Potential Gender Differences of Exenatide in Outpatients With Type 2 Diabetes Mellitus

Type 2 diabetes mellitus (T2DM) represents an emerging burden in developing countries and in developed nations. In Italy, the prevalence was 3.9% in 2001, but in 2007 it reached 4.7%—an increase of 15% in 6 years. Italy’s national health system has to bear the cost of this expensive disease and its chronic complications,¹ among which, cardiovascular (CV) disease represents the leading cause of mortality. Among women with diabetes, neither all-cause nor CV disease mortality declined between 1971 to 1986, and 1988 to 2000. The all-cause mortality rate difference between women with and without diabetes more than doubled (from a difference of 8.3 to 18.2 annual deaths per 1000 persons).²

Furthermore, the effects of some “classical” glucose-lowering therapies on CV events are controversial. Twice-daily injections of exenatide, a GLP-1 receptor agonist (GLP-1 RA), have recently become available for Italian patients with diabetes who did not achieve adequate glycemic control with diet, exercise, and treatment (ie, monotherapy or combination treatment with metformin and a sulphonylurea, or pioglitazone). Exenatide reduces glycated hemoglobin (A1C) and body weight (BW), with improvement in CV risk factors in some but not all patients.³ Few data exist about gender differences concerning the use of exenatide in real-world settings.

STUDY DESIGN AND METHODS

We retrospectively studied 69 outpatients with type 2 diabetes mellitus (34 male and 35 female; mean age = 52.6 years [SD = 10.3]; mean duration of diabetes = 9.4 years [SD = 5.4]) with persistent exenatide treatment, and collected anthropometric (ie, height, BW, body mass index [BMI], waist circumference [WC]), clinical, and metabolic parameters (

Table 1

). In accordance with the Italian Pharmacological Agency, exenatide was added to classical antidiabetic treatment in order to achieve adequate metabolic control.

At baseline, men and women did not differ substantially in preexisting therapy. Thirty-two men were on metformin (6 on metformin alone and 26 in combination with a sulphonylurea); and 2 were on a sulphonylurea (ie, glyburide [15], glimepiride [6], gliclazide [5], gliquidone [1], and repaglinide [1]) monotherapy. Thirty-four women were treated with metformin (10 with metformin alone and 24 in combination with a sulphonylurea); and 1 woman was on a sulphonylurea (ie, glyburide [12], glimepiride [6], gliclazide [6], and repaglinide [1]) monotherapy. One woman treated with a combination of metformin and a sulphonylurea (glyburide) was also in therapy with pioglitazone.

Urine and blood samples were collected in the morning after a 12-hour fast, except for post prandial glucose, which was collected 2 hours after lunch; urine and serum chemistry values were determined in our hospital laboratory with routine methods. Blood pressure (BP) was measured with a certified sphygmomanometer twice each visit after the patients sat for 15 minutes. The mean of the 2 measurements was recorded.

The retrospective study protocol was approved by our independent local ethics committee and was conducted in accordance with the World Medical Association Declaration of Helsinki.⁴

Statistical Analysis

We considered only patients with persistent exenatide treatment (N = 69); of those, 36 (19 men and 17 women) reached a 2-year period of follow-up. We did not consider (a priori exclusion) those who abandoned the prescribed exenatide therapy, whether for side effects or any personal or clinical reason. Metabolic, clinical, and anthropometric parameters were summarized by means of descriptive statistics for all the subjects and by sex.

To highlight gender-related differences at baseline, we adopted a t test for independent data. To describe the variation over time of a given parameter, we used a repeated-measure model considering the baseline and up to 3 follow-up assessments (month 4, 8, and 12 from baseline), without specifying any functional trend for time. Age and diabetes duration were included as covariate into the model. The covariance structure was a compound symmetry; we were not able to find meaningful changes when more complex models were adopted. In

Table 2

, we report the paired t test to assess whether the change from baseline at any given visit was different from 0. We then added a sex

×

visit interaction term to the model and tested whether there was a change at 12 months from baseline for a given parameter differed by gender. The above-mentioned analysis was replicated: 1) excluding those with some missing information at any point in time and 2) adding the 24-month visit for those subjects with a 24-month follow-up period (n = 36). All analyses were conducted using SAS PROC MIXED version 9.1.3 software (SAS Institute, Cary, North Carolina).

RESULTS

Men and women showed typical gender differences in height and high-density lipoprotein cholesterol (where HDL-C is generally higher in women) (Table 1). BMI and low-density lipoprotein cholesterol were both higher in females with diabetes than in males with diabetes (P = .04 and P = .05, respectively).

As shown in Table 2, exenatide rapidly reduced A1C, fasting plasma glucose (FPG), BW, and WC in both men and women (all P <.0001). Systolic and diastolic BP did not significantly change after the first year of observation. When considering gender distribution of A1C, FPG, and BP, we did not observe any difference between males and females; however, both BW and WC were more reduced in women than in men (

Table 3

). Similar findings were obtained when we excluded those subjects with some missing information (data not shown).

