Advertisement

Improved stratification of stage-specific survival for cervical uterine cancer by integrating FDG-PET/CT and MRI for lymph node staging in 2018 FIGO classification

  • Jorun Holm
    Correspondence
    Corresponding author at: Department of Nuclear Medicine, Kloevervaenget 47, Odense University Hospital, DK-5000 Odense C, Denmark.
    Affiliations
    Department of Nuclear Medicine, Odense University Hospital, Odense, Denmark

    Department of Clinical Research, Faculty of Health Science, University of Southern Denmark, Denmark
    Search for articles by this author
  • Oke Gerke
    Affiliations
    Department of Nuclear Medicine, Odense University Hospital, Odense, Denmark

    Department of Clinical Research, Faculty of Health Science, University of Southern Denmark, Denmark
    Search for articles by this author
  • Mie Holm Vilstrup
    Affiliations
    Department of Nuclear Medicine, Odense University Hospital, Odense, Denmark
    Search for articles by this author
  • Diana Spasojevic
    Affiliations
    Department of Radiology, Odense University Hospital, Odense, Denmark
    Search for articles by this author
  • Sara Elisabeth Sponholtz
    Affiliations
    Department of Gynecology and Obstetrics, Odense University Hospital, Odense, Denmark
    Search for articles by this author
  • Kirsten Marie Jochumsen
    Affiliations
    Department of Gynecology and Obstetrics, Odense University Hospital, Odense, Denmark
    Search for articles by this author
  • Anders Thomassen
    Affiliations
    Department of Nuclear Medicine, Odense University Hospital, Odense, Denmark
    Search for articles by this author
  • Author Footnotes
    1 The authors contributed equally to the work and share the last authorship.
    Malene Grubbe Hildebrandt
    Footnotes
    1 The authors contributed equally to the work and share the last authorship.
    Affiliations
    Department of Nuclear Medicine, Odense University Hospital, Odense, Denmark

    Department of Clinical Research, Faculty of Health Science, University of Southern Denmark, Denmark
    Search for articles by this author
  • Author Footnotes
    1 The authors contributed equally to the work and share the last authorship.
    Pernille Tine Jensen
    Footnotes
    1 The authors contributed equally to the work and share the last authorship.
    Affiliations
    Department of Clinical Research, Faculty of Health Science, University of Southern Denmark, Denmark

    Department of Obstetrics and Gynecology, Aarhus University Hospital, Aarhus, Denmark

    Department of Clinical Medicine, Faculty of Health, Aarhus University, Denmark
    Search for articles by this author
  • Author Footnotes
    1 The authors contributed equally to the work and share the last authorship.
Open AccessPublished:September 28, 2022DOI:https://doi.org/10.1016/j.ygyno.2022.09.021

      Highlights

      • One-third of women with CCU were upstaged to stage IIIC by the FIGO-2018 staging system.
      • Lymph node involvement by modern imaging is a significant prognostic factor in stages I-II but not in stage III.
      • The prognostic properties within stage IIIC subgroups stratified by tumor size are variable.
      • The sensitivity of lymph node imaging is low in early-stage CCU, and histological verification is warranted.

      Abstract

      Objective

      Staging carcinoma of the uterine cervix (CCU) by FIGO-2018 suggests stage migration of FIGO-2009 stage I-III patients with lymph node metastasis into FIGO-2018 stage IIIC. We aimed to investigate the prognostic value of lymph node metastases identified by imaging.

      Methods

      We enrolled all patients with biopsy-verified CCU from 2007 to 2016 at Odense University Hospital, Denmark. FDG-PET/CT and MRI were performed before clinical examination in general anesthesia. Disease-specific mortality was compared between women with lymph node-positive and lymph node-negative imaging.

      Results

      In total, 488 patients underwent clinical staging according to FIGO-2009. Lymph node-positive imaging was identified in 146 (30%) patients: 0/36 (0%) in stage IA, 22/195 (11%) in IBI, 14/30 (47%) in IB2, 70/164 (43%) in II and 40/63 (63%) in III. The 5-year cumulative incidence of death due to CCU lymph node-negative vs. lymph node-positive patients was 0.8% vs. 7.1% (p = 0.034) in stage IBI, 0% vs. 34.5% (p = 0.003) in stage IB2, 15.1% vs. 41.4% (p < 0.0001) in stage II, and 33.3% vs. 46.6% (p = 0.28) in stage III by FIGO-2009.

      Conclusions

      One of three women with FIGO-2009 stage I-III CCU had suspected lymph node metastasis on imaging and is upstaged to stage IIIC according to FIGO-2018. The cancer-specific mortality by CCU was significantly lower in the lymph node-negative women stages IBI-II, thus supporting stage migration due to suspected lymph node metastasis. However, the exact prognostic value within stage IIIC is challenged, and future revision of FIGO stages may include new sub-stages.

