Case Report

One 7 yr old female pit bull terrier was referred for evaluation of a swelling in the left jugular region, not hot or painful. Physical examination was unremarkable, except for the presence of the cervical mass. The results of a complete blood count and complete serum biochemical analysis were within normal limits.

A cross-sectional imaging session was performed to define the lesion. Computed tomography (CT)^a scans were acquired before and after iodinated contrast medium injection (iopamiro^b 300 mg/dL administered at the dose of 0.2 mL/kg) showed a solid mass of inhomogeneous structure located in the left dorsal cervical region, dorsolateral to the larynx, connected to the common carotid artery at the level of the bifurcation (Figure 1). The mass was ovoid, 5 cm in diameter, with CT numbers ranging from +22 to +43 HU in direct scans, and from +32 to +184 HU in postcontrast scans (Figure 1).

View Figure

• Download as Powerpoint
• Download Figure

MRI^c was performed to understand better the soft-tissue characteristics of the cervical mass. The lesion was slightly hyperintense relative to adjacent muscles on T2-weighted sequences (Figure 1A). T1-weighted sequences were acquired before and after contrast medium injection (gadodiamide^d 0.5 mmol/kg). On postcontrast MRI and CT scans, the mass showed peripheral early enhancement. Newly formed blood vessels, of small diameter and with a peripheral location, were evident (Figure 1C). Thyroid gland and regional lymph nodes were normal on both CT and MRI.

Imaging characteristics, the anatomical location of the mass, and the visualization of normal thyroid glands made possible a presumptive diagnosis of carotid body paraganglioma.¹ To assess the production of catecholamines, an electrocardiogram with a provocative test, performed by massaging the lesion, was performed. No abnormalities were recognized. Multimodal therapy, with surgical resection and adjuvant radiotherapy and chemotherapy, was proposed.

A marginal excision was planned with the aim of removing the whole mass while sparing the carotid artery. A ventral surgical approach to the neck was made after premedication with methadone^e and medetomidine^f and general anesthesia induced with propofol^g and maintained with isoflurane^h in oxygen. Once we had dissected the superficial cervical fascia at the bifurcation between the maxillary artery and left internal carotid artery, the mass was isolated. The lesion was detached from the vagus-sympathetic trunk and from the carotid artery; only the caudal thyroid artery was ligated and removed. Once we had completed the removal of the mass, individual layers of the surgical wound were closed separately. During the perioperative period, cefazolinⁱ and meloxicam^j were administered. No intra- or postoperative complications were observed.

On histopathological examination, the neoplastic tissue was surrounded by a fibrous capsule and divided into lobules by prominent branching trabeculae of connective tissue, crossed by an abundant network of capillaries. Tumor cells were discrete, from cuboidal to polyhedral, with light eosinophilic, finely granular, or vacuolated cytoplasm; nuclei were round to oval and usually centrally placed; mitotic figures were infrequent. Tumor cells were frequently arranged in small lobules, aligned along and around the thin capillary network. Some tumor cells showed evidence of focal invasion through the peripheral capsule. For immunohistochemical evaluation, 4-μm-thick slices of paraffin-embedded tissue and standard streptavidin-biotin indirect immunoperoxidase were used. Every section was immunostained with mouse monoclonal antibodies to human neuron-specific enolase and human chromogranin A. Based on the tumor’s morphology, immunoreactivity, and anatomical location, the diagnosis of paraganglioma of the carotid body (chemodectoma) was made.

A frameless stereotactic body radiation therapy (SBRT) was planned. For the treatment, a 6 MV linear accelerator^k equipped with a micro-multileaf collimator and an X-ray volume imaging system (XVI) was available.

CT simulation was obtained 20 days after surgery to generate the frameless SBRT with volumetric modulated arc therapy (VMAT) using the Monte Carlo statistic algorithm and the content management and treatment planning system^l (Figure 2). The dog was positioned on a vacuum mattress inserted into a stereotactic cradle^k, and the XVI device was used to verify correct patient positioning.

