Impact of disparity between imaging and pathological tumor size on cancer-specific prognosis among patients with hepatocellular carcinoma

Abstract

Background

The association between preoperative imaging and postoperative pathological tumor size disparity, and cancer-specific survival (CSS) among patients undergoing hepatectomy for hepatocellular carcinoma (HCC) remains unclear. We sought to evaluate this association and identify predictors of size disparity.

Method

Patients undergoing curative-intent hepatectomy for HCC (2000–2023) were identified from an international, multi-institutional database. Size ratio was defined as the ratio of pathological to imaging tumor size. Patients with a size ratio of 0.5–1.5 were classified as “without size disparity,” while patients outside this range were considered “with size disparity.” Multivariable Cox regression was used to identify predictors of CSS, while logistic regression was utilized to determine factors associated with size disparity. For variables identified as significant in multivariable analyses, further evaluation including cutoff determination, were performed using receiver operating characteristic (ROC) analysis.

Results

Among 833 patients, median size ratio was 1.02, with a strong correlation between imaging and pathological tumor sizes (r = 0.87). Size disparity was present in 106 patients (12.7 %); in general, patients had smaller median imaging sizes (2.85 vs. 4.80 cm; p < 0.001) while size on pathology was noted to be larger(both 4.50 cm; p = 0.370). Patients with size disparity had worse 5-year CSS (60.1% vs. 79.0 %; p < 0.001). Multivariable Cox regression identified higher ALBI score (HR:2.56 [1.50–4.37]; p < 0.001), larger pathological tumor size (HR:1.09 [1.03–1.15]; p = 0.001), and size disparity (HR:2.53 [1.37–4.66]; p = 0.002) as independent predictors of CSS. Logistic regression demonstrated that cirrhosis (OR: 2.68 [1.43–5.02]; p = 0.002) and log alpha-fetoprotein (AFP) (OR:1.11 [1.01–1.22]; p = 0.030) were associated with an increased likelihood of size disparity. Cirrhosis and log AFP could be used to stratify patients relative to probability of a size disparity (low-risk:9.9 %, medium-risk:12.2 %, high-risk:17.7 %). The optimal AFP cutoff value was 3928 ng/mL for non-cirrhotic (AUC:0.90) versus 28.9 ng/mL for cirrhotic (AUC:0.74) patients.

Conclusion

Tumor size disparity was associated with worse CSS among patients with HCC undergoing hepatectomy. Size disparity could be predicted preoperatively using cirrhosis status and AFP level, which may help identify high-risk patients who may benefit from more detailed imaging assessments.