Gynecologic Oncology 131 (2013) 689–693

Contents lists available at ScienceDirect

Gynecologic Oncology

j ourna l homepage: www.e lsev ie r .com/ locate /ygyno

Preoperative 18F-FDG PET/CT in the management of advanced epithelialovarian cancer

Robert Fruscio a,⁎, Federica Sina a, Carlotta Dolci b, Mauro Signorelli a, Cinzia Crivellaro b,c, Tiziana Dell'Anna a,Marco Cuzzocrea a, Luca Guerra b, Rodolfo Milani a, Cristina Messa b,c

a Clinic of Obstetrics and Gynecology, University of Milan-Bicocca, San Gerardo Hospital, Monza, Italyb Nuclear Medicine Department, San Gerardo Hospital, Monza, Italyc Fondazione Tecnomed, University of Milan-Bicocca, Milan, Italy

H I G H L I G H T S

• PET/CT is able to identify stage IV EOC patients.• Residual tumor is related to the PET stage.• Response to chemotherapy is poorer in patients with PET stage IV EOC.

⁎ Corresponding author at: Clinics of Obstetrics and GBicocca, San Gerardo Hospital, Via Pergolesi, 33, 209002339433.

E-mail address: robilandia@gmail.com (R. Fruscio).

0090-8258/$ – see front matter © 2013 Elsevier Inc. All rihttp://dx.doi.org/10.1016/j.ygyno.2013.09.024

a b s t r a c t

a r t i c l e i n f o

Article history:

Received 7 August 2013Accepted 22 September 2013Available online 27 September 2013

Keywords:Epithelial ovarian cancer18F-FDG PET/CTDiagnosis

Objective. The introduction of 18-FDG-PET/CT during preoperative evaluation of patients with epithelialovarian cancer (EOC) has led to an increase of the detection of extra-abdominalmetastases. However, the clinicalimpact of this upstage remains unclear.

Methods. Patients with suspected advanced EOC underwent 18-FDG-PET/CT within two weeks prior todebulking surgery.

Results. Between 2006 and 2011 95 patients met the inclusion criteria. Based on the concordance or thediscrepancy of clinical and PET/CT stage, patients were divided into 3 groups (A: clinical and PET III; B: clinicalIII and PET IV; C: clinical and PET IV). Twenty-five patients were upstaged from FIGO stage III to stage IV by

PET/CT. The proportion of patients who achieved a residual tumor b1 cm in group B and C was similar, whereasit was significantly lower compared to group A. Similarly, complete response to adjuvant chemotherapy wasachieved more frequently in patients in group A. PFS was similar in the three groups (17, 17 and 12 months ingroup A, B and C), as well as OS (51, 41 and 35 months).

Conclusions. PET/CT is able to detect distant metastases in EOC patients. The presence of extra-abdominaldisease probably indicates amore aggressive diseasewhich also shows a lower response to standard chemotherapy.However, upstaged patients have a similar prognosis compared to stage III patients, probably becauseintra-abdominal disease is more likely to lead patients to death. This might also explain why residualtumor is the most important prognostic factor for advanced EOC patients.

© 2013 Elsevier Inc. All rights reserved.

Introduction

Ovarian cancer is the most lethal gynecological malignancy.Symptoms are usually vague and the disease is diagnosed, inapproximately 75% of patients, at an advanced stage. Ovariancancer is treated by exploratory laparotomy which has the aimof achieving an optimal debulking, removing as much tumor as

ynecology, University of MilanMonza, Italy. Fax: +39 039

ghts reserved.

possible. In fact, stage and residual tumor after the first surgeryare the most important prognostic factors for advanced stageepithelial ovarian cancer patients [1]. The preoperative evaluationof tumor dissemination, performed with CT or MRI, is crucial forplanning the adequate therapy.

The role of 18-FDG PET/CT in the management of EOC is not wellestablished. It has a high sensitivity and specificity in diagnosing EOC[2]; however, in the great majority of cases, the diagnosis of advanceddisease is adequately obtainable by much less expensive methods,such as measurement of CA125 and ultrasound [3], and only less than10% of the cases require second-stage exams [4]. PET/CT has been dem-onstrated to accurately predict response to neoadjuvant chemotherapy[5] and to detect nodal metastasis in apparently early-stage EOC [6].

