Ashwagandha, Withania somnifera (L.) Dunal, for the prophylaxis against SARS-CoV-2 infection: A multicentric randomized hydroxychloroquine controlled clinical trial in Indian health care workers

Abstract

Background

The study was planned when, the hydroxychloroquine (HCQ) was the only prophylactic agent approved by health authorities in several countries and no prophylactic COVID-19 vaccine was available.

Objective

The present study aimed to evaluate efficacy of Withania Somnifera (L.) Dunal (WS) as a chemoprophylactic and immunomodulatory agent against SARS-CoV-2 infection.

Methods

In a 12 week, randomized, open label, parallel group, two arm, comparative, multicentric, controlled trial compared WS with hydroxychloroquine (HCQ) in health care workers (HCWs). Total 400 HCWs were randomized in 1:1 ratio to receive either oral WS (500 mg for 12 weeks) or HCQ 400 mg (for 7 weeks). The primary outcome was to establish equivalence between WS and HCQ for the proportion of participants contracting SARS-CoV-2 infection.

Results

Seven participants contracted SARS-CoV-2 infection: 5 in WS arm and 2 in HCQ arm. The equivalence between WS and HCQ was established for the proportion difference of participants contracting SARS-CoV-2 infection for per-protocol (PP) (1.6%, 95% CI: −1.08%–4.33%) and in subgroup analysis (ITT, mIIT, non-vaccinated and seronegative).

Notably, the immunomodulatory effect of WS stood scientifically validated by the statistically significant difference in cytokine levels (p < 0.0001) at 12 weeks compared to baseline for Tumor Necrosis Factor (TNF)-alpha, Interleukin (IL)-2, IL-10, IL-17 and Monocyte chemoattractant protein-1 (MCP-1).

Gastrointestinal-related AEs were most frequent (53 in WS and 58 in HCQ). Headache and sneezing were observed only with HCQ. Participant global assessment showed excellent tolerability with both treatment arms.

Conclusion

WS was found equivalent to HCQ as a prophylactic against SARS-CoV-2 infection with no safety concern. WS is thus inferred to be an effective and safe Ayurvedic intervention for prophylaxis against SARS-CoV-2 infection, and also as an immunobooster.

Graphical abstract

Delete this figure

1. Introduction

Background: SARS-COV-2 virus infection (also known as COVID-19) is an acute respiratory tract infection and the overall mortality of COVID-19, due to respiratory complications, is several folds higher than that of influenza. Efforts have focused on developing preventive strategies against the COVID-19. Health care workers (HCWs) and those with a history of close contact with a patient with confirmed SARS-COV-2 virus infection are at high risk of infection. Thus, there is an urgent need to protect these high-risk, vulnerable HCWs by identifying effective and safe chemoprophylaxis against COVID-19. In a comprehensive effort to plug the lacunae in the overall medical and health care management of COVID-19, the AYUSH Government of India (GOI) clinical research program has been implemented for the prophylaxis and management of SARS-COV-2 virus infection. When this study was planned, the hydroxychloroquine (HCQ) was the only prophylactic agent approved by health authorities in several countries and no prophylactic COVID-19 vaccine was available.

Ashwagandha (Withania somnifera [WS]) is a medicinal herb that has been used in the Indian Ayurvedic system of medicine as a Rasayana (tonic) and indigenous medicine for millennia. It is widely prescribed for health benefits such as an immunomodulatory, anticancer, antiviral, anti-inflammatory, and adaptogenic agent [1,2]. The significance of WS in physical as well as in mental strengthening has been first substantiated in the Charaka Samhita and Susruta Samhita [3]. WS has been comprehensively documented for its chemical profile and pharmacological activities. A systematic review of 41 human trials suggested that WS has a potentially large array of therapeutic applications [4]. WS contains Withanolides (particularly withaferin A, D and withanolide G), a group of C28 steroidal lactones, which are attributed for the pharmacological effect of WS roots. It also reported to possess immunostimulating effects in animals [[5], [6], [7]]. The pharmacological profile of WS roots extract was found to be safe, well tolerated, and effective at decreasing stress and anxiety in the stressed individuals [6]. Studies indicate that WS possesses anti-inflammatory, antitumor, antistress, antioxidant, immunomodulatory, hemopoetic, memory enhancing, antiparkinsonian, antivenom and rejuvenating properties, including a positive impact on the endocrine, cardiopulmonary, central nervous system as well as sexual behavior and tolerance [8,9]. WS significantly reduces the stress as well as significantly improves the general health, fatigue, energy level, normal sleep, appetite and happiness. The observed improvement in stress with WS provides further evidence of same traditional and well documented claim of WS benefits in reducing stress in individuals with chronic stress [[10], [11], [12]]. A study also reported an immunostimulatory activity of WS along with significant response of hemolytic antibody towards human erythrocytes [13]. WS is also reported to improve cell-mediated immunity [1].

