SEARCH

SEARCH BY CITATION

Keywords:

  • Canine;
  • Cerebrospinal fluid;
  • Choroid plexus carcinoma;
  • Choroid plexus papilloma;
  • Magnetic resonance imaging;
  • WHO tumor classification

Abstract

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Footnotes
  7. Acknowledgment
  8. References

Background: Choroid plexus tumors (CPTs) comprise approximately 10% of all primary brain tumors in dogs. The clinical utility of magnetic resonance imaging (MRI), cerebrospinal fluid (CSF) analysis, or both in the presumptive diagnosis of CPTs has not been determined.

Objectives: To report MRI and CSF findings in dogs with CPT and determine if there are distinguishing features that allow clinical discrimination between the tumor grades.

Animals: Fifty-six client-owned dogs with naturally occurring CPT.

Methods: Retrospective case series. The inclusion criterion was histologically confirmed CPT. Blinded review of cranial MRI and cisternal CSF analysis was performed.

Results: Thirty-six of 56 dogs had a choroid plexus carcinoma (CPC) and 20 had a choroid plexus papilloma (CPP). Golden Retrievers were overrepresented compared with the hospital population (frequency 3.7 times that expected, confidence interval 95%= 2.0–6.7, P < .0002). Median CSF protein concentration in CPCs (108 mg/dL, range 27–380 mg/dL) was significantly higher than in CPPs (34 mg/dL, range 32–80 mg/dL) (P= .002). Only dogs with CPCs had a CSF protein concentration >80 mg/dL. Cytological evidence of malignancy in CSF was seen in 7 of 15 CPCs. Only CPCs had evidence of intraventricular or subarachnoid metastases on MRI.

Conclusions and Clinical Importance: MRI, CSF analysis or both can help to differentiate between CPPs and CPCs, and may provide valuable prognostic and pretreatment information.

Choroid plexus tumors (CPTs) in dogs account for approximately 10% of all primary intracranial central nervous system (CNS) tumors.1 Since the first report of a choroid plexus papilloma (CPP) in 1936,2 there have been 15 case reports3–17 and 10 case series (n ≤ 16/series)18–27 with a combined total of 98 histologically confirmed cases of CPT where location within the CNS was documented.

Most CPTs occur in middle-aged dogs with an average age at diagnosis of 6 years (range 1–13 years).3,4,6–17,20,22,25–27 There is no reported breed predilection. Male to female ratio varies from 1 : 1 (n = 10)27 to all males in one case series (n = 9).20 Of 55/98 cases in which sex was documented, 64% were males and 36% females.2,4,7–11,13–17,20,22,25,27

CPTs arise from choroid plexus epithelium and the primary mass is usually found in either the lateral, 3rd, and 4th ventricles or the lateral apertures. Of 98 dogs with CPTs, 29% occurred in the lateral ventricle, 22% in the 3rd ventricle, and 49% in the 4th ventricle (Table 1). In 2 dogs, choroid plexus carcinoma (CPC) was disseminated intracranially and the primary site was not identified.11,17 Local microscopic spread within the ventricular system can occur with CPPs22 and distant metastases to the subarachnoid space occur in up to half of the dogs with CPCs.4,8,11,14,17,20,25 There are no reports of extraneural metastases in the dog and it is rare in people.28

Table 1.   Primary CPT ventricular location from the literaturea: CPP (54) versus CPC (13).
CPT GradeVentricular Site
4th3rdLateralTotals
  • a

    Grade presented if documented2,3,5–10,22,24,25 and 2/98 not included as disseminated and primary site not reported.11,17

  • CPC, choroid plexus carcinoma; CPP, choroid plexus papilloma; CPT, choroid plexus tumors.

CPP27 (50%)10 (19%)17 (33%)54 (81%)
CPC9 (69%)3 (23%)1 (8%)13 (19%)
All CPTs47 (49%)21 (22%)28 (29%)96 (100%)

CPTs in people are classified into 2 major grades according to the WHO histological classification system as either Grade I CPP or Grade III CPC, with a much less common Grade II atypical CPP added in 2007.29 The clinical relevance of grading is to predict the biological behavior of the tumor for informed decisions on prognosis and therapeutic planning. Clinical experience using the WHO classification system in people with CPTs demonstrates that tumor grade is the most important prognostic factor and correlates closely with overall survival after surgery.30,31 In a large meta-analysis after surgical removal of CPTs in 857 people, the 5 year overall survival was 87% for patients with CPP, in contrast to only 34% for patients with CPC.31 Several classification schemes have been applied for dogs with CPT.22,32 Of the 98 dogs with CPTs reviewed, 60% were classified as CPPs, 16% CPCs, 19% CPTs without further differentiation, and 5% as atypical CPPs (single case series22) using a variety of classification systems.

