DOI: 10.1148/rg.241035078
(Radiographics. 2004;24:209-223.)
© RSNA, 2004
PET-CT in Recurrent Ovarian Cancer: Initial Observations1
Harpreet K. Pannu, MD,
Robert E. Bristow, MD,
Christian Cohade, MD,
Elliot K. Fishman, MD and
Richard L. Wahl, MD
1 From the Russell H. Morgan Department of Radiology and Radiological Science (H.K.P., C.C., E.K.F., R.L.W.) and the Kelly Gynecologic Oncology Service (R.E.B.), Johns Hopkins Medical Institutions, 600 N Wolfe St, Baltimore, MD 21287. Presented as an education exhibit at the 2002 RSNA scientific assembly. R.L.W. is a stockholder in CTI Molecular Imaging, Knoxville, Tenn; he also has a research agreement with and has received honoraria from GE Medical Systems. Received March 24, 2003; revision requested April 22 and received August 11; accepted August 14. Address correspondence to H.K.P. (e-mail: hpannu1@jhmi.edu).
 |
Abstract
|
|---|
Noninvasive diagnosis of early recurrence of ovarian cancer is challenging due to the small size of peritoneal metastases. Small-volume disease may not be evident at anatomic imaging in patients with elevated serum tumor markers. Functional imaging in the form of positron emission tomography (PET) can help identify patients with recurrent tumor. However, lesion localization for possible surgical treatment is difficult with PET alone. Combined functional-anatomic imaging with fused PET and computed tomographic (CT) scans is feasible and may improve disease detection by increasing radiologic sensitivity and specificity. PET and PET-CT have a potential role in evaluating patients for recurrent ovarian cancer, particularly those with negative CT or magnetic resonance imaging findings and rising tumor marker levels. Fused PET-CT scans obtained with combined scanners can help localize pathologic activity and differentiate this activity from physiologic radiotracer uptake. Combined functional-anatomic imaging can also increase diagnostic confidence at CT. Further study is needed to determine the possible benefits of lesion conspicuity at PET and anatomic localization at CT on fused PET-CT scans.
© RSNA, 2004
Index Terms: Diagnostic radiology, 852.1211, 852.12163 • Dual-modality imaging, PET/CT • Ovary, CT, 852.1211 • Ovary, neoplasms, 852.39 Ovary, PET, 852.12163
|
Introduction
|
|---|
Metastases from ovarian cancer are unlike most other tumors in that they are primarily peritoneal rather than parenchymal in location. Therefore, they usually occur on the surfaces of the viscera rather than as masses within the viscera. These tumor implants can be miliary and isoattenuating relative to the viscera at computed tomography (CT), which makes their detection challenging. A number of approaches are used to detect recurrent metastatic lesions after initial surgery and chemotherapy for ovarian cancer. These approaches include physical examination, determination of serum cancer antigen–125 levels, and imaging. CT, magnetic resonance (MR) imaging, and positron emission tomography (PET) have all been used to evaluate affected patients. Recognized limitations include failure to detect small lesions with all three modalities and misinterpretation of normal physiologic abdominal activity at PET (1,2). However, lesion conspicuity is high at PET due to low background activity. Therefore, there is interest in using PET especially in patients with clinically suspected recurrence but with negative or equivocal anatomic imaging findings (3). Fused PET-CT offers the combined benefits of anatomic and functional imaging.
The PET and CT gantries are mechanically combined, and there is a single imaging table with inline PET-CT scanners (4). This arrangement makes it possible for the patient to maintain a similar position for both studies without significant change in the body curvature or shift of the internal organs, allowing more accurate alignment on the fused scans. External markers or complicated computer fusion algorithms are not necessary for image fusion (4). The volume or content of the stomach, bladder, and colon is also relatively similar because the scans are obtained in one session. Automated intermodality registration results in improved alignment of data to within a few millimeters in three translation dimensions and to within a few degrees in three rotation dimensions (4). In addition, use of CT for attenuation correction decreases the overall time required for PET.
Potential advantages of PET-CT for patients with ovarian cancer include increased lesion conspicuity, anatomic localization of lesions, and differentiation of disease processes from physiologic activity. In this article, we review the nature, symptoms, and treatment of ovarian cancer. We also discuss and illustrate the strengths and limitations of CT, MR imaging, PET, and combined PET-CT in patients with this disease entity. In addition, we describe PET-CT technique and discuss the interpretation of PET scans.
|
Ovarian Cancer
|
|---|
Ovarian cancer usually affects women over the age of 60 years and is the most common gynecologic malignancy to cause death (5,6). Ninety percent of ovarian cancers are sporadic; the remaining 10% are due to inherited syndromes such as breast-ovarian cancer syndrome, Lynch syndrome II, and hereditary site-specific ovarian cancer (7).
