Hematology Grants Workshop—2000

Robert F. Todd, III, M.D., Ph.D.^*

Significant opportunities are currently available to support the fellowship training, career development, and targeted research projects of investigators in the fields of hematology and oncology. Tables1 and 2 provide a brief description of the types of awards offered by the NIH and the Veterans Administration. Tables3,4, and 5 outline research awards available to fellowship trainees (fellowship awards) and to young-to-mid level faculty (new investigator and career development awards) from nonfederal sources. The latter have been extracted from a particularly useful monthly “Hem/Onc Grants Newsletter,” which is available by subscription at http://www.dralanlevineassociates.com.

Donald M. Miller, M.D., Ph.D.^*

Success or failure as an academic scientist depends, ultimately, on the ability to obtain funding for your work. Thus, grant writing is an extremely important but often unappreciated aspect of academic medicine. It is important to develop a personal strategy of grant writing that will optimize your chances of success.

Writing a grant can be regarded as a game of wits with the reviewer. If you (the Principal Investigator, “PI”) are to win, you must convince the reviewer that you 1) are a good scientist; 2) have a good hypothesis; and 3) will be able to do the work that you propose. The reviewer must be convinced that your application is in the top third of the applications that he or she is reviewing in order for it to have a chance for funding. It is important to understand the review process from the reviewer's viewpoint in order to optimize your chances of success. It is also important to understand that not every application will be funded and that you will, as in the case of baseball, be quite fortunate to bat.350. Thus you should be prepared to revise and resubmit. You should also develop a funding strategy which includes submissions to multiple sources of funding. It is a wonderful problem to be required to decide which of two similar applications to decline.

First of all, reviewers are not the enemy. They are, quite likely, your friends. Almost certainly, they want you to be successful. They are scientists with the same day to day concerns as you. Although they have been successful at getting their grants funded, they have almost certainly had grants rejected, as well. Most study section members are “good citizens” who view the many hours that they spend reviewing grant applications as a payback for the service that has benefited their careers. They have a number of grants to review which may range from six to twelve. They are looking for that “outstanding grant” that they will be willing to fight for. You should try to convince them that your application is that one.

The reviewers are likely to be tired when they review your grants and may be not be in a particularly good mood. Therefore it is extremely important to make your grant as easy to read as possible. Irritants such as mispellings and syntax errors should be avoided at all costs. The organization of your application (with appropriate subheadings) and the correct font can make very important contributions to a sense of “competence.” It is also important to recognize the fact that, although the reviewers will read your application carefully, certain elements of the application will have greater importance in deciding their opinion of your ideas. As in most human relationships, first impressions are very important. Thus, the Abstract and Specific Aims sections are generally important areas which each reviewer reads very carefully.

Unless you have thought through your proposed work in great detail, you will not be able to relate these ideas in a clear fashion to your reviewer. The hypothesis should be very clearly stated in the Abstract and the Specific Aims sections. The specific aims should follow very logically from the hypothesis and should address very specific and well-defined goals. They should be stated succinctly and clearly in a manner that gives a good idea of the work that is being proposed. They should be “aims” and not a description of the experiments that will be performed. The number of specific aims should be relatively small, from four to six, in order to avoid the appearance of diffuseness. Subdividing each specific aim into five or six sets of experiments is a reasonable approach, but should be done in the Methods Section, not in the Specific Aims section where it can significantly detract from the a succinct set of aims. Likewise, delineating the experimental methods in the Specific Aims section may be confusing or distracting.

The Background section should not attempt a comprehensive review of the entire area of research. Rather, you should use this section to demonstrate the importance of the work that you propose, to delineate and document the questions that you are addressing and to provide the reviewer with enough information to judge the relative importance of your work. It is important to use the background information to stress the importance of your proposed experiments as they relate to the rest of the field. While you should emphasize differences of your work with other labs in the field, this is not a place to be critical of other workers; the scientific world is a small one and one of these scientists may review your application.

