Proposed article structure
For this topic, a clinically useful structure is: Highlights; Clinical background; Study design and methods; Key results; Interpretation and biological plausibility; Clinical implications; Strengths and limitations; Conclusion; Funding, registration, and citation. This structure fits an observational diabetes-oncology study in which the central issue is how timing of glycaemic exposure relates to long-term cancer outcomes.
Highlights
Higher cumulative HbA1c exposure after diagnosis of type 2 diabetes was associated with higher incident cancer risk in a large territory-wide cohort of Hong Kong Chinese adults.
Hyperglycaemia during the first 2 years after diabetes diagnosis was linked to a 30% to 75% higher relative risk of cancer at any site when HbA1c was at least 53 mmol/mol (7.0%), even after accounting for later glycaemic exposure.
The study’s weighted cumulative exposure analysis suggested that earlier HbA1c exposure contributed more strongly to cancer risk than later exposure, raising the possibility of a biologically important early window after diagnosis.
A modeled 11 mmol/mol (1%) HbA1c reduction at 1 to 2 years was associated with a modest but statistically significant 6% lower cancer risk over a hypothetical 10-year window, whereas reductions after 5 years were not associated with significant risk differences.
Clinical background
Type 2 diabetes is associated with higher risk of several cancers, including liver, pancreas, colorectum, endometrium, breast, and bladder, although the magnitude varies by site and population. The mechanisms remain incompletely defined but likely involve chronic hyperinsulinaemia, insulin resistance, inflammation, altered growth factor signaling, oxidative stress, non-alcoholic fatty liver disease, and shared risk factors such as obesity, aging, smoking, and physical inactivity.
HbA1c is widely used to reflect average glycaemia over approximately 2 to 3 months and is a core therapeutic target in diabetes care. However, whether the timing of glycaemic exposure matters for later cancer development has been uncertain. Most prior studies have focused on average glucose control over long follow-up periods, without examining whether glycaemia soon after diagnosis has a distinct effect independent of later control.
This question is clinically important. Early diabetes management is already emphasized to prevent microvascular complications, and intensive early control may have durable effects on vascular outcomes, sometimes referred to as a legacy effect. Whether a similar phenomenon exists for cancer has been much less clear. The present study addresses that gap by asking not only whether higher HbA1c is associated with cancer risk, but also whether earlier exposure carries greater weight than later exposure.
Study design and methods
Zhang and colleagues conducted a retrospective cohort study using a territory-wide database in Hong Kong. The study included 52,926 Hong Kong Chinese adults with newly diagnosed type 2 diabetes. The principal analyses were performed in slightly smaller subsets depending on available HbA1c data and the modeling approach: 49,978 individuals for analyses of time-weighted mean HbA1c over total follow-up, 39,185 for analyses focused on prespecified early exposure periods, and 49,966 for the weighted cumulative exposure model.
The primary exposure was HbA1c, treated in several complementary ways. First, investigators examined the association between time-weighted mean HbA1c across overall follow-up and incident cancer. Second, they evaluated time-weighted mean HbA1c during specific early windows after diabetes diagnosis, especially the first 2 years and up to the first 5 years. Third, they used a weighted cumulative exposure model to estimate how much historical HbA1c at different time points contributed to subsequent cancer risk. This modeling strategy is particularly relevant when one suspects that exposures at some points in time may be more biologically influential than others.
The main endpoint was incident cancer at any site. Hazard ratios were reported with 95% confidence intervals. The abstract indicates that analyses accounted for subsequent HbA1c when early exposure was studied, which is a notable attempt to isolate the influence of early glycaemic burden rather than simply total glycaemia over time.
Key findings
Overall glycaemic burden and cancer risk
Among 49,978 individuals with newly diagnosed type 2 diabetes, 1,758 cancer events occurred. Each 11 mmol/mol (1%) increase in time-weighted mean HbA1c was associated with a 27% relative increase in risk of cancer at any site, with a hazard ratio of 1.27 (95% CI 1.20 to 1.33). This is a robust association in magnitude and precision for an observational study, and it supports the concept that chronic hyperglycaemia is not merely a marker of diabetes severity but may be linked to oncologic risk in a graded fashion.
