
Chapter 27: An Overview of Research Issues
Files
Loading...
Description
Dr. Hortobagyi explains that recruitment of research nurses and research managers was a key to strengthening the research mission of Breast Medical Oncology. He then moves to a related discussion of conflict of interest, noting monetary dimensions of conflict of interest are only “the tip of the iceberg.” He explains that a Principal Investigator has a vested interest in the success of a clinical trial. The research nurse thus serves as unbiased party to collect and manage data. He explains the decision not to permit principal investigators to look at data before all the results of a trial are in.
Dr. Hortobagyi recalls the controversies at MD Anderson regarding the running of clinical trials, which some researchers believed were unethical. The discussions revealed, however, how difficult it is for a researcher to be unbiased and that the process of generating data needed management to insure that results were unimpeachable.
Dr. Hortobagyi notes the reasons why scientific misconduct was not discussed in the 70s and 80s.
Identifier
HortobagyiGN_05_20130315_C27
Publication Date
3-15-2013
Publisher
The Making Cancer History® Voices Oral History Collection, The University of Texas MD Anderson Cancer Center
City
Houston, Texas
Interview Session
Gabriel Hortobagyi, MD, Oral History Interview, March 15, 2013
Topics Covered
The Interview Subject's Story - The ResearcherMD Anderson History MD Anderson Culture Building/Transforming the Institution Multi-disciplinary Approaches Growth and/or Change The MD Anderson Brand, Reputation Institutional Mission and Values Giving Recognition
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License.
Disciplines
History of Science, Technology, and Medicine | Oncology | Oral History
Transcript
Tacey Ann Rosolowski, PhD:
And what was the piece with research, too? Because you had trials at a certain point that were being run, were you able to—were those research results becoming part of that discussion as well?
Gabriel Hortobagyi, MD:
Yes. So part of these discussions was that sometimes you could get to some agreement about the specific area of controversy, and sometimes you couldn’t. And when we couldn’t, we took advantage of that, and instead of a head-on confrontation and screaming and yelling we would propose—all right, so we have two different opinions about this. Well, this is a perfect opportunity for a clinical trial, and lets have a clinical trial to resolve the issue, and let’s figure out who is right and who—and more importantly, what is of greater benefit to patients. And if it doesn’t make any difference, we should also find that because then we would have two equivalent approaches to a particular problem. With that in mind, then we started to develop clinical trials. Some of our earlier clinical trials were just to integrate chemotherapy into the overall management of advanced breast cancer and inflammatory breast cancer. Later we developed clinical trials to figure out how much benefit radiation therapy brought into the equation, how much benefit immunotherapy with BCG and Corynebacterium parvum, and other nonspecific immunotherapy agents would contribute. Then we started to look at different types of surgery, and eventually from those discussions—which were initially thought to solve internal problems—we eventually became some of the leaders of breast cancer management throughout the country. That was very gratifying. Some of those treatment approaches that were developed here took a decade, sometimes two decades to be adopted elsewhere throughout the country. But essentially everything that we had proposed and everything that we developed and published on has survived the test of time, which is, I think, encouraging. It just reflects well on the—at the risk of sounding self-serving, the really genuine intent of those in those conferences to try to do better and to try to figure out how to best treat patients.
Tacey Ann Rosolowski, PhD:
The development of clinical trials and protocols then also pointed out that in order to do this well, we needed to collect data well and accurately and dispassionately and without the vested interests. So that led us to the recruitment of research nurses and data managers. And in fact, it was our department that developed the concept of research nursing.
Tacey Ann Rosolowski, PhD:
How did that take place?
Gabriel Hortobagyi, MD:
Well, eventually we realized that we didn’t have the time to make sure that every patient on a clinical trial did what needed to be done according to the protocol. The principal investigator didn’t have the time to do all of that, especially in larger trials that had several hundred patients. We needed someone who could babysit each protocol. We couldn’t—we didn’t have the resources nor could we find physicians who would do that fulltime, but we could find trained registered nurses who were ready for the challenge. Eventually they developed the techniques, and they developed the job descriptions, and they helped us develop forms to collect the data, and we developed the standard operating procedures. And then we found funds to pay for them, which was an important consideration, and we haven’t stopped since then. At the very beginning, we had one research nurse for a couple of years, and now we have—I don’t know—fifteen or twenty. Of course throughout the institution, now there are hundreds of research nurses because it’s a critically important task.
