Accurate cytologic diagnosis of vaccine-associated fibrosarcoma is best accomplished as a joint effort between the clinician and the cytologist. For the clinician, good communication includes relaying an accurate history that reveals mass location, size, physical features and duration as well as other relevant clinical findings. If a vaccine-induced sarcoma is suspected, vaccination history is critical information. For the cytologist, good communication means describing the cytologic findings as accurately as possible, clarifying whether the findings are diagnostic versus inconclusive, recommending further diagnostics when needed and suggesting relevant rule outs based on the clinical findings.

Cytologic findings do not differentiate between vaccine-induced fibrosarcoma and other soft tissue sarcomas in the cat. The goal is to obtain a cytologic diagnosis of sarcoma with a reasonable clinical suspicion for the underlying etiology based on clinical history and mass location. The primary use for cytology in this regard then is to distinguish between benign inflammatory injection reactions with associated fibroplasia and malignant mesenchymal neoplasia. Additionally, cytology is useful to monitor for local recurrence of a previously diagnosed fibrosarcoma and to check for distant metastases via lymph node aspiration and ultrasound guided fine needle aspiration of internal organs. Cytology continues to be a rapid, inexpensive and relatively noninvasive means to obtain either preliminary information or, frequently, the final diagnosis. Weaknesses of fine needle aspiration include poor cell exfoliation (especially a problem with mesenchymal tissue) as well as the inability to determine the specific origin of the sarcoma. Concerns over the dreaded "probable mesenchymal neoplasia, cannot fully rule out fibroplasia" can be partially alleviated with active communication between the cytologist and the clinician.

Cytologically, fibrosarcomas are of moderate to high cellularity with cells found both individually and in aggregates. Frequently there is abundant, extracellular, wispy, pink material consistent with extracellular matrix in conjunction with small, pink cytoplasmic granules presumed to be secretory granules. The cells are moderately to markedly pleomorphic, exhibiting both anisocytosis and anisokaryosis. Individual cells are spindyloid to round and plump. More anaplastic tumors frequently have rounder cells with fewer of the typical spindle-shaped cells present. The malignant cells have an increased N:C ratio with lightly basophilic, occasionally vacuolated cytoplasm and indistinct, wispy cytoplasmic borders. Nuclei are round to oval with occasional macronuclei and micronuclei. Chromatin is coarse and open, with single or multiple, prominent, irregular or bizarre nucleoli. An increased number of mitotic figures, especially aberrant mitoses, further support the diagnosis of sarcoma. The most distinguishing cytologic features are the high degree of cellularity, nuclear/nucleolar characteristics of malignancy, and lack of associated inflammation.

References

1. Schultze, A.E., Frank L.A. and K.A. Hahn. 1997. Repeated physical and cytologic characterizations of subcutaneous postvaccinal reactions in cats. American Journal of Veterinary Research 58(7): 719-724.

2. Hendrick, M.J., Goldschmidt, M.H., Shofer, F.S., Wang, Y. and A.P. Somlyo. 1992. Postvaccinal sarcomas in the cat: epidemiology and electron probe microanalytical identification of aluminum. Cancer Research 52: 5391-5394.

Irregular spindle-shaped or stellate fibroblastoid cells with large irregular nuclei containing much euchromatin are ultrastructural features of fibrosarcomas. There are often thin cytoplasmic extensions into the collagenous insterstitium. The spindle cells of fibrosarcomas are arranged in fascicles, and some cells show features of myofibroblasts. Myofibroblasts typically show bundles of actin microfilaments, intercellular attachments, and the occasional presence of an incomplete external lamina on the cell surface. Like fibroblasts, there are subplasmalemmal plaques, abundant, well-developed rough endoplasmic reticulum and irregularly distributed intermediate filaments. In contrast to fibroblasts, myofibroblasts contain long bundles of microfilaments with dense bodies running through the cytoplasm - the smooth muscle differentiation may show pinocytotic vesicles on the cell surface.

Unusual features observed in some feline vaccine-associated sarcomas included needle-like crystalline material, multinucleate giant cells, glycogen, intratumoral endocytosed erthrocytes and hemoglobin crystals, Infiltrating leukocytes included macrophages, neutrophils and lymphocytes. No viruses were seen.

