Chapter 10: At the NIH: Vincristine, a Cure for Childhood Leukemia, and VAMP
In this chapter, Dr. Freireich talks about the development of a drug called vincristine and the work for which he is famous: curing childhood leukemia.
The University of Texas MD Anderson Cancer Center
The Researcher; Overview; Definitions, Explanations, Translations; Discovery, Creativity and Innovation; Discovery and Success; On Research and Researchers; The Professional at Work; Understanding Cancer, the History of Science, Cancer Research; The History of Health Care, Patient Care; Technology and R&D; Patients, Treatment, Survivors
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 3.0 License.
History of Science, Technology, and Medicine | Oncology | Oral History
Emil J Freireich, MD
That was the end of the work on the white cell machine, so we did the platelets. We did combination clinical trials. Now came the big breakthrough, the thing for which I get all the prizes, and that is curing childhood leukemia.
We'd worked out the quantitative stuff. We'd worked out the combination chemotherapy of 2 drugs. One day Irv Johnson from Eli Lilly, who has been my lifetime friend and ended up becoming a vice president, appeared in my office. He said, "We have a drug that I have discovered that we want you to test in children with leukemia."
The drug was a drug called vincristine. Irv Johnson was a little bit like me. As I said, there are people who do regular science. They work and work, they get grants, they get papers, they get famous, and they get to be professors, but they never benefit anybody. Then there are people who are just a pain in the butt who discover things and move the field forward. Irv Johnson was a mover. He was working at Eli Lilly. The vinca alkaloids were taken up by Eli Lilly because they were the major manufacturer of insulin, and insulin was their big product. There was some folklore that the Vinca rosea, which is the periwinkle plant, was used by natives to treat people who had sugar in their urine. They diagnosed the sugar in the urine because the flies like the urine from the people who had this disease; they got very sick. They found out that if you gave them periwinkle they got better. So they decided to pursue that lead. They were trying to treat diabetes with these extracts. It turned out the periwinkle plant manufactures a thousand very complex alkaloids.
Lesley Brunet, MA
Is this the same vinca that we plant in our yards?
Emil J Freireich, MD
Yes, Vinca rosea. That's periwinkle; it's the same plant. It makes a thousand alkaloids. We don't know the evolutionary significance of those alkaloids, but that's true of all the natural products. Virtually all chemicals in antibiotics have come from natural products. There's some reason in nature why they're there, but we don't know why. Why does the mold make penicillin?
They were working on Vinca rosea, and Irv Johnson noted that some of his mice had very low white counts. He got the brilliant idea that this drug might be good for leukemia, so he began to treat leukemia in mice. Sure enough, they found that the crude extract did depress the white counts. Then they transplanted leukemia in identical mice, and they found that it worked. But there was a mystery. It didn't work in the leukemia that we used as our prototype model for developing drugs for man, leukemia L1210. It didn't work at all in L1210. Irv Johnson decided that one leukemia was not enough to screen the compound, so he had developed a number of mouse leukemias, and he had a panel. It turned out the mouse leukemia, which we use all the time today, called P388, was extremely sensitive to the vinca alkaloids. Then they purified these things. There were tons and tons of alkaloids. They got down to one called vincristine, which was present 1 milligram per carload of periwinkle. This was an extremely low concentration but very specific for leukemia. They decided to take it to the clinic. They did a little clinical trial at their small hospital, but they needed a big experience, so they came to NCI. I looked at the data, and I said, "Great idea. Let's meet with Dr. Zubrod."
Lesley Brunet, MA
What year was this?
Emil J Freireich, MD
It would have been about 1960 or '61. So we went to Zubrod, and Zubrod said, "Well, Dr. Freireich, we've spent a million dollars to show that L1210 leukemia predicts. Every drug that works in man works in L1210 leukemia, so this drug is a total bust." I said, "But, Dr. Zubrod, how are we ever going to discover a new drug if we stick to the paradigm that we believe that all the drugs we know came out of L1210? Maybe there are drugs that we don't know that will not work in L1210, and here's one." I said, "Besides, I've got children on the ward right now that are dying, who have no hope for living. What harm is there in doing it?" Zubrod said, "Freireich, on your personal recognizance, I'll let you study this drug." So we went ahead. The first 11 children we treated, 7 of them achieved a complete remission with a single drug. We had kids in total coma who woke up and dying children in remission. It was unbelievably dramatic. We published this paper very quickly. Myron Karon, who was then a pediatrician working on leukemia service, was the senior author. We published this paper, "Vincristine Works in Leukemia." So that was something. I sat down one day, and I said to myself, "Look, we've showed that 6-mercaptopurine and methotrexate both are myelosuppressive. If we give them together, it's better than giving them in sequence." That's interesting.
I have to tell you about one other protocol. I'm missing a step here in logic, and I'll come back to it: Protocol 3. We know that if we add prednisone to this combination, we get the best results of all, because prednisone is not myelosuppressive, and if we add prednisone to 6-MP, you get the full effect of prednisone in 6-MP. It's additive in terms of percent response. We add prednisone to methotrexate; it's additive. Now we add vincristine. Vincristine has no myelosuppression. The toxicity was neurotoxicity. So we could give a full therapeutic dose with no myelosuppression and no steroid effect. So instead of screwing around one at a time, what if we went for the home run? What if we give 6-MP and methotrexate and guess that if we add prednisone we get the full effect and guess that if we add vincristine we get the full effect? We're going to get 4 drugs at a time. Everybody thought I was insane. We had just fought the battle of 2. Now we were going to 4.
