The efficacy and safety of thymosin α1 for sepsis (TESTS): multicentre, double blinded, randomised, placebo controlled, phase 3 trial
BMJ 2025; 388 doi: https://doi.org/10.1136/bmj-2024-082583 (Published 15 January 2025) Cite this as: BMJ 2025;388:e082583Linked Editorial
Immunotherapies for sepsis and the impact of study design
- Jianfeng Wu, chief physician1 2 3,
- Fei Pei, medical doctor1 2,
- Lixin Zhou, professor4,
- Weiqin Li, professor5,
- Renhua Sun, professor6,
- Yimin Li, professor7,
- Zheng Wang, professor8,
- Zhijie He, chief physician9,
- Xiaofei Zhang, associate chief physician10,
- Xiaodong Jin, professor11,
- Yun Long, professor12,
- Wei Cui, chief physician13,
- Chunting Wang, chief physician14,
- Erzhen Chen, professor15,
- Jun Zeng, chief physician16,
- Jing Yan, professor17,
- Qinhan Lin, professor18,
- Feihu Zhou, professor19,
- Lei Huang, chief physician20,
- You Shang, professor21,
- Meili Duan, chief physician22,
- Wei Zheng, chief physician23,
- Duming Zhu, chief physician24,
- Qiuye Kou, chief physician10 25,
- Shihong Zhang, professor26,
- Yin Liu, technician26,
- Chen Yao, professor27,
- Meixia Shang, senior statistician27,
- Sui Peng, professor3,
- Qian Zhou, associate researcher3,
- Kar Keung Cheng, professor28,
- Xiangdong Guan
, professor1 2 - on behalf of the TESTS study collaborator group
- 1Department of Critical Care Medicine, The First Affiliated Hospital, Sun Yat-Sen University, Guangzhou, Guangdong province, China
- 2Guangdong Clinical Research Center for Critical Care Medicine, Guangzhou, Guangdong province, China
- 3Clinical Trials Unit, The First Affiliated Hospital, Sun Yat-sen University, Guangzhou, Guangdong Province, China
- 4Department of Critical Care Medicine, The First People’s Hospital of Foshan, Foshan, Guangdong province, China
- 5Department of Critical Care Medicine, Jinling Hospital, Medical School of Nanjing University, Nanjing, Jiangsu province, China
- 6Emergency and Critical Care Center, Intensive Care Unit, Zhejiang Provincial People's Hospital (Affiliated People’s Hospital, Hangzhou Medical College), Hangzhou, Zhejiang province, China
- 7Department of Critical Care Medicine, State Key Laboratory of Respiratory Diseases, Guangzhou Institute of Respiratory Health, The First Affiliated Hospital of Guangzhou Medical University, Guangzhou, Guangdong province, China
- 8Department of Hepatobiliary Surgery, Department of Surgical Intensive Care Unit, The First Affiliated Hospital of Xi’an Jiaotong University, Xi’an, Shaanxi province, China
- 9Department of Critical Care Medicine, Sun Yat-sen Memorial Hospital, Sun Yat-sen University, Guangzhou, Guangdong province, China
- 10Department of Intensive Care Unit, The Sixth Affiliated Hospital, Sun Yat-Sen University, Guangzhou, Guangdong province, China
- 11Department of Critical Care Medicine, West China Hospital of Sichuan University, Chengdu, Sichuan Province, China
- 12Department of Critical Care Medicine, State Key Laboratory of Complex Severe and Rare Diseases, Peking Union Medical College Hospital, Peking Union Medical College, Chinese Academy of Medical Sciences, Beijing, China
- 13Department of Critical Care Medicine, The Second Affiliated Hospital of Zhejiang University School of Medicine, Hangzhou, Zhejiang province, China
- 14Department of Critical Care Medicine, Shandong Provincial Hospital Affiliated to Shandong First Medical University, Jinan, Shandong province, China
- 15Department of Emergency, Ruijin Hospital, Shanghai JiaoTong University School of Medicine, Shanghai, China
- 16Department of Critical Care Medicine, Guangzhou First People’s Hospital, Guangzhou, Guangdong province, China
- 17Department of Intensive Care, Zhejiang Hospital, Hangzhou, Zhejiang province, China
- 18Department of Critical Care Medicine, Affiliated Qingyuan Hospital, The Sixth Clinical Medical School, Guangzhou Medical University, Qingyuan People’s Hospital, Qingyuan, Guangdong Province, China
- 19Department of Critical Care Medicine, The First Medical Centre, Chinese PLA General Hospital, Beijing, China
- 20Department of Critical Care Medicine, Peking University Shenzhen Hospital, Shenzhen, Guangdong Province, China
- 21Department of Critical Care Medicine, Union Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, Hubei province, China
- 22Department of Critical Care Medicine, Beijing Friendship Hospital, Capital Medical University, Beijing, China
- 23Department of Critical Care Medicine, Zhuhai People’s Hospital, Zhuhai, Guangdong Province, China
- 24Department of Critical Care Medicine, Zhongshan Hospital Fudan University, Shanghai, China
