Objective: To compare FIGO 2009 and FIGO 2018 cervical cancer staging criteria with a focus on stage migration and treatment outcomes. Methods: This study is based on a database cohort of 1282 patients newly diagnosed with cervical cancer from 1997 to 2019. All underwent standard clinical examination and whole-body FDG-PET. Tumor stage was recorded using the FIGO 2009 system, which excluded surgical pathologic, FDG-PET and other advanced imaging findings, and then re-classified to the FIGO 2018 system, including surgical pathologic and imaging findings. Patient management was based on clinical, surgical, and imaging findings. Stage migration and prognosis were evaluated. Results: The distribution per the 2009 staging system was stage I in 593 (46%), stage II in 342 (27%), stage III in 263 (21%), and stage IV in 84 (7%) and the 2018 staging system was stage I in 354 (28%), stage II in 156 (12%), stage III in 601 (47%), and stage IV in 171 (13%). No patients were down-staged. Stage migration occurred in 53% (676/1282) and was attributable to detection of occult lymph node metastasis in 520 (41%), occult distant metastasis in 90 (7%), and tumor size and extent in 66 (5%). The 5-year progression-free survivals (PFS) by FIGO 2009 versus FIGO 2018 were as follows: stage I, 80% vs. 87% (p = 0.02); stage II, 59% vs. 71% (p = 0.002); stage III, 35% vs. 55% (p < 0.001), and stage IV, 20% vs. 16% (p = 0.41). Conclusion: Inclusion of surgical pathologic and imaging findings resulted in upward stage migration in the majority, mostly related to nodal and distant metastasis. While FIGO 2018 improves survival discriminatory ability for stages I and IV patients, survival remains heterogeneous among stage III substages.

Original languageEnglish
Pages (from-to)639-643
Number of pages5
JournalGynecologic oncology
Issue number3
StatePublished - Jun 2020


  • Cancer
  • Cervical
  • FIGO
  • Staging


Dive into the research topics of 'FIGO 2018 staging criteria for cervical cancer: Impact on stage migration and survival'. Together they form a unique fingerprint.

Cite this