Baseline patient characteristics and retention
From May 2015 to December 2016, 12 patients were enrolled in the UC arm and 13 patients in the PRT arm and were followed up for 12 months. At 6 months, one patient in the UC arm was unable to attend follow-up due to hospitalisation (but remained in the study) and one patient in the PRT arm discontinued the study. At 12 months, two patients discontinued the study (one from each arm) and two patients from each arm were excluded from final analysis as ADT was discontinued for more than 3 months prior to their 12-month visit (one from each arm). Thus, at 12 months, 10 patients remained in each arm for analysis (Figure 1).
Adherence to the home-based PRT program was determined by logbooks. Out of the 13 patients, one patient consistently did not comply with filling in his logbook citing poor literacy as the reason, although he reported full compliance with the PRT program. Logbook analysis showed that from baseline to 6 weeks, patients completed a mean of 2.8 sessions/week; 6 weeks to 6 months, 2.3 sessions/week, and from 6 to 12 months, 2.2 sessions/week. At 12 months, 50% of patients managed to comply with stage 3 exercises; 30% continued with stage 2 and 20% continued with stage 1.
At baseline (table 1), UC and PRT patients were well matched in terms of age, blood pressure, weight, body composition, PSA levels, as well as prostate cancer grade and stage. Testosterone levels were significant higher in the UC group (p = 0.02). There were no significant differences between groups in terms of baseline activity level as assessed by pedometer step counts and hours of light, moderate or intense physical activity.
Table 1. Baseline clinical characteristics
Variable
|
UC (n =12)
|
PRT (n = 13)
|
P value
|
Age (years)
|
71.8 ± 1.8
|
69.3 ± 2.3
|
0.14
|
Weight (kg)
|
79.9 ± 2.7
|
87.2 ± 4.7
|
0.03
|
BMI kg/m2
|
28.8 ± 1.1
|
29.7 ± 1.3
|
0.36
|
SBP (mmHg)
|
137 ± 2
|
139 ± 4
|
0.18
|
DBP (mmHg)
|
74 ± 3
|
74 ± 1
|
0.11
|
Gleason score
|
7.6 ± 0.3
|
8.2 ± 0.2
|
0.42
|
Cancer staging
|
|
|
|
Localized (n)
|
9
|
8
|
0.36
|
Biochemical recurrence (n)
|
3
|
3
|
Metastatic (n)
|
0
|
2
|
Previous radiotherapy (n)
|
8
|
5
|
0.39
|
Previous ADT (n)
|
0
|
2
|
0.26
|
Lean body mass (kg)
|
53.1 ± 1.2
|
54.5 ± 2.2
|
0.16
|
LBM (% body weight)
|
66.5 ± 1.8
|
63.2 ± 2.0
|
0.68
|
Fat mass (kg)
|
24.2 ± 2.1
|
29.9 ± 3.2
|
0.12
|
Extracellular water (L)
|
19.0 ± 0.6
|
19.8 ± 0.9
|
0.05
|
BCM (kg)
|
34.1 ± 1.0
|
34.7 ± 1.3
|
0.58
|
Testosterone (nmol/L)
|
16.1 ± 1.1
|
12.5 ± 0.8
|
0.02
|
LH (mIU/mL)
|
7.0 ± 0.7
|
6.3 ± 0.7
|
0.70
|
PSA (ng/mL)
|
11.9 ± 2.6
|
7.7 ± 1.3
|
0.08
|
Step count (number)
|
40172 ± 8502
|
28838 ± 5377
|
0.14
|
Light physical activity (hours)
|
5.9 ± 1.6
|
7.9 ± 2.5
|
0.09
|
Moderate physical activity (hours)
|
3.9 ± 1.8
|
3.3 ± 1.4
|
0.93
|
High physical activity (hours)
|
0.1 ± 0.1
|
0.5 ± 0.3
|
0.05
|
Data are presented as mean ± S.E.M; P-value is for UC vs PRT group; BMI, body mass index; LBM, lean body mass; BCM, body cell mass; LH, luteinizing hormone; PSA, prostate-specific antigen
Response to ADT across the cohort
Following the administration of ADT, as expected, serum testosterone levels significantly decreased (p < 0.0001) by 6 weeks, remaining at castrate levels at 12 months (p < 0.0001) (Figure 2A). Similarly, serum PSA levels fell from 9.7 ± 1.5 ng/mL to 0.4 ± 0.1 ng/mL at 12 months (Figure 2B).
Table 2 summarises the effect of ADT on measured indices throughout the follow-up period. At 6 months, there was a significant change in body composition (p < 0.001) with a decrease in BCM (BCM %) and increase in FM (FM %) as a percentage of total body mass (TBM).