Symptoms of ovarian cancer are nonspecific, and the majority of women have advanced disease at presentation (8). Tumor is limited to the ovaries in stage I disease, extends into the pelvis in stage II disease, extends beyond the pelvis in stage III disease, and involves distant sites or the liver parenchyma in stage IV disease (9). Most patients have stage III or IV disease at diagnosis. The 5-year survival rate for patients with stage III disease is 25%–39% (7).
Metastases occur due to peritoneal, lymphatic, or hematogenous spread of tumor (10), with the peritoneal route being the most common. Peritoneal fluid flows upward from the pelvis to the paracolic gutters and subphrenic regions, carrying tumor cells that implant on the abdominal viscera (11). Common sites of implantation are the pelvis, right hemidiaphragm, liver, right paracolic gutter, bowel, and omentum (12).
An important part of treatment is surgical staging and debulking of ovarian cancer. The abdomen and pelvis are explored to remove the majority of all visible tumor implants. For patients with an initial diagnosis of tumor, the prognosis is better in those with residual subcentimeter lesions after surgery than in those with larger amounts of residual disease (13). After primary cytoreductive surgery, chemotherapy is administered, usually in the form of platinum-based compounds. Patients are monitored for recurrence with physical examination, determination of serum CA-125 levels, and imaging. For patients in whom recurrence is diagnosed, secondary cytoreductive surgery is beneficial if the largest deposit is less than 10 cm and there is no gross tumor after surgery (13).
|
Imaging of Ovarian Cancer
|
|---|
The traditional imaging modalities for evaluating patients for possible recurrence are CT and MR imaging. In prior studies performed with conventional and single–detector row spiral CT, peritoneal metastases were detected in approximately 50% of cases and recurrent tumor in 66.6% (14–20). A more recent study performed with single–detector row CT found that the sensitivity for peritoneal metastases and for subcentimeter lesions was 85%–93% and 25%–50%, respectively (1). Lesion detection is dependent on lesion size and is better for implants larger than 5–10 mm (1,21). Sensitivity for implants is greater than 50% at most sites except the small bowel and mesentery, where lesions can be difficult to appreciate due to partial volume averaging (22). The sensitivity of multi–detector row CT for detecting tumor recurrence has not yet been established.
CT and MR imaging are of equal value in detecting peritoneal metastases from ovarian cancer (16,22–24). In a study by Tempany et al (24), the sensitivity of CT and MR imaging for peritoneal disease in women with a pelvic malignancy was 92% and 95%, respectively. Peritoneal, mesenteric, and bowel metastases can be detected with MR imaging, whose sensitivity for recurrent tumor is 91% (25,26).
Although anatomic imaging is the mainstay for evaluating patients for possible recurrence of ovarian cancer, small implants on visceral surfaces can be difficult to detect with CT and MR imaging. The lesions may not be appreciated due to partial volume averaging or lack of significant differences in attenuation or signal intensity between tumor and normal viscera. A potential advantage of PET—one that has been evaluated in a few studies—is that lesions are more conspicuous relative to minimal background activity due to increased radiotracer uptake in tumor. This phenomenon may help detect metastatic tumor on visceral surfaces and in normal-sized nodes.
The sensitivity of PET for recurrent tumor is higher in patients with suspected relapse than in those without clinical disease (27). Sensitivities ranging from 80% to 100% have been reported in four series involving a total of 113 patients (27–30). However, it is recognized that detection is dependent on lesion size (27,28). PET has a spatial resolution of approximately 6–10 mm; therefore, its sensitivity for lesions less than 1–2 cm in size is lower than that for larger masses (31,32). Omental carcinomatosis due to subcentimeter lesions may not demonstrate sufficient uptake to be detected at PET, even if the disease process is evident at CT (32). In one study of 22 patients with primarily subcentimeter lesions (including microscopic disease), the sensitivity of PET for tumor recurrence was 10% (2). In another study of 31 patients in whom the mean lesion size was 1.1 cm, the patient-based sensitivity of PET was 81% and the lesion-based sensitivity was 45% (32). As with anatomic imaging modalities, microscopic disease can be found at histologic analysis in patients without clinical disease who have negative PET findings and subsequently undergo surgery (31).
The reported specificity of PET for recurrent ovarian cancer ranges from 42% to 100% (2,27–30,32). Three studies that compared PET findings with surgical findings found specificities of 93%, 42%, and 50%, respectively (2,28,32). High specificities of 83% and 100% were reported in two studies that compared PET with surgical or clinical follow-up (27,29).
