In this study, we assessed the adherence and compliance of the 6-month clinical trial by measuring GTP components in blood and urine of the participants. Our previous studies have validated that measurement of blood and urinary GTP components are good biomarkers for determining green tea consumption and GTP intervention in human population studies 
. Among the four major GTP components, all EGCG, EGC, EC, and ECG, were detectable in blood and urine samples 
, but only EGCG and ECG in serum and EC and EGC in urine showed dose-response correlations 
. Results of serum concentrations of EGCG and ECG and urinary excretion of EC and EGC in this study showed significant elevation after GTP intervention and no detectable concentrations in the baseline and the placebo participants over the study period, which further confirmed the excellent adherence and compliance of the overall trial. In addition, these results were consistent with our previous findings, i.e., blood and urinary GTP components are practical and reliable biomarkers for green tea consumption 
In this study, we further assessed the efficacy of GTP supplement and TC exercise for their individual and combined effects on the oxidative DNA damage biomarker, 8-OHdG, in postmenopausal women with osteopenia in the 6-month clinical trial. Results from this study showed that individual GTP supplement, TC exercise, or the combination significantly decreased urinary excretion of 8-OHdG after 1-, 3-, and 6-months intervention in postmenopausal women, as compared with the placebo group, which demonstrated the efficacy of this trial in reducing the oxidative stress biomarker levels.
Several lines of evidence have indicated the role of ROS in induction of osteoporosis 
: ROS are capable of influencing the generation and survival of osteoclasts, osteoblasts, and osteocytes; ROS-activated FoxOs in early mesenchymal progenitors also leads to decreased osteoblastogenesis through disruption of Wnt signaling pathway 
; ROS also increases serum osteopontin and transforming growth factor-β levels in iron overloaded rats, suggesting osteoclast-mediated bone resorption through the receptor activator of nuclear factor-κB/RANK ligand (RANK/RANKL) mediated signaling pathway 
. In postmenopausal women, the reduced estrogen level might be a key factor in boosting the ROS level because estrogen has been reported to be able to diminish production of ROS and stimulate the activity of glutathione reductase 
. The ability of green tea and GTP as potent ROS scavengers has been well recognized as the basis for its biological activity 
. Previous studies by others and ours showed strong evidence of protective effects of GTP on the capacity of bone formation due to its anti-oxidative stress potentials 
. An animal study shows that urinary excretion of 8-OHdG is a reliable indicator of oxidative stress in postmenopausal rats 
. More importantly, urinary 8-OHdG concentration has also been widely used as a DNA oxidative damage biomarker in many human studies 
, . In this study we observed a significant decrease in urinary 8-OHdG concentrations in postmenopausal women appearing at 1- to 6-months after the GTP intervention as compared to the placebo control group, which further suggests evidence of oxidative stress in the study participants. The result of this study is consistent with the previous finding in HBsAg carrier groups, which also showed statistically significant reduced urinary 8-OHdG after 3-months GTP intervention at 500 mg or 1,000 mg/day 
TC exercise has been characterized as a moderate intensity mind-body exercise, coupling muscular activity with an internally directed focus, producing a temporary self-contemplative mental state 
. In this study, the TC exercise significantly decreased urinary 8-OHdG concentrations after 3-months intervention as compared with the control group. The effect of exercise on oxidative stress has been reviewed 
. It was reported that high-intensity exercise increased oxidative stress biomarkers, including 8-OHdG and malondialdehyde (MDA)-modified low-density lipoprotein in humans, whereas moderate exercise tended to decrease both indices of oxidative stress 
. A recent study has also demonstrated that TC exercise stimulated endogenous antioxidant enzymes (superoxide dismutase, SOD) and reduced oxidative damage markers (MDA) in middle-aged adults 
. These reports support our observation that TC, a moderate exercise, reduced oxidative stress in postmenopausal women.
This is the first study to investigate the combined effects of GTP and TC on oxidative DNA damage in postmenopausal women with osteopenia. An additive effect of GTP supplementation and TC exercise on suppression of oxidative stress was found, as evidenced by a 40% reduction of urinary 8-OHdG in the TC group and 60% reduction in the GTP group alone and a 73% reduction was observed in the GTP+TC group. Results of this study further suggest that a combinative intervention strategy, including dietary supplementation and moderate exercise, may be more effective in promoting bone health in postmenopausal women with osteopenia.
The strength of our study was in its design as a randomized and placebo controlled clinical trial that could limit the biases usually introduced through the design and data acquisition. Limitations to this study included a relatively small sample size that might limit the statistical power, such as parameters after 1-month of intervention. A wide inter-individual variation found in concentration of biomarkers measured may be due to the individual variations in GTP metabolism enzymes, endogenous susceptibility factors, and/or DNA repair capacity; the large variation in urinary 8-OHdG levels (43.9±42.1 ng/mg creatinine) has also been reported previously in females 
, which supports this finding in our study. The stratified randomization and non-parametric analysis used could limit potential biases (randomization) and increase the power of detecting the differences (non-parametric model analysis). The intent to treat (ITT) analysis could benefit the comparison; however, considering the comparable compliance and adherence rates for the treatment groups, the effect of dropouts might be minimal. On the other hand, the exclusion of dropouts into statistical analysis may not severely affect the statistical power as an originally estimated 140 participants (34 participants for each group) would be sufficient for a power of about 0.90 for the outcome of urinary 8-OHdG level.
In conclusion, this study provides evidence that 500 mg GTP supplement, Tai Chi exercise, either alone or together, can effectively reduce an oxidative stress biomarker in postmenopausal women with osteopenia. Based on the putative role of oxidative stress in osteoporosis and the observed beneficial effects on bone health as published previously, these intervention strategies hold the potential as an alternative tool to benefit the bone health in postmenopausal women which deserves further long-term human studies.