Abstract
A 13-year old cat had 3-day duration of vomiting, lethargy, and anorexia. A complete blood count revealed a severe neutrophilia (126.9×103 cellsμl). Thoracic radiographs demonstrated a large solitary lung mass. A bone marrow aspirate documented myeloid hyperplasia. A left intercostal thoracotomy was performed and left cranial and caudal lung lobectomies were performed in order to remove the mass in its entirety. Histological diagnosis was squamous cell carcinoma of the lung. Following surgery, the severe neutrophilia began to decrease. It was in the normal range approximately 6 weeks postoperatively. The presence of a primary lung tumor combined with a mature neutrophilia with resolution following surgical resection suggests a paraneoplastic syndrome.
A 13-year-old spayed female domestic shorthair cat was brought to a referring veterinarian for vomiting, lethargy, staggering, and anorexia of 3 days duration. The cat lived indoors with occasional access to the outdoors. Physical examination abnormalities included a rectal body temperature of 104.5 °F (40.3 °C) and a grade III/VI systolic heart murmur. The initial diagnostic plan included complete blood cell count (CBC) and serum biochemical panel. CBC (Table 1) revealed a leukocytosis (85.8×103 cells/μl; reference range: 5.5–19.5×103 cells/μl) with mature neutrophilia (81.1×103 cells/μl; reference range: 2.5–12.5×103 cells/μl). Hematocrit (37%; reference range: 25–45%) was within reference range. No biochemical abnormalities were present.
Hematological values relative to the time of surgery (t=0 days)
The cat was treated with enrofloxacin (Baytril; Bayer Animal Health) at 5 mg/kg PO bid and amoxicillin at 11 mg/kg PO bid. Five days later the owner returned the cat for continued evaluation because there had been no improvement in clinical signs. On CBC, the hematocrit was normal, but a more severe leukocytosis (99.9×103 cells/μl) and mature neutrophilia (79.8×103 cells/μl) were present. Thoracic radiographs revealed a mass on the left caudal lung lobe, and the cat was referred to Colorado State University Veterinary Teaching Hospital (CSU-VTH) for further evaluation.
Upon presentation to CSU-VTH, physical examination was unremarkable except for the grade III/VI systolic murmur. Echocardiogram revealed a mild to moderate mitral insufficiency that was felt to be from hypertrophic cardiomyopathy. Systolic blood pressure was within normal limits at 115 mmHg. Atenolol was prescribed at a dose of 6.25 mg PO q24hr. Hematocrit (31%) and serum biochemical panel were normal but a leukocytosis (128.2×103 cells/μl) with mature neutrophilia (126.9×103 cells/μl) with only rare band cells was observed. Urinalysis was unremarkable.
Repeat thoracic radiographs confirmed a well-defined soft tissue opacity in the left dorsocaudal lung lobe (Fig. 1). The mass measured approximately 4×5 cm on the lateral projection (Fig. 2). The radiographic interpretation of the mass was most consistent with a primary lung tumor. Less likely causes included abscessation or pulmonary granuloma. The cardiac silhouette, pulmonary vasculature and extrathoracic structures were within normal limits.

Ventrodorsal thoracic radiograph demonstrating large, distinct soft tissue opacity in the left caudal lung lobe.

Left lateral thoracic radiograph demonstrating well-defined soft tissue opacity in the caudal lung fields.
Cytological evaluation of a bone marrow aspirate was performed the following day to characterize the marrow elements relative to the circulating neutrophilia. Bone marrow trabeculae were present and hypercellular; megakaryocytes were present in adequate numbers. There was a marked predominance of immature myeloid precursors with both increased promyelocytes and myelocytes. The myeloid to erythroid ratio was 15:1. Rare macrophages were seen. The clinicopathological interpretation was marked myeloid hyperplasia with left shift, consistent with either an extreme neutrophilic leukocytosis or paraneoplastic leukocytosis (Fig. 3).

Bone marrow cytology 10× and 40×. Myeloid hyperplasia demonstrated by the marked predominance of immature myeloid precursors.
In patients with solitary lung masses who can tolerate anesthesia, surgical resection is the treatment of choice; a complete lobectomy is standard but a partial lobectomy is acceptable for very peripherally located tumors (Withrow, 2001). The cat in this report was taken to surgery 7 days after presentation. A left lateral sixth intercostal thoracotomy was used to access the mass and hilar region. Although the mass was confined to the caudal lobe, multiple adhesions were present between the cranial and caudal lobes of the left lung. As a result, both the cranial and caudal lobes of the left lung were excised en-block at the hilus using a 55 mm thoracoabdominal stapler (TA-55 stapler, United States Surgical). The mass was cultured for aerobic and anaerobic bacteria and then placed in 10% buffered formalin for histopathological evaluation. In addition, sternal lymph node and mediastinal biopsies were performed. No visual enlargement of regional lymph nodes was evident.
Histological examination of the mass revealed large areas of necrosis and papilliferous proliferations of neoplastic cells formed by columnar cells found in single to multiple layers. The cells had abundant eosinophilic cytoplasm and large oval nuclei with prominent nucleoli. In some areas, the periphery of the proliferations had larger cells with more polygonal morphology and clear light eosinophilic cytoplasm, occasionally with keratin pearls in the centers of these. There was prominent lymphoplasmacytic inflammation and suppuration throughout the mass and some areas of hemorrhage. On the pleural surface, there were areas of large neoplastic cells with numerous mitotic figures (Fig. 4). The histological diagnosis was poorly differentiated squamous cell carcinoma. Vascular invasion was present as evidenced by sections of hilar tissue containing arteries with neoplastic tissue within their lumens. The mediastinal biopsy contained clotted blood and numerous neutrophils. The sternal lymph node contained neutrophils in the subcapsular and pericortical regions. No neoplastic tissue was observed in either the mediastinum or lymph node. Aerobic and anaerobic cultures of the mass yielded no growth.