      Keywords

      1. Introduction

      The International Federation of Gynecology and Obstetrics (FIGO) committee revised the FIGO staging system for carcinoma of the uterine cervix (CCU) in 2018 [
      • Bhatla N.
      • et al.
      Revised FIGO staging for carcinoma of the cervix uteri.
      ,
      • Bhatla N.
      • et al.
      Revised FIGO staging for carcinoma of the cervix uteri.
      ]. The revised FIGO-2018 classification includes imaging and pathology for staging in contrast to the FIGO-2009 classification, which is based on clinical staging only [
      • Bhatla N.
      • et al.
      Revised FIGO staging for carcinoma of the cervix uteri.
      ,
      • Bhatla N.
      • et al.
      Revised FIGO staging for carcinoma of the cervix uteri.
      ,
      • Pecorelli S.
      • Zigliani L.
      • Odicino F.
      Revised FIGO staging for carcinoma of the cervix.
      ]. The revised FIGO criteria classifies early-stage FIGO I and II into FIGO stage IIIC if retroperitoneal lymph node metastases are identified on imaging or pathology [
      • Bhatla N.
      • et al.
      Revised FIGO staging for carcinoma of the cervix uteri.
      ]. Patients are assigned stage IIIC1 if only pelvic lymph nodes are involved and stage IIIC2 if para-aortic lymph nodes are identified.
      Lymph node metastasis is the main prognostic factor in early-stage cervical carcinomas [
      • Inoue T.
      • Morita K.
      The prognostic significance of number of positive nodes in cervical carcinoma stages IB, IIA, and IIB.
      ]. The accuracy of imaging has been evaluated in several studies and indicated that magnetic resonance imaging (MRI) is superior compared to other imaging modalities to evaluate locally advanced disease [
      • Hricak H.
      • et al.
      Invasive cervical carcinoma: comparison of MR imaging and surgical findings.
      ,
      • Balleyguier C.
      • et al.
      Staging of uterine cervical cancer with MRI: guidelines of the European Society of Urogenital Radiology.
      ], while 2-deoxy-2-[18F]fluoro-d-glucose (FDG) positron emission tomography is superior to demonstrate retroperitoneal lymph nodes and distant metastasis in advanced cancer stages [
      • Rose P.G.
      • et al.
      Positron emission tomography for evaluating Para-aortic nodal metastasis in locally advanced cervical cancer before surgical staging: a surgicopathologic study.
      ,
      • Choi H.J.
      • et al.
      Diagnostic performance of computer tomography, magnetic resonance imaging, and positron emission tomography or positron emission tomography/computer tomography for detection of metastatic lymph nodes in patients with cervical cancer: meta-analysis.
      ,
      • Choi H.J.
      • et al.
      Comparison of the accuracy of magnetic resonance imaging and positron emission tomography/computed tomography in the presurgical detection of lymph node metastases in patients with uterine cervical carcinoma: a prospective study.
      ,
      • Havrilesky L.J.
      • et al.
      FDG-PET for management of cervical and ovarian cancer.
      ,
      • Bjurberg M.
      • et al.
      FDG-PET in cervical cancer: staging, re-staging and follow-up.
      ,
      • Loft A.
      • et al.
      The diagnostic value of PET/CT scanning in patients with cervical cancer: a prospective study.
      ,
      • Yildirim Y.
      • et al.
      Integrated PET/CT for the evaluation of Para-aortic nodal metastasis in locally advanced cervical cancer patients with negative conventional CT findings.
      ]. In early-stage CCU (IA2-IIA1), the sensitivity and specificity of imaging are reported to be low, with sensitivities as low as 0.15 and specificities varying from 0.86 to 1.0 [
      • Loft A.
      • et al.
      The diagnostic value of PET/CT scanning in patients with cervical cancer: a prospective study.
      ,
      • Chou H.H.
      • et al.
      Low value of [18F]-fluoro-2-deoxy-D-glucose positron emission tomography in primary staging of early-stage cervical cancer before radical hysterectomy.
      ,
      • Magne N.
      • et al.
      New trends in the evaluation and treatment of cervix cancer: the role of FDG-PET.
      ,
      • Signorelli M.
      • et al.
      Preoperative staging of cervical cancer: is 18-FDG-PET/CT really effective in patients with early stage disease?.
      ,
      • Wright J.D.
      • et al.
      Preoperative lymph node staging of early-stage cervical carcinoma by [18F]-fluoro-2-deoxy-D-glucose-positron emission tomography.
      ,
      • Reinhardt M.J.
      • et al.
      Metastatic lymph nodes in patients with cervical cancer: detection with MR imaging and FDG PET.
      ,
      • Sironi S.
      • et al.
      Lymph node metastasis in patients with clinical early-stage cervical cancer: detection with integrated FDG PET/CT.
      ,
      • Crivellaro C.
      • et al.
      18F-FDG PET/CT can predict nodal metastases but not recurrence in early stage uterine cervical cancer.
      ,
      • Sponholtz S.E.
      • et al.
      Sentinel lymph node mapping in early-stage cervical cancer - a national prospective multicenter study (SENTIREC trial).
      ], thus questioning the value of imaging in early-stage CCU.
      We aimed to assess the impact of combined FDG-PET/CT and MRI diagnostics of potential lymph node metastases on cancer-specific survival in patients with cervical cancer stages I-IV. We compared stage-stratified groups according to FIGO-2009 and FIGO-2018 classifications. Further, in women who underwent lymph node dissection as part of their surgical treatment, we investigated the accuracy of lymph node involvement by imaging for the FIGO-2009 early-stage CCU (IA2-IIA) with histology as the gold standard.

      2. Materials and methods

      This clinical register study included women with biopsy-verified CCU at Odense University Hospital, a regional hospital center, and one of three national centers that undertakes diagnosis and treatment of CCU. Data were retrospectively collected over 10 years, from 2007 to 2016. The study protocol was approved by the Danish Data Protection Agency (ID number: 14.26574) and the Danish Patient Safety Authority (ID number: 3–3013-651.1). We retrospectively re-staged patients according to image-guided lymph node staging using the revised FIGO-2018 classification [
      • Bhatla N.
      • et al.
      Revised FIGO staging for carcinoma of the cervix uteri.
      ,
      • Bhatla N.
      • et al.
      Revised FIGO staging for carcinoma of the cervix uteri.
      ].

      2.1 Patients

      Women referred to the Department of Gynecology at Odense University Hospital who underwent upfront FDG-PET/CT and MRI before clinical staging in general anesthesia, were eligible for the study.
      Initial data extraction was performed in the local radiology information system with selection of patients referred to FDG-PET/CT and MRI from the Department of Gynecology. The following data were extracted from medical files from the Departments of Gynecology and Oncology: imaging reports, FIGO-stage classification, histopathology, treatment, survival, and cause of death. Patients who did not undergo both imaging modalities were excluded. Hence, patients diagnosed with small volume disease who underwent conization with clear margins for stage IA1 disease were not included in the cohort since these patients did not undergo imaging according to national guidelines.

