Abstract
Background/Aim: This study examined whether metformin can enhance the radiation response in a hepatocellular carcinoma (HCC) xenograft mice model and patient population. Materials and Methods: Huh-7 human HCC-bearing xenograft mice were treated with gamma-ray, metformin, neutron therapy, and their combinations. Tumour growth and lung colonies were assessed. Overall, 145 patients who underwent radiotherapy for HCC were retrospectively analysed. Results: The combinations of gamma-ray and metformin and neutron radiation and metformin inhibited tumour growth and metastatic lung nodule formation when compared to the monotherapy and gamma-ray groups, respectively. In patients who received radiotherapy for HCC, the overall survival rate was higher in the metformin-treated group than in the non-metformin group. Conclusion: Metformin inhibited tumour growth and metastasis in HCC by enhancing the radiation response in animal experiments. Additionally, metformin was also found to be associated with a higher survival outcome in patients with HCC.
Primary liver cancer, including hepatocellular carcinoma (HCC), was estimated as the seventh most commonly diagnosed cancer worldwide; however, it was the third most common cause of cancer death in 2020 (1). Surgical resection, liver transplantation, and radiofrequency ablation (RFA) form the basis of curative treatment for HCC; however, these methods cannot be used in most cases due to advanced stage, poor liver function, comorbidities, or other technical problems (2). Clinical studies have reported favourable outcomes with precise radiation therapy (RT) techniques, such as stereotactic body RT and particle therapy (3-6). However, the prognosis for HCC remains poor owing to frequent recurrence or post-treatment progression (7).
Metformin (1,1-dimethylbiguanide hydrochloride) is the first-line pharmacological treatment for type 2 diabetes and has been shown to be clinically associated with potential antitumor effects (8-10). Recent pre-clinical and pharmacoepidemiological cohort studies have reported improved survival with metformin in HCC (11-14). Furthermore, metformin combined with chemotherapy or RT provides a synergistic benefit against HCC (15-18).
New systemic agents, including targeted agents and immune checkpoint inhibitors, have improved treatment outcomes in HCC (19-22). However, these new agents are expensive and have unsatisfactory therapeutic effects. Furthermore, no agent has been proven effective in reducing recurrence or metastasis in patients with HCC at potentially curative tumour stages (23, 24). Adjuvant treatment options are required to reduce recurrence and improve survival after curative local therapies.
Therefore, we hypothesized that metformin use and treatment outcome for HCC might be associated. We conducted a preclinical study to evaluate the radiation response in an HCC xenograft model after metformin administration, and a retrospective cohort study to validate the effect of metformin on clinical outcomes of HCC after RT. We also assessed the difference in the radiation-enhancement effect of metformin using both high linear energy transfer (LET) radiation (neutron) and lower LET radiation (gamma-ray).
Materials and Methods
Mice and tumour. Balb/c nude (nu/nu) mice (male, 5 weeks old) were purchased from Orient Bio Co. (Seongnam, Korea) and allowed to acclimatize to the new environment for 1 week before the experiments. The housing temperature and relative humidity were maintained at 22±3°C and 50±20%, respectively. The mice were provided with water and food (Purina) ad libitum.
Huh-7 human hepatocellular carcinoma cells were used. Tumour cells were maintained in vitro in Dulbecco’s modified Eagle’s medium supplemented with 10% fetal bovine serum and 1% antibiotics (5 mg penicillin, 5 mg streptomycin, and 10 mg neomycin/ml; Life Technologies, Inc., Carlsbad, CA, USA) in a humidified environment at 37°C with 5% CO2. Exponentially growing huh-7 tumour cells in culture were harvested using trypsin treatment and washed three times with phosphate-buffered saline (PBS). Approximately 1×106 viable cells suspended in 0.05 ml PBS were injected subcutaneously into the right hind leg of male BALB/c nude mice. Five mice were used in each experimental group. Tumour diameters were measured with a calliper, and tumour volumes were calculated using the formula: V=0.523×A×B2, where A represents the longest and B represents the shortest diameter of the tumour. Tumours were used for experiments when they attained a size of 6-7 mm in diameter. All animal experiments were conducted following the protocol approved by the Institutional Animal Care and Use Committee (KIRAMS 2015-0024).
In vivo study using low LET radiation.
Tumour growth delay. Tumour-bearing mice were randomly divided into four groups with 10 mice per group as follows: Control, gamma-ray (RT), metformin (MET), and gamma-ray plus metformin (RT-MET). The mice were lightly anesthetized with tiletamine/zolazepam (Virbac Zoletil™ 50; Virbac Lab., Carros, France), and the tumour-bearing legs were locally irradiated with 15 Gy using a 60Co irradiation unit (Thermatron 780; Atomic Energy of Canada, Ontario, Canada) at a dose rate of 1.3 Gy/min. The MET and RT-MET groups were intraperitoneally injected daily with metformin for 2 weeks (150 mg/kg; ALX-270-432; Enzo Life Sciences, Inc., Farmingdale, NY, USA).
Lung colony assay. Lung metastases developed from the thigh Huh-7 tumours in mice 45-60 days after treatment (25). The anti-metastatic potential of the four treatment strategies was tested using the spontaneous lung metastasis model. When the tumour reached a diameter of 8 mm, mice were randomly assigned to one of four groups, as described above for the tumour growth delay study. Their lungs were harvested at days 45 and 60 after treatment and fixed with Bouin’s solution for counting of the lung nodules under a polarizing microscope (magnification, ×4).
Immunohistochemistry (IHC). At the end of life, the tumours were excised and placed in neutral-buffered formalin, embedded in paraffin, and cut into 4-μm-thick sections. Tissue sections were mounted on silane-coated slides, deparaffinized, and boiled for 10 min in 0.01 M citrate buffer (pH 6.0) for antigen retrieval. The sections were incubated overnight at 4°C in solutions containing appropriate concentrations of antibodies against vascular endothelial growth factor (VEGF) (ab1316; Abcam, San Francisco, CA, USA) and matrix metalloproteinase 9 (MMP9) (ab228402; Abcam). After washing three times with PBS, the sections were incubated with peroxidase reagent and anti-mouse IgG (ImmPRESS™; Vector Laboratories, Inc., Burlingame, CA, USA) for 20 min. Following three washes with PBS, peroxidase-binding sites were stained with diaminobenzidine (Vector Laboratories, Inc.), counterstained with Mayer’s hematoxylin, and examined under a light microscope. The percentages of VEGF-positive and MMP9-positive areas were quantitatively assessed.
In vivo study using high LET radiation.
Tumour growth delay. Tumour-bearing mice were randomly divided into six groups with 10 mice per group as follows: Control, MET, RT, RT-MET, neutron (NT), and neutron plus metformin (NT-MET). The mice were lightly anesthetized with tiletamine/zolazepam (Virbac Zoletil™ 50; Virbac Lab., Carros, France), and the tumour-bearing legs were locally irradiated with 5 Gy at a dose rate of 1.3 Gy/min and 5 Gy using fast neutrons [9.8 MeV, 30-40 keV/μm, produced by the bombardment of beryllium by proton 9Be(p,n)10B, cyclotron (MC-50; Scanditronix, Uppsala, Sweden]. The mice of the MET, RT-MET, and NT-MET groups were intraperitoneally injected daily with metformin for 2 weeks (150 mg/kg).
Lung colony assay. Lung metastases developed from the thigh Huh-7 tumours in mice approximately 35 days post-treatment (20). The anti-metastatic potential of the six treatment strategies was tested using the spontaneous lung metastasis model. When the tumour reached a diameter of 8 mm, mice were randomly assigned to one of the aforementioned six groups. Their lungs were harvested at day 35 post-treatment and fixed with Bouin’s solution for counting of the lung nodules under a polarizing microscope.
Clinical retrospective study.
Patient population. We retrospectively analysed the medical records of 145 patients with HCC who underwent RT for inoperable HCC at a single comprehensive cancer care centre between March 2003 and March 2016. Treatment strategies for these patients were decided through a multidisciplinary liver tumour conference. The indications for RT were unresectable and inoperable tumour, unsuitable for local ablative therapies, an incomplete response to transarterial chemoembolization, tumour confined to the liver without extrahepatic metastasis, and tolerable liver function. This study was approved by the Institutional Review Board (KIRAMS 2021-12-005). All patients provided written informed consent after being informed of the possible benefits and complications of RT. Of the 145 patients with HCC, 41 (28%) had type 2 diabetes and 19 (13%) received metformin during RT. The patients were divided into the metformin group (n=19) and non-metformin group (n=126).
Radiotherapy. The RT technique used for liver cancer has been described previously (3, 26). RT was delivered using a RapidArc (Varian Medical Systems, Palo Alto, CA, USA) or CyberKnife (Accuray Inc., Sunnyvale, CA, USA) machine. Breathing-related tumour motion was controlled by abdominal compression, gating, or tracking techniques. Imaging guidance by orthogonal radiography or on-board computed tomography was used to target the lesion before each treatment session. The total RT doses of 25 to 60 Gy (median, 51 Gy) were delivered in 3 to 6 fractions, and the median biologically effective dose at an α/β of 10 (BED10) was 137.7 Gy (range=45.8-180.0 Gy).
Statistical analysis. In the animal experiments, tumour volumes, numbers of lung nodules, and IHC-positive areas were expressed as means±standard deviations. Student’s t-test was used for comparisons between two groups of mice. A two-way analysis of variance followed by a post hoc test was performed to evaluate the differences in the tumour volumes over time between the treatment groups.
In the clinical retrospective study, the outcomes included overall survival (OS) and progression-free survival (PFS). The treatment response was evaluated using the modified Response Evaluation Criteria in Solid Tumors (27). PFS was defined as living status without any post-treatment progression or metastasis. Data were analysed using a closeout date of March 5, 2020. The OS and PFS were calculated from the start date of RT to the date of death from any cause and until tumour progression at any site or death, respectively. Pearson chi-square or Fisher’s exact test was used to compare the clinical characteristics. Survival rates were calculated using the Kaplan–Meier method, and intergroup comparisons were performed using the log-rank test. All statistical analyses were performed using IBM SPSS Statistics for Windows, version 23.0 (IBM Corp., Armonk, NY, USA). All tests were two-sided, and p-values of 0.05 or less were considered statistically significant.
Results
In vivo study using γ-irradiation. After 26 days of irradiation, the tumour sizes in the control, RT, MET, RT-MET groups were approximately 2,816 mm3, 1,178 mm3, 1,600 mm3, and 535 mm3, respectively. The tumour sizes in the RT, MET, and RT-MET groups had decreased to 41%, 57%, and 13%, respectively (p<0.05), which indicates a radiosensitizing effect of metformin. RT-MET treatment led to a greater reduction in tumour growth (55% and 67%, p<0.05) as compared to the RT and MET groups (Figure 1). At 45 and 60 days after irradiation, the RT, MET, and RT-MET groups showed reduction in the metastatic lung nodules as compared to the control group (p<0.05). After 60 days, the RT-MET group showed greater reduction in the number of metastatic nodules (93% and 76%, p<0.05) as compared to the RT and MET groups (Figure 2). RT-MET significantly reduced VEGF and MMP9 expression compared with that in the control and RT-only group (p<0.05) (Figure 3).
Treatment with radiation or metformin inhibits tumour growth in mice with Huh-7 tumour xenografts. A: Tumour-bearing mice and the removed xenograft tumours in the four experimental groups as follows: (i) control, (ii) radiation, (iii) metformin, and (iv) radiation plus metformin. B: Tumour volumes according to time after treatment in the four groups. Significantly different at p<0.05 from the *control and radiation alone/metformin alone groups, and #control.
Inhibition of metastasis in Huh-7 tumour xenografts. A: Lungs harvested from mice at days 45 or 60 after treatment. B: Number of lung nodules in the four treatment groups. Significantly different at p<0.05 from the *control and radiation-alone/metformin-alone groups at 60 days; †control and radiation alone at days 45 and 60; and #control, at days 45 and 60 after treatment.
Combination of radiation and metformin treatment of mice with Huh-7 hepatocellular carcinoma xenografts reduced the expression of vascular endothelial growth factor (VEGF) (A) and matrix metalloproteinases 9 (MMP9) (B). *Significantly different at p<0.05 from the control and radiation-alone groups.
In vivo study using neutron irradiation. After 25 days of irradiation, the tumour sizes in the control, RT, MET, RT+MET, NT, and NT+MET groups were approximately 2,674, 1,526, 1,844, 1,185, 1,573 and 808 mm3, respectively. After 35 days, the tumour sizes had decreased to approximately 54%, 68%, 45%, 57%, and 31%, respectively (p<0.05), indicative of a radiosensitizing effect of metformin. NT-MET led to greater inhibition of tumour growth (42%, 54%, 31%, and 45%) as compared to the RT, MET, RT-MET, and NT groups (p<0.05) (Figure 4A). After 35 days of irradiation, the number of metastatic lung nodules were reduced in the MET, RT-MET, NT, and NT-MET groups as compared to the control group (p<0.05). NT-MET led to greater reduction in the number of metastases (80%, 56%, 50%, and 54%) as compared to the RT, MET, RT-MET, and NT groups, respectively (p<0.05) (Figure 4B).
