Low-intermediate dose testosterone replacement therapy by different pharmaceutical preparations improves frailty score in elderly hypogonadal hyperglycaemic patients

Abstract

An open-label follow-up study of low-to-intermediate dose testosterone replacement therapy (TRT) was conducted in 64 overweight patients (aged 65–75 years) with late onset hypogonadism (LOH) and increased fasting plasma glucose (FPG). Patients were subdivided into four treatment groups: oral testosterone (T) (T undecanoate, 80 mg/d), transmucosal T (60 mg/d), transdermal T (30 mg/d) or no treatment (control), and evaluated at 0 and 6 months. FPG, hemoglobin (Hb), prostate-specific antigen (PSA) and total T were measured and the Homeostasis Model Assessment of Insulin Resistance (HOMA-IR) index was calculated. Body mass index (BMI), waist circumference, fitness level (6-min walking test), Aging Males’ Symptoms (AMS) scale, handgrip strength and energy expenditure with physical activity (Minnesota questionnaire for Leisure Time Physical Activity (LTPA)) were evaluated and a “frailty score” (based on: grip strength, gait speed and LTPA) was calculated. T levels increased in all treatment groups; the oral T group had values still in the hypogonadal range (5.9 ± 1.1 nmol/L). PSA and Hb concentrations did not change in any group. BMI, waist circumference, FPG and HOMA-IR improved in all T-treated groups after 6 months, with a greater effect seen with transmucosal and transdermal T compared with oral T. This study indicates that low-to-intermediate dose TRT may be safely utilized in LOH patients to ameliorate somatic and psychological frailty symptoms in association with improved anthropometric and glycometabolic parameters in aging, overweight men with LOH and impaired fasting glucose.

• Elderly
• frailty
• hyperglycemia
• overweight
• testosterone treatment

Introduction

Late onset hypogonadism (LOH) is common in men over 50 years of age. Nevertheless, it is often overlooked by general practitioners and sometimes by endocrinologists, since gonadal dysfunction may be believed to be a “normal” part of aging. Decreased libido and erectile dysfunction are indeed often considered as typical symptoms of aging per se, although central nervous system (CNS) dysfunction and muscle-related symptoms may appear as subtle early signs of LOH well before the onset of sexual dysfunction. Moreover, the widespread use of 5-phosphodiesterase inhibitors may extend the interval between symptomatic LOH onset and first referral to the endocrinologist, further delaying LOH diagnosis.

Frank LOH is characterized by very low (<8 nmol/L) circulating total testosterone (T), and its physical and psychological consequences (erectile dysfunction and other sexual symptoms, dyslipidemia, depressed mood, etc.) and also include the features of “frailty”, i.e. sarcopenia, low physical performance, hypomnesia and depressed mood [1–3]. When left untreated, frailty may progress to disability, a serious condition that can eventually lead to dependence (loss of self-sufficiency) and death [4].

The pathophysiological mechanisms underlying LOH are not completely understood but certainly involve central and peripheral mechanisms regulating testicular function. Low circulating T is a strong predictor of frailty in older men [5,6]. The frequent association of obesity and diabetes with LOH might indicate a role of the dysmetabolic condition in LOH patients. Functional damage to Leydig cells caused by high leptin output from hypertrophic adipocytes and low-grade inflammation with consequent release of high levels of cytokines (mostly IL-2, IL-6 and TNF-α) are known peripheral mechanisms occurring in overweight patients [7,8]. In addition, a decreased level of sex hormone-binding globulin may also lead to lower T plasma levels in obese men [9]. Moreover, central mechanisms of impaired hypothalamic-pituitary gonadal regulation are probably involved in LOH, due to enhanced conversion of androgens into estrogens by high aromatase activity within abdominal adipose tissue [8–10] and blunted hypothalamic kisspeptin-releasing neuron activity [11].

Nowadays, the term LOH also includes an intermediate condition characterized by more subtle physical and psychological changes associated with T levels falling within the so called “grey-area” (8 to 12 nmol/L) [1]. The prevalence of this milder form of LOH is as high as 40%–60% in obese elderly men, especially those with impaired fasting glucose (IFG, fasting glucose: 101–125 mg/dL), impaired glucose tolerance (IGT, plasma glucose >140 mg/dL 120 min after 75 g glucose ingestion), or type 2 diabetes mellitus [12].