Feline vaccine-associated sarcomas represent an ongoing problem. A significant component of the dilemma lies in the optimal application and sequencing of treatment to effect local tumor control. Treatment modalities reported to date used alone or in combination in the management of vaccine associated sarcomas include external beam radiation therapy (photons or electrons), surgery and chemotherapy (e.g., doxorubicin). Another treatment option available for the management of microscopic disease postoperatively is high dose rate (HDR) brachytherapy. Ongoing studies are being done to elucidate the optimal dose, course of therapy, and subset of patients likely to benefit from this form of therapy. Ultimately, determination of the efficacy of HDR brachytherapy will require a prospective randomized trial comparing external beam radiation therapy to HDR brachytherapy in the post-operative setting. The following summarizes the basic principles, theory and practice of high dose rate brachytherapy.

The term brachytherapy refers to the placement of radioactive sources into or near the tumor, and sources can be placed either temporarily or permanently. High dose rate brachytherapy utilizes a radiation source that accomplishes treatment in a very short time period and avoids the problems associated with leaving a radioactive source in a patient. This system also entails the use of a remote controlled afterloading system. Remote afterloaders are computer driven and insert and retract the radioactive source, and thereby eliminate the need for personnel to manually load radioactive source(s). Radiation exposure is to the patient alone and not to personnel involved in patient care and treatment as occurs with low dose rate (LDR) brachytherapy.

High dose rate brachytherapy is used in human cancer management in several different settings. HDR brachytherapy can be used to irradiate the tumor using a fractionated course of treatment, or can be used to deliver a boost dose of radiation after a full course of external beam radiation therapy. For a boost dose the implant can be localized to just those tissues that are suspected of harboring residual disease. HDR brachytherapy can also be used for patients that have previously failed combination surgery and external beam radiation therapy. Intra-operative implantation of the tumor bed during surgery is done in the treatment of humans with soft tissue sarcomas and allows more accurate placement of the implants. Studies suggest that the actual treatment be delayed until at least 6 days post-operatively to facilitate wound healing. As a single boost dose this could be done with HDR but there are obvious limitations for a fractionated course of HDR with the applicators removed after completion of the first treatment.

HDR sources were first used in brachytherapy in the 1960's but have only more recently been used in the treatment of veterinary patients. Reports of brachytherapy in the veterinary literature have primarily involved the use of low-dose-rate brachytherapy applications with either temporary (5-7 days) or permanent implantation of radioactive sources. High dose rate brachytherapy became possible with the development of a miniaturized high-activity radiation source in conjunction with a computerized remote controlled delivery system.

High dose rate brachytherapy offers a number of theoretical and practical advantages in the management of vaccine associated sarcomas. The advantages and disadvantages of HDR brachytherapy are outlined in Table 1. The use of brachytherapy to irradiate the tumor bed allows the delivery of a high dose of radiation to residual tumor while minimizing exposure to surrounding normal tissues. Problems encountered with large field irradiation using external beam teletherapy machines such as pneumonitis, and spinal cord damage may be avoided. However there are limitations to the implant size, and in some instances it is more appropriate to use external beam irradiation for large fields. Specifically this will typically entail the use of electrons, as opposed to photons, in the cat whereby the dose to the underlying tissues can be minimized.

The primary advantage of brachytherapy is the ability to precisely deliver a concentrated radiation dose to the tumor but it requires the ability to delimit the tumor tissue and the site must be accessible. Determination of the treatment field requires an accurate knowledge of the surgical field such that all areas of potential residual tumor will be included. It can be problematic to make this determination for patients that first present post-operatively. If all aspects of definitive therapy are carried out at the radiation treatment facility this will improve the likelihood of a positive outcome. The medical, surgical and radiation oncologists all should be involved in treatment planning. Delineation of the treatment field is typically based on a combination of the surgical scar, computed tomographic evaluation of the site pre- and/or post-operatively, or radiographic localization of hemoclips placed at the time of surgery.

