Appropriate and accurate pre-operative imaging in epithelial ovarian cancer patients may allow selection of patients that may mostly benefit either from primary cytoreductive surgery or from neoadjuvant chemotherapy. If the patient is considered suitable for upfront surgery, pre-operative imaging may help in planning the surgical approach, to forecast the operating time, and to estimate the need for other consulting specialists. Currently, computed tomography (CT) imaging is the standard of care for pre-operative evaluation of ovarian cancer patients; however, advanced magnetic resonance imaging (MRI) is emerging as a technique that may overcome the limitations of CT imaging, especially for small peritoneal deposits in difficult-to-resect sites. Positron emission tomography (PET)-CT imaging in the pre-operative setting is currently limited, whereas the use of the new hybrid technique PET-MRI is still under evaluation. Since criteria that may preclude optimal cytoreductive surgery may vary, depending on the aggressiveness of the surgical procedure and surgeon skill, multidisciplinary consensus conferences are the ideal platform to evaluate extent of the disease and surgical strategy.
- ovarian neoplasms
- surgical oncology
Statistics from Altmetric.com
Epithelial ovarian cancer is the fifth cause of cancer death in women in developed countries, with 22 530 new cases estimated in the United States in 2019.1 2 The current standard treatment for epithelial ovarian cancer is primary cytoreductive surgery with complete resection of all macroscopic disease,3 4 followed by platinum-based chemotherapy. Recent studies show that interval cytoreductive surgery after neoadjuvant chemotherapy allows comparable outcomes in patients with International Federation of Gynecology and Obstetrics (FIGO) stage IIIC or IV ovarian cancer.3 5
Appropriate and accurate pre-operative imaging may allow selection of patients that may mostly benefit either from primary cytoreductive surgery or from neoadjuvant chemotherapy. If the patient is considered suitable for upfront surgery, pre-operative imaging may help in planning the surgical approach, to forecast the operating time, and to estimate the need for consulting specialists. Criteria that may preclude optimal cytoreductive surgery may vary, depending on the aggressiveness of the surgical procedure and surgeon skill, as well as on the performance status of the patient.
Therefore, after a detailed assessment of the sites of disease, the possibility of achieving optimal cytoreductive surgery should be evaluated within the multidisciplinary team and then be adapted to the institutional protocols.6
Computed Tomography (CT)
According to the guidelines of the European Society of Urogenital Radiology (ESUR), the imaging modality of choice for pre-operative evaluation of ovarian cancer patients is abdominopelvic computed tomography (CT), extended to the chest when extra-abdominal disease is suspected,7 and sites of disease may be reported according to a quadrant approach.8 In the right upper quadrant, deposits can be better evaluated by CT multiplanar reformatting (Figure 1), especially in the presence of ascites.8 9 The presence of subcapsular implants in the region extending between the Morrison pouch and the inferior vena cava at the level of the right hepatic vein should be carefully evaluated, because of an increased risk of intra-operative bleeding that, according to the institutional experience, may preclude optimal debulking.9 In the left upper quadrant, subcapsular splenic deposits that produce a scalloped appearance of the peripheral parenchyma, and focal hilar implants, should be reported separately, as the latter may require splenectomy. Involvement of the lesser sac, seen on axial images between the splenorenal ligament and the gastrosplenic ligament, should be carefully evaluated because it may preclude optimal cytoreductive surgery.8 9
The evaluation of the upper abdominal quadrants will also include evaluation of the lower thorax, where cardiophrenic lymph nodes and pleural effusion may be detected. Unlike lymphadenopathies in other sites, cardiophrenic lymph nodes in epithelial ovarian cancer are considered pathologic when the short axis is >5 mm.7 According to different institutional protocols, these lymph nodes may or may not preclude optimal debulking.10 Pleural effusion may be either reactive to the presence of carcinomatosis on the abdominal diaphragmatic surface, or it may be indicative of pleural carcinomatosis; therefore, attention should be paid to the presence of nodular contrast-enhanced pleural nodules.