      2.2 Image interpretation

      A specialist in nuclear medicine and a radiology specialist prepared the image reports of the FDG-PET/CT and the MRI. The images and the imaging reports were presented by the imaging specialists and discussed at a Multi-Disciplinary Team (MDT) Conference, and the final reports were available prior to the clinical staging procedure. Disagreement between the PET/CT and MRI reports on suspected lymph node metastases was resolved for the individual patient case at the MDT conference. A joint agreement was reached regarding final judgment of the lymph node as suspicious of metastasis or not by imaging. The final decision from the MDT conference that influenced further staging and treatment decisions was used for dichotomization into image-positive or image-negative metastatic lymph nodes in the present study.

      2.3 Outcome measures

      We analyzed the stage migration to stage IIIC due to potential lymph node metastases and the influence on the cumulative incidence of disease-specific mortality. Women with image-positive metastatic lymph nodes were determined according to FIGO-2009 groups (IA2, IB1, IB2, IIA + B, and IIIA + B) and allocated to stage IIIC according to FIGO-2018. The accuracy of imaging for early-stage CCU was analyzed in a subgroup of women with early-stage CCU who were clinically eligible for surgery with curative intent. This patient group underwent staging with radical pelvic lymphadenectomy. Pelvic lymph node involvement by imaging was determined as decided on the MDT (FDG-PET/CT and MRI combined). Final histopathology was used as the reference standard.
      The tissue from the pelvic lymphatic regions was received at the Department of Pathology in neutral buffered formalin. The individual lymph nodes were identified by thorough dissection, gross sectioned at 2–3 mm intervals and placed in cassettes separately. The paraffin-embedded micro-sections were stained by hematoxylin and eosin, and any metastatic disease was classified according to the WHO classification. The sentinel node mapping technique and histological ultrastaging of lymph nodes were not adopted for the women included in this study.

      2.4 Statistical analysis

      Descriptive statistics were applied according to data type: continuous variables as mean ± standard deviation (SD) or median and range (depending on the symmetry of the empirical data distributions); categorical variables as frequencies and respective percentages. Estimates for diagnostic accuracy measures were supplemented by exact (Clopper-Pearson type) 95% confidence intervals (95% CI) for binomial proportions. Competing-risks regression models were performed according to the method of Fine and Gray [
      • Fine J.P.
      • Gray R.J.
      A proportional hazards model for the subdistribution of a competing risk.
      ], with visual displays using cumulative incidence curves for disease-specific mortality and overall mortality by stage and lymph node involvement (yes/no). Supplementary Kaplan-Meier analyses on ‘cancer-specific’ survival by CCU and on overall survival were done to compare with previous literature findings (Supplementary Figs. S1-S2 and Supplementary Tables S3-S4). The significance level was 5%. All analyses were performed by using STATA/MP 16.1 (StataCorp, College Station, TX 77845, USA).

      3. Results

      A total of 533 patients with primary CCU were included in the study, median age 46 (range 21–94) years. The stage migration from FIGO-2009 stages to FIGO-20018 stages is illustrated in Fig. 1. Patient characteristics are given in Table 1.
      Fig. 1
      Fig. 1Patient allocation from FIGO-2009 to FIGO-2018 stages according to imaging by combined FDG-PET-CT and MRI.
      Table 1Demographic and histological characteristics of women with CCU.
      Women with CCU Number (%)
      FIGO-2009
      All stages533 (100)
       IA1
      For stage IA1 only women with primary conization without free margins were included.
      33 (6.2)
       IA23 (0. 6)
       IB1195 (36.6)
       IB230 (5.6)
       IIA9 (1.7)
       IIB154 (28.9)
       III64 (12.0)
       IV45 (8.4)
      Histology
       Squamous cell carcinoma367 (68.9)
       Adenocarcinoma134 (25.1)
       Adenosquamous carcinoma19 (3.6)
       Other
      Clear cell carcinoma (n = 2), Glassy cell carcinoma (n = 2), Mucinous adenocarcinoma (n = 1), Primary mesonephric carcinoma (n = 1), Sarcomatoid planocellular carcinoma (n = 1), Adenoma malignum (n = 1), Neuroendocrine tumor (n = 5).
      13 (2.4)
      Location of lymph node involvement by imaging in stages I-III146 (27.4)
       Pelvic lymph node involvement (restaged into IIIC1r
      r = radiology.
      )
      112 (21.0)
       Paraaortic lymph node involvement (restaged into IIIC2r
      r = radiology.
      )
      34 (6.4)
      Treatment
      Re-conization or simple hysterectomy33 (6.2)
       Radical hysterectomy/trachelectomy incl. pelvic lymph nodes199 (37.3)
      Subgroup with adjuvant therapy to the radical hysterectomy (External beam radiation + concomitant chemotherapy)71 (13.3)
       Primary radiation and concomitant chemotherapy208 (39.0)
       Primary beam radiation only15 (2.8)
       Primary chemotherapy only4 (0.8)
       Primary palliative care incl. external beam radiation and/or chemotherapy55 (10.3)
       The patient deselected offered treatment19 (3.6)
      Women included in the accuracy study199 (37.3)
      a For stage IA1 only women with primary conization without free margins were included.
      b Clear cell carcinoma (n = 2), Glassy cell carcinoma (n = 2), Mucinous adenocarcinoma (n = 1), Primary mesonephric carcinoma (n = 1), Sarcomatoid planocellular carcinoma (n = 1), Adenoma malignum (n = 1), Neuroendocrine tumor (n = 5).
      c r = radiology.

