Pharmacological interventions for smoking cessation in Type 2 diabetes: A systematic review with meta-analysis and GRADE evaluation

Review Pharmacological interventions for smoking cessation in Type 2 diabetes: A systematic review with meta-analysis and GRADE evaluation Alison Martin a , Giusy Rita Maria La Rosa b,* , Hannah Rice a , Andrea Bertuzzi a , Michal Witkowski a , Erika Anastasi c , Giulio Geraci d , Riccardo Polosa b,d,e , the DiaSmokeFree Working Group 1 a Crystallise, Stanford-le-Hope, UK b Department of Clinical and Experimental Medicine, University of Catania, Catania, Italy c ECLAT Srl, Spin-off of the University of Catania, Catania, Italy d Faculty of Medicine and Surgery, “Kore” University of Enna, Enna, Italy e Center of Excellence for the Acceleration of HArm Reduction (CoEHAR), University of Catania, Italy ARTICLE INFO Keywords: Bupropion GRADE Meta-analysis Nicotine replacement therapy Smoking cessation Systematic review Type 2 diabetes mellitus T2DM Varenicline ABSTRACT Aims: To evaluate the efficacy and safety of pharmacological therapies for smoking cessation in individuals with Type 2 Diabetes Mellitus (T2DM) through a systematic review, meta-analysis, and GRADE evaluation. Methods: PubMed and Scopus were searched on 7 June 2024 using relevant keywords. Randomized controlled trials and prospective cohort studies involving adult smokers with T2DM receiving pharmacological treatments for smoking cessation were included. Data were extracted independently by two reviewers. Random-effects metaanalyses were conducted, and the certainty of evidence was assessed using GRADE. Results: Eighteen studies (19 publications) were included. Pharmacotherapy significantly increased continuous abstinence rates at 12 and 24 weeks (p <0.001), with ORs of 4.17 (95 % CI: 2.71–6.42) and 3.80 (95 % CI: 2.52–5.72), respectively. At 52 weeks, varenicline was more effective than placebo (OR: 2.84, 95 % CI: 1.41–5.69, p =0.003). Adverse events were more frequent with varenicline, but not significantly (OR: 1.40, 95 % CI: 0.98–1.98, p =0.06). Conclusions: Varenicline appears effective for smoking cessation in T2DM, with an acceptable tolerability profile. Bupropion and NRT show potential efficacy. However, most evidence comes from post hoc analyses in which diabetes was not a predefined variable, warranting cautious interpretation. 1. Introduction Diabetes mellitus (DM) is characterized by elevated blood sugar levels caused by defects in insulin secretion, action, or both [1]. As of 2021, it affects 537 million adults worldwide, with the International Diabetes Federation projecting a 46 % increase in cases by 2045 [2]. The most common form, type 2 diabetes mellitus (T2DM), arises from insulin resistance and a gradual decline in β-cell insulin production [3]. Chronic hyperglycemia in T2DM causes vascular damage, leading to macrovascular and microvascular complications. Given that cigarette smoke exposure similarly contributes to vascular damage, endothelial dysfunction, and clotting activation [4–6], it is not surprising that the combined harmful effects of high blood glucose and cigarette smoke likely accelerate vascular complications in patients with diabetes who smoke. Indeed, research has shown that cigarette smoking increases the risk of both microvascular and macrovascular complications, as well as mortality, in patients with T2DM [7,8]. In line with these findings, smoking cessation dramatically reduces this excess risk [8–10]. While limiting exposure to cigarette smoke is a public health priority, it is even more critical for smokers with T2DM. In the general population, smoking quit rates can be doubled or tripled when combining smoking cessation counseling with pharmaceutical interventions including varenicline [11], bupropion [12], nicotine replacement therapy (NRT) [13], and cytisine [14]. However, the efficacy and tolerability of these treatments in individuals with T2DM remain under-researched, and for some interventions, such as * Corresponding author. E-mail address: [email protected] (G.R.M. La Rosa). 1 Members of the DiaSmokeFree Working Group are listed below. Contents lists available at ScienceDirect Diabetes Research and Clinical Practice journal homepage: www.journals.elsevier.com/diabetes-research-and-clinical-practice https://doi.org/10.1016/j.diabres.2025.112202 Received 21 December 2024; Received in revised form 15 April 2025; Accepted 23 April 2025 d i a b e t e s r e s e a r c h a n d c l i n i c a l p r a c t i c e 224 (2025) 112202 Available online 26 April 2025 0168-8227/© 2025 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license ( http://creativecommons.org/licenses/by/4.0/ ). cytisine, efficacy data specifically in patients with diabetes are lacking. To address this gap and support the development of evidence-based treatment guidelines for improving patients’ outcomes, we conducted a systematic review, meta-analysis, and GRADE evaluation of the available evidence. This approach provides a structured method for assessing the quality of evidence and the strength of recommendations for individuals with T2DM who smoke. 2. Methods The present review was reported following the Preferred Reported Items for Systematic Reviews and Meta-analyses (PRISMA) guidelines. The protocol is registered in PROSPERO (CRD42024572377). 