Detection of Primary Lesions
Pooled diagnostic indices for detection of the endometrial cancer lesions (primary lesion detection) were as follows: sensitivity 81.8% (77.9%–85.3%), specificity 89.8% (79.2%–96.2%), LR+ 5.38 (2.58–11.22), LR− 0.21 (0.096–0.47), and DOR 30.22 (9.25–98.71).
Supplementary file 1, http://links.lww.com/IGC/A190, shows the forest plots of pooled sensitivity and specificity for primary lesion detection.
Subgroup analysis for the instrument used in the included studies (PET vs PET/CT) showed 70.4% (62.5%–77.5%) sensitivity for PET and 88.5% (84%–92.2%) sensitivity for PET/CT subgroups.
Lymph Node Staging
Pooled diagnostic indices for lymph node staging are shown in Table 2. Patient basis and region basis data were pooled separately. Diagnostic performances for pelvic and para-aortic nodes were shown separately too.
Supplementary file 2, http://links.lww.com/IGC/A190, shows the forest plots of pooled sensitivity and specificity for lymph node staging.
Subgroup analysis for the instrument used in the included studies (PET [4 studies, 181 patients] vs PET/CT [8 studies, 332 patients]) showed 75.7% (58.8%–88.2%) sensitivity and 92.4% (86.7%–96.1%) specificity for PET group and 68.7% (57.7%–78.2%) sensitivity and 92.7% (90%–94.9%) specificity for PET/CT subgroups.
Pooled diagnostic indices for distant metastases were as follows: sensitivity 95.7% (85.5%–99.5%), specificity 95.4% (92.7%–97.3%), LR+ 12.87 (6.32–26.2), LR− 0.11 (0.04–0.27), and DOR 167.6 (51.3–547). Supplementary file 3, http://links.lww.com/IGC/A190, shows the forest plots of pooled sensitivity and specificity for distant metastasis detection.
Heterogeneity, Publication Bias Evaluation, Threshold Effect
Table 3 shows the results of heterogeneity and publication bias evaluations for primary lesion detection, overall lymph node staging, and distant metastasis detection. Supplementary file 4, http://links.lww.com/IGC/A190, shows the funnel plots of sensitivity and specificity pooling for lymph node staging.
Threshold effect was also evaluated, and correlation coefficients between false-positive and true-positive rates are given in Table 3. Q* and AUC are also provided. Funnel plots and summary receiver operating characteristic curves are not shown here.
Similar to other malignant tumors, proper treatment planning for endometrial cancer is highly dependent on staging of the patients. Complete staging of endometrial cancer (as recommended by the International Federation of Gynecology and Obstetrics [FIGO]) is by surgery, which can have postoperative complications. Several methods such as CT scanning, MRI, and ultrasonography are being used for preoperative imaging of endometrial cancer. However, the diagnostic accuracy of these methods is limited, and surgery is still the cornerstone of staging.35,36
18F-FDG PET has been used for preoperative staging of endometrial cancer in several studies. In the current systematic review, we summarized the available literature in this regard. For convenience, we divided our systematic review into 3 separate parts: detection of primary tumor, lymph node staging, and distant metastasis detection.
Detection of Primary Tumor
Despite high specificity (89.8%), pooled sensitivity for detection of primary endometrial cancer was suboptimal (81.8%). Nonmalignant physiological uptake of 18F-FDG in the normal endometrium is most likely the cause of suboptimal sensitivity. It is reported that 18F-FDG uptake in the endometrial tissue increases in the ovulatory and menstrual phases in the premenopausal women. Even using oral contraceptives can affect 18F-FDG uptake in the endometrium.35
Subgroup analysis regarding the instrument used in the included studies (PET vs PET/CT) showed some improvement in the sensitivity (88.5%), which is most likely due to added value of anatomical information of PET/CT images over PET images alone.
