Feline Sarcoma and Vaccination
Roundtable on the Injection-Site Problem in Cats
March 1999 Issue of "Veterinary Forum"
Philip Bergman, D.V.M., M.S., Dipl. A.C.V.I.M., is in the Department of Cancer Biology, University of Texas, M.D. Anderson Cancer Center, Houston.
Mattie J. Hendrick, V.M.D., Dipl. A.C.V.P., is associate professor and head of the laboratory of Pathology and Toxicology, Department of Pathobiology, University of Pennsylvania School of Veterinary Medicine, Philadelphia.
Dennis Macy, D.V.M., M.S., Dipl. A.C.V.I.M., is President-Elect of the American College of Veterinary Internal Medicine Specialty of Oncology and heads the Preventative Medicine Clinic at Colorado State University, Fort Collins.
Lawrence D. McGill, D.V.M., Ph.D., Dipl. A.C.V.P., is an officer and pathologist of Animal Reference Pathology in Salt Lake City and past president of the Utah Veterinary Medical Association.
Robin M. Starr, D.V.M., M.Ed., has owned a small-animal practice in Colorado Springs and is the chair of the Vaccine-Associated Feline Sarcoma Task Force for the American Animal Hospital Association Foundation.
Kent R. Van Kampen, D.V.M., Ph.D., Dipl. A.C.V.P. (Moderator), is CEO of The Van Kampen Group, Ogden, Utah, and Chairman of the Board of Delft Pharma International.
This roundtable was made possible by an educational grant from Merial.
In the archives of veterinary pathology, feline sarcoma was relatively uncommon until the late 1980s and early 1990s. A malignancy of the connective tissue of cats, these tumors have been associated with an inflammatory reaction to vaccinations. Our panel members are involved in determining the various factors associated with the manifestations of this disease --its pathogenesis, epidemiology, prognosis, and prevention.
Appearance of the Condition in Veterinary Medicine
Kent R. Van Kampen, D.V.M. (Moderator) : When did you first become aware of feline sarcoma?
Mattie J. Hendrick, V.M.D.: In the late 1980s, I noticed in our biopsy service an increase in a unique inflammatory reaction, seemingly associated with vaccinations in dogs and cats. About a year later, I observed that sarcomas were arising at the biopsy sites in cats--but not in dogs. The tumors appeared to be encircled by that same inflammatory response. The two phenomena seemed connected, so I investigated them, epidemiologically at first. We had not seen many vaccine reactions. They had been described in humans and other animals, but our biopsy records showed perhaps one every several months. Now they seemed to be popping up once a week. As I had seen few reactions before, I could not say they were different now, but they were more obvious and stereotypical: They all looked much the same.
Lawrence D. McGill, D.V.M.: Over the years, we had seen some injection-site reactions, but we were not sure what they were. In appearance, they were little granulomas, with many lymphocytes and, often, eosinophils. When Dr. Hendrick's first article appeared (JAVMA. 1991;198:304-305), we began to see the macrophages full of blue-gray material inside or around the reaction. When Dr. Hendrick's letter to the editor was printed (JAVMA. 1991;199:968), we began looking closer and found somewhat lobular and irregular sarcomas, almost invariably with a central area of necrosis and surrounded by an inflammatory process. Currently, we may or may not see the blue-gray macrophages; sometimes they are just foamy, perhaps with some yellowish-brown material--or we cannot find them at all. You may find different types of sarcomas: They may not be fibrosarcomas but osteosarcomas. At the American College of Veterinary Pathology meeting in November 1998, some pathologists noted seeing lymphosarcomas at the same site. We have seen several hemangiosarcomas. D. Glen Esplin, D.V.M., Ph.D., Salt Lake City, described a liposarcoma that metastasized right away, probably before we saw the first biopsy.
Van Kampen: In an attempt to protect them from disease, we have been vaccinating cats for the past 50 years. Why was this disease not seen more frequently?
Hendrick: A number of people have asserted that what happened differently in the mid-1980s was that a new set of vaccines--killed or inactivated vaccines--made their appearance, which seemed to correspond or correlate with the sudden increase in incidence of these tumors. Although it cannot be proven, there is a lot of circumstantial evidence that something changed in the vaccines to increase the natural propensity of felines toward carcinogenesis.
Dennis Macy, D.V.M.: Before 1985, we used modified-live, high-egg-passage rabies vaccines that were not adjuvanted and did not produce local injection-site reactions. In 1985, we saw the introduction of a subcutaneous (SC) adjuvant product for cats and dogs. Because the SC route is less painful, the new vaccine became widely accepted, and ultimately, the modified-live rabies vaccines were no longer available. That same year, a feline leukemia virus (FeLV) vaccination was introduced --again, an inactivated adjuvant vaccine product--and widely accepted by veterinarians in the United States. That product constituted a large part of the market share for the next five years. Thus, for the first time, a material that resulted in inflammation in the SC tissues was widely administered to the cat population.
Van Kampen: We saw an increase in the number and frequency of feline vaccinations due to the availability of new adjuvanted products and new antigens.
The Nature of Injection-Site Sarcomas
Van Kampen: You have described a rather definitive inflammatory reaction at the injection site, associated with the vaccination of cats. When a sarcoma develops at this site, is it consistently the same type and is it always malignant?
McGill: The type of tumor varies, but most are malignant. Often, the same tumor will contain more than one type of sarcoma -- osteosarcoma or fibrosarcoma. You might even find a place with two or three different types of sarcomas. Myxosarcomas are quite common. It is natural that these undifferentiated mesenchymal cells grow into various types of tumors. However, I have not yet seen a mast cell tumor in these sarcomas.
Hendrick: I have not seen any of the round cell tumors, including lymphosarcoma.
McGill: I call the condition "injection-site sarcoma." We cannot yet specifically relate it to vaccines because we are seeing it with other products. We have even seen a tumor that looks identical to surrounding suture material. This year, we have counted 300-320 of those tumors in our laboratory, about twice as many as the first year but typical of the last several years. They are of all categories, and most have been injection-site sarcomas.
Van Kampen: What might the cat's owner observe at an injection site?
Macy: The vast majority of vaccine reactions following the use of these products are dismissed by owners as a mere lump, usually < 1 cm in diameter, if they are noticed at all. Until recently, most vaccines were given in the interscapular space, so sarcomas were somewhat hidden between the shoulder blades, and most owners did not notice them unless the site became abscessed or bothered the cat. If the owner did not notice them, the cat was not presented, and so a vast majority have gone undetected.
Feline Reactivity
Van Kampen: Studies show that implanting particulate matter in the connective tissue of rats over a period of time can evoke a fibrosarcomatous reaction. Do we have evidence that the connective tissue of cats is more reactive than that of other domestic species?
Macy: In terms of reactivity, no. In fact, we have compared the reactivity of rats versus that of cats to the feline vaccines, and they react in a similar manner.
Hendrick: In terms of fibroblast reactivity, I would say that it is high in cats, as it is in rats, while the other mammals that we deal with--dogs, horses, cows, pigs--tend to have less labile fibroblasts. So a cat's granulation tissue response, as compared with that of other domestic animals, is more profound.
Van Kampen: So the connective tissue may hyperreact to the inflammation, and that hyperreaction is associated with tumorous development where injections are made into the SC tissue.
Prevalence
Van Kampen: Have you seen a leveling off of this disease in the last three to four years?
McGill: The first couple of years, the number of cases grew, but it appears to have leveled off-- or fewer veterinarians are reporting it.
Hendrick: The number of cases certainly has not gone down. About three years ago, we hit a peak, when we were seeing sarcomas all the time because we were hypersensitive to the issue, then they leveled off.
