Correlations between metformin and prognosis and adverse reactions in patients undergoing radical cystectomy followed by adjuvant GC chemotherapy for bladder cancer

Abstract

Objective

The objective of this research was to examine the influence of metformin on both prognosis and adverse reactions in patients who have undergone radical cystectomy (RC) and subsequently received adjuvant gemcitabine and cisplatin (GC) chemotherapy for muscle-invasive bladder cancer (MIBC).

Methods

A retrospective evaluation was performed on data from 243 patients who had undergone RC followed by adjuvant GC chemotherapy at the Affiliated Hospital of Xuzhou Medical University, Xuzhou First People’s Hospital, and Xuzhou Third People’s Hospital during the period from April 2014 to April 2024. The subjects were categorized into three categories based on metformin usage: non-diabetic (No DM), type 2 diabetic with metformin use (DM + Metformin), and type 2 diabetic without metformin use (DM-Metformin). Clinical and pathological characteristics were compiled and subjected to analysis. Progression-free survival (PFS) was assessed utilizing the Kaplan–Meier technique, while Cox proportional hazards models were employed for multivariable analysis.

Results

Among the 243 patients, diabetes was present in 68 individuals, of whom 51 were administered metformin. Pairwise comparisons revealed that diabetic patients receiving metformin exhibited significantly higher 3-year PFS rates compared to both diabetic patients not using metformin (23.6% vs. 7.2%, p = 0.007) and non-diabetic patients (23.6% vs. 13.4%, p = 0.034). In the multivariable analysis, metformin use was independently associated with a reduced risk of disease progression, both in the pooled analysis (Met vs. No Met: HR = 0.66, 95% CI 0.45–0.95, p = 0.025) and, most notably, in the direct comparison among diabetic patients (DM + Met vs. DM-Met: HR = 0.34, 95% CI 0.16–0.75, p = 0.007). Moreover, those administered metformin experienced a lower frequency of grade 3 or higher adverse reactions.

Conclusion

This study provides clinical evidence suggesting that metformin use is associated with enhanced progression-free survival and a reduction in adverse reactions in patients receiving adjuvant GC chemotherapy after RC for MIBC. The beneficial association appears to be particularly pronounced in diabetic patients. These findings support the potential of metformin as an adjunctive therapy and justify further investigation in prospective trials.

Background

Bladder cancer (BLCA) is among the most prevalent malignancies of the urinary system, with a persistently increasing global incidence [1, 2]. Muscle-invasive bladder cancer (MIBC) constitutes 20–25% of cases and is characterized by an aggressive clinical course and high metastatic potential, significantly compromising patient survival [3]. Radical cystectomy (RC) is the standard curative treatment, yet postoperative recurrence remains common, affecting approximately 30% of patients within a median of 12 months [1, 3–9]. Adjuvant cisplatin-based chemotherapy, such as the gemcitabine and cisplatin (GC) regimen, can mitigate this risk, improving survival outcomes [10, 11]. However, the associated toxicities frequently impair patients' quality of life and treatment adherence, underscoring the need for strategies that can enhance efficacy while reducing adverse effects [12–16].

Metformin, a first-line medication for type 2 diabetes, has garnered significant interest for its potential anticancer properties [17]. Preclinical studies have demonstrated that metformin can inhibit tumor cell proliferation and migration, primarily through activation of the AMPK pathway and subsequent inhibition of mTOR signaling [18, 19]. It may also exert indirect antitumor effects via systemic metabolic and anti-inflammatory modulation, which have been associated with improved outcomes in several malignancies [20–22]. Notably, evidence suggests metformin may ameliorate chemotherapy-related toxicity and counteract drug resistance [23]. While the combination of metformin with cisplatin-based chemotherapy appears promising, its specific impact in the context of GC chemotherapy for BLCA remains poorly defined. Therefore, this study aimed to evaluate the effect of metformin on prognosis and adverse reactions in a retrospective cohort of MIBC patients who underwent RC followed by adjuvant GC chemotherapy, aiming to clarify its potential role in postoperative treatment.