Inhibition of growth and metastasis in Huh-7 tumour xenografts. A: Tumour volumes according to time after treatment in the six treatment groups. B: Lungs harvested from mice at day 35 after treatment. C: Number of lung nodules in mice from the six treatment groups. The combination of neutron radiation and metformin significantly reduced the tumour volumes and lung metastases (p<0.05) as compared to the groups treated with gamma-rays alone, metformin alone, combination of gamma-rays and metformin, or neutron radiation alone.
Clinical retrospective study of patients with HCC treated with RT. Nineteen patients had been receiving metformin for at least 1 year after RT, and the median daily dose of metformin was 1,000 mg (range=500-2,000 mg). Patient and tumour characteristics of the entire cohort are shown in Table I. There was no significant difference in baseline characteristics between the metformin and non-metformin groups.
Patient and tumor characteristics.
The median follow-up duration after RT for the entire cohort was 31 months (range=3-135 months). The 5-year OS and PFS for patients overall were 45% and 24%, respectively. Six (32%) patients in the metformin group and 74 (59%) in the non-metformin group had died by the time of the last follow-up. The survival analysis indicated that the OS rate in the metformin group was significantly higher than that in the non-metformin group (5-year OS: 68% vs. 41%, p=0.034; Figure 5A). Univariate analysis identified Eastern Cooperative Oncology Group performance status, BED, tumour size, and serum alpha-fetoprotein level as significant prognostic factors for OS (Table II). Twelve (63%) patients in the metformin group and 94 (75%) in the non-metformin group had experienced disease progression at the last follow-up. The survival analysis indicated that the PFS rate of the metformin group was higher than that of the non-metformin group (5-year PFS: 34% vs. 22%, p=0.092; Figure 5B). Univariate analysis identified portal vein tumour thrombus and BED as significant prognostic factors for PFS (Table II).
Kaplan–Meier curves for overall (A) and progression-free (B) survival according to metformin use in patients with hepatocellular carcinoma (p-values by log-rank test).
Prognostic factors.
Discussion
We found that metformin significantly inhibited the growth and metastasis of HCC in vivo. Metformin, in conjunction with RT, enhanced the effect of radiation when compared to RT or metformin alone. The radiosensitizing effect of metformin was shown in experiments using high LET (neutrons) and lower LET (gamma-rays). Moreover, metformin significantly improved survival outcomes in patients with HCC who received RT. These findings suggest that metformin can serve as an adjuvant treatment option in the management of patients with HCC.
RT has been increasingly used in recent years for patients with HCC who are unsuitable for surgical resection, liver transplantation, or RFA (28). RT techniques have evolved with improved target delineation, treatment delivery, dose escalation, and biological effects, while sparing normal tissues. Many clinical studies have reported favourable outcomes in HCC with advanced RT techniques. A phase II multi-centre study of stereotactic body RT reported a 3-year local control rate of 95% and an OS rate of 76% in patients with unresectable HCC (3). A multinational study showed that stereotactic body RT provided better local control than RFA, with comparable toxicities for patients with inoperable HCC (4). A phase III randomized controlled trial demonstrated that proton therapy is comparable to RFA in patients with recurrent HCC (5). Japanese phase II trials reported that carbon-ion RT is safe and effective, even for patients with recurrent or locally advanced HCC (6). The majority of clinical practice guidelines have suggested a potential role of RT in the management of HCC (28-32). However, disease progression in areas other than the treated lesions remains a challenge, necessitating combination treatment strategies involving RT and systemic agents (7).
Metformin has a robust safety profile and low cost and the associated data on its long-term use are available (8). Metformin can inhibit development of and improve the survival outcomes in cancer (8-13). Several clinical studies assessing the effectiveness of metformin in various cancer types are ongoing [summarised in (33)]. Some preclinical and clinical studies have reported similar anticancer effects of metformin and synergistic benefits when used in combination with chemotherapy or RT in HCC (11-17, 34-36). Metformin effectively enhances the therapeutic effect of radiation with a wide range of LETs (17). Our findings from animal experiments and clinical cohort analyses are consistent with previous reported findings, and a combination of RT as the local modality and metformin as a systemic treatment may be an effective therapeutic option.
