EP082/#839 ISO-prognostic clusters in advanced cervix cancer treated with curative intent

Objectives

The new (2018) FIGO staging has led to a total of 11 categories of loco-regionally advanced cervix cancer (LRACC). While incorporating imaging in an improvement to pure clinical staging (2009 FIGO) this had led to more categories which are not necessarily prognostically discrete groups. We aimed toanalyse survival according to 2018 FIGO staging in curatively treated cervix cancer patients and identify iso-prognostic groups based on primary tumour volume and nodal status.

Methods

Patients referred for radiotherapy with curative intent between 1996–2014 were eligible. Baseline clinico-pathological and follow up information was retrieved from an ethics- approved institutional prospective database. Patients were classified to FIGO 2018 staging based on histo-pathology, MRI (for tumour volume and local compartmental spread) and PET (for nodal spread). Kaplan-Meier method was used to estimate survival at five years. Following survival analysis using recognised prognostic factors, iso-prognostic categories were identified and merged to form 5 iso-prognostic clusters.

Results

Seven hundred and forty-four LRACC patients met eligibility criteria were analysed. Median follow-up was 6 years. Iso-prognostic groups (Clusters) cross-tabulated against 2018 FIGO stages shows heterogenous 5 years survival (last column; table 1) across the FIGO stages, as compared with progressively worsening prognosis in the iso-prognostic clusters (last row; table 1).

Conclusions

Prognosis in LRACC depends on the interplay between primary tumour characteristics; type of local spread and nodal disease. A study of survival and patterns of failure according to iso-prognostic clusters would be useful in selection of appropriate treatment modality; estimating survival as well as better patient selection for clinical trials.