Definitive diagnosis of CPT is based on histopathology. Both magnetic resonance imaging (MRI) and cerebrospinal fluid (CSF) analyses are commonly used in the presumptive clinical diagnosis of brain tumors in dogs. A high index of suspicion for a CPT should be considered based on MRI findings of an intraventricular mass at or adjacent to anatomical site(s) of choroid plexus. MRI characteristics typical of CPT have been reported individually or in case series, for 23 cases.15–17,23,24,26 Common findings include intraventricular mass(es) with intense contrast enhancement and ventriculomegaly. CSF analysis in 32 histologically confirmed CPTs15,17,26,33,34 suggest a common finding of markedly increased protein concentration, with a mean protein concentration of 149–157 mg/dL in the 3 largest case series.26,33,34 There are no reports of a definitive diagnosis from cytological examination of CSF in dogs with CPTs. Recently, it has been demonstrated in dogs that a rapid intraoperative diagnosis of CPT can be made from CT-guided stereotactic biopsy samples.35

Treatment options for CPTs in dogs are limited and prognostic data based on either the tumors' behavior or response to treatment is limited. Based on human experience, it is likely that tumor grade is an important factor as management of these tumors advances.31 The aims of this study were to review the findings of MRI and CSF analyses in a large series of canine CPTs and determine if there are distinguishing features that could aid in clinical discrimination between CPCs and CPPs after a presumptive diagnosis of a CPT, and finally to compare the findings of this study with previous reports.

Materials and Methods

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Footnotes
  7. Acknowledgment
  8. References

Animals

This retrospective study was carried out by a computerized database search for dogs examined at the William R. Pritchard Veterinary Medical Teaching Hospital (VMTH), University of California, Davis from 1985 to 2007. The single criterion for inclusion in the study was a histological diagnosis of CPT.

MRI

MRI studies of the head were performed with either a 0.34 Ta or a 1.5 Tb system. A standard spin echo pulse sequence protocol was used and included: noncontiguous 4–5-mm-thick slices with a 0.5 mm interslice gap in the transverse and sagittal plane for T1 weighted (T1W) pre and postcontrast images and in the transverse plane for T2 weighted (T2W) images.36 Additionally, fluid-attenuated inversion recovery (FLAIR) sequences in the transverse plane were performed in all cases imaged after 2004.36 Blinded review of the images was carried out (DRW, EGJ). Interpretation was based on signal intensity of the majority of the primary tumor area and final categorization was from a consensus of opinion. For each primary CPT, the signal intensity relative to cortical gray matter on both the precontrast T1W and T2W imaging sequences was graded as hypointense, isointense, or hyperintense. Signal uniformity was categorized as homogenous or heterogenous. On postcontrast T1W images, enhancement was categorized as present and intense, weak, or absent, and signal uniformity was categorized as either heterogenous or homogenous. The extent of the mass was defined by that area of the image that enhanced after IV contrast administration.c In cases with limited or no enhancement, reasonable estimates of extent took into account spatial changes in relation to normal anatomy and T1W and T2W parenchymal intensity patterns. Apparent peritumoral edema was defined as parenchymal hyperintensity on T2W or FLAIR imaging sequences adjacent to the boundary of only intensely enhancing masses and categorized as either present or absent. When present on T2W images, edema was graded as mild, if it extended ≤10 mm, or moderate marked if it extended >10 mm beyond the tumor margin. The shape of the tumor was interpreted as papilliform, globular, or multiform and irregular. Intracranial metastases were either present or absent and defined as additional intracranial contrast enhancing masses within the ventricular system or subarachnoid space. Ventriculomegaly was categorized subjectively as either present or absent. Periventricular edema was either present or absent and defined as parenchymal hyperintensity adjacent to the ventricular system only on FLAIR images.

CSF

CSF was analyzed as described previously.37 CSF results were only included if complete analysis was available including gross appearance, total protein concentration, red blood cell counts and total nucleated cell counts (TNCC), differential nucleated cell counts, and cytological evaluation. Blinded examination of archived cytological preparations for neoplastic-appearing choroid plexus cells was carried out (WV). Morphological features of malignancy were defined as cells with a round darkly staining nucleus, prominent nucleoli, a high nucleus to cytoplasmic (N/C) ratio, and vacuolated basophilic cytoplasm often with multiple membranous blebs or protrusions.

Histological Classification and Tumor Grading

Consensus histological classification and grading by 2 neuropathologists (RJH, AWB) was carried out on select 5-μm-thick hematoxylin and eosin-stained sections according to the current criteria for the human International WHO histological grading of tumors of the CNS.29 Criteria for CPPs (Grade I) were histological features of normal choroid plexus with minimal histological change with or without minimal local brain invasion, and <2 mitotic figures per 10 high-powered fields (HPF, 400 × magnification with an area of 0.16 mm2). Criteria for atypical CPPs (Grade II) were ≥2 mitoses per 10 HPF, and may also include up to 2 of the following features: increased cellularity, nuclear atypia, loss of papillary pattern, and areas of necrosis, but these features are not required to make a diagnosis of atypical CPP. Criteria for CPCs (Grade III) included at least 4 of the following histological features: ≥5 mitotic figures per 10 HPF, nuclear atypia, multilayering of epithelium, increased cell density, loss of papillary pattern with solid cell sheets, and/or multifocal areas of necrosis. Local aggressive brain invasion is common but is not currently a criterion. At necropsy, the brain and spinal cord were also evaluated for both distant intraventricular spread and for subarachnoid metastases from the primary tumor site.