Although pooled specificity for detection of primary endometrial cancer was high in our meta-analysis, we should be cautious for interpretation. The spectra of patients included in the current meta-analysis were not wide enough for evaluation of specificity. Actually only 4 studies evaluated the specificity,25,29,32,34 and 2 of them included only patients with physiological uterine activity without any endometrial tumoral lesion for this purpose.25,32 Future studies should be conducted on a wide range of patients with factors suggestive of endometrial cancer for better evaluation of specificity.37
Lymph Node Staging
Pelvic and para-aortic lymph node dissection is the recommended method for lymph node staging in endometrial cancer. However, only few endometrial cancer patients have pelvic or para-aortic lymph node involvement, and most patients do not need lymphadenectomy. This fact, in addition to the invasive nature of lymphadenectomy, is the main reason of seeking noninvasive methods for preoperative lymph node staging in endometrial cancer.38 Cross-sectional anatomical imaging such as CT and MRI and sentinel node biopsy have been used for preoperative lymph node staging with various results.39
In the current study, we also evaluated the accuracy of 18F-FDG PET imaging for preoperative lymph node staging in endometrial cancer. A recent systematic review by Chang et al40 also specifically reported on the same topic. Our systematic review has several advantages over the study of Chang et al40: our search strategy was more complete as we could locate 16 studies compared with 7 studies of the report of Chang et al.40 In our study, we took into account the duplicate reports; in contrast, Chang et al40 included both Picchio et al27 and Signorelli et al23 studies, which had duplicate information into their meta-analysis, and this could invalidate their results to some extent. Finally, we evaluated anatomical and region-based reports in detail (pelvic and para-aortic locations separately as well as region-based reports), in contrast to Chang and colleagues’40 study.
The pooled sensitivity of 18F-FDG for detection of lymph node involvement was low (72.3% for patient basis and 64.6% for region basis analyses), which limits its usefulness in lymph node staging of endometrial cancer. The pooled specificity was much higher (92.9% patient basis and 97% region basis analyses), which shows that positive results of 18F-FDG PET in the lymph nodes are much more reliable than the negative ones.
Considering pelvic and para-aortic lymph nodes separately, diagnostic performance of 18F-FDG for para-aortic lymph node staging was better than the pelvic region (higher sensitivity and specificity both).
Subgroup analyses regarding PET versus PET/CT instruments showed higher sensitivity and lower specificity for PET compared with PET/CT studies. However, these differences were not statistically different between PET and PET/CT studies (P = 0.3 for sensitivity and P = 0.8 for specificity).
The previously mentioned results should be interpreted with caution especially for para-aortic lymph nodes. First of all, the number of endometrial cancer patients with involved lymph nodes is low (overall 137 patients in our systematic review), with resulting low precision for sensitivity estimation. Wide confidence intervals in Table 2 (especially for para-aortic lymph node staging) also support this fact. Another word of caution is the criterion standard of lymph node staging in the included studies. Although pelvic lymphadenectomy was used in all but 2 studies29,33 consistently, para-aortic lymphadenectomy was done in selected patients. This can cloud the results with uncertainty and make inferential assumptions difficult to interpret as this bias can considerably change the estimated sensitivity and specificity for para-aortic lymph node staging.
Altogether, sensitivity of 18F-FDG PET imaging is not high enough to justify its routine use for preoperative lymph node staging in endometrial cancer. On the other hand, specificity of 18F-FDG PET imaging is adequate. Further studies with large sample size as well as consistently applied pelvic and para-aortic lymphadenectomy are definitely needed to be able to draw any more specific conclusion.
Diagnostic accuracy of 18F-FDG imaging for detection of distant metastases was very high (both >90%), which shows the potential role in the distant metastases evaluation. However, the number of endometrial cancer patients with distant metastases in the included studies was small (46 patients overall), and this conclusion should be used with caution pending larger multicenter studies results.
Potential locations and pathologies with 18F-FDG uptake that could cause false-positive studies were as follows: benign lung lesions,18,22 bowels,18 second malignancies,19,22,29 colon adenoma,22 and reactive lymph nodes (besides pelvic and para-aortic lesions).22,27
Heterogeneity of included studies is an important issue that should be addressed in all meta-analyses. As shown in Table 3, heterogeneity in our study does not seem to be a major issue except for the sensitivity pooling of primary tumor detection (P values of Cochrane Q tests for lymph node staging and distant metastasis detection were 0.05 and 0.57 for sensitivity and 0.13 and 0.12 for specificity pooling, respectively; for primary tumor detection, the P values were <0.0001 and 0.07 for sensitivity and specificity pooling). In other words, variations of diagnostic indices across included studies were acceptable (although not perfectly low) and do not seem to affect the results of our meta-analysis.
Threshold effect is a major source of heterogeneity in diagnostic studies meta-analyses. As shown in Table 3, this was not an important limitation in the current meta-analysis because AUC and Q* values were all high, and correlation coefficients between false-positive and true-positive rates were low (0, −0.317, and −0.257 with P = 1, 0.29, and 0.62 for primary lesion detection, lymph node staging, and distant metastasis detection, respectively). In other words, diagnostic performance of 18F-FDG PET imaging was not affected with the cutoff value for scan positivity.
Publication bias was another important issue we considered in Table 3. The regression intercepts of Egger et al15 were statistically significant in all but 1 pooling, which shows possible important publication bias. Quantification of publication bias importance using trim-and-fill method resulted in up to 7.4% decrease in pooled diagnostic indices. This means that publication bias; if present, can modify the results of our meta-analysis considerably and is a major limitation for our study.
