• Symposium Overview
• An Emerging Theme
• Gene Expression Profiling, Proteomics and the Future
• Caveats and Concerns
• Avastin and Iressa
• A Biomarker for Chemotherapy Benefit?
• Understanding Hormonal Resistance
• Dose Dense vs. Standard Combination Chemotherapy
• ATAC Trial Update [Arimidex vs. Tamoxifen]
• HER2 News
• Zometa vs. Aredia
• Neoadjuvant Chemo Combo
• To Find Out More
• About Musa Mayer

Symposium Overview

The San Antonio Breast Cancer Symposium is the largest annual medical conference dedicated exclusively to breast cancer research, doubling in size in the five years since I first attended. With nearly 5,000 participants from more than 60 countries at the meeting held on December 11-14, 2002, physicians and other medical professionals were joined by breast cancer researchers, industry people and advocates from around the world who had come to present and listen to the latest research findings. If you follow the research literature, as I do, it's always a thrill to see all the top names in San Antonio every year.

"The objective of the SABCS," the organizers state, "is to provide state-of-the-art information on the experimental biology, etiology, prevention, diagnosis, and therapy of breast cancer and pre-malignant breast disease to an international audience of academic and private physicians and researchers."

Each year, more patients and advocates come to gain important insights into current and future research that they can carry back to their communities of support. The Alamo Breast Cancer Foundation sponsors an excellent scholarship and mentoring program for advocates, which I attended this year. Contact www.alamobreastcancer.org for more information.

For patients and advocates attending for the first time, SABCS can be a somewhat disheartening-not to mention disorienting-experience. One soon realizes that despite the vast amount of research being done, real progress is still painfully slow, as small studies need to be replicated on a larger scale, and outcomes need to be verified before new treatments are ready for the clinic. Study results are often conflicting. Sorting all of this out inevitably takes years-years that many patients don't have. Nevertheless, what attending this conference offers us goes well beyond the usual media sound bytes. It is the stuff of real science, where progress is almost always measured in small, incremental steps, with far more dead ends and wrong turnings than bulls-eyes.

An Emerging Theme

Despite the diversity of content in this year's SABCS, with over 500 posters and dozens of slide presentations, mini-symposia, plenary talks and satellite meetings, a predominant theme emerged, at least for this observer.

Each year, more presentations at the San Antonio Breast Cancer Symposium have focused on the potential of what has been called the "genetic fingerprinting" of breast cancer tumors. This group of technologies examines many interacting genes at once that govern how cancer cells behave, with a goal of individualizing treatment by identifying what scientists have called the "malignant phenotype," in other words, what makes a particular cancer a cancer on a genetic level. Repeatedly, we were reminded that the focus must be on patterns of genetic mutations, rather than on single target genes to be reached by single "magic bullets." In a plenary address, Dr. Robert Nicholson, of Cardiff, Wales, called it "naive" to consider elements within breast cancers separately. As UCLA researcher Dr. Mark Pegram put it: "We can't imagine that we'll have an impact with modifying a single system."

What kinds of systems or patterns are researchers looking at? In one of the satellite symposia, Dr. George Sledge, a researcher from the University of Indiana, offered a conceptual framework I found useful in understanding an extremely complex process. Citing The Hallmarks of Cancer, a pivotal essay by Weinberg and Hanahan (Cell, 2000), Dr. Sledge said that all cancers develop the same capabilities, though each will do so in different ways. "What makes a cancer a cancer," he pointed out, "is also what makes it vulnerable from a therapeutic standpoint. These hallmarks are: unlimited growth; functional immortality-that is to say the ability of the cancer to evade senescence (growing old) and apoptosis (programmed cell death, a normal cell function); the ability of the cancer to invade and to metastasize; the ability both in primary tissues and at distant sites to form new blood vessels (angiogenesis); and finally, genomic elasticity-the basis for drug resistance in most cases." (Parentheses mine) New drugs in development target one or more of these hallmark functions.

