Primary brain tumors are a significant cause of morbidity and mortality in small animal companion animals, particularly in dogs. Although the true incidence of primary brain tumors in dogs is unknown, surveys of necropsy data suggest that the incidence is in the region of 10-20 per 100,000 animals, or 1-3% of all deaths where necropsy was done. For comparison, the incidence of primary malignant brain tumors in people is approximately 6-7 per 100,000. Primary nervous system tumors in dogs account for 60-80% of all such tumors reported in domestic animals (10-20% in cats; 10-20% in other species). Recent advances in imaging technology have transformed the discipline of clinical neurology in terms of the ability to accurately define the location and extent of neoplasia affecting the central and to a lesser degree, peripheral nervous system. Availability of advanced imaging modalities is expanding rapidly so that their use is no longer the reserve of specialist practitioners and university institutions.
Classification of tumors of the central nervous system in humans is based on the definition of presumed cell type of origin together with a grading system or “malignancy scale”. The basic classification is shown below. Not all of the tumor types described for humans have been documented in dogs, and the incidence of specific tumor subtypes and grades can vary considerably between the two species.Tumors of Neuroepithelial Origin
Meningiomas are one of the most common intracranial tumors of the central nervous system (CNS) in dogs comprising about 40% of all primary spontaneous tumors diagnosed at necropsy. Intracranial meningiomas are much more common than intraspinal meningiomas. The median age of affected dogs is around 9-10 years, with no consistent sex predisposition. The Boxer and Golden Retriever have been shown to be overrepresented, and it has been suggested that other dolicocephalic breeds may also be predisposed. The majority of tumors occur over the cerebral convexity or in the olfactory/frontal region. Multiple tumors, metastasis, paranasal, subcutaneous, intaraventricular and retrobulbar tumors are reported rarely.
Canine data. UC Davis VMTH
Oligodendrogliomas (II/III) represent < 10% of human gliomas, however the incidence appears to be much higher in dogs, possibly 30-40%.
Astrocytomas and oligodendrogliomas appear to have a similar rate of occurrence in dogs (incidence varies somewhat between studies). Brachycephalic dogs (Boxers, Bulldogs, Boston terriers) have a higher incidence of glial tumors, particularly oligodendrogliomas which are highly prevalent in the Boxer breed. There is no apparent sex predilection.The majority of tumors occur in animals > 6 years of age, however young animals (<3years) can be affected, and glial tumors can occur in animals as young as 6 months of age.
Both oligodendrogliomas and astrocytomas occur most frequently in the cerebral hemispheres. Oligodendrogliomas occur most commonly in the frontal/temporal lobes, while the lower grade astrocytomas appear to have a predilection for the temporal/piriform lobes. Similar to their human tumor counterparts, canine high grade oligodendrogliomas and astrocytomas commonly invade into the ventricular system, however migration across the corpus callosum (common in humans) is rarely seen
Approximately 50%of all gliomas (astrocytomas and oligodendrogliomas occur in brachycephalic (“squash nosed breeds) such as the Boxer, Boston Terrier and Bulldog!
MRI: The majority of canine oligodendrogliomas are intraaxial masses that are hypointense on T1Wimages and hyperintense on T2W images. Contrast enhancement is often peripherally ring enhancing, and more common with grade III than grade II tumors. Intratumoral calcification, commonly seen as areas of hypointensity in human oligodendrogliomas, is uncommon in canine tumors. Oligodendrogliomas often exhibit marked mass effect , are often relatively well demarcated and can involve ventricular structures.
Typical MR findings with a high grade oligodendroglioma. Ring enhancing, mass effect and ventricular involvement.
Most astrocytomas appear hyeprintense on T2W images and iso to hypointense on T1W images. Grade II tumors can have minimal mass effect and are usually non-contrast enhancing. Anaplastic and grade IV tumors tend to have more mass effect, may be ring enhancing or have patchy enhancement, and invasion of ventricles is common.