In the central abdomen, lympadenopathy, omental deposits, and mesenteric root deposits may be present. The presence of lymph nodes with short axis >10 mm may lead the surgeon to perform lymphadenectomy (Figure 2). Retroperitoneal enlarged lymph nodes above the renal vessels and in the vicinity of the celiac axis may be considered difficult sites to resect. Furthermore, vascular anomalies, such as a retro-aortic left renal vein, a double vena cava or left vena cava, or an adjunctive renal artery (Figure 3), may be visualized in the pre-operative imaging and, when reported, may help to decrease the risk of operative and post-operative complications such as abnormal bleeding, renal injury, and partial renal necrosis.11 Omental deposits are common in ovarian cancer patients and their detection on imaging studies is not a major challenge. However, wide adhesions between the omentum and the small bowel may be difficult to discriminate, especially in thin patients. The CT appearance of mesenteric disease may vary greatly. Extensive mesenteric involvement, causing rigidity and retraction, is a major issue for the surgeon to achieve complete cytoreduction, as well as for the radiologist to diagnose. Indeed, the limits of CT in the detection of small peritoneal, mesenteric, and serosal bowel deposits have been described.12–14 For this reason, many recent studies are considering advanced magnetic resonance imaging (MRI) for assessment of these sites of disease (see 'Advanced MRI' section).
When assessing the pelvis, the radiologist should describe the ovarian masses, including size, morphology, and laterality, with specific description of features of malignancy,7 especially if histologic diagnosis is still not available. However, the presence of small ovaries does not exclude ovarian cancer. Indeed, there is increasing evidence that high-grade serous ovarian cancer, tubal cancer, and primary peritoneal cancer share a common origin from serous tubal intraepithelial cancer,15 and cancer cells may spread to the peritoneal cavity, without enlargement of the ovaries. Intraperitoneal dissemination generates seeding of malignant cells in the Douglas pouch and vescicovaginal spaces. Although rare in epithelial ovarian cancer, pre-operative imaging may enhance the presence of invasion of the pelvic wall (suspected when the tumor lies 3 mm from the pelvic sidewall). Of note, pre-operative imaging may also allow evaluation of other pertinent entities, such as bowel obstruction, hydronephrosis, or venous thrombosis.
Advanced Magnetic Resonance Imaging (MRI)
According to Forstner et al, MRI staging should be reserved for patients with contraindications to iodinated contrast medium administration (eg, renal insufficiency or pregnancy).7 However, given the increasing evidence of the limits of CT in accurate evaluation of peritoneal tumor deposits,16 new imaging techniques, such as whole-body MRI, including functional diffusion-weighted imaging (DWI), are under evaluation for their diagnostic accuracy in all sites of disease, as well as for prediction of resectability in epithelial ovarian cancer patients.17 18 DWI gives information about cellular microstructure through evaluation of water molecule movements. Specifically, differences in diffusion of water molecules among structures rely on different cellularity.19 The lower (restricted) the diffusion, the higher the cellularity and the likelihood of cancer. Some authors used a peritoneal cancer index based on MRI, including DWI and dynamic contrast-enhanced sequences, in 35 patients (five with ovarian cancer) to match with peritoneal carcinomatosis. Radiologic–surgical correlation yielded a high match of tumor sites,20 and later it was shown that MRI, including DWI, was superior to CT in assessment of peritoneal carcinomatosis.21
DWI MRI has shown high accuracy to assess peritoneal metastases and to improve detection of nodal metastases by 17%–21% compared with standard MRI morphologic sequences.17 22 One study has demonstrated, in a small group of patients (n=32), that whole-body DWI MRI has high accuracy for characterizing primary tumors (higher than CT and similar to positron emission tomography (PET)-CT), and for peritoneal staging (higher than both CT and PET-CT).17 Later, the same group demonstrated in a larger cohort (n=161), that whole-body DWI MRI is superior to CT for staging and prediction of incomplete resection.23 Indeed, its accuracy was superior to CT for evaluation of difficult-to-resect distant metastases; for duodenum, stomach, celiac trunk carcinomatosis; for superior mesenteric artery and mesenteric root (Figure 4), as well as for suprarenal retroperitoneal lymphadenopathies.23 MRI performed with fat suppression, delayed contrast-enhanced acquisitions, oral contrast agents, and functional imaging techniques may therefore allow diagnostic sensitivity higher than that achieved by CT. For this reason, in specialized centers where whole-body DWI MR is available, it may be considered as an adjunct or alternative imaging modality for assessing the sites of disease in epithelial ovarian cancer patients.