In microarray analysis, DNA from tumor tissue is placed on a specially prepared gene chip, making it possible to view thousands of genes at once on a glass microscope slide. Using such chips, scientists can tell which genes are expressed, or turned on, in the sample of tissue. This creates a "fingerprint" of an individual tumor, detailing which of the different genes thought to be significant in breast cancer are active or quiescent. From these genetic fingerprints, scientists have begun to detect patterns of gene expression that are associated with prognosis or outcome, as well as predictive factors that indicate sensitivity to specific treatments. Analysis can also be done of the serum (a component of blood), to look at the proteins that are produced-or "expressed"-by genes. This analysis is known as proteomics. These proteins serve as the "messengers" that govern signaling within each cancer cell and determine how it functions.

A few notable examples from the Symposium of how this technology is developing: Scientists at the Eastern Virginia Medical School used a technique called protein chip mass spectrometry to identify unique serum profiling that has the potential of one day yielding a simple blood test for cancer detection. [23] Researchers at the Walter Reed Army Medical Center used high-throughput analysis of breast tissue to identify abnormalities in DNA which might help in determining onset, progression and severity of the disease. [24] NIH scientists are creating a comprehensive database that can help to validate, prioritize and interpret tumor profiling data. [182] Researchers from Sweden's Karolinska Institute screened 300 patients for over 12 thousand genes to obtain patterns of gene expression that could predict not only how patients will do, but also resistance to CMF (Cytoxan, methotrexate, fluorouracil or 5-FU) chemotherapy. [1]

Gene Expression Profiling, Proteomics and the Future

Why is all of this so important, and what are the implications? Some researchers believe that gene expression profiling and proteomics have the potential (but it is still only that, at this early moment in the research) to revolutionize primary breast cancer treatment as we know it, in at least four interconnected ways.

• Prognostic markers. Right now, since we don't know which cancers are most likely to recur and which will almost never recur, most women with invasive breast cancer receive adjuvant (after surgery) chemotherapy and/or hormonal therapy. In the future, using these technologies, we may be able to identify high-risk patients who will need aggressive adjuvant treatment much more reliably than ever before. Just as important, we may be able to identify those who do not need adjuvant treatment. Thus, the majority of patients with favorable profiles indicating with a high degree of certainty that their tumors are extremely unlikely ever to recur will not need any treatment beyond surgery/radiation. The word "cure" may come to have some real meaning for these women, a level of reassurance not currently available. The savings in resources, dollars and human misery would be considerable.

• Predictive markers. Early signs indicate that we may be able one day to predict which treatments are most likely to work on which patients by analyzing gene expression from past patient samples where we already know both the treatments and outcomes. This means we would be able to spare patients toxicity, and tailor treatments individually, which is likely to greatly increase the effectiveness of the treatments we already have. If Herceptin had been used on unselected patients with breast cancer, according to Genentech scientists, it would have been impossible to show any benefit, and the drug would not have been approved. But with the HER2 assays-only recently sorted out-the benefit not only became clear, but was shown to be related to the degree of HER2 over-expression. Imagine one test of tissue (gene expression) or serum (proteomics) that would serve as both a wide-spectrum prognostic indicator of disease outcome and a wide-spectrum predictive indicator of treatment response.

• "Everything old is new again," to quote from researcher George Sledge, of the University of Indiana. Many older chemotherapy drugs, some now abandoned, have shown activity in a few patients, but we had no way of identifying which patients these were, and the toxicities were such that it didn't make sense to keep on using them broadly, with all patients. Even without the development of a single new targeted drug, this potential application of this new technology would offer "new" treatments. An early example of this potential is the discovery of synergism between Herceptin, Taxol and an older chemotherapy drug Paraplatin (carboplatin), leading to an effective combination treatment in advanced breast cancer, reported at SABCS. [439]

• New targets, new combinations. Multiple assays may make possible the development of novel targeted drugs and drug combinations to benefit subgroups of patients with specific patterns of gene expression. This has the potential of paving the way for an understanding of how to effectively combine the new targeted therapies being developed currently, which are unlikely to be effective as single agents, as we've learned already from the trials of Avastin and Iressa. Imagine individually designed "cocktails" of relatively low-toxicity targeted treatments that work in harmony to encourage tumor tissue to behave like normal tissue. Of course, our hope is that metastatic breast cancer, now a progressive, incurable disease in all but a very small minority of patients, could one day become a chronic and manageable disease, enabling normal life-spans for most patients with advanced disease.