Typical low grade (II) diffuse astrocytoma in the temporal/pitriform lobes
Typical grade IV astrocytoma (GBM) with extensive mass effect and effacement of the lateral ventricle
Choroid plexus tumors
In the dog, choroids plexus tumors (CPTs) account for approximately 10% of all primary intracranial central nervous system (CNS) tumors. Most CPTs occur in middle aged dogs with an average age of six years (range 1 to 13 years). Golden Retrievers have been reported to be overrepresented, with no consistent sex predilection. CPTs arise from choroid plexus epithelium and therefore the primary mass is usually found in either the lateral, third, or fourth ventricle and/or in the lateral aperture at the cerebellomedullary pontine angle (CPA).
A. Choroid plexus papilloma (3rd ventricle) with typical uniform contrast enhancement and papilliform structure. Ventriculomegally is also apparent.
B. Choroid plexus carcinoma (3rd ventricle) with metastasis to the subarachnoid space and ventriculomegally.
C. Choroid plexus papilloma (3rd ventricle)with distal dilation of the 4th ventricle
Open brain biopsy: Open biopsy may be indicated when:
Stereotactic brain biopsy
For deep seated lesions, stereotactic procedures are the preferred biopsy technique. Essentially all closed methods rely on the three-dimensional CT (or MRI)-generated coordinates identifying the lesion location. These coordinates are used to plot the optimal trajectory and depth needed for a biopsy needle to reach a target and obtain a diagnostic tissue sample.
Technical impediments exist to the direct application of most human stereotactic systems to dogs and cats. Most commercially available systems use a cumbersome head-frame and localizing system, designed specifically for the human skull, and require dedicated, expensive computer software for the planning phase. Several different systems for image-guided stereotactic brain biopsy have been reported for use in dogs and cats.
Important considerations prior to biopsy procedures include.
In most cases, A small craniotomy (2-mm diameter) is made by means of a twist drill, the dura mater is punctured with an 18-gauge needle, and biopsies may be done with a side-cutting aspirator biopsy needle (Nashold Biopsy Needle, Integra Radionics, Burlington MA) with a 10-mm side opening. On average, two or three specimens are harvested. It is important to biopsy several regions of the intracranial lesion using a single trajectory. Intraoperative samples are assessed by smear techniques (air dried and alcohol fixed) to determine whether satisfactory tissue has been obtained or additional samples are indicated for culture etc. The majority of samples are processed for routine paraffin embedding. Small tissue samples necessitate the use of micromesh cassettes, and agar embedding of samples prior to processing is advisable for many cases to prevent loss of less well defined tissue.
Post procedure CT is essential to confirm that ongoing or significant bleeding has not occurred.
Diagnostic yield is approximately 95%, particularly with neoplastic lesions. Obtaining a specific diagnosis can be more problematic with inflammatory and infectious lesions where defining whether tissue is from the primary lesion or perilesional reactive tissue is sometimes challenging. Grading of tumors may also be difficult due to small sample size, usually resulting in under assessment of a tumor’s malignant potential.
In the majority of cases, analysis of CSF does not result in a definitive diagnosis. However this can be a self fulfilling prophecy! Careful assessment of cytological preparations by a clinical pathologist experienced in looking at CSF preparations critically can result in the identification of neoplastic cells in many cases. This is particularly true for choroids plexus tumors and hematopoetic tumors such as lymphoma and histiocytic sarcoma. Assessment of cytospin preparations using immunophenotyping and clonality assays is also valuable in determining the presence of neoplasia, particularly for lymphoma and histiocytic disease.
In general, conventional therapeutic approaches to brain tumors in people, and animals have involved a combination of surgical debulking/resection, chemotherapy, and radiation therapy. A large body of clinical data exists in human medicine pertaining to the relative efficacy of these therapies for specific tumors, together with the expected prognosis. Very little similar objective information is available for the dog, even relating to the normal progression of brain tumors in the absence of treatment. Small case study series, lack of ante mortem (or post mortem) diagnoses, differing treatment plans, the high degree of variability associated with an end point often defined by euthanasia, and variation in clinical severity at presentation have made the comparison of canine and human data very difficult.
In general, the majority of meningiomas in humans are treated surgically with a mortality rate of less than 10%. Recurring tumors (and aggressive tumors) may be treated with radiation and repeat surgical resection. Human gliomas, grades I-II, II and IV are generally treated with surgical resection followed by radiation and chemotherapy, particularly with high grade tumors The prognosis for these tumors varies from a 15 year survival rate of ~ 15% for Grade I,II tumors to a median survival of 4-16 months for high grade tumors. In fact, despite improvements in surgical techniques, radiation therapy (including radio surgery) and new chemotherapeutic agents, the prognosis for human patients with high grade gliomas has not altered significantly over the last 20 years!