2.1. PICO (Population, Intervention, Comparator, Outcome) The research strategy was based on the PICO approach: Population (P)—adult patients (≥18 years) with T2DM who smoked at baseline; Intervention (I)— pharmacological therapy for smoking cessation (e. g., varenicline, bupropion, NRT); Comparison (C)— placebo, usual care, and counselling; Outcome (O)— Primary: Continuous abstinence rate (CAR) at week 12, 24 and 52; any adverse event. Secondary: Smoking abstinence (7day point prevalence), weight gain post-cessation, and withdrawals due to treatment. 2.2. Search strategy We conducted searches in PubMed and Scopus databases on 7 June 2024, using keywords: “diabetes mellitus, type 2”, “smoking cessation”, “abstinence”, “varenicline”, “nicotine replacement therapy”, “bupropion”. The search syntax is in the Table S1. There were no exclusions by language or time. All included studies were reference checked and citation chased in Google Scholar. Expert recommendations were requested. 2.3. Inclusion criteria Studies were included if they satisfied the following criteria: 1) Randomized controlled studies (RCTs) and prospective cohort studies or meta-analyses of such studies; 2) studies conducted among adult smokers with T2DM or reporting mixed results on both T1DM and T2DM; 3) pharmacological treatment for smoking cessation administered as a standalone intervention; 4) data on control group (including placebo and usual care). 2.4. Exclusion criteria Studies were excluded if they were cross-sectional, case-control, retrospective cohort, non-systematic reviews, editorials, and comments; if data from patients with diabetes were mixed with those from other conditions; and if pharmacological treatment for smoking cessation was provided as optional medication. 2.5. Study selection and data extraction Two reviewers independently screened titles and abstracts to determine full-text assessment needs. Disagreements were resolved by a third reviewer. Full-text articles were then evaluated, and data were entered into a predefined table. The data extraction table captured information on: (1) trial methodology, setting, and duration of follow-up; (2) population characteristics; (3) type of intervention; and (4) analyses and outcomes. 2.6. Risk of bias assessment The risk of bias of outcomes from the included studies was assessed using the revised version of the Cochrane risk of bias tool, RoB 2. The judgments may indicate a risk of bias as “Low”, “High”, or “Some concerns“. 2.7. Meta-analysis A meta-analysis was performed using the ’meta’ package in R, employing the metagen function with the calculated logOR and SE (logOR) as inputs. The meta-analysis was conducted using the randomeffects models, and the results were presented as ORs with 95 % CIs. Heterogeneity was assessed using the I2 statistic, with values over 50 % indicating moderate heterogeneity. Subgroup analyses explored population or treatment differences, while sensitivity analyses assessed the impact of high-risk studies. All analyses were performed using R software (version 4.2.1) and the ’meta’ package (version 7.0). Data extraction involved extracting the published odds ratios (OR) along with their confidence intervals (CIs), as well as the number of patients with the outcome and the total number of patients in both the intervention and comparator groups. For studies where the published OR was not available, the OR was calculated from the n/N data. 2.8. GRADE assessment Using all relevant publications found in the systematic search, a GRADE analysis was conducted in accordance with the Cochrane Guidelines (Supplementary Appendix 1). The GRADE assessment is made for each outcome reported for each treatment comparison. The overall GRADE assessment then informs the determination of the strength of the recommendation in favour of the intervention. 3. Results A total of 540 citations were identified through electronic searches, with 19 additional publications found via citation tracking. After removing duplicates, 542 citations were screened, and 498 were deemed ineligible based on titles and abstracts. Two articles were unavailable. The remaining 42 articles were reviewed, with 23 excluded; the list of excluded studies with reason for exclusion is in Table S2. Ultimately, this leaves 18 studies (19 publications) for inclusion (Fig. 1). Of these, 14 were RCTs from Tonstad’s retrospective meta-analysis [15], though none specifically focused on individuals with diabetes. These RCTs were included for bias assessment, with further analyses referring to Tonstad’s study [15]. Two studies from the EAGLES RCT covered overlapping populations [16,17]. The main characteristics of the included publications are detailed in Table 1. 3.1. Study characteristics One study was conducted in Europe [18], one in Asia [19], and the EAGLES RCT spanned 16 countries [16,17]. Tonstad’s retrospective analysis pooled 15 randomized, placebo-controlled trials across North and South America, Europe, Asia-Pacific, Africa. The follow-up duration across these studies varied from a minimum of 12 weeks to a maximum of 52 weeks [15,18] (Table S3). A total of 1,346 adult smokers with diabetes participated, with T2DM specifically noted in three studies [15,18,19]. All studies included both male and female participants, with the EAGLES RCT including smokers with or without psychiatric diagnoses. Tonstad [15] and Russo [18] assessed varenicline’s efficacy compared to placebo. In the EAGLES RCT, Tønnesen [17] and Rojewski [16] stratified participants by active treatment (i.e., varenicline, bupropion, or NRT) and compared them with placebo. Similarly, A. Martin et al. Diabetes Research and Clinical Practice 224 (2025) 112202 2 Huang’s study [19] compared varenicline, bupropion, and NRT. All studies provided standardized individual counseling to all participants. Continuous abstinence