Intestinal tumours are relatively rare in dogs, accounting for 8 percent of all tumours (Dobson et al., 2002). Lymphoma is the most common intestinal neoplasia in dogs, followed by adenocarcinoma and then leiomyosarcomas or gastrointestinal stromal tumours (GISTs). Although the literature is contradictory, males may be over-represented as some studies have shown male dogs are more likely to develop intestinal lymphoma, adenocarcinoma, leiomyoma and leiomyosarcoma compared to females (Selmic et al., 2019). There are no consistently reported breed predispositions in dogs.
This article will focus on epithelial and mesenchymal tumours of the intestine, highlighting important differences between biological behaviour based on anatomical location. It will cover key clinical signs, how to investigate and diagnose intestinal tumours and the aetiology, treatment and prognosis of specific intestinal tumour types in dogs.
Clinical signs
Clinical signs associated with intestinal tumours mainly include weight loss, diarrhoea, hyporexia and vomiting. These can vary depending on the location of the mass. For example, small intestinal tumours may predominantly present with weight loss, whereas colorectal tumours may be more likely to cause haematochezia and tenesmus. If intestinal obstruction or perforation occurs, then clinical signs of an acute abdomen may follow. Occasionally, clinical signs secondary to paraneoplastic syndromes can occur. In one study, hypoglycaemia attributed to gastrointestinal leiomyosarcoma was reported in over 50 percent of dogs (Cohen et al., 2003). A list of reported paraneoplastic syndromes associated with intestinal tumours is found in Table 1.
| Paraneoplastic syndrome | Tumour type | Reference(s) |
| Alopecia | Intestinal carcinoma | Muller et al., 2002 |
| Neutrophilia | Rectal adenomatous polyps | Thompson et al., 1992 |
| Erythrocytosis | Caecal leiomyosarcoma | Sato et al., 2002 |
| Hypoglycaemia | Smooth muscle tumours | Cohen et al., 2003; Bagley et al., 1996 |
| Nephrogenic diabetes insipidus | Intestinal leiomyosarcoma | Cohen and Post, 1999 |
Diagnosis and investigation
Patients with intestinal tumours should have a minimum database of haematology, biochemistry and urinalysis. Imaging should be performed to further investigate the primary tumour and assess for possible metastatic disease. Accessible lesions should be sampled for cytology and/or histopathology.
Patients with intestinal tumours should have a minimum database of haematology, biochemistry and urinalysis
Physical examination reveals a palpable abdominal mass in up to 50 percent of cases. Rectal examination is crucial as this may aid the identification of rectal masses or strictures/annular lesions. Haematology frequently documents anaemia secondary to gastrointestinal bleeding or chronic inflammation, but melena may not always be present. Neutrophilia is also common. Serum biochemistry demonstrates hypoproteinaemia secondary to malabsorption in up to one-third of cases. Cobalamin levels should also be assessed in patients with small intestinal tumours in case supplementation is required. Other biochemical abnormalities can include elevated alkaline phosphatase, elevated urea (due to gastrointestinal bleeding) and hypoglycaemia as previously discussed.
Diagnostic imaging
Diagnostic imaging to evaluate the primary tumour can be performed via abdominal ultrasound (US) or computed tomography (CT). In most cases, US should be sufficient to identify a focal intestinal lesion, assess regional lymph nodes (LNs) for enlargement and evaluate for visceral metastasis. Additionally, ultrasound can more accurately assess intestinal wall layering compared to CT. This is important as loss of wall layering is significantly associated with neoplasia compared to inflammatory lesions (Penninck et al., 2003).
If the mass is causing luminal obstruction or has perforated, then surgical excision for both diagnostic and therapeutic purposes may be considered
Larger dogs where abdominal ultrasound can be more challenging may benefit from CT. Computed tomography is also more sensitive than ultrasound at detecting and assessing intra-abdominal LNs. Ultrasound-guided aspiration of the mass and any enlarged intra-abdominal LNs and/or other lesions (eg hepatic or splenic nodules) is indicated to try to achieve a definitive diagnosis. If cytology is non-diagnostic or inconclusive, then US-guided needle biopsies (eg Tru-Cut) can be obtained. If the mass is causing luminal obstruction or has perforated, then surgical excision for both diagnostic and therapeutic purposes may be considered.
Staging is indicated in all cases of malignant intestinal neoplasia. As discussed above, the abdomen can be imaged via US or CT, and all accessible abnormalities should be sampled. Most epithelial and mesenchymal intestinal tumours will metastasise to regional LNs and lungs, but as vascular drainage from the gastrointestinal tract is via the portal circulation, metastasis can also be seen in visceral organs, especially the liver. The thorax should be assessed via either CT or radiographs. Computed tomography is most sensitive, and is able to detect pulmonary nodules as small as 1mm compared to 7 to 9mm with radiographs. If radiographs are to be obtained, then taking three-view inflated images is the most sensitive method.
Aetiology, treatment and prognosis of specific intestinal tumour types
Adenocarcinoma
There are no consistently reported breed dispositions for intestinal carcinomas in dogs, although German Shepherd Dogs and collies are over-represented in some reports (Patnaik et al., 1977; Seiler, 1979). Mutations in the adenomatous polyposis coli (APC) tumour suppressor gene are strongly linked to benign colorectal polyp formation in humans and can gradually progress to carcinoma. APC gene mutations have been documented in 70 percent of canine colorectal tumours, suggesting a similar molecular aetiology (Youmans et al., 2012). It has also been reported that in Miniature Dachshunds, a breed predisposed to inflammatory colorectal polyps, malignant transformation of initially benign polyps may occur (Saito et al., 2018).