Improved pre surgical imaging capabilities, together with improved surgical techniques and equipment are likely to lead to some significant improvements in prognosis for veterinary patients, particularly for readily accessible extra axial tumors.
Conventional chemotherapy has advanced very little in both the human and veterinary fields in the last 2 decades. Use of adjunctive chemotherapy such as hydroxyurea in the treatment of meningioma, either following surgery or following recurrence post surgery may be beneficial, however no objective data are available at this time. The only “new” chemotherapeutic agent to be approved for the treatment of human glioma in the last 2 decades is the alkylating agent temozolamide (Temodar), however clinical gains are small. It is unclear whether Temodar offers any significant advantages over standard (much cheaper!) alkylating agents such as CCNU in dogs with gliomas.
Although little data are available to make specific conclusions, radiation therapy is generally accepted to be the most useful adjunctive or sole therapy (where surgery is not possible), particularly for intraaxial tumors. Standard radiation treatment involves 15-20 fractionated doses of radiation over a 3-4 week treatment course, and significant expense. Advances in the ability to deliver radiation to tumors while sparing normal brain (eg intensity modulated radiation therapy IMRT) are likely to result in improved survival, and are becoming available at a limited number of veterinary institutions. More advanced stereotactic radiosurgical techniques such as the Gamma Knife and LINAC knife deliver very high doses of radiation to tumors in a single treatment while sparing normal tissues. These facilities are available to veterinarians at only a small number of research facilities, however this is likely to change in the next several years. Radiation involving a single treatment, and therefore a single anesthesia, has many potential benefits for veterinary patients. Preliminary studies in dogs with brain tumors suggested that single dose radiosurgery can be as effective as standard fractionated radiation treatment in selected tumors.
Molecular diagnostics and Targeted therapy
Over the past 15- 20 years, there has been a large effort to understand the specific molecular abnormalities underlying the development and progression of human primary brain tumors. Many of these abnormalities involve tumor suppressor genes, oncogenes and pathways involved in cell cycle regulation and angiogenesis. This has had an impact in two major ways.
There is little published data documenting the molecular characteristics of canine brain tumors, however several research groups are currently involved in work in several areas including expression and altered regulation of growth factor pathways; tumor suppressor gene function; telomerase activity, gene array expression profiling and chromosomal alterations. The release of the canine genome data will help enormously in promoting this basic research and help ensure that the veterinary profession is able to benefit from current and future advances in human brain tumor therapy, as well as potentially playing an integral part in both basic and applied clinical research.
Delivery of therapeutic agents
A wide variety of methods have been employed to deliver therapeutic agents to brain tumors. Many strategies have involved systemic delivery of agents either orally or intravenously. Some drugs (eg standard chemotherapeutic agents) are relatively non specific with respect to their potential targets, whereas others (eg small molecule tyrosine kinase inhibitors such as Gleevac) may have a more precisely defined target despite the systemic delivery. Even with the use of targeted therapies, the likelihood of significant systemic side effects is a major concern with drugs delivered in this manner. Ability of drugs to cross the blood brain barrier is also a factor that can significantly limit the efficacy of systemically delivered therapies, and many factors such as molecular weight, permeability of vasculature, drug stability and diffusion characteristics as well as tumor related factors are critical to attain effective cellular levels of anti tumor drugs. Local “targeted” delivery of therapies directly into tumor tissue has been advocated as a way to increase both the efficacy of many therapeutic agents whilst at the same time decreasing the likelihood of significant systemic toxicity. Therapeutic agents may be delivered directly at surgery following excision/debulking of tumors, or by stereotactic injection. Recent advances in injection of agents by convection enhanced delivery (CED) (over several hours), have shown great promise, and may allow highly accurate and comprehensive delivery of therapeutics to a defined area of tumor and/or surrounding brain. Preliminary results with CED using a novel chemotherapeutic agent CPT-11 (topoisomerase I inhibitor) in an ongoing clinical trial in dogs with spontaneous gliomas are encouraging, and demonstrate the feasibility of targeted delivery in veterinary patients.