Metformin Use and Survival in Patients with Advanced Extrahepatic Cholangiocarcinoma: A Single-Center Cohort Study in Fuyang, China

Aims Metformin is an oral antidiabetic agent that has been widely prescribed for the treatment of type II diabetes. In recent years, anticancer properties of metformin have been revealed for numerous human malignancies. However, there are few indications available regarding the feasibility and safety of these studies in an advanced extrahepatic cholangiocarcinoma (EHCC) population. This study is aimed at evaluating the feasibility, safety, and value of metformin use and survival in patients with advanced EHCC. Methods All patients with advanced EHCC observed at Fuyang People's Hospital between January 2015 and November 2020 were included in the study. Case data, clinical information, and imaging results were abstracted from the self-administered questionnaire and electronic medical record. All patients were divided into study subjects and control subjects, and the study subjects were given metformin, 0.5 g, three times a day, while control subjects were without metformin. The metformin use and survival time of the subjects were asked by telephone, out-patient, or door-to-door visit, after they left the hospital. Results One hundred and thirty-three study cases and 589 controls were included in the analysis. This study showed that metformin use cannot improve the overall survival rate of patients with advanced EHCC ([95% CI]: -17.05-0.375, t = −1.889, P value = 0.061), but the survival time of patients with drainage treatment from control group (n = 496) was significantly shorter than that of patients with drainage treatment from the study group (n = 113), and the difference was statistically significant (z = −2.230, P value = 0.026). There were significant differences between metformin used before or after the diagnosis of advanced EHCC (OR[95% CI], 3.432[2.617-4.502]; P value = 0.001) in survival time. And there was significant difference between the duration of metformin use and survival prognosis (OR[95% CI], 2.967[1.383-6.368]; P = 0.005). Conclusion Metformin can improve the survival of advanced EHCC patients who underwent drainage treatment, especially for metformin use after diagnosis of advanced EHCC and long duration of metformin.


Introduction
Cholangiocarcinoma (CCA) is categorized as an distal cholangiocarcinoma, perihilar, and intrahepatic [1]. CCA clinical presentation depends on the anatomic location and macroscopic growth pattern [1]. Early CCA often has no special clinical symptoms. Perihilar and distal CCA often have jaundice, which gradually deepens with time; light stool color; dark yellow urine color; and skin pruritus. Intrahepatic CCA is dominated by nonspecific symptoms like abdominal pain, fatigue, weight loss, night sweats, and cachexia; however, cirrhotic patients can be asymptomatic [1]. Histologi-cally, 90% of CCA are adenocarcinomas, while known variants contain signet-ring type, papillary adenocarcinoma, squamous cell carcinoma, clear cell type, oat cell carcinoma, intestinal type adenocarcinoma, and adenosquamous carcinoma [1]. Surgery is the main curative method, whereas stent implantation by endoscopic retrograde cholangiopancreatography, percutaneous transhepatic cholangial drainage, systemic chemotherapy, and radiofrequency ablation are a curative option for advanced CCA [1,2].CCA is a malignant tumor originating in the biliary tree, and it is the second most common cancer after hepatocellular carcinoma [2,3]. The Bertuccio et al. study showed that mortality from extrahepatic (ECC) levelled off or decreased [4], while the EHCC appears to be one of the most rapidly increasing tumors in China [5,6]. Although some treatment approaches were developed as therapeutics for EHCC, the prognosis of patients with unresectable or advanced EHCC is poor [7,8]. More than 50% of cases with jaundice are inoperable at the time of first diagnosis [7].
Metformin is an oral antidiabetic agent that has been widespread prescribed for treatment of type II diabetes [9,10]. This drug lowers hyperglycemia through the inhibition of hepatic glucose production. Compared to normal cells, cancer cells preferentially metabolize glucose to lactate, even in aerobic conditions. Such metabolic alterations not only promote the growth and invasion of tumor cells but also support their chemoresistance [11]. Kaewpitoon et al. suggested that metformin might influence tumorigenesis, both indirectly, through the systemic reduction of insulin levels, and directly, via the induction of energetic stress [12]. A recent epidemiologic survey indicated that metformin use was associated with reduced tumor incidence in patients with type II diabetes [13][14][15][16]. The anticarcinogenic activity of metformin has been attributed to many mechanisms, including the activation of the liver kinase B1 (LKB1)/AMP-activated protein kinase (AMPK) pathway, inhibition of the unfolded protein response, inhibition of protein synthesis, induction of cell cycle arrest and/or apoptosis, activation of the immune system, and potential eradication of cancer stem cells [12,15,17]. LKB1/AMPK pathway activation inhibits the mammalian target of rapamycin (mTOR), which negatively affects protein synthesis in tumor cells [15]. Several in vivo and in vitro studies have demonstrated metformin to inhibit the proliferation of various cancer cell types, including gastric, esophageal, breast, prostate, hepatocellular carcinoma, and colon cancer cells [12]. Anticancer properties of metformin have been revealed for numerous human malignancies including CCA with antiproliferative effects in vitro [3]. Moreover, some studies further found that metformin effectively sensitized CCA cells based on certain chemotherapies [18].
In this study, the effects of metformin on survival and prognosis of patients with advanced EHCC were assessed.

