Toulsie Ramtohul1,2, Mohamed Abdul-Baki3, Manuel Rodrigues4,5, Nathalie Cassoux6,7, Sophie Gardrat5,8, Khadija Ait Rais9, Gaëlle Pierron9, Toufik Bouhadiba10, Vincent Servois11,12, Pascale Mariani10. 1. Department of Radiology, Institut Curie, Paris, PSL Research University, Paris, France. toulsie.ramtohul@curie.fr. 2. INSERM U1288, Laboratoire d'Imagerie Translationnelle en Oncologie, PSL Research University, 91400, Orsay, France. toulsie.ramtohul@curie.fr. 3. Department of Radiology, Institut Curie, Paris, PSL Research University, Paris, France. 4. Department of Medical Oncology, Institut Curie, PSL Research University, Paris, France. 5. INSERM U830, DNA Repair and Uveal Melanoma (D.R.U.M.), Paris, France. 6. Department of Ocular Oncology, Institut Curie, PSL Research University, Paris, France. 7. UMR 144 CNRS, Université de Paris, Paris, France. 8. Department of Biopathology, Institut Curie, PSL Research University, Paris, France. 9. Somatic Genetic Unit, Department of Genetics, Institut Curie, PSL University, Paris, France. 10. Department of Surgical Oncology, Institut Curie, PSL Research University, Paris, France. 11. Department of Radiology, Institut Curie, Paris, PSL Research University, Paris, France. vincent.servois@curie.fr. 12. INSERM U1288, Laboratoire d'Imagerie Translationnelle en Oncologie, PSL Research University, 91400, Orsay, France. vincent.servois@curie.fr.
Abstract
BACKGROUND: Surgical management of liver metastases of uveal melanoma (LMUM) is associated with the best survival rates, especially for patients with a low tumor burden in the liver. The aim was to determine whether the tumor growth rate (TGR0) before liver resection helps predict survival in patients with resectable LMUM. METHODS: This retrospective study included 99 patients with LMUM treated with liver resection between November 2007 and November 2020. TGR0 was expressed as the percentage change in tumor volume over 1 month according to two pretreatment imaging scans. Multivariate Cox analyses identified independent predictors of disease-free survival (DFS) and overall survival (OS). RESULTS: DFS and OS had a statistically significant positive linear relationship (Spearman correlation r = 0.68, p < 0.001). A disease-free interval (DFI) > 24 months and a TGR0 ≤ 50%/month were independent factors associated with better DFS and OS. The 2-component model including TGR0 and DFI had a mean time-dependent area under the curve (AUC) of 0.81 (95% CI, 0.75-0.86) and 0.77 (95% CI, 0.67-0.87), respectively, for predicting DFS and OS. DFI with TGR0 defined three kinetic risk groups that had distinct DFS and OS outcomes (p < 0.001). Cytogenetic alterations at baseline were partially predictive factors of the kinetic risk score based on TGR0 and DFI. DISCUSSION: The assessment of TGR0 improves prognostic stratification by identifying patients at high risk of recurrence and poor survival after liver resection. TGR0 and DFI, reflecting tumor aggressivity, have the potential to be important markers for systemic adjuvant decisions.
BACKGROUND: Surgical management of liver metastases of uveal melanoma (LMUM) is associated with the best survival rates, especially for patients with a low tumor burden in the liver. The aim was to determine whether the tumor growth rate (TGR0) before liver resection helps predict survival in patients with resectable LMUM. METHODS: This retrospective study included 99 patients with LMUM treated with liver resection between November 2007 and November 2020. TGR0 was expressed as the percentage change in tumor volume over 1 month according to two pretreatment imaging scans. Multivariate Cox analyses identified independent predictors of disease-free survival (DFS) and overall survival (OS). RESULTS: DFS and OS had a statistically significant positive linear relationship (Spearman correlation r = 0.68, p < 0.001). A disease-free interval (DFI) > 24 months and a TGR0 ≤ 50%/month were independent factors associated with better DFS and OS. The 2-component model including TGR0 and DFI had a mean time-dependent area under the curve (AUC) of 0.81 (95% CI, 0.75-0.86) and 0.77 (95% CI, 0.67-0.87), respectively, for predicting DFS and OS. DFI with TGR0 defined three kinetic risk groups that had distinct DFS and OS outcomes (p < 0.001). Cytogenetic alterations at baseline were partially predictive factors of the kinetic risk score based on TGR0 and DFI. DISCUSSION: The assessment of TGR0 improves prognostic stratification by identifying patients at high risk of recurrence and poor survival after liver resection. TGR0 and DFI, reflecting tumor aggressivity, have the potential to be important markers for systemic adjuvant decisions.