Methods

Patient cohort

A retrospective evaluation was performed on a cohort of 243 individuals who underwent RC followed by adjuvant GC chemotherapy at the Affiliated Hospital of Xuzhou Medical University, Xuzhou First People’s Hospital, and Xuzhou Third People’s Hospital during the period from April 2014 to April 2024. The inclusion criteria were: (1) individuals diagnosed with muscle-invasive urothelial carcinoma of the bladder who had undergone RC and/or pelvic lymph node dissection; (2) those who had consistently received a minimum of four cycles of adjuvant GC chemotherapy postoperatively(pT2/3/4 stage with concomitant risk factors such as lymphovascular invasion, positive surgical margins, or variant histology); (3) individuals with an Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1. The exclusion criteria included: (1) individuals with histology other than urothelial carcinoma; (2) individuals with concurrent malignancies; (3) those who had undergone neoadjuvant chemotherapy or radiotherapy; (4) individuals with preoperative imaging that revealed metastatic urothelial carcinoma on chest, abdomen, and pelvic computed tomography (CT) or positron emission tomography-CT. Clinical parameters were reviewed through electronic medical records and prescription databases. The assessment of medication usage, including metformin and other oral hypoglycemic agents or insulin, was conducted prior to surgery.

Patient stratification

The study cohort was stratified in a two-step process. First, all patients were divided into two groups based on metformin exposure: the Metformin group (Met group), comprising diabetic patients using metformin (DM + Met), and the Non-Metformin group (No Met group). It is important to note that no non-diabetic patients prescribed metformin were included in this study. Subsequently, patients within the No-Met group were further categorized based on their diabetic status: diabetic patients not using metformin but managed with other glucose-lowering agents (DM-Met group) and non-diabetic patients (No DM group).

Treatment regimen

All patients included in the study had undergone total cystectomy followed by adjuvant chemotherapy with gemcitabine in combination with cisplatin. The treatment protocol consisted of intravenous administration of gemcitabine at a dose of 1250 mg on days 1 and 8, and intravenous cisplatin at a dosage of 70 mg/m² on day 2, with treatments repeated every 21 days for a minimum of four cycles. All included patients underwent postoperative follow-up, with a review at least every 3–4 months in the first year after surgery, once every six months in the second year, and annually thereafter. The follow-up content included physical examination, routine hematological tests, and upper urinary tract and chest CT scans as needed, until disease progression. Disease progression was defined as tumor recurrence in the surgical area, regional lymph nodes, and distant metastasis.Progression-free survival (PFS) was defined as the time from the date of radical cystectomy to the first occurrence of disease progression (local recurrence or distant metastasis), death from any cause, or the last known follow-up.

Pathological assessment

The pathological staging of specimens was carried out according to the TNM classification of the International Union Against Cancer for BLCA and the grading system established by the World Health Organization in 2004. All surgical specimens were processed in accordance with standard pathological protocols.

Clinical and pathological characteristics

The clinical and pathological data collected included gender, age, body mass index (BMI), history of hypertension and smoking, ECOG performance status, tumor stage and grade, number and maximum diameter of tumors, lymph node involvement, surgical method, previous transurethral resection of bladder tumor (TURBT), and the specific number of postoperative chemotherapy cycles administered.

