Efficacy of Cobalt-60 Radiation Therapy for the Treatment of Nasal Cavity Nonkeratinizing Squamous Cell Carcinoma in the Dog

Introduction

Tumors of the nasal cavity and paranasal sinuses are uncommon in dogs, accounting for approximately 1% of all neoplasms.¹ These tumors have a low incidence of clinically detectable metastasis at the time of diagnosis, but they are characterized by progressive local invasion with bony destruction occurring early in the course of disease.^1–3 To date, therapy for these tumors has been directed at local disease utilizing a variety of treatment modalities.^4–9

Radiotherapy has become an integral part of treatment for canine nasal tumors, but prognostic factors that influence response have not yet been clearly identified.⁵⁶ The biological behavior and treatment response of canine intranasal neoplasia are believed to vary with histopathology,⁵¹⁰¹¹ and previous studies have attempted to evaluate the prognostic importance of tumor histopathology with respect to survival, but with conflicting results. It has been suggested that adenocarcinomas and sarcomas have a survival advantage when compared to other carcinomas.⁴ In a more recent study, mean and median survival times for patients with nasal carcinoma versus sarcoma were not statistically different.¹² However, in a study of 77 dogs treated with megavoltage radiation, the overall survival time of dogs with sarcomas was significantly longer than the overall survival time of those with carcinomas.⁵ The apparent discrepancy in the prognostic importance of histopathology for survival time between studies is likely a result of several factors. The various pathological classification systems and the small number of dogs with each histopathological tumor type contribute to the variability between studies. Differences in radiation dose, overall treatment duration, fractionation pattern, dose per fraction, and follow-up information almost certainly affect patient response and survival as well.

Nasal nonkeratinizing squamous cell carcinomas are a rare clinical entity also referred to as transitional carcinomas of respiratory epithelium. These tumors lack keratinization, and the neoplastic cells are mostly spindle cells. The purposes of this study were to specifically evaluate the efficacy of cobalt-60 radiotherapy in the treatment of nasal nonkeratinizing squamous cell carcinomas and to compare this treatment group to historical controls.

Materials and Methods

Six dogs with nasal nonkeratinizing squamous cell carcinomas were treated at the Animal Medical Center in New York, New York, between October of 1990 and November of 1997. Medical records, radiographs, and available computed tomography (CT) scans were reviewed for each animal. Clinical patient evaluation prior to the initiation of radiotherapy included a complete physical examination, hemogram, serum biochemical profile, urinalysis, thoracic radiographs, CT scan, and tissue biopsy for histopathology. Computed tomography scan examination consisted of 5- to 10-mm contiguous slices from the nares to the foramen magnum. Tissue specimens were fixed in 10% buffered formalin and embedded in paraffin, and sections were stained with hematoxylin and eosin. All biopsy specimens were reviewed by one pathologist (Patnaik).

Patients with tumors that were classified as nonkeratinizing squamous cell carcinoma based on the specific criteria defined in the authors’ previous study were candidates for inclusion.¹¹ Briefly, these tumors are characterized by groups and various-sized sheets of cells enclosed by thin, fibrous stroma arranged in a plexiform pattern. The tumor lacks keratinization. The neoplastic cells are mostly spindle cells with distinct cell borders and pale eosinophilic cytoplasm with hyperchromatic nuclei and prominent nucleoli. Numbers of mitotic figures are variable, as is the degree of lymphocytic infiltration seen in these tumors.

Tumors were staged based on CT characteristics according to a modified staging scheme. In a recent study, this staging scheme was demonstrated to more accurately influence relapse-free survival times than the World Health Organization (WHO) staging system.⁵ Stage 1 disease includes unilateral or bilateral neoplasm confined to the nasal passages without extension into the frontal sinuses. Stage 2 disease includes bilateral neoplasm extending into the frontal sinuses with erosion of any bone of the nasal passages.⁵

Computed tomography images were also used for planning radiation therapy. A CT-guided, computer-assisted radiation treatment plan was generated for each patient with the aid of a treatment planning computer. Treatment parameters included a maximum target volume dose of 105% with <15% dose inhomogeneity across the field. All tumor was included within the 95% isodose line. Complex field geometry including wedges, bolus, and greater than two fields were used as necessary to meet the above treatment parameters. Most treatment plans consisted of parallel-opposed fields at 0° and 180° (ventrodorsal and dorsoventral). Radiation field size was also evaluated using equivalent squares. All dogs received cobalt-60 radiation.^a The total dose of radiation delivered was determined by the individual radiation oncologist responsible for the radiation planning at the time of each dog’s treatment. Three dogs were scheduled to receive a total dose of 63 Gy in 21 fractions. One dog in this group did not complete therapy because of progressive disease with a survival time of 30 days. Three dogs were scheduled to receive a total dose of 54 Gy in 18 fractions. One dog in this group did not complete therapy because of progressive disease with a survival time of 60 days. Patients were treated three times weekly over a 6- to 7-week period.

