This Article Full Text (PDF) eLetters: Submit a response to this article Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services E-mail this article link to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints/Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Schein, P. S. Articles by Carter, S. Search for Related Content PubMed PubMed Citation Articles by Schein, P. S. Articles by Carter, S.

The Role of the FDA in the National Cancer Program: Friend or Foe?

^a INCTR (US), Lombardi Cancer Center, Washington, DC, USA; ^b Private Consultant, New York, New York, USA

Correspondence: Philip S. Schein, M.D., INCTR (US), Lombardi Cancer Center, Research Building, Suite E501, 3970 Reservoir Road, NW, Washington, DC 20057, USA. Telephone: 610-527-4605; Fax: 610-527-3810; e-mail: philschein1{at}comcast.net.

Early in 2003, the newly appointed Director of the National Cancer Institute, Dr. Andrew von Eschenbach, defined his goals for the National Cancer Program: "eliminate death and suffering" from cancer by the year 2015 [1]. This was a rather courageous statement and an impressive commitment given the enormity of the challenge and the degree of progress, as measured by changes in patient outcomes, made since the enactment of the National Cancer Act in 1971. In 2003, it is anticipated that approximately 1.3 million new cases of invasive cancer will be diagnosed in the U.S., with over 550,000 deaths. Cancer, as a disease group, accounts for 25% of all deaths in our country and, overall, ranks as the second leading cause of mortality. Moreover, the burden of cancer is a serious international problem, with higher rates of incidence and death projected for the developing world in future years.

We have long since recognized that "cancer" is a generic term for over 100 different diseases, now increased further through molecular subsetting, that may vary in causation, natural history, diagnostic method, and, germane to this discussion, treatment modality. In theory, we need to define specific methods for eradicating each clinical expression of this heterogeneous group of disorders. While progress in prevention and early diagnosis is anticipated, it can be safely assumed that if the goals of Dr. von Eschenbach are to be achieved, significant improvements in therapy, whether they come in the form of adjuvant or advanced disease management, will be required. Central to this issue is the rate at which important new therapies reach the public, for without new tools for cancer treatment, there can be little expectation for success. The basic requirement is that good science, which is certainly plentiful these days, is rapidly and effectively translated into new treatments that are delivered to the public in the form of U.S. Food and Drug Administration (FDA)-approved therapies. There is no substitute. Expanded access to clinical trials, suggested by some as a method for treating the U.S. public in reaction to the current situation, is not a realistic option for the hundreds of thousands of patients who require systemic therapy. The challenges facing the national goal of improved cancer treatment were defined in a more detailed form in a previous manuscript [2]. There have been, however, some important developments since its publication, particularly in regard to FDA policy, that justify a reexamination of some of the critical issues and an assessment as to whether progress is being made.

The issues of the rate and quality of drug approvals for cancer have achieved greater clarity by the publication of an important manuscript in April 2003 from the FDA’s Division of Oncology [3]. The authors of that report provided a summary of the number of cancer drug applications that had been approved over the preceding 13-year period, as well as the end points that had been used for granting marketing authorization. In the article, the authors state that end points other than survival were the basis of approval for 39 of 57 drugs (68%) granted regular approval and for all 14 applications granted accelerated approval during the time frame identified; indeed, tumor response was described as the basis for 12 of the 14 accelerated approvals from 1995–2002 (Table 1). Overall, the general readership might regard the number of approved applications as representing a credible record and one that would counter the general level of criticism of FDA policy expressed in the media, such as the editorial page of the Wall Street Journal.

The authors list 38 NCEs that were approved during the 13-year time frame for a variety of different tumors. Of these 38, four "new drugs" represented older, formerly approved, therapies that were presented in new delivery forms, such as liposomes. One could, however, view these as entirely new agents because of changes in pharmacokinetic profiles and other factors.

Five of the 38 NCEs are not true anticancer agents—drugs that actually attack a cancer—but are better characterized as therapies that can either reduce the symptoms of cancer or better manage the adverse reactions associated with true cancer treatments. These are worthwhile goals but will not in themselves make an impact on outcome statistics. The bottom line is that, during the 13 years of effort and vast expenditures of public and private dollars, an average of three NCEs was approved per year and only 2.5 new agents were approved per year that could be appropriately classified as anticancer agents. For these NCEs, second-line indications (defined as clinical application after a patient has received and failed a prior treatment) predominate in a setting in which success in treatment is known to be limited by tumor resistance and size. Moreover, many of the marketing authorizations were for diseases that are truly orphan indications, such as Kaposi’s sarcoma, hairy cell leukemia, and cutaneous T-cell lymphoma. There were no new approved anticancer therapies for diseases such as head and neck, gastric, and cervical cancers, and there was only one for important tumors such as pancreatic cancer. As newer therapies become more targeted and therefore restricted to specific tumors, or even specific patient subsets within these tumors, the number of approved NCEs will become increasingly important. It is safe to state that the current rate of NCE approval will not allow the goals of the National Cancer Program to be met.