Macy: Could this pattern be due to the fact that cats that entered the vaccine pipeline in 1985, when the causative agents were introduced, have reached their life expectancy, so that we saturated that population and would not expect prevalence to increase?
McGill: Or we have hit the genetic population--the ones that are susceptible.
Hendrick: But we always have new feline patients entering the field. and according to many studies, there is not a long latent period; it can be as short as half a year--or even less.
Macy: We have observed that cats that come from the same litter--even if they grew up in different households--have developed sarcomas. So there appears to be a genetic link--or at least an inherited link in terms of susceptibility.
Van Kampen: What is the prevalence of these reactions now? I have seen a variety of figures in the literature--one in 10,000, and one in 20,000 was seen initially. And now I see figures such as one in 1000 or 2000.
McGill: Dr. Hendrick has mentioned finding five injection-site sarcomas in 2000 cats--a higher rate than some people are finding, which is why we're doing epidemiology studies.
Van Kampen: Is there a cumulative effect of multiple vaccinations ?
Robin M. Starr, D.V.M.: The first study by Philip Kass, D.V.M. (JAVMA. 1993;203:396-405) showed that risk increases with the number of vaccinations administered simultaneously in one site.
Van Kampen: What can we tell clients about the incidence and prevalence of injection-site sarcoma?
Philip Bergman, D.V.M.: Although it is a rare or uncommon phenomenon, its occurrence is still unacceptably high. And to cat owners, it does not matter if the number is one in five million if it is their cat.
Starr: Are the numbers in Dr. Kass' 1993 study still the most reliable? About one sarcoma per 10,000--12 per 100,000 for FeLV vaccines and 10.5 per 100,000 for rabies?
Hendrick: Dr. Kass' study is the most statistically significant and valid. I personally believe that prevalence varies from region to region across the country. A study currently under way will address that problem because this study is multicentered and will be from all parts of the United States and Canada.
The Profession's Response
Van Kampen: Small-animal and especially feline practitioners have indicated concern about this increasing prevalence over the last several years. What has the profession done to address this problem?
Starr: In 1996, a meeting sponsored by the American Animal Hospital Association was held in conjunction with the American College of Veterinary Internal Medicine meeting. There was strong consensus that there was a crucial need for research in this area. The Vaccine-Associated Feline Sarcoma Task Force (VAFSTF) was organized to consolidate resources and fund studies in epidemiology, etiology, and treatment. By May 1998, we had funded six excellent, independently reviewed studies. We also established education of the profession and the public as an objective. We believe we can take some of the credit for the profession's not getting a black eye in the media because we are able to say, "Yes, the profession is working on this problem."
Etiology
Van Kampen: Those of us in the field of oncology know that cancer can be caused by viruses or genetic factors, which include oncogenes. Cancers can also be caused by carcinogenic chemicals or physical agents. We know that mutations occur in the cellular DNA, and these can be associated with cancer. What have you found so far about the etiology of this particular cancer?
Bergman: In the vast majority of cases, we have ruled out any involvement of retroviruses in injection-site sarcomas. Practitioners should know that if they see one of these lesions, they should test the animal for FeLV, as well as feline immunodeficiency virus, because a number of oncologic properties associated with these viruses could suggest the possibility that the tumor is not associated with the vaccination.
Van Kampen: What might the role of oncogenes be?
Hendrick: In some preliminary work with immunohistochemistry, I compared vaccine-associated sarcomas with fibrosarcomas that were not associated with vaccines and found that c-jun, a gene that has been shown to be oncogenic in various species, was overexpressed in tumors associated with vaccination and not in other tumors.
Van Kampen: Do growth factors stimulate the development of these tumors?
Hendrick: It has been shown in a few models that growth factors in general can cause overexpression of some oncogenes and vice versa because some oncogenes code for either growth factors or their receptors.
Van Kampen: Is there hard evidence that chemical or physical agents are involved?