      3.1 Image-positive metastatic lymph nodes

      For women in FIGO-2009 stages I-III, 146 (30%) had image-positive lymph nodes. Patients with image-positive lymph nodes as stated by the PET/CT and MRI imaging reports were all positive on FDG-PET/CT (146/146; 100%) while 84% (122/146) were positive on MRI. Stratified into FIGO-2009 stages, the numbers were 0 (0%) in stage IA, 22 (11%) in stage IB1, 14 (47%) in stage IB2, 70 (43%) in stage II, and 40 (63%) in stage III. Women with stage IV disease by FIGO-2009 classification remained stage IV by FIGO-2018 classification (n = 45).

      3.2 The cumulative incidence of mortality

      The cumulative incidence functions of cancer-specific mortality by CCU and of mortality due to all other causes than CCU are given in Fig. 2 stratified by FIGO stages I-III, with separate curves for the lymph node image-negatives vs. the image-positives. The 5-year cumulative incidence of cancer-specific mortality by CCU for patients with image-negative versus patients with image-positive lymph nodes were: 0.8% vs. 7.1% in stage IB1 patients (p = 0.034), 0% vs. 34.5% in stage IB2 (p = 0.003), 15.1% vs. 41.4% in stage II (p < 0.0001), and 33.3% vs. 46.6% in stage III (p = 0.28). Kaplan-Meier curves for cancer-specific survival (death caused by CCU) and overall survival stratified by stage are given in Supplementary Figs. S1 and S2. Corresponding 5- and 10-year survival probabilities and distribution of events are shown in Supplementary Tables S3 and S4. A comparison of cancer-specific survival by CCU stratified by stage according to FIGO-2009 and 2018, respectively, is given in Fig. 3, illustrating the changes due to lymph node involvement. Fig. 3 highlights that allocating image-positive women with stages I-II to IIIC clearly improved the survival of the remaining women in these stages. All the lymph node positive women in stages I-III were reclassified to the new stage IIIC according to FIGO-2018 (plotted red line in Fig. 3) have lower survival compared to all the other stages I-II by FIGO-2018 classification, but still not inferior to stage IIIA and IIIB. Kaplan-Meier curves for cancer-specific survival within the new stage IIIC stratified by pelvic vs. para-aortic location of lymph node involvement on imaging (stage IIIC1r vs. IIIC2r) are given in Supplementary Figs. S5.
      Fig. 2
      Fig. 2Cumulative mortality incidence curves of cancer-specific mortality by CCU (left column) and mortality by causes other than CCU (right column) for women with FIGO-2009 stage IB1, IB2, II, and III. The curves for each stage are shown for the patients with image-negative (blue curves) vs. image-positive (red curves) lymph node involvement. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
      Fig. 3
      Fig. 3CCU cancer-specific survival plots stratified by stage according to the FIGO-2009 classification (top) and illustration of the same cohort restaged by FIGO-2018 classification with added a separate stage IIIC (red plot) for the image-node-positive patients (bottom). (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

      3.3 Accuracy of FDG-PET/CT and MRI combined for early stages

      Surgery was performed in 199 women. The distribution by stage was: IA2 n = 3; IB1 n = 187; IB2 n = 6 and IIA n = 3. Table 2 shows image-positive and image-negative metastatic lymph node results versus lymph node status on final pathology. The prevalence of lymph node metastases on final pathology was 6% (12/199) in this patient group. The sensitivity, specificity, accuracy, positive and negative predictive values for FDG-PET/CT and MRI combined were 67%, 90%, 89%, and 31% and 98%, respectively (Table 3).
      Table 2Cross-tabulation of image-positive and image-negative metastatic lymph node results versus histology-verified results in the group of women with early-stage cervical cancer who underwent radical pelvic lymph node dissection as part of their treatment. In FIGO-2018, histology overrules imaging.
      Imaging resultNumber of women with positive lymph nodes by histologyNumber of women with negative lymph nodes by histologyTotal
      Number of women with positive lymph nodes on imaging81826
      Number of women with negative lymph nodes on imaging4169173
      Total number of women with lymph node dissection performed12187199
      Table 3Diagnostic accuracy results for early-stage CCU triaged for radical hysterectomy with diagnostic lymph node dissection, histology as gold standard (n = 199).
      Parameter(%)95% Confidence Interval
      Prevalence63–10
      Sensitivity6740–93
      Specificity9086–95
      Positive predictive value3113–38
      Negative predictive value9895–99

