This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Mayer, M. Search for Related Content PubMed PubMed Citation Articles by Mayer, M. Social Bookmarking                            What's this?

Swimming Upstream: An Advocate Reflects on Cancer Research and Social Realities

Musa Mayer

Corresponding author: Musa Mayer, 250 West 82nd St, #42, New York, NY 10024; e-mail: musa{at}echonyc.com

In the years since 1989, when I was diagnosed with breast cancer, the advocacy movement has matured, and many of us have matured within it. It is a little embarrassing now to be confronted with my own naïveté, rereading what I wrote about breast cancer research and drug development a decade or so ago. Not only did I oversimplify the research process, underestimating its daunting challenges, but I was largely unaware of the distorting influence financial incentives can have on research. In the name of hope, I was far too willing to abandon critical thinking and jump onto the latest bandwagon. Like so many, my judgment was easily hijacked by upbeat media sound bites and by what I wanted to believe was true about breast cancer research, rather than the realities, which represent a far more complex and nuanced story.

Years of working with the US Food and Drug Administration (FDA) as a Patient Representative and in the Cancer Drug Patient Consultant Program¹ following the uneasy interaction of regulatory and industry forces have offered some important lessons, as well as some sobering cautionary tales. As a member of the Institute of Medicine's Forum on Drug Discovery, Development, and Translation, I have seen how decisions are made at earlier stages of drug development, and how critical they are to later investments and drug costs. This perspective has served to make me more aware of the ever-enlarging disconnect that lies between what science is making possible, and the current realities in our society.

It is not that seasoned advocates are not excited by the science we learn about at conferences and in the journals. We are, but we wonder how that science can possibly take us where we need to go, given the disparities, inefficiencies, and excesses of our current health care system. We are constantly learning of technical and scientific breakthroughs, inconceivable only a decade ago, the implications of which carry enormous promise. This is the dawn of a new era of personalized medicine, we are told. The end of empirical one-size-fits-all cancer treatment lies just around the next bend. Now that patterns of gene expression have divided breast cancer into several distinct diseases, patients will soon receive only those treatments that will benefit them.

But then, we also read worrisome reports about the inaccuracy of tests for estrogen receptors and HER2 overexpression that have been in use for many years as a way to personalize breast cancer treatment. We hear of significant discrepancies between local and centralized laboratories, and that uniform standards are still lacking for such tests, despite the fact that life-and-death decisions are based on them. And we worry about funding. How can we look forward to a future where personalized medicine becomes standard of care, when only a small fraction of the research dollars are being spent studying and validating biomarkers?

In their latest report, The Pharmaceutical Research and Manufacturers of America, the pharmaceutical industry's trade organization, touts a pipeline of nearly 650 cancer therapies, 79 in breast cancer alone.² Yet, the number of novel drugs approved has been dwindling for years.³ Has the low-hanging fruit of molecular biology already been plucked and brought to market? In meetings, I hear repeatedly that the cost of satisfying the regulatory requirements in drug development has become prohibitive, and that failure rates are still high, even late in development. These costs are passed along to us, the consumers.

Some metastatic breast cancer patients receiving the first of these new treatments are unable to afford even the co-payments. Patient assistance programs are in place and are committed to seeing that no one goes untreated, but not everyone is able to gain access to them. I read frequent news stories about toxicity and safety issues with new drugs on the market, usually accompanied by references to an FDA that is not doing its job of monitoring, and that should be calling for long-term postmarketing studies—although it lacks both the funding and the legal authority to do so. The FDA is also faulted for creating bureaucratic hurdles, and keeping desperately ill patients from access to emerging treatments that might save or extend their lives. Others complain bitterly that the current system of Big Pharma funding the FDA to review their own products is unethical and dangerous. For some, treatments are approved too soon, before we know if they work, and certainly before we know if they are safe. For others, treatments are approved too late, when thousands of patients who might have benefited have died. Some applaud important research breakthroughs, while others lament the incremental, glacial pace of research. Some celebrate research heroes for their innovation and passion, while others see conflicts of interest and narrow careerism wherever they look. Some point with pride to the major research institutions as beacons of hope, offering the best cutting-edge treatments emerging science can provide. Others point with shame to a broken system where widespread failure to meet even the most basic treatment needs of minority and low-income women has resulted in high mortality rates and an ever-increasing disparity in outcomes as death rates from breast cancer continue to decrease for white, affluent, insured women, whose insurance will cover the newest and latest treatments.