- 25Department of Critical Care Medicine, Foersea Life insurance Guangzhou General Hospital, Guangzhou, Guangdong province, China
- 26Department of Laboratory Medicine, The First Affiliated Hospital, Sun Yat-Sen University, Guangzhou, Guangdong province, China
- 27Peking University First Hospital, Beijing, China
- 28Department of Applied Health Sciences, College of Medicine and Health, University of Birmingham, Birmingham, UK
- Correspondence to: X Guan, Department of Critical Care Medicine, The First Affiliated Hospital, Sun Yat-Sen University, Guangzhou, Guangdong province, China guanxd{at}mail.sysu.edu.cn
- Accepted 2 December 2024
Abstract
Objective To evaluate whether the immunomodulatory drug thymosin α1 reduces mortality in adults with sepsis.
Design Multicentre, double blinded, placebo controlled phase 3 trial.
Setting 22 centres in China, September 2016 to December 2020.
Participants 1106 adults aged 18-85 years with a diagnosis of sepsis according to sepsis-3 criteria and randomly assigned in a 1:1 ratio to receive thymosin α1 (n=552) or placebo (n=554). A stratified block method was used for randomisation, and participants were stratified by age (<60 and ≥60 years) and centre.
Interventions Subcutaneous injection of thymosin α1 or placebo every 12 hours for seven days unless discontinued owing to discharge from the intensive care unit, death, or withdrawal of consent.
Main outcome measure The primary outcome was 28 day all cause mortality after randomisation. All analyses were based on a modified intention-to-treat set, including participants who received at least one dose of study drug.
Results Of 1106 adults with sepsis enrolled in the study, 1089 were included in the modified intention-to-treat analyses (thymosin α1 group n=542, placebo group n=547). 28 day all cause mortality occurred in 127 participants (23.4%) in the thymosin α1 group and 132 (24.1%) in the placebo group (hazard ratio 0.99, 95% confidence interval 0.77 to 1.27; P=0.93 with log-rank test). No secondary or safety outcome differed statistically significantly between the two groups. The prespecified subgroup analysis showed a potential differential effect of thymosin α1 on the primary outcome based on age (<60 years: hazard ratio 1.67, 1.04 to 2.67; ≥60 years: 0.81, 0.61 to 1.09; P for interaction=0.01) and diabetes (diabetes: 0.58, 0.35 to 0.99; no diabetes: 1.16, 0.87 to 1.53; P for interaction=0.04).
Conclusions This trial found no clear evidence to suggest that thymosin α1 decreases 28 day all cause mortality in adults with sepsis.
Trial registration ClinicalTrials.gov NCT02867267.
Introduction
Sepsis is a life threatening syndrome of organ dysfunction caused by a dysregulated response to infection.1 The immune system serves as a natural barrier against invasion by pathogenic microorganisms. Sepsis induced immunosuppression, however, leads to a reduction in the number and function of immune cells, such as T cell exhaustion, which weakens the immune system’s ability to eliminate pathogens.2345
Immunomodulation treatment, an important adjunct to use of antibiotics, has become a focus of clinical trials aimed at managing sepsis induced immunosuppression, with several alternative drugs, including thymosin α1, currently under investigation.6789 Thymosin α1, a peptide primarily secreted by the thymus, attenuates chemotherapy induced immune damage to exert a synergistic effect,10 modulates dendritic cells to enhance the antifungal effect of T helper 1 cells,11 and reduces T cell exhaustion to improve lymphopenia.1213 Given its multiple immune effects, thymosin α1 has been used in the treatment of non-small cell lung cancer,1415 melanoma,16 chronic hepatitis,171819 sepsis,20 AIDS,21 and covid-19.13
Previous studies with small patient numbers showed that thymosin α1 improved immune function and reduced mortality in patients with sepsis. A meta-analysis of 10 randomised controlled trials with a total of 530 patients suggested that thymosin α1 might offer a 41% relative reduction in 28 day mortality (22% v 38%, relative risk 0.59, 95% confidence interval (CI) 0.45 to 0.77), but the quality of evidence was low owing to the small sample sizes of the included trials.22 Subsequently, an expert consensus on sepsis immunosuppression recommended the use of thymosin α1 in patients with sepsis, albeit based on limited evidence.23
We conducted a multicentre, randomised, double blinded, placebo controlled clinical trial (TESTS, The Efficacy and Safety of Thymosin α1 for Sepsis) to evaluate the efficacy and safety of thymosin α1 for the treatment of sepsis.