Statistical analysis

All variables were dichotomized, and the distribution of baseline data was evaluated through normality tests. Descriptive statistics were appropriately applied depending on whether variables followed a normal or non-normal distribution. For the comparison of categorical data, Fisher’s exact test was utilized when expected frequencies fell below 5; otherwise, the chi-squared test was employed. All statistical analyses were performed using R software (version 4.2.2). A two-tiered analytical strategy was employed to assess the association between metformin use and progression-free survival (PFS). First, a primary pooled analysis compared the Metformin group (Met group) with the Non-Metformin group (No Met group), as defined in the Patient Cohort section. Second, a supplementary pairwise comparison was conducted among the three original groups (DM + Met, DM-Met, and No DM) to delineate specific effects. Survival distributions were estimated using the Kaplan–Meier method and compared with the log-rank test. And we performed three separate multivariable Cox regression models to make pairwise comparisons between groups: (1) DM + Met vs. No DM; (2) DM + Met vs. DM-Met; (3) DM-Met vs. No DM. Univariable and multivariable Cox proportional hazards models were applied to calculate hazard ratios (HRs) and 95% confidence intervals (CIs), adjusting for clinicopathological variables. The proportional hazards assumption was verified using Schoenfeld residuals. In this study, all P-values were two-sided, with statistical significance set at P < 0.05. All statistical analyses were conducted utilizing R software (version 4.2.2) and MSTATA software.

Results

Clinical and pathological characteristics

The baseline characteristics of the patients, along with their association with metformin usage, are depicted in Table 1. Among the 243 enrolled patients, 68 (28%) were diagnosed with diabetes, with metformin being administered to 51 (75%) of these diabetic individuals. No statistically significant differences were identified in pathological staging, lymph node involvement, pathological grading, surgical approach, TURBT prior to RC, or chemotherapy cycles across the groups, except for a variation in BMI distribution (p = 0.039).

Table 1.

Patient demographics and baseline characteristics

Characteristic	Group				p-value
	Overall, N = 2431	DM, metformin, N = 511	DM, no metformin, N = 171	No DM, N = 1751
Gender					0.52
Female	43 (18%)	11 (22%)	4 (24%)	28 (16%)
Male	200 (82%)	40 (78%)	13 (76%)	147 (84%)
Age(years)					0.12
< 75	215 (88%)	41 (80%)	16 (94%)	158 (90%)
≧75	28 (12%)	10 (20%)	1 (6%)	17 (10%)
BMI					0.0393
>23.0	161 (66%)	29 (57%)	8 (47%)	124 (71%)
≦23.0	82 (34%)	22 (43%)	9 (53%)	51 (29%)
Hypertension					0.42
No	213 (88%)	43 (84%)	14 (82%)	156 (89%)
Yes	30 (12%)	8 (16%)	3 (17.65%)	19 (11%)
Smoking history					0.43
No	104 (43%)	20 (39%)	10 (59%)	74 (42%)
Yes	139 (57%)	31 (61%)	7 (41%)	101 (58%)
ECOG					0.63
0	162 (67%)	31 (61%)	12 (71%)	119 (68%)
1	81 (33%)	20 (39%)	5 (29%)	56 (32%)
Number of tumors					0.32
< 3	223 (92%)	45 (88%)	15 (88%)	163 (93%)
≧3	20 (8%)	6 (12%)	2 (12%)	12 (7%)
Maximum diameter(cm)					0.73
< 3	80 (33%)	15 (29%)	7 (41%)	58 (33%)
≧3	163 (67%)	36 (71%)	10 (59%)	117 (67%)
Pathologic staging(T)					0.72
2	96 (39%)	24 (47%)	6 (35%)	66 (38%)
3	60 (25%)	12 (24%)	5 (29%)	43 (24%)
4	87 (36%)	15 (29%)	6 (36%)	66 (38%)
Lymphatic  invasion(N)					0.53
0	157 (65%)	35 (69%)	9 (53%)	113 (65%)
1	86 (35%)	16 (31%)	8 (47%)	62 (35%)
Pathology grade					0.82
HG	229 (94%)	48 (94%)	17 (100%)	164 (94%)
LG	14 (6%)	3 (6%)	0 (0%)	11 (6%)
Operation4					0.33
1	144 (59%)	28 (55%)	13 (76%)	103 (59%)
2	99 (41%)	23 (45%)	4 (24%)	72 (41%)
TURBT before RC					0.32
No	56 (23%)	16 (31%)	4 (24%)	36 (21%)
Yes	187 (77%)	35 (69%)	13 (76%)	139 (79%)
Chemotherapy cycle					0.82
> 4	23 (10%)	5 (10%)	2 (12%)	16 (9%)
4	219 (90%)	45 (90%)	15 (88%)	159 (91%)