The problem cannot be simply and completely laid at the doorstep of the FDA. The Administration cannot be held accountable for the quality of the science or the level of investment being made to define new treatments for each of the recognized forms of cancer. The latter requires the perspective of the National Cancer Program and the allocation of research and development priorities by the National Cancer Institute (NCI). Specific development projects sponsored by both industry and the NCI have been plagued by problems in clinical study design, and there are instances where development programs lacked sufficient focus to avoid protracted development time frames and associated costs. Companies, in some instances, have not attempted to develop a working relationship with FDA reviewers or heed critical advice when offered by the Administration. The mechanism of Special Protocol Assessment, which serves to allow the parties to enter into agreements, has been underutilized in oncology. As a broad characterization, companies comprising the biotechnology sector, while displaying depth and strength in their programs in basic science and perhaps drug discovery, often lack comparable capabilities or know-how in preclinical development, clinical investigation, and regulatory affairs. A similar profile can be defined for academic medical centers. While we can catalog a long list of impediments to progress, what is required is a recognition that all is not well and that, despite the expenditure of many billions of dollars each year for cancer research by both the public and private sectors, the yield of new and important therapeutics for the country’s second leading cause of death is astonishingly low.

Past generations of anticancer agents, because of the broad, semispecific nature of their mechanisms of action, often proved efficacious in more than one tumor type. These agents (cyclophosphamide, doxorubicin, cytosine arabinoside, etc.), in many instances, serve as the basis for the current practice of oncology and have made notable contributions in the form of long-term, disease-free survival of many forms of cancer when used in combination or combined-modality therapy. Of importance, during the time frames when they achieved marketing authorization (1970s and early 1980s), they were approved based on objective tumor response data, with their optimal applications and contributions to survival determined in postmarketing studies.

In the 1980s, survival became the gold standard for the FDA for approval of drugs, in particular for use as first-line therapy. This has resulted in the need for very large trials with long follow-up times; such studies can become complicated by the heterogeneity of second- and/or third-line therapies that are subsequently employed to deal with the eventual relapse. Additionally, it is likely that, during the years required to complete these pivotal first-line trials, the "standard" or control regimen used for comparison will be replaced by a newer regimen. This has, at times, raised questions regarding the relevance of the results in clinical practice and heavily penalized industry in terms of sunk costs and time. Moreover, many sponsors have abandoned the strategy of developing analogs or new agents to reduce toxicities associated with frequently used cancer therapies because of the large sample size requirements and negative connotations associated with "equivalence" trials. One of the central questions to be addressed is whether the more stringent approval requirements applied in the past decade have resulted in significant improvements in drug selection for marketing authorization, or have they delayed, or even prevented, the entrance of useful agents?

One of the major benefits of the three-decade national investment in the study of cancer has been the emergence of an extraordinary menu of new therapeutic targets. Rational design of new cancer therapeutics, versus the "enlightened empiricism" of the past, is now possible, and with it comes the creation of a new generation of highly specific, molecular-targeted treatments. But with this promise comes an important paradox. While now it may be possible to make cancer treatment more selective, many more drugs will be required and the target patient populations will be increasingly restricted to a subset of patients whose tumors display the required expression of a specific genetic, enzymatic, or receptor abnormality. This is the case with the trastuzumab trials, which have restricted application to the approximately 25% of patients with breast cancer whose tumors overexpress HER-2. This realization, compounded by the recognition that the majority of tumor types qualify for "orphan drug" status based on limited prevalence, has important implications for those who assess the potential market size for a new cancer therapeutic; this is a factor that has not been lost on the industry.

The cost and time for successful drug development have to be considered. As reported by the Tufts Center for the Study of Drug Development, the estimated average cost of the development and approval of a new prescription medication is over $800 million, up from $231 million in 1987, with the largest increases occurring in the cost of clinical development [5]. The time frames for completing development remain long, 10–15 years with current systems and FDA policy. With smaller target patient populations, it can be expected that development times may even increase if the large sample sizes currently required for demonstrating a survival advantage remain a criterion for approval. The biotechnology industry, which has made cancer a dominant theme for its development programs, continues to struggle from undercapitalization and development resources, both qualitative and quantitative, that are often not sufficient to meet the time challenges. This is an important factor that should be considered by the NCI in regard to the importance of this segment of the industry, a true seedbed for innovation, for the fulfillment of national goals.

