Background: Adrenocortical carcinoma (ACC) has an aggressive but variable behaviour. ENSAT tumour stage and Ki67 proliferation index are used to predict clinical outcome but they are limited in distinguishing patients with best outcome. We aimed to investigate the prognostic role of clinical/histopathological parameters alone or in combination according to previously proposed points-based score (mGRAS, Lippert JCEM 2018).
Methods: We assessed 112 patients with histologically-proven ACC recording age, presence of steroid-or mass-related symptoms at diagnosis, ENSAT stage, tumour resection status, Ki67, progression-free survival (PFS), and overall survival (OS). Complete data were available for 72 patients (median age 50yr; 33M:39F; median follow-up 30 months). Each parameters prognostic value was tested by univariate and multivariate analyses. Additionally, we evaluated the prognostic performance of mGRAS as follows: age (<50yr = 0; ≥50yr = 1), symptoms (no = 0; yes = 1), ENSAT stage (12 = 0; 3 = 1; 4 = 2), resection status (R0 = 0; RX = 1; R1 = 2; R2 = 3), and Ki67 (09% = 0; 1019% = 1; ≥20% = 2 points).
Results: Univariate analysis showed prognostic prediction of PFS employing presence of symptoms (P 0.003, X2 8.4), ENSAT stage (P <0.001, X2 32.5), and Ki67 (P 0.02, X2 7.0). These remained significant at multivariate analysis. ENSAT stage and Ki67 also showed significant impact on OS (P 0.002, X2 12.2 and P 0.013, X2 8.6, respectively). The mGRAS score showed superior prognostic stratification by identifying 3 groups with different PFS (median: undefined, 25, and 7 months, respectively, P <0.001, X2 20.9) and OS (median: undefined, 89, and 21 months, P <0.001, X2 20.3).
Conclusion: The prognostic performance of the mGRAS score is superior to that of individual clinical/histopathological parameters. This simple score may guide personalised treatment decisions, e.g. the need for adjuvant therapy and monitoring frequency.