Article Text

Extended report
Comparison of Tripterygium wilfordii Hook F with methotrexate in the treatment of active rheumatoid arthritis (TRIFRA): a randomised, controlled clinical trial
  1. Qian-wen Lv1,
  2. Wen Zhang1,
  3. Qun Shi1,
  4. Wen-jie Zheng1,
  5. Xin Li1,
  6. Hua Chen1,
  7. Qing-jun Wu1,
  8. Wan-lan Jiang1,
  9. Hong-bin Li2,
  10. Lu Gong3,
  11. Wei Wei3,
  12. Hui Liu4,
  13. Ai-jing Liu5,
  14. Hong-tao Jin5,
  15. Jun-xiang Wang6,
  16. Xiu-mei Liu7,
  17. Zhen-bin Li8,
  18. Bin Liu9,
  19. Min Shen1,
  20. Qian Wang1,
  21. Xiang-ni Wu1,
  22. Di Liang1,
  23. Yu-feng Yin1,
  24. Yun-yun Fei1,
  25. Jing-mei Su1,
  26. Li-dan Zhao1,
  27. Ying Jiang1,
  28. Jing Li1,
  29. Fu-lin Tang1,
  30. Feng-chun Zhang1,
  31. Peter E Lipsky10,
  32. Xuan Zhang1
  1. 1Department of Rheumatology and Clinical Immunology, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing, China
  2. 2Department of Rheumatology, Affiliated Hospital of Inner Mongolia Medical College, Huhhot, China
  3. 3Department of Rheumatology, General Hospital of Tianjin Medical University, Tianjin, China
  4. 4Department of Rheumatology, Beijing Dongfang Hospital, Beijing, China
  5. 5Department of Rheumatology, Second Hospital of Hebei Medical University, Hebei, China
  6. 6Department of Rheumatology, Third Hospital of Hebei Medical University, Hebei, China
  7. 7Department of Rheumatology, First Hospital of Shanxi Medical University, Shanxi, China
  8. 8Department of Rheumatology, The Bethune International Heping Hospital of Hebei, Hebei, China
  9. 9Department of Rheumatology, The Affiliated Hospital of QingDao University Medical College, Shandong, China
  10. 10Formerly National Institute of Arthritis and Musculoskeletal and Skin Diseases, National Institutes of Health, Bethesda, MD, USA
  1. Correspondence to Dr Xuan Zhang, Department of Rheumatology and Clinical Immunology, Peking Union Medical College Hospital, Chinese Academy of Medical Sciences and Peking Union Medical College, Beijing 100730, China; zxpumch2003{at}sina.com and Dr Peter E Lipsky, Formerly National Institute of Arthritis and Musculoskeletal and Skin Diseases, National Institutes of Health, 9000 Rockville Pike, Bethesda, MD 20892, USA; peterlipsky@comcast.net

Abstract

Objectives To compare the efficacy and safety of Tripterygium wilfordii Hook F (TwHF) with methotrexate (MTX) in the treatment of active rheumatoid arthritis (RA).

Methods Design: a multicentre, open-label, randomised controlled trial. All patients were assessed by trained investigators who were unaware of the therapeutic regimen. Intervention: 207 patients with active RA were randomly allocated (1:1:1) to treatment with MTX 12.5 mg once a week, or TwHF 20 mg three times a day, or the two in combination. At week 12, if reduction of the 28-joint count Disease Activity Score (DAS28) was <30% in the monotherapy groups, the patient was switched to MTX+TwHF. The primary efficacy point was the proportion of patients achieving an American College of Rheumatology (ACR) 50 response at week 24.

Results 174/207 (84.1%) patients completed 24 weeks of the trial. In an intention-to-treat analysis, the proportion of patients reaching the ACR50 response criteria was 46.4% (32/69), 55.1% (38/69) and 76.8% (53/69), respectively, in the MTX, TwHF and MTX+TwHF groups (TwHF vs MTX monotherapy, p=0.014; MTX+TwHF vs MTX monotherapy, p<0.001). Similar statistically significant patterns at week 24 were found for ACR20, ACR70, clinical Disease Activity Index good responses, EULAR good response, remission rate and low disease activity rate. Significant improvement in the Health Assessment Questionnaire and 36-item Short-Form Health Survey questionnaire scores from baseline to week 24 was seen in each treatment arm (p<0.05), though no significant difference was found among the treatment arms (p>0.05). The result of per-protocol analysis agreed with that seen in the intention-to-treat analysis. Seven, three and five women in the TwHF, MTX and combination groups, respectively, developed irregular menstruation (TwHF vs MTX monotherapy, p=0.216).

