961P - Sodium-glucose co-transporter 2 inhibitors (SGLT2i) improve oncologic outcomes in diabetic patients with hepatocellular carcinoma (HCC): A nationwide database study

Background

Emerging evidence highlights the potential anticarcinogenic effects of SGLT2i. This study explored the impact of SGLT2i on HCC prognosis in diabetic patients.

Methods

We searched the databases of National Health Insurance and Taiwan Cancer Registry for patients with diagnosis of type 2 diabetes mellitus who received curative surgery or ablation for HCC between January 1, 2016 and December 31, 2018. SGLT2i users were defined as those who started SGLT2i either before or within two years following the curative procedure. Patients receiving additional anticancer therapies before SGLT2i use were excluded. We assessed overall survival (OS), cancer-specific survival (CSS), and time to next treatment (TTNT) using SGLT2i use as a time-varying covariate in Cox proportional hazards models.

Results

The study included 3,219 patients; 1042 (32.4%) were female, the median age was 67.0 years, 1385 (43.0%) were seropositive of hepatitis B virus, 1134 (35.2%) were seropositive for hepatitis C virus, and 2943 (91.4%) had Child-Pugh A liver reserve. Barcelona Clinic Liver Cancer stage was 0, A and B in 602 (18.7%), 2073 (64.4%) and 544 (16.9%) patients, respectively. Among all patients, 539 (16.7%) were defined as SGLT2i users (290 [9.2%] started prior to curative procedures, and 249 [8.5%] started within two years afterwards). The 3-year OS rate was higher in SGLT2i users (84.6%) compared to non-users (79.5%). In multivariate analysis, SGLT2i use was an independent predictor for longer OS (hazard ratio [HR], 0.70; 95% confidence interval [CI], 0.54-0.90) and CSS (HR, 0.56; 95% CI, 0.41-0.78), but not TTNT (HR, 0.96; 95% CI, 0.83-1.12). SGLT2i user were 1:1 matched to never users according to propensity scores for demographics, hepatitis etiology, liver reserve, tumor extent, pre-treatment alpha-fetoprotein, comorbidities, and conjunctive medications. In the matched cohort, SGLT2i use significantly predicted longer OS (HR, 0.62; 95% CI, 0.42-0.92), CSS (HR, 0.56; 95% CI, 0.31-0.99), and TTNT (HR,0.71; 95% CI, 0.54-0.93).

Conclusions

SGLT2i use was associated with an improved prognosis of diabetic patients who received curative treatment for HCC.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.