Caveats and Concerns

For every potential benefit, however, there are caveats and areas of significant concern, as these accelerating technologies outstrip our capacity to digest and respond to their implications. Each of these represents important areas for breast cancer advocates.

• Reliability and reproducibility of testing. These are early days for gene profiling and proteomics, and there have already been problems reported with reproducibility between labs. As yet, we have no idea how much of an impediment these and other technical problems may become.

• Pitfalls and challenges in data analysis. A recent article in the Journal of the National Cancer Institute (Simon, et al., Vol.95, No.1, 2003) expressed concerns over the statistical rigor with which these complex data sets are being analyzed. The vast quantity of data generated by gene and protein analysis necessitates extremely sophisticated analytic procedures. Rigorous and appropriate safeguards must be in place, and scientists and statisticians must work in cooperation so that meaningful findings are neither over- or under-represented.

• Ethical and privacy concerns over genetic and proteomic testing. As scientists learn to extract more and more sensitive information from tissue and blood samples, great care should be taken with patients' privacy and confidentiality, now and in the future. At the same time, necessary protections should not overly impede forward scientific progress in the field. Maintaining a dynamic balance between individual needs and the greater good is a matter for thoughtful deliberation among all stakeholders, and potential legislative action as issues are clarified.

• Industry and regulatory barriers. Existing industry and regulatory standards and procedures have been developed in response to older treatment methods, in which standards and expectations were very different. The recent experiences with Iressa and Bexxar (a drug for non-Hodgkin's lymphoma) at the FDA's Oncologic Drugs Advisory Committee represent only the beginnings of an increasing confusion, as more new biological therapies enter the market.

• Access to new technologies. There is concern that we will see effective new tests and treatments priced way beyond the capacity of most patients and/or their insurers to afford them. If the cost of Herceptin treatment and other new drugs is any indication, the use of multiple targeted treatments will be prohibitive until and unless there are changes in the current environment of pharmaceutical research and manufacturing. Still, the potential is certainly there in these technologies for ultimate saving of money and resources, be it through unifying, centralizing and computerizing the testing process, sparing most primary breast cancer patients expensive adjuvant treatment, or applying more specific, individualized drug selection in high-risk and advanced cancers. Industry forces are not likely to support efforts to decrease demands for their products, however, and cooperation, rather than competition, would be required between companies, unlikely in the current anti-regulatory atmosphere.

• Unknowns of genetic manipulation. Our genes perform complex functions, and so it's unlikely that a "cookbook" approach to targeting multiple over- and under-expressing genes will have only the desired effect, without impacting other body processes. No one could have predicted that Herceptin would be associated with heart failure in some patients, for example. There are very likely to be unanticipated negative consequences to treatments based on multiple targeting, unknown to us at this point.

• Loss of research perspective. New ideas are seductive, for researchers and the public alike, so it's important to remember there are still many other approaches worthy of continued funding and interest. In our pursuit of these exciting developments, let's not forget that gene expression profiling is a "big science" idea that may take years to bear fruit. In the meanwhile, we will still need practical solutions for the immediate needs of breast cancer patients.

  Despite all this, the strides already made through understanding only two genetic markers important in breast cancer-ER and HER2-offer a foretaste of what is to come. This is the first concept in years that makes me truly hopeful. Make no mistake, when and if this technology matures-and we figure out ways to use it ethically, safely and equitably-we will see a real paradigm shift away from the unnecessarily toxic one-size-fits-all treatments administered to most breast cancer patients today.

Avastin and Iressa: The "Failure" of Targeted Therapy -- So Far

One of the disappointments at SABCS this year was the overall failure of such targeted therapies as Iressa and Tarceva to show much in the way of treatment benefit, when given to unselected patients, either alone or with chemotherapy drugs. As a result of these and similar results, instead of looking at single genetic factors, scientists are focusing on the "crosstalk" between different genes.