But all is not confusion and contradiction. If there is one overarching theme that is emblematic of what I have learned since my diagnosis, it is the importance of gathering the highest quality evidence of safety and efficacy before full approval of a new drug or broad implementation of a new approach in breast cancer. Whenever we have failed to do this, we have paid a price. Most cancer advocates I have spoken with seem to understand that more patients in the long run will benefit from this painstaking approach, although it is difficult when you work directly, as I do, with people who have metastatic cancer.

To some advocates,⁴ waiting years for trials to be completed seems bureaucratic, wrong-headed, and cruel. They demand that seriously ill patients who have run out of other treatment options should have the right to buy new drugs after phase I trials (when dose-finding and safety studies are typically done in small numbers of patients). Existing provisions for compassionate drug use help too few, they insist, and offer no incentives for companies to make their drugs available before approval. These advocates want experimental drugs on the market, where patients can buy them without FDA say-so, and where companies can sell them, free of concerns over liability. Never mind that these early-stage treatments are far more likely to hurt than to help. Never mind that selling unproven drugs would undermine clinical trial enrollment and incentives for clinical research, or that lowered levels of evidence would open the door to questionable products. Because their message is so simple and emotionally compelling, this viewpoint has gained substantial traction.

Though most advocates can see the flaws in this approach, that does not mean we're happy with the status quo. With all of this new molecular science, surely technology should be able to offer us some other more streamlined method to determine if treatments work. One day this may be so, but today we rely on cell lines, tissue cultures, and animal models—all poor indicators of what will work in people. Despite the enormous effort and cost that goes into selecting the best drugs from among the tens of thousands of candidates, nearly 90% of drugs that enter phase I trials will never reach the market, due to toxicity, lack of efficacy, or both.⁵ Lacking a better alternative, we continue to rely on randomized controlled trials in large groups of patients, which can take years to enroll and years more to measure the outcomes, or end points, in themselves another bottleneck.

Our lack of short-term outcome measures contributes to delays in drugs reaching the marketplace. In a disease such as HIV/AIDS, the viral load can be measured rapidly to determine whether a given drug is effective. Conversely, in breast and other cancers, we have to wait to see if the disease recurs or progresses, or if patients actually live longer. Short-term tumor response is not a reliable predictor of long-term outcome, as we found out to our regret with high-dose chemotherapy with stem-cell or bone marrow transplantation. A validated biomarker that performs as a reliable surrogate for clinically meaningful disease outcomes would revolutionize breast cancer research.

Would not enrolling more patients than the 3% to 5% who currently participate in clinical trials accelerate progress? Certainly, but outside of major research centers, incentives for physicians to refer their patients to trials are largely absent. Sluggish enrollment not only adds to the time it takes for a new treatment to reach the market; it also slows the process through which the clinical use of a drug is refined once it becomes available. Despite the best efforts of the increasingly underfunded National Cancer Institute cooperative group program, our current system is not set up to encourage trial participation, neither from the perspective of overscheduled physicians nor their mistrustful patients.

I am hopeful about one possible scenario that might hasten drugs to market, reduce the huge cost of drug development, yet still uphold levels of evidence. Most of the new biologic therapies work only in small percentages of patients, sometimes fewer than 10%, yet they are typically tested in unselected populations. To detect an effect in such a small number of patients, a phase III trial must be enormous. Why, in most cases, has no biomarker test to select the responders been codeveloped along with the drug?