In dogs, the colon is the most common site for adenocarcinoma, followed by the rectum and small intestine (Patnaik et al., 1980; Church et al., 1987). Colorectal adenocarcinomas demonstrate either a pedunculated, cobblestone or annular morphology, and these forms carry prognostic significance. The risk of metastasis is moderate for small intestinal adenocarcinoma, with one study reporting LN, mesenteric and omental metastasis in 38, 41 and 21 percent of cases, respectively (Smith et al., 2019). The metastatic rate of colorectal adenocarcinoma is poorly defined but appears low.
Treatment and prognosis
The treatment of choice for intestinal carcinoma is surgical excision. Mesenteric LNs should be biopsied regardless of size or appearance, as the metastatic status of these nodes may hold prognostic significance. Any other lesions identified on preoperative imaging or during surgery should also be sampled if this has not already been done. Following surgery, adjuvant chemotherapy does not appear to offer a survival benefit in cases of small intestinal adenocarcinoma (Smith et al., 2019), although could still be considered in cases with metastatic disease.
Mesenteric LNs should be biopsied regardless of size or appearance, as the metastatic status of these nodes may hold prognostic significance
Median survival times (MSTs) range from 7.5 to 17.5 months for dogs with small intestinal adenocarcinoma treated with surgical excision (Smith et al., 2019; Paoloni et al., 2002). The prognostic significance of LN metastasis is unclear as one study demonstrated significantly shorter survival in cases with LN metastasis (3 months) compared to those without (15 months), but this has not been demonstrated in a more recent study (Smith et al., 2019; Crawshaw et al., 1998). The prognosis for adenocarcinoma located in the colorectum is better, with MSTs between 22 and 44 months reported with surgical excision (Church et al., 1987; Morello et al., 2008).
Leiomyosarcoma
Leiomyosarcomas are most frequently found in the stomach, although they can also occur in the small intestine, colon and rectum. They arise from gastrointestinal smooth muscle and must be differentiated from GISTs. Gastrointestinal stromal tumours can appear indistinguishable from leiomyosarcoma on histological examination. Instead, they arise from the interstitial cells of Cajal and do not demonstrate smooth muscle differentiation (Miettinen et al., 2002). Immunohistochemistry is required to differentiate the two. Leiomyosarcomas label positively for smooth muscle actin (SMA) and desmin and are negative for KIT. Gastrointestinal stromal tumours are positive for KIT and DOG1, label weakly for SMA and are negative for desmin. Both can label for vimentin (Hayes et al., 2013; Lopes et al., 2010). The metastatic risk of canine gastrointestinal leiomyosarcoma is generally low, with reported rates of 0 to 17 percent (Hayes et al., 2013; Del Alcazar et al., 2021; Russell et al., 2007).
Treatment and prognosis
Adjuvant doxorubicin-based chemotherapy could be considered in [leiomyosarcoma] cases with metastatic disease or other negative prognostic factors
The treatment of choice for leiomyosarcoma is surgical excision. Median survival times in more recent reports, where leiomyosarcomas are immunohistochemically differentiated from GISTs, are reported as between 7.8 and 18 months (Del Alcazar et al., 2021; Russell et al., 2007). However, incomplete excision is a negative prognostic factor (Del Alcazar et al., 2021; Russell et al., 2007). One study assessing intra-abdominal visceral soft tissue sarcomas identified a mitotic index greater than or equal to 9 and increasing tumour grade to be associated with shorter survival (Linden et al., 2019). Although there is limited literature assessing the use or benefit of chemotherapy, adjuvant doxorubicin-based chemotherapy could be considered in cases with metastatic disease or other negative prognostic factors.
Gastrointestinal stromal tumours
As already discussed, GISTs have a different cellular origin from leiomyosarcomas and can be differentiated via immunohistochemistry. Over 70 percent of GISTs demonstrate mutations in exon 11 of the c-Kit gene (Takanosu et al., 2016). They can occur anywhere within the gastrointestinal tract, although the small intestine and caecum are most common (Hayes et al., 2013; Russell et al., 2007). The metastatic rate for GISTs is reported as between 7 and 32 percent (Del Alcazar et al., 2021; Russell et al., 2007).
Treatment and prognosis
Surgical excision is the primary treatment modality for GISTs, and MSTs between 37.4 and 42.2 months are reported for dogs surviving the perioperative period (Del Alcazar et al., 2021; Russell et al., 2007). Figure 1 shows resection of a caecal GIST. Negative prognostic factors include incomplete excision, a mitotic index greater than nine and a weaker c-Kit staining intensity (Del Alcazar et al., 2021). As most GISTs harbour c-Kit mutations, treatment with tyrosine kinase inhibitors can be considered. Toceranib phosphate (Palladia) can be used in cases of incomplete excision, metastatic disease or non-surgical candidates; a clinical benefit rate of 71 percent has been reported in the gross disease setting (Berger et al., 2018).
Conclusion
Non-lymphomatous intestinal tumours are uncommon in dogs. The main differential diagnoses include adenocarcinoma, leiomyosarcoma and GIST. Leiomyosarcomas and GISTs can appear histologically similar and require immunohistochemistry to differentiate. Achieving a definitive diagnosis is important as treatment recommendations and prognosis differ.