Avastin is Genentech's anti-VEGF drug that recently completed an interim analysis for a Phase III trial in which Xeloda alone was compared with Avasin plus Xeloda, for late stage metastatic breast cancer. VEGF stands for vascular endothelial growth factor, a protein produced by one of the important genes that regulate angiogenesis, a crucial process by which tumors recruit and maintain the blood supply they need in order to grow and thrive. In normal tissue VEGF is thought to be important in childhood development and in wound healing, but in cancers that form solid tumors, angiogenesis is what enables tumors to get larger than the size of a small pea.

A large Phase III trial randomized 462 patients with metastatic breast cancer who had already received prior chemo (anthracycline and taxane) to get Xeloda alone, or Xeloda plus Avastin (bevacizumab). The results were disappointing. Adding Avastin to Xeloda in heavily pretreated late stage metastatic breast cancer patients had little effect. There were some treatment responses, but they didn't translate into any clinical benefit in terms of increasing time to disease progression or survival. Trials are being done in first-line metastatic breast cancer with the hope that this drug could be more effective with earlier stages of the disease. [36]

Iressa is AstraZeneca's tyrosine kinase inhibitor that targets the EGFR or epidermal growth factor receptor or HER1 gene. Recently recommended by the Oncologic Drugs Advisory Committee for FDA approval in lung cancer, the drug demonstrated decidedly mixed results in several trials. The EGFR gene is thought to work in concert with other genes like HER2 to regulate cell growth. The not-yet-approved drug Tarceva, from Genentech, is also an EGFR-inhibitor.

In her presentation of the results of a 63 patient multicenter Phase II trial of Iressa in far-advanced metastatic breast cancer patients, lead investigator Kathy Albain presented findings that showed a single partial response, and several stable disease responses, together adding up to about a 14% response rate. There was significant reduction in bone pain in 42% of patients, however, but this was not an anticipated endpoint of the trial, and was presented only as an indicator of further potential and need for study. Iressa is an oral drug, but is not without side effects, principally a rash and GI symptoms.[20]

Some thoughts about the "failure" of these targeted therapies. Researchers are scrambling to figure out how to use these biological drugs, and the many others that are in the pipeline. Clearly they don't do as well in late-stage metastatic disease when given alone, or in combinations with chemotherapy where there is no clear synergistic rationale for their use. A big part of the problem may well be that they work only in a subset of patients-as Herceptin does-so in order to see a treatment benefit, you would have to be able to select out those patients whose tumors overexpress these factors. This means developing assays, or tests, on tumor tissue, which was a terrific problem for Herceptin that is only now really being fully resolved. If Herceptin had been given to all patients, clinical trials would not have shown a treatment benefit, Genentech scientists maintain.

As we heard at a breakfast meeting held for advocates by Genentech, maker of Avastin, scientists there are scrambling to analyze tumor samples from their Avastin trials, so that they can determine who is most likely to benefit and reanalyze the data. Why wasn't this done to begin with? No company wants to have to invest in the very expensive and lengthy process of developing these assays until and unless a drug is proven effective. Yet, it's seeming more and more that effectiveness with this type of drug can't be demonstrated without identifying the patients who are most likely to respond. Sometimes, there may simply be no direct correlation with gene expression and treatment response if you look at a single drug and a single gene. It's very possible that microarray or proteomics technology, with its promise of screening multiple genes or proteins will provide better answers in a few years. Until then, as Dr. Harold Burstein says, "Absent a better biological marker of tumor dependency on VEGF, it may be difficult to figure out which patients with advanced breast cancer ought to be offered anti-VEGF therapies."

A Biomarker for Chemotherapy Benefit?

Determining which patients are most likely to have recurrences is a crucial goal in breast cancer research, as we've said,-and yet the area of biomarker research has been fraught with contradictory evidence.