It often takes many more years after the approval of a new cancer agent before oncologists have refined the spectrum of tumors for which the new product demonstrates efficacy. In addition, it is common for there to be considerable postapproval development for parameters of dosage and method of administration, often in the form of new combinations that optimize use. During that time, the approved FDA prescribing information cannot keep up with the advances in applications to which practicing oncologists are made aware through the presentations and publications of respected colleagues and cooperative groups. Indeed, it can be posited that an oncologist who restricts the use of a cancer drug to the official, sanctioned prescribing information is likely not practicing state-of-the art cancer care. Getting the new product into the hands of oncologists who have both the training and experience to use the agent safely has traditionally resulted in advances in applications.

From a simple return-on-investment calculation, the large costs of participation in the field of cancer therapeutics coupled with the high risk (the very low rate of new drug approvals) has caused the pharmaceutical industry to question its commitments to this field of investigation relative to others that may provide a more predictable and larger market. Past investments in cancer research have resulted in a vastly improved understanding of cancer biology that can be translated into exciting opportunities for new genomics-based drugs. The much sought after goals of higher degrees of specificity and potential for improved efficacy appear attainable. This is not the time for the industry to reduce its commitment to the field of oncology. That, however, is a risk, and one that must be addressed by the National Cancer Program. While politically incorrect in the current climate of demonization of the industry, the U.S. Congress and the FDA must consider the provision of inducements for participation, such as further tax relief for research investments and longer periods of exclusivity. Greater support of the biotechnology industry in NCI-sponsored programs also must be given careful study, since it is in the interest of the public to ensure that the investments in research and the opportunities so derived are not lost. But the promise of a more efficient system of drug review and approval, coupled with an expectation of realistic criteria for the assessment of therapeutic efficacy, would go far to maintain and perhaps increase private sector commitments to this field of research.

There is evidence that the pharmaceutical industry understands the dilemma and its responsibilities to the national effort, and that it is preparing a response. It was recently announced that an industry group, The CEO Roundtable on Cancer, has met and has under study a proposal to create a research collaboration that would spread the risks and rewards of cancer drug discovery [6, 7]. The model is based in part on Sematech, a research consortium created by the semiconductor industry in the 1980s in the field of computer chip technology. The end result, if such a collaborative effort were established, would be to allow companies to continue to participate in cancer research but at lower individual costs. This assumes that the thorny issues of intellectual property rights and potential government concerns about competitive practices can be overcome. We await the outcome of these ongoing discussions.

Given the impact of cancer on the welfare of the country, of special concern has been the lack of coordination of the total resource available to the government for the development and provision of new anticancer agents to the public. This is occurring even among those agencies that report to the U.S. Department of Health and Human Services (HHS), including the FDA, the NCI, and the Center for Medicare and Medicaid Service. In the 1960s and 1970s, the NCI and FDA worked together very closely. The NCI’s chemotherapy program, which dominated the early years of cancer drug development, helped establish approval guidelines with an FDA that did not have a cancer branch as a separate entity or experienced oncology staff. While this has changed, it can still be posited that the FDA, while serving as the final filter through which all new therapies must flow to reach the public, has not served as an active participant in the National Cancer Program. Instead, the FDA, for presumed reasons of objectivity, has worked in near isolation outside the national effort, receiving the products of the Program’s work and independently judging the merits of each submission. As an example, the FDA has, in the past, viewed the NCI not as a partner but as another drug sponsor to be regulated, in ways comparable with those used for companies in the private sector. There have been calls for greater coordination of the many agencies that comprise the total governmental resource for cancer, not only those resident within the HHS but also the programs in the military and the Veterans’ Administration; the goal would be the removal of layers of bureaucracy to achieve a more seamless system for communication and assignment of priorities for the national effort [2]. The creation of a new post, an Under-Secretary for Cancer in the HHS has been suggested, and not facetiously; the current threat to homeland security as measured by the number of patient deaths per year from cancer relative to those killed by terrorism, for which a cabinet post has been created, more than warrants such a consideration.