Bergman: There is anecdotal evidence that incriminates vaccine adjuvants. Whether this disease is strictly adjuvant-related is controversial. I believe adjuvant is probably involved, based on the evidence that there seems to be a temporal relationship between vaccination and the induction of inflammation. Unfortunately, we have no direct proof; the literature has shown that we are not strictly seeing sarcoma in adjuvanted vaccines. But according to the work of Dr. Macy and myself, inflammation does appear to be an important property for production of a lump. Is inflammation then necessarily antecedent to development of a tumor? We do not know.
Van Kampen: Is there a single causative adjuvant, or are there multiple adjuvants in the vaccines that we use in veterinary medicine?
Bergman: It is fair to say that there are a number of such adjuvants. Dr. Macy coined a term, "secret sauces," mainly because they are essentially proprietary in nature. Dr. Hendrick has investigated the aluminum adjuvant. Many adjuvants are aluminum-based, but a number of other kinds of adjuvant are available, making this task all the more difficult.
Macy: By calling them "injection-site sarcomas" rather than "vaccine-associated sarcomas," we have indicated that anything that produces inflammation can potentially result in the cell cascade discussed by Dr. Bergman and, ultimately, sarcoma development.
Hendrick: Based on several studies, I would say that, almost every vaccine has been associated with a sarcoma, although statistically, the strongest causal relationships were with FeLV and rabies. I know of no study that, says which part of a vaccine causes sarcomas. Some studies have shown which parts of the vaccines cause inflammation--and if we think inflammation is important in sarcoma pathogenesis, then knowing those parts is important.
Van Kampen: Do we have evidence that certain antigens or other carriers might incite the inflammatory response as well as adjuvants?
McGill: Some of those products only have antigen. Early on, we noted that rabies and FeLV are not very antigenic, so they require certain modification to stimulate a strong response, or the vaccines require a large amount of antigen to get them to work.
Van Kampen: What is the transition time from inflammation to malignancy?
Macy: There have been reports of sarcomas developing less than three months after vaccination. I believe that such tumors were already present microscopically as a result of a previous vaccination and perhaps were "up" regulated as a result of the most recent vaccination.
McGill: This is a controversial area because many of us feel the period to tumor development is longer. I caution veterinarians about cases like the one in which the practitioner observed a lump for 12 day after the vaccination, then took it out. That animal is still not immunized. It was too early to take out that lump. We need to wait at least a month to six weeks before taking this step.
Hendrick: It could be that the sarcomas that show up three months after one vaccination are from the vaccine administration the year before.
Macy: From a biologic standpoint, we need a minimum of 25-30 cell-doubling times to obtain something palatable. But doubling times of feline sarcoma virus and Fibrosarcomas have been measured at as few as 12 hours. So biologically, a short period is possible.
Hendrick: Our working hypothesis of the pathogenesis is that something in vaccines causes a persistent inflammatory response, which in the setting of a cat's normal genetic propensities or milieu goes awry. Cells first become dysplastic and probably then neoplastic, resulting in a sarcoma at the site.
Diagnosis
Starr: I wanted to ask the pathologists about concordance if the same tissue is sent to a half-dozen different pathologists. Would the same diagnosis come back? Barbara Powers, D.V.M., Ph.D., Fort Collins, Colorado, stated at the Veterinary Cancer Society (VCS) meeting that nearly all samples show an inflammatory reaction. What is your experience ?
Hendrick: Even though there are many varieties of sarcomas, when you look closely at the cell type, the overall appearance of the tumor is stereotypic. It is well-demarcated but not encapsulated, and little fingerlike projections of tissue emanate from it. It has what appears to be a fibrous capsule around some of it. And almost invariably, it is surrounded by some degree of inflammation. Usually by the time we see them, lymphocytes are forming follicles. One article (J Comp Path. 114;1996:165-175) looked at that feature and the histologic characteristics of these tumors, as well as how they differed from fibrosarcomas in nonvaccine sites and found they had a very stereotypic appearance. The author could define them: The mitotic index was higher, as was the number of lymphocytes.
