Page 1 2 3 4 5 6 7 8 9 10 11 12 13
Q: Do you have any--to what would you attribute this bifurcation that might have taken place between disciplines?
Tullis: I think it was partly that the oncologists were a brand new specialty, they wanted to run with it themselves--as the new child on the block with the new bicycle wants to drive it himself--and they wanted to be their own boss and have their own institution, and made demands of independence and so on that were not consonant with what the American Society felt was appropriate. I think it's as simple as that. I don't think there was any one person that did this, but I think with a little bit more input to a cosmopolitan, ecumenical approach back ten or twelve years ago in the American Society, we might have been able to accommodate a temporary big growth of oncology and then have it come back into hematology as the new observations in oncology ran their course. There are going to be just so many drugs one can use to treat so many tumors, and then you've done it all, and that day, within ten years, I think will occur.
Q: When did this bifurcation happen?
Tullis: There was no single event, or no single date. It just gradually evolved as oncology blossomed, and oncology blossomed, I would say, around 1970.
Q: And was there a new development along with the introduction of new drugs and observations in leukemias, for example that contributed to this?
Tullis: Yes, but I think it was mostly just the vast numbers of persons that suddenly were involved in oncology. It became a very attractive subspecialty, and it's really a driving force in most medical schools not academically, but, economically, for support of hematology. So hematology, really, if you were to go out and say you trained as a hematologist today, at the university of wherever, and set up the practice of hematology, you would have a hard time supporting yourself, because hematology is primarily the basic science--this is what I've been saying over and over now, for several hours--and the clinical aspects of it are nominal. So you have a patient with pernicious anemia. You prescribe B-12, and he's cured for the rest of his life, you see. If you had a patient with a coagulation disease, you identify what it is, you either put him on anticoagulants, if it's too much coagulation, or you put him on support with fractions[?] if there's something missing, and he pretty well manages by himself, with the exception of things like hemophilia--even there, home treatment is now the treatment. So the practitioner in the specialty of hematology doesn't have enough to support himself as an individual. It's the oncology division of departments of hematology and oncology in hospitals, it's the oncology that, because of the huge number of patients who receive chemotherapeutic management for advanced malignant states, support this large group of people that do it. Hematology alone would have a hard time supporting--without grant support for basic science, you see, in hematology--it would have a hard time supporting itself in the university.
Q: Is this in some way, or many ways, similar to the initial developments within hematology, vis-à-vis Cohn’s lab and the amounts of money that were able to flow into work?
Tullis: It might be, yes. Except, nowadays, you see, good science still can get American money from things like the National Institute of Health and some of the private foundations. The good science is still going to be supported; that's the basic science in hematology. It has nothing to do with the man who's, as we say, a practitioner of hematology. On the other hand, in oncology, the practitioner of oncology is in control of vast amounts of money, quite independent of basic research money going into mechanisms of carcinogenesis and mechanisms for the treatment of cancer.
Q: Does self-interest come up in fields such as oncology among practitioners who do control these sums of money, as far as the basic research goes itself?
Tullis: In oncology, you mean.
Q: Yes.
Tullis: I don't know.
Q: I was wondering if that might contribute further to this.
Tullis: I don't know, but it's been a problem, and many institutions have expressed worries about its long term significance.
Q: Would you have any further comments to make about your personal role in ASH?
Tullis: No, I think not. I've enjoyed it. I have no role in it whatsoever now. There's nothing that in a young country like America is more evident than how quickly people pass over the horizon. One of the humorous stories of my life, as I've told many students, having founded the Society, having been the interim president and then the first regular president for a period of a year an a half, I then went to the next meeting after the St. Louis one where Dr. Moore took over. The next meeting, a year later, was in Montreal, and I got there to register, and they said, "What was the name?" And I said, "Tullis." And they said, "Are you a regular member of the Society?" And they couldn't find my name listed on the rolls. My secretary had left it off. You know, like old soldiers, people fade and die very fast in young countries. So, other than having served my standard time that we put into the constitution of five years on the Advisory Committee, and having served now--I shouldn't say that--but all I did afterwards, when they rewrote the constitution the first time, not the second time, I was on the Constitution Committee, and a few things of that sort--but I've had very little official interaction with the American Society.
Q: Just to maintain some perspective at this point--
Tullis: I've gone to all the meetings, but that's all.
Q: To maintain some perspective at this time, is there any way of weighting the comparable, or comparing the influences of, for example, funding in agencies, professional societies, individuals as far as defining a new discipline, such as hematology?