The Ministry of AYUSH (MoA), GoI has recommended WS root powder to improve individual's natural defense system (immunity) against COVID-19 infection [14].

Hence, this study was planned in collaboration with both, allopathic and ayurvedic investigators.

Objectives: To evaluate the potential clinical efficacy and safety of WS as a prophylactic agent against infection with SARS-CoV-02 in HCWs.

2. Methods

2.1. Study design

This was a prospective, randomized, open label, parallel group, two arm, comparative, multicentric, controlled, efficacy clinical trial of 12-week duration with an equivalence design to compare efficacy, safety and other health benefits of WS with HCQ for prophylaxis against COVID-19. The study was conducted between October 2020 to April 2021.

2.2. Materials

The WS test item was gifted by Pharmanza Herbals Pvt Ltd, Gujarat and tested at Council of Scientific and Industrial Research (CSIR). The WS tablets were assessed for quality check and the withanolide content using High Performance Liquid Chromatography (HPLC) at CSIR- Indian Institute of Integrative Medicine (IIIM), Jammu. The data is summarized in Table 1.

Table 1.

Quantification of WS by HPLC.

Marker compound	Percent content (w/w)
Withanone	0.01
Withaferin-A	0.01
12-deoxy withastramonolide	0.06
Withanolide-A	0.06

WS: Withania Somnifera; HPLC: High Performance Liquid Chromatography; w/w: weight by weight.

HCQ (brand name Cartiquin 400) was commercially procured.

2.3. Study protocol and ethical approval

The study protocol was finalized after considering the comments and suggestions from the reviewers. This study was conducted in 8 study centers in India after receiving protocol approval by respective institutional ethics committees of each site as well as central ethics committee by Central Council for Research in Ayurvedic Science, New Delhi.

The protocol was prepared though technical support and guidance from Indian Council of Medical Research (ICMR). This study was conducted in both the types of hospitals, modern medicine and Ayurveda hospitals, and the principal investigators (PIs)/Co-PIs were drawn from both streams. The clinical sites were All India Institute of Ayurveda (AIIA), New Delhi; TN Medical College & BYL Nair Hospital, Mumbai; Institute of Medical Sciences (IMS), Varanasi; Seth G.S. Medical College and K.E.M. Hospital, Mumbai; Grant Government Medical College & Sir J. J. Group of Hospitals, Mumbai; King George's Medical University, Lucknow; Sardar Vallabhbhai Patel Institute of Medical Sciences and Research, Ahmedabad; Employees' State Insurance (ESI) Hospital & PGIMSR, Mumbai.

Trial Registration Number on Clinical Trial Registry of India: CTRI/2020/05/025332.

2.4. Eligibility

HCQ naïve HCWs of either sex between 20 and 69 years of age, who tested negative for COVID-19 by reverse transcription–polymerase chain reaction (RT-PCR) and were willing for follow-up visits for 12 weeks were included into the study. While, HCWs with known hypersensitivity or contraindication to WS, pregnant or lactating women, unwilling to follow study procedures and non-cooperative were excluded. Also, HCWs with uncontrolled medical disease and taking medications that are known to prolonged QT interval on electrocardiogram (ECG) were excluded.