One patient was treated with chemotherapy. Cisplatin^b chemotherapy was used at a dose of 60 mg/m² every 4 weeks for three total treatments. Twelve hours prior to chemotherapy administration, the patient was diuresed with 0.9% sodium chloride crystalloid solution. Immediately prior to chemotherapy administration, dexamethasone^c (0.05 mg/kg body weight), butorphanol^d (0.1 mg/kg body weight), and prochlorperazine^e (0.44 mg/kg body weight) were administered intravenously (IV). After chemotherapy administration, the patient received mannitol (0.5 g/kg body weight IV) and was again diuresed for 12 hours with 0.9% sodium chloride.

Patient follow-up information included telephone communication with the owner (for two dogs) or evaluation at the Animal Medical Center (for four dogs). Repeat thoracic radiographs and CT scans were performed on five of the six dogs. Survival time was calculated from the time of histopathological diagnosis to the date of death. Kaplan-Meier survival analysis was performed on the survival data. Pearson correlation coefficients were utilized to evaluate potential relationships between radiation field size, tumor volume at diagnosis, duration of clinical signs, age, and overall survival time. A Mann-Whitney test was used to compare overall survival times in patients with and without seizures. A P value of <0.05 was considered significant in all cases.

Results

Overall survival times ranged from 30 days to 330 days, with a median of 165 days. Five dogs were either euthanized or died due to their disease, and one dog died due to other causes [Figure 1; see Table].

Five dogs were male, and one was female spayed. The ages ranged from 7 to 10 years (median, 10 years). There were two mixed-breed dogs and four purebred dogs representing four different breeds (golden retriever, keeshond, Doberman pinscher, wirehaired fox terrier). The most common clinical sign was epistaxis, and this was the presenting complaint in four dogs. The only clinical sign displayed by two dogs was seizures. The duration of clinical signs prior to presentation ranged from 30 to 180 days (median, 45 days). The dogs in this study, particularly those that received 63 Gy of radiation, did experience acute-term radiation side effects, including mucositis, dermatitis, and keratoconjunctivitis sicca.

The six dogs in this study had localized disease at presentation, and none developed evidence of metastasis at any time during treatment. All dogs were classified as having stage 2 disease based on a previously published staging system,⁵ and CT scanning revealed bony lysis in all dogs. The two dogs with late-onset seizures had evidence of central nervous system extension on CT examination.

Four of the six dogs had repeat CT scans after completion of radiation therapy. One dog had a repeat CT scan during radiation therapy. One dog did not have a repeat CT scan and was euthanized prior to the completion of radiation therapy, with a survival time of 60 days due to progressive neurological signs. One dog had a partial response to treatment based on repeat CT scan 2 months postradiation therapy, with a median survival time of 150 days. This dog received three doses of cisplatin chemotherapy concurrently with radiation therapy and died due to other nontumor-related causes. Computed tomography scan revealed progressive disease in the four other dogs.

Relative radiation field size was not significantly related to survival time, nor was the tumor volume present at diagnosis. The duration of clinical signs prior to diagnosis was not related to survival time. There was no correlation between the age of the patient and the overall survival time. Total radiation dose delivered was not significantly related to survival time. The median survival time of dogs with seizures was 45 days, compared to 255 days for dogs without seizures. However, this difference did not reach statistical significance (P=0.06). Neither dog with seizures as the presenting complaint completed therapy because of rapidly deteriorating neurological status.

