Besides the primary endpoints of incident type 2 diabetes (diagnosed as 2-hour glucose in the diabetic range, ≥11.1mmol/L on 75g OGTT), and change in post-prandial 2-hour glucose after 2 years of treatment with testosterone undecanoate injections vs. placebo, the following secondary endpoints will be examined:
The reasons for measuring these outcomes are described below.
Normalised blood glucose (2-hour glucose < 7.8mmol/L)
Initiation of antidiabetic pharmacotherapy
Glucose metabolism: fasting plasma glucose, insulin and HbA1c
Anthropometrics: body weight, waist circumference
Body composition: DEXA measurements of whole-body and regional fat and lean body mass
Muscle strength: handgrip
Sex steroid hormone profile: testosterone, DHT, estrone, estradiol and SHBG
Sexual function and lower urinary tract symptoms (LUTS)
Biomarkers for metabolic function: lipids (total cholesterol, LDL, triglycerides, HDL)
Compliance with the lifestyle intervention program
Health care expenditure
Additional markers of glycemic status
The percentage of men with normal glucose levels (2-hour glucose < 7.8 mmol/L), initiation of antidiabetic pharmacotherapy, fasting plasma glucose, insulin and HbA1c.
Mechanisms underlying the effects of testosterone therapy
a. Improved body composition reflected by a decrease in total and/ or abdominal fat mass and increase in lean mass (DXA). Two outcomes related to body composition improvement will also be analysed: muscle strength (measured by hand grip dynamometry) and insulin resistance (by HOMA-IR, based on fasting plasma glucose and insulin).
b. Enhanced adherence to the Weight Watchers® program, as reflected by attendance at groups, use of the online program, or both, and its relationship to weight loss.
c. Increase in physical activity as assessed by the Active Australia questionnaire.
The effects of testosterone therapy may be attributable either to testosterone treatment, improvement in health behaviours and weight loss, or an interaction between testosterone therapy, health behaviours and weight loss.
These outcomes include: erectile function, sexual desire and LUTS, assessed by the International Index of Erectile Function (IIEF-5), Sexual Desire Inventory, and IPSS respectively. These questionnaires were chosen because they were used in the Men in Australia Inflammation Lifestyle, Environment and Stress (MAILES) study.33
Treatment impact on psychosocial factors
a. Health-related quality of life will be assessed by the Short-Form Health Survey (SF-12) to test the hypotheses that Health-related quality of life will improve over time and be greater at study completion for men who had received testosterone therapy vs. placebo for 2 years.
b. Psychosocial function and motivation for lifestyle change will be assessed by the MacArthur Scale of Subjective Social Status, Pearlin’s Personal Mastery Scale, and Sense of Coherence (a factor in determining how well a person manages stress and stays healthy), to test the hypothesis that subjective social status, mastery, and sense of coherence will improve over time in men and be greater at study completion for men who had received testosterone therapy vs. placebo for 2 years.
To determine whether psychosocial measures mediate the impact of testosterone on glycemic status
The hypothesis being tested is that improvements in subjective social status, mastery, and sense of coherence will partially or fully mediate (account for, in the causal pathway) the impact of T on glycaemia.
To determine whether the relationship between testosterone and glycemic status varies by sociodemographic measures
The hypothesis being tested is that greater education, greater household income, higher status occupation, or being married or cohabitating will strengthen the relationship between testosterone and glycemic status and, conversely, lesser values for these measures will be associated with a lesser strength of the relationship between testosterone and glycemic status.
Association of treatment effects with baseline and change in sex steroid concentrations
Circulating testosterone, DHT, estradiol and estrone will be measured by LC-MS/MS at baseline, 18, 66, and 102 weeks. Serum SHBG, follicle stimulating hormone (FSH) and luteinising hormone (LH) will be measured by immunoassay.
Treatment impact on health care expenditure
Costs of prescribed testosterone therapy (Reandron®), costs of hospitalisations, and costs of general practitioner visits will be used to analyse impact of testosterone therapy on health care expenditure.
Cost-effectiveness per incident type 2 diabetes prevented, per death prevented, life year gained and quality-adjusted life year (QALY) gained will be analysed.
Three sub-studies of T4DM will be conducted to determine the effects of testosterone therapy on:
- Bone micro-architecture (measured by peripheral quantitative computed tomography, pQCT), and bone mineral density (measured by DXA) (T4Bone).
- Motivation and behaviour (T4M&B).
- Telomere length (T4Telomeres).
Two additional sub-studies will investigate the effects of extended testosterone therapy for up to 4 years (T4DM run-on), and rate of HPT axis recovery at the end of the initial 2-year treatment period (T4DM run-off), respectively.