Tullis: Ask the question once more.
Q: We spent a fair amount of time talking about both the International Society of Hematology, the American Society of Hematology, and earlier on we spent some time talking about the funding that went in, for example, to Dr. Cohn's lab--
Tullis: The funding of research.
Q: Research. In Dr. Cohn's lab. I was wondering if there are some conclusions that can be drawn from comparable influences of these different institutions in the development of a research program.
Tullis: At the present time, the research money--now, this is not official; this is my view of it. As I view the research money that's passed out by the federal agencies, like the National Institute of Health, and to a lesser extent the Environmental Bureau in the Defense Department, and so on, it has two purposes: first, it identifies quality research, because of its peer review mechanism, which is the best in the world. But then, equally importantly, they try to evaluate the individual who is the P.I., the principal investigator: is he full-time, or is he part-time? If he's full-time, is he young enough that this is part of his career development? And this is the way it should be. I mean, they don't want to give any money to me--why should they? Because my career doesn't need anything. I'm at the end of the career, rather than at the beginning of the career, so that I should have to get my money elsewhere, which I do, for my research, through private foundations and things of this sort, grateful patients, and things of that nature. But the increasing difficulty of getting federal funds is in part robbing the future to support the present. This is a real apprehension to me. It's an apprehension of mine, because I see youngsters, who, as I've watched them come through medical school--for years, I've been a professor here at Harvard, and had to teach these youngsters, in medical school--then have them again as house officers in my program at the hospital, and watch them make their career decisions; ones that twenty-five years ago would go into basic science for a period of ten years, or maybe for life, see the uncertainty of funding that good career investigators constantly face every time their grant comes up for renewal. Even ones doing superb research, every five or ten years are right at the edge of not knowing whether their whole career is going to stop for lack of funding. And this creates a fear that has driven or frightened some of the bright young people out. Now, we haven't felt this yet, because there are enough people still willing to take that risk that America's staying au courant with international knowledge, but I'm worried for fear that ten years, twenty years from now, we will see less volume of good basic knowledge constantly advancing the field of hematology--I shouldn't use the word, hematology--just the field of medical science, because it is not satisfactorily funded as a career for the youngster.
Q: Has this had any effect on the actual content of the research? Have certain areas, due to lack of funding--
Tullis: Yes, in a way, because certain diseases become popular, you see. When it's popular, whatever it is---whether it's cardiac research, or transplantation biology, or what--it's easier to get a grant for a period of time. but that's not major influence, because you'll always find some people that have their own idea of their own subspecialty, their own field, who will plow ahead whether they're adequately funded or not. But I think it inhibits the volume, rather than, so far, the quality.
[tape interruption]
Q: Dr. Tullis, I would like now to discuss the development of your own research program in the context of developments within hematology. Perhaps we could start the discussion by picking up on a point that was raised by L. B. Jacques, that there's been a shift in what he's termed, "the hemostasis paradigm," or conceptual shift, between the 1930s and present-day practices.
Tullis: My own current work in this field is right at the border between neoplasia and hemostasis. I'll get into that in a few minutes, because I've got some concepts of the manner in which platelets interact with tumor cells and lead to local hypercoagulability as part of the metastatic process. That's what I'm currently working on. But let's look at a background, first, of hemostasis and of bleeding in its broadest sense, which includes both coagulation and hemostasis, in order to understand some of the things I'll be referring to in the next few minutes.
Historically, the first person to, in modern times--by modern times, I mean something since the Greeks--to study the kinetics of blood clotting was a man in England, at the same period of time when Benjamin Franklin was over there to the Court of St. James, representing the colonies which were organizing into a confederacy before America was formed as a country. This young man, Mr. Hewson, was a surgeon--they called them, "Mr." in England--at St. Bartholomew's Hospital, and Mr. Hewson, in the afternoons, would pass the marketplace, which still exists, right outside St. Bartholomew's Hospital, in route back to his boarding house, because he was courting the daughter of the woman who owned the boarding house that Mr. Franklin was living in. Hewson saw them, one afternoon, slaughtering a sheep, and apparently when one slaughters a sheep, one cuts the throat, and the sheep immediately bleeds to death and goes into shock as the carotid artery pumps out. He noticed that the blood didn't seem to clot at fast as he thought blood usually clotted when he was dealing with it surgically in the operating theaters at St. Mark's Hospital. So, one day, he took five or six glass vials, and put them on the ground just at the moment when they severed the neck of one of these sheep, and he directed the out flowing blood into each of the five or six vials, and made the observation that the blood which was put in the last vial clotted before the blood which was put in the first vial, which meant that as the sheep was going into the agony of pre-terminal biochemical changes, its blood became more coagulable, hypercoagulable, rather than acoagulable, and he timed all this, and philosophized on what it might mean.