Several papers have been published recently showing a positive effect of testosterone replacement treatment (TRT) on glucose metabolism [13], but the results obtained with TRT in improving the symptoms of frailty are still controversial. While a modest improvement of sexual symptoms with TRT is generally well established, the positive effect of TRT on muscle strength, mood and energy expenditure with physical activity is more controversial [14]. Moreover, the prostate safety of TRT in elderly patients has been questioned. The aim of this study was to verify whether oral, buccal and transdermal TRT administered at a dosage lower than that commonly utilized in young hypogonadal men, improves frailty symptoms in elderly men with visceral adiposity, IFG and LOH.

Patients and methods

From January 2009 to June 2012, 600 elderly male outpatients aged 65 to 75 years with visceral adiposity (waist girth [W] 103–118 cm) and IFG were included in this 6-month open-label study. Among these 600 patients, 76 received a diagnosis of frank LOH, based upon T < 8 nmol/L and the presence of at least three clinical LOH symptoms, as defined by Wu et al. [2]. Inclusion criteria included no previous (during young to middle age) history of hypogonadism, no history of prostatectomy, no symptoms/signs of prostate cancer and no history of urinary retention. Their baseline prostatic specific antigen (PSA) levels were 0.5–5.4 µg/L. They all had a digital rectal examination performed by the same physician and trans-rectal prostate ultrasound scan when needed to confirm the diagnosis of benign prostate hypertrophy (BPH). Moreover, all patients had a FPG >5.56 mmol/L and none had clinically relevant heart failure, kidney disease or liver disease. Sixty-four subjects gave their informed consent to participate in this study. The study was previously approved by our institute’s Ethical Committee. Patients maintained a balanced diet (300 kcal lower than REE. 55% CHO, 30% fat <30% of which saturated, 15% proteins) for 6 months associated with either no specific treatment (control-untreated-group, n = 16) or TRT (n = 48). The TRT group consisted of three different pharmaceutical preparations (T undecanoate 80 mg/day (40 mg b.i.d.) orally [Andriol®, MSD Italia, Rome], n = 15; or transmucosal T, 60 mg/day (30 mg b.i.d.) [Striant 30®, Sandoz, Rome], n = 18; or transdermal T (30 mg/day [Tostrex 2%® gel, Prostrakan S.r.l., Milan], n = 15). Motivational support was provided by a dedicated psychologist at the time of each clinical visit and by a second psychologist who contacted patients at regular intervals to get information and provide help.

At 0 and 6 months (0 m, 6 m) serum was assayed for fasting plasma glucose (FPG, mmol/L; glucose oxidase method), insulin (mU/mL; electro-chemiluminescence immunoassays (ECLIA) – reference standard NIBSC 66/304, Roche Diagnostics, Milano, Italy); haemoglobin (Hb, g/L; automatic cell-Dyn 4000, Abbott Diagnostics, Abbott Park, IL), PSA (µg/L; ECLIA – reference standard NIBSC 66/304) and TT (nmol/L; ECLIA – Testosterone II, Roche Diagnostics). A Homeostasis Model Assessment of Insulin Resistance (HOMA-IR) was used to evaluate IR and was calculated with the following formula: fasting serum insulin (mU/mL) × FPG (mmol/L)/22.5 [15]. At the same time points the following parameters were recorded: body mass index (BMI kg/m²), W, the six-minute walking test (6-MWT expressed in meters) and the Aging Males’ Symptoms (AMS) scale (reference values <27) [16]. The frail phenotype was defined following the five criteria (weight loss, exhaustion, weak grip strength, slow walking speed and low physical activity) described by Fried [17] and identifying frailty when three or more of these criteria are present. In particular, in our study patients, the frailty score ranged from 3 to 4, since nobody had unintentional weight loss. Therefore, the following parameters were evaluated: grip strength (digital hand dynamometer GRIP-D, Takei Scientific Instruments Co, Tokyo, Japan, cut-off set at 25 kg), gait speed (GS, 4.57 meter distance: max accepted time = 6 s for people taller than 173 cm or 7 s for the others), energy expenditure with physical activity (cut-off for sedentary lifestyle set at 269 kcal/d in women and at 382 kcal/d in men, calculated by the Minnesota questionnaire for LTPA or Leisure Time Physical Activity) and fatigue (days/week of fatigue sensation, cut-off set at 2). The Epworth Sleepiness Scale (ESS) was adopted as a screening tool in all patients.