The Preliminary Data section should be used to document your expertise in the area of the application and also to provide evidence to the reviewer that you will be able to accomplish the specific aims. If work presented has been published or submitted, make that clear by providing references in bold or underlined fonts. Do not use this section to present data from every experiment that you have ever done but, rather, attempt to provide a concise and convincing description of your work. If you have important work (and important publications) that are peripherally related to the theme of the application, use them as well. Figures should not be “shrunk” to fit on the pages; it is far better to use a few clear charts or graphs than it is to use figures that can only be discerned with a magnifying glass. Do not be sloppy. The reviewer will likely judge the quality of your proposal, in part, by the quality of your preliminary data. Similarly, don't overinterpret your data or draw conclusions that are not clearly justified. This will lead to distrust on the part of the reviewer and is likely to put him or her into a bad mood.

As with the other sections of the application, the Experimental Methods section of the application must be written very clearly. Each specific aim should be addressed with an individual set of experiments. It is critically important that the specific aims listed in the Abstract, Specific Aims section and the Experimental Methods section are exactly the same. If the specific aims are different in each place, the application is much more difficult to read and much less likely to be funded. Experiments should be outlined in considerable detail, but it is not necessary to provide every detail of every reagent. You will be awarded Brownie points for every mention of “control” and “potential problems.” These should be clearly delineated with appropriate subheadings.

There are several ways in which the reviewer will judge your qualifications as a scientist. You should emphasize your training and experience. Particularly as a young scientist, you should describe the training that you have had and discuss its relevance to the current proposal. If your mentor is a leader in the field, mention this fact. If you have unique skills that make you especially qualified to do the work that you are proposing, emphasize that point. Discuss your productivity. Emphasize publications that are directly relevant to this proposal. Conversely, if there have been periods of low productivity, discuss them. Often there are very valid reasons (clinical training, illnesses, change of scientific direction, etc.) that may obviate reviewer concern about these issues.

It is extremely important to convince the reviewer that your application addresses an important hypothesis and that your proposed experiments will provide important new insights and will test the main hypothesis. Although the importance of your hypothesis will be clear to you, it may not be clear to the reviewer, who likely works in a related but distinct area. You should explain very clearly why this work is important and delineate the important questions of the area that will be answered by your experiments. Write down the underlying hypothesis of your application before you start writing, then place it prominently in the Abstract and Specific Aims sections.

It is important to assume that you know more about your area than the reviewer. Although the good reviewer is likely to be well read and familiar with your area, it is impossible for reviewers to be intimately familiar with the literature of every area they review. This means that you need to outline the basic tenets of your field in a way that emphasizes the importance of your project. Without being condescending, you should outline the basic facts that support the importance of your application. It is extremely important to state very clearly the unique features of your proposed project and the importance of the results. Do not overstate the importance of your work, but emphasize the contribution that you are making.

This is particularly important for young investigators who do not have a long track record of success. Perhaps the most effective way to convince the reviewer that you will be successful is to write a cohesive, clear proposal. Think very carefully about the outline of your application. Make certain that the experiments that you are proposing directly address the specific aims. Make certain that your application does not appear to have been hurriedly written or sloppy. An application that has frequent misspellings or mistakes of organization is very unlikely to make a favorable impression. Likewise, an application which is presented in a very small font or in which the margins abut the edges of the pages is difficult to read and unlikely to do well. Emphasize your experience in the field and document previous productivity in a way that will strongly suggest that you will be successful with this project, as well.

Every scientist, no matter how good, has had grants that have not been funded. It is very disappointing, no matter how senior the investigator, to have this happen. As you plan your strategy, you are likely to feel that you have not been treated fairly and that the reviewer was either biased or did not understand the brilliance of your work. It is very important not to allow these feelings to make their way into the revised application. Rather, you should take the critiques as constructive criticism, whether or not you agree. The letter describing your revisions is a critical part of a revised application and should acknowledge the helpfulness of the reviewers comments, no matter how painful it is to do so. You should address each and every criticism of your application, no matter how small, trivial or insignificant. You should also point out that you have done so. If you disagree with specific comments in the critique, say so respectfully and acknowledge that the reviewer could possibly be correct as well. Do not take an angry or defensive tone. It is likely that at least one of the reviewers of the second application will be the same and insulting his or her intelligence is certain to result in another unfunded application. You should strongly emphasize the amount of work accomplished since the initial submission. If you have accomplished one of the specific aims or if you have done 25% of the work you proposed, say so. Emphasize these important facts by bold or italic fonts to recognize their importance. The best strategy in preparing a revised application will be described in greater detail in Section III.