The mechanism underlying the anticancer or radiosensitizing effect of metformin in HCC is unclear. Pre-clinical data indicate a complex mechanism majorly related to mitochondrial complex I and adenosine monophosphate-activated protein kinase (17, 33). VEGF-mediated angiogenesis has also been proposed as a potential mechanism underlying the effect of metformin, and serum VEGF levels are associated with treatment outcomes in HCC (37, 38). Furthermore, metformin was shown to suppress migration and invasion through the regulation of MMP9 in cancer cells (39). In the present study, IHC staining demonstrated reduced VEGF and MMP9 expression when metformin was used in combination with RT. Metformin has been reported to enhance radiosensitivity in HCC by inhibiting the epithelial-to-mesenchymal transition (16). Here, we observed that metformin not only inhibited tumour growth but also reduced distant metastasis. Our pre-clinical data suggest that the epithelial– mesenchymal transition may be involved in this effect of metformin.
Our study has a few limitations. We did not present a clear mechanism for the radiosensitizing effect of metformin owing to the limitations of animal experiments. However, our research group has proposed adenosine monophosphate-activated protein kinase activation as a potential mechanism associated with radiosensitizers in previous in vitro experiments (17, 34), and we showed the association of metformin with angiogenesis, migration, and invasion processes through IHC staining analysis. Furthermore, the metformin doses used in animals and humans may not be the same. If the dose used in animal experiments is within the range of that commonly prescribed for humans, then metformin can be safe and effective for adjuvant treatment. Furthermore, this was a retrospective study with a limited sample size. The number of patients taking metformin was relatively small, making it difficult to obtain statistical significance. Nevertheless, this study was conducted on a relatively homogeneous group of patients with inoperable HCC, no extrahepatic metastases, and tolerable liver function undergoing curative RT. Additionally, the effects of metformin dose and duration were not sufficiently considered. Metformin is associated with a dose-dependent increase in response to chemoradiotherapy in patients with oesophageal cancer (40). Here, the dosage and duration of administration varied among patients, which might have affected the outcome. However, this might have been corrected for to some extent as all patients had been taking metformin for at least 1 year after RT, and the median daily dose of metformin was 1,000 mg. Lastly, the non-metformin group included both non-diabetic and diabetic patients taking metformin. Diabetes is associated with poorer survival outcomes in patients with HCC (41), and this may affect the comparative analysis of the effect of metformin between non-diabetic patients with better prognosis and diabetic patients with poor prognosis. Nevertheless, we found that metformin usage was significantly associated with the survival outcome of patients treated with RT for HCC.
In conclusion, combination therapy with radiation and metformin inhibited the growth and metastasis of HCC by enhancing the radiation response in vivo. The radiosensitizing effect of metformin was much higher in the neutron-irradiated than in the gamma-ray-irradiated animal HCC model. Additionally, the use of metformin was associated with improved survival outcomes in patients with HCC receiving RT. Our data support the clinical utility of metformin in combination with RT, especially as an adjuvant treatment option after curative local therapies. Although many systemic agents have been developed for the treatment of HCC, none of them has been approved by the Food and Drug Administration for clinical use, and metformin may be a safe, effective, and less cost-intensive adjuvant treatment strategy. Our findings provide theoretical support for future trials and well-designed prospective studies conducted to confirm the effect of metformin in patients with HCC.
Acknowledgements
This work was supported by the Korea Institute of Radiological and Medical Sciences (KIRAMS) grant (No. 50571-2021). This article has been edited to ensure that the language is clear and free of errors by professional editors at Editage, a division of Cactus Communications. This study was presented as a poster at the American Society for Radiation Oncology (ASTRO) 59th Annual Meeting, September 24-27, 2017, San Diego, USA. and at the 1st Annual Conference of the Asia-Oceania Particle Therapy Cooperative Group (PTCOG-AO), December 6-8, 2019, Osaka, Japan.
Footnotes
Authors’ Contributions
WIJ contributed to study conception and design; literature research; data acquisition, analysis, and interpretation; statistical analysis; article preparation and editing. WWK contributed to study design; literature research; animal experiments; data acquisition, analysis, and interpretation; article preparation and article editing. JHJ contributed to the study design; literature research; data interpretation; and article review. JSK contributed to the literature research; and article preparation and review. MSK contributed to study conception and design; literature research; data analysis and interpretation; article preparation and review. All Authors made substantial contributions to drafting and revising the article and final approval of the version to be submitted.
Conflicts of Interest
The Authors declare that they have no competing interests.
- Received November 2, 2021.
- Revision received December 24, 2021.
- Accepted January 11, 2022.
- Copyright © 2022 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
