older patients, shrink down their cancer, do breastconserving surgery, and give them radiotherapy, their chance of getting local recurrence 5 years and beyond is minimal. It’s less than 2% in 5 years and less than 4% in 10 years. It’s remarkably effective treatment. We not only save breasts, but, in the longer term, maintain good local control.
First of all, there always has been a view in the United States that chemotherapy is better. There’s been a fear of endocrine therapy because it takes longer to work. When the patient comes back for follow-up, the cancer is not gone, and that worries the patient, the oncologist, and the surgeon. But, actually, you just need to be patient sometimes. We know, for instance, if we’re going to give adjuvant therapy, we administer the treatment fairly quickly. If you give 6 cycles of adjuvant chemotherapy, it’s over in 18 weeks. You don’t give 18 weeks of tamoxifen or aromatase inhibitors— you give endocrine therapy for 5 years. If you’re going to use endocrine therapy in the neoadjuvant setting to get the benefit, you need to give it for a longer period of time.
I also think it is interesting to consider why people select certain treatments. The primary reason why patients choose specific treatments is because they think the doctor believes that treatment is right for them. I think a lot of oncologists who use endocrine therapy use it because they see it work.
Older patients actually don’t want extensive surgery. Plus, they often aren’t in the best physical condition. They don’t want to lose their breast any more than a younger woman, and they want to be involved a little bit more in the decisions about treatment. If given the option of doing treatment with an aromatase inhibitor, I think a lot of patients would choose that option.
Nevertheless, I think the really big driver here is that oncologists try it and see that it works. I think there’s a fear that the cancer will get bigger. Surgeons are concerned they will lose the opportunity to operate, but once they see how effectively it works, they won’t have that fear. Progression occurs in less than 5% of patients and can be detected early by regular monitoring.
It’s not yet in, although the reason I keep talking about letrozole is because the drug has a product license in the United Kingdom for use in the neoadjuvant setting. As it happens, the other drugs don’t. It’s not that letrozole is that much better than anastrozole, it’s just that it happens to have the product license. It’s now starting to be part of accepted protocols.
Most people get 4 to 8 cycles of neoadjuvant chemotherapy: 4 cycles roughly 3 weeks apart over a 12-week period. We actually use 6 cycles given at 3-week intervals over 18 weeks. Some studies have used 8 cycles, which are given over 24 weeks. So, the time period for neoadjuvant chemotherapy can range between 12 to 24 weeks.
Neoadjuvant endocrine therapy, on the other hand, is given over an average duration of 9 to 10 months.
We did a study where we treated about 250 patients with neoadjuvant endocrine therapy and looked at the time it took the cancers to shrink to a size suitable for breast-conserving surgery. Some patients took letrozole for 2 years before we operated. Because some tumors are really big, shrinkage can be slow.
If you look at some of the pictures I presented at the meeting, you’ll see hugely dramatic changes over time. I’ve seen cases where women present with a tumor that is literally the size of a baseball. We shrink that down with an aromatase inhibitor over 2 years so I can take out the small residual tumor and save the breast.
The first rule of medicine is do no harm. Most people seem to forget that, in their rush to try to benefit the patient, they can do harm. I think we’re starting to learn that sometimes too much treatment is as bad as too little.”
Often the patients have ignored them for a bit because these older women don’t want to come to the hospital for treatment. The idea that patients can go away with a pill for a few weeks until they come to terms with what is going to happen and the knowledge that they will likely need less surgery is very reassuring to them.
Forty percent of all the patients we see with breast cancer are over the age of 70. A lot of these patients aren’t suitable for neoadjuvant chemotherapy, and some are infirm and have other comorbidities. Initially treating with letrozole gives patients time to work out how they’re going to cope with their disease, and gives the oncologist time to talk to the patient and the family to work out what needs to be done to get them fit and prepared for surgery. If you’re going to operate, operate when patients are in the best condition you can get them in; that way, they’re unlikely to develop problems after surgery. The first rule of medicine is do no harm. Most people seem to forget that, in their rush to try to benefit the patient, they can do harm. I think we’re starting to learn that sometimes too much treatment is as bad as too little. Actually, less is more—fewer and fewer surgeries are being done today.