Study Population.
All patients with advanced EHCC observed at Fuyang People's Hospital between January 2015 and November 2020 were included in the study. We searched for EHCC cases using an imaging diagnosis system and electronic medical record system. The diagnosis of advanced EHCC was confirmed by abdominal enhanced CT, abdominal enhanced MR, and PET-CT; all patients were confirmed to transfer (blood vessel, nerve, lymph gland and organ tissue). After screening, 722 patients with confirmed advanced EHCC were included in the analysis. They were divided into study subjects and control subjects, and the study subjects (n = 133) agreed to participate in the study and take metformin, 0.5 g, three times each day, which was produced by Guizhou Shengjitang Pharmaceutical Co., Ltd. Control subjects (n = 589) were without metformin. Cases were matched by sex, age, ethnicity, and residence to subjects who enrolled in the Fuyang People's Hospital between January 2015 and November 2020.This study was approved by the ethics committee of the Fuyang People's Hospital, China.

Clinical
Information. Case data, clinical information, and imaging results were abstracted from the selfadministered questionnaire and electronic medical record. Relevant factors abstracted included diabetes, history of liver disease (HBV or HCV infection), tumor location, radiotherapy, other drugs that may affect tumor (aspirin, regulating immunity, antitumor, and Chinese herbal medicine, etc.), drainage treatment, family history of tumor, drinking history, and smoking history.
We collected the results of tests for HBV and HCV infection for all subjects. HBV infection was defined as a positive hepatitis B surface antigen. And HCV infection was defined as positive HCV antibody or HCV RNA. We abstracted the results of tests for total bilirubin (TBil), transaminase (ALT and AST), bile duct enzyme (ALP and γ-GT), tumor index (AFP, CA199, and Hsp90α), jaundice, and ascites for all cases.
Previous or current use of metformin was ascertained from the questionnaire and physician's notes. And duration of metformin use was ascertained from the follow-up of patients or their families.
The subjects were given metformin, and the medication was recorded in detail; basing on the diagnosis of advanced EHCC was taken as the starting point of observation. The metformin use and survival time of the subjects were asked by telephone, out-patient, or door-to-door visit, after they leave the hospital. Six-month observation was regarded as the end point of the event, and the death was recorded as 0 and the survival was recorded as 1.

Statistical Analysis
t-test was used to compare the data of normal distribution among groups, and mean + SDwas used for expression. Mann-Whitney U test was used to analyze the difference between groups for nonnormal distribution data, and median (IQR) was used for skew distribution data. Pearson chi-square test (χ 2 ) was used to analyze the differences between the groups. Odds ratios (OR) and 95% confidence intervals (95% CI) were used to estimate the relationship between metformin use and survival prognosis.
Kaplan-Meier survival analysis, log-rank tests, and Breslow were performed to analyze the overall survival (OS) of subjects. Box plot was used to compare and analyze the differences of different groups. Statistical analyses were done using SPSS software, version 20.0.All tests were two-sided or Fisher's exact test, with P < 0:05 defined as statistically significant.