Now we may have a new single marker to look at prognosis, in the work of Nadia Harbeck of Munich, Germany. The plasminogen activator (PA) system is thought to play a key role in metastasis. uPA and PAI-1 are validated biomarkers that may have relevance for clinical decision-making. Patients with tumors that are low in expression of these proteins do not need adjuvant treatment, and are likely to be cured by surgery/radiation alone. In her studies, those with high expression of these factors had worse outcomes without chemotherapy. However, in these patients the benefit from chemo is so enhanced, that it equalizes outcome. The FDA is currently considering a test of these markers, manufactured by American Diagnostics. So far at least, this assay has met the highest level of evidence, with no contradictory data. [7, 127]

Understanding Hormonal Resistance

Only of the earliest effective targeted therapies for breast cancer was tamoxifen, which binds to the estrogen receptor (ER) on the surface of cancer cells and blocks estrogenic effects on cell growth. Yet for many patients whose tumors show positive estrogen and/or progesterone receptors (PR or PgR), this and other hormonal treatments either fail to work in the first place, or stop working after a period of time.

Why do 40% of patients with ER+ breast cancer fail to derive benefit from hormonal treatments, and why do patients with an initial response often relapse? Why does tamoxifen sometimes appear to actually promote tumor growth after patients take it for extended periods of time?

A plenary lecture by Dr. Robert Nicholson, of Cardiff, Wales, examined the complex world of growth factors, genes that promote cell growth, as an important cause of why breast cancers develop resistance to hormonal therapy like tamoxifen and the aromatase inhibitors.

An emerging body of knowledge implicates over-expressed growth factors like HER2 and EGFR (as well as HER3 and HER4, MAP, AKT and other tyrosine kinases) in resistance to drugs like tamoxifen and the aromatase inhibitors. You may have heard that women with HER2+ cancers don't respond well to tamoxifen-this is thought to be part of the reason why.

Nicholson said that the signaling of the estrogen receptor (ER) gene interacts with these growth factors in a process often referred to as "crosstalk." He believes that the activation of growth factors is the dominant mechanism by which hormonal treatments fail. Researchers are looking into whether the EGFR inhibitor Iressa may impact HER2+ tumors by working along with Herceptin (which inhibits the HER2 growth factor) to decrease hormone resistance in the one quarter of HER2+ patients who also have ER+ tumors.

A study from Baylor College of Medicine scientists confirmed the presence of this "crosstalk" between hormone and growth factor receptors, pointing the way to possible meaningful drug combinations through a clearer understanding of how these factors interact. [246] Trials investigating these strategies may point the way to more effective use of EGFR inhibitors like Iressa and Tarceva, which has so far been problematic.

If researchers could determine why breast cancers develop resistance to hormonal treatments, and devise therapies to break down this resistance, many, many more lives would be saved in women with ER+ and/or PR+ tumors.

Dose Dense vs. Standard Combination Chemotherapy

This year's biggest media story at the San Antonio conference was about a change in the way that adjuvant chemotherapy should be administered, referred to as "dose dense" treatment. A large study (CALGB and Intergroup 9741) was presented by Memorial Sloan-Kettering Cancer Center researcher Dr. Mark Citron, lead investigator of a clinical trial in which 2,005 women with node-positive primary breast cancer participated. [15]

This trial was designed to examine two basic questions in adjuvant chemotherapy using Adriamycin (A), Taxol (T) and Cytoxan (C). First, would it be better to administer the drugs one at a time (sequentially) rather than the standard way together (concurrently)? Second, would it be better to give the drugs every two weeks rather than the standard every three weeks? The three week cycle was derived because it generally takes that period of time for the bone marrow to recover and counts to come up again, so that that next dose can safely be delivered.

Patients were randomly assigned to four study arms in this trial, all of which got the same amount of all the chemo drugs, but on different schedules. The two sequential arms got four treatments of A, followed by four treatments of T, followed by four treatments of C, either in the standard three-weekly or the dose dense two-weekly intervals. The other two groups got four treatments of A and C together, followed by four treatments of T in either three-week or two-week (dose dense) intervals. Patients who received the two-week administration had to be given Neopogen (filgrastim, or G-CSF) and because of this suffered less neutropenia. Otherwise, side effects were similar in all arms of this trial.