Physiologic activity in the abdomen can lower the specificity of PET (2). Normally, variable activity is seen in the stomach and bowel, whereas faint uptake is seen in the liver and spleen (4). There is also excretion of 2-[fluorine-18]fluoro-2-deoxy-D-glucose (FDG) by the kidneys, which leads to renal, ureteral, and bladder activity. Because ovarian cancer metastases manifest as scattered focal implants, distinguishing between physiologic and pathologic activity at PET can be difficult. In addition, for lesions that are considered to be pathologic, accurate localization for surgical resection is difficult due to a paucity of anatomic landmarks. Digital fusion of PET scans with CT scans allows more accurate diagnosis and localization of lesions (33).
PET-CT combines the benefits of functional and anatomic imaging for recurrent ovarian cancer. There is close approximation of the anatomic and functional images on the fused gantry. Multi–detector row CT scanners are available on combined systems, and a few reports suggest that optimal CT with oral and intravenous contrast material can be performed without interfering with the transmission scan for PET (34,35). Thin-section imaging of the entire abdomen and pelvis can be performed with multi–detector row CT without concern for volume coverage, a capacity that allows multiplanar review for problem solving. However, higher radiation dose is a concern. Fusion with the PET scans highlights areas of potential metastases and allows characterization of equivocal abnormalities, especially in postoperative patients.
Combined information from PET and CT or MR imaging improves overall sensitivity, specificity, and accuracy in detecting tumor recurrence (36). Anatomic imaging can yield true-positive results in patients with false-negative PET findings; conversely, PET findings can be true positive (29). In a case report by Bristow et al involving patients with elevated serum CA-125 levels (3), recurrent tumor was identified at PET in those patients in whom CT findings were negative or equivocal. All patients proved to have small-volume disease at surgery. The PET findings were used to direct secondary cytoreduction. Thus, combining functional and anatomic imaging may be useful in the treatment of patients with clinically suspected recurrence because either test alone may not reveal tumor (3).
The value of PET-CT in detecting ovarian cancer has not yet been established because relatively few scanners are currently available. In a study of eight patients, most of whom had a tumor volume that exceeded 1 cm, the sensitivity of PET-CT for recurrent disease was 62% (37). All patients proved to have recurrent disease at surgery; therefore, specificity could not be determined. Larger studies that correlate radiologic with histologic findings are necessary to determine the sensitivity and specificity of PET-CT.
|
FDG Positron Emission Tomography
|
|---|
PET is usually performed with injection of a radioactively labeled glucose analog in the form of FDG because cancer cells demonstrate increased glucose utilization (4). The initial metabolism of glucose by cancer cells is characterized by overexpression of glucose transporters and hexokinase enzymes, and the FDG is transported into the cell and phosphorylated (4). However, the phosphorylated product cannot be metabolized further and is trapped in the cell. The FDG molecule decays by means of positron emission and has a half-life of approximately 110 minutes.
The emitted positron collides with an electron, and both are annihilated, resulting in the production of two photons. These photons are emitted 180° apart and are detected by opposing detectors to form one count on the image. Millions of counts are combined to form a single image. A short acquisition window (8–16 nsec) is used to ensure that the detected photons are from the same annihilation reaction.
FDG uptake by cancer cells is increased if there are viable cells and the tumor is adequately perfused (4). Uptake can also be acutely increased after chemotherapy or radiation therapy but is decreased if the tumor is extensively necrotic. Besides cancer tissue, other cells with increased glucose metabolism such as neurons, cardiac muscle, exercising skeletal muscle, and inflammatory cells also show increased FDG uptake.
The currently accepted indications for FDG PET are solitary pulmonary nodules, non–small cell lung cancer, colorectal cancer, lymphoma, head and neck cancer, esophageal cancer, breast cancer, melanoma, refractory seizures, and myocardial viability.
|
PET-CT Technique
|
|---|
The patient is asked to fast for 4 hours prior to undergoing PET-CT, and blood sugar levels are checked to ensure that there is no hyperglycemia because FDG uptake in cancer cells is reduced if there are competing unlabeled glucose molecules. Approximately 900 mL of a barium sulfate solution (Readi-cat [1.3% weight-volume barium sulfate suspension]; E-Z-EM, Westbury, NY) is administered orally 1 hour prior to imaging to opacify the bowel for the CT portion of the study. In addition, 15–20 mCi (555–740 MBq) (0.22 mCi/kg body weight) of FDG is administered intravenously 1 hour prior to imaging. Patients sit quietly in a dimly lit room during the uptake phase and are asked to void just prior to imaging. For scanning, patients lie supine with the arms raised above the head on a fused PET-CT scanner with a single gantry and table. The CT and PET scans are obtained with the patient in quiet respiration.
CT is performed prior to PET, and the resulting data are used to generate an attenuation correction map for PET. Five-millimeter-thick sections are obtained at 80 mA (but adjusted for body thickness) and 140 kVp from the skull base to the midthigh. The images are reconstructed with a 512 x 512 matrix and a 50-cm field of view. For fusion with the PET data, images are also reconstructed with a 128 x 128 matrix.
