Insulin glargine: studies of possible cancer link
The European Medicines Agency has advised that no change in recommendations for use is required at present.
Article date: September 2009
Recent observational studies have suggested a possible association between insulin glargine and an increased risk of cancer. The results are not entirely consistent, and can neither confirm nor exclude a relationship between insulin glargine and cancer. Therefore, the European Medicines Agency has advised that no change in recommendations for use is required at present.
Insulin glargine
Insulin glargine (Lantus) is a long-acting insulin analogue—an insulin molecule that has been modified for more sustained effects after injection. It contains arginine residues at positions B31 and B32, together with a glycine substitution at A21. Insulin glargine is licensed for the treatment of adults, adolescents, or children age 6 years or older who have diabetes mellitus, where treatment with insulin is required.
In type 1 diabetes, glargine is usually given once daily as part of a basal bolus regimen. The National Institute for Health and Clinical Excellence recommends neutral protamine Hagedorn (NPH) insulin as first-line therapy for type 2 diabetes, but glargine may be indicated for those who require assistance to administer injections, or who cannot cope with twice-daily injections for other reasons, or who experience troublesome hypoglycaemia.
Diabetes and cancer
Type 2 diabetes is associated with an increased risk of certain types of cancer, including cancer of the breast, colon, and pancreas. These tumours are insulin-responsive in vitro, raising the possibility that insulin might act as a tumour growth factor. Three studies[footnote 1] [footnote 2] [footnote 3] found that metformin was associated with a lower risk of cancer than insulin or sulfonylureas. Cancer diagnosis or mortality are increased in insulin or sulfonylurea users compared with those on metformin, 2 suggesting that metformin may have an antitumour effect;[footnote 4] however, this remains to be confirmed.
Data for cancer risk with insulin glargine
Preclinical data
Some alterations to the insulin molecule increase its trophic effects, as shown in cell-culture systems, typically human mammary epithelial cells. These effects are mediated by prolonged binding to the insulin receptor or increased cross-reactivity with the insulin-like growth factor 1 receptor.4
Insulin glargine is partially degraded at the injection site yielding two bioactive products, suggesting that it acts to some extent as a pro-drug. However, there is substantial variation between individuals’ insulin metabolism, and it is not possible to clearly extrapolate from in vitro to the in vivo situation.4
All new insulins are routinely screened for their effects on cell growth in preclinical evaluation. Some in vitro studies have suggested that insulin glargine may have greater mitogenicity than human insulin[footnote 5] or other insulin analogues. [footnote 6] A 2-year carcinogenicity study in rats and mice found no difference in frequency of mammary tumours between insulin glargine, NPH insulin, and control groups. [footnote 7] However, there was a high overall mortality rate, which may have affected the ability of this study to detect differences in tumour frequency between treatment groups.
Epidemiological studies
After an initial observational study[footnote 8] suggesting a possible association between insulin glargine and an increased risk of cancer, further findings 3Colhoun HM, SDRN Epidemiology Group. Diabetologia published online July 15, 2009; DOI:10.1007/s00125-009-1453-1.,[footnote 9] have been published. These studies assessed the overall risk of cancer in addition to the risk of breast cancer with insulin glargine, compared with risk in other treatment groups (some insulin and some oral therapy). The results of these retrospective observational studies are not entirely consistent, and can neither confirm nor exclude a relationship between insulin glargine and cancer. A summary of the main results is presented in the table below:
Table: Studies of cancer risk for insulin glargine
| Any malignancy, hazard ratio (95% CI) | Breast cancer, hazard ratio (95% CI) | ||||
|---|---|---|---|---|---|
| Hemkens et al8 | 1·18 (1·08–1·28)* | Not assessed | |||
| Currie et al3 | 0·81 (0·59–1·11)† 1·14 (0·84–1·52)‡ | 0·86 (0·42–1·75)§ | |||
| Colhoun et al9 (incident insulin cohort) | 0·87 (0·63–1·21) | 1·47 (0·59–3·64) | |||
| Jonasson et al10 | 1·06 (0·90–1·25) | 1·97 (1·30–3·00) |
| Comparators: *Human insulin alone. †Long-acting human insulin; data derived to ensure consistency of comparators and are not cited in original paper. ‡Biphasic human insulin; data derived to ensure consistency of comparators and are not cited in original paper. §All insulins. | Non-glargine insulin. |
The four studies 3,8,9,10 were relatively short in duration of exposure and observation period to study drug-induced or drug-modified malignances. The mean follow-up for glargine varied from 1·31 years to 2·74 years, whereas the mean follow-up for comparator groups varied from 1·68 years to 3·36 years. A positive association between insulin glargine and any malignancy was found in the study by Hemkens and colleagues8 only after re-analysis of the data for dose-related effects. Jonasson and colleagues10 investigated the incidence of in situ tumours, breast cancer, gastrointestinal cancer, and prostate cancer and found a positive association with use of insulin glargine alone and breast cancer.
Methodological problems identified in these studies included potential exposure misclassification, selection bias, differing choice of comparator group, adjustment for confounding factors, and incomplete information on risk factors. Although the studies controlled for some confounding factors such as age and smoking, most known risk factors for breast cancer (ie, age at menopause, parity, exogenous hormone use, genetic predisposition or family history, body-mass index, and socioeconomic status) were not taken into account in most of the analyses.
Randomised controlled trials
A post-hoc analysis of data from a randomised controlled trial[footnote 10] is more reassuring. This relatively small 5-year trial compared the risk of diabetic retinopathy in patients receiving insulin glargine or NPH insulin; cancer risk was a secondary outcome. 514 patients received insulin glargine and 503 patients received NPH. Three (0·6%) patients in the insulin glargine group developed breast cancer versus five (1%) patients in the NPH group.
Advice from the European Medicines Agency
After a review of the available preclinical and clinical data, the European Medicines Agency has advised that no change in recommendations for use is required at present, and patients being treated with insulin glargine can continue their treatment. However, further research is needed in this area.
Further review and research
A range of options for further research on this issue is being considered. There is a large ongoing randomised controlled trial (called ORIGIN, Outcome Reduction with Initial Glargine Intervention), which may provide additional useful information on this issue.
Further information
European Medicines Agency update on safety of insulin glargine
Diabetes UK website
Post-publication note:
In March 2022, non-functioning links were removed during routine review of older articles.
Article citation: Drug Safety Update Sept 2009, vol 3 issue 2: 3.
-
Evans JM, et al. BMJ 2005; 330: 1304–05. ↩
-
Bowker SL, et al. Diabetes Care 2006; 29: 254–58. ↩
-
Currie CJ, et al. Diabetologia published online July 2, 2009; DOI:10.1007/s00125-009-1440-6. ↩
-
Smith U, Gale E. Diabetologia published online July 14, 2009; DOI:10.1007/s00125-009-1441-5. ↩
-
Weinstein D, et al. Diabetes Metab Res Rev 2009; 25: 41–49. ↩
-
Shukla A, et al. Endocr Relat Cancer 2009; 16: 429–4 ↩
-
Stammberger I, et al. Int J Toxicol 2002; 21: 171–79. ↩
-
Hemkens LG, et al. Diabetologia published online June 30, 2009; DOI:10.1007/s00125-009-1418-4. ↩
-
Jonasson JM, et al. Diabetologia published online July 9, 2009; DOI:10.1007/s00125-009-1444-2. ↩
-
Rosenstock J, et al. Diabetologia published online July 16, 2009; DOI 10.1007/s00125-009-1415-7. ↩