Lower risks of sodium glucose cotransporter 2 (SGLT2) inhibitors compared to dipeptidyl peptidase-4 (DPP4) inhibitors for new-onset non-alcoholic fatty liver disease and hepatocellular carcinoma in type 2 diabetes mellitus: A population-based study ======================================================================================================================================================================================================================================================== * Oscar Hou In Chou * Jing Ning * Raymond Ngai Chiu Chan * Cheuk To Chung * Helen Huang * Kenrick Ng * Edward Christopher Dee * Sharen Lee * Apichat Kaewdech * Tong Liu * Fengshi Jing * Bernard Man Yung Cheung * Gary Tse * Jiandong Zhou ## Abstract **Background** The association between sodium glucose cotransporter 2 inhibitors (SGLT2I) versus dipeptidyl peptidase-4 inhibitors (DPP4I) and the risks of non-alcoholic fatty liver disease (NAFLD) and hepatocellular carcinoma (HCC) are currently unknown. **Methods** This was a retrospective population-based cohort study including type-2 diabetes mellitus (T2DM) patients treated with either SGLT2I or DPP4I between 1st January 2015 and 31st December 2019 in Hong Kong. Patients with concurrent DPP4I and SGLT2I usage were excluded. The primary outcomes were NAFLD and HCC. The secondary outcomes included cancer-related mortality and all-cause mortality. Propensity score matching (1:1 ratio) was performed using the nearest neighbour search. Univariable and multivariable Cox regression was applied to identify significant predictors. Competing risks models and multiple approaches using the propensity score were performed. **Results** This cohort included 62699 patients with T2DM, amongst which 22154 patients were on SGLT2I and 40545 patients were on DPP4I. After matching (44308 patients), 1090 patients developed new-onset NAFLD (Incidence: 4.6; 95% Confidence interval [CI]: 4.3-4.9) and 187 patients developed HCC (Incidence: 0.8; 95% CI: 0.7-0.9). Overall, SGLT2I was associated with lower risks of NAFLD (Hazard ratio [HR]: 0.39; 95% CI: 0.34-0.46), and HCC (HR: 0.46; 95% CI: 0.29-0.72) compared to DPP4I after adjustments. SGLT2I was also associated with lower risks of cancer-related mortality (HR: 0.29; 95% CI: 0.23-0.37) and all-cause mortality (HR: 0.28; 95% CI: 0.25-0.31). However, amongst patients with hepatitis B virus infection, SGLT2I was associated with higher risks of HCC (HR: 3.28; 95% CI: 1.21-8.90). The results were consistent in competing risk models and different matching approaches. **Conclusion** SGLT2I was associated with lower risks of NAFLD, and HCC compared to DPP4I after propensity scores matching and adjustments. **Lay summary** The association between two antidiabetic medications, SGLT2I and DPP4I, and the risks of fatty liver disease and liver cancer have not been explored. In our study, SGLT2I was associated with a lower risk of fatty liver disease and liver cancer compared to DPP4I amongst patients with type 2 diabetes. However, DPP4I was associated with lower risks of liver cancer compared to SGLT2I among patients with hepatitis B virus infection ![Figure1](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2022/08/17/2022.08.16.22278847/F1.medium.gif) [Figure1](http://medrxiv.org/content/early/2022/08/17/2022.08.16.22278847/F1) Keywords * sodium glucose cotransporter 2 inhibitors (SGLT2I) * dipeptidyl peptidase-4 inhibitors (DPP4I) * non-alcoholic fatty liver disease (NAFLD) * hepatocellular carcinoma (HCC) * type-2 diabetes * cancer ## Introduction In the past several decades, the incidence of hepatocellular carcinoma (HCC) has continued to rise [1]. In 2020, HCC was the sixth most common type of cancer worldwide, accounting for most cases of liver cancers and ranked the third leading cause of cancer death [2]. The geographical distribution of disease burden varies significantly, with the highest incidences rates observed in Western Pacific and African regions [3]. Hong Kong has a high rate of HCC due to the prevalence of the hepatitis B virus (HBV) within the territory; HBV accounted for 80% of all cases of HCC from 1992 to 2016 [4]. The prognosis of advanced-stage HCC remains poor as symptoms rarely appear during the early stages of the disease and high-risk patients may not be provided timely surveillance [5]. Patients with chronic hepatitis C (HCV) infection, chronic HBV infection, non-alcoholic fatty liver disease (NAFLD) or alcohol-associated liver disease have a higher likelihood of developing cirrhosis and eventually HCC. [6]. Recently, it has been found that patients with type 2 diabetes mellitus (T2DM) are also associated with a higher propensity to develop HCC. A systemic review and meta-analysis revealed that diabetic patients had a 2.31-fold increased risk of HCC and 2.43-fold increased risk of HCC-related death compared to nondiabetic patients [7]. Meanwhile, antidiabetic medications such as metformin have demonstrated protective effects against the disease [8-10]. This led to the growing interest in exploring the prevent effects of novel antidiabetic medications such as sodium-glucose cotransporter 2 inhibitors (SGLT2I) and dipeptidyl peptidase-4 inhibitors (DPP4I) in HCC. As of now, several anti-diabetic medications show promising anti-tumour effects against HCC. SGLT2I reduces the blood glucose level by blocking the glucose reabsorption at the S1 segment of the proximal convoluted tubules of the kidney. Clinical evidence surrounding the effects of SGLT2I on NAFLD and HCC is relatively scarce. Several studies with a relatively short follow-up suggested that SGLT2I may reduce the risk of NAFLD, which may be indirectly linked to HCC [11-13]. A case report observed