The site is clipped and sterilely prepped as for surgery. The catheters (referred to as applicators) are inserted percutaneously through hollow 16 gauge needles. The applicators are placed typically 1 cm apart, in parallel, and cover the entire tumor bed. Typically, a single plane array of applicators is used to deliver the prescribed dose. The applicators are then attached to connectors that are in turn connected to the remote afterloading system, specifically the head that houses the iridium-192 source. A marker wire is used to define the source positions in the applicators and the information manually entered into the treatment-planning computer. A significant advantage of the HDR remote controlled afterloading system for brachytherapy lies in the adjustable stepping positions and dwell times that allows optimization of dose delivery and distribution with essentially an infinite number of variations possible. Confirmation of applicator position in the patient can be accomplished through orthogonal localization x-ray films with dummy or marker wires in place in the applicators. Dose is then prescribed typically either by prescribing dose relative to a specified distance from each source, or an isodose curve can be drawn around the tumor bed and the dose prescribed to this line. The computer then calculates the dwell time of the source at each point based on the dose prescribed and the current activity of the source. The dose rate for the high activity sources used in HDR is approximately 100 cGy/min. Iridium-192 has a half life of 74.2 days. As the source decays the treatment time increases, and the source is replaced approximately every three months. The activity of a single source used in HDR brachytherapy is on the order of 10 Ci at the time of installation, and requires a shielded room comparable to that required for Cobalt-60 teletherapy units. The outer dimensions of the iridium-192 source are 1 mm in diameter and 5 mm in length. The source is attached to a metal cable that moves the source from the shielded container to the patient through plastic connectors and then into the applicators. HDR machines have multiple channels and one up to 15 or more connectors l meter in length can be simultaneously attached to the HDR unit. Thus, multiple applicators can be placed and connected to the afterloader and the actual treatment is accomplished over a matter of minutes. When the treatment is completed the catheters are removed and the patient recovered and able to return home. The patient does not need to be hospitalized as is necessary for LDR brachytherapy or may be necessary for a fractionated course of external beam radiation therapy when an owner lives any significant distance from the treatment facility.

In using HDR versus LDR brachytherapy one problem that occurs is the decrease in tolerance of late-responding normal tissue due to the increased dose rate. Complications that can arise include normal tissue fibrosis and potential necrosis. This problem in part can be circumvented by delivering dose with HDR using multiple fractions. With LDR brachytherapy normal tissue repair occurs during treatment because of the low dose rate. With HDR brachytherapy, with a high dose rate, repair of normal tissues does not occur which further explains the rationale of fractionated therapy. The typical dose per fraction ranges from 5-7 Gy and four to five total applications are done at weekly intervals.

Metastasis occurs in a subset of cats but local tumor control continues to be the main problem encountered with vaccine associated sarcomas. External beam radiation therapy done either pre- or post-operatively has been reported to result in an appreciable number of local failures often occurring within the irradiated field. HDR brachytherapy in the post-operative setting alone or in combination with external beam radiation may effect improvement in local control rates. Studies will need to be done to elucidate the utility of HDR brachytherapy in the management of this disease.

Selected References

I. Dale RG, Jones B. The clinical radiobiology of brachytherapy. The British Journal of Radiology 1998;71:465-483.

2. Erikson B, Wilson JF. Clinical indications for brachytherapy. Journal of Surgical Oncology 1997;65:218-227.

3. Nag S, Orton C, Thomadsen B. Remote controlled high dose rate brachytherapy. Critical Reviews in Oncology/Hematology 1996;22:127-150.

4. Nickers P, Kunkler I, Scalliet P. Modem brachy-therapy: current state and future prospects. European Journal of Cancer 1997;33:1747-1751.

5. Walker M, Smith JR. Iridium-192: a literature review for further referencing the isotope, its activity units, and dosimetry techniques. Veterinary Radiology 1990;31:281-292.

Computed tomography (CT) and Magnetic Resonance (MR) imaging are an essential part of the diagnostic work-up for treatment planning in people with soft-tissue sarcomas.

In the past year, seven cats with vaccine associated sarcomas were presented to the Veterinary Medical Teaching Hospital, U. C. Davis, for CT and additional diagnostic work-up and treatment assessment. CT evaluations performed after the intravenous administration of iodinated contrast media consistently demonstrated that pre-imaging mass size assessments were underestimated. In addition, patterns of contrast enhancement, including tendril projections along fascial planes and muscle bellies, suggested neoplastic infiltration and/or extensive pericapsular inflammation not appreciated on pre-contrast sequences. The imaging characteristics of these neoplasms contributed in determining which treatment options were most appropriate in each case.