This is a time of great optimism in American science. The tools to address important clinical problems are better than they have been in the past. More than any time in the history of biology and medicine, science is making major contributions to society. The key to participation in this incredible excitement is the ability to develop a funded research program. The development of a personal grant writing style and philosophy can significantly ease the tension associated with this part of academic medicine.

Roy L. Silverstein, M.D.^*

One of the most disappointing events in the life of any scientist is receipt from the NIH of summary statements (sometimes called “pink sheets” in reference to the now abandoned custom of printing them on colored paper) indicating that a grant application is not likely to be funded. It is worth remembering when this happens that most investigators, even well-established leaders in the scientific community, have gone through this humbling experience. It is also worth noting that while, on average, only 20-25% of applications at any given cycle receive a fundable priority score, the ultimate likelihood of any proposal being funded is much higher, approaching 30-40% in some Institutes. In other words, the process of revising and resubmitting an application is worth the time and effort that it entails. This manuscript will review strategies for increasing the effectiveness of grant revision.

Before beginning, we need to understand who is it that is reviewing your grant, how your grant got into their hands, and how scientific priority and funding decisions are made at the NIH. About three-quarters of the Research Project Grants (RPGs) submitted to the NIH are unsolicited, investigator-initiated proposals. These include single project proposals, the most common being the RO1, as well as larger, multi-project proposals, such as the PO1 (program project). Career development awards and training grants, such as the “K” awards, may also be submitted without solicitation. The remainder of the RPGs submitted to the NIH are in response to specific programs initiated by the Institutes and Centers (I/C) of the NIH, so called RFAs (Requests for Applications). On the three major yearly grant application receipt days thousands of packages are dropped at the doorstep of the NIH Center for Scientific Review (CSR), the agency responsible for providing peer review for most proposals.

The first step in the peer-review process is the simplest, but can have important impact on the ultimate prospects of a proposal. After the applications are logged into a database, CSR Referral Officers review the contents of each application using written guidelines and decide which of the 20 Integrated Review Groups (IRG) would be most appropriate for assessment of scientific merit, and which of the 25 I/C would be most appropriate to provide funding. The IRGs are basically clusters of study sections organized thematically to review similar science. It is this step that puts an application into the hands of a specific study section. Table 6 lists the current IRGs. The IRG system was designed to provide at least one appropriate venue for the review of all science relevant to health-related research, and so that the research topics encompassed by each IRG are sufficiently cohesive to allow an external advisory group of scientists for that IRG to judge the content of its entire portfolio. Because of the crucial role that the IRG system plays in maintaining the integrity of the peer review process, a plan was recently developed to revise the system. This was based on a sense within the NIH and the scientific community that the rapid advances in molecular biology, cell biology, and genetics had made the current thematic clusters obsolete, and that the current IRG system was not adequate for providing fair peer review to patient-oriented research and to the new technology-driven research, such as gene expression arrays. The proposed new IRG structure is also outlined in Table 6.

Many if not the majority of grants from members of ASH are assigned to the Cardiovascular Sciences IRG. Within Cardiovascular Science are eight chartered study sections, of which three have most relevance to ASH: Hematology-1 (HEM-1), Hematology-2 (HEM-2), and Pathology A (PTHA). The first two handle a broad array of hematology-related proposals, while the latter reviews, among others, grants involving endothelial cell and vascular biology. A significant number of applications, however, are assigned to other IRGs, including AIDS and Related Disorders, Immunologic Sciences, Genetic Sciences, and Oncologic Sciences, and thus can go to a wide variety of different study sections for peer review. The CSR provides peer review, assigns priority scores and calculates a percentile rank, but the ultimate funding decisions come from the individual Institutes and Centers of the NIH. These decisions are made three times each year, several weeks after all of the study sections complete their work. They usually follow the percentile rankings set by CSR, but they are not compelled to do so and can make decisions moving applications up or down based on their programmatic goals and needs, as defined by the Director and Program Staff. Each I/C also has an Advisory Council, made up of senior scientists, leaders from patient advocacy groups, and other community leaders. This council helps in setting I/C priorities and policies and in making the ultimate funding decisions. Three NIH Institutes provide funding for most of the grants held by ASH members, the National Heart, Lung, and Blood Institute (NHLBI), the National Cancer Institute (NCI), and the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK).