Table 1Patients characteristics (n: 95).

n %

FIGO stageIIIB 8 8IIIC 75 79IV 12 13

HystologySerous 60 63Endometrioid 13 14Clear Cell 1 1Mixed 11 12Undifferentiated 10 10

Grade2 16 173 79 83

Residual tumorb1 cm 37 39N1 cm 58 61

690 R. Fruscio et al. / Gynecologic Oncology 131 (2013) 689–693

Moreover, it has shown satisfactory results in patients with recurrentdisease and negative CA125 and conventional CT imaging [7].

Recently, the introduction of PET/CT in the preoperativemanagementof EOC has led to an increase in the detection of extra-abdominalmetastasis, with the consequent migration of a percentage of patientsfrom FIGO stage III to stage IV [8,9]. However, it is not clear if thissubclinical, radiological extra-abdominal disease has a negativeprognostic impact.

The aim of this retrospective study is to evaluate the diagnosticimprovement of PET/CT in the diagnosis of distant disease in EOCpatients and to define its prediction of the oncological outcome.

Patients and methods

Study population

Patients with suspicious EOCwhose preoperative evaluation includ-ed an abdominal/chest CT scan and a 18FDG-PET/CT and underwentdebulking and staging surgery according to FIGO guidelines in the Clinicof Obstetrics and Gynecology, San Gerardo Hospital, Monza (Italy)between January 2006 and June 2011 were included in this single-centerretrospective study. Patients with distant metastasis beyond theperitoneal cavity or liver parenchymal metastasis detected at CTscan or at PET/CT were considered as FIGO stage IV. Surgery wasperformed within 15 days after the imaging study.

Exclusion criteriawere age b 18 years, early stage EOC, FIGO stage IIIEOC only for nodal localization, histologic diagnosis of mucinous,borderline, non-epithelial or metastatic ovarian cancer.

All patients received a platinum-based chemotherapy after surgery;response was evaluated using the RECIST evaluation criteria [10]. A CTscan was performed in all patients after chemotherapy, coupled with aPET scan in case of preoperative distant PET positive localizations.

The study was approved by the local IRB and all patients signed awritten informed consent.

PET/CT scanning procedure and image analysis

FDG PET/CT studies were acquired by an integrated PET/CT system,either Discovery ST or Discovery 600 (GE Healthcare, Milwaukee, WI,USA). Patients fasted for at least 6 h before the intravenous administra-tion of 300–370 MBqof 18F-FDG. Thenwere orally hydrated (500 mLofwater) during uptake time (approximately 60 min) and were asked toempty their bladder immediately before the scan in order to reducethe urinary bladder that might interfere with the optimal evaluationof the pelvis. After a scout view to define the scan length from skullbase to middle thigh, CT was acquired, during shallow breathing(140 kV, 60–80 mA and 3.75 mmof slice thickness). No oral or intrave-nous contrast media were administered. PET was acquired in 3Dmode.PET images were reconstructed on a 256 × 256 image matrix usingiterative algorithms, and corrected for random scatter. Attenuationcorrection was performed on the basis of CT data.

Image readout was performed on a Xeleris or AW4.4 WorkStation(GE Healthcare, Milwaukee, WI, US), which allows visualization ofPET, CT and fused sections in transverse, coronal and sagittal planes.The images were evaluated by two nuclear medicine physiciansinformed about clinical data of the patient at the moment of thescan. Each focal tracer accumulation deviating from the physiologicaldistribution of the tracer was considered as positive for disease andlocalized using integrated CT images.

Statistical analysis

Absolute and percentage frequencies were used to describecategorical items of patient population while mean values, standarddeviation and range were used for continuous characteristics.

Survival curves were estimatedwith the Kaplan–Meier method. Coxproportional hazards models were used for univariate and multivariateanalyses to test demographic characteristics and clinical features fortheir associations with RFS and OS. The results are expressed as hazardratios (HRs) and their 95% confidence intervals (95% CIs). Fisher's exacttest was used to compare proportions. Statistical significance was setat p b 0.05 for a bilateral test. Analysis was carried out using the SAS(Statistical Analysis System, SAS Institute, Inc., Cary, NC, Version 9.1)software.

Results

Between January 2006 and June 2011 163 patients underwentsurgery for EOC and had a 18-FDG PET/CT scan preoperatively. Sixty-eight patients were excluded from the analysis of this study becausewith FIGO early stage (I-II) EOC (N: 29), received neoadjuvant chemo-therapy (n: 13), had a mucinous histology (n: 2) had a metastatic ornon-epithelial ovarian cancer (n: 17) or had an endometrial syn-chronous tumor (n: 7). Therefore, data of 95 patients were analyzed.