Based on our initial evaluations, a standardized imaging protocol for patients with vaccine associated sarcomas was proposed which addressed patient positioning, CT acquisition parameters and the use of peripheral radiopaque markers. It is recommended that patients with periscapular lesions be positioned in sternal recumbency. In all patients, pre and post-contrast 3 to 5mm transverse images should be obtained with a small field of view using a standard or soft-tissue algorithm. A standard iodinated contrast media labeled for intra-venous use with an equivalence of 400mg iodine per ml given at a dose of I ml per pound was noted to sufficiently enhance the lesions being evaluated. For patients having previously undergone excisional biopsy or attempted mass removal, the use of peripheral radiopaque markers to identify the borders of the resection site is recommended.

Transdermal drug delivery systems permit the slow and continuous absorption of a drug across an intact cutaneous surface, and should produce relatively constant plasma drug concentrations for a prolonged period of time. A transdermal delivery system for fentanyl (Duragesic, Janssen Pharmaceutica, Titusville,NJ) has been licensed for the management of chronic cancer pain in human beings. The rate of release of fentanyl is proportional to the surface area of the patch (25ug/h/10cm2) and the system is available in 4 sizes (25, 50, 75 and 100ug/h). Transdermal administration of fentanyl offers the opportunity to maintain effective analgesic plasma concentrations of the opioid for 72 hours. In human beings and dogs, administration of a transdermal fentanyl patch results in a gradual increase in plasma fentanyl concentrations during the initial 24 hours, followed by a relative steady state plasma concentration until patch removal. After removal of the patch, plasma fentanyl concentrations fall rapidly in dogs and more slowly in human beings. Transdermal fentanyl administration may also be useful in pain management in cats, particularly after major surgical procedures such as en bloc fibrosarcoma resections. Published pharmacokinetic data in cats (Scherk-Nixon JAAHA 1996;32:19-24) indicate that after administration of a 25 ug/h patch, steady state plasma fentanyl concentrations are obtained in 6 to 12 hours, with a mean plasma concentration at steady state of 0.37 ng/ml. However, this study can be criticized in terms of the study population, controls,use of the limit of detection of the assay and the statistical analysis. Re-evaluation of this data indicates that the mean plasma concentration at steady state (18 hours to patch removal) is closer to 0.28 ng/ml. The plasma concentration of fentanyl necessary to produce analgesia in cats is not known, but in humans, concentration ranges of 0.23-1.18 ng/ml and 1-2 ng/ml have been reported to be analgesic. Abstracts of two additional studies in cats have recently been reported. Lee, Hardie and Papich (ACVS 1998) administered a 25 ug/ h patch and reported a mean fentanyl plasma concentration at steady state of 1.88 ng/ml, with a delivery rate of 8.5u.g/h (or 34% of the theoretical drug delivery rate).Yackey,Ilkiw,Pascoeand Tripp reported (International Congress of Veterinary Anesthesia 1997) mean plasma fentanyl concentrations at steady state of 0.20 and0.38ng/ml for 25 and 50ug/h patches respectively. The reason for the marked difference in plasma fentanyl concentrations is unclear. The age and sex of the cats in the former study was not reported, and the site of patch application was different. Additional environmental factors, preparation of the patch application site and the significant individual variation in plasma fentanyl concentrations after transdermal patch application reported in all species may be involved. Clinical studies of the efficacy of transdermal fentanyl administration in cats are lacking. Yackey et al. reported that transdermally administered fentanyl significantly reduced the minimum alveolar concentration of isofluane in cats. Additional pharmacokinetic and clinical studies in cats are warranted. Currently, it is recommended to use a dose rate of 2-4u/kg/h and to place the patch l2 to 24 hours before surgery. Cats should be monitored for signs of insufficient analgesia (and treated for breakthrough pain) and signs of opioid overdosing, such as euphoria, sedation, inappetance, constipation and respiratory depression.

North Carolina Animal Cancer Treatment Program,

Purpose: Review our current treatment protocol for feline injection site fibrosarcoma (FSA) and present a preliminary outcome analysis of cats treated since our previous report (Cronin et al., 1996, Vet Radiol Ultrasound 39:51-56). In that work, we reported a median disease free interval for 33 cats receiving radiation therapy followed by surgical excision of 398 days. The local failure and metastasis rates were 58% and 24%, respectively. However, disease free interval was significantly longer in cats whose surgical margins were free of tumor cells.