View Figure

• Download as Powerpoint
• Download Figure

The prescribed dose of 35 Gy in five fractions was set up with a VMAT plan in a single 360° arc (Table 1). The gross target volume did not exist because the tumor had been resected. The clinical target volume was defined as the volume that was likely to contain the subclinical microscopic malignant disease. By clinical consideration, it was identified as the volume corresponding to the preoperative gross target volume plus undissected rounding tissues and a volume of expansion. The uniform expansion was 2 cm in all directions, and it was justified by the histopathological focal invasion through the peripheral capsule, but it did not account for the organ motion and setup uncertainties. The clinical target volume encompassed pharyngeal lymph nodes, but when superimposition of organs at risk (OARs) occurred, the entire volumes of the OARs were removed. The OARs that were contoured were the larynx, pharynx, esophagus, trachea, tympanic bullae, spinal cord, brain stem, brain, and cerebellum. The dose constraints were derived from human guidelines described in the American Association of Medical Physics Task Group 101 paper (Table 2).² To assess the treatment feasibility, treatment simulation was performed “in air” with the dosimetric check software^m using the Iview amorfous silicon electronic portal imager deviceⁿ. The agreement was parameterized by the gamma function with a dose agreement of 3% and a distance to agreement of 3 mm. The plan was accepted for a gamma <1 in 94% of comparison points. The quality assurance check revealed a −3.46% discrepancy between the planned and the delivery dose at the isocenter (delivery dose: 34.7 Gy).

TABLE 1 Plan Setup Report Generated by Treatment Planning System

View Fullscreen

TABLE 2 Organs at Risk Isoconstraints Resulting from Planning Optimization

View Fullscreen

The patient was repositioned using radiopaque simulation markers with XVI system verification. A 2-mm tolerance level displacement was considered acceptable. For checking interfraction differences, an “on-transit” control was performed using the Dosimetry Check to assess dose matching between the planned distribution obtained from the simulation and the delivered distributions.

After the completion of radiation therapy, systemic chemotherapy with six doses of carboplatin^o at the schedule of 300 mg/m²/mo was administered. Clinical follow-up was conducted over 12 mo. Physical examination was performed 6 and 12 mo after the conclusion of the whole therapy course, and both CT and MRI total-body scans were performed at the 12-mo control. During the follow-up, according to Radiation Therapy Oncology Group and Veterinary Cooperative Oncology Group criteria for radiotoxicity evaluation, acute grade I dermatitis and grade I left-side laryngeal mucositis were recognized.^3,4 One yr after the completion of radiation therapy, no signs of late radiotoxicity or tumor recurrence were observed. According to the VRTOG criteria for toxicity evaluation, during chemotherapy, grade I thrombocytopenia was observed.⁴

Discussion

Chemodectomas or paragangliomas are tumors that arise from the peripheral chemoreceptors, which originate from the neural crest cells and are localized in different anatomical areas, where they serve as sensors that warn of changes in the composition of the blood, partial pressure of oxygen, partial pressure of CO₂, and pH, thereby adjusting the respiration and circulation.⁵ The main chemoreceptor cells are located in the aortic bodies at the carotid bifurcation and near the aortic arch; some other chemoreceptors are also found at the level of the jugular bulb, tympanic cavity, and ganglion nodosum.⁶

Chemodectomas are rare tumors found in dogs, cats, humans, and cattle that grow mainly at the level of the aortic arch and carotid glomus. The etiopathogenic basis includes a genetic predisposition in some dog breeds, such as boxers, Boston terriers, and Pekingese, and prolonged exposure to hypoxia.^5,7,8 The average age of onset in the dog is 8 yr, and males are most affected.⁷

These tumors are rarely functional and secrete catecholamines that can cause ventricular arrhythmias. They more often exert a mass effect, compressing blood vessels and nerves of the neck. They rarely metastasize, but compared with paragangliomas of the glomus aortic, carotid chemodectomas are more often malignant, and in 30% of cases, they can metastasize to regional lymph nodes, lung, liver, pancreas, bones, and kidneys.^9,10 The carotid chemodectomas are usually located near the bifurcation of the carotid artery in the cranial cervical region near the angle of the jaw. The structure, inner growth with vessels and nerves, and vascularity of this tumor make surgical excision particularly difficult. The carotid carcinoma usually is larger than the adenoma and is usually well encapsulated, but tumor cells often infiltrate adjacent tissues.¹¹ In dogs, the diagnosis is based on clinical history, physical examination, imaging techniques, histology, and immunohistochemistry.⁸ Imaging provided by CT or MRI should be considered a noninvasive, accurate, and feasible exam for a presumptive diagnosis. A previous report reviewed cases of carotid body tumors without available CT or MRI scans; of the 13 suspected chemodectomas, only 11 were ultimately confirmed as carotid body tumors.⁵

On CT, the chemodectoma appears as a well-defined mass, with dense soft tissue within the carotid space and intense and homogeneous contrast enhancement (Figure 1C). Large tumors usually are not uniform due to the presence of necrotic and hemorrhagic areas.^6,12 On MRI, carotid body tumors are heterogeneous, iso-hyperintense on T1-weighted spin echo, and heterogeneous hyperintense on T2-weighted spin echo (Figure 1A).^6,12–15 Chemodectomas show a characteristic heterogeneous appearance, the so-called "salt and pepper" aspect, in the T2-weighted spin echo images (Figure 1A), due to the presence of regions with low signal intensity in the mass (high-flow vessels) and regions with high signal intensity (slow flow or hemorrhage).^6,14 A differential diagnosis should be made with reactive or metastatic lymph nodes, abscesses, hyperplasia, or thyroid cancer.¹¹