Statistical Analyses

χ2-Tests of homogeneity were used to compare the breed distributions of dogs in this study to the population of dogs examined at the VMTH during the same study period. The sex and median age of dogs having either a CPP or CPC were compared by a Student's 2-group t-test. Wilcoxon-Mann-Whitney tests were used to compare the distribution of protein concentrations and nucleated cell counts in CSF between CPP and CPC dogs. P-values ≤.05 were considered statistically significant. Descriptive statistics for CSF protein concentration were calculated including sensitivity and specificity with a 95% confidence interval (CI 95%).

Results

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Footnotes
  7. Acknowledgment
  8. References

Animals

Fifty-six dogs, 28 females and 28 males, met the selection criteria for inclusion in the study. There were 20 CPPs in 12 female (none intact) and 8 male (4 intact) dogs, and 36 CPCs in 16 female (1 intact) and 20 male (4 intact) dogs. The sex distribution was not significantly different between tumor types (P= .40). Median age of the dogs with CPP was 5 years (range 3–14 years) and median age of the dogs with CPC was 7 years (range 3–12 years) (P= .015). The most commonly affected breed was the Golden Retriever, 11 of 56 (20%), in which CPTs were found 3.7 times more often than expected (CI 95%= 2.0–6.7, P < .0002) based on the number of dogs admitted to the VMTH during the study period (5.3% of 120,978 cases). No other breed had a substantially higher frequency of CPTs than expected.

MRI

MRI of the brain was performed in 32 dogs (9 CPPs and 23 CPCs). Images were acquired in 15 dogs with a 0.34 Ta system and in 17 dogs with a 1.5 Tb system. The important MRI features included the gross papilliform structure of the mass in 33% (3 of 9) of CPPs (Fig 1) but in none of the CPCs, multiple intraventricular or subarachnoid contrast enhancing masses consistent with metastases in 35% (8 of 23) of CPCs (Fig 2) but in none of the CPPs, and peritumoral edema seen in both tumor grades (Fig 3) (Table 2).

image

Figure 1.  Choroid plexus papilloma: magnetic resonance imaging T1W postcontrast, transverse section. A uniformly contrast-enhancing 3rd ventricular mass with gross papilliform structure seen in a subset of these Grade I tumors is shown.

Download figure to PowerPoint

image

Figure 2.  Choroid plexus carcinoma: magnetic resonance imaging T1W postcontrast, paramedian section. A large contrast-enhancing globular 4th ventricular mass is shown with an additional smaller contrast enhancing subarachnoid mass (arrow) on the ventral surface of the brainstem, consistent with a metastatic lesion.

Download figure to PowerPoint

image

Figure 3.  Choroid plexus carcinoma: magnetic resonance imaging fluid attenuated inversion recovery, transverse section. Marked hyperintensity (multiple arrow heads) predominantly located in peritumoral white matter is present, and is consistent with vasogenic edema secondary to the invasive 3rd ventricular mass (arrow).

Download figure to PowerPoint

Table 2.   Intracranial MRI features in CPPs and CPCs.
 CPP (9)CPC (23)
  1. CPC, choroid plexus carcinoma; CPP, choroid plexus papilloma; MRI, magnetic resonance imaging; FLAIR, fluid-attenuated inversion recovery; T1W, T1 weighted; T2W, T2 weighted.

T1W signal intensity
 Hypointense3 (33%)3 (13%)
 Isointense3 (33%)8 (35%)
 Hyperintense3 (33%)12 (52%)
T1W signal uniformity
 Homogenous5 (46%)14 (61%)
 Heterogenous4 (44%)9 (39%)
Post-contrast T1W signal intensity
 Intense9 (100%)22 (96%)
 No enhancement01 (4%)
Post-contrast T1W signal uniformity
 Homogenous6 (67%)18 (78%)
 Heterogenous3 (33%)5 (22%)
T2W signal intensity
 Isointense1 (11%)3 (13%)
 Hyperintense8 (89%)20 (87%)
T2W signal uniformity
 Homogenous6 (87%)15 (65%)
 Heterogenous3 (13%)8 (35%)
Perilesional edema: present4/9 (44%)16/23 (69%)
 Mild2/4 (50%)7/16 (44%)
 Moderate-marked2/4 (50%)9/16 (56%)
 Absent5/9 (56%)7/23 (31%)
Shape
 Papilliform3 (33%)0
 Globular6 (66%)22 (96%)
 Multiform/linear01 (4%)
Apparent metastasis08 (35%)
Ventriculomegaly: present7/9 (78%)17/23 (74%)
 Rostral to main mass2/7 (29%)11/17 (65%)
 Rostral and caudal5/7 (71%)6/17 (35%)
 Absent2/9 (22%)6/23 (26%)
Periventricular edema (16/32 FLAIR)
 Present6/6 (100%)7/10 (70%)
 Absent03/10 (30%)

In 75% (24 of 32) CPTs there was ventriculomegaly, beyond that expected for the breed. Half of the CPTs located rostrotentorially (lateral or 3rd ventricle) had ventriculomegaly only rostral to the main mass, while the remainder had ventriculomegaly extending caudally through the 4th ventricle, the lateral aperture or both (Fig 4). Infratentorial CPTs were associated with ventriculomegaly of the entire ventricular system in 4 of 10 dogs, while 6 of 10 dogs had dilatation rostral to the main mass. There was hydromyelia in 3 CPCs and 2 CPPs.

image

Figure 4.  Choroid plexus carcinoma: magnetic resonance imaging T1W, postcontrast median section. A markedly enlarged 4th ventricle (arrow) caudal to a single conspicuous contrast-enhancing 3rd ventricular mass is shown. No metastatic obstructive mass lesions were present caudally.