Macy: Subcutaneous location and occurrence in younger cats are also criteria.
Van Kampen: In answer to Dr. Starr's question then, pathologists are rather consistent in their diagnosis of these types of tumors.
Hendrick: If you sent one tumor to different pathologists, you might get different subcategorizations--whether it is a malignant fibrous histiocytoma, an anaplastic fibrosarcoma, or a fibrosarcoma with osteoblastic differentiation. But everyone would recognize it as vaccine-associated sarcoma.
Prognosis and Treatment
Van Kampen: Early on, when we first started to recognize these tumors, we would almost invariably state that they might recur at the site of surgical excision, but they seldom metastasized. Does this outlook still hold?
Bergman: It is generally true, but we are starting to see more papers describing how these sarcomas are metastasizing. The most recent study, done under the auspices of the VCS, showed that 25 or 26 percent of these tumors ended up having metastases. As clinicians have gotten better about adding on radiation and chemotherapy, we are now starting to see increase in survival with multi-modality therapy; unfortunately, we are still seeing a large percentage of cases recur.
McGill: The first time you do surgery on an injection-site sarcoma, you must be vicious because of all those little fingerlike projections. These tumors are not easy to carve out. If you do a "hemicatectomy" the first time, you can save some of these cats.
Bergman: A study recently reported that aggressive surgery either the first or second time is statistically associated with a longer disease-free interval as well as a longer survival time. Do an incisional biopsy, employing a true-cut biopsy instrument, and send a sample to the pathologist. Recall that an incisional biopsy means one by which you do not remove the entire tumor. Then you know what you are dealing with, and it can be appropriately treated.
Macy: Another recommendation is that a lump that has been present longer than three months after vaccination requires biopsy to determine how extensive the first surgery should be. My preference is to do an incisional rather than needle biopsy because even if we have a granuloma, the fibrocytes will look angry, and I would prefer to have a histologic evaluation.
Bergman: In another study investigating this type of tumor, we used nuclear scintigraphy with an agent that in human medicine specifically localizes to inflammation. We were able to show that we could find foci of inflammation, and we proved histologically that these ended up being tumor. Many times, these tumors were 4-5 cm or more away from the actual primary site of vaccination. You see these tentacles all the time histologically, and we can thus surmise that this tumor is moving across the site of least resistance, probably along fascial planes, and setting up another focus or nidus of tumor. This evidence suggests that shelling out the tumor at the time of surgery is very much the wrong approach.
Starr: One of the VAFSTF studies is working with a radiosensitizer to increase the effects of radiation on these tumors. Another study is using Doxil, a stealth liposomal doxorubicin, that cats tolerated better than Adriamycin (doxorubicin) in preliminary studies. Doxil allowed a 40 percent increase in dosage and achieved regression of tumors in 50 percent of patients--a small-number pilot study, but promising. So until we can prevent sarcoma formation, we may be able to increase the odds for a patient's survival.
Recommendations About Vaccination
Van Kampen: We have had many public pronouncements and some feline-association proclamations about how we should vaccinate our cats. What are your thoughts on frequency of feline vaccinations and why?
Macy: I defer to the Feline Practitioner Association': recommendations for the initial series of vaccines in kittens and booster in a year. After that period, we vaccinate every three years. This idea is based in part on work by Fred Scott, D.V.M., Ithaca, New York, who has looked at antibody titers in animals that were vaccinated for a variety of feline diseases and found that protective antibody titers exist far beyond three years. So that time frame would be a safe recommendation for cats.
Starr: For tracking and treatment purposes, the VAFSTF has endorsed different sites for different injections. The task force has not evaluated vaccination intervals. I personally believe that a few years ago, manufacturers would have been heavily criticized if they had made duration-of-immunity claims not backed by challenge.
McGill: There is some question whether antibody titers mean immunity.