¹n (%)

²Fisher's exact test

³Pearson's Chi-squared test

⁴Operations:1 means Laparoscopic Radical Cystectomy with Orthotopic Ileal Neobladder Reconstruction; 2 means Laparoscopic Radical Cystectomy with Cutaneous Ureterostomy (Single or Bilateral)

Kaplan–Meier survival analysis

Over a median follow-up duration of 17 months, disease progression was observed in 174 individuals (72%). The PFS rates at one, two, and three years were 75%, 24%, and 13% for patients without metformin; 85%, 53%, and 24% for diabetic patients with metformin (Table 2). The three-year PFS rate was higher for patients with metformin compared to those who were not (p = 0.023, Fig. 1).

Table 2.

Kaplan–Meier Estimates for Progression- free Survival Rates (95% CI)

Characteristic	N	Event N	1-year	2-year	3-year	p-value1
Overall	243	216	77% (72%, 83%)	29% (24%, 36%)	15% (11%, 21%)
Group	243	216				0.023
No Met			75% (69%, 82%)	24% (19%, 31%)	13% (9%, 19%)
Met			85% (75%, 96%)	53% (39%, 71%)	24% (13%, 43%)

¹Log-rank test

Fig. 1.

Kaplan–Meier curves represent the recurrence-free survival rates of patients after radical surgery with GC adjuvant chemotherapy

Univariable and multivariable analysis

The univariable analysis indicated that, in comparison to the patients without metformin, those administered metformin exhibited a diminished risk of disease recurrence (HR: 0.66, 95% CI 0.46–0.95, p = 0.024). In the multivariable regression analysis, after adjustments were made for age, gender, BMI, hypertension, smoking history, tumor stage, grade, tumor number and maximum diameter, patients receiving metformin continued to show a significant association with reduced PFS when compared to the patients without metformin(HR: 0.66, 95% CI 0.45–0.95, p = 0.025) (Table 3).

Table 3.

univariable and multivariable analysis of influencing factors (Cox regression)

Characteristic	Univariable					Multivariable
	N	Event N	HR1	95% CI1	p-value	N	Event N	HR1	95% CI1	p-value
Gender
Female	43	41	—	—		43	41	—	—
Male	200	175	0.78	0.55, 1.10	0.2	200	175	0.75	0.53, 1.07	0.1
Age(years)
< 75	215	194	—	—		215	194	—	—
≧75	28	22	1.16	0.74, 1.81	0.5	28	22	1.22	0.78, 1.93	0.4
BMI
>23.0	161	144	—	—		161	144	—	—
≦23.0	82	72	1.14	0.86, 1.52	0.4	82	72	1.07	0.80, 1.45	0.6
Hypertension
No	213	190	—	—		213	190	—	—
Yes	30	26	1.06	0.71, 1.60	0.7	30	26	0.85	0.55, 1.31	0.4
Smoking history
No	104	91	—	—		104	91	—	—
Yes	139	125	1.11	0.85, 1.46	0.4	139	125	1.00	0.74, 1.36	0.9
Number of tumors
< 3	223	199	—	—		223	199	—	—
≧3	20	17	0.90	0.55, 1.48	0.7	20	17	0.86	0.50, 1.47	0.5
Maximum diameter(cm)
< 3	80	77	—	—		80	77	—	—
≧3	163	139	1.06	0.80, 1.40	0.7	163	139	0.93	0.69, 1.25	0.6
Pathologic staging(T)
2	96	84	—	—		96	84	—	—
3	60	53	1.35	0.95, 1.92	0.090	60	53	1.30	0.91, 1.87	0.1
4	87	79	1.80	1.31, 2.47	<.001	87	79	1.78	1.27, 2.49	<.001
Lymphatic invasion(N)
0	157	138	—	—		157	138	—	—
1	86	78	1.10	0.83, 1.45	0.5	86	78	1.06	0.79, 1.42	0.7
Pathology grade
HG	14	13	—	—		14	13	—	—
LG	229	203	2.17	1.16, 4.04	0.015	229	203	1.96	1.02, 3.75	0.043
Group
No Met	192	180	—	—		192	180	—	—
Met	51	36	0.66	0.46, 0.95	0.024	51	36	0.66	0.45, 0.95	0.025