spontaneous regression of HCC post-SGLT2I treatment due to the reduction of angiogenesis-related cytokines [14]. DPP4I is an incretin-based antidiabetic drug which inhibits glucagon-like peptide-1 degradation [15]. DPP4I was previously demonstrated to lower the risks of HCC amongst HCV-infected patients in a retrospective cohort study [16]. A case report by Yamamoto *et al*. found a spontaneous regression in HCC after four weeks of DPP4I treatment [17]. Indeed, a meta-analysis demonstrated that DPP4I does not increase the risks of developing overall cancer compared to patients treated with a placebo or other drugs. However, direct comparison between SGLT2I and DPP4I on new-onset NAFLD and HCC remains limited. Therefore, the present study aims to compare the association of SGLT2I versus DPP4I on the risk of new-onset NAFLD and HCC in T2DM patients from Hong Kong. ## Methods ### Study design and population This was a retrospective, territory-wide cohort study of T2DM patients treated with SGLT2I or DPP4I between 1st January 2015, and 31st December 2019, in Hong Kong. Patients were followed up until 31st December 2020, or until death. This study was approved by The Joint Chinese University of Hong Kong–New Territories East Cluster Clinical Research Ethics Committee and complied with the Declaration of Helsinki. The patients were identified from the Clinical Data Analysis and Reporting System (CDARS), a territory-wide database that centralizes patient information from individual local hospitals to establish comprehensive medical data, including clinical characteristics, disease diagnosis, laboratory results, and drug treatment details. The system has been used by local teams in Hong Kong to conduct comparative studies [18-20] and recently by our team to examine adverse cardiovascular outcomes for SGLT2I/DPP4I users [21, 22]. Patients were excluded if they exhibited any of the following criteria: 1) with both DPP4I and SGLT2I use 2) without complete demographics 3) without mortality data 4) less than 18 years old 5) died within 30 days after drug exposure 6) prior HCC 7) new onset HCC development less than 1 year after drug exposure **(Figure 1)**. ![Figure 1.](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2022/08/17/2022.08.16.22278847/F2.medium.gif) [Figure 1.](http://medrxiv.org/content/early/2022/08/17/2022.08.16.22278847/F2) Figure 1. Procedures of data processing IR: Incidence rate; SGLT2I: Sodium-glucose cotransporter-2 inhibitors; DPP4I: Dipeptidyl peptidase-4 inhibitors. Patients’ demographics include gender and age of initial drug use (baseline), clinical and biochemical data were extracted for the present study. Prior comorbidities were extracted using the *International Classification of Diseases Ninth Edition* (ICD-9) codes (**Supplementary Table 1**). Charlson’s standard comorbidity index was also calculated. Both cardiovascular medications and anti-diabetic agents were also extracted. The baseline laboratory examinations, including the complete blood count, renal and liver biochemical tests, and the lipid and glucose profiles were extracted. HBV infection was defined by both the ICD-9 codes for HBV infection and positive HBsAg laboratory results. Hepatitis C virus (HCV) infection was defined by both the ICD-9 codes for HCV infection and positive anti-HCV laboratory results. The standard deviation variability measure for the lipid and glucose profiles were also calculated **(Supplementary Table 2)**. The estimated glomerular filtration rate (eGFR) was calculated using the abbreviated modification of diet in renal disease (MDRD) formula [23]. ### Adverse outcomes and statistical analysis The primary outcome included NAFLD (ICD9: 571.5, 571.8, 571.9) and HCC (ICD9:155). Mortality data were obtained from the Hong Kong Death Registry, a population-based official government registry with the registered death records of all Hong Kong citizens linked to CDARS. Mortality was recorded using the *International Classification of Diseases Tenth Edition* (ICD-10) coding. The endpoint date of interest for eligible patients was the event presentation date. The endpoint for those without primary outcome presentation was the mortality date or the endpoint of the study (31st December 2020). Descriptive statistics are used to summarize baseline clinical and biochemical characteristics of patients with SGLT2I and DPP4I use. For baseline clinical characteristics, the continuous variables were presented as mean (95% confidence interval [CI]/standard deviation [SD])) and the categorical variables were presented as total numbers (percentage). Continuous variables were compared using the two-tailed Mann-Whitney U test, whilst the two-tailed Chi-square test with Yates’ correction was used to test 2×2 contingency data. Propensity score matching with 1:1 ratio for SGLT2I use versus DPP4I use based on demographics, Charlson comorbidity index, prior comorbidities, non-SGLT2I/ DPP4I medications were performed using the nearest neighbour search strategy. We used Stata software (Version 16.0) to conduct the propensity score matching procedures. Baseline characteristics between patients with SGLT2I and DPP4I use before and after matching were compared with standardized mean difference (SMD), with SMD<0.20 regarded as well-balanced between the two groups. Proportional Cox regression models were used to identify significant risk predictors of adverse study outcomes. Subgroup analysis was conducted to identify the risk predictors stratified by age, gender, and HBV infection status. Cause-specific and subdistribution hazard models were conducted to consider possible competing risks. Multiple propensity adjustment approaches were used, including propensity score