Methods
Study design and oversight
Our multicentre trial involved 22 centres in nine provinces across all five geographical regions of China: eastern (Shanghai, Shandong, Jiangsu, Zhejiang), southern (Guangdong), western (Sichuan), northern (Beijing, Shaanxi), and central (Hubei). The trial took place between September 2016 and December 2020 (see supplementary Protocol and SAP file, p3). Written informed consent was obtained from all participants or their legally authorised representatives. An independent data and safety monitoring board oversaw the safety of the trial.
Participants
We included participants aged 18 to 85 years with a diagnosis of sepsis according to Sepsis-3 criteria.1 Patients were excluded if they met any of the following criteria: were pregnant or lactating; had haematological malignancies; had undergone organ or bone marrow transplantation; had an acute phase autoimmune disease or glomerulonephritis; had an allergy or intolerance to thymosin α1; had a history of cardiopulmonary resuscitation within 72 hours before signing the consent form, with incomplete neurological recovery (Glasgow coma scale score ≤8); had a history of radiotherapy, chemotherapy, treatment with immunosuppressive drugs, or continuous treatment with prednisolone >10 mg/day (or an equivalent dose of other steroids) in the past 30 days; had participated in clinical trials related to immunity in the past 30 days; had undrained foci of infection (eg, intra-abdominal infections that cannot be managed through surgery or drainage); had an underlying disease estimated to result in death within 28 days; or the immediate family expressed a wish for life sustaining treatment to be discontinued or end-of-life care instigated. The supplementary appendix lists the inclusion and exclusion criteria.
Randomisation and masking
Using a computer generated block randomisation protocol with a block size of 8, we randomly assigned (1:1) eligible participants to receive either thymosin α1 or matched placebo. Randomised numbering and treatment allocation were done through an interactive web response system, and random assignment was stratified by centre and age (<60 and ≥60 years).
Statisticians not connected with the trial provided the random number tables for drug randomisation, using SAS version 9.4 software based on fixed seed number. They were also responsible for masking of the drug and preparing emergency letters containing individual treatment assignments, which could be opened for medical emergencies when knowledge of treatment allocation was required. The investigators, participants, care providers, and statisticians were all blinded to the assigned treatment.
Procedures
Participants in the intervention group received a subcutaneous injection of 1.6 mg of lyophilised thymosin α1 powder dissolved in 1 mL of sterilised water every 12 hours. The control group received placebo (lyophilised saline) in the same manner. The trial drugs were administered for seven days, unless discontinued owing to discharge from the intensive care unit (ICU), death, or withdrawal of consent. In both study arms, standardised treatment was provided according to the Surviving Sepsis Campaign.24 After enrolment, participants were evaluated according to the study procedures (see supplementary Protocol and SAP file, pp 98-105). A validated electronic clinical data management system was used to collect and manage data from the trial, with data entered on an electronic case report form.
Outcomes
The primary outcome was 28 day all cause mortality after randomisation. Secondary outcomes included incidence of new onset infections within 28 days, 28 day clearance rate of pathogenic microorganisms, duration of ICU and hospital stays, readmission to hospital within 28 days, changes in the sequential organ failure assessment score on day 7, 90 day all cause mortality, mortality on the ICU, mechanical ventilation-free days within 28 days, ICU-free days within 28 days, continuous renal replacement therapy-free days within 28 days, vasopressors-free days within 28 days, and score on the 36-item short form (SF-36) quality of life scale within 90 days. Exploratory outcomes included change in monocyte human leucocyte antigen-DR, neutrophil to lymphocyte ratio, regulatory T cell percentage, and lymphocyte count at 7, 14, and 28 days, compared with baseline values obtained at screening. We monitored adverse events and serious adverse events for safety signals. See supplementary Protocol and SAP file, pp 89 and 90, for details and definitions of all outcomes.