¹HR Hazard Ratio, CI Confidence Interval

Pairwise comparison

To delineate the specific effect of metformin from the underlying status of diabetes, we performed a comprehensive set of pairwise comparisons of progression-free survival (PFS) among the three patient groups. Using non-diabetic patients (No DM, n = 175) as the reference group in a multivariable Cox model, we found that diabetic patients on metformin (DM + Met, n = 51) had a reduced risk of progression (HR = 0.67, 95% CI: 0.46–0.98; p = 0.037, Table 4). In a separate model using diabetic patients not on metformin (DM-Met, n = 17) as the reference, metformin use was powerfully associated with a 66% reduction in progression risk (HR = 0.34, 95% CI: 0.16–0.75; p = 0.007, Table 5). This protective effect was visually apparent in Kaplan–Meier curves (Fig. 2), showing superior 3-year PFS for DM + Met over both DM-Met (23.6% vs. 7.2%; p = 0.007) and No DM (23.6% vs. 13.4%; p = 0.034). Critically, when comparing the two non-metformin groups directly, neither the survival curves (p = 0.2) nor the Cox regression (HR = 1.43 for DM-Met vs. No DM; p = 0.2) showed a significant difference. This pattern of results indicates that the survival benefit is specifically conferred by metformin use, not merely by the presence or absence of diabetes (Tables 6, 7, 8 and 9).

Table 4.

Patient follow-up duration

Group	N	Event N	Median Follow-up, months (IQR)
Overall	243	174 (72%)	17.0 (15.8, 19.0)
DM + Met	51	36 (71%)	26.1 (20.4, 30.9)
DM-Met	17	15 (88%)	14.3 (12.1, 25.7)
No DM	51	123 (70%)	16.2 (14.8, 18.1)

Table 5.

Cox Regression analysis of the effect of metformin on progression-free survival: Diabetic Metformin Users (DM + Met) versus non-diabetic patients (No DM)

Characteristic	N	Event N	HR1	95% CI1	p-value
Before adjustment
No DM	175	165	—	—
DM + Met	51	36	0.68	0.47, 0.97	0.035
After adjustment
No DM	175	165	—	—
DM + Met	51	36	0.67	0.46, 0.98	0.037

¹HR Hazard Ratio, CI Confidence Interval

Fig. 2.

Kaplan–Meier survival analysis for progression-free survival: a pairwise comparison across the three patient groups

Table 6.

Cox Regression analysis of the effect of metformin on progression-free survival in patients with diabetes (DM + Met vs. DM-Met)

Characteristic	N	Event N	HR1	95% CI1	p-value
Before adjustment
DM-Met	17	15	—	—
DM + Met	51	36	0.43	0.23, 0.81	0.009
After adjustment
DM-Met	17	15	—	—
DM + Met	51	36	0.34	0.16, 0.75	0.007

¹HR  Hazard Ratio, CI  Confidence Interval

Table 7.

Cox Regression Analysis of Progression-Free Survival: Diabetic Patients without Metformin (DM-Met) versus Non-Diabetic Patients (No DM)

Characteristic	N	Event N	HR1	95% CI1	p-value
Before adjustment
No DM	175	165	—	—
DM-Met	17	15	1.46	0.86, 2.48	0.2
After adjustment
No DM	175	165	—	—
DM-Met	17	15	1.43	0.81, 2.51	0.2

¹HR Hazard Ratio, CI Confidence Interval

Table 8.