Parametric and non-parametric statistical analysis was performed as needed – according to Gaussian/binomial distribution or to the presence of non-continuous (discrete) variables – using SPSS 13.0 (IBM, Chicago, IL) for the calculation of means, SDs, paired tests and the ANOVA. Data are presented as mean and SDs.

Results

Tables 1 and 2 summarize the results at months 0 and 6 for the anthropometric/metabolic measures and functional parameters, respectively.

Table 1. Anthropometric measures and metabolic parameters of Control (C), Oral testosterone (O), Buccal testosterone (B) and Transdermal testosterone (TT) groups.

		0 months		6 months		6 months versus 0 months	6 months (treated) versus 6 months (control)
Examined parameter	Groups	Mean	SD	Mean	SD	p<	p<
Age (years)	Control (C)	68.5	3.7	–	–	ns	–
	Oral testosterone (O)	70.0	3.1	–	–	ns	ns
	Buccal testosterone (B)	69.3	3.0	–	–	ns	ns
	Transdermal testosterone (TT)	70.1	2.9	–	–	ns	ns
BMI, (kg/m^2)	Control (C)	28.4	1.9	28.9	1.9	ns
	Oral testosterone (O)	27.7	1.3	27.3	1.1	ns	ns
	Buccal testosterone (B)	29.4	2.4	27.9	2.0	0.001	ns
	Transdermal testosterone (TT)	28.6	2.0	27.2	2.2	0.001	ns
Testosterone (nmol/L)	Control (C)	5.8	0.9	6.0	1.4	ns
	Oral testosterone (O)	5.5	0.8	5.9	1.1	0.001	ns
	Buccal testosterone (B)	5.6	0.8	11.4	2.0	0.001	0.001
	Transdermal testosterone (TT)	5.7	0.8	10.6	2.1	0.001	0.001
W (cm)	Control (C)	108.1	3.1	108.8	4.2	ns
	Oral testosterone (O)	107.0	2.9	104.2	3.1	0.001	0.001
	Buccal testosterone (B)	108.1	3.8	103.2	3.5	0.001	0.001
	Transdermal testosterone (TT)	108.4	3.3	104.0	3.6	0.001	0.001
FPG (mmol/L)	Control (C)	6.08	0.33	5.94	0.46	ns
	Oral testosterone (O)	5.98	0.29	5.62	0.34	0.01	ns
	Buccal testosterone (B)	6.19	0.28	5.73	0.43	0.001	ns
	Transdermal testosterone (TT)	6.1	0.31	5.54	0.48	0.001	ns
HOMA-IR	Control (C)	4.2	1.0	3.9	1.3	ns
	Oral testosterone (O)	4.2	0.9	2.8	0.8	0.001	ns
	Buccal testosterone (B)	3.9	0.9	2.4	0.5	0.001	0.001
	Transdermal testosterone (TT)	4.1	1.0	2.2	0.6	0.001	0.001

BMI, Body Mass Index; W, Waist Girth; FPG, Fasting Plasma Glucose; HOMA-IR, Homeostasis Model Assessment of Insulin Resistance; ns, not significant.

Table 2. Functional parameters of Control (C), Oral testosterone (O), Buccal testosterone (B) and Transdermal testosterone (TT) groups.