Table 1 summarizes clinical and pathological data of patients. Meanage of patients was 58 years. The majority of patients had a FIGO stageIII EOC (87%), while 12 patients had a stage IV disease. In details, distantmetastases, detected by CT scan, were localized in the liver (3 patients),in the lung (1), in the pleura (3), in the spleen (1), and in multipleorgans (4).

Serous histotype was the most commonly represented (63%), andmost patients (83%) had a poorly differentiated tumor.

All FIGO stage IV patients had 18-FDG PET positive metastaticdisease, whereas 25 patients were upstaged from FIGO stage III toIV by 18-FDG PET. Therefore patients were divided in 3 groups:group A included patients with concordant FIGO and 18-FDG PETstage III (n: 58), group B patients with FIGO stage III and 18-FDGPET stage IV (n: 25) and group C patients with concordant FIGOand 18-FDG PET stage IV (n: 12).

Upstaging from stage III to IV occurred for the presence ofsupradiaphragmatic disease in 22 patients and for liver parenchymalmetastasis in 3 patients. Detailed metastatic sites for patients in groupB and C are listed in Table 2.

Primary cytoreduction was attempted in all cases; however residualtumor b1 cmwas achieved in 39% of patients, while the remaining 61%had suboptimal cytoreduction after debulking surgery. Residual tumorb1 cm was achieved in a significantly higher proportion of patients ingroup A compared to both group B and C (Table 3, chi-square testp: 0.001). Therewas no difference comparing group B and C (chi-squaretest p: 0.67).

All patients underwent adjuvant polichemotherapy with Carboplatinand Paclitaxel, receiving a median number of 6 cycles. Response to

Table 2Localization of distant metastasis in group B and C.

Group Ba (n: 25)

Supradiaphragmatic metastasis 22b

Lung parenchyma 2Pleura 4Mediastinal nodes 9Axillary nodes 4Supraclavear nodes 4Parasternal nodes 11

Liver parenchymal metastasis 3

Group Cc (n: 12)

Supradiaphragmatic metastasis 7d

Lung parenchyma 3Pleura 5Mediastinal nodes 3Axillary nodes 3Supraclavear nodes 0Parasternal nodes 6

Liver parenchymal metastasis 6Spleen parenchymal metastasis 1

a Metastasis detected by PET/CT;b 10 patients had more than one metastatic site.c Metastasis detected by CT and PET/CT.d 5 patients had more than one metastatic site.

Table 4Association between extra-abdominal FDG uptake and clinical variables.

No Yes p-Value

Residual Tumor b0.001b1 cm 31 6N1 cm 27 31

Grade 0.3752 12 43 50 29

Response to chemotherapy 0.224CR 39 16PR 16 12SD 2 2PD 4 3

Histotype 0.438Serous 38 22Endometrioid 9 4Mixed 6 5Undifferrentiated 8 2Clear Cell 1 0

FIGO clinical stage b0.001III 58 0IV 25 12

Mean CA125 1577 2420 0.192

691R. Fruscio et al. / Gynecologic Oncology 131 (2013) 689–693

chemotherapy is shown in Table 3. The percentage of patients whoachieved a complete response to adjuvant chemotherapywas significant-ly higher in group A compared to group B and C (chi-square test p: 0.03).

The detection by 18-FDG PET of metastatic disease was significantlyassociated only to residual tumor (Table 4, p b 0.001).

After a median follow up of 30 months, the overall and therecurrence-free survival rates were 65.3% and 33%, respectively.

OS and PFS for the three groups are shown in Supplementary Figs. 1and 2, respectively. A total of 33 patients died for EOC: 18/58 (31%) ingroup A, 11/25 (44%) in group B and 4/12 (33%) in group C. MedianOS was 51 months (95% CI 27–74) in group A, 41 (95% CI 15–54) ingroup B and 35 (95% CI 33–56) in group C, and it was not significantlydifferent comparing the three groups (p: 0.71).

A total of 64 patients relapsed (68%), 39/58 (67%) in group A, 16/25(64%) in group B and 9/12 (75%) in group C.Median PFSwas 17 months(95% CI 14–19) in group A, 17 (95% CI 13–20) in group B and 12 (95% CI4–19) in group C. No difference was found comparing the three groups(p: 0.35).