rates was reported at 9–12, 9–24, and 9–52 weeks, verified by breath carbon monoxide (CO) measurements [15–18]. CAR provides a robust measure of long-term cessation success, offering a more accurate reflection of sustained behavior change. One study [19] assessed smoking abstinence at weeks 12 and 24 using selfreported data from the Fagerstr¨ om Test for Nicotine Dependence (FTND) questionnaire. Additionally, Russo’s RCT measured the 7-day point prevalence of abstinence at weeks 12, 24, and 52 [18]. While the 7-day point prevalence is easy to measure and provides immediate data at specific time points, it may not be as robust as CAR and often leads to overestimation of success rates. Safety was investigated in four studies [15–18], weight changes in two [15,18], and only Russo reported withdrawals due to treatmentrelated adverse events [18]. 3.2. Meta-analysis 3.2.1. Study selection A total of 5 studies were deemed suitable for inclusion in the metaanalysis [15–19], of which only 2 contained overlapping data from the EAGLES trial [16,17]. Both studies reported continuous abstinence rates (CAR) at weeks 12 and 24 by treatment arm in patients with diabetes. As the Rojewski’s study included only the 282 participants recruited from the United States [16], while the Tønnesen’s study encompassed the full diabetic subgroup (n =408), we opted to include only the Tønnesen’s data in the primary analysis to avoid duplication and ensure broader coverage [17]. Additionally, one study identified as having a high risk of bias was excluded from the primary analysis but retained for sensitivity analysis [19]. Inclusion of its outcome data (CAR at 24 weeks) in the sensitivity analysis did not alter the overall conclusions of the metaanalysis. Finally, 3 studies [15,17,18] were included in the primary analysis. 3.2.2. Continuous abstinence rate (CAR), week 12 The meta-analysis of CAR at week 12 included 3 studies [15,17,18], with 5 treatment comparisons. The random effects model yielded an OR of 4.17 (95 % CI: 2.71, 6.42), indicating that smoking cessation pharmacotherapy (i.e., varenicline, bupropion, and NRT) was significantly more effective than the comparator (p <0.0001). Heterogeneity statistics was moderate (I2 value of 52.5 %), suggesting some variability in the effect sizes among the included studies (Fig. 2). Varenicline, bupropion, and NRT all provided a similar significant benefit compared to controls. 3.2.3. Continuous abstinence rate (CAR), week 24 The meta-analysis of CAR at week 24 included 3 studies [15,17,18], with 3 treatment comparisons. The random effects model showed an OR of 3.80 (95 % CI: 2.52, 5.72), indicating a similar significant benefit of the interventions compared to controls. (p <0.0001) with low to moderate heterogeneity (I2 value of 22.9 %) (Fig. 3). 3.2.4. Sensitivity analysis Sensitivity analysis (4 studies, 5 comparisons) [15,17–19] confirmed the results with an OR of 3.68 (95 % CI: 2.48, 5.45) indicating a significant benefit of the intervention compared to the comparator (p < 0.0001) and similar low to moderate heterogeneity (I2 =21.1 %) (Fig. 4). At week 24, both primary and sensitivity analyses show that smoking cessation pharmacotherapy significantly increases the odds of continuous abstinence compared to the comparator. Consistent ORs and the inclusion of additional studies in the sensitivity analysis confirm the robustness of the primary analysis results. The low to moderate heterogeneity suggests that the effect size is fairly consistent across studies, further supporting the efficacy of the intervention. Fig. 1. Flow chart of review process. A. Martin et al. Diabetes Research and Clinical Practice 224 (2025) 112202 3 Table 1 Main characteristics of the included publications. Author and Year Setting Sample Size (n) Study Duration Study Followup Age, years mean (SD) Men, n (%) Type 1 Diabetes, n(%) Type 2 Diabetes, n(%) Intervention Control Primary Outcome Huang et al., 2023 Hospital, Taiwan 32 6 months 12th and 24th weeks 56.0 (10.62) 56 (82.4 %) −68 (100 %) 2 month varenicline use plus 4 month coaching Smoking cessation services alone or combined with 2 month varenicline Smoking abstinence (FTND) and smoking reduction (participants’ selfreported number of cigarettes smoked per day) Rojewski et al., 2024 Clinical trial centres, academic centres, and outpatient clinics in 16 countries (EAGLES RCT) 282 12 weeks 12th and 24th weeks NR NR NR Varenicline Bupropion NRT patch Placebo Smoking abstinence (CAR) Russo et al., 2022 Hospital, Italy 300 12 weeks 12th, 24th, and 52nd weeks 57.4 (0.8) 236 (78.66 %) −300 Varenicline (0.5 mg daily for 2–3 days, 0.5 mg twice daily for 4 to 5 days, and 1 mg twice daily for 11 weeks) Placebo Smoking abstinence (CAR) Tønnesen et al., 2022 Clinical trial centres, academic centres, and outpatient clinics in 16 countries (EAGLES RCT) 409 12 weeks 12th and 24th weeks 54.8 (9.4) 191 (46.7) NR NR Varenicline (1 mg twice daily) Bupropion (150 mg twice daily) NRT (patch 21 mg/day) Placebo Smoking abstinence (CAR) Tonstad et al., 2017 Retrospective pooled analysis of 15 RCTs from all continents 323 12 weeks 12th, 24th, and 52th weeks 55.6 (9.35) 242 (74.9) 12 311 Varenicline 1 mg b.i.d Placebo Smoking abstinence (CAR) Legend. b.i.d: bis in die (two times a day); CAR: Continuous Abstinence Rate; FTND: Fagerstr¨ om Test for Nicotine Dependence; NR: Not Reported; NRT: Nicotine Replacement Therapy; RCT: Randomized Controlled Trial. Fig. 2. Forest plot −CAR at 12 weeks. A. Martin et al. Diabetes Research and Clinical Practice 224 (2025) 112202 4 3.2.5. Continuous abstinence rate (CAR), week 52 At week 52, meta-analysis of 2 studies [15,18] with 1 treatment comparison yielded an OR of 2.84 (95 % CI: 1.41, 5.69), showing that varenicline was significantly more effective than the comparator (p = 0.003) with low to moderate heterogeneity (I2 =31.8 %) (Fig. 5). No data were available for bupropion or NRT at this time point. 3.2.6. All adverse events (AE_ALL) The meta-analysis of AE_ALL included 2 studies [15,18], with one treatment comparison. The random effects model showed an OR of 1.40 (95 % CI: 0.98, 1.98), indicating a non-statistical trend towards more adverse events in the varenicline group (p =0.06). Heterogeneity was 0 %, indicating no variation among studies (Fig. 6, Table S4). No data were available for bupropion or NRT. 3.3. Secondary outcomes 3.3.1. Smoking abstinence (7-day point prevalence) The Russo’s study (2022) [18] reported significantly higher smoking abstinence rates in the varenicline group versus placebo at all time points. At week 12, abstinence was 40.0 % vs. 11.0 % (OR 7.67, 95 % CI 3.91–15.05; p <0.001). By week 24, it was 29.0 % vs. 8.3 % (OR 4.44, 95 % CI 2.17–9.07; p <0.001), and at week 52, 23.7 % vs. 9.5 % (OR 3.27, 95 % CI 1.57–6.78; P <0.001). 3.3.2. Weight change In the 2 studies [15,18], weight changes in patients with diabetes treated with varenicline and placebo were consistent. In the Tonstad’s study [15], patients who quit smoking saw identical weight increases of 1.7 kg in both groups, with BMI increases of 0.56 kg/m2 (varenicline) and 0.54 kg/m2 (placebo). Among non-quitters, varenicline users gained slightly more weight (1.3 kg) than placebo users (0.5 kg), with corresponding BMI increases of 0.44 kg/m2 and 0.18 kg/m2, respectively. The Russo’s study [18] showed no significant differences in BMI or waist circumference between groups, supporting the overall finding of minimal weight variation across treatments. 3.3.3. Withdrawals due to treatment In Russo’s study (2022) [18], discontinuation due to adverse events occurred in 4.0 % of patients in the varenicline group (6 cases) and 3.3 % in the placebo group (5 cases). 3.4. Risk of bias Appraisal Table S5 summarizes the risk of bias of the included studies. 3.5. GRADE evaluation of smoking cessation in type 2 diabetes The systematic review identified 6 publications that reported relevant data on separate clinical trials of pharmacological therapies for smoking cessation in people with T2DM. Of these: •Five reported outcome data for a relevant pharmaceutical intervention and comparator in patients with diabetes and were included in the GRADE analysis [15–19]. Two of these reported on different but overlapping populations from the same EAGLES RCT, so have been assessed as one study for the GRADE analysis [16,17]. •One was irretrievable [20]. One of the five included publications was a meta-analysis of clinical trials comparing varenicline to placebo in the general population, with individual patient data for participants with diabetes provided by the study sponsor [15]. We retrieved 14 of the 15 primary publications from the meta-analysis (one was irretrievable). None reported outcomes for the diabetes subgroup, so we included the meta-analysis but assessed the individual studies’ risk of bias. While the primary studies had low bias, the meta-analysis only reported the overall odds ratio for cessation, not study-specific outcomes. In some cases, the primary RCTs reported slightly higher recruitment numbers than the meta-analysis, and one paper reported fewer participants with diabetes than the meta-analysis (47/355 in the varenicline group, 60/359 in the placebo group in the primary study [21]; compared with 58/295 patients in the varenicline group and 68/282 of the placebo group reported to have diabetes in the meta-analysis [15]). There is therefore some uncertainty about how far the results of the meta-analysis can be trusted to be accurate. Using all relevant publications found in the systematic search, a GRADE analysis was conducted in accordance with the Cochrane Guidelines [22]. The outcomes of interest for which there was relevant data were continuous abstinence and adverse event rates. A summary of the GRADE analysis is reported in the Table 2. Fig. 3. Forest plot −CAR at 24 weeks −Primary Analysis. A. Martin et al. Diabetes Research and Clinical Practice 224 (2025) 112202 5 3.6. Adverse events Three studies reported adverse events with varenicline versus placebo: 2 RCTs [17,18] and one meta-analysis of 15 RCTs [15]. Commonly reported adverse events include nausea, headache, and insomnia. Although no statistical analysis was performed, nausea and abnormal dreams occurred in at least 5 % more patients on varenicline compared to placebo. All other adverse events were either reported by only one study or showed less than a 5 % difference between treatment arms (Tables S6, S7). 3.7. Other outcomes In addition to the GRADE analysis outcomes, some publications assessed other outcomes, such as the proportion of participants with < 50 % or ≥50 % smoking reduction, all adverse events, point abstinence rates, and adverse events leading to treatment discontinuation. These were excluded from the GRADE analysis as they appeared in only one high-risk publication, with no comparable results available (Supplementary Appendix 2). 4. Discussion This is the first systematic review and meta-analysis on pharmacological treatments for smoking cessation in patients with T2DM, showing significant benefits of pharmacotherapy on CAR. Tolerability and weight changes were also assessed, with no significant differences between varenicline and placebo. Specifically, the results of this meta-analysis provide robust evidence supporting the efficacy of varenicline as a smoking cessation aid in individuals with T2DM. Pooled data from three key studies [15,17,18] demonstrated that varenicline significantly increased CAR at 12, 24, and Fig. 4. Forest plot −CAR at 24 weeks −Sensitivity Analysis. Fig. 5. Forest plot −CAR at 52 weeks. A. Martin et al. Diabetes Research and Clinical Practice 224 (2025) 112202 6 52 weeks compared to placebo, with odds ratios of 4.17, 3.80, and 2.84, respectively. These findings highlight varenicline’s potential for sustained smoking cessation, crucial for reducing cardiovascular and other smoking-related risks in patients with diabetes [23]. The week 12 results are particularly noteworthy, showing that pharmacotherapy more than quadruples the odds of sustained abstinence during the early, critical phase of cessation, when the risk of relapse is highest [24,25]. The moderate heterogeneity indicates some variability between studies, but the overall effect of pharmacotherapy is consistently positive. By week 24, the benefits of varenicline remained significant consistent with its mechanism of action in targeting nicotinic acetylcholine receptors to reduce cravings and withdrawal symptoms [26]. Low to moderate heterogeneity across studies of all three pharmacotherapies further supports the reliability of these findings. The sensitivity analysis, which included the Huang’s study (2023) [19], confirmed the robustness of the primary outcomes, with similar odds ratios and low heterogeneity, reinforcing the generalizability of the data. By week 52, varenicline’s effect on sustained abstinence, though slightly reduced, remained statistically significant, with no data for bupropion or NRT. This prolonged effect is beneficial for individuals with diabetes, as sustained cessation reduces the risk of cardiovascular events and improved glycemic control over time [8,27]. Low heterogeneity at one year supports varenicline’s consistent efficacy in maintaining cessation. However, declining odds ratios suggest reduced longterm efficacy, possibly due to relapse or rising quit rates in the control group. Over time, behavioral and psychological factors—such as stress, triggers, and habits—become more prominent, increasing the risk of relapse [28]. Smoking relapse is particularly common among people with depression, a condition frequently associated with patients suffering from T2DM [29,30]. The results of this meta-analysis provide limited evidence for NRT and bupropion, as they primarily derive from a single study (EAGLES) [31], reducing confidence in the consistency of these findings. Of note, their effect sizes and confidence intervals are similar to that of Fig. 6. Forest plot −All Adverse Events. Table 2 GRADE analysis for the outcome of continuous abstinence rate (CAR). CAR at week 12 Comparison Initial score Conclusion ROB Consistency Indirectness Imprecision Effect size Dose effect Confounding Final score Bupropion vs placebo 4Bupropion more effective No change No change No change ¡1No change No change No change 3 Varenicline vs placebo 4Varenicline more effective No change No change No change ¡1þ1No change No change 4 NRT vs placebo 4NRT more effective No change No change No change ¡1No change No change No change 3 Varenicline vs bupropion 4Varenicline more effective No change No change No change ¡1No change No change No change 3 Varenicline vs NRT 4No significant difference No change No change No change ¡1No change No change No change 3 Bupropion vs NRT 4No significant difference No change No change No change ¡1No change No change No change 3 CAR at 24 weeks Bupropion vs placebo 4Bupropion more effective No change No change No change ¡1No change No change No change 3 Varenicline vs placebo 4Varenicline more effective No change No change No change ¡1þ1No change No change 4 NRT vs placebo 4NRT more effective No change No change No change ¡1No change No change No change 3 Varenicline vs bupropion 4Varenicline more effective No change No change No change ¡1No change No change No change 3 Varenicline vs NRT 4No significant difference No change No change No change ¡1No change No change No change 3 Bupropion vs NRT 4No significant difference No change No change No change ¡1No change No change No change 3 CAR at 52 weeks Varenicline vs placebo 4Varenicline more effective No change No change No change ¡1þ1No change No change 4 A. Martin et al. Diabetes Research and Clinical Practice 224 (2025) 112202 7 varenicline, which supports the hypothesis that they may also be beneficial. Further studies are needed to confirm the efficacy of these drugs in patients with T2DM. The pooled odds ratio for adverse events with varenicline was 1.40 (95 % CI: 0.98–1.98; p =0.06), a finding that approaches statistical significance and warrants careful interpretation. This consideration is especially important in patients with chronic conditions such as T2DM, where even mild side effects may influence adherence and overall treatment acceptability. Nonetheless, treatment discontinuation rates were not markedly elevated, suggesting that varenicline remained generally acceptable in terms of tolerability and adherence in this population [18]. Moreover, common adverse events, such as nausea and abnormal dreams, were mild and aligned with previous findings in the general population [15]. Additionally, a recent review found no clinically significant interactions between smoking cessation treatments and antidiabetic medications [32]. The inclusion of weight change data from the Tonstad’s [15] and Russo’s [18] studies sheds light on the broader implications of smoking cessation for individuals with diabetes, many of whom are overweight or obese. Weight gain after quitting smoking is a common concern among people with diabetes [33,34], as it can worsen glycemic control and insulin resistance, potentially reducing the benefits of smoking cessation for diabetes and cardiovascular health. Notably, Pani et al. [35] identified weight gain as a predictor of worsening diabetic control. Furthermore, insulin resistance after quitting significantly increased in weight gainers but not in weight maintainers [36]. However, the lack of significant post-cessation weight gain with varenicline is reassuring, suggesting that the metabolic and cardiovascular benefits of quitting smoking for diabetes management are not compromised by weight gain. Considering the significant efficacy and tolerability of varenicline in individuals with T2DM, healthcare professionals can confidently recommend it as a first-line pharmacotherapy for smoking cessation in motivated patients. However, long-term monitoring and support are essential to sustain abstinence after the initial intervention. Additionally, healthcare providers can reassure patients that quitting smoking with varenicline is unlikely to result in significant weight gain, thereby promoting treatment adherence. Despite strong evidence for varenicline, this analysis has limitations. While several high-quality RCTs included data on patients with T2DM, the Tonstad’s meta-analysis [15] used unpublished data from 14 RCTs that did not report outcomes specifically for patients with diabetes. Additionally, these RCTs did not stratify by diabetes status, which may affect the accuracy of varenicline’s efficacy and safety in this subgroup. Our systematic review may have missed studies that reported diabetesspecific outcomes only in full texts. However, even with a broader search, potential issues with randomization not stratified by diabetes status could undermine subgroup analysis reliability. 5. Concluding remarks This meta-analysis shows varenicline as an effective smoking cessation aid for people with T2DM, with sustained abstinence up to 52 weeks and an acceptable tolerability profile, supporting its long-term benefits. Varenicline, combined with counseling, is strongly recommended due to significant reductions in smoking, despite more frequent but mild side effects like nausea and abnormal dreams. Nicotine replacement therapy and bupropion receive a moderate recommendation, limiting confidence in results. While varenicline is supported by multiple studies and longer-term follow-up data, the evidence for bupropion and NRT is primarily drawn from a single trial (EAGLES). As such, recommendations for these interventions should be viewed as preliminary and interpreted with caution until additional supporting data become available. It should be noted that most of the included trials did not stratify participants by diabetes status at the time of randomization. Consequently, conclusions regarding efficacy and safety in individuals with T2DM are derived from subgroup analyses or pooled data, which may limit the direct applicability of the findings. This limitation underscores the need for future studies specifically designed to evaluate outcomes in this population. Strengthening global diabetes education programs to support smoking cessation and exploring alternatives for those struggling to quit is essential. 5.1. Future Directions Pharmacotherapies such as cytisine, along with complete substitution of tobacco cigarettes with combustion-free alternatives like e-cigarettes and heated tobacco products, are currently under investigation for smoking cessation and may offer potential benefits for individuals with T2DM. Cytisine, a partial agonist of the α 4β2 nicotinic receptor—similar to varenicline—has demonstrated comparable 6-month abstinence rates in the general population [14]. However, data specifically evaluating its efficacy and safety in individuals with diabetes are currently lacking. Similarly, e-cigarettes have shown comparable effectiveness to varenicline in recent network meta-analyses [39] and may offer a less harmful alternative for nicotine delivery by reducing exposure to toxicants [37,38]. Although these therapies were not included in the present review, they merit further investigation in diabetes-specific populations. In this context, the ongoing DiaSmoke trial [40] is expected to provide valuable insights into the role of combustion-free alternatives like e-cigarettes and heated tobacco products for smoking cessation in individuals with T2DM. CRediT authorship contribution statement Alison Martin: Writing – review & editing, Writing – original draft, Validation, Supervision, Software, Methodology, Formal analysis, Data curation. Giusy Rita Maria La Rosa: Writing – review & editing, Writing – original draft, Validation, Methodology, Investigation, Data curation, Conceptualization. Hannah Rice: Writing – original draft, Visualization, Validation, Software, Methodology, Formal analysis, Data curation. Andrea Bertuzzi: Writing – original draft, Visualization, Validation, Software, Formal analysis, Data curation. Michal Witkowski: Writing – original draft, Visualization, Software, Resources, Formal analysis, Data curation. Erika Anastasi: Writing – original draft, Visualization, Validation, Resources, Data curation. Giulio Geraci: Writing – review & editing, Visualization, Validation, Data curation. Riccardo Polosa: Writing – review & editing, Writing – original draft, Supervision, Resources, Project administration, Methodology, Funding acquisition, Conceptualization. Funding This research received a grant from Department of Clinical and Experimental Medicine, University of Catania (UPB: 6C725202048/ 2024) and support from ECLAT Srl., a spin-off of the University of Catania. The study sponsor/funder was not involved in the design of the study; the collection, analysis, and interpretation of data; writing the report; and did not impose any restrictions regarding the publication of the report. The authors are solely responsible for the content, selection, and presentation of facts, as well as any opinions expressed in this study. Contribution Statement. R.P. and the DiaSmokeFree Working Group contributed to the design concept and manuscript writing. A.M., H.R., and A.B. contributed to the statistical analysis and wrote the draft. G.R.M.L.R., M.W., E.A., and G.G. were involved in acquisition of data. G. R.M.L.R. performed the search strategy and contributed to database research and manuscript writing. All authors participated in the study concept, reviewed the manuscript, and read and approved the final manuscript. A. Martin et al. Diabetes Research and Clinical Practice 224 (2025) 112202 8 Declaration of competing interest The authors declare the following financial interests/personal relationships which may be considered as potential competing interests: [Giusy Rita Maria La Rosa is a research fellow at University of Catania. She declares no conflict of interest. Alison Martin, Hannah Rice, Andrea Bertuzzi, Michal Witkowski, Erika Anastasi, Giulio Geraci, Agostino Di Ciaula, Tabinda Dugal, Syed Abbas Raza, Anoop Misra, Andre Pascal Kengne, Cristina Russo, Noel Somasundaram, Phuong Le Dinh, and Roberta Sammut declare no conflict of interest. Magdalena Walicka has received honoraria for lectures from Sanofi, AstraZeneca, Teva, Eli Lilly, and has received support for attending meetings from Sanofi, Novo Nordisk, and Berlin Chemie. Riccardo Polosa is full tenured professor of Internal Medicine at the University of Catania (Italy) and Medical Director of the Institute for Internal Medicine and Clinical Immunology at the same University. He has received grants from U-BIOPRED and AIRPROM, Integral Rheumatology & Immunology Specialists Network (IRIS), Foundation for a Global Action to End Smoking (formerly known as Foundation for Smoke-Free World), Pfizer, GlaxoSmithKline, CV Therapeutics, NeuroSearch A/S, Sandoz, Merk Sharp & Dohme, Boehringer Ingelheim, Novartis, Arbi Group Srl., Duska Therapeutics, Forest Laboratories, Ministero dell Universita’ e della Ricerca (MUR) Bando PNRR 3277/2021 (CUP E63C22000900006) and 341/2022 (CUP E63C22002080006), funded by NextGenerationEU of the European Union (EU), and the ministerial grant PON REACT-EU 2021 GREENBando 3411/2021 by Ministero dell Universita’ e (MUR) – PNRR EU Community. He is founder of the Center for Tobacco Prevention and Treatment (CPCT) at the University of Catania and of the Center of Excellence for the Acceleration of Harm Reduction at the same university. He receives consultancy fees from Pfizer, Boehringer Ingelheim, Duska Therapeutics, Forest Laboratories, CV Therapeutics, Sermo Inc., GRG Health, Clarivate Analytics, Guidepoint Expert Network, and GLG Group. He receives textbooks royalties from Elsevier. He is also involved in a patent application for ECLAT Srl. He is a pro bono scientific advisor for Lega Italiana Anti Fumo (LIAF) and the International Network of Nicotine Consumers Organizations (INNCO); and he is Chair of the European Technical Committee for Standardization on “Requirements and test methods for emissions of electronic cigarettes” (CEN/TC 437; WG4); and scientific advisor of the non-profit Foundation RIDE2Med]. Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi. org/10.1016/j.diabres.2025.112202. Data availability The data that support the findings of this study are available on request from the corresponding author, A.M. References [1] American Diabetes Association. Diagnosis and Classification of Diabetes Mellitus (2014) Diabetes Care 37(Supplement_1):S81–90. doi: 10.2337/dc14-S081. [2] International Diabetes Federation. IDF Diabetes Atlas [Internet]. 2021 [cited 2024 Aug 28]. Available from: https://diabetesatlas.org/. [3] American Diabetes Association Professional Practice Committee, ElSayed NA, Aleppo G, Bannuru RR, Bruemmer D, Collins BS, et al. (2024) 2. Diagnosis and Classification of Diabetes: Standards of Care in Diabetes—2024. Diabetes Care 47 (Supplement_1):S20–42. doi: 10.2337/dc24-S002. [4] Caponnetto P, Russo C, Di Maria A, Morjaria JB, Barton S, Guarino F, et al. Circulating endothelial-coagulative activation markers after smoking cessation: a 12-month observational study. Eur J Clin Investigation 2011;41(6):616–26. https://doi.org/10.1111/j.1365-2362.2010.02449.x. [5] Guarino F, Cantarella G, Caruso M, Russo C, Mancuso S, Arcidiacono G, et al. Endothelial activation and injury by cigarette smoke exposure. J Biol Regul Homeost Agents 2011;25(2):259–68. [6] Cacciola RR, Guarino F, Polosa R. Relevance of endothelial-haemostatic dysfunction in cigarette smoking. CMC 2007;14(17):1887–92. https://doi.org/ 10.2174/092986707781058832. [7] Campagna D, Alamo A, Di Pino A, Russo C, Calogero AE, Purrello F, et al. Smoking and diabetes: dangerous liaisons and confusing relationships. Diabetol Metab Syndr 2019;11(1):85. https://doi.org/10.1186/s13098-019-0482-2. [8] Walicka M, Krysi´ nski A, La Rosa GRM, Sun A, Campagna D, Di Ciaula A, et al. Influence of quitting smoking on diabetes-related