Another limitation of our systematic review is the quality of the included studies. As shown in Table 1, not all included studies were of high quality. Most of the included studies were of levels 3 (9 studies) and 4 (2 studies). This can be considered as another limitation for our study.
Because of low sensitivity, diagnostic utility of 18F-FDG PET imaging is limited in primary tumor detection and lymph node staging of endometrial cancer patients. However, high specificities ensure high positive predictive values in these 2 indications. Diagnostic performance of 18F-FDG PET imaging is much better in detection of distant metastases. Larger studies with better design are needed to draw any more definite conclusion.
1. Siegel R, Ward E, Brawley O, et al. Cancer statistics, 2011: the impact of eliminating socioeconomic and racial disparities on premature cancer deaths. CA Cancer J Clin. 2011; 61: 212–236.
2. Zaino RJ, Kurman RJ, Diana KL, et al. Pathologic models to predict outcome for women with endometrial adenocarcinoma: the importance of the distinction between surgical stage and clinical stage—a Gynecologic Oncology Group study. Cancer. 1996; 77: 1115–1121.
3. Boronow RC, Morrow CP, Creasman WT, et al. Surgical staging
in endometrial cancer
: clinical-pathologic findings of a prospective study. Obstet Gynecol. 1984; 63: 825–832.
4. Bhosale P, Iyer R. Diagnostic imaging in gynecologic malignancy. Minerva Ginecol. 2008; 60: 143–154.
5. Dodge JE, Covens AL, Lacchetti C, et al. Preoperative identification of a suspicious adnexal mass: a systematic review
. Gynecol Oncol. 2012; 126: 157–166.
6. Liao LJ, Lo WC, Hsu WL, et al. Detection of cervical lymph node metastasis in head and neck cancer patients with clinically N0 neck-a meta-analysis
comparing different imaging modalities. BMC Cancer. 2012; 12: 236.
7. Sadeghi R, Gholami H, Zakavi SR, et al. Accuracy of 18
F-FDG PET/CT for diagnosing inguinal lymph node involvement in penile squamous cell carcinoma: systematic review
of the literature. Clin Nucl Med. 2012; 37: 436–441.
8. Basu S, Li G, Alavi A. PET and PET-CT imaging of gynecological malignancies: present role and future promise. Expert Rev Anticancer Ther. 2009; 9: 75–96.
9. Kadkhodayan S, Shahriari S, Treglia G, et al. Accuracy of 18-F-FDG PET imaging in the follow up of endometrial cancer
patients: systematic review
of the literature. Gynecol Oncol. 2013; 128: 397–404.
11. Deville WL, Buntinx F, Bouter LM, et al. Conducting systematic reviews of diagnostic studies: didactic guidelines. BMC Med Res Methodol. 2002; 2: 9.
12. DerSimonian R, Laird N. Meta-analysis
in clinical trials. Control Clin Trials. 1986; 7: 177–188.
13. Moses LE, Shapiro D, Littenberg B. Combining independent studies of a diagnostic test into a summary ROC curve: data-analytic approaches and some additional considerations. Stat Med. 1993; 12: 1293–1316.
14. Walter SD. Properties of the summary receiver operating characteristic (SROC) curve for diagnostic test data. Stat Med. 2002; 21: 1237–1256.
15. Egger M, Davey Smith G, Schneider M, et al. Bias in meta-analysis
detected by a simple, graphical test. BMJ. 1997; 315: 629–634.
16. Duval S, Tweedie R. Trim and fill: a simple funnel-plot-based method of testing and adjusting for publication bias in meta-analysis
. Biometrics. 2000; 56: 455–463.
17. Zamora J, Abraira V, Muriel A, et al. Meta-DiSc: a software for meta-analysis
of test accuracy data. BMC Med Res Methodol. 2006; 6: 31.
18. Horowitz NS, Dehdashti F, Herzog TJ, et al. Prospective evaluation of FDG-PET for detecting pelvic and para-aortic lymph node metastasis in uterine corpus cancer. Gynecol Oncol. 2004; 95: 546–551.
19. Suzuki R, Miyagi E, Takahashi N, et al. Validity of positron emission tomography using fluoro-2-deoxyglucose for the preoperative evaluation of endometrial cancer
. Int J Gynecol Cancer. 2007; 17: 890–896.
20. Nayot D, Kwon JS, Carey MS, et al. Does preoperative positron emission tomography with computed tomography predict nodal status in endometrial cancer
? A pilot study. Curr Oncol. 2008; 15: 123–125.
