Link between gestational diabetes during pregnancy and breast cancer risk: a review and analysis

This study was conducted with the requirements of Preferred Reporting Items for Systematic Review and Meta-Analyses guideline (33). And it was registered on the International Prospective Register of Systematic Reviews (ID: CRD420251032589).

We conducted a comprehensive search for original studies on the association between GDM and breast cancer using PubMed, Embase, Scopus, the Cochrane Library, and Web of Science. The search covered the time frame from the inception of each database until April 14, 2025. The search terms consisted of both subject headings and free-text terms. The search strategy employed for PubMed was as follows: (((“Diabetes, Gestational”[Mesh]) OR (((Gestational Diabetes Mellitus) OR (GDM)) OR (Diabetes, Pregnancy-Induced))) AND ((“Neoplasms”[Mesh]) OR (((Cancer*) OR (Tumor*)) OR (Carcinoma*)))) AND (Breast OR mammary gland). This strategy was adapted for use in the other databases, with terminology adjustments made according to each database’s specific syntax and indexing system. Meanwhile, manual searches were carried out based on the reference lists of relevant studies.

Eligible studies must meet the following criteria:

(1) The study design must be a cohort or case-control study.

(2) The study should focus on the association between GDM and breast cancer risk.

(3) The study should report odds ratios (OR), relative risks (RR), or hazard ratios (HR) with their corresponding 95% confidence intervals (CI) or provide sufficient data to calculate the effect size between GDM and breast cancer.

(4) The article must be published in English.

The search results from various databases were imported into Endnote X9 software for deduplication and literature management. To ensure data accuracy and objectivity, two independent reviewers (JL and JZL) screened the titles and abstracts of the retrieved literature based on pre-set inclusion criteria. Full texts were obtained and further screened for studies that passed the initial screening to determine the final included studies. During the screening process, any disagreements between the two reviewers were resolved through discussion to reach a consensus; if necessary, a third reviewer (JJ) would participate in the discussion and provide an arbitration opinion.

This study strictly adhered to the PRISMA statement for data extraction to ensure the systematic approach of the research methodology. Two reviewers (JL and JZL) independently extracted data using a predefined data extraction form, while a third author (JJ) cross-checked the accuracy of the results. The extracted data included Author (year), Country, Study design, Age (GDM/non-GDM), Sample size, RR (95% CI), OR (95% CI), HR (95% CI), Follow-up time (years), Quality (scores), and Adjustment factors.

This study evaluated the quality of the included studies using The Newcastle-Ottawa Quality Assessment Scale. The Newcastle-Ottawa Quality Assessment Scale is a tool specifically designed to assess the quality of both cohort and case-control studies, enabling the evaluation of the quality of each study (34). It consists of 8 items organized into three domains: selection of the study groups, comparability of the groups, and assessment of exposure/outcome. The maximum score on this scale is 9 points. Studies scoring less than 4 points are considered low quality, those scoring between 4 and 6 points are of moderate quality, and those scoring between 7 and 9 points are deemed high quality (34).

To evaluate the association between GDM and breast cancer, we equated all RR and OR to HR (35) and then conducted a meta-analysis of HR and its 95% CI. We used the Q test to assess heterogeneity among studies, with a significance level set at P = 0.1. Subsequently, we determined the degree of heterogeneity based on the I² statistic: if I² < 50%, indicating non-significant heterogeneity, we applied a fixed-effects model; if I² ≥ 50%, suggesting significant statistical heterogeneity, we chose a random-effects model (36). Additional analyses were conducted, including a sensitivity analysis using the leave-one-out method (37) and an assessment of publication bias by observing funnel plot symmetry and calculating Begg’s and Egger’s test values (38, 39). Data were processed using Stata 17.0 statistical software, with P < 0.05 indicating statistical significance.

There were no inconsistencies with the pre-registration protocol.

