Insulin Glargine/Breast Cancer Link Seen Again in Type 2 Diabetes

Marlene Busko

October 18, 2017

In a large observational study, women with type 2 diabetes who received long-acting insulin glargine (Lantus, Sanofi) had a 1.4-fold increased risk of breast cancer compared with women who were given intermediate-acting neutral protamine Hagedorn (NPH) insulin during roughly 4 years and up to 12 years of follow-up.

In contrast, those who received insulin detemir (Levemir, Novo Nordisk) did not have any increased risk of breast cancer.

Of note, the breast-cancer signal with insulin glargine was only significant among prior insulin users and not new users.

And this signal does not mean clinicians should change clinical practice without a review by regulatory agencies, caution the investigators, led by Jennifer W Wu, MD, of McGill University, Montreal, Quebec, who published their paper in the Journal of Clinical Oncology.

"Despite these findings, the benefits and risk of insulin glargine must be considered by drug regulatory agencies before any changes in clinical practice can be made," they conclude.

The fact that the increased risk of breast cancer was only seen in women with prior exposure to insulin, "an unusual group of type 2 [diabetes] women," suggests "the safety signals may be nuanced, with specific types of insulin," said Craig Currie, PhD, an epidemiologist from Cardiff University, Wales, who was not involved with the study.

Nevertheless the "findings add to an increasing body of evidence that questions the safety of insulin in people with type 2 diabetes more generally," he told Medscape Medical News in an email.

New Study as Long-Acting Insulins Have Been Available for Longer

The issue of cancer risk associated with insulin, which is a potential growth factor, isn't new and has reared its head before in relation to insulin glargine in particular.

Several observational studies have looked at whether insulin glargine is associated with an increased risk of breast cancer and have come up with conflicting results, Dr Wu and colleagues explain as background.

Moreover, in the randomized Outcomes Reduction Insulin Glargine Intervention (ORIGIN) trial, which had adjudicated cancer outcomes reported in 2013, only roughly 4000 of the 12,500 patients were women, and there were only 56 cases of breast cancer during follow-up, which was "insufficient power for site-specific cancers such as breast and also...too short a follow-up for the necessary latency."

Thus "to date, the US Food and Drug Administration finds the evidence is inconclusive and suggests that more epidemiologic data are needed," they observe.

So they aimed to assess the risk of breast cancer with insulin analogs, "now that these insulins have been on the market for a longer time."

They identified 22,395 women with type 2 diabetes in the UK Clinical Practice Research Datalink (CPRD) who had not been prescribed insulin before age 40, had not had gestational diabetes or any cancer, but had received at least one prescription of insulin glargine, detemir, or NPH during 2002–2012.

On average, the women received 5.4 to 5.8 insulin prescriptions each year.

During follow-up, 108 of 9549 women who received NPH, 176 of 9575 women who received insulin glargine, and 37 of 3271 women who received insulin detemir developed breast cancer.

The incidence rates of breast cancer were 35.1, 48.7, and 14.8 per 1000 person-years for women who received NPH, glargine, and detemir, respectively.

New Insulin Users Not at Risk?

Those who received insulin glargine for 5 or more years had a roughly twofold increased risk of breast cancer compared with the reference group (women who received NPH) and women who were switched to insulin glargine from another insulin had an approximately 1.5-fold increased risk of developing breast cancer.

Risk of Breast Cancer, Insulin Glargine vs NPH Users*

Insulin glargine users HR (95% CI)
Overall users 1.44 (1.11–1.85)
> 5 y of insulin glargine 2.23 (1.32–3.77)
> 30 prescriptions 2.29 (1.26–4.16)
Prior insulin users 1.53 (1.10–2.12)
New insulin users 1.18 (0.77–1.81)
*During a mean 4.4-year follow-up; adjusted for age, study entry, alcohol use, smoking, BMI, HbA1c, diabetes duration, prior insulin use, duration of insulin use, Charlson comorbidity score, noninsulin diabetic medications, other medications

The risk of breast cancer was not significantly higher in the insulin detemir group compared with NPH users (HR, 1.17; 95% CI 0.77–1.77). However, this result is somewhat inconclusive due to the smaller number of women using this newer insulin and the shorter time that it has been available, Dr Wu and colleagues write.

Riccardo Perfetti, MD, PhD, VP for medical affairs diabetes, Sanofi, told Medscape Medical News in an email: "The results of the new-user analysis in this study provided additional evidence that [in new users] insulin glargine is not associated with an increased risk of breast cancer, even for exposure longer than 5 years."

However, Dr Wu and colleagues note that among new users, the upper limit of the confidence interval was 1.81 (meaning that some of the women had this increased risk), and the numbers may have been too small to detect risk, since few women in the comparator NPH group (< 30%) were new insulin users.

Too Much Confounding? Weak Signal With No Cause for Concern

The new trial is "interesting, but I don't think it can rule out confounding," lead author of the ORIGIN trial, Hertzel C Gerstein, MD, from McMaster University, in Hamilton, Ontario, who was not involved with the current study, told Medscape Medical News.

"Is it the insulin, or somehow the type of person who was prescribed glargine was different from the type of person who was prescribed NPH or detemir? We don't know," he said.

Also, the study did not adjust for age at menarche and menopause, parity, age at first birth, breast feeding, "and in particular, family history of breast cancer," Dr Perfetti stressed.

The researchers acknowledge that unaccounted-for confounders might explain some of the findings. "However, this is unlikely because only a strong unmeasured confounder, with major imbalance between insulin glargine and NPH, would be needed to bias the hazard ratio."

Nevertheless, "I would not say that this study raises any real concern in my mind," Dr Gerstein said, noting it was an observational study with a weak signal.

"If it were a strong risk relationship, fivefold, 10-fold, then we'd be having a different discussion."

"In the International Epidemiology Study and the ORIGIN trial, a total of 1,595,400 person-years of exposure to insulin were evaluated, and no association with cancer has been found," Dr Perfetti pointed out.

Upcoming Study From US Databases

"Moreover," Dr Perfetti added, "new results, currently in press, assessing the risk of breast cancer in women treated with insulin glargine and human NPH insulin in new users and all-users on two large US database, analyzing ~300,000 patients over 12 that glargine was not associated with an increased risk of breast cancer compared with NPH for long-term exposure."

But according to Dr Currie, to truly resolve the question of whether insulin glargine increases the risk of breast cancer "would require a trial of analogue basal insulins vs alternative insulin regimens vs non-insulin glucose-lowering medications."

The study was partly funded by the Canadian Institutes of Health Research and the Canadian Foundation for Innovation. Dr Wu was a recipient of a Canadian Institutes of Health Research Doctoral Research award at the time the study was being conducted. Disclosures for the coauthors are listed in the paper. Dr Currie has no relevant financial relationships. Dr Gerstein has received research grant support from Sanofi, Lilly, AstraZeneca and Merck, and honoraria for speaking or consulting from Sanofi, Novo Nordisk, Lilly, Boehringer Ingelheim, AstraZeneca, Merck, and Abbott.Sanofi sponsored the ORIGIN trial.

J Clin Oncol. Published online September 27, 2017. Abstract

For more diabetes and endocrinology news, follow us on Twitter and on Facebook.