Cohort studycited by the Tayebat archivefull text checked A-INS-06 in-depth review
Glargine and breast cancer in UK women
Long-term effects of insulin glargine on the risk of breast cancer.
Overview
Suissa S et al.. Diabetologia 2011 · PMID 21614572 · DOI 10.1007/s00125-011-2190-9
No increase in breast cancer during the first five years; a possible increase later in a subgroup.
Does the study support the specific claim? Partly · Methodological quality: moderate
Plain-language summary
Study explained simply
- What did the researchers want to find out?
Long-term glargine use and breast cancer
- How was the study done?
Matched cohort in UK GPRD
- How many people, animals or samples were studied?
15,227 (4,579 glargine); Women >=40 with type 2 diabetes on insulin
- What was actually found?
Overall HR 1.0; no increase in first 5 years; HR 2.7 after >5 years only in women already on insulin before glargine (10 cases).
- How large was the observed effect?
Overall HR 1.0 (0.7-1.4); first 5 years HR 0.9 (0.7-1.3).
- What do the results mean in everyday life?
Possible long-term question, not established.
- What are the study's limitations?
Long-term signal based on 10 cases in one subgroup; chance possible; no dose data; some baseline differences (e.g. HRT use).
- What does it say about Elawady's specific claim?
Mixed for C-INS-06.
- Do the results apply to Germany and Europe?
UK data.
- What can explicitly not be concluded from this study?
Causation.
Scientific in-depth review
1. Full reference
- Authors
- Suissa S, Azoulay L, Dell'Aniello S, Evans M, Vora J, Pollak M
- Journal
- Diabetologia
- Year
- 2011
- Identifier
- PMID 21614572 DOI 10.1007/s00125-011-2190-9
- Verification
- verified – PubMed E-utilities efetch, 2026-10-25
- Retracted
- no
- Corrections
- none
- Extent checked
- full text
- Conflicts of interest and funding
Funded by Canadian Institutes of Health Research and Canadian Foundation for Innovation. S. Suissa and M. Pollak participated in advisory meetings organised by Sanofi-Aventis (glargine manufacturer); M. Pollak consulted for Novo Nordisk; formal declaration states no duality of interest; PI independent of commercial funders.
2. Research question
Long-term glargine use and breast cancer
3. Study design and population
- Study design
- Cohort study – Population-based cohort study in the UK General Practice Research Database (GPRD): glargine-exposed cohort vs reference other-insulin cohort, prevalent users matched on birth-year stratum, calendar time and duration of prior insulin use; intention-to-treat exposure; time-dependent Cox model by duration of use.
- Population
Women aged >=40 years with type 2 diabetes and >=1 insulin prescription between 1 September 2002 and 31 December 2006 (mean age 65); excluded gestational diabetes, type 1 diabetes, prior breast cancer.
- Country
UK
- Period
Cohort entry September 2002 - December 2006; follow-up to 31 December 2009 (up to 8 years).
- Sample size
15,227 women: 4,579 glargine (1,604 first-time, 2,975 prevalent insulin users) and 10,648 other insulin (3,086 first-time, 7,562 prevalent); 246 breast cancers (66 in glargine cohort, 180 in comparison cohort).
- Inclusion and exclusion criteria
As described in the abstract; details not available in abstract.
4. Methods and endpoints
- Exposure / intervention
Insulin glargine (intention-to-treat from first prescription); dose not analysable (prescriptions mostly 'as directed').
- Comparator
Other insulins (mainly human and analogue biphasic, human long-acting, short-acting).
- Endpoints
Incident breast cancer (validated algorithm, >95% confirmed in 100-case review).
- Statistical methods
Cox proportional hazards adjusted for age, alcohol, smoking, obesity, HbA1c, diabetes duration, prior insulin duration, oophorectomy, prior cancer, HRT, sulfonylureas, thiazolidinediones, metformin, statins; missing-indicator categories for BMI, smoking, HbA1c; duration-of-use strata (<1, 1-3, 3-5, >5 years); cubic spline; sensitivity analyses (lag 1 year, excluding prior cancer). Powered (>80%) for HR 1.5.
5. Actual results
- Results
Overall incidence 4.1 (95% CI 3.6-4.7) per 1,000 per year. Glargine vs other insulin overall adjusted HR 1.0 (0.7-1.4); first 5 years HR 0.9 (0.7-1.3); >5 years HR 1.8 (0.8-4.0; 10 vs 17 cases). First-time insulin users: HR 0.8 (0.3-2.1); no cases after >5 years (393 person-years). Prevalent insulin users: overall HR 1.0 (0.7-1.5); >5 years HR 2.7 (1.1-6.5; 10 cases, 1,400 person-years); sensitivity: excluding first year 2.5 (1.0-6.1), excluding prior cancer 3.9 (1.5-10.5). P=0.031 for the 2.7 estimate, from one of two stratified analyses.
- Absolute risks
Breast cancer rate per 1,000 per year: glargine 3.5 vs other insulins 4.4 overall; after >5 years 5.6 vs 3.8 (all), 7.1 vs 3.1 (prevalent insulin users).