Methods: The medical records of cats treated since November 1994 were reviewed. Our recommendation for cats without metastatic disease is radiation therapy (48 Gray, 16 daily fractions of 3 Gray = XRT), followed in 2-3 weeks by excision that has evolved over the years to include excision of vertebral spinous processes, scapulectomy or, hemipelvectomy if needed. If possible, adjunctive chemotherapy is administered. The owners of some cats declined some or all of these treatments; thus, some cats received palliative treatment only. Because these data are preliminary, we were unable to test the effect of (1) the tumor having been excised but recurrent at the time of presentation, (2) obtaining "clean" surgical margins, or (3) chemotherapy on local control or metastasis. We were also unable to conduct multivariate analysis.

Results & Discussion: Since November 1994, we have treated 168 cats with a histologic diagnosis of sarcoma. There were 9 oral cavity, 11 non-injection site and 148 injection site sarcoma. There was no difference in age between cats with injection and non-injection site tumors. The median age of cats with injection site FSA was 9 years (mean = 9.2 + 3.1). There was no detectable sex predilection. There were 7 cervical, 89 thorax/scapular, 36 lumbar/flank, and 16 proximal extremity injection site tumors. Forty-three cats were not treated. Ten cats had surgery only. Palliative radiation was administered to one cat. Full course XRT was administered to 93 cats and 80 had surgery after radiation. Chemotherapy, during + after radiation, was administered to 3 cats receiving full course XRT and 22 receiving XRT + surgery. Chemotherapy consisted of carboplatin, doxorubicin, and mitoxantrone in 25, 10, and 5 cats, respectively.

Treatment failure event, defined as local recurrence or metastasis, was observed in 33/148 cats; some cats had both local recurrence and metastasis. Metastasis was recorded in 13/148 cats; however, it is likely that metastases were under-reported in this preliminary analysis. Local recurrence was observed in 22/148 cats. There was no correlation between failure and tumor location. The median time to treatment failure event was 204 days. In cats receiving chemotherapy, the failure rate was 22% whereas in cats not receiving chemotherapy, it was 43% (Fisher's exact test was not significant). In this preliminary analysis, duration of local control data were not available in cats that had not failed; therefore, disease-free interval could not be determined.

Our data suggest injection site fibrosarcoma may be treated effectively using radiation + surgery. Thorough data analysis is required to determine whether rate of failure is decreased by chemotherapy.

A 17 year old, spayed female, DSH cat weighing 6 lb was brought to a veterinarian with a complaint of fleas. The cat was given a 40 mg subcutaneous injection of PROGRAM® 6 Month Injectable for Cats (lufenuron) in the interscapular area. One month later the owner noticed a lump at the injection site. Six months after the injection the cat was returned to the veterinarian. Examination revealed a 7 cm subcutaneous mass in the interscapular area. A biopsy was taken and submitted for histopatholoigcal examination. At surgery the mass was described as firm, lobulated and well encapsulated.

Histopathological examination revealed the mass to be composed of intersecting bundles and tangled sheets of spindle-shaped, neoplastic fibroblasts. At the peripheral margins of the mass were aggregates of small lymphocytes. The diagnosis was fibrosarcoma.

The owner declined surgical excision or other treatment of the tumor. The mass continued to enlarge and the cat was euthanized 3.5 months later. A necropsy was performed. No gross nor microscopic evidence of metastasis of the tumor to internal organs was found.

In a single case it is difficult, if not impossible, to prove a cause and effect relationship between an injection and subsequent tumor development. The cat had not been vaccinated at this site in the year prior to the lufenuron injection but the prior vaccination history was unknown. Since vaccine-associated sarcomas are reported to occur up to 3.5 years after vaccination, influence from prior vaccinations in the area cannot be ruled out.

However, a lump in the area was first noticed by the owner immediately (within one month) following the lufenuron injection. Under adverse reactions in the package insert for lufenuron, one of the adverse reactions listed is "injection site lumps/granulomas." The section further states, "Histologic examination of one cat's injection site lump showed evidence of inflammation surrounding an area of necrosis with marked proliferation of fibrous connective tissue. In another cat, granulomatous inflammation was noted which included non-pleomorphic fibro-cytes and fibroplasia." This description is similar to the necrotizing granulomas report at vaccine injection sites (Hendrick and Dunagan, JAVMA, 198:304, 1991), some of which, in rare cases, apparently progress to vaccine-associated sarcomas (Hendrick et al., Cancer Research, 52:5391, 1992).