Table 5 shows the results of the univariate andmultivariate analyses.Residual tumor b1 cm significantly affected PFS in the univariateanalysis (HR 0.6; 95%CI 0.3–0.9, p b 0.05) and showed a non-significanttrend toward a protective effect on recurrence in themultivariate analysis

Table 3Prognostic factors and oncological outcome.

FIGO III/PETIII

FIGO III/PETIV

FIGO IV/PETIV

58 61% 25 26% 12 13%

n % n % n %

Residual tumorb1 cm 31 53% 5 20% 1 8%N1 cm 27 47% 20 80% 11 92%

Carcinosis 41 71% 19 76% 5 42%CA125 (mean) 1622 2538 1875Response to chemotherapy

CR 39 67% 11 44% 5 42%PR 12 21% 12 48% 4 33%SD 2 3% 2 8% 0 0%PD 4 7% 0 0% 3 25%

Median PFS (months) 17 17 12Median OS (months) 51 41 35

(HR 0.7; 95% CI 0.3–1.0, p: 0.06). OS was affected by residual tumor bothin the univariate (HR0.4; 95%CI 0.2–0.9, p b 0.05) and in themultivariatemodel (HR 0.4; 95% CI 0.1–0.9, p b 0.05). The detection by 18-FDG PET ofmetastatic disease did not significantly affect neither PFS nor OS.

Discussion

The introduction of 18FDG PET/CT has led, in many cases, to animprovement of the diagnosis and management of cancer patients.In the gynecological field, PET/CT has a role in the preoperative eval-uation of high risk endometrial cancer [11,12] and cervical cancerpatients [13,14]. However, the role of PET/CT in the preoperativeevaluation of EOC is controversial.

In this study we confirm the observation that PET/CT performedpreoperatively is able to detect a higher proportion of patients withextra-abdominal localization compared to conventional CT [8,9]. Evenif the standard treatment for both stages has no differences, consistingin upfront surgery and subsequent platinum based polichemotherapy[15,16], the correct localization of metastatic disease is important bothin cases with resectable metastatic disease to achieve, when possible, anoptimal residual tumor and, in all other cases, to assess the response tochemotherapy between cycles and at the end of the course.

In our population, conventional imaging detected a rate of 13% ofFIGO stage IV ovarian cancer, which is comparable to data commonlyreported in literature [17,18]. However, a remarkably high proportionof patients (26%) has been upstaged by PET/CT, raising the global per-centage of stage IV patients to nearly 40%. Metastatic sites recognizedby PET/CT causing upstaging were predominantly supradiaphragmatic,and this data confirms the recently published findings of Hynninen[8] and of Baats [19], who reported a percentage of mediastinal

Table 5Univariate and multivariate analyses for recurrence-free and overall survival.

Univariate Multivariate

HR 95% CI p-Value HR 95% CI p-Value

Progression-free survivalResidual tumora 0.6 0.3–0.9 b0.05 0.7 0.3–1.0 0.6PET stageb 0.7 0.4–1.1 0.2 0.8 0.5–1.4 0.4

Overall survivalResidual tumora 0.4 0.2–0.9 b0.05 0.4 0.1–0.9 b0.05PET stageb 0.8 0.4–1.6 0.5 0.9 0.5–2.1 0.4

a b1 cm versus N1 cm.b Stage III versus IV.

692 R. Fruscio et al. / Gynecologic Oncology 131 (2013) 689–693

involvement in 37.7% of patients. Since the lymphatic drainagepathway of the diaphragm, involved in the great majority of casesof epithelial ovarian cancer with an abdominal spread of the disease,is known to go toward the lymphatic duct through the retrosternalnodes [20,21], this observation seems realistic. The prevalence of extra-abdominal localization of the disease in our study seems to confirm thatPET/CT is more accurate in finding nodal metastasis [22] and thoracicinvolvement compared to historical data obtained with conventionalimaging such as CT and MRI. A possible explanation of this observationcould be that, in the majority of patients in group B, supradiaphragmaticnodes detected by PET/CT, but not by conventional CT,were not increasedin size, but had a deposit of metastatic disease between 5 and 10 mm.

Since we wanted to investigate if the detection of distant metastasisby PET/CT could be relevant for patients' prognosis, we tried to correlatepreoperative findings with pathological data of the tumor and with theoncological outcome of the patients. Upon univariate and multivariateanalyses, only residual tumor correlated with OS and PFS was afterdebulking surgery. This result confirms the recent observation publishedby Risum [23]. The absence of metastatic disease at PET/CT was notassociated with a statistically significant survival, although medianOS decreased from group A (51 months) to group B (41 months)and C (35 months). This might suggest that a larger population isneeded to detect differences in survival.