Download figure to PowerPoint

In 16 dogs, a FLAIR sequence was obtained. In all 6 CPPs and in 7 of 10 CPCs, there was periventricular hyperintensity consistent with edema. This was always bilateral (with or without asymmetry) at the level of the cavity of the olfactory bulb (the most rostral extension of the rostral horn of the lateral ventricle) or dorsolateral margins of the lateral ventricles and was separate from the contrast-enhancing mass(es). Subjectively, there was no apparent difference in severity for either tumor grade. In addition, 1 or more intratumoral cystic structures (2–6 mm diameter) were seen in 5 of 32 (16%) CPTs, comprising 2 CPPs and 3 CPCs.

CSF

Twenty-five CSF samples were examined from 7 dogs with CPPs and 18 dogs with CPCs. All samples had increased protein concentration (reference range <25 mg/dL).37 The median and range of CSF protein concentration was as follows for all CPTs, CPPs, and CPCs, respectively: 80 mg/dL (27–380 mg/dL), 34 mg/dL (32–80 mg/dL), and 108 mg/dL (27–380 mg/dL). The median protein concentration was significantly higher in CPCs than in CPPs (P= .002). A protein concentration >80 mg/dL was only seen in CPCs. Using the maximum CSF value of 80 mg/dL for CPPs as a cutoff for distinguishing between the 2 tumor types yielded a sensitivity for diagnosing CPCs of 0.67 (CI 95%= 0.41–0.87) and a specificity of 1.00 (CI 95%= 0.59–1.00).

All CPPs and 50% of the CPCs had an elevated TNCC (reference range <5 cells/μL). Median and range for CPPs and CPCs was 13 cells/μL (5–66 cells/μL) and 6 cells/μL (1–260 cells/μL), respectively. Both tumor grades often had a mixed pleocytosis with a variable nucleated cell distribution. There was no significant difference in nucleated cell counts between the CPCs and the CPPs (P= .26). Cytological reevaluation of 21 archived CSF samples revealed that 7 of 15 (47%) CPCs had CSF that contained single cells, or rows, or clusters of cells with morphological features of neoplastic choroid plexus epithelium (Fig 5a–c).

image

Figure 5.  Cerebrospinal fluid cytology from choroid plexus carcinomas. (a) Single neoplastic tumor cell with large round nucleus and abundant basophilic cytoplasm with some cytoplasmic blebbing and a normal macrophage. Scale bar = 12 μm. (b) Row of neoplastic epithelial cells with an adherent macrophage. Scale bar = 10 μm. (c) Cluster of neoplastic epithelial cells, round nuclei with high N/C ratio with basophilic cytoplasm sometimes with blebs. Scale bar = 12 μm. Cytocentrifuge preparations. Wright's stain.

Download figure to PowerPoint

Histological Classification and Tumor Grading

Histopathological data from the 56 tumors were obtained from 50 necropsies (18 CPP, 32 CPC), 6 surgical excisional biopsies (2 CPP, 3 CPC), and 1 CT-guided stereotactic biopsy procedure (CPC). There were 20 (36%) CPPs and 36 (64%) CPCs. None met the criteria for atypical CPPs. The location of the CPPs is summarized in Table 3. For the CPPs (Fig 6), the mitotic index varied from 0 to 1 per 10 HPF. In addition, 1 CPP had nuclear atypia and multilayering, another had foci of necrosis, and 2 others had areas of loss of papillary differentiation, with no other abnormal histological features. Some degree of mineralization occurred in 9 (47%) CPPs and minimal local brain invasion adjacent to the tumor was seen in 5 cases (25%). Only 1 CPP had a single small subarachnoid metastasis. Disseminated intraventricular or subarachnoid spread was not detected in any other CPP.

Table 3.   Primary CPT ventricular location: CPP (20) versus CPC (36).
CPT GradeVentricular Site
4th3rdLateralTotals
  1. CPC, choroid plexus carcinoma; CPP, choroid plexus papilloma; CPT, choroid plexus tumors.

CPP8 (40%)9 (45%)3 (15%)20 (36%)
CPC18 (50%)11 (31%)7 (19%)36 (64%)
All CPTs26 (46%)20 (36%)10 (18%)56 (100%)
image

Figure 6.  Brain: choroid plexus papilloma Grade I. Note architectural resemblance to normal choroid plexus histology although papilliform structure attenuated. Hematoxylin and eosin stain. Scale bar = 70 μm.

