Clinical Research That Can Change Clinical Practice
For many years in his clinical practice, and continuing to this day, Dr. Stephen Grobmyer, an associate professor of surgery, has been faced with a difficult clinical dilemma. When a patient is referred to him for evaluation and treatment of breast cancer, the diagnosis is often based on a biopsy that was taken by surgical excision ("open biopsy"). Although such biopsies involve a surgical procedure in an operating room setting, they are problematic from the standpoint of diagnosis and incomplete from the standpoint of treatment.
Diagnosis from open biopsy is problematic because it often creates the need for more surgery. On the basis of an open biopsy, the surgeon does not know the spatial orientation of the tissue in the patient relative to the cancer cells seen on the slide. This creates ambiguity about where residual tumor may reside in the patient — more surgery is needed. Moreover, while the open biopsy might excise the entire tumor, such “treatment” is often incomplete if the margins of the biopsy are not completely clear of tumor cells, and/or if the histologic subtype of the tumor indicates that additional surgery, such as sentinel lymph node biopsy, is needed. All of this adds up to the need for a second surgical procedure.
As early as 2005, reports in the literature on state-of-the-art diagnosis and treatment of breast cancer (Silverstein MJ et al., J Am Coll Surg 2005; 201:586-97) indicated that optimal management of an image-detected breast abnormality was needle biopsy ("minimally invasive breast biopsy," or MIBB). In 2006, this view was affirmed in the official consensus statement of the American Society of Breast Surgeons (http://www.breastsurgeons.org). And yet, Dr. Grobmyer and his breast surgery colleagues nationally are continuing to see large numbers of patients referred with a diagnosis based on open biopsy.
About two years ago, Dr. Grobmyer decided to quantify the problem in the form of a research study, testing the hypothesis that “despite the advantages of MIBB, open surgical biopsy continues to be widely and excessively used for the initial management of suspicious breast lesions in Florida.” Findings from this study showed that the overall rate of open surgical biopsy in Florida was about 30 percent, in contrast to the expected rate of 5 percent to 10 percent, and that reduction in the rate of open biopsy from 30 percent to 10 percent could be associated with a charge reduction of over $37 million per year in Florida. These findings were reported first in February 2010 at the annual meeting of the Southeastern Surgical Congress, and were published online Feb. 2, 2011 in The American Journal of Surgery.
This story was reported by The Gainesville Sun and many other newspapers, including The New York Times, on Saturday, Feb. 20. Later that day, the story was No. 4 on the list of 10 most-popular stories e-mailed off NYTimes.com. It also aired recently on NBC Nightly News with Brian Williams. If this level of interest is any indication, this study has the potential for significant clinical and public health impact.
There are at least three important lessons to be learned from this story of how a clinical dilemma can be turned into a research study that can change clinical practice.
The first is that almost all contemporary research is conducted by a team of investigators, brought together by a common interest, propelled by the excitement of the research question, and comprising individuals with complimentary backgrounds and areas of expertise. Dr. Grobmyer turned to Luke Gutwein, M.D., now a general surgery resident from the Department of Surgery, to take the lead in coordinating data collection and other operational aspects of the study. Dr. Gutwein embraced the study’s hypothesis and methodology, and ultimately presented this work at the Southeastern Surgical Congress and became first author on the published paper. Darwin Ang, M.D., M.P.H., an assistant professor of surgery, also joined the study, bringing expertise in health services research and outcome assessment. Julia Marshall, M.D., an assistant professor of radiology, brought an important dimension from her perspective as director of the UF Women’s Imaging Program. Other researchers involved in the project include Huazhi Liu, M.S., a statistics researcher in the department of surgery; Steven Hochwald, M.D., an associate professor and chief of surgical oncology; and Edward Copeland III, M.D., a distinguished professor of surgery.
Second, it is not necessary for research to be supported by a grant from the National Institutes of Health or similar funding agency to be important in shaping clinical practice. Valuable information can be gained from analytic, observational studies that have limited funding. In this case, Dr. Ang identified a database from the Florida Agency for Health Care Administration that would allow the team to test its hypothesis by applying a retrospective cohort study design. Unlike previous studies, in which only limited information could be inferred from databases that were restricted to cancer cases, the AHCA database contained information on all patients undergoing breast biopsy, whether open or MIBB, independent of whether the final diagnosis proved to be cancer. During the five-year study period, data on 172,342 breast biopsy procedures were available for analysis. A small grant (about $1,000) from the Department of Surgery was all that was needed to purchase the database from AHCA. With these results in hand, these data can now serve as the basis for submission of grant applications that can fund future efforts of this team.