Next, PET is performed on a dedicated PET scanner with a 5-minute emission acquisition per imaging level. The images are acquired in a caudocranial direction from the midthigh to the skull base. The CT transmission map is used for attenuation correction, and the PET images are reconstructed with a 128 x 128 matrix, an ordered subset expectation maximum iterative reconstruction algorithm (two iterations, 28 subsets), an 8-mm gaussian filter, and a 50-cm field of view.
Attenuation correction on the PET scans is needed for fusion with the CT scans. Attenuation correction compensates for differing activity in deep versus superficial lesions. Photons from deep lesions are attenuated to a greater degree, and there can be geometric distortion on the PET scans if correction is not performed.
|
Interpretation of PET Scans
|
|---|
Once FDG has been injected, it accumulates in the brain and myocardium (38). Myocardial uptake is often intense in patients who have not been fasting but is variable in patients who have fasted for 4–18 hours (Fig 1). FDG is excreted by the kidneys, and intense activity is seen in the ureters and bladder (Fig 2). Pooling of activity in the calices can simulate lesions (38). Voiding by the patient prior to undergoing imaging reduces bladder activity. Less intense uptake occurs in the liver, spleen, and bone marrow, and blood pool activity decreases during the 60-minute period after injection.
In the gastrointestinal tract, activity in the stomach is common and can be identified by its configuration. Normal colonic and small bowel activity can also occur in fasting patients (Fig 3) (38). In the musculoskeletal system, activity is seen in skeletal muscle groups that have been used a few hours before or immediately after FDG injection. Similarly, postinjection use of extraocular muscles and of laryngeal muscles for speaking increases uptake in these areas. Other areas of normal uptake are the thyroid gland, endometrium, breast, and testes.
Focal uptake of FDG can also be seen in benign conditions, including healing fracture, granulomatous disease (eg, tuberculosis, sarcoidosis), inflammatory or degenerative joint disease, abscess, and pancreatitis (38,39). Granulation tissue associated with ostomy sites also demonstrates increased FDG uptake.
PET scans are interpreted qualitatively or quantitatively for pathologically increased uptake, with qualitative assessment being the more common method. The images are reviewed on a workstation in the axial, coronal, and sagittal planes with a varying gray scale and rotating views (4). Areas of tumor activity are identified depending on the experience of the reviewer. Activity in areas where there is no expected physiologic uptake is abnormal. In areas of physiologic uptake, an intensity of activity that is visually greater than would be expected is considered abnormal.
Quantitative evaluation of areas of increased activity can be performed on attenuation-corrected scans. The standardized uptake value (SUV) of a lesion is calculated to determine if the lesion is more likely to be benign or malignant and is defined as follows:
where dose in tissue is in millicuries per milliliter, injected dose is in millicuries, and patient weight is in grams. The SUV is dependent on many variables, including body mass and the region of interest, and is higher in obese patients and with smaller regions of interest (4). Therefore, the SUV is not often used for diagnosis but is commonly used to follow up treatment response. SUVs tend to be higher in tumors than in benign lesions: the higher the SUV of a mass, the more likely the mass is to be malignant. For ovarian cancer, SUVs can range up to 6.5 or more, with those over 3.25 being of greatest concern (4).
|
PET-CT Appearance of Ovarian Cancer Metastases
|
|---|
At CT, peritoneal implants manifest as nodular soft-tissue masses that can coalesce to form plaques that coat the viscera. The implants can enhance with intravenous contrast material or calcify. Some implants mimic loculated fluid because they are hypoattenuating (9). There is nodular thickening of the diaphragm, mesentery, and omentum. Involvement of solid organs like the liver and spleen results in scalloping of the surface by masses that are hypoattenuating relative to the normal parenchyma. Soft-tissue masses on the bowel can tether the loops and result in bowel obstruction, the most common type of morbidity secondary to ovarian cancer (40). In the pelvis, implants lie on the surface of the sigmoid colon and rectum, sigmoid mesocolon, perirectal tissues, bladder, and cul-de-sac. Nodes greater than 1 cm in short-axis diameter are considered abnormal.
At PET, metastatic lesions appear as globular foci of intensely increased activity. Diagnosis is aided by recognizing activity in areas where implants are commonly located.
At PET-CT, regions of increased activity seen at PET can be localized on the CT scans, and lesions seen at CT can be evaluated for pathologic activity on the PET scans. Metastatic lesions seen at CT demonstrate increased activity on the fused PET-CT scans (Figs 4, 5). In mixed cystic and solid lesions, uptake is seen in the solid component but not in the necrotic portion (Fig 6).