Download figure to PowerPoint

The location of the primary tumor for the CPCs is summarized in Table 3. The mitotic index ranged from 1 to 18 per 10 HPF with a mean of 4 per 10 HPF. There were abnormal mitotic figures in 8 (22%), nuclear atypia in 34 (94%) (Fig 7), focal areas of necrosis in 24 (67%), increased cell density in 27 (75%), loss of papillary pattern in 35 (97%) (Fig 8), and cellular multilayering in 30 (83%) (Fig 9). Mineralization of varying severity was seen in 17 cases (50%). Of 28 necropsied dogs with CPC, 26 (92%) had aggressive local brain invasion at the primary tumor site (Fig 10). In 14 (53%) of these locally invasive tumors, there was also extraventricular spread to the subarachnoid space of the brain and/or spinal cord (Fig 11). Intraventricular spread distant to the site of the tumor was detected in 16 (57%). Seven cases (19%) had spinal cord involvement.

image

Figure 7.  Brain: choroid plexus carcinoma Grade III. Marked nuclear atypia, mitotic figures loss of normal epithelial cell orientation. Hematoxylin and eosin stain. Scale bar = 28 μm.

Download figure to PowerPoint

image

Figure 8.  Brain: choroid plexus carcinoma Grade III. There is marked loss of papillary structures replaced with sheets of neoplastic cells. Hematoxylin and eosin stain. Scale bar = 45 μm.

Download figure to PowerPoint

image

Figure 9.  Brain: choroid plexus carcinoma Grade III. Marked piling up and disorganization of neoplastic cells with nuclear atypia and abnormal mitotic figures. Hematoxylin and eosin stain. Scale bar = 15 μm.

Download figure to PowerPoint

image

Figure 10.  Brain: choroid plexus carcinoma Grade III. Aggressive invasion of brain parenchyma adjacent to the intraventricular tumor. Hematoxylin and eosin stain. Scale bar = 180 μm.

Download figure to PowerPoint

image

Figure 11.  Spinal cord: choroid plexus carcinoma Grade III. Multiple metastatic tumor sites in the subarachnoid space of the spinal cord from the primary tumor located in the 4th ventricle. Hematoxylin and eosin stain. Scale bar = 200 μm.

Download figure to PowerPoint

Discussion

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Footnotes
  7. Acknowledgment
  8. References

Findings of this retrospective study demonstrate that different histological grades of CPT correlate with distinct differences in both CSF analysis and MRI characteristics. CPCs had a substantially higher median CSF protein concentration than CPPs whereas a CSF protein concentration >80 mg/dL occurred only in CPCs. Of novel diagnostic importance was the cytological evidence of malignancy identified in the CSF in 45% of dogs with CPCs. Subarachnoid or distant intraventricular metastasis was restricted to CPCs and were detected with MRI in a third of cases.

We chose to classify and grade these tumors by means of the current criteria established by the human international WHO histological classification system as either Grade I CPPs, Grade II atypical CPPs, or Grade III CPCs.29 This decision was based on the striking similarities between canine and human CPTs and extends the previous approach of the 1999 WHO histological classification of tumors of the nervous system of domestic animals,32 which was based on the 1993 human international WHO classification system,38 the latter since updated 3 times. As yet, no prospective data are available to validate a classification system in dogs, in contrast to that well demonstrated in people. The histological grading system used in this study demonstrated the predictive accuracy of either distant intraventricular spread or subarachnoid metastases for CPC. In addition, aggressive brain invasion occurred only in CPCs (92%), as is also reported in people.29 However, parenchymal invasion is not a criterion for the human WHO grading system for CPTs. Prospective multicenter studies will be necessary to provide data allowing validation and modification of the proposed classification system based on clinical outcome.

CPTs occur most commonly in middle-aged dogs and dogs with CPCs were substantially older than those with CPPs. There was an equal male to female ratio for CPTs in this case series as has been reported27; however, overall, the previous literature indicates a 2 : 1 male predominance. Breed predisposition has not been previously recognized, but we found Golden Retrievers to be 3.7 times more likely to be diagnosed with CPT than that expected from the VMTH case population breed distribution.

The MRI findings in this study are similar to those described in both canine and human CPTs.15–17,23,24,26,27,39,40 CPTs are predominantly iso-hyperintense on both T1W and T2W images whereas contrast enhancement of the mass(es) on T1W images is generally intense and homogenous. Signal heterogeneity is reported on MRI of CPTs of all grades with associated histopathological features including cyst formation,17 mineralization, hemorrhage, increased vascularity, necrosis, and even osseous or adipose metaplasia.40 In this study, the causes of MRI signal heterogeneities were not established, primarily because it was not possible retrospectively to directly compare MRI and histopathological sections.

In our study, the most important MRI finding to assist in clinical discrimination between the tumor grades were intraventricular or subarachnoid metastases detected in 35% of CPCs. Evidence of metastasis within the CNS on MRI is seen in approximately 45% of humans with CPC and is associated with a poor prognosis.40 Although evidence of CNS metastasis on MRI has been documented in people with CPPs, this is very rare41 and has not been reported in dogs. Thus, MRI detection of CNS metastases may help to distinguish CPCs from CPPs and provide valuable prognostic information. Furthermore, MRI of the spinal cord may be indicated based on the fact that at necropsy 19% of CPCs had evidence (gross, microscopic, or both) of spinal cord metasteses.