      4. Discussion

      We applied the FIGO-2018 stage classification in a large population of women undergoing clinical staging supplemented by FDG-PET/CT and MRI who were primarily staged according to FIGO-2009. We evaluated the prognostic impact of a multidisciplinary judgment of lymph node-positive or negative imaging. One-third of the population was upstaged due to image-positive lymph nodes according to FIGO-2018. This was reflected in a corresponding significant effect on survival outcomes; women in FIGO-2018 stages I-III showed improved survival while those with image positive lymph nodes independent of prior FIGO-2009 stage showed survival outcomes at the same level as women with stage IIIA and IIIB disease. This supports the revised FIGO-2018 classification system as regards imaging for lymph node staging and confirms the allocation of the image-node-positive stages I and II patients to a more advanced stage than by the FIGO-2009 classification. The significant difference in survival by imaging pelvic and para-aortic lymph node involvement illustrated in Supplementary Fig. S5 supports the subdivision of FIGO-2018 stage IIIC by stages IIIC1r and IIIC2r (p < 0.0001). There was no significant difference in cumulative incidence mortality between the image lymph node-negative and node-positive group in stage III A + B patients.
      The frequency of histologically verified lymph node metastasis in early-stage and locally advanced cervical cancer has been reported to vary between 5 and 30%, in stages ranging from IA2-II [
      • Loft A.
      • et al.
      The diagnostic value of PET/CT scanning in patients with cervical cancer: a prospective study.
      ,
      • Chou H.H.
      • et al.
      Low value of [18F]-fluoro-2-deoxy-D-glucose positron emission tomography in primary staging of early-stage cervical cancer before radical hysterectomy.
      ,
      • Signorelli M.
      • et al.
      Preoperative staging of cervical cancer: is 18-FDG-PET/CT really effective in patients with early stage disease?.
      ,
      • Wright J.D.
      • et al.
      Preoperative lymph node staging of early-stage cervical carcinoma by [18F]-fluoro-2-deoxy-D-glucose-positron emission tomography.
      ,
      • Reinhardt M.J.
      • et al.
      Metastatic lymph nodes in patients with cervical cancer: detection with MR imaging and FDG PET.
      ,
      • Sironi S.
      • et al.
      Lymph node metastasis in patients with clinical early-stage cervical cancer: detection with integrated FDG PET/CT.
      ,
      • Crivellaro C.
      • et al.
      18F-FDG PET/CT can predict nodal metastases but not recurrence in early stage uterine cervical cancer.
      ]. Recent Chinese and American studies restaging patients with CCU by the revised FIGO-2018 classification found that 17–37% were upstaged from stage IB1, IB2, and IIA to stage IIIC based on suspected lymph node metastasis on imaging only [
      • Yan D.D.
      • et al.
      Prognostic value of the 2018 FIGO staging system for cervical cancer patients with surgical risk factors.
      ,
      • Grigsby P.W.
      • et al.
      FIGO 2018 staging criteria for cervical cancer: impact on stage migration and survival.
      ,
      • Wright J.D.
      • et al.
      Prognostic performance of the 2018 International Federation of Gynecology and Obstetrics Cervical Cancer Staging Guidelines.
      ]. This is similar to the distribution in our study with a clear overweight of stage migration in women with FIGO-2009 stage > IB1, i.e. with larger tumor >4 cm or parametrial involvement. These studies all conclude that the new FIGO staging system appears to be useful for predicting prognosis using imaging for preoperative staging in the early stages of CCU, and approve that both tumor size and lymph node involvement are a part of the CCU staging in the FIGO-2018 classification [
      • Grigsby P.W.
      • et al.
      FIGO 2018 staging criteria for cervical cancer: impact on stage migration and survival.
      ,
      • Wright J.D.
      • et al.
      Prognostic performance of the 2018 International Federation of Gynecology and Obstetrics Cervical Cancer Staging Guidelines.
      ]. Although using various imaging in the study by Wright [
      • Wright J.D.
      • et al.
      Prognostic performance of the 2018 International Federation of Gynecology and Obstetrics Cervical Cancer Staging Guidelines.
      ], and partly including FDG-PET without the CT counterpart in the study by Grigsby [
      • Grigsby P.W.
      • et al.
      FIGO 2018 staging criteria for cervical cancer: impact on stage migration and survival.
      ], both these studies find the same prognostic results as in our study: improved prognosis for the lymph node negative women within each FIGO-2018 stage I-II, but not for the FIGO-2018 stage III group.
      The American studies display concern for the prognostic heterogeneity within the new stage IIIC and suspect that increasing tumor size within stage IIIC could reflect a successive worse prognosis regardless of the lymph node status [
      • Grigsby P.W.
      • et al.
      FIGO 2018 staging criteria for cervical cancer: impact on stage migration and survival.
      ,
      • Wright J.D.
      • et al.
      Prognostic performance of the 2018 International Federation of Gynecology and Obstetrics Cervical Cancer Staging Guidelines.
      ]. In our study, pooling the imaging lymph node-positive patients in original stages I-II and stage III to the new stage IIIC, regardless of the tumor size and/or local invasion, gave a survival curve that was similar, but not inferior to stages II A + B (red and blue curve in Fig. 3). This questions if the new FIGO-2018 stage IIIC correctly reflects different prognostic groups within stage III. Regarding the stage-specific cumulative mortality rates by FIGO-2009 (Fig. 2), mortality increases with tumor size in stages I-III. We suspect that the collective survival for FIGO-2018 re-staged IIIC patients is positively influenced by women with smaller tumors. It gives cause for concern whether the FIGO-2009 stage I-II lymph node positives are correctly allocated in the advanced FIGO-2018 stage IIIC, or in a future FIGO revision should be allocated in a separate stage, e.g. stage IC and/or IIC.
      Several studies have evaluated the accuracy of preoperative imaging with FDG-PET, FDG-PET/CT, or MRI in patients with CCU [
      • Rose P.G.
      • et al.
      Positron emission tomography for evaluating Para-aortic nodal metastasis in locally advanced cervical cancer before surgical staging: a surgicopathologic study.
      ,
      • Choi H.J.
      • et al.
      Diagnostic performance of computer tomography, magnetic resonance imaging, and positron emission tomography or positron emission tomography/computer tomography for detection of metastatic lymph nodes in patients with cervical cancer: meta-analysis.
      ,
      • Choi H.J.
      • et al.
      Comparison of the accuracy of magnetic resonance imaging and positron emission tomography/computed tomography in the presurgical detection of lymph node metastases in patients with uterine cervical carcinoma: a prospective study.
      ,
      • Havrilesky L.J.
      • et al.
      FDG-PET for management of cervical and ovarian cancer.
      ,
      • Bjurberg M.
      • et al.
      FDG-PET in cervical cancer: staging, re-staging and follow-up.
      ,
      • Loft A.
      • et al.
      The diagnostic value of PET/CT scanning in patients with cervical cancer: a prospective study.
      ,
      • Yildirim Y.
      • et al.
      Integrated PET/CT for the evaluation of Para-aortic nodal metastasis in locally advanced cervical cancer patients with negative conventional CT findings.
      ,
      • Chou H.H.
      • et al.
      Low value of [18F]-fluoro-2-deoxy-D-glucose positron emission tomography in primary staging of early-stage cervical cancer before radical hysterectomy.
      ,
      • Magne N.
      • et al.
      New trends in the evaluation and treatment of cervix cancer: the role of FDG-PET.
      ,
      • Signorelli M.
      • et al.
      Preoperative staging of cervical cancer: is 18-FDG-PET/CT really effective in patients with early stage disease?.
      ,
      • Wright J.D.
      • et al.
      Preoperative lymph node staging of early-stage cervical carcinoma by [18F]-fluoro-2-deoxy-D-glucose-positron emission tomography.
      ,
      • Reinhardt M.J.
      • et al.
      Metastatic lymph nodes in patients with cervical cancer: detection with MR imaging and FDG PET.
      ,
      • Sironi S.
      • et al.
      Lymph node metastasis in patients with clinical early-stage cervical cancer: detection with integrated FDG PET/CT.
      ,
      • Crivellaro C.
      • et al.
      18F-FDG PET/CT can predict nodal metastases but not recurrence in early stage uterine cervical cancer.