Similarly, there was a significant increase in fasting glucose (p < 0.001), HOMA- IR (p = 0.01), as well as hepatic and muscle insulin resistance (p < 0.01 and p = 0.02, respectively). There was a significant reduction in insulin sensitivity as represented by the Matsuda Index (p < 0.001). Bone mineral density also significantly decreased at the lumbar spine (p = 0.046) as did muscle strength, with a reduction in right hand grip (p = 0.003), left hand grip (p = 0.04) and lower limb strength (p = 0.01). These changes persisted at 12 months. At study completion, there was a -3.6 ± 0.4 % (p <0.001) decrease in BCM and 3.8 ± 0.5 % (p < 0.001) increase in FM. Fasting insulin increased by 6.1 ± 2.1 IU/L (p <0.001). There was a reduction in physical function as reflected by a prolongation in the TUGT by 0.4 ± 0.1 sec (p = 0.02) and a decrease in co-ordinated stability (p < 0.001). Quality of life also fell with a significant reduction in the QLQ-C30 summary score by -3.6 ± 1.3 (p = 0.007).
The effect of ADT on growth factors and adipokines
At 6 weeks, there was a significant reduction in BDNF levels from 8208 ± 1268 pg/mL to 6126 ± 1038 pg/mL (p = 0.003) but significance was not maintained at 6 weeks and 6 months (Figure 3A). At 12 months, there was a significant increase in serum adiponectin compared to baseline (p = 0.006) (Figure 3C) (p < 0.001). There were no significant changes in serum IGFBP-2 and IGF-1 following treatment with ADT. However, there were significant increases in serum leptin (p < 0.001) (Figure 3B), IGFBP-3 (p = 0.02) (Figure 4A) and the IGF-1: IGFBP-3 ratio at 12 months (p = 0.02) (Figure 4B). There was also a negative correlation between serum PSA levels and IGFBP-3 across all time points (R2 = 0.07, p = 0.01) (Figure 4C).
The effect of PRT – results of the randomised intervention
Body Composition
The effect of PRT on changes in body composition is shown in Figure 2. At 6 weeks and 6 months, there was a trend towards a greater reduction in LBM % and BCM % in the UC versus the PRT group but this did not reach statistical significance. However, at 12 months there was a preservation of both LBM % by 2.7 ± 0.8 % (p = 0.001) and BCM % by 1.9 ± 0.8 % (p = 0.02) (Figure 5A) in the PRT versus UC group. Similarly, there was a trend towards a larger gain in FM % at 6 weeks and 6 months in the UC group but this was not statistically significant. At 12 months, patients in the PRT group experienced less of a gain in fat mass, with a -3.1 ± 1.0 % (p = 0.002) difference between the two groups (Figure 5B). When assessing regional fat mass, the biggest difference between groups was observed with trunk mass (p = 0.02) (Table 3).
Glucose Tolerance and Insulin Resistance
At 6 weeks, there was a decrease in the Matsuda Index in the UC group by -0.3 ± 0.5 (p = 0.47) and significant increase in the PRT by 2.2 ± 0.7 (p = 0.009) with an overall difference between groups of 2.5 ± 0.8 (p = 0.004) (Table 4). However, this significant early difference was not maintained at 6 and 12 months. There were no significant differences between groups in terms of plasma insulin and glucose levels, HOMA-IR, disposition index, or liver and muscle insulin resistance.
Blood biomarkers
PRT did not have any effect on blood pressure, liver function or serum lipids at the various time points. There were no differences between groups in terms of PSA levels, IGF-1, IGF-binding proteins, leptin, adiponectin and other measured biomarkers (Table 5).
BMD and physical function
There were no differences in BMD of the femoral neck or lumbar spine between the UC and PRT groups at each of the times points (Table 6). Similarly, PRT had no significant effect on muscle strength, balance, or submaximal VO2 (Table 6). However, there was a significant increase in physical activity levels in the PRT group as measured by step count compared to the UC group at 12 months (p = 0.02) (Table 6). There were no differences in physical activity levels between groups.
Quality of Life
The effect of PRT on HRQOL is summarised in Table 7. At 6 weeks, there was a reduction in the SF36v2 general health score in the UC group and an increase in the PRT group, with a significant difference of 5.0 ± 2.4 (p = 0.04) between the two groups. At 6 months, there were significantly greater reductions in the EORTC QLQ-C30 score for global health status (p = 0.02), SF36v2 scores for vitality (p = 0.02) and social functioning (p = 0.03) in the UC compared to the PRT group. In addition, the EORTC QLQ-C30 score for pain was significantly higher in the UC compared to the PRT group (p = 0.02). At 12 months, the SF36v2 scores for vitality and mental health both improved in the PRT group, as opposed to a reduction in the UC group (Figure 6A and 6B).
Intervention safety and adherence
In terms of adverse events, one patient developed right shoulder pain (rotator cuff tendonitis) at the completion of the study, requiring physiotherapy.