Two other MRI findings that might help distinguish between the tumor grades (although there was much overlap) were perilesional edema and tumor shape. CPCs more commonly had perilesional edema (69 versus 44%) but this was not statistically significant. In people, up to 73% of CPCs have intensity changes consistent with apparent edema on MRI.40 Although most CPTs had a globular shape on MRI, 33% of CPPs had a grossly papilliform shape, which was not seen in any CPCs. Loss of gross papilliform pattern is also recognized in people as a possible distinguishing feature between CPPs and CPCs.42 Other MRI findings that may assist in differentiating CPCs from CPPs in people, but were not clearly identified in this case series, include decreased tumor enhancement on postcontrast T1W images, lack of defined tumor margins with or without apparent parenchymal invasion, and increased vascular flow voids consistent with increased tumor vascularity.40,42

Ventriculomegaly, documented by MRI, occurred in 75% of dogs in this study. Ventriculomegaly is difficult to define objectively and no definitive criteria have been established in the dog, although attempts have been made to define normal canine lateral ventricle size with both ultrasound43,44 and MRI.45–48 Interestingly, in this study, ventriculomegaly extended caudal to the mass in 46% of cases and was occasionally associated with marked hydromelia. Additionally, on FLAIR sequences, 13 of 16 (18%) CPTs had apparent periventricular edema consistent with increased pressure within the ventricular system and subependymal leakage of CSF.49 In none of these dogs at necropsy was there obstruction because of another more caudal mass. The exact cause of this caudal ventriculomegaly was not determined but it does not appear to be purely obstructive. In people with CPTs, hydrocephalus is very common and is considered to occur principally because of obstruction of CSF pathways and/or poor CSF resorption.42 Tumor mass, blood products, tumor products, metastases, ependymitis, arachnoiditis, and protein deposition all probably contribute to reduced absorption and therefore accumulation of CSF.42 In some cases, overproduction of CSF may occur, although this is not commonly documented.42,50

All 25 CSF samples had increased protein concentrations irrespective of the histological grade of the tumor. Increased CSF protein concentration is a common finding in CPTs in dogs7,15,17,26,33,34 and is usually higher than for other canine brain tumor types despite much individual overlap.26,33,34 Results from 3 previous studies of CPT in dogs with the highest numbers of CSF samples (n = 30) were compared with those from this study.26,33,34 In all but one previous case, the total CSF protein concentration was increased (mean of 149–157 mg/dL),26,33,34 with an individual maximum range to 356 mg/dL,35 which is similar to the findings of this study. The increased CSF protein concentration is considered to be primarily because of both altered endothelial cell permeability and integrity of the basement membrane.51 In this series, the median protein concentration in CPCs was significantly higher than in CPPs (P= .002) and only CPCs had a protein concentration >80 mg/dL. This difference has not been previously recognized in CPTs in either dogs or people. Thus, measurement of CSF protein concentration may be a useful diagnostic tool and may also guide clinical differentiation between the major tumor grades.

The ability to diagnose a CPT in a dog by CSF cytological analysis has not been reported previously. Archived CSF cytospin slides from 47% dogs with CPC contained neoplastic cells with characteristic morphological features observed in human CPTs.52 These 7 tumors histopathologically had disseminated intraventricular spread while 5 also had subarachnoid metastases. In people, a diagnosis from cytological evaluation of CSF is commonly made for various tumors that involve the leptomeninges or ventricular system (eg, lymphoid tumors, medulloblastomas, metastatic carcinomas, and CPTs).53,54 Factors influencing the likelihood of tumor cell identification in CSF include observer expertise, the tumor type and capacity to exfoliate, the site and frequency of sampling, and method of cytological concentration.55,56

Footnotes

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Footnotes
  7. Acknowledgment
  8. References

a0.3 T Signa, General Electric Medical Systems, Milwaukee, WI

b1.5 T Signa, General Electric Medical Systems

cMagnevist, 469.01 mg gadopentatate/mL, 0.1 mmol/kg dose, Berlex Laboratories, Wayne, NJ

Acknowledgment

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Footnotes
  7. Acknowledgment
  8. References

We thank John Doval for assistance with the illustrations.