View larger version (130K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4a. Metastatic disease in a 46-year-old woman with a prior history of micropapillary serous ovarian cancer and an elevated serum CA-125 level (>8,000 U/mL). (a) CT scan demonstrates partially calcified masses in the left side of the pelvis (arrows) near the bowel and superior to the bladder. (b) Fused PET-CT scan shows increased activity in the area of the pelvic masses (arrows), a finding that is compatible with tumor.
|
|
View larger version (94K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4b. Metastatic disease in a 46-year-old woman with a prior history of micropapillary serous ovarian cancer and an elevated serum CA-125 level (>8,000 U/mL). (a) CT scan demonstrates partially calcified masses in the left side of the pelvis (arrows) near the bowel and superior to the bladder. (b) Fused PET-CT scan shows increased activity in the area of the pelvic masses (arrows), a finding that is compatible with tumor.
|
|
View larger version (150K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6a. Metastatic disease in a 57-year-old woman with a history of moderately differentiated ovarian carcinoma with serous and endometrioid features. (a) CT scan demonstrates a mass (arrow) with cystic central and solid peripheral components. (b) Fused PET-CT scan shows areas of increased activity (arrow) that are compatible with tumor.
|
|
View larger version (106K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6b. Metastatic disease in a 57-year-old woman with a history of moderately differentiated ovarian carcinoma with serous and endometrioid features. (a) CT scan demonstrates a mass (arrow) with cystic central and solid peripheral components. (b) Fused PET-CT scan shows areas of increased activity (arrow) that are compatible with tumor.
|
|
Implants on or near bowel loops can be difficult to detect or to distinguish from normal bowel at CT. Increased activity in such lesions at PET draws attention to their presence. In our experience, increased bowel uptake with no obvious mass at CT has occasionally proved to represent tumor nodules at surgery (Figs 7, 8). If small nodules are visible at CT, they can be assessed at PET for increased activity to confirm their malignant nature (Fig 9). Differentiation between bowel implants and adjacent unopacified bowel is also easier when the implants demonstrate increased activity (Fig 10). Increased uptake in pleural effusions suggests metastatic disease to the chest (Fig 11). Thoracic metastases are usually present in the setting of abdominal disease. Affected patients most commonly have pleural effusions and adenopathy. As expected, metastatic mediastinal adenopathy demonstrates increased activity (Fig 12).
View larger version (152K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7a. Tumor implants in a 59-year-old woman with a history of moderately differentiated serous carcinoma. (a) CT scan shows small nodules in the sigmoid mesocolon (arrow) but no definite mass in the right side of the pelvis. (b) Fused PET-CT scan shows activity in the sigmoid nodules and superimposed on the cecum (arrows). Surgery revealed tumor at both sites.
|
|
View larger version (108K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7b. Tumor implants in a 59-year-old woman with a history of moderately differentiated serous carcinoma. (a) CT scan shows small nodules in the sigmoid mesocolon (arrow) but no definite mass in the right side of the pelvis. (b) Fused PET-CT scan shows activity in the sigmoid nodules and superimposed on the cecum (arrows). Surgery revealed tumor at both sites.
|
|
View larger version (153K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8a. Tumor implants in the same patient as in Figure 7. (a) CT scan shows minimal nodularity (arrow) that is adjacent to the unopacified transverse colon and is difficult to distinguish from adjacent bowel. No obvious mass is identified. (b) PET scan shows a small area of increased uptake in the midabdomen (arrow). (c) Fused PET-CT scan helps localize the increased activity seen at PET to the transverse colon (arrow), where small tumor nodules were found at surgery. (Fig 8 reprinted, with permission, from reference 41.)
|
|
View larger version (65K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8b. Tumor implants in the same patient as in Figure 7. (a) CT scan shows minimal nodularity (arrow) that is adjacent to the unopacified transverse colon and is difficult to distinguish from adjacent bowel. No obvious mass is identified. (b) PET scan shows a small area of increased uptake in the midabdomen (arrow). (c) Fused PET-CT scan helps localize the increased activity seen at PET to the transverse colon (arrow), where small tumor nodules were found at surgery. (Fig 8 reprinted, with permission, from reference 41.)
|
|
View larger version (114K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8c. Tumor implants in the same patient as in Figure 7. (a) CT scan shows minimal nodularity (arrow) that is adjacent to the unopacified transverse colon and is difficult to distinguish from adjacent bowel. No obvious mass is identified. (b) PET scan shows a small area of increased uptake in the midabdomen (arrow). (c) Fused PET-CT scan helps localize the increased activity seen at PET to the transverse colon (arrow), where small tumor nodules were found at surgery. (Fig 8 reprinted, with permission, from reference 41.)
|
|
View larger version (165K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9a. Tumor implants in a 55-year-old woman with ovarian cancer and a rising serum CA-125 level. (a) CT scan demonstrates minimal nodularity in the left paracolic gutter (arrow), a finding that is suggestive of implants. (b) PET scan shows a linear area of slightly increased uptake in the same region (arrow) that may represent either normal bowel or implants. (c) Fused PET-CT scan demonstrates an area of increased activity (arrow) that corresponds to the nodules seen at CT and is suggestive of tumor implants.
