Chapter 06: Investigating Remission Percentages, Expanding Research Interest in Leukemia, and Developing Databases and New Research into Blood Malignancies

Chapter 06: Investigating Remission Percentages, Expanding Research Interest in Leukemia, and Developing Databases and New Research into Blood Malignancies

Files

Loading...

Media is loading
 

Description

Dr. Keating tells the story of setting up the first large scale database at MD Anderson in the mid-seventies. He begins with his suspicion that remission results were inflated leading to a natural history study he conducted to gather actual percentages. He describes the database of nine thousand acute patients and databases compiled for chronic lymphocytic leukemia and all the acute leukemias. He describes the studies that came out of the database project.

He next explains why the medical community was resistant to believing the information they presented.

Dr. Keating then discusses Dr. Emil J Freireich’s view of randomized trials and his belief that MD Anderson should develop new concepts and leave it to other institutions to confirm these ideas. Dr. Keating describes Drs. Frei and Freireich and goes on to talk about the latter’s innovative platelet transfusions to treat childhood leukemia. He explains how Dr. Freireich came to be accused of fraud and the outcome.

Identifier

KeatingM_01_20140513_C06

Publication Date

5-13-2014

City

Houston, Texas

Topics Covered

The Interview Subject's Story - The Researcher; MD Anderson Snapshot; Professional Practice; Contributions to MD Anderson; MD Anderson Impact; MD Anderson History; Beyond the Institution; Understanding the Institution; Portraits; Controversies; Ethics; Discovery and Success; Patients; Patients, Treatment, Survivors

Creative Commons License

Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 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:

So what was the research project that you embarked on in your second year?

Michael Keating, MD:

Well, in my second year, because I was already a faculty associate, I didn’t have a specific research project, but I will tell you that there were reports that we were getting a 70 percent complete remission rate in acute leukemia. I had been on the Inpatient Service for six months and I didn’t see 70 percent, and I had decided that I couldn’t challenge these heroes of leukemia research who were running these protocols and say, “Your protocol has got inflated results,” so I decided to do a natural history study. So I went and had a look at 325 patients that had received leukemia treatment over a four-year time period, and that was my first big database that was set up here. So that started my interest in cataloging what treatment people got and their outcome, so that at the end of my time on the acute leukemia part of it, we had built up a database so that every acute and chronic leukemia patient that had come to the institution from 1965 onwards we had catalogued, and we still have, I think, the only truly complete database in the institution—

Tacey Ann Rosolowski, PhD:

Oh, wow.

Michael Keating, MD:

—so that we now have in the chronic lymphocytic leukemia, for example, we have a nine-thousand-patient database which we update every six months so that we now can look at trends in different comparison to one treatment versus another, and men versus women and particular genetics sometimes, etc.—

Tacey Ann Rosolowski, PhD:

Wow.

Michael Keating, MD:

—just by sitting down and getting spreadsheets together. And it’s a unique database in the world. We’ve done the same with chronic myeloma leukemia and all of the acute leukemias, so we have a good way of recording all these things. And this is, I think, part of what they want to do with “big data,” to be able to get the information on everyone that comes in. The weakness of “big data” is that it’s okay, you collect all the data of things that happened at MD Anderson, but you don’t keep track of what happens when the patients go home. And we have so many patients that are referred from out of state or out of country, etc., that we have to develop ways of getting that cataloged, and that’s a challenge which is making the leaders of that group a little intimidated, because it’s hard.

Tacey Ann Rosolowski, PhD:

Now, you mentioned that you began this database because you wanted to check or challenge that 70 percent. What did you find?

Michael Keating, MD:

We found that it was about 54 percent, which was still the best in the world. And one of the other features was that we found that after doing the four years from 1973 onwards, that I decided to go backwards all the way to 1965, and then I found that there were a number of patients that were still alive for long periods of time. So in 1978 when I came back to Anderson, I used that information to look at the potential cure of acute leukemia in adults, and we had twenty patients that had been alive and in complete remission for more than five years, and that at that time there were only sixteen others in the world that had been reported.

Tacey Ann Rosolowski, PhD:

Wow.

Michael Keating, MD:

So there was a lot of disbelief that you could actually cure the disease, and it was just the tenacity and the understanding of pharmacology and attention to detail that made that possible, that had challenged the rest of the world to catch up.

Tacey Ann Rosolowski, PhD:

Would you like to talk more about the studies that came from that, or, I mean, because we can do it in your research [unclear].

Michael Keating, MD:

No, that’s fine. The studies that came from it was identification of what they call prognostic factors. So we found that older people didn’t do as well. If someone had a slow onset of a blood disorder before it became full-blown acute leukemia, that was a bad thing. There were the height of the white blood count and a few other parameters. Dr. Ed Gehan is the biostatistician that was mainly working with Dr. Freireich on the database and the clinical trials. He and a woman, Terry Smith, who was the program analyst that was working with him, the two of them taught me how to do multivariate analysis, and we had one of the few multivariate analyses that had been done at that time. And what we were able to do was to plug in the characteristics of patients and calculate the likelihood of them responding to conventional treatment, and it might go from 3 percent, if a patient had a lot of bad characteristics, up to 95 percent for the good characteristics, so that every patient, you had an expectation, and so that after twenty or twenty-five patients, you would be able to find out if a new treatment was matching the expectations, exceeding it, or failing to achieve what we wanted to achieve. There’s now a whole litany of approaches to prognostic and predictive factors. But one of the things was that Dr. Richard Simon at the National Cancer Institute wrote a summary of predictive models, and he said that the leukemia predictive model in acute leukemia was far and away the best that had ever been described because, (a), we’d proved that it applied prospectively, and we were allocating people to new research protocols if they were unlikely to respond to the standard, so that we set up that if someone had less than a 20 percent likelihood of responding, we started something new on those patients, and if it beat the expectations, we went and included the 20 to 40 percent and kept on matching up until it worked or didn’t work.