stratification [24], propensity score matching with inverse probability of treatment weighting [25] and propensity score matching with stable inverse probability weighting [26]. The hazard ratio (HR), 95% CI and P-value were reported. Statistical significance is defined as P-value < 0.05. All statistical analyses were performed with RStudio software (Version: 1.1.456) and Python (Version: 3.6). ## Results ### Baseline characteristics This was a retrospective, territory-wide cohort study of 76147 patients with T2DM treated with SGLT2I/DPP4I between 1st January 2015 and 31st December 2019 in Hong Kong. Patients during the aforementioned period were enrolled and followed up until 31st December 2020 or until their deaths. Patients with both DPP4I and SGLT2I use (N=12858), without complete demographics (N=17), without mortality data (N=13), less than 18 years old (N=135), died within 30 days after drug exposure (N=295), prior HCC (N=84), new onset HCC development less than 1 year after drug exposure (N=46) were excluded **(Figure 1)**. After exclusion, this study included a total of 62699 patients with T2DM (mean age: 63.3 years old [SD: 12.9]; 55.18% males). 22154 patients (35.33%) used SGLT2Is, and 40545 patients (64.67%) used DPP4Is. The DPP4I and SGLT2I cohorts were comparable after matching **(Supplementary Figure 1)**. In the matched cohort, 1090 (2.46%) patients developed NAFLD, and 187 patients (0.42%) developed HCC. The characteristics of patients are shown in **Table 1, Supplementary Table 3 and 4**. View this table: [Table 1.](http://medrxiv.org/content/early/2022/08/17/2022.08.16.22278847/T1) Table 1. Baseline and clinical characteristics of patients with SGLT2I v.s. DPP4I use before and after propensity score matching (1:1). *for SMD>0.2; SD: standard deviation; SGLT2I: sodium glucose cotransporter-2 inhibitor; DPP4I: dipeptidyl peptidase-4 inhibitor; ACEI: Angiotensin-converting enzyme inhibitors; ARBs: angiotensin receptor blockers; MDRD: modification of diet in renal disease; NAFLD: Non-alcoholic fatty liver disease; # indicated the difference between SGLT2I users and DPP4I users. Over a total follow-up of 236856.5 person-years, the incidence rate (IR) of NAFLD was lower amongst SGLT2I users (IR: 3.4; 95% CI: 3.0-3.7) compared to DPP4I (IR: 5.9; 95% CI: 5.5-6.4) after propensity score matching **(Table 2)**. Meanwhile, after a follow-up of 239941.9 person-year, the incidence of HCC was lower amongst SGLT2I users (IR: 0.3; 95% CI: 0.2-0.4) compared to DPP4I users (IR: 1.3; 95% CI: 1.1-1.5). SGLT2I users (IR: 1.1; 95% CI: 0.9-1.3) also had a lower incidence of cancer-related mortality than DPP4I users (IR: 6.6; 95% CI: 6.1-7.1); the incidence of all-cause mortality was also lower amongst SGLT2I users (IR: 5.0; 95% CI: 4.6-5.4) than DPP4I users (IR: 24.4; 95% CI: 23.6-25.4). View this table: [Table 2.](http://medrxiv.org/content/early/2022/08/17/2022.08.16.22278847/T2) Table 2. Annualized incidence rate of primary and secondary outcomes in patient cohort after 1:1 propensity score matching. NAFLD: Non-alcoholic fatty liver disease; IR: Incidence rate; SGLT2I: sodium glucose cotransporter-2 inhibitor; DPP4I: dipeptidyl peptidase-4 inhibitor ### Significant predictors of the study outcomes Univariable Cox regression identified the significant risk factors for NAFLD and HCC before and after propensity score matching (1:1) **(Supplementary Table 5)**. In the multivariable Cox models, SGLT2I was associated with lower risks of NAFLD (Hazard ratio [HR]: 0.39; 95% CI: 0.34-0.46) and HCC (HR: 0.46; 95% CI: 0.29-0.72) after adjustments for significant demographics, past comorbidities, non-SGLT2I/DPP4I medications, abbreviated MDRD, fasting glucose, HbA1c, and duration from earliest diabetes mellitus date to initial drug exposure date. SGLT2I was also associated with lower risks of cancer-related mortality (HR: 0.29; 95% CI: 0.23-0.37) and all-cause mortality (HR: 0.28; 95% CI: 0.25-0.31) upon adjustments. The cumulative incidence curves stratified by SGLT2I versus DPP4I demonstrated that SGLT2I was associated with a lower cumulative hazard for NAFLD, HCC, cancer-related mortality and all-cause mortality after matching **(Figure 2)**. ![Figure 2.](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2022/08/17/2022.08.16.22278847/F3.medium.gif) [Figure 2.](http://medrxiv.org/content/early/2022/08/17/2022.08.16.22278847/F3) Figure 2. Cumulative incidence curves for all-cause mortality, cancer-related mortality, new onset HCC, new onset NAFLD stratified by drug exposure of SGLT2I and DPP4I before and after propensity score matching (1:1) ### Subgroup analysis and sensitivity analyses In the subgroup analysis, SGLT2I was associated lower cumulative incidence of NAFLD and HCC regardless of gender **(Figure 3 and Supplementary Figure 2)**. SGLT2I was associated with reduced risks of NAFLD compared to DPP4I particularly amongst patients younger than 65 years old regardless of history of hypertension and ischaemic heart diseases. SGLT2I was also associated with lower risks of HCC amongst patients older than 65 years old, without history of hypertension or ischaemic diseases. Amongst HBV negative patients, SGLT2I was associated with lower risks of NAFLD (HR: 0.35; 95% CI: 0.30-0.42), HCC (HR: 0.30; 95% CI: 0.18-0.50), cancer-related mortality (HR: 0.32; 95% CI: 0.25-0.41), and all-cause mortality (HR: 0.30; 95% CI: 0.27-0.34). Meanwhile, amongst HBV-positive patients, SGLT2I was associated with an insignificant risk of NAFLD (HR: 1.21; 95% CI: 0.69-2.13) and higher risks of HCC (HR: 3.28; 95% CI: 1.21-8.90) compared to DPP4I **(Supplementary Table 6)**. However, SGLT2I was associated with lower risk of cancer-related mortality (HR: 0.08; 95% CI: 