Discussion

Results of this study indicate that nonkeratinizing squamous cell carcinoma of the nasal cavity in dogs is an aggressive tumor that responds poorly to radiotherapy, with a median survival time of 5.5 months. This is considerably shorter than previously reported median survival times for dogs with nasal tumors that receive cobalt radiotherapy, which vary from 8.1 months to 16.5 months depending upon the study reviewed.⁴⁶¹⁰ Only one study of which the authors are aware, isolates nonkeratinizing squamous cell carcinomas into a group separate from other carcinomas. In that study, the authors classify nonkeratinizing squamous cell carcinomas as transitional carcinomas, and the median survival time with cobalt radiotherapy was 9.4 months.⁵ These dogs were treated with cobalt radiotherapy to a total dose of 48 Gy. The dogs in that study had both stage 1 and stage 2 disease, unlike the dogs in this study that had only stage 2 disease.

Tumor control was achieved in only one dog in the present study. This dog had evidence of a partial response to therapy on a repeat CT scan 2 months after completion of radiation therapy. All other dogs had progressive disease. This dog also received concurrent chemotherapy. However, it is impossible for us to draw any definitive conclusions concerning the efficacy of chemotherapy combined with radiation therapy based on data from one dog. At the time of diagnosis, all six tumors were large, infiltrative, and locally destructive and were classified as stage 2 tumors. The failure of radiotherapy in the treatment of these tumors may be a result of several factors. Extended total treatment times of >4 to 5 weeks may result in decreased tumor control due to increased tumor cell repopulation. The inherent shape of the canine nasal cavity makes homogeneous dose distribution difficult. Additionally, megavoltage therapy requires a dose buildup tissue zone that is interrupted by air in the sinuses surrounding the treatment field.¹³ Another important factor to consider is that large, bulky nasal tumors are postangiogenic lesions. Angiogenic vessels, unlike vessels of normal tissue, are highly irregular, tortuous, leaky, have arteriovenous shunting, and blind ends.¹⁴ As a result, blood flow is sluggish and irregular. Due to the nature of postangiogenic vasculature, areas of hypoxia build up within the tumor. When tissue hypoxia is present, there is resistance to radiotherapy because of decreased cellular radiation sensitivity.¹⁵¹⁶ Oxygen is necessary for tissue damage to occur from radiation therapy. In hypoxia, at least three times the amount of radiation is required to produce the same effect as in oxygenated tissue.¹⁷ Also, hypoxia inhibits cell proliferation. Cells that are actively proliferating are the most sensitive to radiation therapy.¹⁶ In large, bulky, postangiogenic tumors with areas of hypoxia and decreased cell proliferation, it is unlikely that radiation therapy alone will control disease.

Five of the six dogs in this study were males. A male predominance for the development of canine nasal tumors has been reported¹⁸ as well as a slight male predominance for the development of nasal tumors in humans.¹⁹ Four of the six dogs in this study were purebred dogs. It has been previously reported that mixed-breed dogs display the same risk as all breeds combined, suggesting the influence of environmental factors in the development of this tumor.²⁰ The significance of the overrepresentation of male and purebred dogs in the current study is unknown.

Seizures were the primary presenting complaint in two of the six dogs in this study. The authors’ data suggests that dogs with central nervous system involvement may have shortened survival times and a poorer prognosis, although statistical significance was not achieved. Additional study with larger numbers of patients will be necessary to further define this potential difference.

Several factors often contribute to small sample size in studies examining canine nasal cavity tumors, and they limit the conclusions that may be drawn; the current study is no exception. Many owners may decline therapy or elect palliative rather than definitive radiation therapy for these large, stage 2 tumors. Also, nasal cavity nonkeratinizing squamous cell carcinomas are not common tumors,¹¹ and histopathological classification of nasal tumors in the dog is problematic. Squamous cell carcinomas can have varying degrees of differentiation.²¹ When the tumors are poorly differentiated, recognition and categorization can become difficult. In humans, nonkeratinizing squamous cell carcinomas, transitional carcinomas, and respiratory carcinomas are all grouped under nonkeratinizing squamous cell carcinoma.²²¹

Conclusion

The goal of this retrospective study was to determine if cobalt-60 radiotherapy was effective in the treatment of nasal nonkeratinizing squamous cell carcinomas. Based upon the results of this study, the median survival time for nasal nonkeratinizing squamous cell carcinoma is less than previously reported historical controls for carcinomas as a group. Central nervous system involvement may further shorten survival times. An optimal protocol for the treatment of these tumors has yet to be developed, and continued evaluation of individual tumor types treated with uniform treatment protocols is necessary to allow for dependable conclusions regarding both biological behavior and treatment response.