Correlation analysis of the impact on chemotherapy-related adverse reactions (Pairwise Comparison)

	No DM	DM, metformin	DM, no metformin
	(N = 175)	(N = 51)	(N = 17)
Responders	70 (40%)	11 (22%)	6 (35%)
Non-Responders	105 (60%)	40 (78%)	11 (65%)
Unstratified Response Analysis
Risk Difference, RD (%) (95% CI)		−18.43(−31.85, 5.01)	−4.71 (−28.55, 19.14)
Risk Ratio, RR (95% CI)		0.54 (0.31—0.94)	0.88 (0.45—1.72)
Odds Ratio, OR (95% CI)		0.41 (0.20—0.86)	0.82 (0.29—2.31)
p-value (Chi-squared test)		0.016	0.7
p-value (Fisher's exact test)		0.019	0.8

Table 9.

Logistic regression for chemotherapy-related adverse reactions analysis

Characteristic	OR1	95% CI1	p-value
Treatment (unadjusted)
No DM	—	—
DM, metformin	0.41	0.19, 0.83	0.018
DM, no metformin	0.82	0.27, 2.25	0.7
Treatment (adjusted)
No DM	—	—
DM, metformin	0.34	0.14, 0.75	0.011
DM, no metformin	0.84	0.26, 2.53	0.8

¹OR Odds Ratio, CI Confidence Interval

Impact on chemotherapy toxicities

During the assessment of chemotherapy-induced adverse reactions, gastrointestinal symptoms such as nausea and vomiting, alongside hematological toxicities like leukopenia and thrombocytopenia, were the most frequently observed adverse effects across the three patient cohorts. Grade 3 or more severe hematological toxicities were reported in 70 out of 175 non-diabetic patients (40%), 11 out of 51 diabetic patients on metformin (22%), and 6 out of 17 diabetic patients not on metformin (35%) (Table 4). Additionally, patients on metformin exhibited a lower incidence of grade 3 or higher hematological adverse events compared to those not on metformin (22% vs. 37%). When compared to non-diabetic patients, those receiving metformin demonstrated a reduced incidence of grade 3 or higher adverse reactions (p = 0.016), whereas the presence of diabetes among non-metformin users did not appear to correlate with an increased incidence of severe adverse reactions (p > 0.05). Even after adjusting for variables such as gender, age, BMI, hypertension, smoking history, tumor characteristics, and chemotherapy cycles, metformin use remained associated with a lower incidence of grade 3 or 4 adverse reactions (p = 0.011) (Table 5).

Discussion

The outcomes of this study demonstrate that the use of metformin enhances PFS in patients receiving adjuvant GC chemotherapy following RC for BLCA. This finding highlights the potential significance of metformin in cancer therapy, particularly within the adjuvant context post-BLCA surgery. Although the specific mechanisms of metformin were not directly examined in this study, both the existing literature and our results imply that metformin may exert its anticancer and toxicity-reducing effects through several pathways. Beyond the previously mentioned mechanisms, including inhibition of tumor cell proliferation, promotion of apoptosis, and suppression of invasion and metastasis, metformin might also influence the tumor microenvironment, regulate the expression of inflammatory cytokines, and ultimately enhance its anticancer properties. Additionally, metformin could potentially mitigate the deleterious effects of chemotherapy by modulating energy metabolism and cellular signaling pathways.

One of the essential mechanisms by which metformin exerts its anticancer effects is through the activation of the AMPK pathway, subsequently leading to the suppression of the mTOR signaling pathway. This cascade results in the inhibition of tumor cell proliferation, the induction of apoptosis, and the reduction of tumor growth, thereby manifesting its anticancer efficacy [24, 25]. Additionally, metformin has the capacity to alter the tumor microenvironment by suppressing the activity of tumor-associated fibroblasts and inhibiting tumor angiogenesis, which may mitigate tumor aggressiveness and metastatic potential [26]. Its anti-inflammatory and immunomodulatory properties, demonstrated by the reduction of pro-inflammatory cytokine secretion and the enhancement of CD⁸⁺ T cell activity, further contribute to the suppression of tumor growth [27]. Moreover, the antioxidant capabilities of metformin might enhance patient tolerance to chemotherapy by diminishing oxidative stress and decreasing DNA damage, thereby potentially improving prognosis [28].