The answer may lie more with misaligned financial incentives than with unanswered scientific questions. Obviously, companies are looking for blockbuster drugs to earn billions, not niche drugs for small groups of patients. Investing in science that will make their customer pool smaller and sales dwindle will not appeal to shareholders. Though this may seem like a mere technical or procedural issue, its implications have reverberated throughout recent drug development.

The most striking example of a short-sighted untargeted strategy was that of gefitinib (Iressa; AstraZeneca, Wilmington, DE), an epidermal growth factor receptor inhibitor developed for the treatment of non–small-cell lung cancer, after dramatic responses in a few patients became evident in early studies. Gefitinib was given an accelerated approval by the FDA on the basis of a few responses in two small uncontrolled phase II studies, after an emotional hearing in which patients with advanced lung cancer testified about how this drug had given them back their lives.⁶

After gefitinib was approved, two things became clear: 90% of patients had little or no benefit from the drug, but the 10% who did saw dramatic disease regression. If you had lung cancer, there was no way of telling if you would respond or not. Within the year, however, two teams of scientists had isolated the particular gene mutation that identified the super-responders to gefitinib,⁷ raising the question of why AstraZeneca had not undertaken this research during the drug's development. Given that gefitinib had received an accelerated approval based on a small number of responses, another randomized trial was required by the FDA for confirmation. Completed by necessity overseas in countries where the drug was not available, this trial found that gefitinib plus chemotherapy was no better than chemotherapy alone. But where were the super-responders? They were still there, but their results were diluted in the much larger pool of nonresponders. There were simply not enough of them to achieve statistical significance. Aside from patients currently responding to the drug, gefitinib was taken off the market.

The same scenario would likely have occurred with trastuzumab (Herceptin; Genentech, South San Francisco, CA) had that drug been tested in an unselected metastatic breast cancer population, had Dennis Slamon and Genentech not grasped the essential wisdom and necessity of codeveloping a predictive test.⁸ By selecting the minority who could benefit, even that initial imperfect test was good enough to select for responders and thereby greatly decrease the size and cost of randomized controlled trials needed for approval, and the length of time that such a trial would take. Trastuzumab proved even better in primary breast cancer. When the results were presented at the American Society of Clinical Oncology Annual Meeting in 2005, the results of the adjuvant trastuzumab trials were dramatic enough to elicit gasps and cheers from the normally reserved standing-room–only audience of oncologists. Those data showed the largest benefit since adjuvant tamoxifen, our first truly targeted therapy for a selected population.^9,10

If drug developers learn from the cautionary tale of gefitinib, and from the success of trastuzumab, and if they worry less about marketing to all patients and more about how to make good drugs in conjunction with tests to find the patients who will benefit from them, committing resources to biomarker discovery and embracing enrichment designs for drug trials, we will be a few big steps closer to our goal of personalized medicine.

Other approaches to making cancer research smarter and more efficient are under intense discussion at health policy tables. Managing information in more intelligent ways is essential in the genomic age. So is using electronic medical records to monitor drug safety postmarketing, rather than the inadequate passive system (Adverse Event Reporting System), which the FDA relies on today.¹¹ Informatics and data mining will yield smarter hypotheses to test in clinical trials.

Economic incentive is a powerful engine, producing innovation, but also increasing consumption. Unlike every other industry, health care is not held accountable for outcomes. We pay for products and services whether or not they actually work. Only when we can accurately monitor results will this begin to change. Huge databases such as Medicare, Veterans Affairs, and managed care groups such as Kaiser and others can be used to help inform much of health care and make it accountable. Movement in this direction will depend on overcoming entrenched fears of health care rationing, however.

