A Review of the Department of Defense's Program for Breast Cancer Research (1997)

The committee is favorably impressed with the Breast Cancer Research Program (BCRP) as implemented by the Army and believes it should be continued. Despite initial skepticism by the scientific community, the BCRP team overcame hurdles related to tight time frames and unfamiliarity with administration of a large peer-reviewed multidisciplinary research program. In the view of the committee, the BCRP has succeeded in establishing a fair peer review system and a broad-based research portfolio by stimulating scientists from a wide range of disciplines to participate as applicants, reviewers, and advisers.

The BCRP fills a unique niche among public and private funding sources for breast cancer research. It is not duplicative of other programs and is a promising vehicle for forging new ideas and scientific breakthroughs in the nation's fight against breast cancer.

Among the most outstanding features of the program are the flexible approach taken for setting priorities annually; the involvement of breast cancer survivors (consumers) in the grant peer review process; the level of commitment and diligence of the individuals who serve the program in various capacities; the commitment and support of the program director; the low administrative costs that allow the greatest share of funding resources to be awarded as grants; the use of outside experts for evaluation; and the unwavering respect and advocacy for this program among breast cancer advocacy organizations nationwide.

During its first two years (i.e., FY 1993/1994), the program was established and managed according to both the spirit and letter of the 1993 IOM report. Those responsible for the organization and management of the program deserve special commendation for the ingenuity and resourcefulness that forged a structure and processes that are for the most part working well despite a series of limiting circumstances.

The peer review system was established in record time, although not without some difficulty. The IP is to be commended for recognizing weaknesses in the first year's procedure and recommending that the Army rebid the contract. United Information Systems (UIS) was selected.

In the third year of the program (FY 1995), two consumers (breast cancer survivors nominated by an advocacy group) were placed on each scientific peer review panel, an innovation now being evaluated. Meanwhile, testimony the committee heard from consumers and other peer review panel and IP members indicates that most observers have found the participation of consumers to be a very positive aspect of the BCRP peer review process, and one that may serve as a model for other peer review systems.

The additional years of funding that began with FY 1994 presented a considerable challenge to the leadership of the program because IOM (1993) did not specifically address the possibility of additional funds. The fact that the program is funded for only one year at a time has understandably hampered the ability of the program managers to plan for the longer-term. For example, it has prevented the establishment of standing primary review panels, resulting in lack of standardization of priority scores across the ad hoc panels. Year-to-year funding has also produced too short a time frame between the publication of the announcement of each grant cycle by a Broad Agency Announcement (BAA) and the deadline for grant applications, and exacted an unduly heavy toll in time and energy on those involved in the various stages of the process.

Based on abstracts of projects funded in the 1993/1994 and 1995 cycles, the committee determined that the portfolio covers science that is responsive to the range of six questions posed in the 1993 IOM report. The distribution of funds was such that the majority supported basic molecular and cellular biology of breast cancer with far less going to epidemiological, psychosocial, and health services research. No inherent bias was apparent, though, insofar as the number of funded proposals was proportionate to the number of applications received for each discipline. Reliable methods to measure the success of the BCRP investment are not yet in place. In addition, it was considered premature for this committee to evaluate the quality of the portfolio of funded projects, since most funded projects are not complete and progress reports were not available to the committee.

The committee is concerned about the wide range of, and sometimes conflicting, responsibilities currently placed on the IP as a result of the lack of

scientific infrastructure within the Army. It recognizes a need for independent evaluation of the function of both tiers of review by an oversight group outside the Army.

1. Continue the Army's BCRP and make efforts to obtain multi-year authorization of and funding for it. Longer-term stability would allow longer-range programmatic planning, establishment of standing peer review panels, and implementation of more efficient and effective grants administration procedures (e.g., more timely release of the BAA, recruitment of appropriate reviewers, and optimization of review assignments). This could be achieved through either incorporation of the program into the annual DOD budget or multi-year authorization of funding by Congress.

2. Develop and implement a plan with benchmarks and appropriate tools to measure achievements and progress towards goals of the BCRP annually and over time. This would allow an evaluation of the effectiveness of the different funding mechanisms, with particular emphasis on IDEA grants (e.g., have the IDEAs generated new avenues of research or provided major breakthroughs) and recruitment and training grants. Elements of the process could include examination of records of publications and presentations, success by investigators in obtaining other grant support relevant to breast cancer, and identification and tracking of investigators who were recruited into breast cancer research by BCRP funding. Program evaluation should also measure achievements of the programmatic aims outlined in the 1993 IOM report.

3. Consider establishing a permanent non-Army oversight committee that is independent of both the IP and the contractors. Since responsibility for recommendations on policy setting and executive functions both rest with the IP, some members of the committee agreed that a separate mechanism for oversight and evaluation of the BCRP should be established. For other committee members, the fact that the IP has responsibilities in both areas was of lesser concern since no evidence was detected that the IP had failed to meet or had abused its responsibility. Despite differing views on the committee regarding the need for a group to oversee the work of the IP and the BCRP in general, the majority of this committee agreed to recommend the establishment of a relatively small permanent oversight group that would be responsible for quality assurance and program evaluation activities. This group would include scientists and clinicians experienced in directing research programs, widely

respected leaders in cancer research, as well as a consumer representative. Members could come from academic, medical, and other relevant organizations. The group would report directly to the BCRP Director and would have access to all information needed to oversee and rigorously evaluate the program in an ongoing fashion.

1. Establish measures to ensure the continuation of the current strength of the Integration Panel. The committee believes that the IP represents a new and imaginative concept in planning and monitoring a research grants program. By functioning as a second-tier (programmatic) review and council, and reporting to contractors and predominantly nonscientific administrators within the Army, the IP wields considerable power in deciding investment strategies and funding policy. The unquestionable success of the IP is the result of the high level of dedication and professional excellence of its members. The committee is concerned that it may be difficult to continue to recruit individuals with both the expertise and the level of commitment needed to sustain the wide range of current responsibilities of the IP.

   The committee believes that it is important to maintain the current high status within the research community that serving on the IP confers. In part, this will be sustained by continuing to accord a high level of responsibility to the Panel. However, the workload of individual IP members should be reduced where possible. For example, if the program's funding is stabilized, tasks such as development of program announcements and proposal formats, orientation of executive secretaries and development of new investment strategies may not need to be revisited by the IP every year. However, the program's unique flexibility should be protected as the program matures.

   The amount of work taken on by individual IP members should be flexible to ensure continued willingness to participate and diversity with respect to area of expertise, gender, ethnicity, and the mix of junior and senior investigators. The committee recommends that a broad range of perspectives continue to be represented on the IP, from both the research and consumer advocacy communities.

2. Spell out in more detail in the BAA the types of proposals sought, the programmatic evaluation criteria, and exclusionary parameters. The concepts of "innovation" and "translatability," espoused in the 1996 funding cycle, need to be developed and defined more extensively. Clarity of definitions, in the minds of applicants, peer reviewers, and IP members, is essential for reaching the programmatic goals envisioned. The BAA should be explicit in

inviting proposals in currently underfunded areas of epidemiology, psychological, social, and quality of life issues, and health care delivery research.

3. Lengthen the time between release of the BAA and the deadline for submission of proposals. This would require shortening the time between appropriation and release of funds from the DOD to the BCRP. This recommendation is especially important for large multidisciplinary proposals that require coordination between a number of basic and clinical researchers.

4. Increase the time between receipt of applications and first-tier peer review panel meetings. This would facilitate assignments of applications to the most appropriate panels and recruitment of the best and most appropriate ad hoc reviewers. Special emphasis panels may need to be constituted to deal with diverse emerging research directions and multidisciplinary proposals.

5. Communicate detailed information about consumer participation in the BCRP peer review process to the scientific community. This is an innovative experiment that is currently being evaluated by a questionnaire study, the outcome of which will be of great interest to other private and public funding agencies.

6. Move toward establishing standing review panels. Include some of the same peer reviewers on consecutive committees to increase reviewer familiarity with the procedures and goals of the program and to provide more consistency in rating patterns.

7. Improve feedback to applicants whose applications were not funded. To dispel myths about "secret criteria" supposedly used for funding BCRP proposals, communicate the fact that scores and percentiles carry different weights in the BCRP's ad hoc review system as compared to those used by other funding agencies. IP decisions not to fund applications within the funding range (and to fund applications below the funding range) should be fully documented and the rationale should be communicated to applicants.

8. Establish a procedure for resubmission of unfunded applications. Proposals that have been revised according to the previous scientific peer reviewers' critiques provided to the applicant should be eligible for resubmission in the next funding cycle. Responsiveness to the previous critique should be made an evaluation criterion.

9. Establish a procedure for competitive renewal applications. In the framework of a long-term BCRP, successful projects should be considered for continued funding. In particular, this would allow the BCRP to capitalize on successful IDEA grants as well as other types of awards. In the review of renewal applications, past progress made while receiving BCRP support should be taken into account as one of the scoring criteria.

10. Revise the application process to make it less cumbersome. To reduce the workload of applicants and Army personnel, the Army should

consider accepting institutional assurances in the areas of human and animal use and laboratory and environmental safety, in the same way other federal funding agencies do.

11. Reduce the time it takes between funding recommendation by the IP and actual awarding of funds to the investigator's institution. Streamlining of award and contract negotiations could be accomplished by appointing a program officer dedicated to the BCRP and by increasing the number of staff positions.

12. Streamline the annual reporting process and allow awardees more flexibility in changing experimental design and methodology. It seems counterintuitive to fund a 3-year Innovative Developmental and Exploratory Award (IDEA) that is by nature high-risk and open-ended, and yet manage it like a contract with close monitoring of adherence to a statement of work that was defined at the time of the award. Since no preliminary data are required for these awards, the results of initial experiments and/or progress made by others in the field may suggest a more promising research strategy or more appropriate methodology to attain the original goals of the funded proposal.

13. Allow awardees flexibility in use of funds across spending categories. This would allow the optimal use of available money toward reaching the goals of the project.

The 1993 IOM report identified six questions on the causation, prevention, screening, detection, diagnosis, and optimal treatment of and recovery from breast cancer that were to be used as a framework for breast cancer research. The report recommended that research projects funded under the program be directed toward answering one or more of those six fundamental questions. The committee notes that 50% of the funding to date has gone to address the first two questions, and reiterates the continuing importance of the other questions.

The committee finds that the six fundamental questions remain a useful framework for elaborating its recommendations for future research emphasis, as follows:

1. What genetic alterations are involved in the origin and progression of breast cancer?

2. What are the changes in cellular and molecular functions that account for the development and progression of breast cancer? The first two questions address a single fundamental issue, the identification of the cellular events involved in the pathogenesis of breast cancer. The identification and

characterization of the genes involved in breast cancer initiation and progression, including invasion and metastasis, will facilitate study of the basic physiology and biochemistry of the normal breast, because it will become possible to assess the role of these genes in normal breast development and function.

Breast cancer is caused by multiple genetic changes, some of which initiate the malignant process and some of which are responsible for tumor progression, including invasion and metastasis. Thus, studies to understand the mechanisms involved in tumor initiation and progression, the sequential steps from normalcy to malignancy in the breast, and the biochemical and biological functions of the relevant gene products present great opportunities for the development of new approaches to control this disease. Such studies may result in the development of diagnostic tools capable of identifying heritable and acquired changes that can be detected before the cells become invasive, or even in the premalignant phase, and also in knowledge of the likelihood of an in situ cancer's progressing to invasion. Furthermore, novel therapies capable of eliminating or terminally differentiating breast cells carrying the genetic changes predisposing to malignancy could be developed. The development of such gene therapy requires a better understanding of the genetic and immunological basis of breast cancer, with the vaccine approach to prevention and treatment facilitated by knowledge of the new altered gene products and peptides expressed in cancer cells.. Innovation and progress in any one of the areas noted here depends on progress in other diverse areas.

3. How can endogenous and exogenous risk factors for breast cancer be explained at the molecular level? The challenge to epidemiology is to move beyond examination of traditional risk factors to basic and applied investigations using genetic information to assess both risk and prognosis factors. Knowledge of the genes involved in the complex cascade of events leading to tumor development and progression will not, by itself, tell us how best to intervene in the process. The goal should be a complete understanding of the natural history of breast cancer through molecular epidemiological research. Studies of interactions of genetic and environmental or other nongenetic factors should be given high priority. This work will require close collaboration of clinical and basic scientists. The natural history of breast cancer and factors that influence prognosis need to be understood at both a histological and a molecular level. Epidemiological studies should evaluate new and existing risk factors at the molecular level with emphasis on hormonal, geographic, and family history variables. Emphasis should be placed on identification of new factors whose molecular mechanisms explain cancer risks not explained by know risk factors. There is an ongoing need for methodological research (including biostatistical modeling), investigations into measurements of exposure, intermediate markers

of carcinogenic processes, and sources of bias that can affect new types of studies.