      ]. The majority of these accuracy studies have focused on comparing the accuracy of individual imaging modalities instead of their combined use, as is often used in the clinical setting. In the present study, we combined FDG-PET/CT and MRI in all patients, but still found only modest accuracy results for the 199 patients with early-stage disease from whom a sufficient reference standard was obtained by surgical staging of pelvic lymph nodes. We found a moderate sensitivity of 67% but a higher specificity of 90%, comparable with the accuracy found in earlier studies [
      • Loft A.
      • et al.
      The diagnostic value of PET/CT scanning in patients with cervical cancer: a prospective study.
      ,
      • Chou H.H.
      • et al.
      Low value of [18F]-fluoro-2-deoxy-D-glucose positron emission tomography in primary staging of early-stage cervical cancer before radical hysterectomy.
      ,
      • Magne N.
      • et al.
      New trends in the evaluation and treatment of cervix cancer: the role of FDG-PET.
      ,
      • Signorelli M.
      • et al.
      Preoperative staging of cervical cancer: is 18-FDG-PET/CT really effective in patients with early stage disease?.
      ,
      • Wright J.D.
      • et al.
      Preoperative lymph node staging of early-stage cervical carcinoma by [18F]-fluoro-2-deoxy-D-glucose-positron emission tomography.
      ,
      • Reinhardt M.J.
      • et al.
      Metastatic lymph nodes in patients with cervical cancer: detection with MR imaging and FDG PET.
      ,
      • Sironi S.
      • et al.
      Lymph node metastasis in patients with clinical early-stage cervical cancer: detection with integrated FDG PET/CT.
      ,
      • Crivellaro C.
      • et al.
      18F-FDG PET/CT can predict nodal metastases but not recurrence in early stage uterine cervical cancer.
      ]. With a very low prevalence of lymph node metastasis in this patient group, mainly FIGO-2009 stage IB1 (tumor <4 cm), this resulted in a very low positive predictive value (31%). In the SENTIREC study published recently by our group, we used up-to-date imaging and sentinel node mapping in all patients [
      • Sponholtz S.E.
      • et al.
      Sentinel lymph node mapping in early-stage cervical cancer - a national prospective multicenter study (SENTIREC trial).
      ]. In this setting, we reported a sensitivity of FDG-PET/CT as low as 14.8% in CCU patients with tumor size <4 cm [
      • Sponholtz S.E.
      • et al.
      Sentinel lymph node mapping in early-stage cervical cancer - a national prospective multicenter study (SENTIREC trial).
      ]. The reference standard was histologic examination performed as ultrastaging using a sentinel node algorithm, and this may have increased the validity of the histological reference standard. The lower sensitivity may be more correctly reflected in the SENTIREC study, while the histologic examination of pelvic lymph nodes that was performed as clinical practice in the time period of our study, may be less valid and falsely cause a higher sensitivity.
      The low prevalence of lymph node metastases combined with the low accuracy and predictive values comparable to findings in earlier studies may suggest omitting pre-operative advanced imaging in patients with FIGO-2018 stage IA-IB2. Due to the clearly valuable prognostic value of imaging positive lymph node metastases also in early-stage cervical cancer, histological confirmation of the lymph node staging is still of essence in early-stage CCU. These considerations are not addressed in detail in the FIGO-2018 classification [
      • Bhatla N.
      • et al.
      Revised FIGO staging for carcinoma of the cervix uteri.
      ,
      • Bhatla N.
      • et al.
      Revised FIGO staging for carcinoma of the cervix uteri.
      ]. The European Guidelines of management of CCU address this problem, by suggesting a two-step procedure of staging the FDG−/imaging-positive nodes in the early stages, before deciding on final stage and treatment [
      • Cibula D.
      • et al.
      The European Society of Gynaecological Oncology/European Society for Radiotherapy and Oncology/European Society of Pathology guidelines for the management of patients with cervical cancer.
      ]. This is a rather comprehensive surgical procedure, with a potentially high risk of complications. We suggest sentinel node mapping with ultrastaging procedures for improved lymph node staging accuracy in early-stage CCU [
      • Sponholtz S.E.
      • et al.
      Sentinel lymph node mapping in early-stage cervical cancer - a national prospective multicenter study (SENTIREC trial).
      ].
      We have no valid reference standard for the higher stages (II and III) in the present study. These patients undergo chemo-radiotherapy with curative intent, and confirmatory lymph node biopsies are optional. The accuracy measures of imaging are likely to be higher in patients with advanced disease who present with a higher prevalence of lymph node metastases based on former literature testing the accuracy of FDG-PET in both early and advanced CCU [
      • Rose P.G.
      • et al.
      Positron emission tomography for evaluating Para-aortic nodal metastasis in locally advanced cervical cancer before surgical staging: a surgicopathologic study.
      ,
      • Choi H.J.
      • et al.
      Diagnostic performance of computer tomography, magnetic resonance imaging, and positron emission tomography or positron emission tomography/computer tomography for detection of metastatic lymph nodes in patients with cervical cancer: meta-analysis.
      ,
      • Choi H.J.
      • et al.
      Comparison of the accuracy of magnetic resonance imaging and positron emission tomography/computed tomography in the presurgical detection of lymph node metastases in patients with uterine cervical carcinoma: a prospective study.
      ,
      • Havrilesky L.J.
      • et al.
      FDG-PET for management of cervical and ovarian cancer.
      ,
      • Bjurberg M.
      • et al.
      FDG-PET in cervical cancer: staging, re-staging and follow-up.
      ,
      • Loft A.
      • et al.
      The diagnostic value of PET/CT scanning in patients with cervical cancer: a prospective study.
      ,
      • Yildirim Y.
      • et al.
      Integrated PET/CT for the evaluation of Para-aortic nodal metastasis in locally advanced cervical cancer patients with negative conventional CT findings.
      ]. The prognostic value of imaging with FDG-PET/CT positive lymph nodes in women with locally advanced disease is clearly reflected in a deterioration of the survival outcomes in our study. This is comparable to findings in studies by Hansen and Cegla [
      • Hansen H.V.
      • et al.
      Survival outcomes in patients with cervical cancer after inclusion of PET/CT in staging procedures.
      ,
      • Cegla P.
      • et al.
      Influence of biological parameters assessed in [18F]FDG PET/CT on overall survival in cervical Cancer patients.
      ], who also tested survival outcomes in CCU based on stage assessed with FDG-PET/CT. This confirms the clinical value of pre-operative advanced imaging with FDG-PET/CT in patients with locally advanced cervical cancer.
      We analyzed a rather large group of patients, representative of the daily clinical practice, and used high technology imaging with FDG-PET/CT and MRI. We recognize limitations such as this being a single-center retrospective study and our results may not apply to other populations. We only tested the prognostic values on restaging from FIGO-2009 to FIGO-2018 by lymph nodes on imaging. We did not collect data on the number of lymph nodes by imaging to calculate whether increasing numbers would influence the survival within stage IIIC. We did not collect data on sociodemographic factors or race because this type of data is not systematically entered into the patient's files and The Danish Healthcare System entertain free and equal access for all citizens regardless of socioeconomic means. It is interesting that despite the smaller size of our study compared to prior studies [
      • Grigsby P.W.
      • et al.
      FIGO 2018 staging criteria for cervical cancer: impact on stage migration and survival.
      ,
      • Wright J.D.
      • et al.
      Prognostic performance of the 2018 International Federation of Gynecology and Obstetrics Cervical Cancer Staging Guidelines.
      ], and likely with different sociodemographic profiles between the populations, the prognostic results are very similar. The present study does not address the clinical impact in terms of changed treatment strategies associated with the revised classification system, because these data were not available. However, this would be of uttermost interest in future prospective studies. Future studies should be designed to test the influence of changed treatment strategies for stage IIIC on survival outcomes stratified by the impact of tumor size and explore any potential difference in treatment strategies and survival outcomes for stage IIIC1r vs. IIIC2r including numbers of lymph nodes.