The assignment of a grant to both study section and I/C is an interactive process involving CSR Referral Officers, study section Scientific Review Administrators (SRAs), and I/C program representatives. This decision is based mainly on the title of the proposal and the abstract, but the Referral Office also seriously considers written requests from applicants for both study section and Institute assignments. All you have to do as an applicant is include a cover letter with the application indicating your preferences. The CSR maintains a website, which can be accessed from the NIH home page (www.nih.gov; click on “Institutes and Centers”), that maintains a list of all study sections, their rosters and the dates of their meetings. It is definitely worth reviewing this site to get a feel for the constituency of the study sections and which ones might be most hospitable to your proposal. Careful attention to the title and abstract can also help steer a grant to a desired study section and I/C. For example, if you are studying the structure-function relationships of a particular transcription factor, putting the words “hematopoiesis” or “blood cell development” in the title will most likely direct the grant to one of the HEM study sections and to NHLBI or NIDDK, rather than a gene regulation study section and to the Institute of General Medical Science.

A typical chartered CSR study section is composed of 18-20 individuals, nominated by the SRA from among active, NIH-funded researchers in the biomedical community. Members meet three times each year, reviewing as many as 80 applications each time, and serve for multi-year terms. Service on a study section is thus a major commitment, and in my experience members take their commitment very seriously. Ideally the combined knowledge of the group spans the diversity of subject matter assigned to that study section, but of course this is not always possible and the study section's membership is supplemented by temporary “ad hoc” members and written outside opinions. In some instances, Special Emphasis Panels (SEPs) are formed by CSR on an ad hoc basis to review applications requiring special expertise, or due to special circumstances (such as when a conflict of interest occurs). SEPs may also be formed by the individual I/Cs to review I/C initiated RFAs.

After applications are assigned to a study section, the SRA reads them, checks for completion, and assigns them to study section members best suited to review that specific application. At least two, but usually three members are assigned each grant for written review (one primary and one or two secondary) and one or more additional members assigned as discussants. Discussants can prepare a brief written review, but it is not required. All members of the study section receive all applications (with the exception of those for which that member is in conflict) approximately six weeks before the meeting, but in my experience, most members read only their assigned grants and just skim the remainder, looking at the titles, abstracts and specific aims pages.

The CSR has set the following goals for the review process:

• Set a high standard for scientific excellence, facilitating the selection of the best scientists and ideas on any given topic for NIH support.

• Optimize the benefits that can be gained from the progress of science and its contributions to health.

• Encourage innovation and risk taking.

• Exercise fairness.

• Be monitored continuously.

• Be clearly explained to both scientists and the general public.

To effect these goals reviewers are instructed to consider and comment specifically on the following issues and to use these criteria to generate a priority score:

• The overall significance of the research proposal; i.e. if the project succeeds what impact will it have on human disease and on our understanding of basic life processes; or will important new technologies emerge.

• Innovation in aims and approach.

• Qualifications of the PI. For new investigators the previous training environment is considered. For all investigators, productivity as assessed by peer reviewed publications is considered. A small number of high impact, high quality publications is more important that an larger number of publications in lower impact journals.

• The richness of the research environment.

• The experimental approach. This is basically whether the studies described in the application are feasible and whether the proposed experiments will successfully address the aims. Among the issues that the reviewers address is the specific experience of the PI with the proposed experiments, whether there is preliminary data to support each of the aims, and will the proposed experiments, if successful, allow definitive conclusions. Here, issues of study design, controls, pitfalls, interpretation of results and alternative strategies are addressed. In summary, the proposal should be important, feasible and novel.

All study section meetings follow the same general format. They usually last two days and take place in a hotel near the NIH. Members sit around a large conference table with the chairperson and the SRA, who jointly conduct the meeting, sitting together at the head of the table. Representatives from the various relevant I/C are usually present as observers and sit behind the conference table. A week or so before the meeting the SRA asks study section members for a list of R01 applications believed not to rank in the top half of scientific merit. At the beginning of the meeting a collated list is reviewed and a final list of “streamlined” applications established. These are not scored or discussed at the meeting, but reviewers' written critiques are provided to the applicant. In general, any one member of the study section can vote to remove an application from the “streamlined” list. The advantage of this system is that it allows more time at the meeting for discussion of applications judged to have scientific merit more likely to be funded and, since summary statements are not prepared, the applicant receives the critiques faster and has more time to prepare a revised application. It is important to remember that “streamlining” is not equivalent to disapproval; many streamlined applications achieve a fundable priority with careful revision.

The bulk of the time at the study section meeting is spent discussing the individual applications. Members come to the meetings with their written reviews. The chair usually asks each reviewer to indicate their enthusiasm for the proposal and then asks the primary reviewer to provide a brief summary of the proposal and a more detailed summary of his or her review, concentrating on the strengths and weaknesses. If all reviewers are in agreement in their initial assessment of enthusiasm, the chair will then ask the secondary reviewer and discussants to make additional brief points. If there is some disagreement, they will be asked to provide more detailed comments justifying their differences of opinion. After these comments have been made the chair will open discussion to the entire study section. A serious attempt is made to raise all appropriate issues and arrive at a consensus priority score. After a period of discussion, ranging from 10-30 minutes, the budget is discussed and human and animal rights issues raised. The reviewers are then asked to consider the overall priority independently of budget, but in practice, extremes of budget at the high and low ends can influence the enthusiasm for the project. Finally the chair re-polls the reviewers to determine if the discussion has changed their initial level of enthusiasm. Following this a secret written vote is taken.

The scale of scores is based on criteria defined by CSR (Table 7), and the study section members are strongly encouraged to “spread” their scores over the entire range. For RO1 applications, with streamlining of the bottom half, the scores should be spread form 1 to 3. In practice each study section develops its own “culture” of scoring, so that when the scores are averaged after the meeting, the percentile rank is calculated based on the entire pool of grants reviewed by that committee over the last few cycles. This means that similar priority scores from different study sections can result in surprisingly different percentile ranks. After the SRA and staff complete the summary statements for all of the applications (usually about 6 weeks), they are sent to the applicants and to the appropriate I/C for funding consideration. At this point the I/C program officials become the applicant's link to the NIH with regard to disposition of the application.

In my own experience on four different chartered CSR study sections and numerous Special Emphasis Panels, the review process is extremely fair and careful attention is paid to every proposal. Although the time spent on each application at the actual meeting is relatively brief, in the weeks before the meeting most reviewers spend at least 6-8 hours carefully considering each of the applications assigned to them as primary or secondary reviewer. The general tone of most study sections is positive, with a focus on identifying the strengths of each application. With applications that will probably fall outside the payline the reviewers generally make a serious effort to identify the most important and correctable weaknesses. The members of the study section are usually strong advocates for research in their fields and would like to see as many grants funded as possible.

As mentioned previously, despite the hard work of every grant writer and the best intentions of every study section, most applications will not receive a fundable priority score. As a result, virtually all successful investigators must become adept at the process of revising and resubmitting applications. At most study section meetings, as many as a quarter to a third of applications are re-submissions. As of October 1996 the NIH allows only two attempts at revision (A1 and A2) and only two years beyond the date of receipt of the initial unamended application. This decision was not meant to be punitive, but to recognize the historical reality that applications beyond A2 were rarely successful, and even when successful, were almost never followed by successful renewal. After three unsuccessful attempts, the NIH feels that the investigator would be best served by taking a fresh start at his or her research plan.

When preparing a re-submission there are several rules of thumb to follow. As with all applications, it is extremely important to pay attention to the basic rules of grants-manship. Remember that most members of the study section are busy senior scientists. While they are very willing to invest several hours on your application, if your grant is difficult to read your chances are diminished. Therefore it is very important to make your application easy to read and free of typographical errors. Use a readable font of reasonable size for a tired, middle-aged reviewer. A proportionally spaced, 12-point type, such as Times New Roman or Arial, is easiest on the eye. Be generous in your spacing; skip lines between paragraphs and don't crowd figures into the text. With a 25-page, single-spaced application the extra information “squeezed” in by crowding margins and paragraphs and using a small typeface is counterproductive. Figures must be high quality; take advantage of modern computer word processing and printing technologies and use color if appropriate—everyone else does! Place the figures into the body of the proposal, adjacent to the text that refers to the figure if at all possible. Most busy reviewers do not want to search through appendices or through pages of text to find a figure. Provide high quality originals for all of the submitted copies of your application so that all of the reviewers have the highest quality figures. Make figure legends as brief as possible. This can be done by providing a brief title for the figure that contains the conclusion drawn from the figure, by making sure the graphs and images are well labeled, and by describing them carefully in the text. Remember not to use small typeface in the figures, as they will not photocopy well.