The results were surprising. The simple change of administering the chemo more often had dramatic positive effects.

Every-two-weekly dose-dense regimens reduced the rate of recurrence by 26% and the death rate by 31% compared with the every-three-weekly regimens. Absolute disease-free survival of patients on the dose-dense regimens was 82% by contrast with 75% with the three-weekly regimens. After four years, 92% of patients on the dose dense therapy were still alive, compared with 90% on the conventional therapy. There was no difference in disease-free survival or overall survival with the sequential vs. combination regimens.

In addition, the dose-dense group finished their treatment in two-thirds the time. The combination AC then T every two-weeks arm took only 16 weeks to complete, while the sequential every three-weeks arm took 36 weeks.

True, it's only one study (through a large, very well-designed one) and the results are early because the time of follow-up is relatively short for an adjuvant study, but some oncologists were talking in San Antonio about how this was going to change their practice immediately. Of course, many others will wait for other studies to confirm these findings, and to see if the results hold up over time. One cautionary note, though: unless drug costs are reduced, the added expense of Neupogen (or Neulasta, its once-a-cycle much more expensive form) may equal or exceed the cost of the chemo drugs, which could pose a major obstacle in access.

This study represents the second significant example in 2002 where a simple change in administration of adjuvant treatment improved disease-free survival and lowered the death rate. The first was presented at the American Society of Clinical Oncology (ASCO) meeting last May, when Dr. Kathy Albain presented conclusive evidence that patients did better when tamoxifen was administered following chemotherapy, instead of at the same time. It should be noted that these improvements in treatment, that are saving lives today, do not depend upon decades of development of new drugs and technologies.

ATAC Update

Last year in San Antonio, the preliminary results of the ATAC trial were presented, in which the drug Arimidex was compared with Tamoxifen for the adjuvant treatment of early breast cancer. Arimidex is an aromatase inhibitor, which decreases circulating estrogen in postmenopausal women by inhibiting the enzyme aromatase which converts androgens into estrogens, the principal source when the ovaries have shut down. The ovaries are the main source prior to menopause, which is why aromatase inhibitors can only be given to premenopausal women if they have ovarian ablation, either surgically through oopherectomy, or chemically, through a drug called Zoladex (goserelin). As mentioned earlier, tamoxifen, a selective estrogen receptor modulator (SERM), works in an entirely different way, by selectively competing with estrogen for the estrogen receptors on the cancer cell.

ATAC stands for Arimidex, Tamoxifen Alone or in Combination. ATAC is a multi-center, randomized, double-blind study involving 9,366 primary breast cancer patients from 380 cancer centers in 21 countries. ATAC was initiated to answer the question of whether Arimidex (1 mg daily) was more effective than tamoxifen (20 mg daily) in breast cancer patients after they had completed their surgery (and chemotherapy, if warranted). A third arm of this trial was designed to look at whether a combination of both drugs was better than either drug alone, but has been discontinued because the outcomes were the same as the tamoxifen arm, but with more toxicity. Patients were to follow their treatment regimens for five years. The endpoints (measured outcomes) of the trial are disease-free survival and safety, as well as distant recurrence and survival. They are also looking at quality of life and at a variety of side effects.

Last year, after 33 months of follow-up, the risk for recurrence was 22% lower in the Arimidex only group who were ER+ or PR+ (75% of the patients). It's important to remember here that tamoxifen already lowers risk of recurrence by about half, so these were impressive results. However they represented an absolute difference of only a couple of percentage points in rates of recurrence, and no difference (yet) in mortality. More impressive still was the reduction in risk of recurrence in contralateral breast cancer (a new primary cancer in the other breast), which was 60% lower than tamoxifen, though the absolute numbers are very small.

By comparison with tamoxifen, Arimidex-only patients have a lowered risk of weight gain, vaginal bleeding, endometrial cancer, hot flashes, strokes and clotting disorders. However, there was an increase over tamoxifen in musculoskeletal problems caused by Arimidex, particularly arthralgias (muscular aches and pains, sometimes debilitating) and fractures. Since Arimidex works by removing circulating estrogen, there are real concerns about the long-term effects of estrogen depletion on the body, in bone, of course-but also on cognitive processes.