Only time will tell if there will be a significant association between lufenuron injections and development of post-injection sarcomas. But be on the lookout.

The task force originated when American Animal Hospital Association (AAHA) members expressed concerns about cats developing aggressive sarcomas at vaccination sites. AAHA responded by hosting a meeting in San Antonio during the ACVIM annual meeting in June 1996. It became apparent that many leaders in veterinary medicine shared this concern. The consensus by the researchers and clinicians present was that there was a crucial need for research in this area.

Over the summer of 1996, four organizations, AAHA, American Veterinary Medical Association (AVMA), Veterinary Cancer Society (VCS), and American Association of Feline Practitioners (AAFP), joined forces and created the steering committee of the Vaccine-Associated Feline Sarcoma Task Force. We divided the research objectives into in three areas - epidemiology-pathology, etiology, and treatment. We also recognized there was great potential for the vaccine-associated sarcoma to become a public relations disaster for the profession. To address this, we established an objective of providing reliable education - always first to the profession and then to the public.

The VCS has been an integral part of the Task Force effort since its inception. The academics who chair the sub-groups in each area were selected during the October 1996 VCS meeting. Their task was to further define the needs in each of the four areas. Dr. Mattie Hendrick chairs the epidemiology-pathology sub-group, Dr. Barbara Kitchell the etiology sub-group, Dr. Dennis Macy the treatment sub-group, and Dr. Jim Richards the education-communication sub-group.

The steering committee believes the task force must be a forum for all perspectives on this issue. We contacted the Animal Health Institute (AHI), an organization that represents nearly all the vaccine manufacturers in the USA, to obtain representation for vaccine manufacturers. Similarly, the Animal and Plant Health Inspection Service (APHIS), the regulatory agency that licenses all animal vaccines in the US, was contacted and agreed to provide a representative.

The task force had its first formal meeting in November of 1996 to select a chair, review recommendations and establish objectives and priorities for each of the four subgroups. We developed vaccination site recommendations, based in part on AAFP and California VMA recommendations, and decided to prepare a client brochure on Vaccine-Associated Feline Sarcomas (VAFSs).

At subsequent meeting and teleconferences, the Task Force determined that study proposals would be reviewed and scored by independent scientists from AAHA and the AVMA's Council on Research. The first request for proposals was distributed in May 1997, and the Task Force met in November 1997 to review the scored proposals and match them with our anticipated funding. A second request for proposals was distributed in May 1998 and later revised, with submissions due February 1.

· Phil Kass and his colleagues at five major centers will be evaluating a multitude of risk factors, comparing the occurrence in cats with sarcomas to that in cats with basal cell tumors. By early May 1999, Phil had over 1000 cases and unknowns, with a similar number of basal cell tumor controls enrolled in his study. They have an excellent participation rate in obtaining histories from case veterinarians, and are continuing their interviews.

· Kanjilal et al at the University of Minnesota has established a comprehensive tissue bank of sarcomas, blood, and normal appearing margins. A number of sites of polymorphism in the p53 gene have been identified, including in the morphologically normal areas adjacent to excised sarcomas. Preliminary work on identifying signature mutations and correlations to clinical outcomes has been completed.

· MacEwen and Radinsky have established 14 cell lines from vaccine-associated and non vaccine-associated feline sarcomas, and initiated evaluations of levels of growth factors HGF, IGF and PDGF and response of the cell lines to the growth factors. They have also determined the tumorigenic potential of several of the cell lines.

· Hendrick has initiated investigation of tumorigenesis by evaluating the role of local lymphocytes. Preliminary data suggest T-lymphocytes tend to infiltrate the sarcomas, while B-lymphocytes do not. Ongoing work will conduct molecular analysis of growth factors and their receptors.

· Vail and colleagues in a multi-center study are comparing the efficacy of stealth doxorubicin and doxorubicin both before and after surgery. Case accrual is ahead of schedule with 74 enrollees. The stealth form Doxil® was found to have a delayed renal toxicity at an increased dosage of 1.5 mg/kg. Studies will continue at a lower dosage of 1.0 mg/kg.

· Hohenhaus is investigating a radiosensitizer to test the hypothesis that enhancing the oxygenation of the tumor will enhance the cell kill rate after radiation therapy. Due to the departure of AMC's radiation oncologist, the study is proceeding more slowly than anticipated.