Statistical analysis
We determined the sample size to detect superiority in the primary outcome on the basis of data from a previous trial and assumed a 27% 28 day mortality rate for the intervention group and 35% for the control group, with a one sided type I error of 0.025 and power of 80%.20 Considering a dropout rate of 5%, we determined that 553 participants (1106 in total) would be required in each group. Continuous variables were analysed using either an independent sample t test or the Wilcoxon rank-sum test between two groups. Paired group comparisons were made using a paired t test or the Wilcoxon signed-rank test. Categorical variables were analysed using the χ2 test or Fisher’s exact test, and the Cochran-Mantel-Haenszel test was used to calculate the between group difference according to stratified factors. Kaplan-Meier estimates were used for time-to-event analyses, and log-rank tests were used for comparisons. The primary outcome was analysed using a Cox proportional hazards model adjusted for centre and age and assessed based on Schoenfeld residuals. The mixed effects model for repeated measures was used to analyse the dynamic changes in immune markers. Subgroup analyses were conducted by age (a stratified factor), sex, and chronic conditions (hypertension, coronary heart disease, chronic obstructive pulmonary disease, diabetes, chronic kidney disease, and solid malignant tumour). Tests for interaction between treatment and subgroups were conducted by including multiplicative interaction terms in the Cox proportional hazards models. Confidence intervals and P values were not adjusted for multiple testing, except for interaction P values of subgroup analyses. We used Bonferroni correction to adjust for multiple testing in subgroup analyses. All outcomes were analysed in the modified intention-to-treat set, which included participants who were randomised and received at least one dose of study drug. To assess the superiority of the primary outcome, we examined whether the upper limits of the confidence intervals exceeded zero. P values were two sided, and we considered P<0.05 to indicate statistical significance, except for the primary outcome. The primary outcome was not imputed because no data were missing after study completion. Missing data for other outcomes were replaced using regression multiple imputation under the assumption of monotonic patterns. All statistical analyses were prespecified in the statistical analysis plan and independently conducted using SAS version 9.4 (see supplementary Protocol and SAP file, pp 139-177).
Patient and public involvement
Limitations in funding and expertise restricted our ability to conduct focus groups for patient and public involvement. However, our investigators’ clinical experience with patients played an important role in informing the study’s design and rationale. The involvement of patients in the later phases of the trial, as well as the statistical analysis after the trial’s conclusion, was affected by the covid-19 pandemic.
Results
Baseline characteristics
From September 2016 to December 2020, a total of 1106 participants with sepsis were enrolled in the trial, 552 of whom were randomised to the thymosin α1 group and 554 to the placebo group (fig 1 and supplementary appendix table S1). After randomisation, 17 participants (10 in the thymosin α1 group and seven in the placebo group) were excluded because they had not received any study drug before withdrawing consent, and 1089 participants were included in the modified intention-to-treat analysis (see supplementary appendix table S2). The daily number of participants who received the study drugs was similar between the two groups (see supplementary appendix table S3).
Personal and clinical characteristics were well balanced between the two groups at baseline (table 1), including indicators for severity of illness and immune function. The median age was 65 years (interquartile range (IQR) 52-73), and 68.9% of participants (750/1089) were men. The most common site of infections was lung (31.6%). A total of 1046 patients (96.1%) received antibiotics at screening (table 1 and see supplementary appendix table S4). The median acute physiology and chronic health evaluation-II (APACHE-II) scores and sequential organ failure assessment (SOFA) scores at screening were 14 (IQR 10-19) and 7 (5-10), respectively, indicating disease severity. Before randomisation, 76.8% of participants (n=836) received mechanical ventilation for a median of eight (IQR 4-16) days, 61.9% (n=674) received vasopressors for a median of five (2-10) days, and 16.9% (n=184) received continuous renal replacement therapy for a median of six (3-14) days.