The biggest issue raised about giving patients Arimidex rather than tamoxifen is the complete lack of long term data about efficacy, long-term side effects, and the optimal duration of treatment. As for switching midstream from tamoxifen to Arimidex, since the third or combination arm of this trial actually did slightly less well and was shut down, there is some concern that Arimidex may actually promote an agonist (stimulatory) effect in tamoxifen. Right now there isn't much evidence to believe to support giving these drugs sequentially, and some concerns.. An ASCO panel was convened last May, headed by Dr. Eric Winer of Dana-Farber, that recommended waiting for longer-term results on efficacy and safety before changing the standard of care. The interim ATAC results were published in British journal, The Lancet (Lancet 2002; 359: 2131-39)

This year at the SABCS, Dr. Amon Buzdar, of MD Anderson, presented an update of the ATAC trial, after a median follow-up of 47 months, on behalf of the ATAC trialists group. Basically, last year's findings have held up. Disease-free survival in the hormone-receptor positive group was 89% for the Arimidex group, and 86.1% for the tamoxifen group, an absolute benefit of 2.9%. The decrease in contralateral cancer is somewhat less now, but still in favor of Arimidex. The recurrences reported include both local and distant, however, and these have not yet been analyzed for statistical significance, but there were 222 distant recurrences in 3116 patients in the tamoxifen arm, 195 distant recurrences in 3125 patients in the Arimidex arm.

No difference in efficacy has yet been demonstrated between Arimidex and tamoxifen for women who also have chemotherapy. According to data presented later by Per Lønning, from Norway, on response and resistance to aromatase inhibitors, there's a theoretical suspicion that Arimidex may work better in women who have a lower BMI (body mass index), while tamoxifen may work better for women with higher BMI. [13]

A talk by Matthew Ellis, from Duke, on the clinical use of aromatase inhibitors, dealt at some length with the comparative side effects of the two drugs. Ellis affirmed that Arimidex users appear to have a higher risk of osteoporosis/fracture, neurocognitive problems (with memory, attention, word-finding, etc.), muskulo-skeletal syndrome (aches and pains in soft tissues and joints), hyperlipidemia (elevated cholesterol), and possibly heart disease. By contrast, Ellis reported that tamoxifen users have an increased risk of deep vein thrombosis, stroke and endometrial cancer. It should be noted that the serious risks to both these drugs are quite low, and the risk-benefit ratio for tamoxifen as breast cancer treatment was estimated at last year's SABCS to be 30 to 1, meaning that tamoxifen prevents 30 deaths for every one that it causes. The figures for Arimidex are almost certainly even more favorable.

Finally, Ellis and others cautioned about a potential interaction between HER2+ tumors and tamoxifen, in which the hormonal drug may actually have a stimulatory effect in the present of the HER2 gene, which governs growth, as well as in overexpressed EGFR, or HER1, another growth factor.

HER2 News

More HER2 testing confusion. Testing for HER2 status continues to pose problems. HER2 is a gene that promotes cell growth, thought to be in part responsible for more aggressive, rapidly growing tumors in around 25-30% of breast cancer patients. Tests are done at diagnosis, on stored tumor samples, or more recently on serum (blood). For several years data has been accumulating indicating that FISH (fluourescence in-situ hybridization) testing, which looks at gene amplification, is more accurate than IHC (immunohistochemistry, eg. DAKO HerCeptest), which looks at protein overexpression, in predicting which patients will respond to the drug Herceptin.

Local laboratories may fail to identify as much as 20% of patients found to be HER2+ by central labs, according to two studies, one by and international team of Herceptin researchers, the other by the Laboratory Corporation of America and Genentech [238, 235]. IHC testing is less reliable and reproducible than FISH testing, which is more consistant across laboratory settings. IMPATH researchers found that 21% of patients testing +2 on IHC tested positive on FISH, while less than 5% of IHC +1 or 0 tested positive on FISH. Less than 10% of IHC 3+ cases are negative for FISH. [236]

The take home message? Make sure your test is done by a lab that processes at least a hundred breast cancer patients each year, and if your score is 0, +1 or +2 on IHC testing, the results should probably be confirmed by FISH.