4. How can investigators use what is known about the genetic and cellular changes in breast cancer patients to improve prevention, detection, diagnosis, treatment, and follow-up care? Knowledge of a woman's genetic makeup should facilitate the determination of whether she would benefit from a particular treatment and of what her chances would be for good health and quality of life. Studies to determine the optimal way to counsel women with genotypes that place them at risk will assist in developing informed consent procedures for testing and methods for effectively communicating test results. Implementation of preventive measures in high-risk women requires the full understanding of the natural history of breast cancer and the efficacy of various interventions, stratified by genotype information.

   Multi-institutional, randomized, and controlled clinical trials should precede the widespread clinical application of promising clinical research. Long-term outcome studies based on established clinical trial principles and statistical methods should be continued to validate (or not) final outcome—for example, mortality. The outcome studies should include quality of life and risk tolerance issues. Finally, there is a need to periodically update systematic reviews of these trials.

   Furthermore, since 1993, women with breast cancer have had increasing influence in discussions relating to the direction and content of breast cancer research; and they will continue to do so. For example, in testimony to this IOM committee, consumers have asked for additional research in the areas of prevention and treatment of lymphedema, long-term effects of axillary node dissection, living with metastatic disease and treatment for it, hormone replacement therapy for menopause, detection and prevention measures for women with inherited susceptibility to breast cancer, and weight management.

   Complementary and alternative medicine interventions should be subjected to the same standards of testing as traditional interventions. About one-third of Americans are using complementary and alternative medicine, and breast cancer patients are particularly interested in these approaches, despite the widespread negative views held by physicians trained in the Western world.

5. What is the impact of risk, disease, treatment, and ongoing care on the psychosocial and clinical outcomes of breast cancer patients and their families? Behavioral, psychological, and social research has focused increasingly on race, ethnic, and cultural differences, and the psychological effects of genetic testing for breast cancer susceptibility. Work in these areas should continue where gaps remain. There is increasing recognition of the importance of survivorship issues, especially because growing numbers of women are living longer with the disease. Survivorship issues are encompassed under the rubric of "health-related quality of life" research. Studies are needed

to better understand how breast cancer and its treatment influence women's evaluation of the quality of their lives and which variables are most influential in terms of diminishing or improving the health-related quality of life for breast cancer survivors and their families. Thus, there is continuing concern with improving knowledge of the range of disease and treatment consequences that occur such as body image, depression, early menopause, the psychological impact of long-term treatments, the impact of breast cancer on family and caregivers, economic hardship (e.g., loss of earnings, treatment costs), functional limitations (e.g., sexual and physical), and social role disability. Studies of disability prevention are also essential for maximizing the breast cancer survivors' ability to participate in valued social roles and activities.

6. How can investigators define and identify techniques for delivering effective and cost-effective health care to all women to prevent, detect, diagnose, treat, and facilitate recovery from breast cancer? The IOM (1993) outlined a number of targets for health services research including: barriers to state-of-the-art health care, health care seeking behavior, patient treatment preferences, and barriers and inducements to participation in clinical trials. These topics remain important. Other areas for investigation have emerged, including access to care, patterns of utilization of health services, patient–provider communication, provider education and behavior, economic and cost analyses, issues relating to policy setting and guidelines, and health care delivery systems.

Use of computer information systems is increasingly important in patient tracking, tissue bank administration, networking genetic information, and facilitating enrollment in clinical trials. These systems require additional investigation prior to widespread implementation because of confidentiality and acceptability issues.

Studies regarding ethnic, cultural, and personal differences in health beliefs and health care seeking behavior will yield important information for those providing care and setting policy. Also necessary is accurate, reliable, unbiased information on direct and indirect costs associated with genetic testing, prevention strategies, screening and diagnostic techniques, or a given treatment; such information is a critical component of realistic health care planning and delivery. An area of urgent importance is the effect of managed care on breast cancer screening, detection, treatment, and follow-up. There is concern about the trade-off between quality and cost of health care.