complications: a scoping review with a systematic search strategy. Diabetes Metab Syndr 2024;18(5):103044. https://doi.org/10.1016/j.dsx.2024.103044. [9] Pan A, Wang Y, Talaei M, Hu FB. Relation of smoking with total mortality and cardiovascular events among patients with diabetes mellitus: a meta-analysis and systematic review. Circulation 2015;132(19):1795–804. https://doi.org/10.1161/ CIRCULATIONAHA.115.017926. [10] Qin R, Chen T, Lou Q, Yu D. Excess risk of mortality and cardiovascular events associated with smoking among patients with diabetes: meta-analysis of observational prospective studies. Int J Cardiol 2013;167(2):342–50. https://doi. org/10.1016/j.ijcard.2011.12.100. [11] Livingstone-Banks J, Fanshawe TR, Thomas KH, Theodoulou A, Hajizadeh A, Hartman L, et al. Nicotine receptor partial agonists for smoking cessation. Cochrane Tobacco Addiction Group, editor. Cochrane Database Syst Rev 2023;5 (5):CD006103. https://doi.org/10.1002/14651858.CD006103.pub8. [12] Hajizadeh A, Howes S, Theodoulou A, Klemperer E, Hartmann-Boyce J, Livingstone-Banks J, et al. Antidepressants for smoking cessation. Cochrane Tobacco Addiction Group, editor. Cochrane Database Syst Rev 2023;5(5): CD000031. https://doi.org/10.1002/14651858.CD000031.pub6. [13] Meng Y, Xiang S, Qu L, Li Y. The efficacy and acceptability of pharmacological monotherapies and e-cigarette on smoking cessation: a systemic review and network meta-analysis. Front Public Health 2024;12:1361186. https://doi.org/ 10.3389/fpubh.2024.1361186. [14] Puljevi´ c C, Stjepanovi´ c D, Meciar I, Kang H, Chan G, Morphett K, et al. Systematic review and meta-analyses of cytisine to support tobacco cessation. Addiction 2024; 119(10):1713–25. https://doi.org/10.1111/add.16592. [15] Tonstad S, Lawrence D. Varenicline in smokers with diabetes: a pooled analysis of 15 randomized, placebo-controlled studies of varenicline. J Diabetes Invest 2017;8 (1):93–100. https://doi.org/10.1111/jdi.12543. [16] Rojewski AM, Palmer AM, Baker NL, Toll BA. Smoking cessation pharmacotherapy efficacy in comorbid medical populations: secondary analysis of the evaluating adverse events in a global smoking cessation study (EAGLES) randomized clinical trial. Nicotine Tob Res 2024;26(1):31–8. https://doi.org/10.1093/ntr/ntad126. [17] Tønnesen P, Lawrence D, Tonstad S. Medication-assisted quit rates in participants with smoking-relateddiseases in EAGLES: post hoc analyses of a double-blind, randomized, placebo-controlled clinical trial. Tob Induc Dis 2022;20:1–11. https:// doi.org/10.18332/tid/146567. [18] Russo C, Walicka M, Caponnetto P, Cibella F, Maglia M, Alamo A, et al. Efficacy and safety of varenicline for smoking cessation in patients with type 2 diabetes: a randomized clinical trial. JAMA Netw Open 2022;5(6):e2217709. https://doi.org/ 10.1001/jamanetworkopen.2022.17709. [19] Huang LC, Chang YT, Lin CL, Chen RY, Bai CH. Efficacy of health coaching in smoking cessation and promoting the use of oral smoking cessation drugs in patients with type 2 diabetes: a randomized controlled trial. Int J Environ Res Public Health 2023;20(6):4994. https://doi.org/10.3390/ijerph20064994. [20] Brath H, Lasar D, Buchh¨ ausl I, K¨ astenbauer T, Binter E. Intensified smoking cessation for diabetic patients–preliminary results. Acta Med Austriaca 1999;26(5): 163–7. [21] Rigotti NA, Pipe AL, Benowitz NL, Arteaga C, Garza D, Tonstad S. Efficacy and safety of varenicline for smoking cessation in patients with cardiovascular disease: a randomized trial. Circulation 2010;121(2):221–9. https://doi.org/10.1161/ CIRCULATIONAHA.109.869008. [22] Ryan R, Hill S. How to GRADE the quality of the evidence. Cochrane Consumers and Communication Group 2016. [23] Yang Y, Peng N, Chen G, et al. Interaction between smoking and diabetes in relation to subsequent risk of cardiovascular events. Cardiovasc Diabetol 2022;21 (1):14. https://doi.org/10.1186/s12933-022-01447-2. [24] Herd N, Borland R, Hyland A. Predictors of smoking relapse by duration of abstinence: findings from the International Tobacco Control (ITC) Four Country Survey. Addiction 2009;104(12):2088–99. https://doi.org/10.1111/j.13600443.2009.02732.x. [25] Prochaska JJ, Benowitz NL. The past, present, and future of nicotine addiction therapy. Annu Rev Med 2016;67:467–86. https://doi.org/10.1146/annurev-med111314-033712. [26] McCaul ME, Wand GS, Kuwabara H, Dannals RF, Wong D, Xu X. The relationship of varenicline agonism of α 4β2 nicotinic acetylcholine receptors and nicotine-induced dopamine release in nicotine-dependent humans. Nicotine Tob Res 2020;22(6): 892–9. https://doi.org/10.1093/ntr/ntz080. [27] Chen H-J, Huang W-H, Chan H-L, Hwang L-C. Improvement in cardiometabolic risk factors during smoking cessation treatment in patients with type 2 diabetes: a retrospective cohort study. Diabetes Metab Syndr Obes 2021;14:1695–702. https://doi.org/10.2147/DMSO.S303446. [28] Caponnetto P, Keller E, Bruno CM, Polosa R. Handling relapse in smoking cessation: strategies and recommendations. Intern Emerg Med 2013;8(1):7–12. https://doi.org/10.1007/s11739-012-0864-z. [29] Anderson RJ, Freedland KE, Clouse RE, Lustman PJ. The prevalence of comorbid depression in adults with diabetes. Diabetes Care 2001;24(6):1069–78. https://doi. org/10.2337/diacare.24.6.1069. [30] Holt RIG, De Groot M, Lucki I, Hunter CM, Sartorius N, Golden SH. NIDDK international conference report on diabetes and depression: current understanding and future directions. Diabetes Care 2014;37(8):2067–77. https://doi.org/ 10.2337/dc13-2134. A. Martin et al. Diabetes Research and Clinical Practice 224 (2025) 112202 9