0.05-0.19) and all-cause mortality (HR: 0.09; 95% CI: 0.04-0.12) among those patients. ![Figure 3A.](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2022/08/17/2022.08.16.22278847/F4.medium.gif) [Figure 3A.](http://medrxiv.org/content/early/2022/08/17/2022.08.16.22278847/F4) Figure 3A. Subgroup analyses for SGLT2I v.s. DPP4I exposure predict new onset NAFLD in the matched cohort. SGLT2I: Sodium-glucose cotransporter-2 inhibitors; DPP4I: Dipeptidyl peptidase-4 inhibitors. ![Figure 3B.](http://medrxiv.org/https://www.medrxiv.org/content/medrxiv/early/2022/08/17/2022.08.16.22278847/F5.medium.gif) [Figure 3B.](http://medrxiv.org/content/early/2022/08/17/2022.08.16.22278847/F5) Figure 3B. Subgroup analyses for SGLT2I v.s. DPP4I exposure predict new onset HCC in the matched cohort. SGLT2I: Sodium-glucose cotransporter-2 inhibitors; DPP4I: Dipeptidyl peptidase-4 inhibitors. The interaction effects with abbreviated MDRD demonstrated that SGLT2I was associated with lower risks of NAFLD, HCC, cancer-related mortality, and all-cause mortality across different abbreviated MDRD **(Supplementary Figure 3A)**. Meanwhile, as the neutrophil-to-lymphocyte ratio was lowered, the risks for NAFLD increased for DPP4I users while the risks for HCC, cancer-related mortality and all-cause mortality decreased (**Supplementary Figure 3B)**. The SGLT2I was also associated with lower risks of NAFLD, HCC, cancer-related mortality, and all-cause mortality across all aspartate aminotransferase-to-platelet ratio and fibrosis-4 index **(Supplementary Figure 3C and 3D)**. Sensitivity analyses were performed to confirm the predictive ability of the models. SGLT2I was associated with lower risks of NAFLD (HR: 0.45; 95% CI: 0.35-0.64) and HCC (HR: 0.42; 95% CI: 0.31-0.54) compared to DPP4I in the cause-specific hazard. SGLT2I was also associated with lower risks of NAFLD (HR: 0.56; 95% CI: 0.39-0.82) and HCC (HR: 0.53; 95% CI: 0.41-0.75) compared to DPP4I in the subdistribution model **(Table 4)**. SGLT2I also was associated with lower risks of new-onset NAFLD, HCC, cancer-related mortality, and all-cause mortality after different propensity score approaches (all P-values <0.0001). View this table: [Table 3.](http://medrxiv.org/content/early/2022/08/17/2022.08.16.22278847/T3) Table 3. Multivariate Cox regression models with adjustments to predict primary and secondary outcomes in the matched cohort. *for p≤ 0.05, ** for p ≤ 0.01, \***| for p ≤ 0.001; HR: hazard ratio; CI: confidence interval; SGLT2I: sodium glucose cotransporter-2 inhibitor; DPP4I: dipeptidyl peptidase-4 inhibitor; Model 1 adjusted for significant demographics. Model 2 adjusted for significant demographics, and past comorbidities. Model 3 adjusted for significant demographics, past comorbidities, and non-SGLT2I/DPP4I medications. Model 4 adjusted for significant demographics, past comorbidities, non-SGLT2I/DPP4I medications, abbreviated MDRD, fasting glucose, HbA1c, and duration from earliest diabetes mellitus date to initial drug exposure date. ^ Patients with prior NAFLD were excluded in the multivariate Cox regression models to predict new onset NAFLD. View this table: [Table 4.](http://medrxiv.org/content/early/2022/08/17/2022.08.16.22278847/T4) Table 4. Sensitivity analyses for SGLT2I v.s. DPP4I exposure predict new onset NAFLD and new onset HCC in the matched cohort. *for p≤ 0.05, ** for p ≤ 0.01, \***| for p ≤ 0.001; SGLT2I: Sodium-glucose cotransporter-2 inhibitors; DPP4I: Dipeptidyl peptidase-4 inhibitors; HR: hazard ratio; CI: confidence interval; PS: propensity score; IPTW: inverse probability of treatment weighting, SIPTW: stable inverse probability of treatment weighting. ## Discussion In this territory-wide retrospective cohort study, we used real-world data from routine clinical practice to compare the association between SGLT2I versus DPP4I and NAFLD and HCC. Our findings demonstrated that SGLT2I was associated with 58% lower risk of NAFLD and 55% lower risk of HCC than DPP4I users. However, amongst patients infected with HBV, DPP4I was associated with lower risks of HCC compared to SGLT2I. ### Comparison with previous studies T2DM is a metabolic syndrome characterized by hyperinsulinemia and contributes as a comorbidity in NAFLD, subsequently increasing the risk of HCC development [27, 28]. The prevalence of NAFLD in T2DM patients ranges from 29.6% to 87.1% [29]. T2DM patients have a 2.5-fold increased risk of developing HCC, while NAFLD further increases the risks [30, 31]. As the prevalence of NAFLD continues to be on the rise, managing diabetes becomes more pressing considering the risks of developing HCC. Antidiabetic agents such as metformin, thiazolidinediones, and GLP-1 analogues have been shown to improve the pathological manifestations of NAFLD and HCC in T2DM [32, 33]. The protective effects of SGLT2I on the cardiovascular system are well-established [34-37]. For hepatic diseases, the literature generally supports the notion that SGLT2I and DPP4I are beneficial in hepatic diseases but lack direct comparisons. The results from our study suggest that SGLT2I may have a lower risk of NAFLD and HCC development compared to DPP4I. Furthermore, the reduction in NAFLD and hepatocellular carcinoma might be one of the contributing factors to the reduction in all-cause mortality by reducing cancer-related mortality. It was suggested that a significant proportion of HCC developed amongst patients without cirrhosis [38]. In our results, SGLT2I reduced the risks of NAFLD and HCC risks across all baseline severity of liver fibrosis and cirrhosis, as indirectly reflected by the aspartate aminotransferase-to-platelet ratio and fibrosis-4 index interaction. Multiple studies have found that SGLT2I