References

  1. Top of page
  2. Abstract
  3. Materials and Methods
  4. Results
  5. Discussion
  6. Footnotes
  7. Acknowledgment
  8. References
  • 1
    Koestner A, Higgins RJ. Tumors of the nervous system. In: MeutenDJ, ed. Tumors of Domestic Animals, 4th ed. Ames, IA: Iowa State Press; 2002:709712.
  • 2
    Milks HJ, Olafson P. Primary brain tumors in small animals. Cornell Vet 1936;26:159170.
  • 3
    Cotchin E. Primary carcinoma of the choroid plexus of the fourth ventricle in a dog. J Pathol Bacteriol 1953;65:257258.
  • 4
    McGrath JT. Neurologic Examination of the Dog with Clinico-Pathologic Observations, 2nd ed. Philadelphia, PA: Lea and Febiger; 1960:164165, 182pp.
  • 5
    Nashold BS, Gills JP, Moore GT. Papilloma of choroid plexus in a Weimaraner dog. J Neurosurg 1961;18:250251.
  • 6
    Loomis LN, Keep JM. Choroid plexus papilloma in a dog. Br Vet J 1964;120:336339.
  • 7
    Kurtz HJ, Hanlon GF. Choroid plexus papilloma in a dog. Vet Pathol 1971;8:9195.
  • 8
    Zaki FA, Kay WJ. Carcinoma of the choroid plexus in a dog. J Am Vet Med Assoc 1971;164:11951197.
  • 9
    Palmer AC, Malonowski W, Barnett KC. Clinical signs including papilloedema associated with brain tumors in twenty-one dogs. J Small Anim Pract 1974;15:359386.
  • 10
    Indrieri RJ, Holliday TA, Selcer RR, et al. Choroid plexus papilloma associated with prolonged signs of vestibular dysfunction in a young dog. J Am Anim Hosp Assoc 1980;16:263268.
  • 11
    Patnaik AK, Erlandson RA, Lieberman PH, et al. Choroid plexus carcinoma with meningeal carcinomatosis in a dog. Vet Pathol 1980;17:381385.
  • 12
    Whitbread TJ, Skerritt GC. A case of choroid plexus papilloma in a dog. Anatomy histology embryology. Rep World Assoc Vet Anatomists 1980;9:372.
  • 13
    Cheiner M, Gosselin Y, Teuscher E, et al. Paradoxic vestibular syndrome associated with choroid plexus papilloma in a dog. J Am Vet Med Assoc 1983;182:6667.
  • 14
    Wilson RB, Holscher MA, West WR. Choroid plexus carcinoma in a dog. J Comp Pathol 1989;100:323326.
  • 15
    Hammer AS, Couto CG, Getzy D, et al. Magnetic resonance imaging in a dog with choroid plexus carcinoma. J Small Anim Pract 1990;31:341344.
  • 16
    Ohashi F, Kotani T, Onishi T, et al. Magnetic resonance imaging in a dog with choroid plexus carcinoma. J Vet Med Sci 1993;55:875876.
  • 17
    Lipsitz D, Levitski RE, Chauvet AE. Magnetic resonance imaging of a choroid plexus carcinoma and meningeal carcinomatosis in a dog. Vet Radiol Ultrasound 1999;40:246250.
  • 18
    Verwer MAJ, Van Den Akker S. Intracranial tumors in dogs. Proc XVIth Int Vet Congr 1959;289291.
  • 19
    Luginbühl H, Fankhauser R, McGrath JT. Spontaneous neoplasms of the nervous system in animals. In: KrayenbuhlH, MaspesP, SweetW, eds. Progress in Neurological Surgery, Vol. 2. Chicago, IL: Karger, Basel; 1968:107111.
  • 20
    Zaki FA, Nafe LA. Choroid plexus tumors in the dog. J Am Vet Med Assoc 1980;176:328330.
  • 21
    Turrel JM, Fike JR, LeCouteur RA, et al. Computed tomography characteristics of primary brain tumors in 50 dogs. J Am Vet Med Assoc 1986;188:851856.
  • 22
    Ribas JL, Mena H, Braund KG, et al. A histologic and immunocytochemical study of choroid plexus tumors of the dog. Vet Pathol 1989;26:5564.
  • 23
    Thomas WB, Wheeler SJ, Kramer R, et al. Magnetic resonance imaging features of primary brain tumors in dogs. Vet Radiol Ultrasound 1996;37:2027.
  • 24
    Kraft SL, Gavin PR, DeHann C, et al. Retrospective review of 50 canine intracranial tumors evaluated by magnetic resonance imaging. J Vet Intern Med 1997;11:218225.
  • 25
    Cantile C, Campani D, Menicagli M, et al. Pathological and immunohistochemical studies of choroid plexus carcinoma of the dog. J Comp Path 2002;126:183193.
  • 26
    Snyder JM, Shofer FS, Van Winkle TJ, et al. Canine intracranial primary neoplasia: 173 cases (1986–2003). J Vet Intern Med 2006;20:669675.
  • 27
    Thankey K, Faissler D, Kavirayani A, et al. Clinical presentation and outcome in dogs with histologically confirmed choroid plexus papillomas. Proc ACVIM Forum 2006; Abstract 266:782783.
  • 28
    Valladares JB, Perry RH, Kalbag RM. Malignant choroid plexus papilloma with extraneural metastasis. J Neurosurg 1980;52:251255.
  • 29
    Aguzzi A, Brandner S, Paulus W. Choroid plexus tumors. In: LouisDN, OhgakiH, WiestlerOD, CaveneeWK, eds. World Health Organization Classification of Tumours: Pathology and Genetics of Tumours of the Nervous System. Lyon, France: IARC Press; 2007:8486.
  • 30