Investing research dollars wisely while placing patients first, is crucial. In 2006, Mitch Dowsett of the Royal Marsden Cancer Hospital in London and several colleagues from Europe and the United States asked breast cancer clinicians and scientists around the world to prioritize their most pressing research questions, offering them a list of some 70 from which to choose.¹²

By a large margin, the 420 who responded chose as their most important research priority, the identification of molecular signatures to select patients who could be spared chemotherapy. It is not news to advocates that women with breast cancer are overtreated. I was diagnosed in 1989, the year that adjuvant chemotherapy for all invasive breast cancers larger than 1 cm became standard of care. During the decade that followed, the disaster of high-dose chemotherapy made it clear that more is not always better. In fact, much of the real progress in breast cancer treatment seems to have followed the path of "less is more," as treatments have become less disfiguring, less damaging, and less toxic. But we are swimming upstream, against the strong current of an industry that will always be asking us to add something new, something more, to current treatments.

So, it is heartening that the second most important research question was the identification of molecular features to choose the optimal chemotherapy regimen. Since the American Society of Clinical Oncology Annual Meeting in June 2007, we have been aware of unpublished data that the preferential benefit of anthracycline-based chemotherapy over other regimens is confined to those whose tumors coexpress HER2 and TOPOII, a finding with tremendous clinical implications, given the toxicities of doxorubicin.^13,14 A concerted effort to analyze tissue both retrospectively and prospectively will undoubtedly yield many more such patterns of response that will enhance chemotherapy effectiveness and avoid the toxicity of treatments that do not help.

Although we have not been formally surveyed, I believe many advocates share with these breast cancer physicians and scientists a sense of what matters most in research. Perhaps we are already witnessing the beginning of the end of the paradigm of empirical treatment and clinical trials design, and the beginning of a new kind of medicine that spares patients toxicity, and our healthcare system the cost, of treatments people do not need and cannot afford. What will these new models be? And why are we not more excited about them?

In a recent speech, Mark McClellan, the former director of both the FDA and the Center for Medicare and Medicaid Services, put his finger on the current climate: "Our healthcare system is delivering more and more in terms of health. The potential of personalized medicine, and more prevention-oriented medicine, is even greater for the future. Yet, something that should be regarded as good news by everybody in the country is generally regarded with fear and dread, that even if these better treatments come along, they won’t be able to afford them. They won’t be able to take advantage of them. So, instead of enthusiasm, we have a lot of frustration with the public, with employers, with other healthcare payers, and with physicians about where our health care system is, despite the fact that the promise for the future is tremendous."¹⁵

One question we must ask ourselves is whether we as a society have the will to enact what every other developed country in the world has already offered its citizens: universal health care. What good will our innovations ultimately be if they benefit only the wealthy and privileged? To sustain innovation, what we need are fundamental shifts in social priorities. These questions belong around the policy table, in the halls of Congress, and at the voting booth.

Former Surgeon General David Satcher spoke to this question: "The U.S. health system spends far more on the ‘technology’ of care than on achieving equity in its delivery," he observed. "Achieving equity may do more for health than perfecting the technology of care"¹⁶—a sobering thought. According to Satcher's data, five times as many lives could have been saved during the 1990s by equalizing the mortality rates of whites and African Americans, than were saved by all the medical advances made during that decade.

However questionable our collective will for social change may be at this moment in time, there is still reason for hope. The death rates from many cancers are decreasing, if slowly. Scientific understanding is advancing remarkably, with no end in sight to what is possible. Bringing forth those possibilities will be more challenging than I, in my naïveté, ever thought it could be. So much must change, yet change is so difficult. Day to day, I am reassured by quiet, incremental victories, as the women with metastatic breast cancer with whom I work are living longer, and living better, than a decade ago.

Even as we continue to swim upstream, I still cherish a vision of a world where science, medicine, and commerce work in alignment for the benefit of informed patients and public, and where each person has full access to the health care she or he needs—no less and no more.

AUTHOR'S DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

The author indicated no potential conflicts of interest.

NOTES

Musa Mayer is a writer, advocate, and breast cancer survivor. She can be contacted through her website for women with metastatic breast cancer at www.AdvancedBC.org.