Baseline characteristics of participants with sepsis randomly assigned to receive thymosin α1 or placebo. Values are number (percentage) unless stated otherwise
Primary outcome
In the modified intention-to-treat set, all cause mortality within 28 days of randomisation occurred in 127 participants (23.4%) in the thymosin α1 group and 132 (24.1%) in the placebo group (hazard ratio 0.99, 95% CI 0.77 to 1.27, P=0.93) (table 2 and fig 2). The 90 day all cause mortality was not significantly different between the thymosin α1 group and placebo group (31.0% v 32.4%, hazard ratio 0.94, 0.76 to 1.16, P=0.54) (table 2 and fig 2). Sensitivity analysis further confirmed the findings for the primary outcome (supplementary appendix table S5).
Outcomes of modified intention-to-treat population in patients with sepsis randomised to receive thymosin α1 or placebo. Values are number (percentage) unless stated otherwise
Kaplan-Meier estimates of time-to-event for death at 28 days (primary outcome) and 90 days from randomisation. Log-rank tests were used for comparisons and the hazard ratios for relative risks of the primary outcome between the two groups, along with corresponding 95% CIs, were calculated using a Cox regression model adjusted for centre and age. The model assumption was assessed using Schoenfeld residuals. CI=confidence interval
Subgroup analysis
Subgroup analysis was performed in eight subgroups (fig 3). Among these subgroups, we observed heterogeneity for the treatment effect by age: in the thymosin α1 group, 28 day mortality was higher in participants younger than 60 years (1.67, 1.04 to 2.67); for participants aged 60 and older, the hazard ratio was 0.81 (0.61 to 1.09, P for interaction=0.01) (fig 3 and supplementary figure S1). Post hoc analysis suggested that thymosin α1 might not have increased 28 day mortality in participants younger than 60 (1.45, 0.89 to 2.36, P for interaction=0.04) after adjusting for organ support before randomisation (supplementary appendix table S6). For chronic conditions, participants with diabetes in the thymosin α1 group had a lower 28 day mortality than participants with diabetes in the placebo group (0.58, 0.35 to 0.99, P for interaction=0.04). Additionally, in participants with hypertension (0.71, 0.49 to 1.04, P for interaction=0.06) and coronary heart disease (0.47, 0.21 to 1.01, P for interaction=0.06) the trend was consistent with those aged 60 and older and those with diabetes.
Subgroup analyses. A Bonferroni threshold for significance for overall type I error of 0.05 was P=0.006. The HRs for relative risk of primary outcome of the two groups and associated 95% CIs were calculated with a Cox regression model adjusting for centre and age. An interaction term was added between treatment and subgroup in the Cox regression model. CI=confidence interval; COPD=chronic obstructive pulmonary disease
Secondary outcomes
No statistically significant differences were found between the thymosin α1 group and placebo group for secondary outcomes, including change in SOFA score, new onset infections, mechanical ventilation-free days within 28 day, vasopressors-free days within 28 days, and continuous renal replacement therapy-free days within 28 days (table 2). These results were consistent with the original analysis accounting for missing data (supplementary appendix table S7).
At the time of enrolment, nearly all participants with sepsis (1088/1089, 99.9%) had at least one abnormality in an immune marker, and about seven out of 10 participants (747/1089, 68.6%) had three or more (supplementary appendix table S8). Exploratory analyses of immune markers showed that after a week of treatment both study arms showed a similar change in monocyte human leucocyte antigen-DR, lymphocyte count, and regulatory T cell percentage (table 2). However, the absolute change (group difference −1.7, 95% CI −3.0 to −0.6, P=0.02) and percentage change (group difference 9%, 2% to 15%, P=0.006) of neutrophil to lymphocyte ratio in the thymosin α1 group were greater than in the placebo group (table 2). This result was consistent when missing data were infilled (supplementary appendix table S7). As for longitudinal dynamic change of these markers, an ancillary finding was a significant improvement in monocyte human leucocyte antigen-DR, lymphocyte count, and neutrophil to lymphocyte ratio over the 28 day follow-up period, but no between group differences between thymosin α1 and placebo (fig 4 and supplementary appendix figure S2).