Van Kampen: There are at least two facets of immunity: cell-mediated and antibody-mediated. Antibody protection alone might be sufficiently effective. But within the United States Department of Agriculture guidelines and the regulations for licensing, most vaccines require proof of immunization either by challenge, as in the case of rabies vaccines, or by persistent antibody titers over a period of time. It is based on those kinds of studies that duration of immunity is actually made part of the label, and frequency of vaccination is recommended based on duration of immunity. So whenever those recommendations are included on the label, studies have demonstrated that that agent is effective in the majority of the kinds of animals to be vaccinated. When we make recommendations beyond that label, we should have similar data to support them so that as a profession we do not open ourselves up to liability if, in fact, an animal contracts an infectious disease for which we failed to recommend vaccination. If this sarcoma complex is associated with injection-site reactions, should we discontinue vaccination of cats? Does tht risk outweigh the reward or vice versa?
Macy: I do not believe we should stop vaccinating, but if accept that at least some inflammation is a necessary antecedent to sarcoma development, we should strive to develop vaccines that do not produce inflammation.
Van Kampen: What about location of injection? Most vaccines today are administered SC, whereas we gave many of the older, modified-live viral vaccines intramuscularly (IM).
Macy: We have examined a few products to determine whether the inflammation is different with the SC or IM route and found no difference. An IM site tends to hide the inflammation, and unfortunately, if the tumor develops at an IM site, the veterinarian will not be able to detect it until too late. So if we administer a vaccine IM, we do not detect the cancer until late in its biologic history, and we have less chance to cure it. For this reason, I opt for the SC administration
Bergman: An important resource for finding these recommendations is available at www. avma.org/vafstf.
Macy: We have to be cautious about what we put underneath the skin of cats--whether vaccines or any other substance that has potential to produce chronic inflammation. While these sarcomas are uncommon, if we are to prevent them, we just have to say no to putting things under the skin more often than is necessary for the patient.
McGill: If you use repositol materials, such as vaccines, make sure you separate them by time and space if possible.
Starr: Vaccinations should be removed from the status of a routine automatic procedure and returned to that of a medical procedure. We must look at the history and the lifestyle of the patient and make a decision with the risk factor of sarcoma in mind.
Bergman: When you believe you have an injection-site sarcoma case, inform the vaccine's manufacturer because these sarcomas are grossly underreported to manufacturers. Second, report it to the United States Pharmacopoeia, (800) 4USP-PRN. Providing this information will help the profession solve the problem.
Toward Prevention
Van Kampen: What can vaccine manufacturers do to aid in the prevention of this disease? There is, of course, the concern for research money. But from a manufacturer's standpoint, anything that can be done to reduce reactivity at the injection site is a positive step.
McGill: If all the manufacturers got rid of as much of the reaction as possible, I am now of the opinion that we still would not get rid of the tumor. We would probably decrease it to the level that we are seeing in the dog, which is minimal.
Starr: Reducing the local reaction rates of existing products seems to be an important step. However, for a manufacturer starting a new research program, we still have a black hole in data from three weeks' postvaccination to the time of tumor formation. I do not think we know enough about the etiology to say, for example, that if you use one of the new designer adjuvants or a novel delivery system, you will solve the sarcoma problem. We need to find out more about the causes before manufacturers have a clear direction. To help find these causes, VAFSTF decided that for 1998 studies, we would consider joint ventures between the vaccine manufacturers and an independent researcher. We will only fund nonprofit organizations. But perhaps a company, with guarantees of confidentiality, can reveal proprietary information to researchers for evaluation. Manufacturers have information that we--the profession and the university researchers-- do not. We need to think of ways that information can be shared without putting proprietary formulas at risk.
Van Kampen: We have an old saying in pathology-- that the study of things caused must precede the study of the cause of things. We seem to be locked into that situation now with injection-site sarcomas. We are making some progress, but we are still not to the point where we can say we are certain of the cause of this disease.
To return to Sylvia's Cyber Kitty Condo just scratch her banner below....