Adapted from a plenary address given at the California Breast Cancer Rese-arch Symposium Conference, Los Angeles, CA, September 8, 2007.

Author's disclosures of potential conflicts of interest and author contributions are found at the end of this article.

REFERENCES

1. US Food and Drug Administration: Patient Involvement. http://www.fda.gov/oashi/patrep/pindex.html

2. Pharmaceutical Research and Manufacturers of America: 2006 Survey: Medicines in Development for Cancer. http://www.phrma.org/files/PhRMA%20NM%20Cancer%200901.pdf

3. United States Food and Drug Administration: Innovation/Stagnation: Challenge and Opportunity on the Critical Path to New Products, 2004. http://www.fda.gov/oc/initiatives/criticalpath/whitepaper.html

4. Abigail Alliance: For Better Access to Developmental Drugs. http://abigail-alliance.org

5. Okie S: Access before approval—A right to take experimental drugs. N Engl J Med 355:437-440, 2006[Free Full Text]

6. United States Food and Drug Administration: Transcript of the FDA Oncologic Drugs Advisory Committee meeting, Tuesday, September 24, 2002. http://www.fda.gov/ohrms/dockets/ac/02/transcripts/3894T1.htm

7. Lynch TJ: Activating mutations in the epidermal growth factor receptor underlying responsiveness of non-small-cell lung cancer to gefitinib. N Engl J Med 350:2129-2139, 2004[Abstract/Free Full Text]

8. Press M, Seelig S: Lessons Learned From the Development of a Diagnostic to Predict Response to Herceptin, Targeted Medicine Conference Report, 2004. www.targetedmed.com/flash/lib/pdf/Targeted%20Medicine%20Report.swf

9. Romond EH, Perez EA, Bryant J, et al: Trastuzumab plus adjuvant chemotherapy for operable HER2-positive breast cancer. N Engl J Med 353:1673-1684, 2005[Abstract/Free Full Text]

10. Piccart-Gebhart MJ, Procter M, Leyland-Jones B, et al: Trastuzumab after adjuvant chemotherapy in HER2-positive breast cancer. N Engl J Med 353:1659-1672, 2005[Abstract/Free Full Text]

11. US Food and Drug Administration: Adverse Event Reporting System (AERS). http://www.fda.gov/cder/aers/default.htm

12. The Top Ten Programme: Report on findings and outcomes Prepared for Prof. Mitch Dowsett by Phase IV Communications, 2007, based on Top Ten Priorities in Translational Research. http://www.toptenresearch.org

13. Bazell R: Breast cancer treatment may fail most women. NBC News, June 5, 2007. http://www.msnbc.msn.com/id/19048185/

14. National Breast Cancer Coalition: Press release, NBCCF urges the oncology community to re-assess the use of anthracycline-based chemotherapy in the adjuvant treatment of breast cancer, May 2007. http://www.stopbreastcancer.org/bin/index.asp?strid=944&btnid=1&depid=20

15. McClellan M: Brookings Institution Conference, 2006. The Practice and Potential of Medicine: How to Close the Gap. http://www.brookings.edu/comm/events/20061215.htm

16. Woolf SH, Johnson RE, Fryer GE, et al: The health impact of resolving racial disparities: An analysis of us mortality data. Am J Public Health 94:2078-2081, 2004[Abstract/Free Full Text]

Submitted September 20, 2007; accepted October 22, 2007.

CiteULike    Complore    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

This article has been cited by other articles:

				U. Schuklenk and C. Lowry Terminal illness and access to Phase 1 experimental agents, surgeries and devices: reviewing the ethical arguments Br. Med. Bull., March 1, 2009; 89(1): 7 - 22. [Abstract] [Full Text] [PDF]

This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Mayer, M. Search for Related Content PubMed PubMed Citation Articles by Mayer, M. Social Bookmarking                            What's this?