Dynamic change of immune markers in participants with sepsis. Mixed effects model for repeated measures was used to analyse the absolute changes in immune markers: mHLA-DR, lymphocyte count, neutrophil to lymphocyte ratio, and regulatory T cells from screening period to 28 day follow-up. This model included measurements of immune markers on days 7, 14, and 28 as dependent variables (the proportion of regulatory T cells was only measured on day 7), whereas immune marker levels at screening, centre, and age group served as independent variables. mHLA-DR=monocyte human leucocyte antigen-DR
Safety
Overall, 730 participants (67.0%) experienced at least one adverse event (360 participants (66.4%) in the thymosin α1 group and 370 (67.6%) in the placebo group; group difference −1.2%, 95% CI −6.8% to 4.4%, P=0.70), and 305 participants (28.0%) experienced at least one serious adverse event (145 participants (26.8%) in the thymosin α1 group and 160 (29.3%) in the placebo group; group difference −2.5%, −7.8% to 2.8%, P=0.38) within 90 days of follow-up (table 2 and supplementary appendix tables S9 and S10). The most common adverse effects were anaemia (10.7%), followed by fever (9.6%), abdominal distension (5.4%), and coagulation disorders (4.8%). No unexpected serious adverse events related to thymosin α1 occurred during the study.
Discussion
This multicentre trial found no conclusive evidence that thymosin α1 reduces 28 day mortality in adults with sepsis. The drug’s good safety profile was, however, validated. Additionally, thymosin α1 might have beneficial effects in patients aged 60 years and older and those with chronic conditions.
Comparison with other studies
The immunomodulatory drug thymosin α1 reduces T cell exhaustion and preserves the number and function of effector T cells, thereby exerting a sustained effect in eliminating pathogenic microorganisms.1213 A meta-analysis of previous trials of thymosin α1 showed a 41% reduction in 28 day mortality, but these trials had important limitations.22 In the current study, the failure to observe efficacy from thymosin α1 use can be explained in several ways. Firstly, previous studies were small and positive results might have been chance findings. Methodological issues were also present, including lack of masking and imbalance between trial arms in key patient characteristics. Secondly, sepsis is a highly heterogeneous disease, with complex and diverse pathophysiological processes, as well as diverse immune states. Therefore, improving the overall survival rate of all patients with sepsis may be unrealistic with a single agent. Thirdly, despite the sample size of the TESTS trial being nearly twice that of the combined total from the previous 10 trials, a modest reduction in mortality may still be possible. The broader confidence intervals include values that could be statistically significant, highlighting the need for further research. Finally, it is possible that the dosage of thymosin α1 was inappropriate. In the treatment of patients with cancers and viral hepatitis, the conventional therapeutic dose of thymosin α1 is 1.6 mg twice weekly, although the half life of the drug is 1.5 hours.25 In patients with sepsis, a previous study reported that treatment with 3.2 mg of thymosin α1 daily for one week in combination with anti-inflammatory drugs reduced morbidity and mortality, whereas 1.6 mg daily was ineffective.26 Therefore, in the current study we chose 1.6 mg twice daily for one week.
Clinical implications and rationale
Age is an important factor influencing immune system function, particularly in patients with sepsis when age related changes can affect treatment responses.2728 Our findings indicate that thymosin α1 use may be more effective in patients aged 60 years and older and in those with chronic conditions such as diabetes, hypertension, and coronary heart disease. As a multifunctional immunomodulatory drug, thymosin α1 has shown efficacy in alleviating T cell exhaustion, a key phenotype associated with impaired immune function in patients aged 60 years and older and those with chronic conditions.122930 This T cell exhaustion compromises the ability of patients with sepsis to clear pathogens, resulting in poor prognoses. Thus, the rationale for administering thymosin α1 to these patients is well supported. Conversely, we did not observe similar efficacy in patients younger than 60 years with sepsis. Although younger patients seemed to experience worse outcomes, part of this difference in mortality may be due to discrepancies in baseline clinical characteristics between the thymosin α1 group and placebo group. It may be worthwhile for future studies on thymosin α1 for the treatment of sepsis to focus on patients aged 60 years and older and those with chronic conditions.