Positron Emission Tomography (PET)-CT and PET-MRI
PET-CT is currently not considered for routine pre-operative evaluation of patients with epithelial ovarian cancer. Although fluorodeoxyglucose (FDG) PET-CT can detect lymph node and distant metastasis in ovarian and other cancers24 with high accuracy, and it may be superior to CT for nodal and distant metastases in the staging of ovarian cancer, its role is limited for local tumoral staging.25 Its usefulness in this setting is indeed reserved to an improved characterization of enlarged or normal-sized pelvic and retroperitoneal lymph nodes,26 which are not the main part of pre-operative staging in epithelial ovarian cancer.
PET-MRI is an emerging modality that combines the superior soft tissue contrast of MRI in an adaptable field of view, with the facility of detecting small but metabolically active lymph nodes on FDG PET, without the use of the radiation associated with CT.27 28 This examination assesses the locoregional extent of pelvic tumor and evaluates the entire body for nodal, peritoneal, and skeletal metastases. In PET-MRI scanners, during the PET acquisition, MRI sequences such as whole-body Dixon images, the anatomically descriptive half-Fourier acquisition single-shot turbo spin-echo images, and fluid-sensitive inversion recovery images and DWI, are acquired simultaneously. A dedicated pelvic MRI examination follows and includes dynamic intravenous gadolinium administration. Times on the current scanners are long (approximately 60–90 min).29 One concern with MRI and, thus, PET-MRI is chest (in particular lung) metastases. In hybrid imaging, PET’s poor spatial resolution is compensated by the overlaid anatomic imaging modality; but MRI has poorer spatial resolution in the chest compared with CT and, therefore, lung metastases may be missed.30 31 PET-MRI is not part of current guidelines given the novelty of the modality. However, when guidelines do suggest PET and MRI separately, PET-MRI is a possible indication. PET-MRI is still an emerging modality, and its potential role in pre-operative evaluation of epithelial ovarian cancer remains to be explored. Indications, advantages, and limitations for each of the abovementioned imaging modalities are summarized in Table 1.
Prediction of Resectability
The detection of sites of disease can be challenging for radiologists because of the known limits of imaging techniques, including (but not limited to) evaluation of mesenteric retraction, wide adhesions, and small, disseminated disease on the bowel walls. However, definition of resectability is not a general and reproducible model because it may depend on surgeon experience, on anesthesia support, and on departmental policies. A prospective, non-randomized, multicenter trial including 350 patients demonstrated that three clinical (age >60 years; CA125 ≥500 U/m; American Society of Anesthesiologists (ASA) score 3–4) and six radiologic criteria (suprarenal retroperitoneal lymph nodes; diffuse small bowel adhesions/thickening; lesions >1 cm in the small bowel mesentery; in the root of the superior mesenteric artery; in perisplenic area; in lesser sac) were significantly associated with suboptimal cytoreductive surgery.32 Later, the same authors performed a secondary analysis, looking at the ability to predict any residual disease.33 In this second analysis, they demonstrated that the same aforementioned three clinical criteria, and eight different radiologic criteria (lesions in the root of the superior mesenteric artery; splenic hilum/ligaments; lesser sac >1 cm, gastrohepatic ligament/porta hepatis, gallbladder fossa/intersegmental fissure; suprarenal retroperitoneal lymph nodes, small bowel adhesions/thickening, and moderate-severe ascites) were significantly associated with the presence of any residual disease.33 With the purpose of constructing a model for prediction of residual disease in epithelial ovarian cancer, Janco et al demonstrated that age, absence of ascites, omental cake, and diffuse peritoneal thickening on CT scan imaging independently predicted complete cytoreduction; whereas performance status, presence of diffuse peritoneal thickening and lymphadenopathy on pre-operative CT imaging were predictors of suboptimal cytoreduction.34 Nelson et al demonstrated that optimal cytoreduction was accomplished in 23/24 patients with disease scored at CT as cytoreducible.35 In a study published in 2004, two radiologists not aware of the operative outcome, retrospectively evaluated 87 pre-operative CT scans for 17 criteria indicating disease extent. The authors found that a model based on diffuse peritoneal thickening and ascites had a positive predictive value (PPV) of 68%, 52% sensitivity, and was associated with a low rate of optimal cytoreduction.36 Borley et al analyzed radiological predictors associated with cytoreductive surgery success and requirement for bowel resection by logistic regression models. In their study, the presence of lung metastases >7 mm, pleural effusion, deposits >10 mm in size on large and small bowel mesentery, and infrarenal paraaortic lymph node metastases were associated with low success rate with cytoreductive surgery.37 In a study performed on 34 patients and comparing DW MRI and exploratory laparoscopy, Espada et al demonstrated that DW MRI is an accurate technique for depicting intra-abdominal sites of implants and predicting surgical outcome in ovarian cancer with a high predictive value similar to exploratory laparoscopy.38
