2159 - Prognostic Value of Pre- and Post-Treatment PET-CT Metrics after Definitive Chemoradiation for Anal Canal Cancer
Presenter(s)
R. M. Narasimhan1, Y. Wang2, A. Vera3, Y. Yang4, I. C. Ogobuiro5, R. Kuker6, K. R. Padgett7, and A. H. Wolfson8; 1University of Miami Miller School of Medicine, Miami, FL, 2University of Miami - Sylvester Cancer Center, Coral Gables, FL, 3Virginia Commonwealth University Health, Department of Radiation Oncology, Richmond, VA, 4University of Miami, Miami, FL, United States, 5Department of Radiation Oncology, University of Miami, Sylvester Comprehensive Cancer Center, Miami, FL, 6Jackson Memorial Hospital, University of Miami Health System, MIAMI, FL, 7Department of Radiation Oncology, Division of Radiation Physics, University of Miami/Sylvester Comprehensive Cancer Center, Miami, FL, 8Department of Radiation Oncology, University of Miami/Sylvester Comprehensive Cancer Center, Miami, FL
Purpose/Objective(s):
Post-treatment PET-CT is generally obtained after definitive chemoradiation (CRT) for locally advanced anal canal cancer (LAACC) patients (pts), yet its prognostic utility remains uncertain. We hypothesized that quantitative PET-CT metrics, including adapted Positron Emission Tomography Response Criteria in Solid Tumors (PERCIST) measurements based on peak standardized uptake value normalized to lean body mass (SULpeak), predict overall survival (OS), and that pretreatment metabolic parameters stratify risk better than post-treatment response alone in LAACC pts treated with CRT.Materials/Methods:
We conducted a retrospective cohort study of LAACC pts who underwent definitive CRT and pre and post-treatment fluorodeoxyglucose (FDG) PET-CT. Quantitative PET metrics were extracted for primary tumors, including size, mean SUL, total lesion glycolysis (TLG), metabolic tumor volume (MTV), and SULpeak. OS was estimated via Kaplan-Meier methods. Cox proportional hazards regression with MaxStat methodology identified optimal cutpoints and evaluated associations between FDG PET-derived variables and OS, with constraints applied to ensure statistical robustness and avoid edge effects given sample size.Results:
Forty-five pts met inclusion criteria (median age 57.7 years), with median follow-up time of 2.6 months between CRT completion and posttreatment imaging. On univariable Cox analysis, lean body mass (HR 3.10, 95% CI 1.01–9.50; p=0.049; cutoff 48.32 kg; N=44) and pretreatment variables including primary tumor mean SUL (HR 4.93, 95% CI 1.40–17.37; p=0.013; cutoff 4.64; N=41), size (HR 4.01, 95% CI 1.22–13.17; p=0.022; cutoff 45.75 mm; N=42), TLG (HR 3.68, 95% CI 1.11–12.14; p=0.033; cutoff 239.39; N=41), and SULpeak (HR 3.31, 95% CI 1.02–10.72; p=0.046; cutoff 8.19; N=42) were significantly correlated with OS. Of post-treatment variables, higher weight (HR 4.94, 95% CI 1.06–22.96; p=0.041; cutoff 60.42 kg; N=45) and higher lean body mass (HR 4.82, 95% CI 1.06–21.90; p=0.042; cutoff 44.50 kg; N=44) predicted worse survival. No post-treatment PET-CT metrics showed statistical significance in predicting OS. In subgroup analyses of pts with high pretreatment mean SUL, a greater reduction in mean SUL post CRT trended toward improved OS, though not significant (n=10, ROC-AUC=0.88, log-rank p=0.205).Conclusion:
In our cohort, many pre-treatment PET-CT metabolic metrics strongly predicted OS; post-treatment PET-CT response alone did not reliably predict outcomes. These findings suggest that baseline tumor metabolic activity may inform prognosis more reliably than post-treatment metrics. Larger studies incorporating prospectively standardized PET metrics and multivariable modeling may improve pre-CRT risk stratification and post-CRT surveillance in pts with curable LAACC.