We’ve identified genes that change within 2 weeks of starting a drug like letrozole. These genes can actually predict whether a patient is going to get a substantial reduction in the tumor. We’re now validating that on the second series where we’ve treated about 350 patients with neoadjuvant letrozole and have obtained tissue samples during their treatment.
As I’ve explained before, if you give a drug for 2 weeks before surgery, by the time of surgery, knowing what that drug has done to the cancer might also help oncologists make informed decisions about whether to continue that drug alone, whether to switch drugs, or whether to add some other treatment. We believe that dynamic changes— how a cancer responds to a drug—are probably more important than where the cancer originated.
The other thing we’re interested in is specifically looking at HER2-positive cancers treated with endocrine therapy. We want to analyze how the pathways that change in HER2-positive and HER2-negative cancers differ.
The other work we’ve done in studying the biology of these tumors involves a group of patients treated with adjuvant tamoxifen who have recurred either within 3 years or after 10 years. We have tumor tissue for these patients from 20 years ago. We’ve looked at that tumor as a diagnosis and we’ve identified pathways. There are a large number of redundant pathways in those that relapse early that are important, and there are different pathways in those that relapse late. We have validated that finding in a second set of cancers and plan to present that data at the San Antonio Breast Cancer Symposium later this year.
We think there may be as few as 4 genes that ultimately identify 80% of ER-positive cancers that will respond to treatment. Some of these genes relate to proliferation; others are so-called “stromal genes.” The reason that the stromal genes increase is because once the cancer dies, what fills the void is fibrosis and scarring. We’ve got to work toward developing a relatively simple test that allows us to say to our patients, “You would do very well on this drug. You don’t need anything else. You don’t need chemotherapy.” The trouble is that there’s a very big void for those who don’t do very well. That’s why we and many others are doing studies trying to identify what pathways are active in these resistant cancers.
Exclusive interview with
We are also looking at proteins in the immunohistochemistry that are related to these gene pathways. Having identified the genes that change, we’re now trying to develop a simple test like OncoType.
However, my concern with OncoType and MammaPrint is that they currently take what we already know and they measure it. OncoType measures ER, progesterone, HER2, and proliferation and some other genes very accurately, and combines them all together. That gives you a probability, which is not enormously helpful to the individual. What we want to know is how long the patient is going to survive and what treatment will make them survive longer. The current tests don’t do that yet.
What we need is a test that would help determine which drug is going to work for individual patients. That’s why we’re trying to identify the pathways that are actually in resistant cancers so that we can either combine that drug with other drugs or try different approaches. Essentially, what we’re trying to do is find how a cancer responds and get more information than just analyzing it at the time of diagnosis.
Part of the problem facing community oncologists is that not all the new drugs have reached them yet, and they may not get them for a long time. I think the big challenge is how to combine these drugs.
One of the things that was discussed at the meeting is how to kill resistant cancers. Well, you could give patients 21 different drugs and knock off every pathway, but this approach would knock off the pathways in normal cells too. So, getting that combination is going to kill the cancer, but it may also do considerable harm to the patient.
The reason that antibiotics are effective is that they exploit huge weaknesses in the bacteria. There’s a difference between treating bacteria and normal cells. With chemotherapy, for instance, we’re trying to hit the cancer’s main strength: its increased growth rate. But the more effective a drug is at hitting proliferation, the more side effects you get. We’ve got to get much smarter by targeting treatments. The great news about the HER2 story is that it shows that if you find the target, you’re going to get the benefits. The difficulty is in finding new targets.
I think it’s going to be a long, tough road ahead with lots of studies and lots of blind alleys. So, the community oncologists just have to optimally apply what we already know—and be careful not to overtreat.