Tacey Ann Rosolowski, PhD:
At that time forty years ago, there was not the concept of conflict of interest. Nobody talked about conflict of interest. While the public and politicians talk about conflict of interest in monetary terms, I think that that is just the tip of the iceberg because physicians make a decent living, so it’s hard to have a physician become biased by giving him $2000 honorarium or something like that. But a principal investigator has a vested interest in the success of his clinical trial and unconsciously can do a number of things that will compromise the factual outcome of a clinical trial.
Tacey Ann Rosolowski, PhD:
The second role of a nurse or a second person who doesn’t have that vested interest is to provide that sort of unbiased approach to data collection and data management. It also makes sure that it is not the principal investigator who actually collects data and perhaps would have the opportunity to alter the data. It is a totally unbiased person who just collects that data and protects that data. Because soon after we started to recruit research nurses, we also came to the agreement that before the trial is completed, the investigator shouldn’t actually look at the data except for what needed to be attended to on a day-to-day basis. I think it was an early development in clinical trials that was very important.
Tacey Ann Rosolowski, PhD:
And the reason for that was—was what—not to have a look before the trial is created—completed?
Gabriel Hortobagyi, MD:
In the 1970s there was a major discussion—a major controversy in oncology about how to do clinical trials. Some oncologists—many oncologists, I would say, especially many statisticians—considered that the only reliable way to get information was to do randomized blinded clinical trials because that would avoid the assignment of treatments based on physician bias or perhaps patient bias and preference and whatnot. And there was another side, and we were part of that within this institution, and several of the leaders of the institution at that time considered that doing a randomized trial was sort of unacceptable, and it was an unethical use of patients, and they could do that very well without the control group because they were unbiased, and they were pure anyway.
Tacey Ann Rosolowski, PhD:
During those discussions, many of which were very public discussions, it became apparent that it is actually difficult to be unbiased because we all are led to some extent by our belief systems and our cultural background, and that much of science is based on evidence, but much of science is not. Sometimes when you develop a hypothesis, you come to believe that hypothesis even before you have proven it. If your career depends on the success of your hypothesis, then it’s very easy to be misled and to fool oneself into accepting a conclusion before reaching that conclusion. Since I was much younger at that time and I still hadn’t completely accepted either of those positions, I started to see some of the weaknesses of each argument, especially the weakness of the absolute embracing of either side. But I did see—and so did some of my close colleagues—that there needed to be some protection of the process of generating the best evidence because we are all imperfect human beings, and with the best of intentions we can actually mislead ourselves into areas that we didn’t want to go.
Tacey Ann Rosolowski, PhD:
That led us into the process of trying to develop data management and data collection methods that wouldn’t be viewed as unimpeachable by outsiders. It didn’t happen overnight, of course. Some of the earliest research I did based on retrospective studies—with what I know today I would say, “My God, how did I do that? How come I didn’t know better?” But it was done with the best of intentions at that time. And I know that others within the institution who are passionate about their research also did it with the best of intentions. At that time, certainly scientific misconduct hadn’t been talked about or discovered. Today it is a fairly prominent aspect of what we talk about because there have been some very highly publicized cases of fraud—scientific fraud or misconduct. At that time, I don’t think it was much of a topic of conversation either because it hadn’t been discovered or it was uncommon or because there were actually far fewer scientists. There was less competition for resources, and perhaps people didn’t feel the need to misquote or develop erroneous things. So that’s how we developed our research group.
Recommended Citation
Hortobagyi, Gabriel N. MD and Rosolowski, Tacey A. PhD, "Chapter 27: An Overview of Research Issues" (2013). Interview Chapters. 1132.
https://openworks.mdanderson.org/mchv_interviewchapters/1132
Conditions Governing Access
Open