Figure 1 illustrates the selection process and reasons for exclusion in this study. We retrieved 1010 articles from five databases: PubMed, Embase, Scopus, Cochrane Library, and Web of Science. In this study, 343 duplicate articles were removed using the automatic tools of Endnote X9 and manual efforts. Subsequently, 580 studies were excluded based on their titles and abstracts, leaving 87 articles for further evaluation. Two studies were excluded due to the inability to obtain the full text, and the remaining 85 studies progressed to the full-text evaluation stage. After full-text evaluation, 67 studies were excluded because their outcome measures were not relevant to the theme of this study, or complete target data could not be obtained. Ultimately, 18 studies met the inclusion criteria and were included. A related citation tracking search and supplementation were also conducted (5 studies). After applying the inclusion and exclusion criteria, one additional study was included. Finally, 19 studies (17, 25–32, 40–49) were obtained for meta-analysis.

This meta-analysis included 19 studies, consisting of 14 cohort studies and five case-control studies, spanning multiple countries and regions such as Canada, the United States, Israel, South Korea, Taiwan (China), and Denmark. The sample size of the included studies ranged from a minimum of 630 cases to a maximum of 990,572 cases, with a total sample size exceeding 3 million. These studies employed hazard ratios (HR), odds ratios (OR), or relative risks (RR) to evaluate the strength of the association between GDM and breast cancer. Four studies indicated a reduced risk of breast cancer with GDM, five studies showed an increased risk, and ten studies found no significant association. Except for one study rated as moderate quality (5 scores), all other studies were of high quality (≥7 scores). All studies adjusted for confounding factors, covering multiple dimensions such as age, parity, BMI, pregnancy-related factors, long-term lifestyle, and disease history. More information about the main results of each study is presented in Table 1.

We conducted a systematic review and meta-analysis to evaluate the association between GDM and breast cancer risk by including 19 cohort or case-control studies. The results (Figure 2) showed no significant association between GDM and the risk of developing breast cancer (HR=1.03, 95%CI: 0.92-1.15).

We conducted subgroup analyses based on the included studies’ region, study design type, and follow-up duration. In the regional subgroup (Figure 3), results from 11 studies in North America showed that GDM could reduce the risk of breast cancer (HR=0.89, 95%CI: 0.84-0.95). Conversely, findings from six studies in Asia indicated that GDM increased the risk of breast cancer (HR=1.23, 95%CI: 1.15-1.31).

In the study design subgroup (Figure 4), the pooled analysis of 14 cohort studies and five case-control studies both demonstrated no association between GDM and the risk of breast cancer (HR=1.02, 95%CI: 0.97-1.06; HR=0.87, 95%CI: 0.72-1.06).

In the follow-up duration subgroup (Figure 5), the combined results from five studies with short-term follow-up and nine studies with long-term follow-up revealed no significant association between GDM and the risk of breast cancer (HR=0.98, 95%CI: 0.92-1.04; HR=1.04, 95%CI: 0.99-1.10).

As shown in Figure 6, the sensitivity analysis revealed that the pooled results remained robust after excluding any individual study.

The funnel plot (Figure 7), Begg’s test (Z = 0.63, P = 0.529), and Egger’s test (T = 0.21, P = 0.835) (Supplementary Figure S1) provided additional evidence supporting the absence of publication bias in our meta-analysis summary results.

Currently, there is controversy regarding mechanistic studies on the relationship between GDM and breast cancer risk. From a metabolic perspective, the unique state of insulin resistance in women with GDM may affect the growth conditions of tumor cells by altering the microenvironment of breast tissue (17). High estrogen and progesterone levels during pregnancy can induce breast cell differentiation, which may provide a protective effect, reducing the sensitivity of breast epithelial cells to carcinogenic factors (2). Additionally, women with GDM often require strict glycemic control and lifestyle interventions, such as dietary adjustments and moderate exercise. These measures indirectly influence breast cancer risk by improving the overall metabolic state (52). It is worth noting that breastfeeding after GDM may play a significant role, as it promotes terminal differentiation of breast epithelial cells and extends the recovery period of hormone exposure. This biological change may have a long-term protective effect on breast tissue (53, 54). Some studies have also found that changes in specific metabolites associated with GDM (such as adiponectin) can affect tumorigenesis by regulating inflammatory responses and cell proliferation pathways (55, 56).