In an attempt to quantify the degree of peritoneal carcinomatosis, a number of scores have been considered. Sugarbaker first described the peritoneal cancer index (PCI) to evaluate peritoneal carcinomatosis in patients with colon cancer and mesothelioma. Using this score, the peritoneal cavity is divided into 13 regions and for each region the size of the largest tumor nodule is measured. In the absence of any lesions, the score is 0, for lesions <0.5 cm the score is 1, between 0.5 and 5 cm the score is 2, and if >5 cm the score is 3. The maximum score achievable is therefore 39. More recently, some authors have considered the PCI to describe peritoneal spread in advanced ovarian cancer. Llueca et al investigated the predicitive prognostic value of PCI in a series of 110 patients, demonstrating high accuracy about tumor burden and distribution, and indeed a high PCI value was associated with a lower survival rate (>10).39
The use of PCI applied to CT (CT-PCI) was considered by Diaz-Gil et al as a feasible and valid tool for evaluating 5-year survival.40 Another study evaluated the importance of PCI-CT in the selection of patients for cytoreductive surgery, whereby patients with a score >15 were recommended neoadjuvant chemotherapy. In this study, the authors concluded that PCI-CT and laparoscopy represent the best option to predict the operability in patients with advanced ovarian cancer.41 More recently, Ahmed et al showed high accuracy of CT imaging, with routine use of a standardized PCI form, for a better comprehensive multiregional analysis. CT and laparoscopy correctly detected peritoneal carcinomatosis in 88.2% and 90.6% of patients, respectively.42 Llueca et al performed a pilot study in 110 patients, where CT and laparoscopic data were used to determine PCI and lesion size score. The authors constructed two-by-two contingency tables and two predictive models. Each model included three risk score levels; one included operative PCI, the other did not. Their models predicted suboptimal or complete and optimal cytoreductive surgery with a sensitivity of 83% and 69%, respectively. The authors also showed that a PCI >20 was a major risk factor for inability of complete resection.43
Despite advances in surgical techniques, bowel involvement still represents a major limitation for optimal cytoreduction. Thus, signs of bowel and mesenteric involvement have to be carefully analyzed at pre-operative imaging. Since small peritoneal deposits (<5 mm) are difficult to see on CT imaging, laparoscopy has been considered in the pre-operative assessment of ovarian cancer.44 In a prospective comparative study with surgery as the reference standard, whole-body MRI using DWI was superior to CT and to PET-CT imaging in the assessment of bowel serosal and mesenteric disease. Furthermore, metastases outside the abdomen were detected at a similar rate to PET-CT imaging.23 Conversely, another comparative study found no significant differences between MRI, CT, and PET/CT for staging. However, PET/CT was more accurate for supradiaphragmatic metastases.45
In summary, pre-operative imaging may represent a roadmap for surgery, alerting the surgeon to the presence of difficult-to-resect sites or anatomic abnormalities. Consequently, surgery might be properly planned (eg, need for a surgical consultant; time of anesthesia, operating room time). Currently, CT imaging is the standard of care for evaluating patients with ovarian cancer, although this modality shows limitations in assessment of small peritoneal deposits on bowel serosa and mesentery. Advanced MRI is emerging as the technique that may overcome the limitations of CT imaging, this being due primarily to optimal soft tissue contrast in morphologic sequences, that may be merged with functional diffusion-weighted sequences for optimal carcinomatosis evaluation, even in difficult-to-resect sites of disease. PET-CT use in pre-operative imaging is currently limited, whereas the usefulness of the new hybrid technique PET-MRI is still under evaluation. Multidisciplinary consensus conferences are the ideal forum to evaluate patient-related factors, extent of the disease, and surgical technical issues. This will eventually warrant a reduction of under-treatments, as well as a reduction of complications due to over-treatments, thus increasing patient survival and quality of life.
Contributors SR, LM, and FDG contributed substantially to the conception and design of the work. SR, LM, MDG, and AP drafted the manuscript and revised it critically for important intellectual content. SR, MDG, LM, AP, and FDG approved the final version prior to submission.
Funding The authors have not declared a specific grant for this research from any funding agency in the public, commercial or not-for-profit sectors.
Competing interests None declared.
Patient consent for publication Not required.
Provenance and peer review Not commissioned; externally peer reviewed.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.