Furthermore, focusing our attention on patients with persistent exenatide treatment for 24 months (

Figure

), we recorded stable and significant reductions of A1C and FPG, with a progressive decline of BW (—4.4 kg, –5.5 kg, –5.9 kg, and –6.7 kg, respectively, after 4, 8, 12, and 24 months; P <.0001 vs basal value at each time of observation). Once again, dichotomizing our patients by gender, we did not find any differences in the reduction of A1C and FPG, although BW lowering was more rapid and pronounced in women compared with men (

Table 4

).

DISCUSSION

Women constitute 51.45% (approximately 30 million) of the Italian population. Currently, diabetes is more frequent among Italian females than males (5% prevalence vs 4.6%), and it is estimated that nearly 3 million Italians suffer from this metabolic disease.⁵ The prevalence of T2DM increases proportionally with age, and the difference in life expectancy may account for the different prevalence of diabetes between genders, with the average lifespan being 6 years longer in women than in men.

Furthermore, CV disease is the leading cause of death among women in Italy, with >40% of deaths attributed to this disease, and accounting for 50% of deaths in females 65 years or older. For this reason, conditions in women require particular attention as far as both prevention and therapy are concerned. Gender differences may also be relevant in the management of T2DM, as data from the Diabetes Prevention Program showed some sex-specific differences,⁶ while other population trials have underlined the differences between men and women in the effects of various CV risk factors.^7,8 Additionally, the relative risk of fatal coronary heart disease associated with diabetes was found to be 50% higher in women than in men.⁹ In a study of outpatients with diabetes,³ women with T2DM had a worse CV risk profile and achieved therapeutic goals less frequently than did men.

“Add-on” therapy with exenatide may provide an advantage, as suggested by some studies^10,11 in which adjunctive exenatide treatment resulted in sustained improvement in glucose metabolism and anthropometric parameters, as well as in CV risk factors. In our study, we observed a uniform reduction of A1C and a decrease of FPG in the group treated with exenatide, even if not at levels reaching the desired therapeutic targets (Table 2; Figure). However, the observed A1C reduction was even better than that recorded by the Italian report on new antidiabetic drugs (Δ = 0.93%).¹² Although we cannot exclude that our findings could at least in part be explained by lifestyle differences, gender-related psychological/body image issues, and the fact that females with diabetes had significantly higher BMIs at baseline compared with the male patients, we observed that BW and WC reductions were more pronounced in women than in men after 12 months of exenatide treatment (Table 3). Our patients showed a reduction of BW with a significant gender difference (—7.6 kg in women and –3.4 kg in men; P = .0001) (Table 2); women also had a greater reduction of waistline (—6.5 cm) in comparison with men (–3.6 cm; P = .02) (Table 3).

In those completing 24 months of treatment, we did not record any significant gender difference in the reduction of A1C or of FPG (Figure; Table 4). However, on weight measures, females showed a reduction of —8.7 kg versus –3.7 kg in men (P = .002) (Table 4). Our real-world data, while confirming the acceptable glycemic efficacy of exenatide in both sexes, demonstrated better anthropometric response in women with diabetes.

A recent retrospective meta-analysis of 16 studies, including 2067 patients with diabetes treated with exenatide, did not show any gender difference regarding glycemic control, although the observed weight loss was significantly higher in females than in males.¹³ To potentially explain the observed gender difference in our patients, it was suggested that in normal-weight individuals, the elicited secretion of endogenous GLP-1 by means of fiber-rich nutrients is greater in women.¹⁴

Pharmacologically, possible interference with previous classic antidiabetic treatment appears unlikely, because similar proportions of men and women were treated with metformin as monotherapy, and metformin and a sulphonylurea in combination. Moreover, at the end of the observation (24-month completers), we registered the suspension of a sulphonylurea in 6 men and in 6 women, and a sulphonylurea dose-reduction in 9 men and in 2 women; however, men demonstrated no advantage of this reduction in terms of a major decrease in BW.

We did not add exenatide to insulin-treated T2DM patients because of a specific contraindication defined by the Italian regulatory system. However, the addition of GLP-1 analogues to insulin in patients with T2DM should be associated with a reduction in A1C, BW, and insulin dose, along with a low risk of hypoglycemia and high treatment satisfaction, as reported by a recent Swedish study.¹⁵

According to efficacy-based guidelines for prevention of CV disease in females, weight management (among other recommendations) has to be implemented for all women, because obesity is one of the most important risk factors for developing T2DM, and because the presence of diabetes increases a woman’s risk for CV disease.¹⁶ GLP-1 RAs (eg, exenatide twice daily, exenatide once a week, liraglutide, lixisenatide) and dipeptidyl peptidase-4 (DPP-4) inhibitors (eg, sitagliptin, vildagliptin, saxagliptin, linagliptin, alogliptin), presently available in Italy, are new drugs that can significantly reduce A1C if given in monotherapy or in combination with other antidiabetic drugs. These incretin-based therapies have very low risk for the development of hypoglycemia, and either decrease BW (GLP-1 analogues) or have a neutral effect on it (DPP-4 inhibitors).¹⁷ The low risk of hypoglycemia is a very important advantage because, according to the Action to Control Cardiovascular Risk in Diabetes (ACCORD) Study,¹⁸ a higher mortality is associated with hypoglycemia in both intensive glycemic and conventional treatment arms. Therefore, when there is a need to intensify glucose control, hypoglycemia may increase CV risk and counter the benefits of reducing hyperglycemia.

