A Bio-Imaging Signature as a Predictor of Clinical Outcomes in Locally Advanced Pancreatic Cancer.
Cancers (Basel)
; 12(8)2020 Jul 23.
Article
in En
| MEDLINE
| ID: mdl-32717967
ABSTRACT
Purpose:
To evaluate the predictive value of 18F-FDG PET/CT semiquantitative parameters of the primary tumour and CA 19-9 levels assessed before treatment in patients with locally advanced pancreatic cancer (LAPC).Methods:
Among one-hundred twenty patients with LAPC treated at our institution with initial chemotherapy followed by curative chemoradiotherapy (CRT) from July 2013 to January 2019, a secondary analysis with baseline 18F-FDG PET/CT was conducted in fifty-eight patients. Pre-treatment CA 19-9 level and the maximum standardized uptake value (SUVmax), metabolic tumour volume (MTV) and total lesion glycolysis (TLG) of primary tumour were measured. The receiving operating characteristics (ROC) analysis was performed to define the cut-off point of SUVmax, MTV, TLG and CA 19-9 values to use in prediction of early progression (EP), local progression (LP) and overall survival (OS). Areas under the curve (AUCs) were assessed for all variables. Post-test probability was calculated to evaluate the advantage for parameters combination.Results:
For EP, CA 19-9 level > 698 U/mL resulted the best marker to identify patient at higher risk with OR of 5.96 (95% CI, 1.66-19.47; p = 0.005) and a Positive Predictive Value (PPV) of 61%. For LP, the most significant parameter was TLG (OR 9.75, 95% CI, 1.64-57.87, p = 0.012), with PPV of 83%. For OS, the most significant parameter was MTV (OR 3.12, 95% CI, 0.9-10.83, p = 0.07) with PPV of 88%. Adding consecutively each of the other parameters, PPV to identify patients at risk resulted further increased (>90%).Conclusions:
Pre-treatment CA 19-9 level, as well as MTV and TLG values of primary tumour at baseline 18F-FDG PET/CT and their combination, may represent significant predictors of EP, LP and OS in LAPC patients.
Full text:
1
Collection:
01-internacional
Database:
MEDLINE
Type of study:
Prognostic_studies
/
Risk_factors_studies
Language:
En
Journal:
Cancers (Basel)
Year:
2020
Document type:
Article
Affiliation country: