Abstract
Background
With an increase in morbidity of patients with type 2 diabetes mellitus (T2DM), to investigate the impact of inadequate prescription practices and medication habits on tumor-related risk factors.
Methods
We analyzed data from 2114 patients with T2DM treated at our hospital from January 2019 to September 2022. The Medication Adherence Report Scale-5 items (MARS-5) and medication possession ratio (MPR) were used to assess medication use. Variables considered included diabetes duration, glycated hemoglobin (HbA1c) level, hypoglycemic drug regimens, and instances of irrational drug use. Univariate analysis and multivariate logistic regression were employed to identify risk factors associated with drug use in both the tumor and control groups, with a specific focus on lung cancer.
Results
A total of 766 (36.23%) malignant tumors were identified, including 257 (33.55%) in the respiratory system, with 246 (32.11%) cases of lung cancer. Univariate analysis revealed that the use of insulin, combination of metformin with insulin and other oral hypoglycemic agents (OHAs), and instances of irrational drug use were significantly associated with malignant tumors in patients with T2DM (P < 0.05). Multivariate analysis: HbA1c ≥ 7% was no longer associated with higher tumor(P > 0.05), the irrational drug use is risk factors for high incidence of malignant tumors, (P < 0.001), which included factors such as insufficient dosing (odds ratio [OR]: 4.348; 95% confidence interval [CI]: 2.709, 6.978), not combined medication (OR: 4.801; 95% CI: 3.484, 6.615), frequent medication changes (OR: 6.056; 95% CI: 4.185, 8.763), poor medication compliance (OR: 3.716; 95% CI: 2.667, 5.179), and wrong timing of administration (OR: 3.792; 95% CI: 2.687, 5.353). Similar in lung cancer.
Conclusion
Irrational drug use is a significant risk factor for the development of malignant tumors, particularly lung cancer, is't correlated with abnormal HbA1c levels.
Keywords: Type 2 diabetes mellitus, Irrational drug use, Risk factors, Lung cancer, HbA1c
1. Background
Type 2 diabetes mellitus (T2DM) is widely recognized as a significant chronic metabolic disease.1 The rising prevalence of T2DM in recent decades has placed a considerable burden on global healthcare systems.2 Growing evidence links T2DM to a range of serious complications, including diabetic nephropathy, cardiovascular disease, and malignant tumors.3, 4, 5, 6 With ongoing economic development and shifts in lifestyle, diabetes and cancer have emerged as leading causes of mortality, following cardiovascular disease.7., 8, 9 A noteworthy aspect of T2DM is at 1.5greater risk of developing cancer and of dying from it.3 In china, lung cancer ranks first, and the deaths due to high fasting plasma glucose has been increasing year by year.10The adoption of a healthy lifestyle could substantially reduce premature mortality due to type 2 diabetes mellitus and its complications (e.g., tumor), are related to the wider socioeconomic and cultural environment.11
T2DM is characterized by two significant pathophysiological abnormalities: insulin resistance and inadequate insulin secretion. These factors can stimulate tumor development by affecting multiple signaling pathways. Notably, the insulin and insulin-like growth factor 1 (IGF-1) pathways, along with inflammatory responses and reactive oxygen species production,12 can facilitate tumor cell proliferation. Observational studies have shown a heightened risk of liver, pancreatic, and uterine cancers among individuals with T2DM.13., 14, 15, 16 While T2DM is recognized as a potential risk factor for various cancers,17 the relationship between drug use and cancer risk remains underexplored. Therapeutic interventions targeting hyperglycemia can improve insulin resistance associated with hyperinsulinemia. Additionally, specific anti-diabetic medications, such as metformin, sulfonylureas, and thiazolidinediones, may significantly influence tumorigenesis.3
Recent findings have indicated that metformin, which has demonstrated better tolerance(such as: gastrointestinal reactions can be relieved with the prolongation of treatment) in managing blood sugar levels, and reduce the 20–30% risk of tumorigenesis, somewhat greater benefit for cancer-related mortality, whereas exogenous insulin could increase that risk.18 Moreover, non-adherence medication use among patients with diabetes is directly linked to higher colorectal cancer mortality,19 and nonadherence to treatment poses an increased risk of all-cause hospitalizations for patients with cancer.20 Metformin is associated with a decreased risk of cancer incidence, a dose-response relationship was noted,21 and thiazolidinediones are considered beneficial except for bladder cancer; insulin significantly increases the overall incidence of cancer, same in the first and second generation sulfonylurea drugs.22, 23. Therefore, the rational use of hypoglycemic medications is crucial in mitigating cancer risk.
In this study, we aim to investigate the risks of tumorigenesis associated with the irrational use of hypoglycemic drugs. We will identify the factors influencing total tumor and lung cancer incidence in patients with T2DM to provide insights for optimizing medication usage.
2. Methods
2.1. Patients and study design
We conducted a review of the electronic medical records to collect data on patients with T2DM in The Affiliated Hospital of Xuzhou Medical University (Single-center) from January 2019 to September 2023. The Medication Adherence Report Scale-5 items (MARS-5) was used to quantify the impact of diabetes on patients' physical, psychological, social functions, treatment management and overall quality of life, and to gather information on prescribed treatment regimens and evaluate medication adherence, along with the medication possession ratio (MPR), the investigation continue until: 2024-12-30. MPR assessed concurrently with MARS-5 responses: after 21 days of outpatient treatment and medication administration, a follow-up visit was conducted, 90 days later to observe the medication behavior. Extract prescription data from electronic pharmacy records or medical insurance databases. Exclusion rule: If the medication is provided by the hospital during hospitalization, it will not be included in the MPR.
The study included two cohorts: the case group, which consisted of individuals with T2DM who developed secondary malignancies, and the control group, comprised solely of patients with T2DM. If a member of the control group developed a tumor during the study period, they were reclassified into the case group. We recorded various clinical parameters, including age, sex, height, weight, body mass index (BMI), glycated hemoglobin (HbA1c) levels(measure at least once every three months), duration of T2DM, type of hypoglycemic medications used, and instances of irrational drug use. Out of a total of 4867 patients with T2DM, 2114 had available HbA1c values for analysis. A flow diagram outlining patient selection is presented in Fig. 1. A cross-sectional study data was conducted to analyze the risk factors associated with drug usage, secondary malignancies and lung cancer.
Fig. 1.
Flow diagram for patient selection in the analysis.
Inclusion criteria for the study included a confirmed diagnosis of T2DM, age > 18 years, a pathological diagnosis of a malignant tumor occurring after T2DM diagnosis, having received anti-glucose therapy for at least 1 year, and demonstrating good cooperation.
Patients were excluded if they had type 1 diabetes or other forms of diabetes, such as gestational diabetes, a minimum diabetes duration and so on; if their malignant tumors were diagnosed prior to T2DM; if they exhibited non-cooperation with the investigation or had missing baseline data; if their conditions were deemed unsuitable for inclusion, such as: refusing to cooperate with diabetes treatment for personal reasons.
2.2. Diagnostic criteria
T2DM was diagnosed according to the medical system records.An HbA1c level < 7.0% was recommended. The normal BMI range is defined as 18.5–23.9 kg/m2, with values <18.5 kg/m2 classified as lean, ≥24 kg/m2 as overweight, and ≥ 28 kg/m2as obese.24 Malignant tumors were diagnosed based on pathological evidence, in conjunction with clinical manifestations, imaging studies, and cytological findings.
2.3. Exposure to hypoglycemic regimens
The hypoglycemic drugs used for diabetes management were classified into three main categories for the purposes of this study: metformin, insulin, and oral hypoglycemic agents (OHAs), which include sulfonylureas, glinides, thiazolidinediones, DPP-4 inhibitors, α-glucosidase inhibitors, and SGLT-2 inhibitors. The patients were grouped into seven subsets based on the nature of hypoglycemic regimens: metformin only, insulin only, other OHAs only, metformin + insulin, metformin + other OHAs, insulin + other OHAs, and metformin + other OHAs + insulin.
2.4. Irrational drug use criteria
Irrational drug use was defined in accordance with the guidelines established by the 1985 WHO Expert Meeting on Rational Drug Use25 and the Diabetes Advocacy: Standards of Medical Care in Diabetes-2022.”26 This type of use was identified when the same irrational drug use behavior occurs cumulated for a period of >3 months. Specific criteria for identifying irrational drug use included insufficient dosing, which involved inadequate prescribed doses without self-monitoring of blood sugar; non-combination medication, where a lack of necessary combinations led to suboptimal effects; frequent medication changes, characterized by the number of newly added or replaced drugs within 3 months exceeds 3 types; poor medication compliance, defined as missed doses or inconsistent intake(Medication possession Ratio, MPR also give a reminder); and wrong timing of administration. Ultimately, data collection was achieved through hospital prescription research, medical history records, MARS-5, MPR and patient interviews, analyze and sort out cases of irrational drug use, input into the excel sheet.
2.5. Statistical analysis
Data integrity was ensured through the identification and removal of outliers via two quality control channels. Statistical analyses were performed using IBM SPSS Statistics (version 26, IBM Corp., Armonk, NY), and figures were created with GraphPad Prism software (version 7.0, GraphPad Inc., San Diego, CA, USA). Continuous variables were reported as means with standard deviations and analyzed using independent-samples t-tests or one-way analysis of variance (ANOVA), while categorical variables were expressed as frequency rates or percentages and analyzed using the Chi-square test or Fisher's exact test. The Wilcoxon rank-sum test was used for data that were not normally distributed.
Binary logistic regression was employed with secondary malignant tumors in T2DM as the dependent variable, while various factors served as independent variables. Significant factors were further analyzed through multivariate logistic regression to assess their risk. Odds ratios (ORs) and 95% confidence intervals (CIs) were calculated to estimate the risk of malignant tumors in patients with T2DM. A P value <0.05 was considered statistically significant, with all P values being two-tailed.
2.6. Ethics approval
All procedures involving human participants were conducted in accordance with the ethical standards of the institutional and/or national research committee, as well as the 1964 Helsinki Declaration and its later amendments. This study received approval from the ethics committee of the Affiliated Hospital of Xuzhou Medical University (Ethical Code: XYFY2022-KL429–01). All patients were unrelated, and informed consent was obtained from each participant.
3. Results
3.1. Patient characteristics
The baseline characteristics of the 766 patients with malignant tumors, including 246 with lung cancer, and 1348 controls are summarized (Table 1). There were no statistically significant differences in sex, age (years), BMI in kg/m2, or duration of T2DM (years) among the two tumor groups and controls (P > 0.05; see Table 1). The distribution of the 766 malignant tumors was primarily in the respiratory system (257 cases, 33.55%), digestive system (248 cases, 32.38%), female reproductive system (71 cases, 9.27%), lymphatic system (50 cases, 6.53%), urinary system (50 cases, 6.53%), endocrine system (38 cases, 4.96%), blood system (28 cases, 3.65%), and other systems (24 cases, 3.13%) (Fig. 2). The five most common malignant tumors were lung cancer (246 cases, 32.11%), liver cancer (81 cases, 10.57%), colorectal cancer (56 cases, 7.31%), gastric carcinoma (41 cases, 5.35%), and thyroid cancer (38 cases, 4.96%).
Table 1.
Baseline characteristics and univariate analysis of the total tumor, lung cancer, and control groups in patients with T2DM.
| Characteristic | Controls |
Case |
aP value | Lung cancer |
bP value |
|---|---|---|---|---|---|
|
n = 1348 (%) |
(Total tumor) |
n = 246 (%) |
|||
| n = 766 (%) | |||||
| Sex | 0.989 | 0.971 | |||
| Female | 429 (31.82) | 244 (31.85) | 78 (31.71) | ||
| Male | 919 (68.18) | 522 (68.15) | 168 (68.29) | ||
| Age (Years) | 0.67 | 0.805 | |||
| ≤60 | 383 (28.41) | 211 (27.55) | 68 (27.64) | ||
| >60 | 965 (71.59) | 555 (72.45) | 178 (72.36) | ||
| BMI (kg/m2) | 0.946 | 0.996 | |||
| <24 | 575 (42.66) | 323 (42.17) | 105 (42.68) | ||
| 24–27.9 | 562 (41.69) | 325 (42.43) | 103 (41.87) | ||
| ≥28 | 211 (15.65) | 118 (15.40) | 38 (15.45) | ||
| T2DM duration (Years) | 0.32 | 0.318 | |||
| ≤5 | 559 (41.47) | 331 (43.21) | 114 (46.34) | ||
| 6–10 | 345 (25.59) | 207 (27.02) | 61 (24.80) | ||
| ≥10 | 444 (32.94) | 228 (29.77) | 71 (28.86) | ||
| Hypoglycemic drug regimen | |||||
| Only metformin | 285 (21.14) | 134 (17.49) | 0.043 | 47 (19.11) | 0.469 |
| Only insulin | 105 (7.79) | 157 (20.50) | <0.001⁎ | 44 (17.89) | <0.001⁎ |
| Only other OHAs | 173 (12.83) | 83 (10.84) | 0.905 | 24 (9.76) | 0.52 |
| Metformin + insulin | 100 (7.4) | 43 (5.61) | 0.671 | 12 (4.88) | 0.353 |
| Metformin + other OHAs | 371 (27.52) | 230 (30.02) | 0.039⁎ | 81 (32.93) | 0.159 |
| Insulin + other OHAs | 141 (10.46) | 49 (6.40) | 0.123 | 13 (5.28) | 0.075 |
| Metformin + insulin + other OHAs | 137 (10.16) | 37 (4.83) | 0.009⁎ | 12 (4.88) | 0.059 |
| HbA1c% | <0.001⁎ | <0.001⁎ | |||
| <7.0% | 607 (45.03) | 218 (28.46) | 53 (21.54) | ||
| ≥7.0% | 741 (54.97) | 548 (71.54) | 193 (78.46) | ||
Note: *: Univariate analysis, significant variables (P < 0.05).
a: P value is the comparison between the case (total tumor) and control groups.
b: P value is the comparison between the lung cancer and control groups.
Fig. 2.
Pie chart illustrating the distribution of malignant tumors among patients with T2DM.
*: In the 257 (33.55%) respiratory system cases, lung cancer was the most prevalent, accounting for 246 (32.11%) cases.
**: Other systems included oropharyngeal cancer (11 cases [1.44%]), brain cancer (6 [0.78%]), skin cancer (3 [0.39%]), cardiac cancer (3 [0.39%]), and tongue cancer (1 [0.13%]).
Univariate analysis indicated that the insulin-only regimen was a potential risk factor (P < 0.001). When patients were administered metformin + insulin, the risk diminished (P > 0.05), maybe the protective effect of metformin (P = 0.043), or for further analysis. However, the risk persisted with metformin + other OHAs (P = 0.039) and with metformin + insulin + other OHAs (P = 0.009). For lung cancer specifically, the insulin regimen was again identified as a risk factor (P < 0.001), while no risks were associated with other hypoglycemic regimens, including only other OHAs, metformin + insulin, metformin + other OHAs, insulin + other OHAs, and metformin + insulin + other OHAs. Additionally, HbA1c% was associated with the risk of tumor development in both overall sample and patients with lung cancer (P < 0.001; Table 1).
3.2. Chi-square monitoring analysis of risk factors
In a comparison of irrational drug use, Chi-square monitoring analysis revealed that HbA1c% was significantly lower in the total cases (P < 0.001; Fig. 3a), with rational drug use exhibiting the lowest values (P < 0.001; Fig. 3a). Irrational drug use was more prevalent in the case group and among patients with lung cancer than in the control group, establishing it as a risk factor(P < 0.001; Fig. 3b): specific types of irrational drug use included insufficient dosing (58 [7.57%] cases in the total tumor group and 19 [7.72%] in the lung cancer group versus 49 [3.64%] in controls); non-combination medication (210 [27.42%] total cases and 68 [27.64%] lung cancer cases versus 182 [13.50%] controls); frequent medication changes (113 [14.75%] total cases and 44 [17.89%] lung cancer cases versus 120 [8.90%] controls); poor medication compliance (135 [17.62%] total cases and 44 [17.89%] lung cancer cases versus 189 [14.02%] controls); and incorrect timing of administration (109 [14.23%] total cases and 34 [13.82%] lung cancer cases versus 142 [10.53%] controls).
Fig. 3.
Distribution of HbA1c and irrational drug use in patients with T2DM.
*: a: Compared with rational drug use, HbA1c < 7.0% of irrational drug use and total cases (HbA1c% group) was higher (P < 0.001).
b: Compared with the control group, the case group (total tumor) showed a higher incidence of irrational drug use (P < 0.05).
3.3. Logistic regression analysis of total tumor
Multivariate logistic regression analysis identified four risk factors associated with malignancy in the cohort of 766 total tumors compared with 1348 controls. As shown in Fig. 4, HbA1c ≥ 7.0% did not show a significant association (OR: 0.940, P = 0.624). In terms of hypoglycemic drug regimens, the insulin-only regimen wasn't identified as a risk factor (OR: 1.105, P = 0.718), whereas only metformin (OR: 0.417, P < 0.001), other OHAs (OR: 0.410, P < 0.001), metformin + insulin (OR: 0.358, P = 0.024), insulin + other OHAs (OR: 0.236, P < 0.001), metformin + insulin + other OHAs (OR: 0.215, P < 0.001) reduced the risk. The following forms of irrational drug use were associated with a heightened risk of secondary malignancies in patients with T2DM: insufficient dosing (OR: 4.348, P < 0.001), non-combination medication (OR: 4.801, P < 0.001), frequent medication changes (OR: 6.056, P < 0.001), poor medication compliance (OR: 3.716, P < 0.001), and incorrect timing of administration (OR: 3.792, P < 0.001)(Fig. 4).
Fig. 4.
Multivariate logistic regression analysis in patients with T2DM with a secondary malignant tumor.
*: Significant variables in the logistic regression analysis (P < 0.05).
Abbreviations: OR: odds ratio; CI: confidence interval.
3.4. Logistic regression analysis of lung cancer
Multiple logistic regression analysis comparing 246 patients with lung cancer with 1348 controls indicated that HbA1c ≥ 7.0% were not significant (OR: 1.368, P = 0.119). The insulin-only regimen didn't remain a risk factor (OR: 0.785, P = 0.518). Furthermore, various forms of irrational drug use were identified as independent risk factors for lung cancer, similar to those identified for total tumors: insufficient dosing (OR: 4.574, P < 0.001), non-combination medication (OR: 5.221, P < 0.001), frequent medication changes (OR: 7.515, P < 0.001), poor medication compliance (OR: 3.716, P < 0.001), and incorrect timing of administration (OR: 3.671, P < 0.001)(Fig. 5).
Fig. 5.
Multivariate logistic regression analysis identifying the risk factors for T2DM with secondary lung cancer.
* Significant variables in the logistic regression analysis (P < 0.05).
Abbreviations: OR: odds ratio; CI: confidence interval.
4. Discussion
Numerous studies have identified a potential correlation between T2DM and certain cancers, as these diseases are often diagnosed concurrently in the same individual.27 For instance, the risk of developing liver28 and pancreatic cancer29 significantly escalates in patients diagnosed with T2DM. In this study, we also observed an elevated incidence of tumors, including lung cancer and gastrointestinal tumors, particularly among older patients and those with poor medication compliance. The hyperglycemic state may be exacerbated by inappropriate medication practices, highlighting the importance of considering prescription habits as integral risk factors and the need for proactive prevention measures.
Therapeutic interventions targeting hyperglycemia can improve insulin resistance associated with hyperinsulinemia. Additionally, specific anti-diabetic medications, such as metformin, sulfonylureas, and thiazolidinediones, may significantly influence tumorigenesis.3 Our regression analysis identified hypoglycemic regimens and irrational drug use as key factors associated with malignancy. These findings have substantial implications for the management of T2DM and strategies aimed at preventing malignant tumors.
First, Chi-square test analysis found that HbA1c ≥ 7.0% was directly associated with an increased risk of malignant tumors, consistent with previous findings.30 Multiple logistic regression analysis HbA1c ≥ 7.0% were not significant not only in total tumor but also in lung cancer. Under the combined influence of multiple factors, this risk factor has not been manifested. The occurrence of tumors is related to multiple factors, simply attributed to an abnormally elevated HbA1c is not appropriate. Further research is needed to explore the heterogeneity of biological mechanisms and sample characteristics that may explain this association.
Secondly, the choice of hypoglycemic drug regimens appears to influence the risk of malignant tumors. While there is considerable interest in the potential of metformin to reduce cancer risk, several studies have reported clear correlation between metformin use and the risk of cancer,31, 32 in line with our findings. Meanwhile, the combination of metformin with other drugs is associated with a lower cancer risk in our study. Specifically, patients receiving triple therapy (metformin + insulin + other OHAs) exhibited a trend toward reduced cancer risk, metformin + insulin or combined with other OHAs also showed a significant negative correlation. Patients using combination drug regimen may have more severe conditions themselves, may develop more complications, and require more aggressive treatment plans. At the same time, such patients may also pay more attention to health management in other aspects, such as diet control and exercise. These factors may work together to play a certain inhibitory role in the occurrence and development of complications, thereby reducing the overall risk. Additionally, insulin alone didn't appears to be associated with an increased risk of secondary cancers, both in the overall patient population and specifically among patients with lung cancer. This finding didn't align with previous research33. Insulin can promote cell proliferation by activating a series of signaling pathways such as PI3K/Akt, and regulate apoptosis-related proteins like Bcl-2 family proteins to inhibit cell apoptosis,34 insulin resistance and hyperinsulinemia can also weaken the surveillance function of the immune system and promote tumor occurrence.35 But the harmful of insulin monotherapy were not seen in this study, behavioral habits and individual physique may offset correlation.
Given the mechanisms of drug action, rational drug use is crucial for helping patients maintain optimal blood glucose levels, thereby creating an environment less conducive to the survival of cancer cells.36 We identified irrational drug use behaviors—such as insufficient dosing, non-combination medication, frequent changes in medication, poor compliance, and incorrect timing of administration—as significant risk factors for secondary malignant tumors in patients with T2DM. However, an increase in HbA1c > 7% does not indicate a high risk of total tumors and lung cancer. More comprehensive examinations are needed for higher levels of HbA1c. The use of medications is the most crucial factor among the influencing factors(such as: unstable and highly fluctuating blood sugar levels). This might be due to the promotion of other mechanisms by this behavior.
Lung cancer was the most prevalent among these tumors, echoing previous research.37 Furthermore, hyperglycemia-induced oxidative stress has been shown to intensify the spread and development of lung cancer in diabetic mouse models,38 prompting us to analyze irrational drug use risk separately. The prevalence of irrational drug use is evident not only in overall tumor patients but also in those with lung cancer, underscoring the need to optimize treatment regimens, enhance patient education, and improve compliance. To effectively mitigate the risk of tumorigenesis—a multifaceted process regulated by genes and signaling pathways and influenced by behavior and other factors—reinforcing patient behaviors, enhancing physician prescribing practices, and establishing clear medication guidelines are imperative.
Not similar with previous studies,37, 38 our study didn't show high HbA1c levels increased the risk of total tumors and lung cancer, but firstly confirmed the single medication behavior and its minimum cumulative duration (3 months) contributed. So achieving timely screening, diagnosis, and appropriate treatment requires deliberate attention to drug use habits within the population, rational use of antihyperglycemic medications, minimizing missed doses, and reducing frequent changes in medication can help maintain normal blood sugar levels and reduce the potential risk of tumor progression.
This study was a single-center research. During the study, although we employed the MARS-5, implemented dual quality control measures, and utilized both univariate and multivariate analyses to control for confounding factors, there are still many shortcomings, such as patients' comorbidities have not been adequately controlled, other OHAs contain too many types of drugs, and long-term follow-up and larger sample sizes are still necessary to further confirm these effects.
Supplementary.
The MPR data can be found in Supplementary Table 1. It can be seen that the MPR of the rational drug use group is higher; not combined (insulin resistance but alone, etc) had a poor effect, frequent dressing change and wrong timing of administration is no difference. However, due to the fact that some patients have hoarded drugs or missed doses, and the medication behaviors of specific patients have also changed before and after the tumor diagnosis, etc. Because of this situation, it cannot directly indicate the impact of individual irrational drug use behaviors on cancer. MPR is an objective condition of survey, and serves as a reference for compliance, the medication behavior and compliance of individual patient are judged combine with the MARS-5 research questionnaire. Ultimately, the final classifications of irrational drug use behavior, are divided into 5 categories (1 = Insufficient dose; 2 = Not combined (insulin resistance but alone, etc) had a poor effect; 3 = Frequent dressing change; 4 = Poor medication compliance (miss taking medicine); 5 = Incorrect administration timing.), the rest are rational drug use (0 = Rational use of drugs).
Usually, when the ROC curve approaches the upper left corner, the True Positive Rate (TPR) is relatively higher and the False Positive Rate (FPR) is relatively lower, the area is larger, means the model can correctly identify positive class samples while rarely misjudging negative class samples as positive class samples, and has a strong classification ability. In our research, the ROC curve of the total tumor and lung cancer risk prediction model is 0.720 and 0.743 respectively (Supplementary Fig. 1). Between 0.7 and 0.9 indicates that the model has a good discriminatory ability and certain clinical application value, but there is still room for improvement and refinement.
Supplementary Fig. 1.
ROC curve of the malignant risk prediction model secondary to T2DM.a: The malignant tumor risk prediction model on the ROC curve was 0.720 (95% CI: 0.698, 0.742; P < 0.001). The maximum value of the Youden index was 0.341, and the sensitivity and specificity were 80.3% and 53.8%, respectively. b: The area under the ROC curve of lung cancer was notably higher at 0.743 (95% CI: 0.712, 0.775; P < 0.001). The maximum value of the Youden index was 0.384, and the sensitivity and specificity were 80.5% and 57.9%, respectively.
5. Conclusion
Based on at least one year of assessment and research on the cases collected from January 2019 to September 2023. It can be analyzed that: inadequate prescribed doses without self-monitoring of blood sugar; non-combination medication, where a lack of necessary combinations led to suboptimal effects; frequent medication changes, characterized by a high frequency of alterations in 3 monthes; poor medication compliance, defined as missed doses or inconsistent intake; incorrect timing of administration, may pose a significant risk of developing tumors in patients with diabetes, such as: total tumor and lung cancer. However, this kind of risk cannot be predicted by HbA1c ≥ 7.0%, which reminds us of the importance of good medication behavior. It can not only control blood sugar well, delay diseases, but also prevent complications and reduce the risk of tumor.
The following are the supplementary data related to this article.
However, due to the fact that some patients have hoarded drugs or missed doses, and the medication behaviors of specific patients have also changed before and after the tumor diagnosis, etc. Because of this situation, it cannot directly indicate the impact of individual irrational drug use behaviors on cancer. MPR is an objective condition of survey, and serves as a reference for compliance, the medication behavior and compliance of individual patient are judged combine with the MARS-5 research questionnaire. Ultimately, the final classifications of irrational drug use behavior, are divided into 5 categories (1 = Insufficient dose; 2 = Not combined (insulin resistance but alone, etc) had a poor effect; 3=Frequent dressing change; 4= Poor medication compliance (miss taking medicine); 5 = Incorrect administration timing.), the rest are rational drug use (0 = Rational use of drugs) in Supplementary material 1.
The MPR data can be found in Supplementary Table 1. It can be seen that the MPR of the rational drug use group is higher; not combined (insulin resistance but alone, etc) had a poor effect (P < 0.001), frequent dressing change and wrong timing of administration is no difference(P > 0.05).
CRediT authorship contribution statement
Lili Hu: Writing – original draft, Project administration, Investigation, Funding acquisition, Conceptualization. Mengjiao Wang: Writing – original draft, Methodology, Investigation, Formal analysis, Data curation. Haiqing Zhou: Resources, Methodology, Investigation, Formal analysis, Data curation. Tao Wang: Writing – review & editing, Visualization, Supervision, Data curation. Juan Wen: Resources, Investigation, Data curation. Qian Li: Software, Resources, Formal analysis, Data curation. Wei Jin: Resources, Investigation. Nan Zhou: Software, Methodology, Data curation.
Consent for publication
No financial or other relevant conflicts of interest to disclose.
Ethics approval and consent to participate
All procedures performed in studies involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. This study was approved by the ethics committee of the Affiliated Hospital of Xuzhou Medical University, the ethical code is: XYFY2022-KL429–01. The patients were unrelated and all consented.
Funding
This work was supported by the fund project of Xuzhou Science and Technology Social Development Medical and Health General Project (KC22263), the Jiangsu Research Hospital Association Fund (H202041) and Jiangsu Province 2023 drug clinical comprehensive evaluation project (WJ20221507).
Declaration of competing interest
The authors declare that there are no conflicts of interest regarding the publication of this manuscript. No relevant financial relationships could be perceived to influence the research outcomes presented in this manuscript. All authors have contributed to the study design, data collection, analysis, and interpretation, and have approved the final version of the manuscript for submission. If any potential conflicts arise in the future, the authors commit to disclosing them promptly in accordance with ethical guidelines.
Acknowledgments
The authors wish to thank the affiliated hospital of Xuzhou Medical university for their support and assistance in completing this study.
References
- 1.Jain Shelesh, Saraf Swarnlata. Type 2 diabetes mellitus—its global prevalence and therapeutic strategies[J] Diabetes Metab. Syndr. Clin. Res. Rev. 2010;4(1):48–56. doi: 10.1016/j.dsx.2008.04.011. [DOI] [Google Scholar]
- 2.Lovic D., Piperidou A., Zografou I., Grassos H., Pittaras A., Manolis A. The growing epidemic of diabetes mellitus[J] Curr. Vasc. Pharmacol. 2020;18(2):104–109. doi: 10.2174/1570161117666190405165911. [DOI] [PubMed] [Google Scholar]
- 3.Rey-Reñones C., Baena-Díez J.M., Aguilar-Palacio I., Miquel C., Grau M. Type 2 diabetes mellitus and Cancer: epidemiology, physiopathology and prevention[J] Biomedicines. 2021;9(10):1429. doi: 10.3390/biomedicines9101429. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Duan W., Shen X., Lei J., et al. Hyperglycemia, a neglected factor during cancer progression. Biomed. Res. Int. 2014;2014 doi: 10.1155/2014/461917. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Janssen J. Hyperinsulinemia and its pivotal role in aging, obesity, type 2 diabetes, cardiovascular disease and Cancer. Int. J. Mol. Sci. 2021;22(15) doi: 10.3390/ijms22157797. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Zimmet P., Alberti K.G., Magliano D.J., Bennett P.H. Diabetes mellitus statistics on prevalence and mortality: facts and fallacies. Nat. Rev. Endocrinol. 2016;12(10):616–622. doi: 10.1038/nrendo.2016.105. [DOI] [PubMed] [Google Scholar]
- 7.Torre LA, Siegel RL, Ward EM, Jemal A. Global cancer incidence and mortality rates and trends--an update. Cancer Epidemiol Biomarkers Prev 2016;25(1): 16–27. doi: 10.1158/1055-9965.EPI-15-0578. [DOI] [PubMed]
- 8.Tsilidis K.K., Kasimis J.C., Lopez D.S., Ntzani E.E., Ioannidis J.P. Type 2 diabetes and cancer: umbrella review of meta-analyses of observational studies. BMJ. 2015;350 doi: 10.1136/bmj.g7607. [DOI] [PubMed] [Google Scholar]
- 9.Yang L., Shao J., Bian Y., et al. Prevalence of type 2 diabetes mellitus among inland residents in China (2000-2014): a meta-analysis. J Diabetes Investig. 2016;7(6):845–852. doi: 10.1111/jdi.12514. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Li L, Zhang X, Jiang A, et al. Disease burden of lung cancer attributable to metabolic and behavioral risks in China and globally from 1990 to 2021. BMC Public Health. 2025;25(1):911. doi: 10.1186/s12889-025-21941-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Li Y., Pan A., Wang D.D., et al. Impact of healthy lifestyle factors on life expectancies in the US population. Circulation. 2018;138:345–355. doi: 10.1161/CIRCULATIONAHA.117.032047. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Ferroni P, Riondino S, Buonomo O, Palmirotta R, Guadagni F, Roselli M Type 2 diabetes and breast cancer: the interplay between impaired glucose metabolism and oxidant stress. Oxidative Med. Cell. Longev. 2015;2015 doi: 10.1155/2015/183928. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Bjornsdottir HH, Rawshani A, Rawshani A, et al. A national observation study of cancer incidence and mortality risks in type 2 diabetes compared to the background population over time. Sci. Rep. 2020;10(1):17376. doi: 10.1038/s41598-020-73668-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Kang C., LeRoith D., Gallagher E.J. Diabetes, obesity, and breast Cancer. Endocrinology. 2018;159(11):3801–3812. doi: 10.1210/en.2018-00574. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Shlomai G., Neel B., LeRoith D., Gallagher E.J. Type 2 diabetes mellitus and Cancer: the role of pharmacotherapy. J. Clin. Oncol. 2016;34(35):4261–4269. doi: 10.1200/JCO.2016.67.4044. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Siegel RL, Miller KD, Fuchs HE, Jemal A Cancer statistics, 2021. CA Cancer J. Clin. 2021;71(1):7–33. doi: 10.3322/caac.21654. [DOI] [PubMed] [Google Scholar]
- 17.Yilmaz Kavcar SR, Acikgoz A, Ergor G, Unal B. Evaluating the relationship between diabetes and cancer in a cohort. Asian Pac. J. Cancer Prev. 2022;23(4): 1223–1229. doi:10.31557/apjcp.2022.23.4.1223. [DOI] [PMC free article] [PubMed]
- 18.Søndergaard CS, Esquivel PN, Dalamaga M, Magkos F Use of antihyperglycemic drugs and risk of cancer in patients with diabetes. Curr. Oncol. Rep. 2023;25(1):29–40. doi: 10.1007/s11912-022-01344-7. [DOI] [PubMed] [Google Scholar]
- 19.Choe S., Lee J., Park J.W., et al. Prognosis of patients with colorectal Cancer with diabetes according to medication adherence: a population-based cohort study. Cancer Epidemiol. Biomarkers Prev. 2020;29(6):1120–1127. doi: 10.1158/1055-9965.EPI-19-1455. [DOI] [PubMed] [Google Scholar]
- 20.Tan X., Feng X., Chang J., Higa G., Wang L., Leslie D. Oral antidiabetic drug use and associated health outcomes in cancer patients. J. Clin. Pharm. Ther. 2016;41(5):524–531. doi: 10.1111/jcpt.12430. [DOI] [PubMed] [Google Scholar]
- 21.Decensi A., Puntoni M., Goodwin P., et al. Metformin and cancer risk in diabetic patients: a systematic review and meta-analysis. Cancer Prev. Res. (Phila.) 2010;3(11):1451–1461. doi: 10.1158/1940-6207. [DOI] [PubMed] [Google Scholar]
- 22.Chang C.H., Lin J.W., Wu L.C., Lai M.S., Chuang L.M. Oral insulin secretagogues, insulin, and cancer risk in type 2 diabetes mellitus. J. Clin. Endocrinol. Metab. 2012;97(7):E1170–E1175. doi: 10.1210/jc.2012-1162. [DOI] [PubMed] [Google Scholar]