Conclusions TwHF monotherapy was not inferior to, and MTX+TwHF was better than, MTX monotherapy in controlling disease activity in patients with active RA.

Trial registration number NCT01613079.

View Full Text

Statistics from Altmetric.com

Video abstract

Introduction

Rheumatoid arthritis (RA) is a chronic autoimmune/inflammatory disease, which results in joint damage and loss of physical function. It affects 0.5–1.0% of adults in industrialised countries and causes absence from work and economic loss.1 ,2 Disease-modifying antirheumatic drugs (DMARDs), including biological agents, are used to treat RA. However, for a proportion of patients this treatment does not produce an adequate response. Moreover, the extensive application of biological therapy is restricted because of its relatively high cost.3

Tripterygium wilfordii Hook F (TwHF) is used in traditional Chinese medicine for the treatment of joint pain, fever, chills, oedema and local inflammation. The chloroform–methanol extract of the roots of TwHF has been investigated as a potential treatment for autoimmune diseases and malignancies.1 ,4–6 In China, TwHF is approved and often used to treat RA. Recently, extracts of TwHF have also been tested in the West, with good efficacy noted.7 ,8

More than 300 compounds have been identified from extracts of TwHF. Of these, many are diterpenoids, three of which dominate the chemical profile and the medicinal chemistry of TwHF, including triptolide, tripdiolide and triptonide.9 It has been shown that the extracts of TwHF have anti-inflammatory and immunosuppressive activities both in vivo and in vitro.10 ,11 Many of the anti-inflammatory and immunoregulatory activities of extracts of TwHF relate to the ability of the major diterpenoids to suppress the transcription of cytokine and other proinflammatory genes.12 ,13 Importantly, three randomised controlled clinical trials have indicated the clinical efficacy of the extracts of TwHF for patients with RA compared with placebo or sulfasalazine,7 ,8 ,14 ,15 although most of these trials were conducted in a small number of patients. Importantly, none of these studies compared the efficacy of TwHF with methotrexate (MTX), the most commonly used DMARD in Western countries.

In Peking Union Medical College Hospital (PUMCH), the tertiary referral centre in China, we treat more than 30 000 patients with RA each year and of these, about two-thirds are treated with TwHF, in most cases in combination with MTX, because of the low cost of TwHF (around US$10 a month). Importantly, in ‘real-world’ clinical practice, we have observed the considerable effectiveness of the MTX+TwHF combination,16 but this efficacy has not been studied in randomised controlled trials (RCTs). This prompted us to conduct the TRIFRA (Comparison of Tripterygium wilfordii Hook F with methotrexate in the treatment of active rheumatoid arthritis) study, an investigator-initiated, 24-week, multicentre RCT to determine whether TwHF is as effective as MTX and also whether the MTX+TwHF combination might be better than MTX alone. The risk of reversible amenorrhoea is of great concern when TwHF is used for the treatment of patients with RA, so potential subjects were informed of this risk and monitored carefully for the possible development of this adverse event. This TRIFRA trial also evaluated other aspects of the safety profile of TwHF in patients with RA.

Methods

Design overview

This open label 24-week RCT was conducted between May 2012 and June 2013. Written informed consent was obtained from each patient participating in this study. The protocol was approved by the institutional ethical review board. This trial is registered with ClinicalTrials.gov (NCT01613079).

Setting and participants

All participants were recruited from nine general hospitals with divisions of rheumatology in China (PUMCH, Affiliated Hospital of Inner Mongolia Medical College, General Hospital of Tianjin Medical University, Second Hospital of Hebei Medical University, Beijing Dongfang Hospital, Third Hospital of Hebei Medical University, First Hospital of Shanxi Medical University, The Bethune International Heping Hospital of Hebei and The Affiliated Hospital of QingDao University Medical College).

To be eligible for this trial, patients had to meet the following criteria: (1) 18–65 years of age; (2) diagnosed with RA as determined by meeting the 2010 ACR/EULAR classification criteria and having had RA for at least 6 weeks17; (3) at least three swollen joints (SJC) and five tender joints (TJC); (4) erythrocyte sedimentation rate (ESR)>28 mm/h or C-reactive protein (CRP)>20 mg/L.