|
|
View larger version (117K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9b. Tumor implants in a 55-year-old woman with ovarian cancer and a rising serum CA-125 level. (a) CT scan demonstrates minimal nodularity in the left paracolic gutter (arrow), a finding that is suggestive of implants. (b) PET scan shows a linear area of slightly increased uptake in the same region (arrow) that may represent either normal bowel or implants. (c) Fused PET-CT scan demonstrates an area of increased activity (arrow) that corresponds to the nodules seen at CT and is suggestive of tumor implants.
|
|
View larger version (110K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9c. Tumor implants in a 55-year-old woman with ovarian cancer and a rising serum CA-125 level. (a) CT scan demonstrates minimal nodularity in the left paracolic gutter (arrow), a finding that is suggestive of implants. (b) PET scan shows a linear area of slightly increased uptake in the same region (arrow) that may represent either normal bowel or implants. (c) Fused PET-CT scan demonstrates an area of increased activity (arrow) that corresponds to the nodules seen at CT and is suggestive of tumor implants.
|
|
View larger version (134K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 10a. Midabdominal tumor in a 46-year-old woman with a history of micropapillary serous ovarian cancer and an elevated serum CA-125 level (same patient as in Fig 4). (a) CT scan shows inhomogeneous distribution of oral contrast material, resulting in some small bowel loops being unopacified. (b) PET scan demonstrates an abdominal mass with increased uptake (arrow). (c) Fused PET-CT scan helps localize the mass (arrow) and allows differentiation of the tumor from adjacent bowel loops.
|
|
View larger version (117K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 10b. Midabdominal tumor in a 46-year-old woman with a history of micropapillary serous ovarian cancer and an elevated serum CA-125 level (same patient as in Fig 4). (a) CT scan shows inhomogeneous distribution of oral contrast material, resulting in some small bowel loops being unopacified. (b) PET scan demonstrates an abdominal mass with increased uptake (arrow). (c) Fused PET-CT scan helps localize the mass (arrow) and allows differentiation of the tumor from adjacent bowel loops.
|
|
View larger version (115K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 10c. Midabdominal tumor in a 46-year-old woman with a history of micropapillary serous ovarian cancer and an elevated serum CA-125 level (same patient as in Fig 4). (a) CT scan shows inhomogeneous distribution of oral contrast material, resulting in some small bowel loops being unopacified. (b) PET scan demonstrates an abdominal mass with increased uptake (arrow). (c) Fused PET-CT scan helps localize the mass (arrow) and allows differentiation of the tumor from adjacent bowel loops.
|
|
View larger version (122K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 11a. Metastatic disease to the chest in a 46-year-old woman with a history of micropapillary serous ovarian cancer and an elevated serum CA-125 level (same patient as in Fig 4). (a) CT scan shows a small right-sided pleural effusion and pleural thickening (arrow). (b) Fused PET-CT scan shows increased uptake in the right-sided pleura (arrow), a finding that is suggestive of malignancy.
|
|
View larger version (121K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 11b. Metastatic disease to the chest in a 46-year-old woman with a history of micropapillary serous ovarian cancer and an elevated serum CA-125 level (same patient as in Fig 4). (a) CT scan shows a small right-sided pleural effusion and pleural thickening (arrow). (b) Fused PET-CT scan shows increased uptake in the right-sided pleura (arrow), a finding that is suggestive of malignancy.
|
|
View larger version (99K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 12a. Metastatic mediastinal adenopathy in a 59-year-old woman with a history of poorly differentiated ovarian carcinoma with serous features. Fused PET-CT scans demonstrate enlarged nodes with increased uptake in the right paratracheal, bilateral hilar, subcarinal, and prevascular regions (arrows). The patient also had increasing abdominal disease at CT and a rising serum CA-125 level.
|
|
View larger version (93K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 12b. Metastatic mediastinal adenopathy in a 59-year-old woman with a history of poorly differentiated ovarian carcinoma with serous features. Fused PET-CT scans demonstrate enlarged nodes with increased uptake in the right paratracheal, bilateral hilar, subcarinal, and prevascular regions (arrows). The patient also had increasing abdominal disease at CT and a rising serum CA-125 level.
|
|
Areas of increased activity at PET are localized on the fused PET-CT scans, allowing peritoneal lesions involving bowel to be distinguished from abnormal nodes (Fig 13). Pathologic activity is distinguished from physiologic bowel activity, and uptake in surgical incisions or ostomies is diagnosed with increased confidence on the fused scans (Figs 9, 14).