Early glycaemic exposure appeared especially important
The most clinically provocative finding was that HbA1c during the first 2 years after diagnosis carried an independent association with later cancer. Compared with a time-weighted mean HbA1c below 53 mmol/mol (7.0%), HbA1c at or above 53 mmol/mol (7.0%) during this early period was associated with a 30% to 75% higher relative risk of cancer at any site, depending on the specific HbA1c category analyzed. Importantly, this relationship persisted even after adjustment for subsequent HbA1c, implying that early exposure was not simply acting as a proxy for poor control throughout follow-up.
The signal strengthened when longer durations of early exposure were considered. Across the first 5 years after diagnosis, higher HbA1c exposure was associated with relative risk increases ranging from 51% to 213%. Although the abstract does not provide all category-specific cut points and confidence intervals, the trend suggests a dose-duration relationship: both higher glucose and longer exposure early in the disease course were linked to greater cancer risk.
Earlier exposure had more weight than later exposure
The weighted cumulative exposure model adds a mechanistic and temporal layer to the findings. Earlier elevated HbA1c values contributed more strongly to cancer risk than later elevations. In other words, two individuals with similar overall glycaemic burden might not carry the same cancer risk if one had worse control near diagnosis and improved later, while the other achieved good early control but worsened later. This is the essence of the paper’s main message and aligns with the authors’ interpretation that early glycaemic management may have lasting implications.
Potential benefit of early HbA1c reduction
The model further suggested that lowering HbA1c by 11 mmol/mol (1%) at 1 to 2 years after diagnosis was associated with a 6% relative reduction in cancer risk over a hypothetical 10-year period, with a hazard ratio of 0.94 (95% CI 0.91 to 0.98). By contrast, equivalent HbA1c reductions after 5 years were not associated with statistically significant differences in risk. This effect size is modest in absolute terms, but at the population level it may be meaningful, especially given the high global prevalence of type 2 diabetes.
Interpretation and biological plausibility
Several mechanisms could explain why hyperglycaemia early after type 2 diabetes diagnosis might matter disproportionately. One possibility is that the early disease period reflects a broader metabolic milieu that includes not only elevated glucose but also pronounced insulin resistance, compensatory hyperinsulinaemia, lipid abnormalities, chronic low-grade inflammation, and hepatic steatosis. These processes can promote cell proliferation, DNA damage, and growth factor signaling through pathways such as insulin and insulin-like growth factor 1 signaling.
Another possibility is that early diabetes may accelerate growth of preclinical neoplastic lesions already present at diagnosis. If this is true, then the first years after diabetes onset could represent a window during which metabolic improvement alters the trajectory of occult tumor biology. This would fit the observation that earlier HbA1c reduction appeared more relevant than later reduction.
It is also possible that HbA1c acts partly as an integrated marker of overall metabolic health, treatment adherence, access to care, and competing lifestyle factors. Even with careful adjustment, observational studies cannot fully separate the direct effect of glycaemia from correlated exposures such as obesity, smoking, alcohol use, or medication patterns. Therefore, the results should not be interpreted as proving that glucose alone causes cancer. Rather, they support the idea that early dysglycaemia identifies, and may contribute to, a higher-risk oncometabolic state.
Clinical implications
For clinicians, the study reinforces the importance of prompt and sustained glycaemic management soon after type 2 diabetes is diagnosed. Current diabetes guidelines already advocate individualized but generally early treatment intensification to reduce symptomatic hyperglycaemia and prevent microvascular complications. These data suggest that there may also be a longer-term cancer prevention rationale for avoiding prolonged early HbA1c elevation.
That said, the study does not justify aggressive glucose lowering at any cost, nor does it establish a cancer-specific HbA1c target independent of existing standards. Overly intensive treatment, especially in older adults or those with multimorbidity, may increase hypoglycaemia risk without clear net benefit. The practical message is more balanced: therapeutic inertia in the first years after diagnosis may have consequences beyond classical diabetes complications, and achieving reasonable HbA1c control early may be especially valuable.
The findings may also influence risk communication. Patients often understand the relationship between glucose and eye, kidney, or nerve disease, but may be less aware of links between diabetes and cancer. Discussing metabolic health in a broader framework may improve motivation and long-term engagement, particularly when framed around overall health rather than fear.
Strengths and limitations
Major strengths
The study’s strengths include its large sample size, real-world territory-wide dataset, substantial number of cancer events, and sophisticated exposure modeling. Using time-weighted mean HbA1c and a weighted cumulative exposure framework is methodologically stronger than relying on a single baseline value. The attempt to separate early glycaemic exposure from later exposure addresses an important and often overlooked question.