21. Kitajima K, Murakami K, Yamasaki E, et al. Accuracy of 18
F-FDG PET/CT in detecting pelvic and paraaortic lymph node metastasis in patients with endometrial cancer
. AJR Am J Roentgenol. 2008; 190: 1652–1658.
22. Park JY, Kim EN, Kim DY, et al. Comparison of the validity of magnetic resonance imaging and positron emission tomography/computed tomography in the preoperative evaluation of patients with uterine corpus cancer. Gynecol Oncol. 2008; 108: 486–492.
23. Signorelli M, Guerra L, Buda A, et al. Role of the integrated FDG PET/CT in the surgical management of patients with high risk clinical early stage endometrial cancer
: detection of pelvic nodal metastases. Gynecol Oncol. 2009; 115: 231–235.
24. Inubashiri E, Hata K, Kanenishi K, et al. Positron emission tomography with the glucose analog [F]-fluoro-2-deoxy-D-glucose for evaluating pelvic lymph node metastasis in uterine corpus cancer: comparison with CT and MRI findings. J Obstet Gynaecol Res. 2009; 35: 26–34.
25. Tsujikawa T, Yoshida Y, Kudo T, et al. Functional images reflect aggressiveness of endometrial carcinoma: estrogen receptor expression combined with 18
F-FDG PET. J Nucl Med. 2009; 50: 1598–1604.
26. Klar M, Meyer PT, Hancke K, et al. Evaluation of FDG-PET for detecting lymph node metastasis in uterine corpus cancer. Anticancer Res. 2010; 30: 3787–3790.
27. Picchio M, Mangili G, Samanes Gajate AM, et al. High-grade endometrial cancer
: value of [(18)F]FDG PET/CT in preoperative staging
. Nucl Med Commun. 2010; 31: 506–512.
28. Sang LH, Kim SK, Kim TS. The accuracy of nodal staging
with preoperative 18
F-FDG PET compared with MRI in patients with endometrial cancer
. J Nucl Med. 2010; 51: 15.
29. Suga T, Nakamoto Y, Saga T, et al. Clinical value of FDG-PET for preoperative evaluation of endometrial cancer
. Ann Nucl Med. 2011; 25: 269–275.
30. Lee HJ, Ahn BC, Hong CM, et al. Preoperative risk stratification using (18)F-FDG PET/CT in women with endometrial cancer
. Nuklearmedizin. 2011; 50: 204–213.
31. Nakamura K, Hongo A, Kodama J, et al. The measurement of SUVmax of the primary tumor is predictive of prognosis for patients with endometrial cancer
. Gynecol Oncol. 2011; 123: 82–87.
32. Kim HJ, Kang WJ, Cho A, et al. Comparison of FDG PET-CT and MRI in lymph node staging
of endometrial cancer
. J Nucl Med. 2011; 52: 1854.
33. Antonsen SL, Jensen LN, Loft A, et al. MRI, PET/CT and ultrasound in the preoperative staging
of endometrial cancer
—a multicenter prospective comparative study. Gynecol Oncol. 2013; 128: 300–308.
34. Antonsen SL, Loft A, Fisker R, et al. SUVmax of (18)FDG PET/CT as a predictor of high-risk endometrial cancer
patients. Gynecol Oncol. 2013; 129: 298–303.
35. Iyer RB, Balachandran A, Devine CE. PET/CT and cross sectional imaging of gynecologic malignancy. Cancer Imaging. 2007; 7 spec no. A: S130–S138.
36. Avril N, Gourtsoyianni S, Reznek R. Gynecological cancers. Methods Mol Biol. 2011; 727: 171–189.
37. Kitajima K, Murakami K, Kaji Y, et al. Spectrum of FDG PET/CT findings of uterine tumors. AJR Am J Roentgenol. 2010; 195: 737–743.
38. Ayhan A, Celik H, Dursun P. Lymphatic mapping and sentinel node biopsy in gynecological cancers: a critical review of the literature. World J Surg Oncol. 2008; 6: 53.
39. Selman TJ, Mann CH, Zamora J, et al. A systematic review
of tests for lymph node status in primary endometrial cancer
. BMC Womens Health. 2008; 8: 8.
40. Chang MC, Chen JH, Liang JA, et al. 18
F-FDG PET or PET/CT for detection of metastatic lymph nodes in patients with endometrial cancer
: a systematic review
. Eur J Radiol. 2012; 81: 3511–3517.
18F-FDG PET; Endometrial cancer; Staging; Systematic review; Meta-analysis
Supplemental Digital Content
© 2013 by the International Gynecologic Cancer Society and the European Society of Gynaecological Oncology.