Patients were excluded if they had a history of chronic serious infection, any current infection or any type of cancer. Pregnant or lactating women were excluded. All patients were informed of the risk of infertility and patients with a desire to have children were excluded. Patients who had received DMARDs or biological therapy within 3 months before participating in this study were excluded. Patients were allowed to continue to receive non-steroidal anti-inflammatory drugs and stable dosage of oral corticosteroids (≤10 mg prednisone or equivalent/day) during the study.

Randomisation and interventions

A computer-generated randomised code was used to assign patients with RA in a 1:1:1 ratio to one of three treatment arms to receive oral TwHF pills 20 mg three times a day; or MTX starting from 7.5 mg once a week and increasing to 12.5 mg once a week (0.20–0.25 mg/kg) within 4 weeks, with folic acid 10 mg on the day after each MTX administration; or TwHF plus MTX at the same dosage as above. At week 12, if patients in the MTX or TwHF monotherapy groups showed a reduction of the 28-joint count Disease Activity Score (DAS28) ≥30%, then the monotherapy was continued; otherwise the patients were switched to MTX+TwHF combination therapy. If the patients in the combination therapy group had a reduction of DAS28<30% at week 12, they were allowed to leave the trial and defined as treatment failures due to lack of efficacy. The flow chart of the study is shown in figure 1. In this study, the TwHF used was all from the same batch produced by Zhejiang DND pharmaceutical company, in which triptolide (C20H24O5), the major immunosuppressive anti-inflammatory diterpenoid, was 1.2 μg/10 mg, and wilforlide (C30H46O3), an anti-inflammatory triterpene, was 36.6 μg/10 mg.

Figure 1

Patient disposition. MTX, methotrexate; qw, once a week; RA, rheumatoid arthritis; TwHF, Tripterygium wilfordii Hook F.

Outcomes and measurements

Because this was an investigator-initiated trial and placebo tablets were not available, the treating doctors and patients were not blinded to the medication, but the clinical parameters and outcomes were assessed at baseline, weeks 4, 12 and 24 by different trained evaluators who were unaware of the specific therapeutic regimen.

The primary efficacy point was the proportion of patients achieving an ACR50 (the American College of Rheumatology criteria) response at week 24. To meet the criteria, patients had to have ≥50% improvement in both tender and swollen joints (28 tender and 28 swollen joints were assessed) and ≥50% improvement in three or more of the following: the evaluator's or patient's assessment of global health status, the patient's assessment of pain on a visual analogue scale, the patient's assessment of function using the Health Assessment Questionnaire (HAQ) and the ESR or serum CRP level.

Secondary efficacy points were the proportion of patients achieving ACR20 and ACR70 responses, EULAR good or moderate response, clinical Disease Activity Index (cDAI) good response (defined as achieving ≥50% improvement in the cDAI, or cDAI ≤2.8),18 ,19 clinical remission (defined as DAS28 <2.6) and low disease activity (LDA, defined as DAS28 <3.2) at week 24, as well as a change in HAQ or 36-item Short-Form Health Survey questionnaire (SF-36) scores.

The safety p profile was also recorded on each visit.

Statistical analysis

A non-inferiority test was conducted to compare the treatment efficacy of TwHF monotherapy and MTX monotherapy. A sample size of 207 subjects (69 patients in each group) was estimated to provide at least 80% power to detect a non-inferiority margin of 10% in the proportion of TwHF- and MTX-treated patients achieving an ACR50 response at week 24, at a 5% level of significance. In addition, the combination treatment with MTX+TwHF was compared with MTX monotherapy for potential superiority using the χ2 test. Intention-to-treat (ITT) and per-protocol (PP) analyses were performed with the ITT analysis as the primary outcome analysis. The ITT analysis included all patients who received at least one dose of the originally allocated treatment drug. The PP calculation included only the participants who finished the originally allocated treatment schedule without violating the inclusion/exclusion criteria. Those patients who withdrew from the trial prematurely or switched to the combination group were considered to be missing data, which were calculated using the last observation carried forward imputation method when performing the ITT analysis.

Continuous data are presented as mean (SD) or median (25th–75th centiles). Categorical data are presented as numbers (n) or proportions (%). Differences between groups were analysed for significance using a one-way analysis of variance F test or a non-parametric test (continuous data) and χ2 test (categorical data). All analysis was computed using SPSS statistics V.17.0 and SAS V.9.1. The statistical analysis plan is provided as online supplementary text.