Metformin has been observed to potentially enhance survival outcomes across various cancer types. For example, in the case of prostate cancer, the use of metformin has been correlated with a notable improvement in overall survival (OS), reflected by a pooled HR of 0.79 (95% CI 0.63–0.98) when compared to non-users [29]. In gastric cancer, particularly among diabetic patients who have undergone gastrectomy, metformin has been demonstrated to extend cancer-specific survival (CSS), with the most substantial benefits seen in stage III gastric cancer patients [30]. Regarding BLCA, several studies have investigated the clinical effects of metformin. For instance, one study highlighted a significant link between metformin usage and enhanced OS in patients with non-muscle-invasive bladder cancer (NMIBC), where the most favorable outcomes were identified in diabetic patients utilizing metformin [31]. Rieken et al. conducted two studies revealing that, in comparison to non-diabetic patients, those using metformin displayed a reduced risk of disease recurrence in NMIBC (HR: 0.50, 95% CI: 0.27–0.94, p = 0.03) and a decreased risk of disease progression (HR: 0.61, 95% CI 0.37–0.98, p = 0.04), cancer-specific mortality (HR: 0.56, 95% CI 0.33–0.97, p = 0.04), and all-cause mortality (HR: 0.54, 95% CI 0.33–0.88, p = 0.01) in MIBC patients following RC [32, 33]. Furthermore, another study identified a significant link between metformin intake and improved RFS, PFS, and CSS in BLCA patients [34].

Recent investigations have concentrated on the potential synergistic interactions between metformin and various chemotherapeutic agents, notably cisplatin and gemcitabine, in the context of BLCA treatment. Metformin has been found to enhance the cytotoxic effects of these chemotherapy drugs, thereby further inhibiting the proliferation of tumor cells. For example, the combination of metformin with gemcitabine has been shown to produce significant synergistic antitumor effects, thereby increasing the sensitivity of BLCA cells to chemotherapy while simultaneously reducing drug resistance [35]. Additionally, metformin is posited to mitigate adverse reactions related to chemotherapy, particularly those induced by cisplatin and gemcitabine. Research indicates that metformin may decrease toxicity to normal tissues by modulating intracellular energy metabolism and enhancing antioxidant stress responses. This protective effect is not limited to the mitigation of common hematological toxicities but may also play a crucial role in alleviating chemotherapy-associated gastrointestinal symptoms and systemic discomfort [36].

However, certain studies have observed that prolonged metformin usage in BLCA patients may correlate with reduced OS, disease-specific survival, and RFS, indicating possible variations in metformin’s effects across distinct patient populations [37]. Although the efficacy of metformin may not be pronounced in specific cancer types, such as renal and head and neck cancers, where no statistically significant link with enhanced survival outcomes has been detected [38], it still possesses the potential to extend PFS and CSS in particular cancer types, offering valuable insights for its role in cancer treatment.

Notably, this study assessed the influence of metformin on chemotherapy-related adverse reactions, demonstrating that patients administered metformin exhibited.

lower incidences of grade 3 or higher hematological adverse reactions as well as overall adverse reactions when compared to non-users (21.57% vs. 37.14% and 21.57% vs. 40%, respectively). This discovery not only highlights metformin’s potential in mitigating chemotherapy toxicity but also systematically unveils this effect for the first time in BLCA patients undergoing postoperative chemotherapy, suggesting new avenues for future clinical treatment strategies. Through the utilization of a large patient cohort and robust statistical analysis, this study not only affirms the potential benefits of metformin in adjuvant BLCA therapy but also sheds light on its clinical relevance and protective effects. Compared to prior research, the results of this study demonstrate greater clinical significance and wider applicability [39, 40].

