WHI Informs on Links Between Diabetes, Metformin, and Cancer
A new analysis from the Women’s Health Initiative (WHI) shows that, compared with women without diabetes, adult postmenopausal women with type 2 diabetes are at a significant 13% increased risk of invasive cancer and have a 46% higher risk for cancer death.
The latest WHI data, reported in the April 15 issue of theInternational Journal of Cancer, also suggest that long-term metformin therapy — the first go-to medicine for type 2 diabetes — may be linked with a reduced risk of cancer mortality.
In women with carcinomas who were taking metformin to treat their type 2 diabetes, the risk of dying from cancer was about the same as it was for women without diabetes, but the risk of death was 45% higher for women with cancer and diabetes treated with nonmetformin medications.
“Our findings suggest that diabetes remains a risk factor for cancer and cancer death, and metformin therapy, compared with other diabetes medications, may have a role in the prevention or management of diabetes-associated cancer,” lead author Zhihong Gong, PhD, assistant professor of oncology at the Roswell Park Cancer Institute, Buffalo, New York, told Medscape Medical News.
These novel results from the WHI add to a growing body of evidence from meta-analyses and observational studies that has correlated diabetes with an increased risk for cancer as well as linking use of metformin to a reduced risk of cancer-related death in diabetes patients.
“This is a good population-based study and confirms what has been noted in other studies linking diabetes with invasive cancers,” endocrinologist Edward Horton, MD, professor of medicine at Harvard Medical School and a senior investigator at the Joslin Diabetes Center, Boston, Massachusetts, toldMedscape Medical News.
“However, it will not totally settle the question on the association of metformin and reduced risk for cancer death in all people with diabetes,” he added.
And another endocrinologist warns about interpretation of this type of work. “Such studies [as the WHI] need to be viewed with caution,” Vivian A Fonseca, MD, Tullis–Tulane Alumni Chair in Diabetes and chief of the section of endocrinology at Tulane University, New Orleans, told Medscape Medical News .
Dr Fonseca explained that studies linking diabetes and cancer and metformin use with decreased cancer mortality are confounded by selection bias and lack of randomization.
“A randomized study that could conclusively show that metformin is associated with less cancer-related mortality in patients with diabetes would take a long time to complete,” he stressed.
The WHI: Linking Diabetes, Cancer, and Cancer Mortality
The WHI was started in 1991 and was designed to address major health issues related to morbidity and mortality in postmenopausal women between 50 and 79 years of age. In this report, Dr Gong and researchers looked at the link between diabetes and cancer in 145,826 participants.
This analysis represents one of the largest prospective cohort studies to examine associations of diabetes and diabetes medications with risk of cancer and cancer mortality among postmenopausal women in the United States. All the data were collected at baseline and updated during the long-term follow-up (median, 15 years).
“Most other studies have a much shorter follow-up,” Dr Gong noted.
Women with a prior cancer diagnosis and those who had diabetes diagnosed younger than 21 years of age (to remove likely type 1 diabetes) were excluded.
At baseline, women completed a self-administered questionnaire to collect information on medical history, and these were updated semiannually or annually.
The WHI researchers adjusted for variables such as age, ethnicity, education, smoking, aspirin use, and history of hyperlipidemia to determine whether diabetes and use of diabetes medications were associated with an increased risk for cancer and cancer-related death.
At baseline, 8484 participants (5.8%) had diabetes. Of the 145,826 women, 82% were white, 9% black, and 4% Hispanic.
Of the 8484 women with diabetes at baseline, 1100 were on metformin and 4106 were on other medications for diabetes, which were not specified. Compared with other diabetes medications, metformin usage was higher in whites (64.2% vs 60.7%) and lower in blacks (18.8% vs 26.4%).
At a median of 15 years, 24,796 women had a diabetes diagnosis, and 16,248 had invasive cancer. Women with diabetes were at a 13% increased risk of total invasive cancer (hazard ratio [HR], 1.13) compared with those without diabetes (P < .001).
Women with diabetes were specifically at a 37% increased risk (HR, 1.37) for cancers of the digestive organs and peritoneum compared with those without diabetes (P < .001). Significantly increased risk was also seen for cancers of the colon and endometrium, and there was over a twofold higher risk of liver cancer.