Results

Patient characteristics

Altogether 269 patients with RA were screened initially and 207 eligible patients were finally enrolled in the study. They were randomly assigned to one of the three treatment arms: 69 patients received MTX monotherapy, 69 patients TwHF monotherapy and 69 patients received the combination therapy. As shown in figure 1, 86%, 86% and 81% of patients in the MTX, TwHF and combination groups completed the 24 weeks’ evaluation.

Baseline demographic and clinical characteristics of patients participating in the trial are summarised in table 1. All patients had active disease, as reflected by the number of TJC and SJC, ESR and CRP values and high DAS28 scores and there was no statistically significant differences among the three treatment groups (p>0.05). About 17.4% of patients were taking oral prednisone at baseline, at a dosage of <10 mg/day. Most patients (203 cases, 98.1%) had never been treated with DMARDs or biological reagents and only two patients had previously received MTX.

Table 1

Baseline demographic and clinical characteristics

Clinical efficacy

In an ITT analysis, at week 24, 46.4% (32/69), 55.1% (38/69) and 76.8% (53/69) of patients who received MTX, TwHF and combination therapy, respectively, reached at least 50% improvement of disease activity as determined by the ACR50 response criteria. A non-inferiority test was performed comparing the treatment efficacy of TwHF and MTX monotherapy and a significant difference was seen (pnon-inferiority=0.014). The result persisted after performing an additional sensitivity analysis assuming no treatment effect for those switched patients (MTX vs TwHF, pnon-inferiority=0.031). Similar statistically significant patterns at week 24 were seen for ACR20, ACR70, cDAI good responses, EULAR good response, remission rate (DAS28<2.6) and LDA rate (DAS28<3.2) between the TwHF and MTX monotherapy groups (TwHF vs MTX) (ACR20: 72.5% vs 63.8%; ACR70: 30.4% vs 23.2%; cDAI good response: 65.2% vs 52.2%; EULAR good response: 47.8% vs 26.1%; remission rate: 31.9% vs 20.3%; and LDA rate: 44.9% vs 27.5%) (p<0.05) (figure 2). In addition, there were significant differences between the combination therapy and the MTX monotherapy, indicating that treatment with TwHF plus MTX was more effective than MTX monotherapy as shown by ACR20, ACR50, ACR70, cDAI good response, EULAR good response, remission rate and LDA rate at week 24 (ACR 20: 92.8% vs 63.8%; ACR50: 76.8% vs 46.4%; ACR70: 43.5% vs 23.2%; cDAI good response: 87.0% vs 52.2%; EULAR good response: 58.0% vs 26.1%; DAS28<2.6: 49.3% vs 20.3%; and LDA rate: 55.1% vs 27.5%) (p<0.05) (figure 2).

Figure 2

Primary and secondary efficacy measures over time in intention-to-treat (ITT) analysis. *The number within each bar represents the patients who reached the response criteria in each group. The percentage of response was calculated with the denominator of total enrolled patients (69 for each group). Those patients who withdrew from the trial prematurely or switched to the combination group were classed as missing data, which were calculated using the last observation carried forward imputation method when performing the ITT analysis. **The p values between the methotrexate (MTX) group and the Tripterygium wilfordii Hook F (TwHF) group were calculated using the non-inferiority test. Comparison of the combination treatment and MTX monotherapy was calculated using the χ2 test. ACR, American College of Rheumatology; cDAI, clinical Disease Activity Index.

There was no statistically significant difference in the mean change of DAS28 from baseline to week 24 (ΔDAS28) between TwHF monotherapy (−2.99) and MTX monotherapy (−2.74) (p=0.348, table 2), whereas patients receiving combination therapy had a significantly greater decrease in DAS28 (−3.41) than patients receiving MTX monotherapy (p=0.032). No significant difference was found in the HAQ improvement from baseline to week 24 among patients receiving TwHF monotherapy, MTX monotherapy or combination therapy (mean improvement 0.88, 0.78 and 1.06 respectively; p>0.05). A similar non-significant pattern was found for SF-36 (mean improvement 252, 229 and 285, respectively; p>0.05).

Table 2

Clinical and laboratory measures of the three groups at each visit

We also carried out a PP analysis that calculated data only from the patients who completed 24 weeks’ evaluation. The ACR50 response was 59.2% (29/49), 65.3% (32/49) and 80.4% (45/56), respectively, in patients who received MTX monotherapy, TwHF monotherapy and combination therapy. The result of the PP analysis agreed with that found in the ITT analysis (pnon-inferiority=0.049 for TwHF vs MTX; p=0.018 for TwHF+MTX vs MTX). Similar statistically significant patterns were also seen for ACR20, ACR70, EULAR good response, cDAI good response, clinical remission and LDA rate at week 24 (figure 3).

