Ovarian cancer accounts for 3% of all cases of cancer in females; however, it has the highest mortality of all gynecological cancers (1). Most patients have advanced stage disease at presentation due to a paucity and incidious onset of symptoms. Initial treatment consists of cytoreductive surgery and adjuvant platinum-based cytotoxic chemotherapy (1,2). Despite high response after initial treatment, 20–30% of patients with early-stage disease (stage IA–IIA) and up to 75% of patients with advanced disease (IIB–IV) present with recurrence within two years (2,3).

Imaging examinations, including ultrasound (US), computed tomography (CT) and magnetic resonance imaging (MRI), should be performed during patient follow-up when there is clinical suspicion of ovarian cancer recurrence or CA-125 elevation. A systematic review and meta-analysis by Gu et al(4) evaluated CA-125 levels, positron emission tomography (PET) alone, PET/CT, CT and MRI in diagnosing recurrent ovarian carcinoma. CA-125 levels had the highest pooled specificity (93%), while PET/CT had the highest pooled sensitivity (91%). CT (sensitivity, 79%; specificity, 84%) and MRI (sensitivity, 75%; specificity, 78%) had a similar diagnostic performance.

While the level of CA-125 has been shown to be a sensitive marker for tumor recurrence and levels may rise 3 to 6 months before there is clinically apparent disease, it does not provide information concerning the size and distribution of the lesions (3–5). Levels may also increase in a number of benign conditions, not being a specific marker for ovarian cancer, and a number of patients with relapse of disease present with normal CA-125 levels (4,5). CT has low sensitivity for detecting disease recurrence, probably due to its inability to detect small peritoneal implants and normal-sized lymph node metastases (4,6). Pelvic MRI is useful for the evaluation of local recurrence of disease, however, it has low specificity due to post-surgical anatomical alterations (4,7).

2-deoxy-2-(¹⁸F)-fluoro-D-glucose (¹⁸FDG) PET/CT may play an important role in ovarian cancer recurrence, as the metabolic tracer is able to increase lesion detection, the fusion of metabolic and anatomical imaging aids the determination of the exact location of disease and it is capable of surveying the whole body. Several studies have examined the performance of PET/CT scanning in patients with recurrent ovarian cancer (4,8–14).

An Australian prospective, multi-center cohort study of 90 females (14) assessed the impact of ¹⁸FDG PET/CT in the management of patients with suspected recurrent ovarian cancer and evaluated information provided by ¹⁸FDG PET/CT in this clinical context. PET/CT detected 168 additional and unsuspected sites of disease in 61 patients (67.8%). The management was changed in 53 patients (58.9%) based on PET/CT scan findings. PET/CT was superior to abdominal and pelvic CT in the detection of nodal, peritoneal and subcapsular liver disease and it also allowed the identification of patients whose disease was likely to progress within 12 months. The authors suggested that PET/CT should be the preferred imaging modality in patients with suspected ovarian carcinoma recurrence.

The aim of the present study was to evaluate the use of ¹⁸FDG PET/CT in patients with suspicion of ovarian cancer recurrence and describe the distribution of metastasis.

A total of 42 patients were diagnosed with recurrence of ovarian cancer after surgery or during clinical follow-up. Three patients remained free of disease during clinical follow-up. CA-125 levels were raised in a total of 34 patients, 14 patients had clinical suspicion of recurrence and 23 presented with alterations on US, CT or MRI. There were 11 patients with raised CA-125 levels and normal imaging examinations. The characteristics of the patients according to the PET/CT findings are shown in Table II.

¹⁸FDG PET/CT scan was positive in all 42 patients who were confirmed to have recurrence of disease. ¹⁸FDG PET/CT scan was negative in 3 patients, all free from disease during follow-up, with normal CA-125 levels and no evidence of disease on imaging examinations. One of the patients without ovarian cancer recurrence presented with focal abnormal uptake in the right thyroid lobe (SUV, 17.0) and a new primary tumor was diagnosed following surgery.

There were 11 patients with elevated CA-125 levels and normal conventional imaging, all with positive PET/CT findings. However, of the 11 patients with normal CA-125 levels, eight presented with a positive PET/CT scan (Fig. 1). Four of those patients had localized pelvic or abdominal disease (two of which were amenable to surgical resection) and 4 had supra-diaphragmatic lymph node metastasis. Patients with elevated CA-125 levels tended to have a more disseminated disease.

Overall, lymph nodes were the most frequent site of relapse of disease (Table III), being localized to the pelvic/abdominal region in 30 patients (66.7%) and the thoracic region in 16 (35.6%). Six patients had internal mammary metastasis, with surgical resection in two patients. Peritoneal implants were found in 27 patients (60%). Distant sites of metastasis included the liver (n=6), spleen (n=2), pleura (n=2), lung (n=2) and bone (n=2). However, with regard to the subgroup of 11 patients with elevated CA-125 levels and normal conventional imaging, peritoneal implants were the most frequent site of relapse of disease (n=9), followed by lymph nodes, most being in the pelvic/abdominal region (n=7).

PET/CT found unsuspected lesions in 20 out of 45 patients (44.4%), most being supra-diaphragmatic lymph node metastases or normal sized abdominal lymph nodes with abnormal ¹⁸FDG uptake (Fig. 2). There was no statistical difference when comparing the SUVs of lesions from different tumor types (P=0.6683; Table IV).