Optimal cytoreduction, as already reported by other authors [24],was reached in a significant lower proportion of patients with PET/CTstage IV disease, despite the surgical effort being similar in all patientsand not influenced by the spread of the disease. We acknowledge that,in patientswith a preoperative diagnosis ofmetastatic disease, surgeonsmight have been unconsciously less inclined to perform an aggressivesurgery, and this might be an explanation of the significant differenceobserved. On the other hand this observation, coupled with the signifi-cantly lower percentage of response to chemotherapy in this group ofpatients, might suggest that PET/CT is able to identify patients with amore aggressive disease. We currently do not have any reliable clinicalor histopathological predictor of EOC behavior; however the existenceof different gene and miRNA expression profiles correlating with theoncological outcome of clinically and histopathologically similar EOCpatients has recently been demonstrated and reported in several articles[25–27]. Moreover, a correlation between clinical outcome, molecularexpression profile and the PET/CT findings has recently been reportedin many types of cancers, such as colon [28], liver [29], lung [30] andbreast cancer [31,32], but not for EOC yet. These reports are extremelyinteresting as thepossibility to convergemolecular profiles andmolecularimaging in a “whole picture” open up the possibility to individualizepatient-specific treatments and adjust the therapeutic regimensrationally.

We acknowledge some drawbacks in our study that could limit thestrength of the conclusions.

First of all, the retrospective design of this study and the relativelylow number of patients considered might have determined thestatistical significance in the survival analysis.

Another limitation of our study is that we could not provide ahistological confirmation of PET positivity, as the removal of the involvedthoracic nodes is not part of the standard staging procedures and wasconsidered too invasive without a counterbalanced proven diagnosticimprovement. However, none of our patients had positive historyof pulmonary disease (e.g. tuberculosis, pulmonary infections) thatcould lead to a false positive finding due to an inflammatory condi-tion. Moreover the fact that many patients had more than onesupradiaphragmatic site suspected for metastatic disease and thatpositivity disappeared during chemotherapy might suggest that thefindings were true.

Another factor that should be taken into account is the absolute lowpercentage of patient who achieved a complete cytoreduction afterdebulking surgery, as residual tumor is the most important recognizedfactor determining the oncological outcome of ovarian cancer patients.

We started a more aggressive surgical protocol in our clinic nearly24 months ago, reaching a percentage of optimally debulked patientsof approximately 70%; however these patients have not been includedin this study because they do not have the minimum required follow-up of 24 months. However, the fact that our population has been treat-ed in a single center guarantees a uniformity of the therapeutic decisionon each patient, comparable surgical procedures and imaging studyinterpretation.

In conclusion we confirm that PET/CT is nowadays the mostpowerful diagnostic tool for the detection of the presence of meta-bolically active disease outside the abdomen in patients with advancedEOC. A preoperative distant PET/CT positivity might be considered as afunctional marker of the aggressive behavior of the tumor, but it mightbe clinically not crucial, since fatal events leading to the death of patients,as bowel perforation or occlusion, are a result of the acquired or intrinsicchemoresistance and progression of the abdominal disease, rather than ofextra-abdominal localizations.

Given the poor prognosis of these patients, finding a correlationbetween functional imaging and the molecular mechanisms underlyingthe aggressiveness of the tumor could allow both the selection of patientswho could benefit from amore aggressive surgical treatment and theidentification of new therapeutic targets specific for this high riskpopulation of patients.

Supplementary data to this article can be found online at http://dx.doi.org/10.1016/j.ygyno.2013.09.024.

Conflict of interestAll the authors have no possible conflict of interest to disclose.

References

[1] Bristow RE, Tomacruz RS, Armstrong DK, Trimble EL, Montz FJ. Survival effect ofmaximal cytoreductive surgery for advanced ovarian carcinoma during the platinumera: a meta-analysis. J Clin Oncol Mar. 1 2002;20(5):1248–59.

[2] Risum S, Høgdall C, Loft A, Berthelsen AK, Høgdall E, Nedergaard L, et al. Thediagnostic value of PET/CT for primary ovarian cancer — a prospective study.Gynecol Oncol Apr. 2007;105(1):145–9.