lowered the risk of NAFLD with pathological reductions of steatosis on histological examinations [13, 39-41]. In one study, SGLT2I reduced the risk for death and improved the survival of T2DM veterans with cirrhosis when compared to DPP4I [42]. Furthermore, SGLT2I also reduced the body mass index significantly and reduced the hepatocarcinogenesis for NAFLD [43]. On the other hand, while DPP4I studies have also been shown to reduce the risk of HCC in T2DM patients with chronic HCV infection [16, 44], data assessing the association with the risk of NAFLD are mixed [45-47]. A systematic review demonstrated no hepatic benefit associated with DPP4I in patients with hepatic steatosis but significant risk reductions in patients receiving SGLT2I [48]. Chronic HBV infection is still prevalent in Asia despite universal vaccination for individuals over 20 years [38]. The co-presence of steatohepatitis and HBV infection significantly increased the risks of HCC and death [49]. While SGLT2I lowered the risks of HCC amongst patients without HBV infection, SGLT2I was associated with higher risks of HCC (HR: 3.28; 95% CI: 1.21-8.90) compared to DPP4I amongst patients infected with HBV **(Supplementary Table 6)**. While HBV infection may contribute to more severe fibrosis, SGLT2I was associated with lower risks of HCC across all baseline severity of liver fibrosis. However, the effect can be due to dipeptidyl peptidase-4 (DPP4). DPP4 is an important molecule involved in the development of HCC, such that inhibition of DPP4 may help prevent HCC amongst HCV-infected patients. Indeed, the serum DPP4 level was previously suggested to be elevated amongst viral hepatitis patients [50]. However, the underlying mechanism mediating this finding in patients with T2DM and coinfection with HBV remains unclear. Future research is needed. ### Potential underlying mechanisms Several mechanisms have been proposed to explain the relationship between SGLT2I and NAFLD and HCC. It was hypothesized that SGLT2I inhibited de novo lipogenesis by inhibiting the expression of the FAS gene involved in fatty acid biosynthesis, which decreases fatty acid production and reduces steatosis [51]. Besides, SGLT2I has demonstrated anti-inflammatory and anti-steatosis properties preventing the progression of NAFLD to HCC [52]. Furthermore, SGLT2 receptors are highly expressed in liver tumours due to their increased demand for glucose for ATP synthesis and overall growth [53]. Meanwhile, DPP4I was also suggested to reduce the risks of HCC via several mechanisms. It was previously suggested that the GLP-1 hormone might help ameliorate liver fat accumulation and prevent the progression of NAFLD [54]. It also reduces HCC through the activation of lymphocyte chemotaxis and downregulation of the pentose phosphate pathway [55, 56]. However, there are also some conflicting results which suggest DPP4I may play a role in the progression of non-alcoholic steatohepatitis-related HCC by suppressing p62 and Keap1 [22]. Future research is needed to confirm the effects of SGLT2I and DPP4I in HCC. ### Clinical implications Given the importance of NAFLD and HCC in T2DM, [57], there is a need to investigate how SGLT2I and DPP4I may modify the risks for these diseases. The present study used data from routine clinical practice, which may influence the choice of second-line antidiabetic therapy in T2DM patients in terms of the hepatic disease risks. The findings of our study show that SGLTI and DPP4I may help prevent NAFLD and HCC compared to DPP4I. It particularly reduced the risks of NAFLD regardless of gender, prior history of hypertension and ischaemic heart diseases. However, amongst HBV-positive patients, DPP4I may be a better option to prevent the new-onset HCC. Generally, SGLT2I were found to reduce the risk of malignancies. SGLT2I patient groups had a lower risk of haematological and urinary tract malignancies in a nationwide study conducted by *Rokszin et al*. [48]. By exploring the association of SGLT2I and DPP4I in HCC, we add to the growing body of evidence supporting the use of antidiabetic agents in preventing NAFLD and potentially HCC. Further investigations are needed to confirm the causation relationship between SGLT2I and DPP4I with HCC, especially amongst HBV-positive patients. ### Limitations This study had several limitations. Firstly, given the observational nature of this study, there is inherent under-coding, coding errors, and missing data leading to information bias. The retrospective design necessitates the presentation of associations but not causal links between SGLT2I/DPP4I use and the risk of new-onset NAFLD and HCC. Secondly, medication adherence can only be assessed indirectly through prescription refills but not through direct measurement of drug exposure. Thirdly, residual and post-baseline confounding may be present despite robust propensity-matching, particularly with the unavailability of information on NAFLD and HCC risk factors such as smoking, obesity, and the potential overlooked alcohol consumption. As such, lipid profile was included in an attempt to account for obesity in an indirect way. Last but not least, the duration of drug exposure has not been controlled, which may affect their risk against the study outcomes. Time-varying analysis was also not included in this study. ## Conclusion SGLT2I was associated with lower risks of NAFLD and HCC compared to DPP4I after propensity scores matching and adjustments. SGLT2I was also associated with lower risks of cancer-related mortality and all-cause mortality compared to DPP4I. However, amongst patients infected with HBV, DPP4I was associated with lower risks of HCC compared to SGLT2I. The results supported the need for further