Monocyte human leucocyte antigen-DR expression, regulatory T cell percentage, lymphocyte count, and neutrophil to lymphocyte ratio are commonly used as markers of immune status in patients with sepsis.31323334 Based on these markers, almost all patients in this study had immune dysregulation at enrolment, and improvement in these markers was observed in both groups after treatment. Monocyte human leucocyte antigen-DR are rarely assessed in multicentre studies because their detection requires the use of flow cytometry. This study initially aimed to reduce heterogeneity between laboratories through standardised quantification. Despite this effort, unlike in previous studies using a single laboratory, we did not observe an increase in monocyte human leucocyte antigen-DR levels in response to thymosin α1 treatment.35 This lack of observed increase may partly be attributed to the high variability among different laboratories, which may have masked any true treatment related differences between groups. The neutrophil to lymphocyte ratio reflects the balance between innate and adaptive immunity and is a cheap biomarker and easy to obtain.3637 A meta-analysis showed that a high neutrophil to lymphocyte ratio was a reliable immune biomarker for predicting poor prognosis in patients with sepsis.38 Consequently, a more pronounced reduction in the ratio after thymosin α1 treatment is supportive of a role of thymosin α1 in maintaining immune homoeostasis of sepsis.
Limitations of this study
Several limitations should be considered when interpreting our findings. Firstly, owing to the high heterogeneity of sepsis, it is difficult to achieve consistent efficacy with a single treatment. Although we attempted to identify potential study populations during the trial design, previous studies lacked relevant evidence. To deal with this issue, we predefined subgroups, hoping to identify potential beneficiary subpopulations for thymosin α1 treatment through this trial. Secondly, our study may have been underpowered. Our sample size calculation was guided by data from the largest published trial at the time we initiated the current study. The former found outcome rates of 26% and 35% in the thymosin α1 and placebo groups, respectively. We powered our study based on a difference of 8% (27% v 35%).20 In retrospect, this might have been too ambitious. In the event, the observed effect was close to 1, with a wide confidence interval (hazard ratio 0.99, 95% CI 0.77 to 1.27). Therefore, further research might still be warranted, although based on the results from this trial there is no strong evidence to support the presence of an effect. Thirdly, this study was not able to determine the precise times from onset of sepsis to diagnosis because of individual variability and disease progression. To estimate the interval from sepsis diagnosis to randomisation, we collected data on the duration of organ support before randomisation. This suggests that most enrolled patients were likely not in the early stages of sepsis.
Conclusions
This multicentre trial found no conclusive evidence that thymosin α1 reduces 28 day mortality in adults with sepsis. The drug’s good safety profile in the treatment of sepsis was, however, validated. Additionally, thymosin α1 might have beneficial effects in patients aged 60 and older and those with chronic conditions. Future research might consider focusing on these patients to clarify potential therapeutic advantages.
What is already known on this topic
The dysregulated response to infection in sepsis plays a crucial pathophysiological mechanism, making immunomodulatory treatment an appealing option
Thymosin α1, a peptide with pleiotropic immune effects during infections, is a promising immunomodulatory agent
Studies suggest thymosin α1 might reduce 28 day mortality in patients with sepsis, but the quality of evidence is low owing to small sample sizes and methodological issues
What this study adds
This study found no conclusive evidence that thymosin α1 reduces 28 day mortality in adults with sepsis
Thymosin α1 did, however, show good safety in the treatment of sepsis
Future research on thymosin α1 in sepsis should address the heterogeneity of the disease, particularly focusing on patients aged 60 and older and those with chronic conditions
Ethics statements
Ethical approval
This study was approved by the ethics committee of The First Affiliated Hospital of Sun Yat-sen University (2016007). The study protocol received approval from the ethics committees of all participating centres, adhering to local laws and the Declaration of Helsinki.
Data availability statement
Requests for data sharing can be sent to peif3@mail.sysu.edu.cn, wujianf@mail.sysu.edu.cn, or guanxd@mail.sysu.edu.cn. Only deidentified data will be made available to academic researchers upon reasonable request. Data will be available once all planned analyses have been completed and published or presented.