Much will depend on future FDA policy, which is being watched very carefully these days with the appointment of a new and dynamic Commissioner, Dr. Mark McClellan. Dr. McClellan appears to have developed an important line of communication and apparent coordination with the Director of the NCI, Dr. Andrew von Eschenbach. On May 30, 2003, the NCI and FDA announced "that they will establish a multi-part Interagency Agreement to enhance the efficiency of clinical research and the scientific evaluation of new cancer medications" [8]. Quite possibly, the past concerns will be addressed if the staff of each organization will now work in full cooperation in the process of drug development and review. There is emerging evidence to support such a contention. Recently, FDA Oncology Division Director, Dr. Richard Pazdur, announced a critical revision in the policy for accelerated approval of anticancer agents [9]. The importance of accelerated approval, a mechanism that should in our interpretation be the dominant form of approval for drugs for cancer and other life-threatening diseases, is that new drugs may gain approval based on the submission of "surrogate end points" such as response rate and time to tumor progression. Full, unrestricted approval, particularly for first-line indications, had required survival outcomes that necessitated many additional years of follow-up in controlled trials. Previously, this mechanism had a major drawback: the first drug that fulfilled an "unmet medical need" would be granted such approval, often with early and modest evidence of efficacy, and would effectively block accelerated approval of other drugs for the same indication, even those that appeared superior in efficacy and/or safety. Dr. Pazdur announced that a product that received accelerated approval would not be regarded as "available," and hence fulfill the unmet medical need criterion, until the completion of a postmarketing trial that confirmed clinical benefits, which has been a code word for survival improvement. This new interpretation overcomes one of the most important impediments that beset the system: a much larger number of new drugs achieving initial marketing authorization. Additionally, the pace of review and approval of new biologics, many of which are regarded as offering as much promise as targeted treatments, has also been addressed with the recent consolidation of specific functions of the Center for Biologics Evaluation and Research into the Center for Drug Evaluation and Research [10]. The expectation is that the review programs for biologics and drugs would be more efficient and consistent in the application of standards of efficacy and safety.

The most important evidence of a possible change in policy and procedures has been the approval of four anticancer drugs since May of 2003: gefitinib on May 5 after a positive recommendation from the Oncologic Drugs Advisory Committee in September, 2002; bortezomib on May 13; imatinib mesylate on May 20, and tositumomab and iodine I 131 tositumomab on June 30. All four of these therapies were approved under the accelerated mechanism (Table 1). Three of the recent approvals represent NCEs. If this is the beginning of a new trend, then it bodes well for an increased rate at which new cancer therapies achieve marketing authorization—a much welcomed change. Investigators, drug sponsors, and patients will be tracking the actions of the FDA to determine whether a new policy has been truly adopted, as evidenced by the efficacy end points required for approval and rate at which new drugs are brought to the public.

The multibillion-dollar public-private investment in the National Cancer Program and its mandate to reduce the morbidity and mortality inflicted by this disease group will require a bold commitment to exploit the promise of current scientific advances and fulfill the expectations and hopes of cancer patients and their families. One important component of this commitment must be the timely delivery of a much larger number of cancer therapies and preventatives. Let us hope that the new spirit of cooperation between the FDA and NCI and the trend toward streamlining of policies will result in an effective response to the urgent public need that cancer presents as the second-leading cause of death in the U.S. and a growing international problem.

REFERENCES

1. Kaiser J. NCI goal aims for cancer victory by 2015. Science 2003;299:1297–1298.
2. Schein PS. The case for a new national program for the development of cancer therapeutics. J Clin Oncol 2001;19:3142–3153.[Free Full Text]
3. Johnson JR, Williams G, Pazdur R. End points for United States Food and Drug Administration approval of oncology drugs. J Clin Oncol 2003;21:1404–1411.[Abstract/Free Full Text]
4. Tufts Center for the Study of Drug Development. Tufts Center for the Study of Drug Development pegs cost of a new prescription medicine at $802 million. Available at: http://csdd.tufts.edu/NewsEvents. Accessed November 30, 2001.
5. Holmer AF. Medicines in development for cancer: 395 new medicines in development offer hope in the War on Cancer. Pharmaceutical Research and Manufacturers of America. May 2003.
6. Warner S. Collaborations against cancer: pharmaceutical companies work with their rivals to produce cancer drugs. Scientist 2003;17:48–50.
7. Pollack A. An early step on collaboration on cancer drugs. New York Times. section C, page 1, column 3, May 13, 2003.
8. FDA. News Release. NCI and FDA announce joint program to streamline cancer drug development. May 30, 2003.
9. Usdin S. Fixing accelerated approval. BioCentury. June 9, 2003:A13.
10. Coghill K. CBER, CDER consolidation plan goes into effect today. BioWorld Today 2003;14:1.

This article has been cited by other articles:

				P. S. Schein and B. Scheffler Barriers to efficient development of cancer therapeutics. Clin. Cancer Res., June 1, 2006; 12(11): 3243 - 3248. [Full Text] [PDF]

				E. K. Rowinsky Curtailing the High Rate of Late-Stage Attrition of Investigational Therapeutics Against Unprecedented Targets in Patients with Lung and Other Malignancies Clin. Cancer Res., June 15, 2004; 10(12): 4220S - 4226S. [Abstract] [Full Text] [PDF]

This Article Full Text (PDF) eLetters: Submit a response to this article Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Services E-mail this article link to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Reprints/Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Schein, P. S. Articles by Carter, S. Search for Related Content PubMed PubMed Citation Articles by Schein, P. S. Articles by Carter, S.