View larger version (124K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 13a. Malignant iliac nodes in a 46-year-old woman with a history of micropapillary serous ovarian cancer (same patient as in Fig 4). (a) PET scan shows increased uptake in the left side of the pelvis (arrow) that may represent either a bowel implant or a node. (b) CT scan demonstrates enlarged left iliac nodes (arrow). (c) Fused PET-CT scan shows activity in the left iliac nodes (arrow), which proved to be malignant at surgery. Note the misregistration due to bladder activity in the right side of the pelvis.
|
|
View larger version (159K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 13b. Malignant iliac nodes in a 46-year-old woman with a history of micropapillary serous ovarian cancer (same patient as in Fig 4). (a) PET scan shows increased uptake in the left side of the pelvis (arrow) that may represent either a bowel implant or a node. (b) CT scan demonstrates enlarged left iliac nodes (arrow). (c) Fused PET-CT scan shows activity in the left iliac nodes (arrow), which proved to be malignant at surgery. Note the misregistration due to bladder activity in the right side of the pelvis.
|
|
View larger version (112K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 13c. Malignant iliac nodes in a 46-year-old woman with a history of micropapillary serous ovarian cancer (same patient as in Fig 4). (a) PET scan shows increased uptake in the left side of the pelvis (arrow) that may represent either a bowel implant or a node. (b) CT scan demonstrates enlarged left iliac nodes (arrow). (c) Fused PET-CT scan shows activity in the left iliac nodes (arrow), which proved to be malignant at surgery. Note the misregistration due to bladder activity in the right side of the pelvis.
|
|
View larger version (173K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 14a. Increased uptake in a surgical incision in a 73-year-old woman who had undergone primary surgery for adenosarcoma of the ovary 4 weeks earlier. (a) CT scan demonstrates a surgical scar in the abdominal wall (arrow). (b, c) Axial (b) and sagittal (c) fused PET-CT scans demonstrate increased uptake in the abdominal wall (arrow), a finding that corresponds to the scar seen at CT. The increased activity is due to inflammatory change secondary to surgery.
|
|
View larger version (114K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 14b. Increased uptake in a surgical incision in a 73-year-old woman who had undergone primary surgery for adenosarcoma of the ovary 4 weeks earlier. (a) CT scan demonstrates a surgical scar in the abdominal wall (arrow). (b, c) Axial (b) and sagittal (c) fused PET-CT scans demonstrate increased uptake in the abdominal wall (arrow), a finding that corresponds to the scar seen at CT. The increased activity is due to inflammatory change secondary to surgery.
|
|
View larger version (60K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 14c. Increased uptake in a surgical incision in a 73-year-old woman who had undergone primary surgery for adenosarcoma of the ovary 4 weeks earlier. (a) CT scan demonstrates a surgical scar in the abdominal wall (arrow). (b, c) Axial (b) and sagittal (c) fused PET-CT scans demonstrate increased uptake in the abdominal wall (arrow), a finding that corresponds to the scar seen at CT. The increased activity is due to inflammatory change secondary to surgery.
|
|
However, there are also pitfalls associated with PET-CT. Misregistration can result from physiologic patient activity that occurs during the relatively long acquisition time (5 minutes per bed position) for PET. Although the external anatomy is aligned to within a few millimeters on the fused scan, respiratory motion and bowel peristalsis can result in a mismatch between the location of a lesion at PET and its location at CT (Fig 15). Misregistration due to physiologic bladder activity can manifest at PET as pelvic activity that is located superior to the actual location of the bladder as seen on the corresponding CT scans (Fig 16). This finding may obscure pathologic pelvic activity, with lesions being missed due to overlap by bladder activity. Physiologic activity tends to be most intense in the bladder, ureters, and renal collecting system because the radiotracer is excreted by the kidneys; therefore, it is recommended that the patient void prior to undergoing imaging. Pooling of contrast material can occur in the upper pole calices, in tortuous ureters, and in bladder diverticula, thereby mimicking disease (38). Another pitfall involves the gastrointestinal tract, where normal activity may be misinterpreted as pathologic or vice versa, especially if the activity is focal. FDG uptake can appear focal in the stomach and right colon (38). Finally, normal activity or lesions can appear larger on a PET scan than on the corresponding CT scan due to the "blooming" of intense activity that occurs at PET (Fig 17).