I mentioned the role of Mr. Franklin just as an aside, because shortly after Hewson made this observation--he was only twenty-three or twenty-four years of age, he infected himself accidentally while doing an autopsy and died of sepsis, and his young wife had apparently became enamored of Mr. Franklin, followed Mr. Franklin afterwards--Mr. Franklin went back to Philadelphia, where the American Constitution was written, and started the Hewson family, which is a famous medical family in America, with the child of the man who had died. But Franklin was very much her supporter, and confidant.
Hewson's original observation prompted people to think that there was a dynamic element to clotting, which they later referred to as enzymatic ways of changing blood from a liquid to a solid when anything took place that deranged the circulatory integrity. But it stayed pretty much at that level, until about the 1900s when, unfortunately--I say, unfortunately, from the standpoint of research, but fortunately from the standpoint of medical treatment--Lewisohn, at Mt. Sinai Hospital in New York, conceived of collecting blood into an anticoagulant solution, like citrate or oxylate, to block the action of calcium, which was shown to keep the blood from clotting. The moment they did that, they could collect blood for transfusion purposes without using direct transfusions, but they simultaneously were able to traumatize the blood so badly that there was nothing hemostatically left in it. The platelets were damaged, the coagulation proteins were activated and damaged, but one didn't see it because the blood didn't clot.
I mention this because, if you look carefully at the literature of the late nineteenth century, you'll see the first reports suggesting that platelets, these strange little plate-like objects which were deficient in the blood of people that tended to bleed, could be replaced by transfusion. But, of course, those were direct transfusions of freshly collected blood, not anticoagulated, and the only way they could do the direct transfusion was to use what was called a Kimpton tube. Empirically, they had observed that if they took a glass tube and coated it with wax that the blood could be collected, and particularly if they shielded it from losing the sealed tube, the blood could be collected and would remain liquid for a half-hour or more, rather than clotting like Hewson's sheep. And they reported that blood so collected would replace the missing platelets in people with platelet deficiency diseases, and stop the bleeding. Then, if you look at the literature twenty years later, you'll see very astute clinicians saying that these old observations of the people in the late nineteenth century were poor observations because blood transfusion would not replace the platelets. Well, now, the reason it wouldn't replace the platelets is there were no viable platelets left in the blood. The platelets in turn were essential to give something to the endothelium which is essential for the endothelium to remain impermeable to colloid solutions like plasma. In the platelet, that's one of the functions it fulfills. This substance was called "Substance P" in Switzerland after permeability, and then much work was done in an attempt to isolate what it was, but it was a part of the platelets. We then found another one of the functions of platelets, because of the increasing knowledge of adhesion and aggregation of platelets was to form a plug, which, if one damaged the vessel wall by either touching it with an electrode, or cutting it, or anything of that sort, the blood would begin to leak out of the capillary bed where it was damaged, but a platelet plug would form and, if you will, cement the area. Then a third function of the platelet, totally independent of that, we found was the supply of what had been called PF 3, platelet factor three, which was the precursor of thromboplastin, that was being generated in the plasma. So all of this was worked out and evolved in the last ten or twenty years as part of hemostasis. If you have qualitative abnormalities in your platelets where some of these functions are lost, or if you have quantitative deficiencies in your platelets, where there aren't enough of them, these three mechanisms are deranged--and I'm speaking of just the grossest aspects of it; there are some aspects that are too technical to go into.
Q: Is this--just to ask a question at this point--is this the reason that platelets--in the metaphor you used, platelet as sponge, for example--would this be in reference to that?
Tullis: I wouldn't use the term platelet as a sponge; I'd use the term, platelet as a truck. It's a system for delivering things which never was thought of before, and the platelets, tiny as they are, deliver and carry a lot of important monoamines, and things of this sort, to local areas in the body--serotonin is one example, catecholamine is another example, histamine is another example--and this is quite independent of their role in protecting the integrity of the vascular bed. But probably their primary role is protection of the vascular impermeability. If you look at hemostatis phylogenetically, if you go down the scale, you get back to the Limulus, which, as you know, has no clotting system whatsoever. All it has is giant platelets that plug up the holes when you damage the circulatory system. There are all gradations in between the Limulus and the human, where nature has supplied proteins like fibrinogen, to strengthen the hemostatic plug and maintain it.
