Validation of an independent prognostic value of the asphericity of FDG uptake in non-small cell cancer patients undergoing treatment in curative intent


Validation of an independent prognostic value of the asphericity of FDG uptake in non-small cell cancer patients undergoing treatment in curative intent

Rogasch, J.; Chibolela, C.; Frost, N.; Wedel, F.; Hofheinz, F.; Rückert, J.; Neudecker, J.; Böhmer, D.; von Laffert, M.; Amthauer, H.; Furth, C.

Ziel/Aim:

In patients with non-small cell lung cancer (NSCLC) undergoing treatment with curative intent, the asphericity (ASP) of the primary tumor’s metabolic tumor volume (MTV) has been demonstrated as a prognostic factor. This study aimed at validation in an independent cohort with sufficient sample size.

Methodik/Methods:

Retrospective study in 313 NSCLC patients (203 men; median age, 67 [41-87] a) undergoing FDG-PET/CT with the same scanner prior to treatment in curative intent (always including resection of the primary tumor). 137 patients had UICC stage I, 79 patients stage II and 97 patients stage III disease (adenocarcinoma [ADC], 153; squamous cell carcinoma [SCC], 143, other, 17). Delineation of primary tumor MTV with semiautomated background-adapted threshold relative to its SUVmax. Univariable Cox regression for progression-free (PFS) and overall survival (OS) for PET parameters (MTV, ASP, SUVmax, SUVmean), clinical (UICC stage I vs. II vs. III), histological (SCC vs. ADC) and treatment variables (Rx/1 vs. R0 resection, chemotherapy yes/no, radiotherapy yes/no). Multivariable Cox of parameters significant in univariate Cox.

Ergebnisse/Results:

Events (progression, relapse, death) occurred in 169/313 patients, 139 patients died (median follow-up in survivors, 37 months). In multivariable Cox for OS, ASP >33.3% (hazard ratio [HR], 1.53 [95%-confidence interval, 1.02-2.3]), Rx/1 vs. R0 resection (HR, 2.47 [1.5-4.2]) and SCC vs. ADC (HR, 1.53 [1.1-2.2]) were significant. Log-rank test for ASP was significant at any cut-off from 18% upwards. Among separate UICC stages, ASP was only prognostic for OS in stage II (optimal, >19.5%; median OS, 33 vs. 59 months; p<0.01). In multivariable Cox for PFS, ASP >21.2% (HR, 1.75 [1.2-2.5]) and Rx/1 vs. R0 (HR, 2.48 [1.5-4.1]) were significant. Log-rank test for ASP was significant at any cut-off from 10-60%.

Schlussfolgerungen/Conclusions:

ASP was validated as an independent predictor of PFS and OS in NSCLC patients with curative treatment intent. Subdividing UICC stages, ASP remained prognostic in stage II.

  • Lecture (Conference)
    Jahrestagung der DGN, 03.-06.04.2019, Bremen, Deutschland

Permalink: https://www.hzdr.de/publications/Publ-29094
Publ.-Id: 29094