Third, this research study demonstrates that thoughtful reflection on common clinical practice, aided by inquisitiveness and the right environment, can lead to important research that can change clinical practice. As those New York Times e-mails are circulated, and the findings are further disseminated in other forums, women in the future who have an imaging study suggestive of a suspicious lesion, as well as the clinicians who are providing advice on next steps, will become familiar with the findings of Drs. Gutwein, Grobmyer and their UF colleagues. Across time, the percentage of breast biopsies that are obtained by MIBB will increase and the percentage that are obtained by open surgical excision will decline and patients will benefit.
Throughout the Health Science Center, clinicians in a variety of fields reflecting our six colleges deal with patient care issues on a daily basis about which they undoubtedly have questions. It’s one thing to be taught that something is done a certain way ... because that's the way it’s done. With experience, however, it is natural to ask why is it done that way. Think of each step in every clinical decision — each procedure, drug, imaging or laboratory test, or device — and ask: Is this step truly needed? Is it known to have true efficacy? Is it the least invasive and most cost-effective option? Will it be associated with a high false-positive rate that will lead to a large number of additional unnecessary procedures? What are its implications for patient satisfaction relative to whatever benefit it provides?
I have no doubt that every active clinician reading this, if he or she paused for a moment to reflect on these questions in relation to their clinical field, can identify a potential research project. The next step for studies that are not externally funded — the difficult step executed so well by Dr. Grobmyer and his colleagues — is to select the right clinical research question that has substantial significance and can be answered with a feasible study design in a relatively short period of time at reasonable cost. An additional criterion might be the generation of data that can serve as the basis for a grant application to support next steps in the research story.
Dr. Grobmyer and his team are already hard at work questioning another “standard” clinical procedure in the practice of surgical oncology — the use of punch biopsy to diagnose melanoma. For many generations, the teaching of surgeons, dermatologists and others involved in the diagnosis and management of melanoma was that the diagnosis “must” be made by punch biopsy. While this is a minor surgical procedure, it requires local anesthesia and a suture; therefore, it might not be used as often as it should by primary care providers and others who notice a lesion suspicious of melanoma. The only improvements in outcomes of patients with melanoma over the last several decades have resulted from early detection and resultant early treatment. Anything that would further promote early diagnosis, therefore, would be a major clinical advance. Dr. Grobmyer and his colleagues from UF (Steven Hochwald, M.D.; Rony Francois; Vladimir Vincek, M.D., Ph.D.; Christina Mitchell, M.D.; Ann Church, M.D.; and Edward Copeland, M.D.) and the Moffitt Cancer Center (Jonathan Zager, M.D., Suroosh Marzban, Kimberly Law, Ashley Davis, Jane Messina, M.D., and Vernon Sondak, M.D.) have performed a study of patients treated both at UF and Moffitt showing that an initial “shave” biopsy, which is much simpler for a variety of clinicians to perform than a punch biopsy, is reliable and accurate in the overwhelming majority of cases (97%). The results were presented at the Southern Surgical Association meeting in December and will be published in the May issue of the Journal of the American College of Surgeons.
These studies are excellent examples of translational research. Two types of translational research were initially defined by the Institute of Medicine’s Clinical Research Roundtable (Woolf, S.H. JAMA 2008;299: 211-213): T1 — “the transfer of new understandings of disease mechanisms gained in the laboratory into the development of new methods for diagnosis, therapy, and prevention and their first testing in humans;” and T2 — “the translation of results from clinical studies into everyday clinical practice and health decision making.” More recently, others have parsed out from T2 a “T3” type of research, which focuses on the translation of evidence-based guidelines into health practice through delivery, dissemination and diffusion research. Depending on your viewpoint, the UF breast biopsy study can be considered an example of T2 or T3 translational research. Either way, the most compelling element of this story is the inestimable value of astute clinicians and colleagues working together to improve the practice of medicine and the health of our patients.
David S. Guzick, M.D., Ph.D.
Senior Vice President, Health Affairs
President, UF&Shands Health System