Cytology is often not conclusive. Immunohistochemical demonstration of positive chromogranin A and vimentin reactivity is pathognomonic for neuroendocrine tissue.⁸ Surgery is the best treatment. However, the feasibility of a complete surgical excision is limited to the early stages of the disease. In the case series of Obradovich et al., 10 dogs had surgery, and 9 of them died or developed neurological deficits.⁵ As reported in humans, in advanced stages, the infiltrative growth makes surgery complex and risky.^5,16,17

In human medicine, to prevent possible iatrogenic injuries, embolic treatment and radiation alone or in combination are proposed.¹⁸ In patients in weak clinical condition or affected by unresectable masses, palliative treatment with octreotide is performed.¹⁹ To assess a chemodectoma’s resectability, the classification system of Shamblin is used.^5,13–15 Stage I tumors can be easily removed because they are small and have poor adhesion to the vessels; stage II tumors are intimately associated with the arterial wall and they can be removed with careful subadventitial dissection; stage III tumors strongly adhere to the vessels and require resection of the internal or external carotid artery.⁵

For tumors not amenable to surgery, partially excised tumors, or high-risk patients, radiotherapy is recommended.⁵ Chemodectomas were considered radioresistant tumors, but the supposed concept of their radioresistance is actually not supported by most of the recent literature.^16,19 Moreover, past radiotherapy techniques for paraganglioma implied high exposure of the nerves and bones to the radiation, resulting in considerable radiation-induced complications, unlike newer radiotherapy image-guided or volumetric techniques that allow better sparing of these tissues.¹⁶

There is little literature on chemodectoma radiotherapy in dogs and cats. In the 11 cases of Obradovich et al., only 2 dogs underwent radiotherapy. The hypofractionated treatment consisted of 40 Gy in 10 fractions for one dog and 48 Gy in 12 fractions for the other. No technical details about equipment and treatment planning were available.⁵

SBRT has shown therapeutic efficacy with increased delivered dose and is actually the treatment of choice for unresectable tumors or in an adjuvant setting to improve local disease control.^19,20 In particular, the arc techniques minimize the dose to the surrounding tissues and thus enable the safe delivery of a higher dose to the tumor.¹⁹ SBRT dose fractionation in human patients with paragangliomas ranges from 16 to 25 Gy (median 20.5 Gy) in 1–5 fractions on consecutive days.²¹

In this study, we decided to use stereotactic radiation therapy for the eradication of possible residual disease. The total dose of 35 Gy delivered in five fractions prescribed to the 95% isodose was calculated (Figure 2) to distribute the maximum dose to the tumor with a biological effect that could be tolerated without side effects to the surrounding structures, as reported in American Association of Medical Physics Task Group 101. To compare human regimens to our regimen, the biologically effective dose (BED) was used.²² The BED is calculated using the formula BED = tD[1+D/(α/β)], in which tD = total dose, D = dose per fraction, and n = number of fractions, and it is expressed in Gy_α/β. The BED formula has not been established for treatment effectiveness of SBRT because of the lack of experimental validity when large doses per fraction and short overall treatment times have been used. However, the BED formula currently serves as a useful model for biological comparison of different fractionations, particularly for adverse reactions. An α/β of 3 Gy is generally applied for late-responding tissues and late radiotoxic effects, whereas a value of 10 Gy is applied for acutely responding tissues and acute radiotoxic effects. The BED of a typical SBRT protocol for human paragangliomas (three fractions of 6.7 Gy for a total dose of 20 Gy given on consecutive days) has a Gy₃ value of 66.7 Gy and a Gy₁₀ value of 34 Gy.²¹ The BED of our study (35 Gy in five fractions on consecutive days) had a Gy₃ value of 116.7 Gy and a Gy₁₀ value of 59.5 Gy. Such doses and the type of fractionation allowed excellent local control of the disease and at the same time allowed the preservation of the surrounding critical structures.

To the best of our knowledge, this is the first report of a canine carotid chemodectoma successfully treated with a multimodal approach, in particular with VMAT adjuvant postsurgical stereotactic radiotherapy. VMAT allows better sparing of normal tissues and the delivery of high doses to the target, with a higher disease control probability.

Conclusion

Multimodal therapy of chemodectomas may be an effective treatment option when microscopic extracapsular extension is documented by histopathology and/or when marginal excision is performed. In particular, SBRT with VMAT represents a potential technique developed in providing disease control and producing slight radio-induced side effects. Further evaluation and multiple studies on canine patients are needed to confirm these results.