Now, I mentioned that my personal work had to do with sort of an interface between cancer and clotting.
Q: When did this work date from?
Tullis: Well, it dates back to about 1970. I've only had preliminary publications in the field. I publish now only when I am absolutely certain. I've always thought I was certain, but the older you get, the more cautious you get. I'm just ready to publish now some manuscripts in this field. But the field dates back to the observations of a young surgeon by the name of Cliffton working at the Memorial Hospital in New York City. He observed that if he heparinized an animal before injecting tumor cells into the portal circulation, he cut down on the 'take' of the tumor--the number of positive 'takes' that he had. He knew from the work of Dr. Warren here at the Deaconess, who had been my chief at one time and for whom I have great respect, how many tumor cells it took to establish a metastasis with a transplantable animal tumor. Shields Warren had studied this because of his interest in tumor biology. The young surgeon found that if he heparinized an animal, as I say, and then repeated Warren's experiments, the animals did not get the tumor. So the anticoagulant appeared to be doing something to prevent the growth of local metastases.
Immediately, everybody said, “Well, heparin has lots of other chemical effects, and these are probably what kept the tumor from growing. It's not the anticoagulant effect, but rather that heparin blocks a translation of what we call messenger RNA, for example, and that would interfere with protein synthesis in a broad basic way. But the Russians, then, did some work where they found out if they defibrinated animals by using snake venom, that the defibrinated animal, which doesn't clot, also resists transferable animal tumors the same way. This was followed, a few years later, maybe in the middle 1960s, by an epidemiologist in England by the name of Allen, who found that people who had been put on coumadin anticoagulation for heart attacks, when matched with controls of people with similar heart attacks and similar age, sex, and background, who had not been given anticoagulants for their heart attacks, that the incidence of death from cancer was sixfold less in those that were on coumadin than it was in those who had not received anticoagulants. So this made it look like something of an undescribed nature was influencing cancer metastases by cutting down on either the rate of growth of the tumor or its ability to set up housekeeping in other parts of the body of an anticoagulated individual. I should say that the work thus far--this other work, to which I'm referring in the literature--has not led to a lot of concrete data vis-à-vis therapeutic effectiveness of such an approach, and I'm not discussing it from that standpoint, but merely from the historical aspects of the development of the science and then finally my own personal involvement in it. But there was a lot of work instituted in different countries simultaneously because of this. I think the best thing that can be said is that it's now been proven quite definitely that, certain animal tumors, particularly in mice, rabbits and rats, are favorably suppressed by anticoagulant therapy by diverse means: either removing fibrinogen, or giving heparin, or suppressing prothrombin, or even suppressing platelet function by giving aspirin. But the work at the human level, from the therapeutic standpoint, really hasn't progressed very much, although a good multi-institutional study that's sponsored at Dartmouth University has recently published some data to suggest that with certain advanced malignancies, particularly carcinoma of the lung, if one anticoagulates the patient at the same time as starting the standard chemotherapy protocol, and that those patients who receive both the standard chemotherapy protocol, plus anticoagulants, live longer than the ones who get the same chemotherapy, without anticoagulants. But the beneficial effects are still relatively minor, and that's not the reason I wanted to mention this.
My own interest in this field began because of platelet work. Platelets as part of hemostasis, and, secondarily, as part of clotting, seem to play a role, and the role that we more or less stumbled on was the following. Because of this work that I've mentioned from the other countries, I thought that the logical way to approach it would be to take tumor cells, fresh from the operating room--fresh specimens--and bring them here into the laboratory, analyze the cells, and see what coagulant or anticoagulant substances I could recover physiologically from the cells. What we did was take growths like carcinoma of the colon, where one can take normal cells identified right adjacent to the tumor from the same specimen, so that one had normal and abnormal cells from a single individual, and we would bring these here, and wash the cells, make single-cell suspensions, then homogenize, and so on, and try to identify the clotting proteins. Well, to our surprise, what we didn't find is easier to discuss than what we did find. There was no prothrombin, there was no hemophilic factor, there was no fibrinogen nor accelerator globulin. There was none of the usual clotting proteins, with the one exception of thromboplastin, which was also being worked on by Dr. Gordon, a superb investigator at the University of Colorado, who was identifying the chemical nature of tumor-derived thromboplastin from tissue cultures of melanoma cells. But, in our studies here, we didn't do that.