- Authors' conclusion
Breast cancer risk not increased during the first 5 years of glargine use; longer-term use may increase this risk, particularly in women with long-standing insulin use before glargine; prospective studies would be more definitive but probably impractical; role of chance cannot be dismissed.
6. Strengths
- Up to 8 years follow-up, longer than earlier studies
- Matching on calendar time and prior insulin duration; separation of first-time and prevalent users
- Adjustment for BMI, smoking, alcohol, HbA1c, HRT and other breast cancer risk factors
- Validated outcome algorithm; intention-to-treat avoids reverse causation from stopping glargine
7. Weaknesses
- Positive finding rests on 10 cases in one subgroup (>5 years, prevalent users) with wide CI; p=0.031 from repeated stratified analyses (authors acknowledge chance)
- Overall and first-5-year results null; no cases at all >5 years in first-time glargine users
- No dose data; missing-indicator handling of BMI/smoking/HbA1c
- Prevalent glargine users differed (leaner, higher HbA1c, more HRT use 29.6% vs 19.4%) - residual confounding possible
- Unmeasured diet, physical activity, ethnicity, mammography screening
8. Risk of bias
- Orienting tool
- ROBINS-E
- Type of assessment
- structured assessment based on the full text by a single reviewer (no formal double review)
- Full formal assessment
- no
- Overall risk of bias
- some concerns
- Rationale
Single-reviewer ROBINS-E-oriented judgement from the full text; formal assessment needs two independent reviewers. Overall rating unchanged (some_concerns). Main concern is chance in a small subgroup rather than systematic bias.
| Domain | Judgement | Rationale |
|---|---|---|
| Confounding | some concerns | Extensive adjustment incl. BMI, HbA1c, HRT; but imbalances in prevalent users and missing-indicator method; screening not measured. |
| Exposure measurement | some concerns | Prescription-based ITT exposure; no dose; discontinuation/switching ignored. |
| Selection of participants | low | Population-based GPRD cohort with matching on time and prior insulin duration. |
| Post-exposure interventions | some concerns | Co-medication with other insulins in glargine cohort (short-acting analogues, biphasic). |
| Missing data | some concerns | BMI, smoking, HbA1c partly missing, handled by missing category. |
| Outcome measurement | low | Validated breast-cancer algorithm with >95% confirmation. |
| Selection of reported result | some concerns | Abstract emphasises the significant subgroup (>5 years, prevalent users) among multiple duration x stratum comparisons. |
Full-text review (25 October 2026): Cohort breakdown (first-time vs prevalent), case counts (66 vs 180), overall HR 1.0 (0.7-1.4), stratum results and sensitivity HRs added; absolute rates by group and duration added; COI completed (CIHR/CFI funding, Sanofi advisory participation); ROBINS-E domains added, overall unchanged (some_concerns).
9. Assessment of the statistical interpretation
The main analysis is null (HR 1.0, 0.7-1.4). The long-term signal is imprecise (10 exposed cases) and confined to one of two strata; the >5-year estimate in all women is non-significant (HR 1.8, 0.8-4.0). Hypothesis-generating only.
10. Relevance to the specific claim
Mixed for C-INS-06.
11. Independent contrary findings
12. Transferability to Germany and Europe
UK data.
13. Overall methodological judgement
No overall or short-term association of glargine with breast cancer; an imprecise long-term subgroup signal that requires confirmation. Does not establish that long-term glargine use causes breast cancer.
| Multidimensional assessment | |
|---|---|
| Authenticity of the source | verified |
| Methodological quality | moderate |
| Risk of bias | some concerns |
| Precision of results | low |
| Directness | direct |
| Reproducibility | not replicated |
| Consistency with the overall evidence | inconsistent |
| Regional transferability | high |
| Adequacy of the interpretation (by the archive) | not verifiable |
14. Open questions
None beyond those stated.
15. Linked fact checks
- Insulin glargine increases cancer risk in a dose-dependent way (and long-term use increases breast cancer). – Studies with mixed or unclear results
Use by the Tayebat archive
Note: according to the archive, the studies were compiled by the archive operators. Whether Dr. Elawady himself cited this study is not established.
- What the archive says about the study
Listed in the archive bibliography for 'insulin-injections' without its own interpretation; used as general support for the claim that insulin intensification harms.
- A. What was actually studied?
Long-term glargine use and breast cancer
- B. What did the authors find?
Overall HR 1.0 (0.7-1.4); first 5 years HR 0.9 (0.7-1.3); >5 years HR 1.8 (0.8-4.0); prevalent insulin users >5 years HR 2.7 (1.1-6.5) based on 10 cases; first-time users no increase.
- C. What conclusions do the authors themselves draw?
No increase in first 5 years; longer use may increase risk, particularly with long prior insulin use.
- D. What does the archive derive from it?
Implicit support for long-term glargine-cancer concern.
- E. Is this derivation scientifically justified?
Partly: the full text shows an overall null result; the long-term signal is based on 10 cases in a subgroup and the authors state chance cannot be dismissed.
- Adequacy of the interpretation (by the archive)
- not verifiable