Pay attention to your prose! Be clear and concise, avoiding jargon and long lists of abbreviations. Try to make your application “flow” in a logical manner. It is best to make your Specific Aims section brief. There should be a clear statement that outlines the broad goals of the proposal and the hypotheses that drive it, followed by a list of specific aims that is not simply a list of the planned experiments but rather outlines the conceptual framework that organizes the proposal. While it is safe to assume that the reviewers will be sophisticated and knowledgeable scientists, it is a mistake to believe that they will be absolute experts in your own specific field. Therefore the goals for the Introduction and Background sections are similar to those of a good, brief scientific review article; you want to excite the readers about your area of science and educate them as to the key open questions in the field. In addition, you want to convince the reviewers that your hypotheses are sound and that they are well justified by your own published work, that of others in your field, and your unpublished preliminary data. Since assessment of “significance” is a key component of the review process, be explicit in your text as to what makes your proposal significant.

The goals of the Preliminary Data section are two-fold. First, high quality preliminary data convinces the reviewers that the proposal is feasible and that the PI has the experience necessary to perform key experiments. Thus it is important to include preliminary data that supports the use of the key technologies in the experimental plan. As an example, if a key part of the plan is to study a “knock-out” mouse, it is important to show the reviewer that you have direct experience with this technology and that you are well on your way to having the mouse in hand; i.e. the gene is cloned, targeting vector made, and embryonal stem (ES) cells made. If the mouse is absolutely critical to the proposal, reviewers will probably want to see even more progress. Most reviewers expect to see preliminary data to support the feasibility of each of the specific aims. Second, the preliminary data should support the hypotheses that drive the proposal. Even if the proposed studies are elegant, feasible, and likely to succeed in your laboratory, the reviewers need to be convinced that the hypotheses are sound and significant.

The NIH, at the discretion of the study section SRA, allows applicants to submit supplementary preliminary data after the initial submission date. If you chose to do this, you should contact the SRA first to determine the proper procedure for that study section. You should submit new data as early as possible, preferably before the applications are mailed to the reviewers (usually 6 weeks before the meeting). No reviewer is happy to receive new information after he or she has already spent 6-8 hours reviewing an application and writing the critique. Supplementary data should be brief, important, and critical to the application; most successful applications are not made so by the addition of new last minute data.

The Experimental Design section of the grant is perhaps the most common place where the inexperienced grant writer stumbles. This should not be a “cookbook” type list of experimental procedures. The best way to show that an experimental technique is feasible is to provide convincing preliminary data, not a detailed description of the methodology. Reviewers are looking for an outline of a rational experimental approach with attention to controls, interpretation of expected results, frank discussions of pitfalls and limitations, and consideration of alternative approaches. Since assessment of “innovation” is a critical component of the review process, it is useful to explicitly state in your proposal where your approach is particularly novel.

The scientific community is a small one; there are few secrets. If your career status has changed since the initial submission or if the status of key collaborators has changed, it is wise to inform the study section of this (through the SRA). For example, if you accepted a position at a new institution, you should inform the committee of what this means in terms of your laboratory facilities, time to commit to the project, etc.

It is very important to take the criticisms and suggestions of the study section seriously and not to get hung up on concerns that you have “enemies” or competitors on the committee, or that your application went to the “wrong” study section, or that the reviewers made mistakes. Remember that the priority score reflects the consensus of at least three senior scientists who carefully read and reviewed your application, and for non-streamlined applications, additional input from an experienced group of grant reviewers at the meeting. If the critiques really do reflect mistakes in interpretation and understanding of your application, then the most reasonable conclusion is that your proposal was difficult to read and understand.

When the study section reviews an amended application the reviewers pay close attention to how you have responded to the previous critiques. They will not have your old application, but they will have copies of the summary statements and critiques. A successful re-submission always includes gracious and specific responses to the written critiques. Defiant re-submissions with only cosmetic changes and superficial attention to the critiques almost never succeed, no matter how eloquent the rebuttal. A thoughtful response that includes new data and that makes additions and deletions to the Experimental Approach section as suggested by the study section flatters the reviewers and is more likely to result in improvement in priority score. The introduction to the amended application should make it clear to the reviewers what was deleted, what is new, and how the proposal has been otherwise modified. Some applicants use high lighters or margin symbols to flag important changes and additions.