Moreover, incretin drugs offer considerable promise as a means of targeting the primary defect in T2DM: exenatide (and other GLP-1 RAs) exert positive beneficial effects on the β-cells, thereby limiting the disease progression and the development of its complications.¹⁹ Furthermore, exenatide improved glycemic control, reduced BW, and was associated with a lower risk of CV events in comparison with other hypoglycemic agents.²⁰ In a recent observation from another Italian group,²¹ target glycemic response was achieved in a significantly higher proportion of males than females, while target weight loss—arbitrarily expressed as 1-year percent loss ≥75th percentile in the whole population (8.5%)&mdash;was more often achieved among females at 8 months (28% vs 15%; χ² = 8.04; P = .004) and 12 months (33% vs 17%; χ²= 10.98; P = .0009). After substituting body weight with BMI, the results did not change.

These observations highlight a gender difference in the anthropometric evolution during exenatide treatment. We observed a similar A1C evolution in both genders with diabetes, with a more favorable reduction in BW and WC in women, confirming the observation of the Italian group²¹ and that in the analysis by Pencek and colleagues,¹³ in which the final weight loss observed in patients was likewise significantly higher in females than in males. However, although the evolution of glycemic control and BW loss after 12 months of exenatide therapy should be predicted in different ways between genders in T2DM,²¹ the intrinsic mechanisms of weight loss obtained with GLP-1 receptor agonists are not fully understood. The higher mean BMI observed in women (Table 1) may have contributed to a greater weight loss after exenatide, but this difference is not reported in the population of Anichini and colleagues.²¹ Moreover, GLP-1 RAs decrease the rate of gastric emptying, reduce appetite, and promote satiety, but they may also provide changes in energy expenditure and in leptin sensitivity, or cause nausea resulting in decreased food intake—factors all leading to adipose tissue reduction.

Lately, short-term exenatide treatment has been associated with modest weight loss and decreased WC in a cohort of obese nondiabetic women.²² Recent data suggest that, during a meal, GLP-1 can simultaneously exert an incretin effect on insulin secretion and a protective effect on endothelial function, reasonably controlling oxidative stress generation.²³ The prospect of a CV protection in patients treated with exenatide is very encouraging because these patients usually have other comorbidities in addition to T2DM. Women actually have fewer basal risk factors than men, but according to an Italian report, they are more sedentary and are diagnosed to have diabetes 3 years later than men.²⁴ Furthermore, a recent study reports that women are more obese than men at the time of diagnosis of diabetes.²⁵ So, we may hypothesize that females should reveal a greater sensitivity to reducing weight after exenatide therapy.

Limitations and Strengths

There are several limitations of this study, primarily the small number of studied subjects and the inclusion of only “completers” in the analysis. Another limitation is the study’s retrospective observational design and the lack of a control group. Possible confounding factors to be considered in the results’ evaluation are increased physical activity or changes in eating habits due to initial favorable results obtained with the novel therapy, which could partially explain the noted effects. General side effects such as nausea and vomiting were not studied in detail since we considered only patients with persistent exenatide treatment; those who abandoned the prescribed therapy were excluded a priori. Moreover, we were not able to record hypoglycemic events, once again because of the retrospective design of the study. On the other hand, in our real-world study population, we did not observe any major adverse event such as pancreatitis or pancreatic cancer.

A strength of the study is the quite long period of observation, and the opportunity to stratify our patients by gender, partially confirming and reinforcing previous observations.^13,21 Finally, it is to be considered that this investigation took clinical settings into account; new incretin-based therapies cannot, however, be detached from lifestyle modification. Preventing CV disease is fundamental in both sexes, especially in women who tend to be less physically active than men. The most dangerous risk factor for women, in Italy, is not considering CV disease as pathology particularly relevant to females. Treatment strategies should be improved in both sexes, but women with diabetes should receive a more aggressive therapy, especially when CV complications are present.

CONCLUSIONS

Persistent exenatide treatment as “add-on” therapy in T2DM may show a gender difference in terms of clinical results. In our clinical practice, as a result of the study, we suggest exenatide as the drug of choice in obese or overweight patients suffering from T2DM, with the likelihood that women with diabetes should experience a more prominent therapeutic effect (in comparison with men) on some of their CV risk factors but, due to the observational retrospective design of the present study, any suggested gender difference during treatment with GLP-1 RAs needs confirmation by trials designed specifically to address this topic.