Figure 3

Primary and secondary efficacy measures over time in the per-protocol analysis. *The number within each bar represents the patients who reached the response criteria in each group. The percentage of response was calculated with the denominator of patients who finished the originally allocated treatment schedule (49, 49, 56 for the methotrexate (MTX), the Tripterygium wilfordii Hook F (TwHF) and the combination groups, respectively). **The p values between the MTX group and the TwHF group were calculated using the non-inferiority test. Comparison of the combination treatment and MTX monotherapy was calculated using the χ2 test.

A significant ESR reduction was achieved in the TwHF and the combination groups at week 12, whereas in the MTX group a significant reduction in ESR was not seen until week 24. At week 4, the TwHF group had greater improvement than the MTX group in the change in ESR (p=0.04).

Four patients in the MTX monotherapy group (5.8%) withdrew as a result of lack of efficacy compared with four (5.8%) in the TwHF monotherapy and three (4.3%) in the combination therapy groups. Seven patients withdrew because of severe adverse events, including three (4.3%), one (1.4%) and three (4.3%) each from the MTX, TwHF and combination groups. Nine patients discontinued the trial because of a protocol violation (one, three, five cases each from the MTX, TwHF and the combination groups, respectively). Another 10 patients each in the MTX (14.5%) and TwHF (14.5%) monotherapy groups were switched to MTX+TwHF treatment at week 12 owing to a clinically insufficient response (defined as a reduction of DAS28 <30%). After switching to the combination, nine (90%) patients each from the MTX and TwHF groups reached an ACR20 response at week 24, five (50%) patients in the MTX group and seven (70%) patients in the TwHF group achieved ACR50 responses, four (40%) patients in the MTX group and two (20%) patients in the TwHF group achieved ACR70 responses.

Safety and tolerability

Table 3 listed all adverse events occurring during the trial. About 52.7% of all patients experienced adverse events, which were seen in 46.4%, 62.3% and 49.3% patients receiving TwHF, MTX and TwHF+MTX, respectively (p=0.136). The most common adverse events were gastrointestinal, which occurred in 29.0%, 43.5% and 34.8% patients receiving TwHF, MTX and TwHF+MTX, respectively (p=0.202). Despite the high frequency, most gastrointestinal events, were mild and 93.2% of patients with gastrointestinal events completed 24 weeks’ treatment.

Table 3

Adverse events of all participants.

One patient receiving TwHF monotherapy, three patients each receiving MTX monotherapy and TwHF+MTX discontinued the study because of severe adverse events, with no statistically significant differences among the treatment groups (p=0.554). The patient from the TwHF group discontinued the study because of an increase in alanine aminotransferase (ALT 92 U/L, upper limit of normal (ULN) 40 U/L). In the MTX group, two patients discontinued the study because of severe nausea, vomiting, diarrhoea and loss of appetite; the other patient had severe rash and oral ulcer. In the combination group, the severe adverse events included tuberculous pleuritis in one patient, a significant increase of ALT in another patient (ALT 89 U/L) and a gastrointestinal event in the third patient.

Fifteen (8.8%) of all female patients (n=170, 69 of whom were premenopausal female) developed irregular menstruation during the 24 weeks’ trial, including seven in the TwHF group, three in the MTX group and five in the combination group (TwHF monotherapy vs MTX monotherapy, p=0.216).

This TRIFRA study showed that TwHF monotherapy was not inferior to, and combination therapy of MTX+TwHF was better than, MTX monotherapy in controlling disease activity in patients with active RA. These efficacy results confirm the results from the clinical effectiveness assessment and indicate that the combination of TwHF and MTX is a safe and efficacious treatment for patients with active RA.

Discussion

The results of the TRIFRA trial indicate that TwHF is effective for the treatment of active RA. Importantly, most patients in this trial (98.0%) were DMARD naïve, indicating that TwHF is effective in this group of patients with RA. Previous studies7 ,8 ,14 had evaluated patients who had more advanced disease and had been treated with multiple DMARDs, so these are the first data indicating that TwHF is effective in DMARD-naïve patients with RA. At week 24, TwHF monotherapy resulted in significant improvement of disease activity, including pain assessment, the patient's and physician's global assessment, TJC, SJC, ESR, CRP, and HAQ and SF-36 scores. MTX monotherapy and TwHF monotherapy had similar efficacy as shown by ACR20, ACR50 and ACR70 response criteria, EULAR and cDAI good response criteria, as well as DAS28 remission criteria and LDA rate. The efficacy of TwHF was not inferior to that of MTX, and MTX+TwHF combination therapy was more effective than MTX monotherapy in treating active RA.