Acknowledgments
TESTS study collaborator group
The First Affiliated Hospital, Sun Yat-Sen University: Bin Ouyang, Changjie Cai, Minying Chen, Yongjun Liu, Zimeng Liu, Cuiping Wang, Yao Nie, Yihao Liu, Bin Li; The First People’s Hospital of Foshan: Xinhua Qiang, Fengyun Wang, Ping Xiao, Wei Lin; Jinling Hospital, Medical School of Nanjing University: Lu Ke, Jing Zhou, Qi Yang, Wenjian Mao; Zhejiang Provincial People’s Hospital: Jingquan Liu, Fangxiao Gong, Liannv Qiu; The First Affiliated Hospital of Guangzhou Medical University: Xiaoqing Liu, Runhua Li; The First Affiliated Hospital of Xi’an Jiaotong University: Ying Di, Jie Gao; Sun Yat-sen Memorial Hospital: Weichao Li, Zijun Zou; The Sixth Affiliated Hospital, Sun Yat-Sen University: Yiming Shi; West China Hospital of Sichuan University: Yao Chen, Chenggong Hu, Lin Yan, Weihua Feng; The Second Affiliated Hospital of Zhejiang University School of Medicine: Linlin Du, Tiancha Huang; Peking Union Medical College Hospital: Yi Chi, Wenyan Ding, Chunmei Huang, Wei Wu; Shandong Provincial Hospital Affiliated to Shandong First Medical University: Jicheng Zhang, Zijian Tai, Yan Liu; Ruijin Hospital, Shanghai JiaoTong University School of Medicine: Ying Chen, Zhitao Yang, Jing Lu; Guangzhou First People’s Hospital: Xiaoqin Wu, Wei Su; Zhejiang Hospital: Liang Xu, Yanfei Shen; Qingyuan People’s Hospital: Gang Luo, Weiliang Wu; The First Medical Centre, Chinese PLA General Hospital: Xin Hu, Zhi Mao; Peking University Shenzhen Hospital: Yingxin Lin, Weixing Zhang; Union Hospital, Tongji Medical College, Huazhong University of Science and Technology: Jiqian Xu, Xiaojing Zou; Beijing Friendship Hospital: Deyuan Zhi, Xiao Zhou; Zhuhai People’s Hospital: Wei Wang, Jiwen Zhong; Zhongshan Hospital Fudan University: Ming Zhong, Lizhen Xuan.
Footnotes
Contributors: JW, FP, and XG conceived the study. JW, FP, and XG designed the study, with input from all other authors. JW, FP, LZ, WL, RS, Y Li, ZW, ZH, XZ, XJ, Y Long, WC, CW, EC, JZ, JY, QL, FZ, LH, YS, MD, WZ, DZ, CY, and SP contributed equally to this work and are joint first authors. FP and JW wrote the original manuscript. JW, FP, LZ, WL, RS, Y Li, ZW, ZH, XZ, XJ, WC, Y Long, CW, EC, JZ, JY, QL, FZ, LH, YS, MD, WZ, DZ, QK, and XG contributed to data acquisition and review of the final manuscript. SZ and Y Liu coordinated the laboratory component. CY, SP, MS, and QZ contributed to the statistical analysis of the protocol and manuscript. JW, FP and KC revised the manuscript. FP, JW, and XG are joint corresponding authors and the guarantors. All authors interpreted the data and critically reviewed the manuscript. All authors revised the report and read and approved the final version before submission. The corresponding authors attest that all listed authors meet authorship criteria and that no others meeting the criteria have been omitted.
Funding: The trial was funded by the Sun Yat-sen University Clinical Research Program 5010 (grant No 2019002 and No 2024006), the Guangdong Clinical Research Center for Critical Care Medicine (grant No 2020B1111170005), and SciClone Pharmaceuticals. The funders were not involved in the trial’s design, data collection, interpretation, or manuscript preparation.
Competing interests: All authors have completed the ICMJE uniform disclosure form at https://www.icmje.org/disclosureof-interest/ and declare: support from the Sun Yat-sen University Clinical Research Program 5010, the Guangdong Clinical Research Center for Critical Care Medicine, and SciClone Pharmaceuticals for the submitted work; XG, JW, WL, ZH, Y Li, DZ, MZ, JY, WC, MD, ZW, and QL received grants from SciClone Pharmaceuticals; WL, JZ, and LK received consultancy fees from SciClone Pharmaceuticals, Jiangsu Nhwa Pharmaceutical, and Nutrici; no financial relationships with any organisations that might have an interest in the submitted work in the previous three years; no other relationships or activities that could appear to have influenced the submitted work.
Transparency: The lead authors (the manuscript’s guarantors) confirm that this manuscript accurately, honestly, and transparently represents the study being described. It ensures that no critical aspects of the study have been left out, and any deviations from the original study plan (and, if applicable, registration) have been clarified.
Dissemination to participants and related patient and public communities: Results were communicated to patients who expressed an interest during visits. Outcomes will be disseminated through press releases, academic conferences, and social media.
Provenance and peer review: Not commissioned; externally peer reviewed.
This is an Open Access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.