Maurie Markman, M.D.
Former Vice President
Clinical Research
The University of Texas, MD Anderson Cancer Center
Hello, I am Dr. Maurie Markman, former Vice President for Clinical Research at University of Texas MD Anderson Cancer Center and I am going to be speaking on the topic of clinical trials.
The objectives of this lecture are to talk about the various endpoints that are used potentially in cancer clinical trials, distinguish between the various phases of the trials, and very importantly discuss some of the important ethical principles that are specific to the conduct of clinical research.
The first question is: What is a clinical trial? Although different definitions can be used, it is a methodical investigation of the outcome of an intervention used to treat a disease or clinical condition in patients. Although it clearly does include the treatment of a condition, for example cancer, which is what we are going to be talking about largely, but it is to look at this in a systematic way to answer a question that would potentially help future patients or members of society in the future. The conduct of clinical trials requires rigorous methodology and design; I am going to talk about some of that in a few moments, as well as expertise to monitor treatment outcomes. Very importantly, and we will discuss this in some detail, very important ethical principles must be strictly observed in the conduct of clinical trials.
The next question is, why bother? Why conduct clinical trials? There are number of reasons: It is critically important to look for effective therapies in diseases where no effective therapy exists currently. To improve the effectiveness of available therapies. To reduce the side effects or toxicities of treatment while maintaining the effectiveness of the treatment. Very importantly, when we already have an established treatment for a particular disease, it seems promising to see if we can expand the application of that particular treatment strategy into other disease categories.
What about endpoints in clinical trials? What are the goals? What are you trying to accomplish? It is important to state that when you design a clinical trial, the endpoints must be prospectively defined within the protocol. What that means is you state at the beginning what you will consider to be a positive outcome and what you would consider to be a negative outcome because what you do not want to have happen is you have done all this work and then someone says well, you know, let's call it positive, because we saw this. No, no that is not the way to do research. You say: we think, for example, in this particular trial, that if 25% of the patients have responded that we would consider that to be a positive outcome and of course, if you achieve that, that is great. If you do not achieve that, the outcome would be negative. You do not define positive trial after the trial is done. A variety of endpoints can be used, as I mentioned, response to treatment that could be, for example, the cancer has gone away completely. You cannot measure it, maybe it has shrunk by 50%, or maybe we are not actually looking at an actual size of a tumor that is either felt on physical exam or by radiographic imaging, maybe we are using a biomarker, maybe we are using a blood test, and that is a perfectly reasonable endpoint in certain situations. Other endpoints might include the duration of response, does it last two months, six months, three years, the time the disease is stable before it actually progresses, side effects of the treatment are important endpoints. The overall quality of life associated with patients treated in a trial. Increasingly, we are looking at issues of the cost effectiveness of the treatment. Not only is it positive from the point of view what it does to the condition, but what about the cost and certainly the cost relative to other treatments that one might have for a condition. Finally, certainly in the case of cancer, overall survival is a critically important endpoint.
A variety of measures for effectiveness and toxicity have been defined. Tumor Response. We classically use what are known as RECIST criteria. These are criteria agreed upon by the International Cancer Community led by the National Cancer Institute, very specific criteria of what is a response. Improved time to progression or overall survival. This is particularly when you are doing a randomized trial you are comparing treatment A to treatment B or treatment A to B to C. For toxicity, we have a variety of agreed-upon elements that specifically define what is a serious adverse event versus a mild adverse event or moderate adverse event that has been agreed upon. This particular criteria that is used is a National Cancer Institute Common Terminology Criteria for Adverse Events.
There are a variety of ways of measuring endpoints. One can look at physical exam and various imaging studies. Increasingly, that includes functional imaging, commonly the PET scan, which can look at not only the size of the tumor potentially, but also whether there is evidence of active dividing cancer, for example. A variety of laboratory tests, and here again I will include biomarkers, toxicity questionnaires, quality of life scales and, again, in the sort of discussion of cost effectiveness, we are beginning to increasingly look at financial analysis and life projection analysis.
So in designing a trial, there are a variety of issues that one must consider even before you put pen to paper and write the study. What is the hypothesis that you are trying to test? What is the scientific or clinical background supporting that hypothesis? You want to have a detailed treatment outline. How you are going to treat patients? How often do they have to come in? When do you repeat tests? What are those tests? What if toxicity is observed, the blood count goes too low, the patient develops mouth sores? How long do you wait before for you treat them again? How do you modify the dose? Maybe if you have several drugs, which doses do you modify? All of that has to be considered before the trial can be submitted for review. Then very importantly, as I mentioned, what are the endpoints in the trial? You want to describe the population that you are going to treat in the trial very carefully and of course you want to have a detailed statistical design that allows you to test the hypothesis and potentially answer the question with valid endpoints.
Turning to the different types of trials, these are trials that are obviously very relevant for cancer, but they are also relevant for other conditions. Phase 1 studies, or sometimes people use the term early stage trials; their primary focus is on safety, defining the safety of a particular regimen, combination chemotherapy, new biological agents with standard therapies. What you are looking at here is, is it safe? What are the doses to be given? What are the schedules to be used? If you look at the pharmacokinetics, you hopefully will get some preliminary evidence for the biological activity of a drug or combination. Hopefully you will actually receive real genuine benefit, but the primary focus is again: define safety.
When you get to phase 2 trials, you have hopefully defined the appropriate doses and schedules to move forward, so everybody could be treated in the same way, and in the phase 2 setting you are looking at the evidence of biological activity of a particular strategy, generally, certainly in anticancer therapies, we are looking at a response rate, the percentage of patients whose tumor shrinks to certain level that will define biological activity, and we, of course, in the phase 2 setting, are looking at this in a very specific setting, particularly tumor type, meaning that, we would not just treat patients with breast cancer, we would treat patients with breast cancer that have received one prior regimen and there is a tumor that we measured on radiographic imaging, for example. So the patients have very similar characteristics, so that when you, at the end of the day, have response rate, you could say this is the response rate in this specific setting.
If phase 2 data are promising, what we will then do to really find evidence of benefit of a new treatment is to do a randomized comparison of some experimental strategy to a standard treatment option. This is known as the randomized phase 3 trial. There can be variable endpoints. They can include overall survival, progression-free survival, response rate, reduced toxicity, improved quality of life, but again it is a comparison one particular treatment to a standard option to see if the experimental option whatever it may be is superior or may be equally effective, but, for example, less toxicity is observed. Finally, after a particular regimen has been shown to be effective, and often the parlance here is the drug is approved by the FDA, but the question is: what is the toxicity observed and the efficacy observed when you are outside of the very well defined patient populations of certain characteristics? So what happens when you get it into the community and you are treating patients who might be older or who have comorbid medical conditions? They might have heart disease or kidney disease in addition to their cancer, so what happens in the community setting? Such trials are often referred to as phase 4 studies.
In the context of clinical trials, there are number of critically important elements. Of course there are the patients, clinical investigators, research nurses, data managers, pharmacists, that is certainly the case in drug studies. We have to figure out how to process the data and store the data. We clearly have to have biostatistical support to allow this trial to be properly run and properly analyzed and of course administrative support as well as clinical and laboratory support facilities for the conduct of a trial.
Turning to the critically important issues of the ethics of clinical research, these are briefly outlined on this slide, and I cannot emphasize enough the importance of the conduct of ethical research. In the case of clinical research, while all of the items listed on the slide are important, the single most important issue is that of the ethical principle of autonomy, and that means very simply the patients must freely, and I have to emphasize the word freely or voluntarily, elect to participate in the trial. Coercion is not acceptable in anyway. The patients must be given information that allows them to make a free and voluntary decision to participate in the trial. Clinical research is not the same thing as clinical care. Clinical care is provided to patients with serious conditions, of course, informed consent is required there as well, but a patient has to be told that clinical research is voluntary. Although the purpose of that research may of course be to help the patient, there is also information being generated and gathered in the conduct of that trial and the patient needs to be informed and they need to be informed of the issues that are additional, in other words extra biopsies that are obtained and extra tests that are done. They need to be given all of this information. What are the known risks associated with the research or lack of knowledge of the particular side effects of an agent? That is what you are trying to learn. The patients need to be told all of this information and then must make an informed, but voluntary, choice. Beneficence is another important principle, of course it is important that physicians in particular, but of course all members of the healthcare team, understand that there needs to be some potential benefit to the patient, or at least there is no evidence that we know that they will be harmful, that is the non-maleficence. We must be concerned about the risk of harm and make sure that risk is acceptable and realistic relative to the potential benefit. The final principle is that of justice. The conduct of the trial must be fair and critically not exclude a patient population based on non-clinical considerations. For example, I think today one would say excluding individuals who are elderly simply because they are elderly is unacceptable or other again non-clinical considerations should not be used in making decisions whether individuals should or should not be given the opportunity to participate in the clinical trial.
One of the ways to ensure the ethical conduct of research is to have independent review of those research projects and that is generally done by what are known as institutional review boards, sometimes they are called institutional review committees. These groups are responsible for ensuring the ethical conduct of human subjects' research. One of the things they do of course is responsible for approving an informed consent document that clearly states in clear language understandable to the potential research subjects, the purpose of the study, how the study is being performed, potential risks and, critically important, available alternative options because if there are options, meaning there are standard drugs that might be considered, these options must be discussed with the patient. The informed consent document also needs to inform the patients of their rights and who they may contact for concerns related to the study or the treatment program.
When the trial is completed, there are ways to disseminate the information and this is a critical part of the trial and is also part of the ethics of the trial. You inform a patient they are going to be participating in research to generate knowledge that will help future patients. The way that knowledge is disseminated is obviously through a variety of strategies including peer-reviewed publications, abstracts that are presented at national/regional meetings, obviously the IRB needs to be informed of this. In the case of cancer trials, of course, often sponsorship for many of these trials comes from the National Cancer Institute, drug studies or device studies are often under the review of the FDA and, of course, there is a sponsoring company that will get this information and critically important that the results of trials be presented in public databases such as clinicaltrials.gov. That is the way the public will know that these trials have been conducted and the results of the trials will be freely available to the public.
Who is responsible for the conduct of the clinical trials? This can obviously include a single institution. It might even be a single investigator. It might be multiple institutions working together on a single trial. Then, of course, there are cooperative groups. In the cancer arena, there are groups that are funded by the National Cancer Institute and, of course, studiesmay be sponsored and run by pharmaceutical or biotech companies. These can even be multinational efforts today.
So, in conclusion, it is important to state that clinical trials are essential in identifying effective therapies and in expanding the application of treatment strategies for various cancers. It is critical that these studies have prospectively defined clinical endpoints, and these can include response, time of disease progression, survival, as well as quality of life factors. Finally, and critically important, patients must be free to elect to participate in clinical trials and to understand the potential risks of participation as well as alternative strategies. I thank you for your attention and we would certainly be very interested in knowing if this presentation has been of value to you, and we hope that it has increased your understanding of both the need for clinical trials and the ethical conduct of clinical trials. Thank you very much.
©2011 The University of Texas MD Anderson Cancer Center
1515 Holcombe Blvd, Houston, TX 77030
1-800-392-1611 (USA) / 1-713-792-6161