Effective Herceptin-Chemo Combinations. The results of the first randomized trial to look at the results of adding a platinum-based chemotherapy to Herceptin were presented at SABCS. This US Oncology Phase III study of 194 HER2+ patients confirmed that adding Carboplatin onto the already active Herceptin+Taxol combination improved time to disease progression from 6.9 months to 11.2 months, and more patients responded to treatment (52% vs. 36%). Median survival, 33.5 months for the Herceptin+Taxol arm, had not been reached yet by the arm that added on Carboplatin. In HER2 +3 patients only, the results were more pronounced. However, there was more toxicity to the three drug regimen. Most patients had neutropenia, and some had leukopenia and thrombocytopenia, but researchers described this as "transient abnormalities that did not affect the patients' risk of infection, fever or bleeding." [35] A smaller study from the Minnie Pearl Cancer Research Network also found a benefit from this three drug combination, with a response rate of 89% for selected patients. [439]

Other studies showed that Herceptin was active when combined with several other chemotherapy drugs, among them Navelbine (vinorelbine), Gemzar (gemcitabine) and Taxotere (docetaxel) [431-7] Many women with HER2+ metastatic breast cancer stay on Herceptin as they move from one chemo to another. Reassuring news from a British Columbia Cancer Agency study confirmed the feasibility of second -line Herceptin-chemo combinations, with continued benefit and no undue toxicities. [440]

Zometa vs. Aredia

Bone is a common metastatic site in breast cancer, and until the wide use of bisphosphonates in the last five years, fractures and other complications were common. A large multicenter international study, presented by Dr. Robert Coleman of the UK, compared the two most commonly used bisphoshonates in metastatic breast cancer spread to the bone. 1130 patients with advanced breast cancer and at least one bone metastasis, were randomized to receive either Zometa (zoledronic acid) or Aredia (pamidronate).

After two years, researchers found that Zometa had been more effective than Aredia in reducing fractures in patients with bone metastases. Bone pain was significantly decreased as well, reducing the need for palliative radiation. Zometa's other advantages over Aredia, demonstrated prior to this study, are shorter infusion time, as well as greater effectiveness in controlling hypercalcemia, a late complication of bone metastases. [355]

Neoadjuvant Chemo Combo

This small 27 patient multicenter trial in high-risk stage IIA-IIIB patients caught my eye because of the dramatic clinical and pathological responses to this drug combination. These preliminary numbers are small because this is an ongoing trial that is still accruing patients. In this regimen, neoadjuvant Taxotere and Navelbine were followed by surgery, and then by standard AC chemo. 97% of patients had either partial or complete tumor response to this regimen, 50% had their tumors disappear on clinical examination, and 36% had a complete pathological response, meaning that when they did have their surgery after chemo, there was no evidence of tumor. There was fever and neutropenia, but no serious infections. The researchers conclude that "Taxotere and Navelbine administered in a dose intense and dose dense fashion in the neoadjuvant setting is among the most active regimes ever tested." [162]

To Find Out More

The San Antonio Breast Cancer Symposium has a virtual website, where slides and audio of some of the plenary presentations are available. www.sabcs.org.

A searchable database of will allow you to read the SABCS abstracts, listed here in [brackets]. www.sabcs.org/AbstractOnline/index.asp.

Two good medically-oriented sources for information about the SABCS can be found at wwww.medscape.com. and wwww.docguide.com.

About Musa Mayer

Musa Mayer is a 13 year survivor, advocate and author of the forthcoming book, After Breast Cancer: Answers to the Questions You're Afraid to Ask (O'Reilly, April 2003), as well as Advanced Breast Cancer: A Guide to Living with Metastatic Disease (O'Reilly, 1998) and Examining Myself: One Woman's Story of Breast Cancer Treatment and Recovery (Faber, 1994).