      5. Conclusion

      One out of three women with FIGO-2009 stage I-III CCU had suspected lymph node metastasis on imaging and was upstaged to stage IIIC according to FIGO-2018. The cancer-specific mortality by CCU was significantly lower in the imaging lymph node-negative women stages IBI-II, thus supporting stage migration due to lymph node metastasis. However, the exact prognostic value within the FIGO-2018 stage IIIC is challenged and future revision of FIGO stages may include substages of stage IIIC or other stages for the lymph node positives with small tumor size. Histological confirmation of image positive lymph nodes is still of essence in early-stage cervical cancer due to the clearly valuable prognostic value of lymph node metastases in the early stages. Here, we recommend sentinel lymph node mapping with concomitant removal of image positive lymph nodes.
      Image 1
      Supplementary Fig. S1. Kaplan-Meier plots for CCU cancer-specific survival for FIGO-2009 stage IB1, IB2, II and III patients, separated by image-negative vs. image-positive lymph nodes. Blue: N0 by imaging. Red: N+ by imaging.
      Image 2
      Supplementary Fig. S2. Kaplan-Meier plots for overall survival for FIGO-2009 stage IB1, IB2, II and III patients, separated by image-negative vs. image-positive lymph nodes. Blue: N0 by imaging. Red: N+ by imaging.
      Image 5
      Supplementary Fig. S5. Cancer-specific survival stratified by substages IIIC1r vs. IIIC2r, representing pelvic vs. para-aortic lymph node involvement by imaging.

      CRediT authorship contribution statement

      Jorun Holm: Visualization, Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Validation, Writing – original draft, Writing – review & editing. Oke Gerke: Data curation, Formal analysis, Methodology, Validation, Writing – original draft, Writing – review & editing. Mie Holm Vilstrup: Methodology, Validation, Writing – review & editing. Diana Spasojevic: Validation, Writing – review & editing. Sara Elisabeth Sponholtz: Validation, Writing – review & editing. Kirsten Marie Jochumsen: Visualization, Conceptualization, Validation, Writing – review & editing. Anders Thomassen: Visualization, Conceptualization, Validation, Writing – review & editing. Malene Grubbe Hildebrandt: Visualization, Conceptualization, Investigation, Methodology, Project administration, Funding acquisition, Supervision, Validation, Writing – original draft, Writing – review & editing. Pernille Tine Jensen: Investigation, Conceptualization, Methodology, Supervision, Validation, Writing – original draft, Writing – review & editing.

      Declaration of Competing Interest

      There are no conflicts of interest to disclose.