Tacey Ann Rosolowski, PhD:

It’s starting to sound like kind of the general contours of personalized care.

Michael Keating, MD:

It is, because we’re at that point beginning to realize that were some genetic subsets that were unique. In fact, in the twenty patients that were the five-year complete remission for five years plus, we found that six of them had some abnormal cells called eosinophils, and another six had a cytogenetic characteristic called –C+D+E-G, because this was before the chromosomes were given numbers one to twenty-two in the text chromosomes, because I was just grouping them into roughly the same size, the big ones, the intermediate, and the littles. And it ends up that both of those particular subsets that you can recognize just by genetics and looking under the microscope were the harbingers of what they call the Core Binding Factor patients, and they’re certainly the most curable and they’re the ones that are the most curable with a single drug called Cytosine, or Arabinoside or Ara-C. We were then able to define some others that died of bleeding episodes that were the promyelocytic leukemias, etc., so from this you’re able to compartmentalize them into what happened to them clinically and described why they died; if they relapsed, what happened to them; and how many of them had disease that spread to the brain and how many didn’t.

Tacey Ann Rosolowski, PhD:

Why was there resistance outside the institution or maybe even within the institution to this information and what it clearly suggested and enabled you to do with patients?

Michael Keating, MD:

I think that there was always a lot of jealousy of the MD Anderson for two reasons. One is that it went from a very primitive leukemia program in 1965 when Dr. Freireich came with Dr. Frei. I think J, or J Freireich, saw perhaps about single-digit numbers in his first year, and they were usually people that had been damaged by treatment by their own disease, etc., up until the point where they became the biggest leukemia program in the country. But in those days, Dr. Freireich was militantly against randomized trials, and he was actually antagonistic to a lot of people inside the institution and outside.

Tacey Ann Rosolowski, PhD:

What was his objection about the randomized trials?

Michael Keating, MD:

It takes too long to figure out whether things work or they don’t work, and this was why he liked the concept of the expectation for every patient, because if you look at the structure of a randomized trial and you have two treatments, the possibility is that it will be better or worse for each of the two drugs, which means that you have to have four degrees of possibilities, and it’s a lot easier if you just say we know the expectation; we just have to do one. The other thing that was annoying to the people outside was that J always thought that the Anderson should be developing new concepts which would be confirmed in the cooperative groups, but once we developed a new concept, we’d move on to the next new concept and let the other institutions in SWOG to see whether it worked or it didn’t work. We then went away from entering outpatients on the studies. So it was an elitist concept that “We are better than you,” and this antagonized a bunch of people, so we eventually got kicked out of the cooperative group and—

Tacey Ann Rosolowski, PhD:

And I’m not sure I know it. What is that cooperative group?

Michael Keating, MD:

The Southwest Oncology Group was set up by Frei and Freireich when they came down to Texas, because there was nothing like it down here. There was the Cancer and Leukemia Group B, which had been established up in the Northeast because Frei and Freireich were at the National Cancer Institute. Memorial Sloan-Kettering and Mount Sinai were in New York, so they had this cluster up there, but there’s nothing in the South.

Tacey Ann Rosolowski, PhD:

Right. I didn’t realize they set it up.

Michael Keating, MD:

Yeah. And so they were very imaginative people and risk takers. They would try things that no one else had ever tried before. And Freireich was at the National Cancer Institute. All these kids that he was treating—he was assigned to treat childhood leukemia—they were all dying of hemorrhage, and he thought that it was because there were not enough platelets. So he was working on collecting platelets, and you’d get whole blood and you’d separate out the platelets and you’d have to count them on little glass stage— [telephone interruption]

Michael Keating, MD:

And he was doing all this manually and spending a lot of time doing it, and he was saying that the platelet transfusions were preventing the children from having fatal hemorrhages. But he was accused of knowing which children were getting the transfusions because it wasn’t a double blind thing. So the Blood Bank up at the National Cancer Institute thought that it was a waste of blood from him collecting all this stuff, etc., and he was accused basically of fraud, so that he had to defend that. Dr. Gordon Zubrod was in charge of the National Cancer Institute chemotherapy branch at that time, so he held this investigation, and at the end of it, he said, “I think Dr. Freireich’s right, so keep sending him the platelets.” So nowadays if you were accused of fraud, there’d be congressional hearings, everything would grind to a halt, etc., there’d be terrible charges that were thrown at you, but Zubrod just had the personal principles that he said, “Okay, this guy’s trying to do something important, and time will tell whether it’s right or not, and if we just stop it now, we’ll never know the answer.” So people forget that platelet transfusions are a relatively new experience. It was, I mean, in the sixties that they were developed. One of the things that impressed me was that J said, “Well, my religious background is fluid,” and he said but the thing that impressed him was that every time he traveled with Zubrod, the first thing that he asked when he checked into the hotel was, “Where’s the nearest Catholic church?” and he went to Mass every morning. And I said, “Well, I’m not ever going to be that good, but it’s nice to know that someone can.” So it’s some sort of goal that you can put in place. So it was this sense of the balance between your career and your belief system and your family structure and all of those things. They’re still the most difficult balancing act I think we have, and the whole burnout situation, which is now being discussed very actively, is certainly part of that.

Conditions Governing Access

Open

Chapter 06: Investigating Remission Percentages, Expanding Research Interest in Leukemia, and Developing Databases and New Research into Blood Malignancies

Share

COinS