Compared with MTX monotherapy, patients receiving TwHF monotherapy showed a more rapid reduction of the ESR (table 2); this phenomenon was previously reported in a systematic review.20 It is reported that TwHF may inhibit the expression of proinflammatory genes, such as those for interleukin 2, inducible nitric oxide synthase, tumour necrosis factor α, cyclo-oxygenase-2 and interferon γ.11 The significant decrease of ESR is probably related to the anti-inflammatory activity of TwHF, which has been shown in both laboratory experiments and clinical trials.10 ,21

A safety evaluation of the study demonstrated that the frequency of total adverse events and severe adverse events of TwHF monotherapy was not significantly higher than that of MTX monotherapy, except for a slightly increased frequency of irregular menstruation. In the 1980s, extracts of TwHF were found to have an antifertility effect, but a number of studies have shown that these effects are usually reversible and more common in perimenopausal women.22–26 It was found that the serum levels of follicle-stimulating hormone, luteinising hormone and oestradiol decreased significantly after 12 weeks’ treatment with TwHF extracts and some of the patients receiving TwHF extracts had amenorrhoea; this was reversed when treatment was stopped.25 In addition, Tao and Lipsky reported that the number of spermatozoa in men decreased and the sperm viability declined after receiving TwHF extracts.5 The inhibition of T-type Ca (2+) influx may be responsible for the antifertility activity.24 The effects of TwHF on reproductive hormones and the results of the safety evaluation of our study suggest that TwHF treatment should be used in postmenopausal female patients and in patients who are no longer interested in having children.

Our study has several limitations. First, the study was an open-label clinical trial. Because this was an investigator-initiated trial and placebo tablets were not available, the treating doctors and patients were not blinded to medication. To make the study more objective, the clinical outcomes were assessed by different trained evaluators who were unaware of the specific therapeutic regimen. To be completely objective, a double-blind RCT would be necessary in the future. Second, as it was a 24-week clinical trial, the radiological progression of the affected joints was not evaluated. We will continue to follow-up this cohort of patients and compare the radiological progression when the patients have completed 2 years of treatment. Finally, the dose of MTX was limited to 12.5 mg/week. This is standard in Asia, although it is common in the West to use higher doses. Whether the results would have been altered if the patients had received higher doses of MTX is unknown, although this seems unlikely as the levels of responsiveness in this trial were much greater than previously reported in similar patients given MTX monotherapy.27

Acknowledgments

Zhejiang DND had no role in the study design or in the collection, analysis, or interpretation of the data, the writing of the manuscript or the decision to submit the manuscript for publication. The Peking Union Medical College Hospital and the study investigators are solely responsible for all data collection and management, statistical analysis, manuscript development and the decision to publish these results.

References

View Abstract

Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.

    Files in this Data Supplement:

  • Press release

    Press release

    Files in this Data Supplement:

Footnotes

  • Handling editor Tore K Kvien

  • Q-wL, WZ, QS and W-jZ contributed equally to this manuscript.

  • Contributors XZ, HC, WZ, W-jZ conceived and designed the study. PEL advised on the design of the study and draft of the paper. XZ, WZ, QS, W-jZ, Q-jW, H-bL, LG, WW, HL, A-jL, H-tJ, J-xW, X-mL, Z-bL, BL, MS, QW, Y-yF, J-mS, L-dZ, YJ, JL, F-lT, F-cZ recruited the patients for this study, with most of the recruiting work done by XZ, WZ, QS, W-jZ, Q-jW. Q-wL, W-lJ, X-nW, XL, DL, Y-fY evaluated the participants and recorded information about them. Q-wL and XL analysed the data. Q-wL drafted the paper. XZ and PEL edited and revised the paper.

  • Funding This work was supported by the grants from the National Natural Science Foundation of China (81325019, 81172859 and 81273312), Beijing Municipal Natural Science Foundation (7141008), the Research Special Fund for Public Welfare Industry of Health (20120217), the Capital Health Research and Development of Special Fund (2011-4001-02) and the National laboratory Special Fund (2060204).

  • Competing interests None.

  • Ethics approval Approved by the institutional ethical review board.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.