Key limitations
As an observational retrospective cohort study, residual confounding remains the main limitation. The abstract does not list all covariates used in adjustment, and factors such as body mass index, waist circumference, smoking, alcohol consumption, diet, physical activity, socioeconomic status, diabetes duration before clinical diagnosis, and use of specific glucose-lowering medications may influence both HbA1c and cancer risk.
Reverse causation is another concern, especially for cancers such as pancreatic cancer that may alter glycaemia before diagnosis. Newly diagnosed diabetes can sometimes be a manifestation of occult malignancy rather than the cause of it. Careful lag analyses can mitigate this issue, but they are not described in the abstract.
Generalizability also deserves caution. The cohort consisted of Hong Kong Chinese adults, and associations may differ in populations with different cancer epidemiology, obesity prevalence, treatment patterns, or genetic backgrounds. In addition, the endpoint was cancer at any site. Site-specific analyses would be particularly helpful because hyperglycaemia may have stronger relationships with some cancers than others.
Finally, the modeled estimate that a 1% HbA1c reduction at 1 to 2 years lowers cancer risk by 6% over 10 years is hypothesis-generating rather than a direct causal effect from an intervention trial. Clinicians should not interpret it as proof that glucose lowering itself will necessarily produce that exact risk reduction.
How this fits with existing evidence
Previous epidemiologic studies and umbrella reviews have shown that type 2 diabetes is associated with elevated risk of multiple cancers. However, evidence specifically linking temporal glycaemic patterns to cancer has been less consistent than evidence for vascular outcomes. This study advances the field by suggesting an exposure-timing effect, conceptually similar to the legacy effect described in diabetes complications research.
Whether future studies confirm a true early glycaemic window for cancer risk will depend on replication in other populations, site-specific analyses, and ideally causal inference approaches that better address time-varying confounding. Nonetheless, the present data are biologically plausible and clinically relevant enough to merit attention.
Conclusion
This large real-world cohort study suggests that in newly diagnosed type 2 diabetes, hyperglycaemic exposure is associated with higher long-term cancer risk, and that the timing of exposure matters. Elevated HbA1c soon after diagnosis appeared to confer greater risk than similar exposure later in the disease course, even when subsequent glycaemia was taken into account. For practicing clinicians, the message is not to pursue indiscriminate intensive therapy, but to recognize that early, effective, and durable glycaemic management may have benefits extending beyond traditional diabetes endpoints. For researchers, the findings open an important line of inquiry into whether there is an oncometabolic legacy effect in diabetes care.
Funding, registration, and citation
Funding: Not reported in the abstract provided. Readers should consult the full Diabetologia article for complete funding and conflict-of-interest disclosures.
ClinicalTrials.gov: Not applicable based on the information provided, as this was a retrospective observational cohort study rather than an interventional trial.
Citation: Zhang X, Yang A, Wu H, Lau ESH, Shi M, Fan B, Chow E, Kong APS, Chan JCN, Ma RCW, Luk AOY. Early glycaemic exposure and cancer risk in people with newly diagnosed type 2 diabetes. Diabetologia. 2026-06-05. PMID: 42247170. https://pubmed.ncbi.nlm.nih.gov/42247170/
Selected references
Zhang X, Yang A, Wu H, Lau ESH, Shi M, Fan B, Chow E, Kong APS, Chan JCN, Ma RCW, Luk AOY. Early glycaemic exposure and cancer risk in people with newly diagnosed type 2 diabetes. Diabetologia. 2026-06-05. PMID: 42247170.
Pearson-Stuttard J, Papadimitriou N, Markozannes G, Cividini S, Kakourou A, Gill D, et al. Type 2 diabetes and cancer: An umbrella review of observational and Mendelian randomization studies. Cancer Epidemiol Biomarkers Prev. 2021;30(6):1218-1228.
American Diabetes Association Professional Practice Committee. 6. Glycemic goals and hypoglycemia: Standards of Care in Diabetes—2026. Diabetes Care. 2026;49(Suppl 1). Please verify final volume, issue, and page details in the published guideline version.
Giovannucci E, Harlan DM, Archer MC, Bergenstal RM, Gapstur SM, Habel LA, et al. Diabetes and cancer: A consensus report. CA Cancer J Clin. 2010;60(4):207-221.