[3] Timmerman D, Van Calster B, Testa AC, Guerriero S, Fischerova D, Lissoni AA,et al. Ovarian cancer prediction in adnexal masses using ultrasound-basedlogistic regression models: a temporal and external validation study by theIOTA group. Ultrasound Obstet Gynecol Aug. 2010;36(2):226–34.

[4] Valentin L, Ameye L, Savelli L, Fruscio R, Leone FP, Czekierdowski A, et al. Adnexalmasses difficult to classify as benign or malignant using subjective assessment ofgray-scale and Doppler ultrasound findings: logistic regression models do nothelp. Ultrasound Obstet Gynecol Oct. 2011;38(4):456–65.

[5] Avril N, Sassen S, Schmalfeldt B, Naehrig J, Rutke S,WeberWA, et al. KuhnWPredictionof response to neoadjuvant chemotherapy by sequential F-18-fluorodeoxyglucosepositron emission tomography in patients with advanced-stage ovarian cancer. J ClinOncol Oct. 20 2005;23(30):7445–53.

[6] Signorelli M, Guerra L, Pirovano C, Crivellaro C, Fruscio R, Buda A, et al. Detectionof nodal metastases by 18F-FDG PET/CT in apparent early stage ovarian cancer: aprospective study. Gynecol Oncol Aug. 27 2013. http://dx.doi.org/10.1016/j.ygyno.2013.08.022 [S0090-8258(13)01107-4, Epub ahead of print].

[7] Thrall MM, DeLoia JA, Gallion H, Avril N. Clinical use of combined positron emissiontomography and computed tomography (FDG-PET/CT) in recurrent ovarian cancer.Gynecol Oncol Apr. 2007;105(1):17–22.

[8] Hynninen J, AuranenA, CarpénO,DeanK, SeppänenM, Kemppainen J, et al. FDGPET/CTin staging of advanced epithelial ovarian cancer: frequency of supradiaphragmaticlymph node metastasis challenges the traditional pattern of disease spread. GynecolOncol Jul. 2012;126(1):64–8.

[9] Risum S, Høgdall C, Loft A, Berthelsen AK, Høgdall E, Nedergaard L, et al. Does theuse of diagnostic PET/CT cause stage migration in patients with primaryadvanced ovarian cancer? Gynecol Oncol Mar. 2010;116(3):395–8.

[10] Eisenhauer EA, Therasse P, Bogaerts J, Schwartz LH, Sargent D, Ford R, et al. Newresponse evaluation criteria in solid tumours: revised RECIST guideline (version 1.1).Eur J Cancer Jan. 2009;45(2):228–47.

[11] Signorelli M, Guerra L, Buda A, Picchio M, Mangili G, Dell'Anna T, et al. Role of theintegrated FDG PET/CT in the surgical management of patients with high risk clinicalearly stage endometrial cancer: detection of pelvic nodal metastases. Gynecol OncolNov. 2009;115(2):231–5.

[12] Crivellaro C, Signorelli M, Guerra L, De Ponti E, Pirovano C, Fruscio R, et al. Tailoringsystematic lymphadenectomy in high-risk clinical early stage endometrial cancer:the role of 18F-FDG PET/CT. Gynecol Oncol Aug. 2013;130(2):306–11.

[13] Signorelli M, Guerra L, Montanelli L, Crivellaro C, Buda A, Dell'Anna T, et al.Preoperative staging of cervical cancer: is 18-FDG-PET/CT really effective inpatients with early stage disease? Gynecol Oncol Nov. 2011;123(2):236–40.

693R. Fruscio et al. / Gynecologic Oncology 131 (2013) 689–693

[14] Crivellaro C, Signorelli M, Guerra L, De Ponti E, Buda A, Dolci C, et al. 18F-FDG PET/CTcan predict nodal metastases but not recurrence in early stage uterine cervicalcancer. Gynecol Oncol Oct. 2012;127(1):131–5.

[15] Curtin JP, Malik R, Venkatraman ES, Barakat RR, HoskinsWJ. Stage IV ovarian cancer:impact of surgical debulking. Gynecol Oncol Jan. 1997;64(1):9–12.

[16] Liu PC, Benjamin I, Morgan MA, King SA, Mikuta JJ, Rubin SC. Effect of surgicaldebulking on survival in stage IV ovarian cancer. Gynecol Oncol Jan. 1997;64(1):4–8.