View larger version (121K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 15a. Misregistration due to physiologic activity in a 57-year-old woman with a history of moderately differentiated ovarian carcinoma with serous and endometrioid features (same patient as in Fig 6). (a) PET scan shows increased uptake in the left side of the abdomen (arrow). (b) CT scan shows a small peritoneal implant in the left paracolic gutter (arrow). (c) Fused PET-CT scan shows activity superimposed on the stomach (arrow) and just medial to the actual location of the implant (arrowhead). The misregistration is due to peristalsis or respiratory motion. Activity in the right kidney is normal.
|
|
View larger version (158K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 15b. Misregistration due to physiologic activity in a 57-year-old woman with a history of moderately differentiated ovarian carcinoma with serous and endometrioid features (same patient as in Fig 6). (a) PET scan shows increased uptake in the left side of the abdomen (arrow). (b) CT scan shows a small peritoneal implant in the left paracolic gutter (arrow). (c) Fused PET-CT scan shows activity superimposed on the stomach (arrow) and just medial to the actual location of the implant (arrowhead). The misregistration is due to peristalsis or respiratory motion. Activity in the right kidney is normal.
|
|
View larger version (106K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 15c. Misregistration due to physiologic activity in a 57-year-old woman with a history of moderately differentiated ovarian carcinoma with serous and endometrioid features (same patient as in Fig 6). (a) PET scan shows increased uptake in the left side of the abdomen (arrow). (b) CT scan shows a small peritoneal implant in the left paracolic gutter (arrow). (c) Fused PET-CT scan shows activity superimposed on the stomach (arrow) and just medial to the actual location of the implant (arrowhead). The misregistration is due to peristalsis or respiratory motion. Activity in the right kidney is normal.
|
|
View larger version (165K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 16a. Misregistration due to physiologic bladder activity in a 58-year-old woman with intermediate to poorly differentiated Sertoli-Leydig cell tumor with heterologous elements. (a) CT scan shows a cystic mass in the left side of the pelvis (arrow). (b) Fused PET-CT scan shows activity in the right side of the pelvis (arrow). (c) CT scan obtained inferior to a reveals that the pelvic activity seen in b represents bladder activity secondary to misregistration due to respiratory motion and bladder filling (arrow). The lesion in the left side of the pelvis (cf a) represents a postoperative fluid collection and shows no activity.
|
|
View larger version (104K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 16b. Misregistration due to physiologic bladder activity in a 58-year-old woman with intermediate to poorly differentiated Sertoli-Leydig cell tumor with heterologous elements. (a) CT scan shows a cystic mass in the left side of the pelvis (arrow). (b) Fused PET-CT scan shows activity in the right side of the pelvis (arrow). (c) CT scan obtained inferior to a reveals that the pelvic activity seen in b represents bladder activity secondary to misregistration due to respiratory motion and bladder filling (arrow). The lesion in the left side of the pelvis (cf a) represents a postoperative fluid collection and shows no activity.
|
|
View larger version (159K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 16c. Misregistration due to physiologic bladder activity in a 58-year-old woman with intermediate to poorly differentiated Sertoli-Leydig cell tumor with heterologous elements. (a) CT scan shows a cystic mass in the left side of the pelvis (arrow). (b) Fused PET-CT scan shows activity in the right side of the pelvis (arrow). (c) CT scan obtained inferior to a reveals that the pelvic activity seen in b represents bladder activity secondary to misregistration due to respiratory motion and bladder filling (arrow). The lesion in the left side of the pelvis (cf a) represents a postoperative fluid collection and shows no activity.
|
|
View larger version (174K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 17a. Ureteral "enlargement" at PET. (a) On a CT scan, both ureters are normal in size (arrows). (b) PET scan shows intense but normal activity in the ureters (arrows), which causes the ureters to appear larger than they really are. (c) Fused PET-CT scan shows ureteric activity superimposed on the left psoas muscle (arrow).
|
|
View larger version (90K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 17b. Ureteral "enlargement" at PET. (a) On a CT scan, both ureters are normal in size (arrows). (b) PET scan shows intense but normal activity in the ureters (arrows), which causes the ureters to appear larger than they really are. (c) Fused PET-CT scan shows ureteric activity superimposed on the left psoas muscle (arrow).
|
|
View larger version (117K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 17c. Ureteral "enlargement" at PET. (a) On a CT scan, both ureters are normal in size (arrows). (b) PET scan shows intense but normal activity in the ureters (arrows), which causes the ureters to appear larger than they really are. (c) Fused PET-CT scan shows ureteric activity superimposed on the left psoas muscle (arrow).
|
|
|
Conclusions
|
|---|
PET and PET-CT have a potential role in evaluating patients for recurrent ovarian cancer, particularly those with negative CT or MR imaging findings and rising tumor marker levels. Fused PET-CT scans obtained with combined scanners are highly accurate. A combination of anatomic and functional imaging can be used to diagnose disease with increased confidence at CT. Fused PET-CT scans can help localize pathologic activity and differentiate this activity from physiologic uptake. Additional investigation is necessary to determine the possible benefits of lesion conspicuity at PET and anatomic localization at CT on combined PET-CT scans.
|
Footnotes
|
|---|
Abbreviations: FDG = 2-[fluorine-18]fluoro-2-deoxy-D-glucose,
SUV = standardized uptake value
|
References
|
|---|
Top
Abstract
Introduction