      References

        • Bhatla N.
        • et al.
        Revised FIGO staging for carcinoma of the cervix uteri.
        Int. J. Gynaecol. Obstet. 2019; 145: 129-135
        • Bhatla N.
        • et al.
        Revised FIGO staging for carcinoma of the cervix uteri.
        Int. J. Gynaecol. Obstet. 2019; 145: 129-135
        • Pecorelli S.
        • Zigliani L.
        • Odicino F.
        Revised FIGO staging for carcinoma of the cervix.
        Int. J. Gynaecol. Obstet. 2009; 105: 107-108
        • Inoue T.
        • Morita K.
        The prognostic significance of number of positive nodes in cervical carcinoma stages IB, IIA, and IIB.
        Cancer. 1990; 65: 1923-1927
        • Hricak H.
        • et al.
        Invasive cervical carcinoma: comparison of MR imaging and surgical findings.
        Radiology. 1988; 166: 623-631
        • Balleyguier C.
        • et al.
        Staging of uterine cervical cancer with MRI: guidelines of the European Society of Urogenital Radiology.
        Eur. Radiol. 2011; 21: 1102-1110
        • Rose P.G.
        • et al.
        Positron emission tomography for evaluating Para-aortic nodal metastasis in locally advanced cervical cancer before surgical staging: a surgicopathologic study.
        J. Clin. Oncol. 1999; 17: 41-45
        • Choi H.J.
        • et al.
        Diagnostic performance of computer tomography, magnetic resonance imaging, and positron emission tomography or positron emission tomography/computer tomography for detection of metastatic lymph nodes in patients with cervical cancer: meta-analysis.
        Cancer Sci. 2010; 101: 1471-1479
        • Choi H.J.
        • et al.
        Comparison of the accuracy of magnetic resonance imaging and positron emission tomography/computed tomography in the presurgical detection of lymph node metastases in patients with uterine cervical carcinoma: a prospective study.
        Cancer. 2006; 106: 914-922
        • Havrilesky L.J.
        • et al.
        FDG-PET for management of cervical and ovarian cancer.
        Gynecol. Oncol. 2005; 97: 183-191
        • Bjurberg M.
        • et al.
        FDG-PET in cervical cancer: staging, re-staging and follow-up.
        Acta Obstet. Gynecol. Scand. 2007; 86: 1385-1391
        • Loft A.
        • et al.
        The diagnostic value of PET/CT scanning in patients with cervical cancer: a prospective study.
        Gynecol. Oncol. 2007; 106: 29-34
        • Yildirim Y.
        • et al.
        Integrated PET/CT for the evaluation of Para-aortic nodal metastasis in locally advanced cervical cancer patients with negative conventional CT findings.
        Gynecol. Oncol. 2008; 108: 154-159
        • Chou H.H.
        • et al.
        Low value of [18F]-fluoro-2-deoxy-D-glucose positron emission tomography in primary staging of early-stage cervical cancer before radical hysterectomy.
        J. Clin. Oncol. 2006; 24: 123-128
        • Magne N.
        • et al.
        New trends in the evaluation and treatment of cervix cancer: the role of FDG-PET.
        Cancer Treat. Rev. 2008; 34: 671-681
        • Signorelli M.
        • et al.
        Preoperative staging of cervical cancer: is 18-FDG-PET/CT really effective in patients with early stage disease?.
        Gynecol. Oncol. 2011; 123: 236-240
        • Wright J.D.
        • et al.
        Preoperative lymph node staging of early-stage cervical carcinoma by [18F]-fluoro-2-deoxy-D-glucose-positron emission tomography.
        Cancer. 2005; 104: 2484-2491
        • Reinhardt M.J.
        • et al.
        Metastatic lymph nodes in patients with cervical cancer: detection with MR imaging and FDG PET.
        Radiology. 2001; 218: 776-782
        • Sironi S.
        • et al.
        Lymph node metastasis in patients with clinical early-stage cervical cancer: detection with integrated FDG PET/CT.
        Radiology. 2006; 238: 272-279
        • Crivellaro C.
        • et al.
        18F-FDG PET/CT can predict nodal metastases but not recurrence in early stage uterine cervical cancer.
        Gynecol. Oncol. 2012; 127: 131-135
        • Sponholtz S.E.
        • et al.
        Sentinel lymph node mapping in early-stage cervical cancer - a national prospective multicenter study (SENTIREC trial).
        Gynecol. Oncol. 2021; 162: 546-554
        • Fine J.P.
        • Gray R.J.
        A proportional hazards model for the subdistribution of a competing risk.
        J. Am. Stat. Assoc. 1999; 94: 496-509
        • Yan D.D.
        • et al.
        Prognostic value of the 2018 FIGO staging system for cervical cancer patients with surgical risk factors.
        Cancer Manag. Res. 2019; 11: 5473-5480
        • Grigsby P.W.
        • et al.
        FIGO 2018 staging criteria for cervical cancer: impact on stage migration and survival.
        Gynecol. Oncol. 2020; 157: 639-643
        • Wright J.D.
        • et al.
        Prognostic performance of the 2018 International Federation of Gynecology and Obstetrics Cervical Cancer Staging Guidelines.
        Obstet. Gynecol. 2019; 134: 49-57
        • Cibula D.
        • et al.
        The European Society of Gynaecological Oncology/European Society for Radiotherapy and Oncology/European Society of Pathology guidelines for the management of patients with cervical cancer.
        Radiother. Oncol. 2018; 127: 404-416
        • Hansen H.V.
        • et al.
        Survival outcomes in patients with cervical cancer after inclusion of PET/CT in staging procedures.
        Eur. J. Nucl. Med. Mol. Imaging. 2015; 42: 1833-1839
        • Cegla P.
        • et al.
        Influence of biological parameters assessed in [18F]FDG PET/CT on overall survival in cervical Cancer patients.
        Clin. Nucl. Med. 2019; 44: 860-863