[17] Heintz AP, Odicino F, Maisonneuve P, Quinn MA, Benedet JL, Creasman WT, et al.Carcinoma of the ovary. FIGO 6th annual report on the results of treatment ingynecological cancer. Int J Gynaecol Obstet 2006;95:161–92.

[18] Bonnefoi H, A'Hern RP, Fisher C, Macfarlane V, Barton D, Blake P, et al. Naturalhistory of stage IV epithelial ovarian cancer. J Clin Oncol 1999;17:767–75.

[19] Baats AS, Hugonnet F, Huchon C, Bensaid C, Pierquet-Ghazzar N, Faraggi M, et al.Prognostic significance of mediastinal 18F-FDG uptake in PET/CT in advancedovarian cancer. Eur J Nucl Med Mol Imaging Mar. 2012;39(3):474–80.

[20] Souilamas R, Hidden G, Riquet M. Mediastinal lymphatic efferents from thediaphragm. Surg Radiol Anat 2001;23:159–62.

[21] Okiemy G, Foucault C, Avisse C, Hidden G, Riquet M. Lymphatic drainage of thediaphragmatic pleura to the peritracheobronchial lymph nodes. Surg Radiol Anat2003;25:32–5.

[22] Yuan Y, Gu ZX, Tao XF, Liu SY. Computer tomography, magnetic resonance imaging,and positron emission tomography or positron emission tomography/computertomography for detection of metastatic lymph nodes in patients with ovariancancer: a meta-analysis. Eur J Radiol May 2012;81(5):1002–6.

[23] Risum S, Loft A, Engelholm SA, Høgdall E, Berthelsen AK, Nedergaard L, et al. Positronemission tomography/computed tomography predictors of overall survival in stageIIIC/IV ovarian cancer. Int J Gynecol Cancer Sep. 2012;22(7):1163–9.

[24] Risum S, Høgdall C, Loft A, Berthelsen AK, Høgdall E, Nedergaard L, et al. Prediction ofsuboptimal primary cytoreduction in primary ovarian cancer with combined positron

emission tomography/computed tomography — a prospective study. Gynecol OncolFeb. 2008;108(2):265–70.

[25] Marchini S, Cavalieri D, Fruscio R, Calura E, Garavaglia D, Nerini IF, et al. Associationbetween miR-200c and the survival of patients with stage I epithelial ovariancancer: a retrospective study of two independent tumour tissue collections. LancetOncol Mar. 2011;12(3):273–85.

[26] Marchini S, Mariani P, Chiorino G, Marrazzo E, Bonomi R, Fruscio R, et al.Analysis of gene expression in early-stage ovarian cancer. Clin Cancer ResDec. 1 2008;14(23):7850–60.

[27] Marchini S, Marabese M, Marrazzo E, Mariani P, Cattaneo D, Fossati R, et al.DeltaNp63 expression is associated with poor survival in ovarian cancer. AnnOncol Mar. 2008;19(3):501–7.

[28] Strauss LG, Koczan D, Klippel S, Pan L, Cheng C, Willis S, et al. Impact ofangiogenesis-related gene expression on the tracer kinetics of 18F-FDG incolorectal tumors. J Nucl Med Aug. 2008;49(8):1238–44.

[29] Segal E, Sirlin CB, Ooi C, Adler AS, Gollub J, Chen X, et al. Decoding global geneexpression programs in liver cancer by noninvasive imaging. Nat Biotechnol Jun.2007;25(6):675–80.

[30] Guo J, Higashi K, Ueda Y, Ishigaki Y, Sakuma T, Oguchi M, et al. VEGF-A and itsisoform VEGF121 mRNA expression measured by quantitative real-time RT-PCR:correlation with F-18 FDG uptake and aggressiveness of lung adenocarcinoma:preliminary study. Ann Nucl Med Jan. 2011;25(1):29–36.

[31] Palaskas N, Larson SM, Schultz N, Komisopoulou E, Wong J, Rohle D, et al. 18F-fluorodeoxy-glucose positron emission tomography marks MYC-overexpressinghuman basal-like breast cancers. Cancer Res Aug. 1 2011;71(15):5164–74.

[32] Osborne JR, Port E, Gonen M, Doane A, Yeung H, Gerald W, et al. 18F-FDG PET oflocally invasive breast cancer and association of estrogen receptor status withstandardized uptake value: microarray and immunohistochemical analysis. J NuclMed Apr. 2010;51(4):543–50.