    Wolff JEA, Sajedi M, Brant R, et al. Choroid plexus tumors. Br J Cancer 2002;87:10861091.
  • 31
    Wrede B, Liu P, Ater J, Wolff JEA. Chemotherapy improves the survival of patients with choroid plexus carcinoma: A meta-analysis of individual cases with choroid plexus tumors. J Neurooncol 2007;85:345351.
  • 32
    Koestner A, Bilzer T, Fatzer R, et al. Histological Classification of Tumors of the Nervous System of Domestic Animals. Washington, DC: AFIP; 1999:2324.
  • 33
    Bailey CS, Higgins RJ. Characteristics of cisternal cerebrospinal fluid associated with primary brain tumors in the dog: A retrospective study. J Am Vet Med Assoc 1986;188:414417.
  • 34
    Moore MP, Gavin PR, Bagley RS, et al. Cerebrospinal fluid analysis in dogs with intracranial tumors. Proc ACVIM Forum 1994;8:917921.
  • 35
    Vernau KM, Higgins RJ, Bollen AW, et al. Primary canine and feline nervous system tumors: Intraoperative diagnosis using the smear technique. Vet Pathol 2001;38:4757.
  • 36
    Sturges BK, Dickinson PJ, Bollen AW, et al. Magnetic resonance imaging and histological classification of intracranial meningiomas in 112 dogs. J Vet Intern Med 2008;22:586595.
  • 37
    Dickinson PJ, Sturges BK, Kass PH, LeCouteur RA. Characteristics of cisternal cerebrospinal fluid associated with intracranial meningiomas in dogs: 56 cases (1985–2004). J Am Vet Med Assoc 2006;5:564567.
  • 38
    KliehuesP, BurgerPC, ScheithauerBW, eds. Histological Typing of Tumours of the Central Nervous System. Berlin, Germany: WHO Springer-Verlag; 1993:2021.
  • 39
    Coates TL, Hinshaw DB Jr., Peckman N, et al. Pediatric choroid plexus neoplasms: MR, CT, and pathologic correlation. Radiology 173:8188.
  • 40
    Meyers SP, Khademain ZP, Chuang SH, et al. Choroid plexus carcinomas in children: MRI features and patient outcomes. Neuroradiology 2004;46:770780.
  • 41
    McCall T, Binning M, Blumenthal DT, et al. Variations of disseminated choroid plexus papilloma: 2 case reports and a review of the literature. Surg Neurol 2006;66:6267.
  • 42
    Ellenbogen RG, Scott RM. Choroid plexus tumors. In: KayeAH, LawsERJr., eds. Brain Tumors: An Encyclopedic Approach, 2nd ed. London, UK: Churchill Livingstone; 2001:551562.
  • 43
    Hudson JA, Simpson ST, Buxton DF, et al. Ultrasonographic diagnosis of canine hydrocephalus. Vet Radio1 1990;31:5058.
  • 44
    Spaulding KA, Sharp NJH. Ultrasonographic imaging of the lateral cerebral ventricles in the dog. Vet Radiol 1990;131:5964.
  • 45
    De Haan CE, Kraft SL, Gavin PR, et al. Normal variation in the size of the lateral ventricles of the Labrador Retriever dog as assessed by magnetic resonance imaging. Vet Radiol Ultrasound 1994;35:8386.
  • 46
    Kii S, Uzuka Y, Taura Y, et al. Magnetic resonance imaging of the lateral ventricles in Beagle-type dogs. Vet Radiol Ultrasound 1997;38:430433.
  • 47
    Vite CH, Insko EK, Schotland HM, et al. Quantification of cerebral ventricular volume in English bulldogs. Vet Radiol Ultrasound 1997;38:437443.
  • 48
    Vullo T, Korenman E, Manzo RP, et al. Diagnosis of cerebral ventriculomegaly in normal adult beagles using quantitative MRI. Vet Radiol Ultrasound 1997;38:277281.
  • 49
    Mosely ME, Sawyer-Glover A, Kucharaczyk J. Magnetic resonance imaging principles and techniques. In: LatchawRE, KucharaczykJ, MoseleyM, eds. Imaging of the Nervous System: Diagnostic and Therapeutic Applications. Vol. 2. Philadelphia, PA: Moseby; 2004;10631087.
  • 50
    Milhorat TH, Hammock MK, Davis DA, et al. Choroid plexus papilloma. I. Proof of cerebrospinal fluid overproduction. Childs Brain 1976;2:273289.
  • 51
    McConnell H, Bianchine J. Cerebrospinal Fluid in Neurology and Psychiatry. New York, NY: Chapman and Hall Medical; 1994:132135.
  • 52
    Bigner SH, Johnston WW. The cytopathology of cerebrospinal fluid. II. Metastatic cancer, meningeal carcinomatosis and primary central nervous system neoplasms. Acta Cytol 1981;25:461479.
  • 53
    Glass PJ, Melamed M, Chernik NL, Posner JB. Malignant cells in cerebrospinal fluid (CSF): The meaning of a positive CSF cytology. Neurology 1979;29:13691375.
  • 54
    Pryson RA, Fischler DF. Cerebrospinal fluid cytology: an 11-year experience with 5951 specimens. Arch Pathol Lab Med 1998;122:4751.
  • 55
    Olson ME, Chernik NL, Posner JB. Infiltration of the leptomeninges by systemic cancer: A clinical and pathologic study. Arch Neurol 1974;30:122137.
  • 56
    Rogers LR, Duchesneau PM, Nunez C. Comparison of cisternal and lumbar CSF in leptomeningeal metasteses. Neurology 1992;42:12391241.