evaluation in the prospective setting. ## Supporting information Supplementary Table 1 [[supplements/278847_file03.docx]](pending:yes) ## Data Availability An anonymised version without identifiable or personal information is available from the corresponding authors upon reasonable request for research purposes. ## Conflicts of Interest None. ## Funding source ECD is funded in part through the Cancer Center Support Grant from the National Cancer Institute (P30 CA008748). The other authors received no funding for the research, authorship, and/or publication of this article. ## Availability of data and materials An anonymised version without identifiable or personal information is available from the corresponding authors upon reasonable request for research purposes. ## Acknowledgements None. ## Footnotes * * Co-first authors * Received August 16, 2022. * Revision received August 16, 2022. * Accepted August 17, 2022. * © 2022, Posted by Cold Spring Harbor Laboratory This pre-print is available under a Creative Commons License (Attribution-NonCommercial-NoDerivs 4.0 International), CC BY-NC-ND 4.0, as described at [http://creativecommons.org/licenses/by-nc-nd/4.0/](http://creativecommons.org/licenses/by-nc-nd/4.0/) ## Reference 1. [1].Petrick JL, Kelly SP, Altekruse SF, McGlynn KA, Rosenberg PS. Future of Hepatocellular Carcinoma Incidence in the United States Forecast Through 2030. J Clin Oncol 2016;34:1787–1794. [Abstract/FREE Full Text](http://medrxiv.org/lookup/ijlink/YTozOntzOjQ6InBhdGgiO3M6MTQ6Ii9sb29rdXAvaWpsaW5rIjtzOjU6InF1ZXJ5IjthOjQ6e3M6ODoibGlua1R5cGUiO3M6NDoiQUJTVCI7czoxMToiam91cm5hbENvZGUiO3M6MzoiamNvIjtzOjU6InJlc2lkIjtzOjEwOiIzNC8xNS8xNzg3IjtzOjQ6ImF0b20iO3M6NTA6Ii9tZWRyeGl2L2Vhcmx5LzIwMjIvMDgvMTcvMjAyMi4wOC4xNi4yMjI3ODg0Ny5hdG9tIjt9czo4OiJmcmFnbWVudCI7czowOiIiO30=) 2. [2].Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries. CA: A Cancer Journal for Clinicians 2021;71:209–249. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.3322/caac.21660&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=http://www.n&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2022%2F08%2F17%2F2022.08.16.22278847.atom) 3. [3].Sayiner M, Golabi P, Younossi ZM. Disease Burden of Hepatocellular Carcinoma: A Global Perspective. Dig Dis Sci 2019;64:910–917. [PubMed](http://medrxiv.org/lookup/external-ref?access_num=http://www.n&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2022%2F08%2F17%2F2022.08.16.22278847.atom) 4. [4].Chui AMN, Yau TCC, Cheung TT. An overview in management of hepatocellular carcinoma in Hong Kong using the Hong Kong Liver Cancer (HKLC) staging system. Glob Health Med 2020;2:312–318. 5. [5].Lai CL, Lam KC, Wong KP, Wu PC, Todd D. Clinical features of hepatocellular carcinoma: review of 211 patients in Hong Kong. Cancer 1981;47:2746–2755. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1002/1097-0142(19810601)47:11<2746::AID-CNCR2820471134>3.0.CO;2-K&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=6266641&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2022%2F08%2F17%2F2022.08.16.22278847.atom) [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=A1981LS34100031&link_type=ISI) 6. [6].Kuang XJ, Jia RR, Huo RR, Yu JJ, Wang JJ, Xiang BD, et al. Systematic review of risk factors of hepatocellular carcinoma after hepatitis B surface antigen seroclearance. J Viral Hepat 2018;25:1026–1037. [PubMed](http://medrxiv.org/lookup/external-ref?access_num=http://www.n&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2022%2F08%2F17%2F2022.08.16.22278847.atom) 7. [7].Wang P, Kang D, Cao W, Wang Y, Liu Z. Diabetes mellitus and risk of hepatocellular carcinoma: a systematic review and meta-analysis. Diabetes Metab Res Rev 2012;28:109–122. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1002/dmrr.1291&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=21898753&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2022%2F08%2F17%2F2022.08.16.22278847.atom) 8. [8].Wainwright P, Scorletti E, Byrne CD. Type 2 Diabetes and Hepatocellular Carcinoma: Risk Factors and Pathogenesis. Curr Diab Rep 2017;17:20. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1007/s11892-017-0851-x&link_type=DOI) 9. [9].Zhang ZJ, Zheng ZJ, Shi R, Su Q, Jiang Q, Kip KE. Metformin for liver cancer prevention in patients with type 2 diabetes: a systematic review and meta-analysis. J Clin Endocrinol Metab 2012;97:2347–2353. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1210/jc.2012-1267&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=22523334&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2022%2F08%2F17%2F2022.08.16.22278847.atom) [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000306286100058&link_type=ISI) 10. [10].Cunha V, Cotrim HP, Rocha R, Carvalho K, Lins-Kusterer L. Metformin in the prevention of hepatocellular carcinoma in diabetic patients: A systematic review. Ann Hepatol 2020;19:232–237. 11. [11].Kontana A, Tziomalos K. Role of sodium-glucose co-transporter-2 inhibitors in the management of nonalcoholic fatty liver disease. World J Gastroenterol 2019;25:3664–3668. 12. [12].Eriksson JW, Lundkvist P, Jansson PA, Johansson L, Kvarnstrom M, Moris L, et al. Effects of dapagliflozin and n-3 carboxylic acids on non-alcoholic fatty liver disease in people with type 2 diabetes: a double-blind randomised placebo-controlled study. Diabetologia 2018;61:1923–1934. [PubMed](http://medrxiv.org/lookup/external-ref?access_num=http://www.n&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2022%2F08%2F17%2F2022.08.16.22278847.atom) 13. [13].Shao S-C, Kuo L-T, Chien R-N, Hung M-J, Lai EC-C. SGLT2 inhibitors in patients with type 2 diabetes with non-alcoholic fatty liver diseases: an umbrella review of systematic reviews. BMJ Open Diabetes Research & Care 2020;8:e001956. 