Revisiting preoperative CT tumor size for predicting adrenocortical carcinoma risk: a multicenter registry analysis of 614 adrenalectomies

Adrenocortical carcinoma (ACC) is a rare but highly aggressive malignancy requiring prompt recognition and surgical intervention. Preoperative determination of malignancy risk relies heavily on imaging, yet optimal tumor size cutoffs for surgical decision-making remain debated. This study aimed to evaluate the predictive value of preoperative tumor size for adrenocortical carcinoma (ACC) and to define an exploratory size threshold for malignancy risk stratification using a large multicenter database. A retrospective analysis was conducted using the Turkish Uro-Oncology Association Adrenal Database. Of 1,100 screened patients, 614 with complete clinico-pathological data were included; those with adrenal metastases and malignancies other than ACC were excluded. Patients were categorized into ACC and non-ACC groups. Clinical characteristics, tumor size, and surgical outcomes were compared. Receiver operating characteristic (ROC) analysis and multivariable binary logistic regression were performed to assess independent predictors of ACC and determine the optimal diagnostic cut-off threshold. Among 614 patients, ACC was identified in 32 cases (5.2%). The risk of ACC increased progressively across tumor size strata, rising from 1.8% in lesions < 4.0 cm to 5.5% in lesions 4.0–6.0 cm, and reaching 19.5% in lesions ≥ 6.0 cm. Preoperative CT-measured tumor size was significantly larger in ACC cases ( p < 0.001), whereas age, sex, laterality, incidental detection, and hormonal activity showed no significant univariate differences. ROC analysis revealed that tumor diameter was strongly associated with ACC (AUC: 0.814, 95% CI: 0.738–0.890, p < 0.001). An exploratory cutoff value of 45.5 mm yielded a sensitivity of 71.9% (95% CI: 53.3%–86.3%), specificity of 80.1% (95% CI: 76.6%–83.2%), positive predictive value of 16.5%, negative predictive value of 98.1%, and overall accuracy of 79.6%. In multivariable logistic regression, continuous tumor size remained an independent predictor of ACC (aOR: 1.017 per mm, 95% CI: 1.002–1.031, p = 0.024), while a tumor diameter ≥ 45.5 mm demonstrated a statistically significant 4.5-fold increase in malignancy risk (aOR: 4.516, 95% CI: 1.543–13.220, p = 0.006). Preoperative CT tumor size is a useful independent predictor for discriminating Non-ACC adrenal masses from ACC, with lesions exceeding 45.5 mm conferring a significantly higher risk of malignancy (aOR: 4.516). Rather than replacing current guidelines, our multicenter findings provide robust empirical validation for the established 40 mm threshold, which demonstrates high sensitivity (78.1%) and negative predictive value (98.2%) as a safe clinical trigger. The identified 45.5-mm threshold serves strictly as an exploratory, complementary risk-stratification tool rather than a stand-alone criterion for malignancy or absolute surgical decision-making. While open adrenalectomy remains the standard for confirmed or highly suspected ACC, laparoscopic surgery should be reserved for carefully selected smaller lesions without signs of local invasion in experienced centers.