Lesley Brunet, MA
How would you be able to tell the difference between the prednisone and the vincristine?
Emil J Freireich, MD
They have totally different dose-limiting side effects. They have different biologies. Prednisone affects the blood glucose. Vincristine neurotoxicity and prednisone had totally different mechanisms. We knew that this arrests cells in metaphase. Prednisone causes lysis in T cells. We presumed that they would be additive, not synergistic.
We didn't know what to do with this, so I wrote a protocol. We called it VAMP. This is the most famous acronym in the history of chemotherapy. It was the first time anyone had used these acronyms. Now they're universal. All chemotherapy is identified by initials. VAMP was vincristine, amethopterin, mercaptopurine, and prednisone. And not only did we have the idea of combining all 4 drugs at one time, but we had a number of really important new ideas.
First of all, when I proposed this, Dr. Zubrod was very upset. He said, "This is ridiculous. We know how to do the randomized trials. You have to do this first and this first. Then you have to do this." I said, "Dr. Zubrod, I've got children dying on the ward. We just got this vincristine. We've got to move ahead aggressively." "Well, everybody worries about us experimenting on children." "But these children are hopeless." So we went through this with the IRB, Dr. Zubrod, and Dr. Mider. They finally allowed me to treat brand-new, untreated children with this regimen. This is the most famous chemotherapy in the history of chemotherapy. We treated 17 consecutive children, and 16 of those 17 children responded immediately, in 2 or 3 weeks.
I said to myself, "Wow! This is something." The purpose of combination is to avoid resistance. Could we possibly, with this enormously potent combination, cure leukemia? No one had ever proposed that any systemic cancer in man could be cured. There was only one systemic cancer that had ever been cured, and that was choriocarcinoma, which is an allograft. It was cured by M.C. Li. It's a disease of the placenta, not of the host, so immune and transplant. Leukemia is a childhood disease; it's a cancer, widespread and metastatic. Everybody thought I was insane. "You can't cure cancer. The only thing is radiation and surgery."
It was a long battle. We had meetings with the parents. I had to get the parents to agree to try this. "Look, I can give you chemotherapy, and they're going to relapse. Then I'll do it again, and they're going to relapse. What if we go all out? It's like a transplant. What if we take a chance? We might kill them with this combination, but we're going to try to cure them." So these 17 children, instead of just putting them in remission, while they were in remission we treated them aggressively. And then we stopped. You can imagine the tension on the ward. Every day, all the parents were asking, "How is Joe?" "How is Sam?" "How is Fred?" We had these 17 children and all their parents. I used to meet with the parents every Tuesday and Thursday to be sure they didn't die of anxiety. By the time I left NCI, we published a paper and clearly made the claim that we had cured leukemia. It was based on what's called the Kaplan-Meier plot, which makes a projection based on the available evidence. Of course, that prediction turned out to be true in 2001. This was 1964.
Now, I told you about Protocol 1 and Protocol 2 in the cooperative group. We started the first cooperative group. We did a protocol that I senior-authored. It was my most famous publication. Protocol 3 is another citation classic. In Protocol 3, the idea was based on this: We said, "Look, we have 3 drugs that work—6MP, methotrexate, and prednisone. We've learned that if we give them together, it's more effective than giving them in sequence. But we don't know what to do for the children in remission, because once the blood and bone marrow is normal and there's no leukemia, if we don't do anything, it comes back. If we treat them continuously, we prolong remission; but it always comes back."
So we got the idea, which is now universally used in chemotherapy, of minimal residual disease. We think they have residual leukemia, and it was based on the fact that the leukemia that came back was identical to the leukemia they had before, so even though we couldn't measure it, we did some fancy calculations and we wrote a paper on it. Based on the rate of regression, we did an estimate. We counted the number of cells per gram of tissue, we estimated how many you kill with each treatment, and we did a number of calculations. We figured out that if this kills x number of cells, and we did it 4 times, and it was just as effective each time, we'd get to zero.
So we said, "The thing to do is, we have to treat these children when they're in remission." So we designed an experiment; it was the most beautiful experiment. It was a prospective, randomized, placebo-controlled trial. We treated all children in the cooperative group with prednisone. That gives you about a 60 percent remission rate. These children are now free of disease. Once they were free of disease, we randomized them to 2 treatments. Treatment number 1 was to give them 6-mercaptopurine when there was no disease. The other was to receive a placebo. The difference in these 2 things was dramatic. The reason this study was so important and is a citation classic is it was the first adjuvant therapy. It was the first time we treated people who had no disease on the assumption that they had disease and proved it, based on the duration of disease-free period.
This was the first adjuvant chemotherapy study in the world. The median duration of remission with 6-MP maintenance was about 10 months or so. By that time, 95 percent of the children on placebo had relapsed. Now, the children who received placebo and relapsed were subsequently treated with 6-MP for induction so that the overall survival was not terribly worse; it was almost as good. It was a little bit better to give the 6-MP during remission.
While this study was going on, vincristine was brewing, so now you can see the logic. We went from combinations of 2 drugs. We proved that treatment when the patients were in remission could prolong remission. We now had a powerful 4-drug combination that induced remission, and we said, "Wow! What if we gave that as an adjuvant? Could we cure children?" And we did. So that's the other important step.
Freireich, Emil J. M.D. and Brunet, Lesley W., "Chapter 10: At the NIH: Vincristine, a Cure for Childhood Leukemia, and VAMP" (2001). Interview Chapters. 158.