14. [14].Kawaguchi T, Nakano D, Okamura S, Shimose S, Hayakawa M, Niizeki T, et al. Spontaneous regression of hepatocellular carcinoma with reduction in angiogenesis-related cytokines after treatment with sodium-glucose cotransporter 2 inhibitor in a cirrhotic patient with diabetes mellitus. Hepatol Res 2019;49:479–486. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1111/hepr.13247&link_type=DOI) 15. [15].Zhao M, Chen J, Yuan Y, Zou Z, Lai X, Rahmani DM, et al. Dipeptidyl peptidase-4 inhibitors and cancer risk in patients with type 2 diabetes: a meta-analysis of randomized clinical trials. Scientific Reports 2017;7:8273. 16. [16].Hsu WH, Sue SP, Liang HL, Tseng CW, Lin HC, Wen WL, et al. Dipeptidyl Peptidase 4 Inhibitors Decrease the Risk of Hepatocellular Carcinoma in Patients With Chronic Hepatitis C Infection and Type 2 Diabetes Mellitus: A Nationwide Study in Taiwan. Front Public Health 2021;9:711723. 17. [17].Yamamoto S, Tokuhara T, Nishikawa M, Nishizawa S, Nishioka T, Nozawa A, et al. Spontaneous regression of hepatocellular carcinoma after improving diabetes mellitus: possibly responsible for immune system. Kanzo 2012;53:167–174. 18. [18].Chou OHI, Zhou J, Lee TTL, Kot T, Lee S, Wai AKC, et al. Comparisons of the risk of myopericarditis between COVID-19 patients and individuals receiving COVID-19 vaccines: a population-based study. Clin Res Cardiol 2022:1–6. 19. [19].Zhou J, Chou OHI, Wong KHG, Lee S, Leung KSK, Liu T, et al. Development of an Electronic Frailty Index for Predicting Mortality and Complications Analysis in Pulmonary Hypertension Using Random Survival Forest Model. Front Cardiovasc Med 2022;9:735906. 20. [20].Chou OHI, Zhou J, Mui JV, Satti DI, Lee TTL, Lee S, et al. Lower risks of new-onset acute pancreatitis and pancreatic cancer in sodium glucose cotransporter 2 (SGLT2) inhibitors compared to dipeptidyl peptidase-4 (DPP4) inhibitors: a propensity score-matched study with competing risk analysis. medRxiv 2022:2022.2005.2027.22275702. 21. [21].Lee S, Zhou J, Leung KSK, Wai AKC, Jeevaratnam K, King E, et al. Comparison of Sodium-Glucose Cotransporter-2 Inhibitor and Dipeptidyl Peptidase-4 Inhibitor on the Risks of New-Onset Atrial Fibrillation, Stroke and Mortality in Diabetic Patients: A Propensity Score-Matched Study in Hong Kong. Cardiovasc Drugs Ther 2022. 22. [22].Zhou J, Lee S, Leung KSK, Wai AKC, Liu T, Liu Y, et al. Incident heart failure and myocardial infarction in sodium-glucose cotransporter-2 vs. dipeptidyl peptidase-4 inhibitor users. ESC Heart Fail 2022;9:1388–1399. 23. [23].Soliman AR, Fathy A, Khashab S, Shaheen N. Comparison of abbreviated modification of diet in renal disease formula (aMDRD) and the Cockroft-Gault adjusted for body surface (aCG) equations in stable renal transplant patients and living kidney donors. Ren Fail 2013;35:94–97. 24. [24].Austin PC. An Introduction to Propensity Score Methods for Reducing the Effects of Confounding in Observational Studies. Multivariate Behav Res 2011;46:399–424. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1080/00273171.2011.568786&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=21818162&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2022%2F08%2F17%2F2022.08.16.22278847.atom) [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000291533400002&link_type=ISI) 25. [25].Austin PC, Stuart EA. Moving towards best practice when using inverse probability of treatment weighting (IPTW) using the propensity score to estimate causal treatment effects in observational studies. Stat Med 2015;34:3661–3679. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1002/sim.6607&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=26238958&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2022%2F08%2F17%2F2022.08.16.22278847.atom) 26. [26].Avagyan V, Vansteelandt S. Stable inverse probability weighting estimation for longitudinal studies. Scandinavian Journal of Statistics 2021;48:1046–1067. 27. [27].Zhang C, Liu S, Yang M. Hepatocellular Carcinoma and Obesity, Type 2 Diabetes Mellitus, Cardiovascular Disease: Causing Factors, Molecular Links, and Treatment Options. Front Endocrinol (Lausanne) 2021;12:808526. 28. [28].Chan KE, Koh TJL, Tang ASP, Quek J, Yong JN, Tay P, et al. Global Prevalence and Clinical Characteristics of Metabolic Associated Fatty Liver Disease. A Meta-Analysis and Systematic Review of 10,739,607 Individuals. J Clin Endocrinol Metab 2022. 29. [29].Dai W, Ye L, Liu A, Wen SW, Deng J, Wu X, et al. Prevalence of nonalcoholic fatty liver disease in patients with type 2 diabetes mellitus: A meta-analysis. Medicine (Baltimore) 2017;96:e8179. 30. [30].Mantovani A, Targher G. Type 2 diabetes mellitus and risk of hepatocellular carcinoma: spotlight on nonalcoholic fatty liver disease. Ann Transl Med 2017;5:270. 31. [31].Huang DQ, El-Serag HB, Loomba R. Global epidemiology of NAFLD-related HCC: trends, predictions, risk factors and prevention. Nature Reviews Gastroenterology & Hepatology 2021;18:223–238. 32. [32].Tacelli M, Celsa C, Magro B, Giannetti A, Pennisi G, Spatola F, et al. Antidiabetic Drugs in NAFLD: The Accomplishment of Two Goals at Once? Pharmaceuticals (Basel) 2018;11. 33. [33].Kramer JR, Natarajan Y, Dai J, Yu X, Li L, El-Serag HB, et al. Effect of diabetes medications and glycemic control on risk of hepatocellular cancer in patients with nonalcoholic fatty liver disease. Hepatology 2022;75:1420–1428. 34. [34].Sfairopoulos D, Zhang N, Wang Y, Chen Z, Letsas KP, Tse G, et al. Association between sodium-glucose cotransporter-2 inhibitors and risk of sudden cardiac death or ventricular arrhythmias: a meta-analysis of randomized controlled trials. Europace 2022;24:20–30. 