While our findings suggest a potential benefit of metformin, we acknowledge the inherent limitations of a retrospective analysis. As rightly noted, a randomized controlled trial (RCT) would be required to definitively establish a causal effect and quantify the efficacy of adding metformin to standard adjuvant GC chemotherapy. Although a dedicated phase III RCT in this specific setting may face practical challenges in patient accrual and design, our results provide the robust preliminary evidence necessary to justify and inform the design of such prospective studies, or smaller-scale phase II investigations. Until then, the decision to use metformin in diabetic patients undergoing this treatment should be made individually, considering glucose control needs and in consultation with the treating oncologist.

There are several limitations must be recognized. Firstly, the retrospective nature of this study introduces the possibility of selection bias, particularly with regard to patient inclusion criteria and metformin usage. Given that all participants were sourced from a single medical center or comparable healthcare environments, the external validity of these findings may be constrained, making it challenging to generalize them to diverse patient populations or different healthcare settings. Furthermore, the relatively short duration of follow-up, while suggesting metformin’s potential advantages at the three-year mark, may not offer a comprehensive assessment of long-term outcomes. Despite efforts to control for confounding factors through multivariable analysis, there remains the possibility that unmeasured or unknown variables could have impacted the results. Future research should incorporate prospective randomized controlled trials to validate these findings and explore metformin’s mechanisms in BLCA treatment more thoroughly, thereby providing a stronger scientific foundation for clinical practice.

In conclusion, this study provides preliminary clinical evidence supporting the potential utility of metformin in the adjuvant setting for bladder cancer. Our findings, derived from a retrospective cohort and strengthened by direct pairwise comparisons, suggest an association between metformin use and improved survival outcomes alongside a more favorable toxicity profile. Given its widespread availability, established safety record, and low cost, metformin represents a promising candidate for further clinical evaluation. Future prospective studies are warranted to validate these findings, elucidate the underlying mechanisms, and define optimal dosing and patient selection criteria—particularly focusing on its role in diabetic populations—to fully assess its therapeutic potential.

Conclusions

This retrospective analysis of 243 patients undergoing radical cystectomy followed by adjuvant GC chemotherapy for bladder cancer suggests potential benefits associated with metformin use. Our findings indicate that in diabetic patients, metformin use was associated with improved PFS and a lower incidence of severe adverse reactions.

This study provides preliminary evidence highlighting the potential of metformin as an adjunctive therapy and lends support for its consideration in postoperative treatment strategies. Given its favorable safety profile, low cost, and wide availability, metformin represents a promising candidate for further clinical evaluation in this setting.

Abbreviations

GC

Gemcitabine and cisplatin

MIBC

Muscle-invasive bladder cancer

RC

Radical cystectomy

PFS

Progression-free survival

Authors’ contributions

Lei Zhang and Jingyi Cao are the co-first authors who mainly wrote the manuscript text. Chong Han conducted the investigation and data compilation. Jingwen Zhang performed the formal analysis of this paper. Yiwen Lui was mainly responsible for data validation. Hailong Li, as the corresponding author, participated in the review and editing of the manuscript, project management, fund acquisition, and provided supervision and guidance.

Funding

National Nature Science Foundation of China Youth Project (Project approval number: 81502193).

Data availability

The datasets generated and/or analyzed during the current study are not publicly available due to privacy concerns and ethical restrictions related to patient confidentiality. However, the corresponding author can be contacted (justinlee719@163.com or zlei2023@163.com) for reasonable requests related to the study design, data collection and analysis methods, provided they comply with the ethical guidelines and regulations governing the use of human subject data.

Declarations

Ethics approval and consent participate

This retrospective study was conducted in accordance with the Declaration of Helsinki and approved by the Ethics Committee of The Affiliated Hospital of Xuzhou Medical University (XYFY2022-KL340). Since the study involved the review of existing medical records and did not require direct patient contact or intervention, the need for informed consent from individual patients was waived by the ethics committee. However, all patient data were anonymized to ensure patient confidentiality and privacy.

Consent for publication

The authors confirm that all authors have reviewed and approved the final version of the manuscript for publication. All authors agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. There are no conflicts of interest to disclose that could potentially influence the publication of this manuscript.

Competing interests

The authors declare no competing interests.