However, there's a lot of empiricism in research, as you know, and there was an assay that we had developed in our laboratory, called the Platelet Anti-Thrombin Assay, which is still an experimental tool, because it hasn't yet been applied enough at the clinical level to know what its parameters are. But when Dr. Kioaki Watanabi was with me as a graduate student for three years, in the late 1970s and early 1980s, we asked him to set up the Platelet Antithrombin Assay, based on an observation of one of my younger Tullis associates, Dr. Francis Chao, who had identified platelet anti-thrombin and which we'd published data about. So, we had this Assay in the laboratory, and I said one day to the technician after we had found that cancer cells had no significant amount of coagulant protein, I said, "Well, for heaven's sakes, let's dump some of the tumor cells into a Platelet Antithrombin Assay and substitute the tumor cells for the platelets to see if they have any anticoagulant properties," and to our amazement we found that tumors from the colon--and we later found tumors from the pancreas, and certain other organisms--bind thrombin, and bind it very avidly. They have both high-affinity and low-affinity binding sites for thrombin, apparently on the tumor cell surfaces. Thrombin is the principal coagulant substance that's generated when blood clots, and the question, obviously, was, well, why do they bind thrombin, because if there's a receptor there, there has to be a reason for it, because obviously the internal aspects of the cell are affected by activation from the ligand.
Thrombin is a peculiar enzyme. In addition to its role in conversion of fibrinogen to fibrin, which is its principal role in the blood, it also is a very potent mitogenic substance. This has been shown in tissue cultures. It stimulates rates of growth, the cell division. It may even be mutagenic, although that has not been as well established as its mitogenicity. So, we philosophized that maybe certain cells of the body had binding sites to begin with and if you activated these often enough and already had a hypercoagulable state, such as, if you will, cigarette smoking, or estrogen use, or stress, or many of the other things that turn on clotting, like the sheep outside at St. Bartholomew's hospital, that this type of mechanism would make the temporary presence of thrombin cause the tumor cells to break down and liberate more thromboplastin, this would create for a microsecond longer the presence of additional, molecularly active, thrombin to bind to the receptor sites on the cells, and set up an increased rate of tumor cell division. Not that it started the tumor, simply that if a tumor was already there because of something else--genetics, radiation, you name it--that this would be a propagating force for a self-supporting, autocatalyzing process to increase rates of growth, and division, and dissemination. Now, having found that these tumor cells appeared to have thrombin binding sites, the next question was, well, where did the substances come from that did this, and what were the substances? So we looked at the platelets from the same individuals in the next series of studies and found out that the thrombin binding sites on the platelets, which is what we had set up the Platelet Antithrombin Assay to measure over the years, are decreased in these same patients where the tumor cells have more binding sites for thrombin. So, the question was, first of all, what was the substance, and, secondly, had a transfer taken place from one cell strain to another? With respect to what is the substance, our study is attempting to identify by gelelectrophoresis, and so on, the protein bands that are in the membrane of the platelets. The data thus far suggests that there is a deficiency of glycoprotein l-b from the surface of the platelets, and it's particularly the glycocalicine part of this Protein l-b that is lost from the platelet of these people with advanced cancer and is present on their tumor cells.
Now, the answer to the second question, how does it get from one cell to the other, I don't know. I have two theses, but we're just now starting some experiments on this. One of the theses is that it's a direct cellular interaction. Two researchers in England, Vale and Warren--that's a different Warren than Shields Warren--published some electron photomicrographs a few years ago that show quite clearly that platelets interact with the surface of tumor cells as part of the lodgment of a metastasis in the vascular bed. So one of the possibilities I'm entertaining is that the platelets actually transfer by direct physical adherence temporarily to the tumor cell, part of the membrane of the platelet to the tumor cell, instead of to the endothelium, and that this, then, is the site which will bind the thrombin. Another possibility is that it's indirect and mediated by activation of proteases within the platelet leaking out from the platelets having been activated through coagulation mechanisms, or particularly fibrinolytic mechanisms, having these proteases leak out from the platelets and from the tumor cells--because tumor cells are full of proteases--and that these, in turn, have split off sialyic acid, like Protein 1-b, splits of glycocalicine, and that this then is free in the plasma and later picked up on the tumor cell.
Now, I don't know which of these it is. We're just starting some in vitro experiments, and trying to do this thing. Come back in five years; I might have an answer for that.
Page 1 2 3 4 5 6 7 8 9 10 11 12 13
©2008 Columbia University
back to Dr. Tullis's profile
back to top