35. [35].Han SJ, Ha KH, Lee N, Kim DJ. Effectiveness and safety of sodium-glucose co-transporter-2 inhibitors compared with dipeptidyl peptidase-4 inhibitors in older adults with type 2 diabetes: A nationwide population-based study. Diabetes Obes Metab 2021;23:682–691. 36. [36].Bazoukis G, Papadatos SS, Thomopoulos C, Tse G, Cheilidis S, Tsioufis K, et al. Impact of SGLT2 inhibitors on major clinical events and safety outcomes in heart failure patients: a meta-analysis of randomized clinical trials. J Geriatr Cardiol 2021;18:783–795. 37. [37].Paolisso P, Bergamaschi L, Santulli G, Gallinoro E, Cesaro A, Gragnano F, et al. Infarct size, inflammatory burden, and admission hyperglycemia in diabetic patients with acute myocardial infarction treated with SGLT2-inhibitors: a multicenter international registry. Cardiovasc Diabetol 2022;21:77. 38. [38].Yip TC, Lee HW, Chan WK, Wong GL, Wong VW. Asian perspective on NAFLD-associated HCC. J Hepatol 2022;76:726–734. 39. [39].Wei Q, Xu X, Guo L, Li J, Li L. Effect of SGLT2 Inhibitors on Type 2 Diabetes Mellitus With Non-Alcoholic Fatty Liver Disease: A Meta-Analysis of Randomized Controlled Trials. Front Endocrinol (Lausanne) 2021;12:635556. 40. [40].Akuta N, Kawamura Y, Fujiyama S, Saito S, Muraishi N, Sezaki H, et al. Favorable impact of long-term SGLT2 inhibitor for NAFLD complicated by diabetes mellitus: A 5-year follow-up study. Hepatol Commun 2022. 41. [41].Arai T, Atsukawa M, Tsubota A, Mikami S, Ono H, Kawano T, et al. Effect of sodium-glucose cotransporter 2 inhibitor in patients with non-alcoholic fatty liver disease and type 2 diabetes mellitus: a propensity score-matched analysis of real-world data. Ther Adv Endocrinol Metab 2021;12:20420188211000243. 42. [42].Saffo S, Kaplan DE, Mahmud N, Serper M, John BV, Ross JS, et al. Impact of SGLT2 inhibitors in comparison with DPP4 inhibitors on ascites and death in veterans with cirrhosis on metformin. Diabetes Obes Metab 2021;23:2402–2408. 43. [43].Ng CH, Lin SY, Chin YH, Lee MH, Syn N, Goh XL, et al. Antidiabetic Medications for Type 2 Diabetics with Nonalcoholic Fatty Liver Disease: Evidence From a Network Meta-Analysis of Randomized Controlled Trials. Endocr Pract 2022;28:223–230. 44. [44].Hollande C, Boussier J, Mottez E, Bondet V, Buivan TP, Charbit B, et al. Safety of sitagliptin in treatment of hepatocellular carcinoma in chronic liver disease patients. Liver Cancer International 2021;2:73–81. 45. [45].Hattori S, Nomoto K, Suzuki T, Hayashi S. Beneficial effect of omarigliptin on diabetic patients with non-alcoholic fatty liver disease/non-alcoholic steatohepatitis. Diabetology & Metabolic Syndrome 2021;13:28. 46. [46].Fu ZD, Cai XL, Yang WJ, Zhao MM, Li R, Li YF. Novel glucose-lowering drugs for non-alcoholic fatty liver disease. World J Diabetes 2021;12:84–97. 47. [47].Dos Santos LR, Duarte ML, Peccin MS, Gagliardi ART, Melnik T. Dipeptidyl Peptidase IV Inhibitors for Nonalcoholic Fatty Liver Disease - Systematic Review and Metanalysis. Curr Diabetes Rev 2021;17:e101120187811. 48. [48].Dougherty JA, Guirguis E, Thornby KA. A Systematic Review of Newer Antidiabetic Agents in the Treatment of Nonalcoholic Fatty Liver Disease. Ann Pharmacother 2021;55:65–79. 49. [49].Choi HSJ, Brouwer WP, Zanjir WMR, de Man RA, Feld JJ, Hansen BE, et al. Nonalcoholic Steatohepatitis Is Associated With Liver-Related Outcomes and All-Cause Mortality in Chronic Hepatitis B. Hepatology 2020;71:539–548. 50. [50].Itou M, Kawaguchi T, Taniguchi E, Sata M. Dipeptidyl peptidase-4: a key player in chronic liver disease. World J Gastroenterol 2013;19:2298–2306. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.3748/wjg.v19.i15.2298&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=23613622&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2022%2F08%2F17%2F2022.08.16.22278847.atom) [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000318592000002&link_type=ISI) 51. [51].Jensen-Urstad AP, Semenkovich CF. Fatty acid synthase and liver triglyceride metabolism: housekeeper or messenger? Biochim Biophys Acta 2012;1821:747–753. [CrossRef](http://medrxiv.org/lookup/external-ref?access_num=10.1016/j.bbalip.2011.09.017&link_type=DOI) [PubMed](http://medrxiv.org/lookup/external-ref?access_num=22009142&link_type=MED&atom=%2Fmedrxiv%2Fearly%2F2022%2F08%2F17%2F2022.08.16.22278847.atom) [Web of Science](http://medrxiv.org/lookup/external-ref?access_num=000303028400005&link_type=ISI) 52. [52].Jojima T, Wakamatsu S, Kase M, Iijima T, Maejima Y, Shimomura K, et al. The SGLT2 Inhibitor Canagliflozin Prevents Carcinogenesis in a Mouse Model of Diabetes and Non-Alcoholic Steatohepatitis-Related Hepatocarcinogenesis: Association with SGLT2 Expression in Hepatocellular Carcinoma. Int J Mol Sci 2019;20. 53. [53].Ganapathy V, Thangaraju M, Prasad PD. Nutrient transporters in cancer: Relevance to Warburg hypothesis and beyond. Pharmacology & Therapeutics 2009;121:29–40. 54. [54].Valdecantos MP, Pardo V, Ruiz L, Castro-Sánchez L, Lanzón B, Fernández-Millán E, et al. A novel glucagon-like peptide 1/glucagon receptor dual agonist improves steatohepatitis and liver regeneration in mice. Hepatology 2017;65:950–968. 55. [55].Nishina S, Yamauchi A, Kawaguchi T, Kaku K, Goto M, Sasaki K, et al. Dipeptidyl Peptidase 4 Inhibitors Reduce Hepatocellular Carcinoma by Activating Lymphocyte Chemotaxis in Mice. Cell Mol Gastroenterol Hepatol 2019;7:115–134. 56. [56].Kawaguchi T, Nakano D, Koga H, Torimura T. Effects of a DPP4 Inhibitor on Progression of NASH-related HCC and the p62/ Keap1/Nrf2-Pentose Phosphate Pathway in a Mouse Model. Liver Cancer 2019;8:359–372. 57. [57].De Souza A, Irfan K, Masud F, Saif MW. Diabetes Type 2 and Pancreatic Cancer: A History Unfolding. JOP 2016;17:144–148.