A264 ADHERENCE TO ENHANCED POST-TREATMENT SURVEILLANCE IS ASSOCIATED WITH INCREASED DETECTION OF EARLY STAGE RECURRENCE AFTER RADIOFREQUENCY ABLATION BUT NOT SURGICAL MANAGEMENT OF HEPATOCELLULAR CARCINOMA

Abstract

Background: Little is known about the optimal surveillance schedule after curative intent treatment of hepatocellular carcinoma (HCC). Current guidelines advocate cross-sectional imaging at 3-6 months intervals for the first 2 years, but are not backed by strong evidence of benefit. In Calgary, we perform MRI one month after radiofrequency ablation (RFA) and 3-6 months after surgery. Thereafter we alternate contrast enhanced ultrasound and MRI every 3 months for two years and then every six months for another 3 years. Aims: We conducted a retrospective study to investigate the impact of surveillance intensity on clinical outcomes following curative HCC treatment. Methods: From a combined surgery and interventional radiology database in Calgary (2012-2015), 124 unique HCC patients receiving post-treatment surveillance after successful resection (n=46) or RFA (n=78) were identified after excluding patients who received transarterial chemoembolization (TACE) (n=43), failed to achieve radiological remission (n=38) or died within three months (n=10). Baseline characteristics, imaging frequency, and clinical outcomes were collected. A surveillance rate defined as sum of actual number of images performed divided by the sum of the expected number of images in the defined surveillance period for each subject was calculated. Results: Baseline characteristics including age, gender and liver disease were similar in both groups. The surgery group had a higher percent of Barcelona Clinic Liver Cancer (BCLC) staging 0 (p=0.00052), non-cirrhotic (p=0.0045), and first HCC (p=0.00027) then the RFA group. The two-year disease-free survival and five-year mortality rate on Kaplan-Meier analysis, as well as mean surveillance rate (89±15% vs 84±21%) were comparable. Interestingly, BCLC B/C recurrence (n=8) was significantly associated with a lower surveillance rate as compared to BCLC 0/A (n=32) recurrence in RFA patients (76±22% vs 92±14%, p=0.015), but this was not observed in the surgical cohort. Comparison of mean imaging interval (120±47 days vs 82±29 days, p =0.038) and time to recurrence (591±245 days vs 399±222 days, p=0.048) between BCLC B/C and 0/A recurrence also differed in the RFA cohort. Univariant analysis did not identify other predictors of more significant recurrence. Conclusions: In conclusion, high intensity post-treatment surveillance of HCC patients after locoregional therapy, but not surgical resection, appears to associated with detection of recurrence at an earlier stage. Ongoing follow-up will determine if this is associated with a survival benefit.