A total of 12 patients (26%) died during follow-up (mean ± SD, 15.3±7.9 months after examination; range, 6–35 months). Five of these patients had disseminated abdominal and supra-diaphragmatic disease, while 7 had disease limited to the pelvic/abdominal region.

PET/CT correctly diagnosed patients with suspected ovarian cancer recurrence. All patients with elevated CA-125 levels and normal conventional imaging had positive PET/CT scan. However, most patients with normal CA-125 levels in this series presented with positive PET/CT scan. Lymph nodes were the most frequent site of relapse of disease, most being in the pelvic/abdominal region and others in the thoracic region. Peritoneal implants were found in more than half of patients. Distant sites of metastasis included the liver, spleen, pleura, lung and bone. PET/CT detected unsuspected lesions in almost half of the patients, most being supra-diaphragmatic lymph node metastasis.

Our population had a high pre-test probability of disseminated disease, given that 18/45 patients (40%) had previously had relapse of ovarian cancer and the referral to PET/CT was to evaluate the progression of the disease and to restage. The advantage of PET/CT in this clinical setting was ability to evaluate the whole body that may aid the correct selection of patients who are amenable to surgical resection.

Most of our findings are in accordance with those previously described in the literature, with the exception of the high prevalence of supra-diaphragmatic lymph node metastases (7–14). Iagaru et al(9) retrospectively evaluated 43 patients with ovarian carcinoma and described the distribution of extra-pelvic metastases in 19 patients, 5 of which (11.6%) had supra-diaphragmatic lymph node involvement. A prospective, multi-center Australian study (14) showed that 14 out of 90 patients (15%) presented with disease above the diaphragm.

In the Australian study (14), PET/CT findings changed the management plan with medium to high impact in 58% of patients, due to its ability to detect unsuspected lesions and the advantage of the whole body evaluation.

The change in management based on PET/CT was also previously described by Simcock et al(8). The authors prospectively evaluated 56 females with ovarian cancer who underwent PET/CT scan for suspicion of recurrence or surveillance with no evidence of diseacse. PET/CT altered the apparent disease distribution in 40 scans (61%), with a high impact on management plan in 32 patients (57%).

Numerous clinicians routinely measure the level of CA-125 since it is often the first evidence of ovarian cancer recurrence and may rise 3 to 6 months before clinical evidence of disease (3–5). However, without localized disease, there is no rationale to initiate treatment based on a laboratory test alone and this can cause considerable patient anxiety. Therefore, it is important to have an accurate method that is able to both diagnose and localize the recurrence of disease and aid the selection of patients amenable to surgical resection.

Recent meta-analyses evaluated CT, MRI, PET and PET/CT for the detection of metastatic lymph nodes in patients with ovarian cancer (15). PET and PET/CT were a more accurate modality for lymph node metastasis detection, with a global pooled sensitivity of 73.2% and a specificity of 96.7%. However, the greater specificity of PET or PET/CT compared with those of CT or MRI was statistically insignificant. CT and MRI showed similar diagnostic performance, with pooled sensitivity of 42.6 and 54.7% and pooled specificity of 95.0 and 88.3%, respectively.

PET/contrast-enhanced CT in the same study (15) showed a sensitivity of 84.4% and specificity of 97.4%, which was better than non-contrast-enhanced PET/CT. Certain authors believe that the CT portion of the study should be performed without contrast (8). Those who defend the use of contrast usually perform a non-contrast-enhanced CT before the diagnostic contrast-enhanced CT, for the purposes of attenuation correction of the PET images. Yau et al(16) showed that application of intravenous contrast does not interfere with the diagnostic value of PET/CT when contrast-enhanced CT is used for attenuation correction purposes.

The PET/CT evaluation of pelvic and abdominal regions may be challenging due to urinary excretion and bladder concentration of ¹⁸FDG. Contrast material may aid the distinguishing of vessels and urethers from small nodal disease, which can result in better sensitivity of the PET/CT scan. This may be of particular importance in patients with ovarian cancer, since most metastases involve the pelvic and abdominal lymph nodes or implants. Certain authors (17) suggest the use of diuretics and dual time imaging as another way of improving the sensitivity of the examination for detection of pelvic lesions.

A limitation of our study is that there was no pathological confirmation of all the sites of abnormal ¹⁸FDG uptake. However, the confirmation of all the sites would not have been ethical solely for the purpose of validation of PET/CT findings. Accurate surgical assessment of pelvic and retroperitoneal lymph nodes is difficult, and surgery appears to be an unreliable gold standard, with disease recurrence in a third of females with negative surgical findings (18). We agree with Simcock et al(8) and believe that the course of disease and clinical outcomes may more accurately validate PET/CT data.

Another limitation was that we did not have data concerning the treatment plans prior to the PET/CT, therefore, it was not possible to evaluate the change in management in our study. However, PET/CT revealed unsuspected lesions in 44.4% of our patients, which is in accordance with previously published data.

There is no evidence that PET/CT improves the overall survival of patients diagnosed with ovarian cancer recurrence. However, the whole body examination shows the extent of the disease. This may aid the correct restaging of patients considered for further treatment.

In conclusion, ¹⁸FDG PET/CT was an accurate and useful tool for diagnosing ovarian cancer recurrence. The advantage of a whole body scan and metabolic imaging is that it may aid the detection of additional sites of disease. Supra-diaphragmatic disease in this series of patients with suspicion of ovarian cancer recurrence was more frequent than previously described.