Page 142 - Withrow and MacEwen's Small Animal Clinical Oncology, 6th Edition

CHAPTER 6 Diagnostic Imaging in Oncology 121

mast cell tumor in a small group of dogs. 112 Further specific uses
of FDG and use of targeted agents will increase the usefulness
of PET/CT in the future. The utility of PET/MRI in veterinary
VetBooks.ir medicine is undergoing investigation.

Lymph Node Assessment

Assessing LNs for evidence of metastasis is an important part of
staging cancer patients. B-mode ultrasound, CT, and MRI pro-
vide information on the size, shape, and echogenicity of LNs,
providing a good initial assessment provided the LN is accessible.
For example, only the medial iliac and cranial aspect of the inter-
nal iliac LNs are typically visible with abdominal ultrasound; as a
result, ultrasonography is a less sensitive imaging modality than
cross-sectional imaging techniques, such as CT and MRI, for the
detection of nodal metastasis from perineal tumors where the
sacral node is likely to be the draining node (e.g., apocrine gland
anal sac adenocarcinomas). 52,113
Using absolute LN size to determine malignancy can be prob-
lematic because of the substantial variation in dog breed sizes and
the fact that metastatic LNs are not always enlarged. Many studies
have looked at normalizing LN size to account for this variation,
• Fig. 6.14 A dorsal plane reconstruction of the head obtained 3 minutes but with variable success. 114,115 CT assessment of sternal LNs sug-
after injection of a contrast medium around a rostral right mandibular mass gest that a ratio of sternal LN height to the height of the sec-
demonstrates multiple lymphatic vessels draining into the right mandibular ond sternebra of >1 and a precontrast attenuation of 37.5 HU
lymph node (arrow) identifying this as the sentinel lymph node. It is impor- or greater is highly predictive of malignancy. 115 The presence of
tant to note that further cytologic or histopathologic assessment of the micrometastasis is also common and may not result in an absolute
sentinel lymph node is required to determine whether there is evidence of change in LN size. A change in the shape of an LN from oval
4
metastasis within this lymph node. to round, as indicated by a short axis-to-long axis ratio of >0.7,
and loss of definition of the LN hilus are seen more frequently
with malignant than reactive LNs. 116,117 The signal characteris-
bone-specific radiopharmaceutical that enables generation of tics of LNs should also be considered as, in combination with
highly detailed skeletal images that are superior for detection of other parameters, this may be helpful in differentiating benign
bone metastasis compared with planar or single-photon emission from malignant LNs. On ultrasound, both reactive and malignant
computerized tomography (SPECT) imaging with Tc99m-methyl LNs tend to be hypoechoic and may have a heterogeneous echo-
diphosphonate (Tc99m-MDP). To date, the only reported use genicity, but there is no consistent pattern to differentiate these
95
of NaF F-18 is in equine orthopedic disease and in four normal processes. 116,118 MRI perinodal contrast enhancement is seen
immature dogs. 96,97 Labeling F-18 with the thymidine analog FLT more commonly in cats and dogs with reactive lymphadenitis. 119
18
(3′-deoxy-3′[ F]fluorothymidine) allows the imaging of proliferat- Metastatic disease to the LNs often alters the vascularity and
ing tissues, including neoplastic tissues. FLT reflects DNA synthesis these changes may be evident with Doppler ultrasonography.
and was successful in detecting initial disease, response to therapy, The blood supply to normal LNs is predominantly centered at
and predicting relapse in a small group of dogs with lymphoma. the hilus, whereas malignant LNs trend to having a peripheral
98
PET/CT functional imaging of tumor hypoxia using dogs with or mixed peripheral and hilar distribution with a greater number
spontaneously arising nasal tumors injected with copper(II)-diace- of vessels. 26,117,118 Displacement of the central hilar vessels, more
tyl-bis(N -methylthiosemicarbazone) (Cu-ATSM) has also been than one central vessel, pericapsular vessels, and peripheral vessels
4
performed and evaluated for radiation boost planning. 99 that do not connect with the central vessel have all been associated
29
The baseline metabolic rate of normal tissues must be known with malignancy. Vascular resistance may increase in malignant
to accurately interpret PET images. The standardized uptake value LNs as a result of vascular compression from tumor cells, and
(SUV) is a semiquantitative measure of FDG uptake that is directly this can be detected by measuring resistive index and pulsatility
proportional to the metabolic activity. 100 Increased SUV measure- index. 120 Increases in these parameters are suggestive of malig-
ment correlates to subjective increased hypermetabolic activity vis- nancy when seen in combination with other sonographic find-
ible on PET images. No specific cut-offs have been established for ings. 26,116 Fundamentally, however, there is no one sonographic
malignancy in veterinary medicine, whereas in human medicine measurement that is predictive of malignancy; using a combina-
an SUV max >2.5 is predictive of malignancy in patients with soli- tion of size, shape, and vascularity of the LN, in addition to fine-
tary pulmonary nodules. 101 However, multiple studies have found needle aspirate cytology of the LN, is recommended to increase
that visual interpretation by an experienced reader is equivalent to the predictive value for detection of metastatic disease. 120
or superior compared with SUV measurements. 102–104 Detection of the sentinel or draining LN is very important
Multiple tumor types in dogs and cats have been shown to be in the clinical staging and determining the treatment options
hypermetabolic on PET or PET/CT, including but not limited and prognosis for animals and people with cancer (see Chap-
to lymphoma, mast cell tumor, osteosarcoma, oral squamous cell ter 9). Assessment of LNs is based on size, shape, and contrast
carcinoma, mammary carcinoma, fibrosarcoma, and primary lung enhancement, but there is wide variation in accuracy using these
tumors. 98,105–111 F-18 FDG was shown to correlate with grade of parameters due in part to the subjectivity of evaluation. 4,118