However, there are also studies suggesting that GDM may increase the risk of breast cancer. Firstly, patients with GDM exhibit significant insulin resistance and hyperinsulinemia. Insulin and its growth factors (such as IGF-1) can promote the proliferation of breast epithelial cells and inhibit apoptosis by activating signaling pathways like PI3K/Akt and MAPK, thereby increasing the risk of carcinogenesis (17, 57). For GDM patients carrying breast cancer genetic susceptibility genes (such as germline mutations in BRCA1/2), hyperinsulinemia, and chronic inflammation may further impair the DNA damage repair capacity through the PI3K/Akt pathway (58, 59). Secondly, the chronic hyperglycemic state associated with GDM can lead to oxidative stress and the accumulation of advanced glycation end products (AGEs), which can induce DNA damage and genomic instability (60, 61). Furthermore, the abnormal secretion of inflammatory factors (such as IL-6 and TNF-α) from adipose tissue in women with GDM can create a tumor-promoting microenvironment. At the same time, elevated estrogen and progesterone levels during pregnancy may synergistically promote the development of breast cancer through hormonal receptor pathways (21, 62). It is important to note that some patients may develop type 2 diabetes mellitus (T2DM) after GDM, and the accompanying metabolic syndrome (such as obesity and dyslipidemia) may further exacerbate the risk of breast cancer by altering adipose factors (such as imbalances in the leptin/adiponectin ratio) (57, 63). These conflicting mechanisms suggest that the impact of GDM on breast cancer may involve complex metabolic memory effects and individual differences (60, 64).

The subgroup analysis in this study revealed a trend toward reduced breast cancer risk in women with GDM in North America. In contrast, a significant positive association between GDM and increased breast cancer risk was observed in Asia. These results may be attributed to variations in diagnostic criteria, accessibility to healthcare services (including medical interventions), regional lifestyle, body mass index (BMI) cutoff values, breastfeeding practices, and genetic factors.

In North America, the potential association between GDM and reduced breast cancer risk may be attributed to the following factors: Firstly, regarding diagnostic criteria and healthcare services, North America adopts the International Association of Diabetes and Pregnancy Study Groups (IADPSG) diagnostic criteria for GDM, enabling accurate patient identification and targeted management (65). Concurrently, the monitoring and intervention for postpartum metabolic abnormalities (e.g., insulin resistance, obesity) in GDM patients are more comprehensive and systematic. Through lifestyle modifications such as dietary control and physical activity, the long-term risk of metabolic disorders is mitigated, thereby reducing potential breast cancer-promoting factors (17). Furthermore, the high accessibility of healthcare services in North America ensures that patients receive timely professional advice and treatment, which facilitates better disease management (17). Secondly, regarding obesity, screening, and breastfeeding, The high obesity rate among North American women, combined with higher BMI thresholds, leads to increased clinical attention toward a larger cohort of obese females. As a marker of metabolic aberration, GDM prompts earlier initiation of breast cancer screening (e.g., mammography), enabling early lesion detection and statistically manifesting as “risk reduction” (21, 57). Additionally, the relatively prevalent breastfeeding practice in North America promotes terminal differentiation of mammary epithelial cells. It prolongs the recovery period from hormonal exposure, conferring long-term protective effects on breast tissue and reducing breast cancer risk. Thirdly, from a genetic perspective, Genetic polymorphisms associated with GDM (e.g., TCF7L2, IRS1) in North American populations (particularly those of European ancestry) (66) may intersect with breast cancer protective pathways (e.g., estrogen metabolism), counteracting the carcinogenic effects of hyperglycemia (2, 17).