		0 months		6 months		6 months versus 0 months	6 months (treated) versus 6 months (control)
Examined parameter	Groups	Mean	SD	Mean	SD	p<	p<
AMS score	Control (C)	46.8	12.2	46.1	12.5	ns
	Oral testosterone (O)	48.3	9.9	43.6	9.3	0.001	ns
	Buccal testosterone (B)	46.4	7.5	37.7	7.0	0.001	0.001
	Transdermal testosterone (TT)	47.2	8.1	35.9	8.3	0.001	0.001
Handgrip strength (kg)	Control (C)	26.8	1.5	27.1	2.3	ns
	Oral testosterone (O)	26.8	1.5	28.9	1.6	0.001	0.02
	Buccal testosterone (B)	25.7	2.2	32.4	6.3	0.001	0.001
	Transdermal testosterone (TT)	26.1	1.8	33.1	6.3	0.001	0.001
6MWT (m)	Control (C)	269.6	23.5	264.6	20.1	ns
	Oral testosterone (O)	274.3	17.2	325.7	35.0	0.001	0.02
	Buccal testosterone (B)	276.4	33.0	352.3	34.9	0.001	0.001
	Transdermal testosterone (TT)	271.3	25.7	349.1	35.2	0.001	0.001
LTPA (kCal/d)	Control (C)	371.2	60.0	395.0	85.5	ns
	Oral testosterone (O)	357.5	62.0	432.8	61.6	0.001	0.01
	Buccal testosterone (B)	357.1	76.1	532.2	129.3	0.001	0.001
	Transdermal testosterone (TT)	360.3	63.2	491.6	98.1	0.001	0.001
Frailty score	Control (C)	3.2	0.5	3.5	0.6	ns
	Oral testosterone (O)	3.5	0.6	2.4	0.6	0.001	0.05
	Buccal testosterone (B)	3.9	0.6	0.9	0.6	0.001	0.001
	Transdermal testosterone (TT)	3.6	0.6	1.0	0.5	0.001	0.001

AMS, Aging Males’ Symptoms rating scale; 6-MWT, Six-Minute Walking Test; LTPA, Leisure Time Physical Activity; ns, not significant.

Baseline levels were not different among the four groups. No change occurred in PSA in any groups during the observation period (C, 4.2 ± 0.7 versus 4.2 ± 0.9 µg/L; O, 1.5 ± 1.0 versus 1.6 ± 1.0 µg/L; B, 0.7 ± 0.3 versus 1.0 ± 0.8 µg/L; TT, 0.9 ± 0.4 versus 0.9 ± 0.6 at 0 m and 6 m, respectively) or Hb levels (C, 130.2 ± 8 versus 129 ± 7 g/L; O, 129 ± 9 versus 133 ± 10 g/L; B, 132 ± 9 versus 134 ± 12 g/L; TT, 129 ± 9 versus 133 ± 11 g/L at 0 m and 6 m, respectively). Conversely, as clearly shown in Table 1, T circulating levels increased significantly in all TRT groups (p < 0.001) reaching significantly higher levels in B and TT than in O (p < 0.001), while remaining unchanged in group C.

An improvement in antropometric (BMI and W reduction) and metabolic parameters (FPG and HOMA-IR reduction) was observed in the T-treated groups. This improvement was more pronounced and highly significant (p < 0.001, see Table 1) with transmucosal and transdermal delivery.

All of the examined functional parameters were significantly improved (p < 0.001) in the three TRT groups. Finally, the Frailty Score, unchanged in the control group, significantly decreased in all three TRT groups (p < 0.001), with a more consistent reduction occurring in the B and TT groups. ESS did not score higher than 9 in any patients either before or after TRT.

Discussion

The results of this open-label follow-up study indicate that low-to-intermediate dose TRT delivered using three different pharmaceutical preparations, and administered for 6 months, produced the following effects in hyperglycemic overweight elderly men with frank LOH: (i) decrease of FPG, HOMA-IR, BMI and W, (ii) improvement of AMS scale, physical performance as evaluated by LTPA, and frailty score, and (iii) no increase in PSA or Hb.

The positive effect exerted by TRT on glucose metabolism in our patients with fasting hyperglycemia and altered HOMA-IR is in agreement with previous studies in LOH patients with overt diabetes and metabolic syndrome [8,18–21]. In particular in a multicenter, prospective, randomized, double-blind, placebo-controlled study, the TIMES2 study [8], transdermal 2% testosterone gel was evaluated in 220 hypogonadal men with type 2 diabetes and/or metabolic syndrome and produced a significant improvement in glycemic control (HOMA-IR reduction in the overall population by 15.2% at 6 months).

Different mechanisms are implicated in T’s effect on glucose metabolism and obesity [22]. We have shown in this study that TRT, even at low-to-intermediate doses, results in a significant reduction of waist circumference and BMI in LOH patients. The increased energy expenditure observed in the treated patients might have contributed to an improvement in glucose utilization by the muscles and a reduction in fat mass.