Patient Characteristics.
One hundred and thirty-three study cases and 589 controls were included in the analysis. 2 Gastroenterology Research and Practice

Gastroenterology Research and Practice
Tables 1 and 2 summarize the baseline characteristics and risk factors for the study and control groups that may affect the survival and prognosis. Demographics were comparable between groups. Trend analysis showed that there were no differences in baseline characteristics (P value > 0.05). And trend analysis showed that there were no differences in laboratory results (P value > 0.05).

Study Subject Characteristics.
In the study group, 6 cases were treated with metformin before EHCC because of diabetes, while 127 cases were given metformin after EHCC. In the study group (n = 133), there was no significant difference between the age when metformin was started and survival time ( (Figure 1). As expected, the survival time of patients with drainage treatment from the control group (n = 496) was significantly shorter than that of patients with drainage treatment from the study group (n = 113), and the difference was statistically significant (z = −2:230, P value = 0.026) ( Table 4 and Figure 2). There was no significant difference in the survival time between patients without drainage treatment from the study group (n = 20) and patients without drainage treatment from the control group (n = 93) ([95% CI]: -9.012-13.442; P value = 0.697) (Figure 3). Compared to countryside patients, the survival time of town patients from the study group (n = 59) and the control group (n = 291) was significantly longer (101:03 ± 44:94 vs. 132:56 ± 44:59; 100:84 ± 41:27 vs. 112:66 ± 36:96), and the difference was statistically significant (P value < 0.01), Table 4.

The Value of Metformin Use in Feasibility and Safety.
This study showed that the number of patients who actively withdrew from using metformin due to intolerance was only 7 (5.26%), while survival time was shorter (Table 5).

Discussion
EHCC is a highly aggressive epithelial malignancy and usually has a poor prognosis because of the insensitivity to therapies and difficulty in detection [19], particularly for advanced EHCC. The diagnosis of EHCC is very complex and usually requires a combination of clinical symptoms, endoscopic techniques, imaging techniques, and cytopathological tests. In recent years [3], metformin has received growing attention due to its promising anticancer potential observed in many human tumors. A number of epidemiologic studies showed that metformin use in patients with diabetes was associated with a decreased incidence of various cancers, including CCA, gastroenterological cancers, pancre-atic cancer, and breast cancer [16]. To our knowledge, this is the first time to study the relationship between metformin use and the survival in advanced EHCC. This study showed that metformin use cannot improve the overall survival rate of patients with advanced EHCC ([95% CI]: -17.05-0.375; t = −1:889, P value = 0.061, Figure 1), but the survival time of patients with drainage treatment using metformin was significantly longer than that of patients without metformin (z = −2:230, P value = 0.026). In recent decades, study showed that treatment with the antidiabetic drug metformin has been recently associated with decreased incidence of intrahepatic CCA. Metformin reverts the mesenchymal and epithelial-to-mesenchymal transition (EMT) traits in intrahepatic CCA by activating AMPK-FOXO3-related pathways suggesting it might have therapeutic implications [20]. Metformin treatment reverses EMT and downregulates the proteolytic enzyme matrix metalloproteinase (MMP-2), resulting in suppression of CCA cell migration and invasion. Some studies [10,20] demonstrated that metformin exerted antitumoral effects by (1) inhibiting adenosine deaminase that converts AMP into IMP, resulting in AMP accumulation with a subsequent activation of AMPK; (2) activating AMPK that plays a role in cellular energy homeostasis [21]; (3) blocking the mitochondrial respiratory chain complex (NADH dehydrogenase) that impairs ATP synthesis and increasing the AMP/ATP ratio [15]; and (4) metformin targeting the AMPK/mTORC1 pathway in cholangiocarcinoma cells [9,22]. Trinh et al. and Saengboonmee et al. [3,23] studies showed that metformin exposure significantly reduced cancer cell proliferation, migration, and invasion [9], possibly involving the signal transducers and activators of the transcription 3(STAT3) pathway and nuclear factor-kappa B (NF-κB) pathway and reversal of EMT marker expression. STAT3 plays important roles in cancer development and progression, and its expression was associated with shorter survival of patients with CCA. And they further suggest that metformin may be useful for CCA management.
Complexes of Cdk6 and Cdk4 with cyclin D1 are required for G1 phase progression [13]; however, complexes of Cdk2 with cyclin E are required for the G1 to S transition [21]. Metformin has been demonstrated to downregulate cyclin D1 in various tumor cell lines, including stomach, colon, liver, breast, and prostate cancer lines [12]. The findings shown here indicate that these major cell cycle regulators (Cdk4,cyclin D1, and phosphorylated Rb) may be intracellular targets of the metformin-mediated antiproliferative effect in people CCA cell lines. Metformin has been demonstrated to alter the phosphorylation of many proteins, including c-Src, β-catenin, CREB, Chk2, and Akt, in various cell lines. Fujimori et al. [13] findings indicate that metformin inhibits people CCA cell proliferation and cancer growth, potentially by suppressing cell cycle-related molecules through miRNA alterations. In the present work [15], Zhang et al. demonstrated that metformin treatment profoundly suppressed proliferations of two human CCA cell lines (QBC939 and MZ-CHA-1) in dose-dependent ways. Through comparing metformin-induced changes of metabolite levels between the CCA cells and normal HUVEC cells,

5
Gastroenterology Research and Practice they indicate that metformin profoundly aggravate the Warburg effect and promote glycolysis in CAA cells [11]. In the Tang et al. study, they found that metformin could suppress the Warburg effect in CCA, which promotes oxidative phosphorylation and decreases aerobic glycolysis, thus making CCA cells vulnerable to chemotherapy [11]. Moreover, metformin specifically increases UDP-GlcNAc and BCAAs, indicating the occurrence of autophagy and cell cycle arrest in metformin-treated CAA cells. Ling et al. showed that met-formin sensitizes arsenic trioxide to suppress intrahepatic cholangiocarcinoma through the regulation of AMPK/p38 MAPK-ERK3/mTORC1 pathways [18]. Metformin altered the miRNA (mir124, 182, and 27b; let7b, 221, and 181a) expression to inhibit tumor proliferation [2].
This study showed that the patients who had used metformin before the diagnosis of advanced EHCC could not improve the survival, which may be related to advanced EHCC tolerance or noninsensitivity to metformin (  [9]. Metformin intake after starting chemotherapy can improve the clinical outcome in advanced cholangiocarcinomas [24]. Metformin could change the metabolic status of cancer cells and reverse the Warburg effect via the inhibition of lactate dehydrogenase A(LDHA), which was overexpressed in CCA tissues and indicated a shorter survival time [11]. However, a study showed that [25] the survival of forty-nine patients who continued taking metformin after CCA diagnosis was not different from that of one hundred and sixty-five patients never taking metformin ( Figure 2: The box-plot distribution and comparative analysis of the survival time between patients with drainage treatment from study group (n = 113) and patients with drainage treatment from control group (n = 496) was different (z = −2:230, P value = 0.026). The Kaplan-Meier method was used to estimate survival rate and compare survival curve of patients with drainage treatment from two groups: the curve of control group was below, and the curve of study group was above, which showed that metformin use can improve survival rate. Log-rank tests and Breslow were performed to check; P value was less than 0.05.  6 Gastroenterology Research and Practice use before CCA diagnosis (n = 79) also did not affect survival. So, metformin did not improve the survival of CCA patients with diabetes mellitus. Our study is consistent with that of Yang et al., but the data of the above study is too few, which needs multicenter verification. This study also showed that the survival time of patients living in town is higher than patients living in the countryside, whether metformin is used or not, which may be related to the patients' cultural literacy, attention to the disease, scientific and effective modern medical intervention, and the patients' affordability.
In conclusion, we elucidated that metformin can improve the survival and prognosis of advanced EHCC patients who have undergone drainage treatment. Metformin is an inexpensive drug, and its use has been proven safe without severe adverse effects in people. Thus, our findings show that the use of metformin might be beneficial for advanced EHCC patients who have undergone drainage treatment and might be a potential therapeutic agent for the treatment of EHCC (Table 5).

Conclusion
In conclusion, this research article, for the first time, reports the use of metformin in advanced EHCC patients. The results demonstrate that metformin can improve the survival and prognosis of advanced EHCC patients who have undergone drainage treatment. It is a feasible, practical, and safe therapeutic agent for the treatment of advanced EHCC.

Data Availability
The data used to support the findings of this study are available from the corresponding author upon request.