The first step in planning a re-submission is to read the reviews very carefully. I recommend making a list of all the specific criticisms, both positive and negative. In the introduction to the revised application it is always good to summarize the positive comments and thank the reviewers for them. Since there is always turnover of personnel on the study section, this serves to inform any new reviewers of the good feelings of the last set of reviewers. It is also wise to thank the reviewers for their help in identifying the weaknesses of your proposal. Once you have generated a list of the specific issues raised by the study section, it is useful to separate the minor technical issues from the rest. Even though a large part of the critique may be made up of these minor points, they are really not as important as the “big picture” issues. They should all be carefully addressed, but attention to these should not be the major focus of the re-submission.

The remaining criticisms should be ranked in order of importance, remembering that the key elements of the review process that determine priority score are “Significance,” “Innovation,” and “Approach.” It can be very helpful to ask one or more colleagues with recent study section experience to review your critiques. As mentioned before, nearly all scientists have been through this, so do not be embarrassed by the implicit “rejection” of the critiques. The language of the critiques tends to be understated and often to use terminology that has more meaning to the peers on the study section than to the applicant. An experienced colleague or mentor can help sort out these subtle meanings and help identify the critical changes that need to be made and new data that needs to be obtained to push the application into the fundable range. Another good source of information is the SRA. If the grant was not streamlined, the SRA can often provide a general sense of how the application was received by the committee, what issues generated the most discussion, and how enthusiastic the committee would be towards a re-submission. Institute program officers, if present at the study section, can also provide similar insights.

Among the more common criticisms that appear in summary statements are that the proposal is “too ambitious” or that it is “overly descriptive.” The first is especially relevant to younger investigators. It is easy to see how this happens; a scientist just getting started is excited about his or her work, anxious to do as much as possible, and perhaps a little fearful that reviewers will find the application weak if all possible studies aren't considered. The reviewers, on the other hand, see the proposal as diffuse and perhaps reflective of an inability of the investigator to make critical decisions and to prioritize experiments. If this criticism is raised it is essential to take it seriously, cutting where suggested by the reviewers and focusing on the most significant and most feasible aspects of the proposal. The flip side of this coin is represented by proposals that are judged to be overly narrow in scope. Here the reviewers are concerned that the focus is too narrow and that the studies will not contribute enough significant new knowledge to the field. This is a more difficult problem to address and often requires substantial new approaches on the part of the applicant; i.e. new aims and new preliminary data. Advice from senior colleagues can be very useful in this regard.

The criticism of a proposal as “overly descriptive” reflects the desire of the NIH to support research that reveals the fundamental mechanisms of human disease and/or basic life processes. This type of insight rarely comes from descriptive studies, no matter how well performed. If this criticism appears in the summary statements the reviewers will expect to see significant re-working of the amended application with the addition of new aims that are more mechanistic in nature and new preliminary data to support those aims. As with grants considered too narrow in focus, grants judged as overly descriptive may reflect hesitancy on the part of the investigator to ask the big and important questions. Finding a new collaborator to help in the design and execution of critical mechanistic studies is one strategy to address this criticism.

Another common criticism is that the approach of the proposal is a “fishing” expedition and is not hypothesis-driven. This comes up in particular with proposals involving gene expression arrays and yeast 2 hybrid screens. These are perfectly acceptable methods, it is just very important to try to frame the studies in the context of a hypothesis and, more importantly, to outline a clear and detailed experimental approach designed to interpret any information obtained from these screens. The study section will want to know how you will verify true positives, prioritize findings if there are multiple positives, and convert these discoveries to mechanistic insights.

Perhaps the most important advice for any applicant is to finish the proposal early and send it to an experienced colleague for rigorous criticism. After living with a grant for days and weeks, it is easy to lose perspective and to see clarity where others may see obfuscation. As chief of a large academic division, I am happy to pay a small honorarium to outside consultants to obtain a formal pre-review in the NIH style of any application from our faculty; I am sure that most Department Chairs and Division Chiefs feel likewise.

There are many complaints about the NIH peer review system. This is not surprising given that the majority of proposals reviewed at any given study section are judged to have priority too low for funding. On the other hand, a much larger number are ultimately funded and a large number of investigators are continuously funded through most of their careers. Persistence, flexibility, thick skin, and willingness to pay attention to friendly and wellmeant criticism will go a long way towards success.