Clinically, the most significant result of this study is the improvement with TRT of functional parameters in LOH patients. The longer distance covered during the six-minute walking test by treated patients in our study reflects an improvement in cardiorespiratory function as well as the observed increase in skeletal muscle mass and strength. The higher LTPA energy expenditure was also strongly linked to these improvements and, together with reduced fatigue commonly experienced with training, contributed to reduced risk of frailty, as reflected by the observed decrease in the frailty score [23]. Metabolic diseases and LOH are both known risk factors for frailty. Dysmetabolism can lead to frailty by increasing cardiovascular risk and causing cognitive defects, while LOH weakens muscle and bones as well as central nervous system performance, thus causing memory and mood disturbances among the rest [24]. The significant decrease in frailty score observed after six months of TRT in our study was clinically impressive given that it was obtained with low-to-intermediate doses of TRT, which produced circulating T levels in the low to low-normal range. Interestingly, transmucosal and transdermal T preparations resulted in higher serum T levels than the oral T preparation, probably because of the differing pharmacokinetic profiles of the former compared to the latter, which is known to produce sharp plasma peaks with lower inter-peak T levels [25].

In agreement with our data, Yamaguchi [26] demonstrated in a cohort of 56 LOH patients that TRT improved both AMS scale and androgen levels. A large placebo-controlled, multicenter study compared the effect of three different doses (80, 160 and 320 mg/day) of oral T undecanoate for 12 months in LOH patients and found no significant changes in the total AMS score in treated patients compared to placebo controls, although there was a consistent trend for improvement across the three doses tested [14]. A significant improvement was observed relative to placebo in the sexual symptoms domain of AMS score in the 160 mg/day group [14]. The lack of significant effect on AMS score in this previous study, and thus the discrepancy with the study here presented, was probably due to the specific patient examined in the previous study, since patients in the multicenter study had a lower BMI than those in our study, their glycemic status was not evaluated, and the mean baseline circulating T level (12.8 nmol/L) was at least two times greater than in our patients and was above the lower limit of the normal range, usually fixed at 10 nmol/L and generally recognized as being 12 nmol/L by all relevant scientific societies [1]. Additionally, the sample size of the previous multicentre study may have also not been large enough to detect a significant treatment effect on AMS score. TRT proved to be effective on a number of clinically meaningful domains of the quality of life assessment (sexual and cognitive functions and muscular strength). It appears therefore that TRT treatment is suitable to improve multiple frailty symptoms, although evidence-based information about the duration of treatment and the long-term benefits and risks of TRT in aging men is, however, still to be fully clarified [27]. It should be noted that our study shows some limitations due to the specific design adopted, such as the lack of randomization and of a placebo group.

The safety of TRT substitution in the elderly is an important consideration for physicians. Although elevation of the serum T to the normal range has not been shown to result in an increased risk of developing prostate cancer, the stimulatory effects of T on existing prostate cancer are well established. Therefore, a careful screening of the prostate should be undertaken prior to starting TRT (PSA evaluation, digital rectal exploration and eventually trans-rectal prostate ultrasound scan). Several studies [28–30], including our own, have demonstrated that TRT does not result in a significant increase in serum PSA levels. Moreover, the incidence of cancer in LOH patients on TRT is not different from that in the general population [30]. On the other hand, a protective effect of T on metabolic syndrome-related inflammation has been recently shown in rabbit prostate through a newly discovered mechanism, which might be proven to be active in the human as well in the future [31]. Polyglobulia, another risk factor of TRT, was not present in our patients; this side effect was not expected in our experimental setting, since it typically occurs only when very high doses of testosterone are chronically administered to healthy young subjects for doping purposes [32,33]. On the contrary, in LOH patients who usually present with low-normal Hb concentrations (mostly less than 130 g/L), TRT might be expected to produce an improvement in Hb concentration.

In summary, the results of this study indicate that low-to-intermediate dose TRT improves somatic and psychological frailty symptoms in association with improved anthropometric and glycometabolic parameters in overweight aging men with LOH and IFG. Further investigations are warranted to verify the effects of low-to-intermediate dose TRT over a longer period of time.

Declaration of interest

The authors report no declarations of interest.

The publication of the present paper has been made possible by research funds provided by the Italian Space Agency (ASI) through INRCA, Rome, research contract I/010/11/0 for the study of the endocrine and metabolic effects of isolation-confinement (experiment MARS 500).