Diabetes and Cancer, Metformin vs Other Diabetes Medications
However, the risk of invasive cancer, although significant, did not differ between women who used metformin (HR, 1.13) or those who used other diabetes medications (HR, 1.10).
“In this regard, observations from the WHI differ from other meta-analyses, which noted a decreased risk for the incidence of cancer in patients with diabetes who are treated with metformin,” Dr Horton pointed out.
But these meta-analyses have generally included relatively small and short-term studies that were not adequately powered to see significant differences in cancer incidence or metformin effects, he added.
Cancer mortality was also 46% higher in women with diabetes (HR, 1.46), with a 44% increased risk for death from colorectal cancer (HR, 1.44), for example.
The WHI researchers suggest that long-term use of metformin may reduce the risk of cancer-related death.
There was a lower risk of death from cancer for metformin users compared with users of other medications (HR, 1.08 vs 1.45; P = .007), relative to women without diabetes, and risk of death from breast cancer was also lower among metformin users compared with those using nonmetformin drugs (HR, 0.50 vs 1.29; P = .05).
“The association between diabetes and cancer and metformin reducing cancer risk is a hot topic, and studies of the use of metformin in cancer prevention are ongoing,” Dr Horton noted.
But Does Metformin Really Reduce Cancer Deaths?
Dr Fonseca explained to Medscape Medical News that it is difficult, in the WHI, to determine whether metformin may reduce the risk of cancer mortality in a realistic sense.
He feels that the message this kind of study sends is that other drugs used to treat diabetes are “bad,” which is not true.
“Metformin is standard therapy and is typically used in relatively healthy individuals, while other medications are used in patients with progressive diabetes; in short, in sicker patients,” he noted.
Dr Horton agreed. “Indeed, metformin is contraindicated in patients with significantly impaired kidney function,” he said.
Sicker patients have additional comorbidities that may well contribute to the risk of cancer and cancer mortality, Dr Fonseca pointed out.
“Randomized studies have not shown that other drugs used to treat diabetes are associated with a higher risk for cancer mortality,” he added.
The WHI researchers contend that complex factors shared by diabetes and other cancers may have common underlying, interrelated, pathophysiologies that make it difficult to differentiate individual contributions to overall risk.
Dr Gong told Medscape Medical News that the study did not have information on comorbidities and that “the metformin group” was on metformin alone or in combination with other diabetes medications.
“Specific information on the other diabetes medication was not noted,” she said.
Insulin resistance and excess circulating insulin are likely suspects underlying the link between diabetes mellitus and cancer, so metformin, which sensitizes tissues to insulin, may have a role in the prevention or management of diabetes-associated cancer, the researchers suggest.
But both Drs Fonseca and Horton point out that the pathophysiology linking diabetes to increased cancer risk is purely speculative, as is the reasoning as to how metformin may have an impact.
And prospective randomized trials have not yet shown that metformin use is conclusively associated with lower risk for cancer incidence or mortality in people with diabetes, Dr Horton indicated.
“Currently, there are no definitive data that will settle the question of whether metformin decreases cancer incidence or mortality in all people with diabetes,” he said.
DPPOS Study May Help Answer Question in 5 Years’ Time
A long-term study is indeed required to determine the effects of prolonged metformin use, and theDiabetes Prevention Program Outcome Study (DPPOS) may help provide some of the answers in the next 5 years after a follow-up of approximately 25 years, Dr Horton indicated.
DPPOS is an extension of the Diabetes Prevention Program (DPP), which randomized patients with impaired glucose tolerance (prediabetes) to three arms: a control (placebo); an intensive lifestyle-modification arm; and a metformin arm, where individuals received metformin as a preventive measure.
The original metformin group from DPP going into DPPOS continued receiving metformin as a study drug, with the majority of the original cohort continuing to take 80% of the assigned medication.
“We anticipate that by 2021, when the DPPOS has a 25-year follow-up, we will have a unique opportunity to look at the long-term treatment of metformin on the incidence of cancer and cancer mortality,” Dr Horton said.
And the metformin effect noted in diabetes has not gone unnoticed by the research community. It has spurred a plethora of clinical studies to determine its effects across many solid tumors.
Int J Cancer. 2016;138:1915–1927