Lung histology 10× and 40×. Section of lung through neoplastic mass. Papilliferous proliferation of neoplastic cells with abundant eosinophilic cytoplasm and large oval nuclei with prominent nucleoli. Occasional keratin pearls and numerous mitotic figures are also present.
The tentative diagnosis was squamous cell carcinoma of the lung with paraneoplastic leukocytosis. Adjunctive chemotherapy was recommended. In a very limited number of cases of animals with primary lung tumors, multidrug chemotherapy has been tried with some benefit (Withrow, 2001). In this case the owner opted to treat with feldene (Piroxicam; Pfizer Animal Health), a non-steroidal anti-inflammatory drug that has been shown to be useful in the treatment of canine squamous cell carcinoma (Schmidt et al., 2001). Radiation has been of little use for treatment of primary lung tumors in cats, as the large doses of radiation required to kill tumor cells result in significant pathology to normal lung tissue (Withrow, 2001).
CBC 8 days after surgery, revealed that the leukocytosis had decreased (50.5×103 cells/μl), as had the neutrophilia (44.4×103 cells/μl) (Table 1). Thirteen days after surgery, the leukocytosis decreased to 30.6×103 cells/μl; the neutrophilia to 22.0×103 cells/μl. Within 6 weeks after surgery the leukogram was normal with a total white cell count of 18.0×103 cells/μl and a total neutrophil count of 11.9×103 cells/μl.
Based on the findings of the bone marrow aspirate, granulocytic leukemia and metastatic were considered unlikely to have caused the mature neutrophilia. Likewise, there was no evidence of infection or necrosis in the lung tissue or elsewhere in the cat's body to elicit such an extreme inflammatory process from the bone marrow. The presence of a primary lung tumor combined with a mature neutrophilia with resolution following surgical resection suggests a paraneoplastic syndrome. Definitive confirmation of a paraneoplastic syndrome requires derivation of tumor cell line and demonstration of G-CSF or GM-CSF production from the tumor, which the authors were not able to perform (Hocking et al., 1983; Sato et al., 1989; Sharkey et al., 1996).
Paraneoplastic leukocytosis is a marked, mature neutrophilia that cannot be attributed to inflammatory disease or to a primary hematopoietic neoplasm. A monocytosis may also be present (Bunn and Ridgeway, 2001). Stimulation of bone marrow by necrotic and/or infected tumor masses, bone marrow metastasis and production of colony stimulating factors have all been proposed as mechanisms for paraneoplastic leukocytosis (Lappin and Latimer, 1988). The production of colony stimulating factors by the tumor is the most widely accepted hypothesis. In humans, cancer producing granulocyte colony-stimulating factor (G-CSF) and granulocyte-macrophage colony-stimulating factor (GM-CSF) have been well documented (Sato et al., 1989; Furihata et al., 1996; Kato et al., 1999). The degree of leukocytosis correlates well with tumor burden (Kitamura et al., 1989). Although not yet demonstrated, this same mechanism has been postulated in dogs and cats (Sharkey et al., 1996). In this cat, necrosis was confined to the neoplasm, no evidence of infection existed, and the bone marrow showed no evidence of metastasis. For these reasons, tumor production of G-CSF and/or GM-CSF is thought to be the most likely cause of paraneoplastic leukocytosis (Sharkey et al., 1996). Paraneoplastic leukocytosis is often associated with hypercalcemia and with squamous cell carcinoma of the lung in people; this tumor is the most common cause of paraneoplastic hypercalcemia in humans. Calcium was measured twice in this cat, values both times falling within reference range. Ionized calcium was not evaluated in this cat. To the best of our knowledge, this is the first report of paraneoplastic leukocytosis associated with pulmonary squamous cell carcinoma in the cat.
Two prognostic factors for survival times in cats following surgical resection of primary lung tumors have been described: histological morphology of the tumor and size of tracheobronchial lymph nodes. Histological morphology of the tumor is significantly associated with survival time whereas size of tracheobronchial lymph nodes appears to be of prognostic value (Hahn and McEntee, 1998). Mean survival time of cats following surgical resection of poorly differentiated primary lung tumors is 75 days (Hahn and McEntee, 1998). The cat described in this report died 74 days postoperatively from undetermined causes.