- 23.Dąbrowski M. Diabetes, antidiabetic medications and cancer risk in type 2 diabetes: focus on SGLT-2 inhibitors. Int. J. Mol. Sci. 2021 Feb 7;22(4):1680. doi: 10.3390/ijms22041680. [DOI] [PMC free article] [PubMed]
- 24.The Chinese Working Group on Obesity Guidelines for prevention and control of overweight and obesity in China (excerpts) Acta Nutrimenta Sin. 2004;1:1–4. [Google Scholar]
- 25.Onference of Experts on the Rational Use of Drugs (1985: Nairobi) In: The Rational Use of Drugs: Report of the Conference of Experts, Nairobi, 25–9 November 1985. World Health Organization, editor. 1987. https://iris.who.int/handle/10665/37174 [Google Scholar]
- 26.Standards of medical care in Diabetes-2022 abridged for primary care providers. Clin. Diabetes. 2022;40(1):10–38. doi: 10.2337/cd22-as01. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Giovannucci E., Harlan D.M., Archer M.C., et al. Diabetes and cancer: a consensus report. Diabetes Care. 2010;33(7):1674–1685. doi: 10.2337/dc10-0666. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Tan Y., Wei S., Zhang W., Yang J., Yang J., Yan L. Type 2 diabetes mellitus increases the risk of hepatocellular carcinoma in subjects with chronic hepatitis B virus infection: a meta-analysis and systematic review. Cancer Manag. Res. 2019;11:705–713. doi: 10.2147/cmar.S188238. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Li D., Tang H., Hassan M.M., Holly E.A., Bracci P.M., Silverman D.T. Diabetes and risk of pancreatic cancer: a pooled analysis of three large case-control studies. Cancer Causes Control. 2011;22(2):189–197. doi: 10.1007/s10552-010-9686-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Castro M.J., Jiménez J.M., López M., et al. Impact of preoperative total proteins and glycated hemoglobin on recurrences after early colorectal cancer. Nutrients. 2021;13(2) doi: 10.3390/nu13020711. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Hua Y., Zheng Y., Yao Y., Jia R., Ge S., Zhuang A. Metformin and cancer hallmarks: shedding new lights on therapeutic repurposing. J. Transl. Med. 2023;21(1) doi: 10.1186/s12967-023-04263-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.O’Connor L., Bailey-Whyte M., Bhattacharya M., et al. Association of metformin use and cancer incidence: a systematic review and meta-analysis. J. Natl. Cancer Inst. 2024;116(4):518–529. doi: 10.1093/jnci/djae021. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Liu X.L., Wu H., Zhao L.G., Xu H.L., Zhang W., Xiang Y.B. Association between insulin therapy and risk of liver cancer among diabetics: a meta-analysis of epidemiological studies. Eur. J. Gastroenterol. Hepatol. 2018;30(1):1–8. doi: 10.1097/meg.0000000000001001. [DOI] [PubMed] [Google Scholar]
- 34.Jacobo-Tovar E., Medel-Sánchez A., Durán-Castillo C., Guardado-Mendoza R. Insulin resistance in cancer risk and prognosis. Semin. Cancer Biol. 2025;114:73–87. doi: 10.1016/j.semcancer.2025.06.006. [DOI] [PubMed] [Google Scholar]
- 35.Burnol A.F., Morzyglod L., Popineau L. Dialogue entre les voies de signalisation de l’insuline et les voies de prolifération cellulaire [Cross-talk between insulin signaling and cell proliferation pathways] Ann Endocrinol (Paris) 2013;74(2):74–78. doi: 10.1016/j.ando.2013.02.003. [DOI] [PubMed] [Google Scholar]
- 36.Chiefari E., Mirabelli M., La Vignera S., et al. Insulin resistance and cancer: in search for a causal link. Int. J. Mol. Sci. 2021;22(20) doi: 10.3390/ijms222011137. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37.Cao M., Li H., Sun D., Chen W. Cancer burden of major cancers in China: a need for sustainable actions. Cancer Commun. (Lond.) 2020;40(5):205–210. doi: 10.1002/cac2.12025. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Ikemura M., Nishikawa M., Kusamori K., Fukuoka M., Yamashita F., Hashida M. Pivotal role of oxidative stress in tumor metastasis under diabetic conditions in mice. J. Control. Release. 2013;170(2):191–197. doi: 10.1016/j.jconrel.2013.05.028. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
However, due to the fact that some patients have hoarded drugs or missed doses, and the medication behaviors of specific patients have also changed before and after the tumor diagnosis, etc. Because of this situation, it cannot directly indicate the impact of individual irrational drug use behaviors on cancer. MPR is an objective condition of survey, and serves as a reference for compliance, the medication behavior and compliance of individual patient are judged combine with the MARS-5 research questionnaire. Ultimately, the final classifications of irrational drug use behavior, are divided into 5 categories (1 = Insufficient dose; 2 = Not combined (insulin resistance but alone, etc) had a poor effect; 3=Frequent dressing change; 4= Poor medication compliance (miss taking medicine); 5 = Incorrect administration timing.), the rest are rational drug use (0 = Rational use of drugs) in Supplementary material 1.
The MPR data can be found in Supplementary Table 1. It can be seen that the MPR of the rational drug use group is higher; not combined (insulin resistance but alone, etc) had a poor effect (P < 0.001), frequent dressing change and wrong timing of administration is no difference(P > 0.05).