In Asian populations, the potential association between GDM and increased breast cancer risk may be explained by the following mechanisms: First, regarding diagnostic criteria and healthcare services, the diagnostic thresholds for GDM in Asian populations are relatively lenient (67, 68), potentially including more mild hyperglycemia cases. The metabolic abnormalities in these cases often receive insufficient intervention, which may lead to epigenetic carcinogenic effects (69, 70). Additionally, the accessibility of healthcare services is suboptimal in some Asian regions, making it difficult for GDM patients to obtain timely and comprehensive medical care, thereby compromising disease management (17). Second, due to the persistent impact of metabolic dysfunction, Asian GDM patients exhibit higher rates of progression to type 2 diabetes mellitus (T2DM) postpartum (71), frequently accompanied by more severe insulin resistance and chronic inflammatory states (72). These factors collectively promote tumor growth through activation of the PI3K/Akt/mTOR signaling pathway (71). Third, differences in body composition distribution (obesity): Asian women exhibit higher proportions of visceral adipose tissue. Given that Asian populations have lower BMI cutoff values (73), even within normal BMI ranges, visceral fat accumulation following GDM may exacerbate abnormalities in adipokine (e.g., leptin) secretion, thereby creating a carcinogenic microenvironment (2, 22). Fourth, screening and intervention delays: In some Asian regions, inadequate long-term follow-up of GDM patients fails to effectively manage glucose metabolism disorders, leading to the persistent accumulation of hyperglycemia-related DNA oxidative damage (27). Fifth, breastfeeding practices: Cultural and occupational factors and other socioenvironmental factors in certain Asian regions result in suboptimal breastfeeding practices, preventing the full realization of the lactation-associated reduction in cancer risk, consequently elevating breast cancer incidence (54).

In conclusion, although we have explored the impact of GDM on the development of breast cancer, the existing research results have not elucidated the specific mechanisms of the association between GDM and breast cancer. Therefore, more basic and clinical studies are needed to clarify the relationship between GDM and the risk of breast cancer.

Our meta-analysis has the following limitations:

(1) All included studies were observational and may have been subject to confounding factors and biases. In addition, both the case-control study and the cohort study are observational studies with a relatively low level of evidence; therefore, the quality of evidence derived from our findings is limited.

(2) Differences in the definition and diagnosis codes for GDM among studies could potentially affect the accuracy of the results.

(3) The heterogeneity of the outcome measures was relatively high, and the sources of this heterogeneity were not fully explained.

(4) Due to the limited number of included studies, the subgroup analysis was dominated by studies from China and Korea, leading to insufficient regional representation. Larger sample size studies are needed for further validation in the future.

Despite these limitations, our meta-analysis has several notable strengths:

(1) This study strictly followed the PRISMA guidelines for systematic searching, screening, and data extraction. The process was ensured to be objective through independent double-blind reviews and third-party arbitration. Additionally, the Newcastle-Ottawa Scale was used to assess the quality of the included studies, ensuring high methodological reliability overall.

(2) Subgroup analyses were conducted to explore the effects of region, study design, and follow-up duration. Regional differences were identified as key moderators, providing new directions for future research and comprehensive data integration.

(3) Both Begg’s and Egger’s tests did not reveal significant publication bias, and the funnel plot demonstrated good symmetry, indicating that small sample studies less influenced the results. Sensitivity analysis showed that the main effect estimates were robust and reliable.

(4) This study also explored the potential mechanisms underlying the association between GDM and breast cancer, providing a stronger theoretical foundation for the research conclusions.

In clinical practice, attention should be paid to the regional differences in the association between GDM and breast cancer risk. Given the observed association in Asian populations with GDM, for Asian patients with GDM, especially those who progress to T2DM or have visceral fat accumulation after childbirth, early screening for breast cancer (such as regular breast ultrasound and mammography) should be strengthened, and the postpartum metabolic follow-up period should be extended. Meanwhile, regardless of the region, lifestyle interventions (such as a low-carbohydrate diet and regular exercise) should be intensified for GDM patients after childbirth to improve insulin resistance and chronic inflammatory status, and breastfeeding should be encouraged to exert its potential protective effects on breast tissue. Future clinical studies can focus on long-term metabolic trajectory monitoring after childbirth in GDM patients and the application of biomarkers in breast cancer risk prediction, providing a scientific basis for individualized prevention strategies.

The original contributions presented in the study are included in the article/. Further inquiries can be directed to the corresponding author. 1