2.5. Study procedures and interventions

After written informed consent, 400 eligible HCQ naïve HCWs were randomized using permuted block randomization schedule in 1:1 allocation ratio to receive either WS or HCQ. A permuted block randomization list was divided in 20 strata among 8 sites, with each stratum comprising of 20 participants with 10 participants, each, to receive WS and HCQ in randomly chosen sequences. The participants had the autonomy to make a choice in selecting the study treatment using the Zelen model. In accordance with this, participants who were randomized to receive HCQ, could chose not to accept HCQ and opt for WS and vice versa at the time of randomization.

Participants in the WS arm took 2 tablets of 250 mg root aqueous extract of WS orally, administered twice daily following meals for 12 weeks while participants in the HCQ arm took 400 mg HCQ twice a day with a meal on Day 1; subsequently, 400 mg once a week with a meal for 7 weeks (prescribed treatment for high-risk HCWs as per ICMR Advisory) with continued follow-up till 12 weeks (please refer Appendix 1).

Study visits were planned every 4 weeks till 12 weeks and each visit involved vitals and clinical examination, ayurvedic evaluation by ayurvedic physician and nasal/oropharyngeal sample collection for RT-PCR testing. Information regarding safety (including monitoring of AEs), medication adherence and side effects were collected throughout the study. The assessment of general physical health, psychosocial health, and Quality of Life (QOL) was performed by using validated WHO Quality of Life-BREF (WHO QOL-BREF) and Health Related-Behavior Habit and Fitness (HR-BHF) questionnaire was evaluated based on visual analogue scale (VAS). Ayurvedic prakruti analysis (for kapha, pitta and vata) was performed using Ayurveda Case Record Form, which is a protocol defined prakriti Questionnaire. Qualitative assessment for anti-SARS-CoV-2 antibodies such as IgG and IgM and selected cytokine assay was performed.

Participants who developed confirmed SARS-CoV-2 infection were withdrawn from the study and a positive RT-PCR report on nasal/oropharyngeal sample was considered as an AE since all participants were RT-PCR negative at the time of randomization.

2.6. Outcome measures

The primary outcome was the proportion of SARS-CoV-2 infection free participants and the proportion of participants contracting SARS-CoV-2 infection as confirmed by real time RT-PCR test on nose/mouth swab at any time during the study period. Secondary outcomes include drug related AE, drug tolerability, assessment of WHO QOL Bref and HR-BHF questionnaire.

2.7. Sample size

The trial was designed with expected 7% and 14% infection at any time during the study period with HCQ and WS, respectively. The sample size was calculated using standard statistical method for equivalence design study: both manual and by a software (COMPARE II) with alpha of 2.5%, 80% power and equivalence margin of 15% (7.5% two sided), which provided the conservative estimate of 124 participants per arm. By considering the dropouts and compensating for power reduction due to Zelen model, final sample size of 200 participants in each arm was established for this study.

2.8. Statistical analysis

The analysis was performed on pooled data from the study sites. The statistical data was analyzed and performed by SAS® version 9.4. The statistical analysis was performed in accordance with the approved study protocol, including data analysis of intention to treatment (ITT) and per protocol (PP) population. All randomized participants in PP population completed 12- weeks of study treatment without any major deviation while participants in ITT population completed at least 4 weeks of intervention and follow up visit. Participants in modified intent to treat (mITT) population received at least 1 dose of study medications. All randomized participants who had taken at least 80% of the study medications were considered as compliant. A comparison between WS and HCQ was performed for the incidence of SARS-CoV-2 infection and proportion of participants contracting SARS-CoV-2 infection during the study. The equivalence analysis between two treatments was demonstrated based on whether the 95% confidence interval (CI) for proportion difference for SARS-CoV-2 infected and infection free participants were within the predefined equivalence margin of 15% (7.5% two sided, −0.075, +0.075) using Two one sided tests (TOST). The analysis of safety based on incidence of AEs was included as secondary outcome.

Complete data were collected for each study visits on paper source and transcribed into the electronic CRF (eCRF). This eCRF data along with source data was verified by the staff and quality assurance (QA) personnel of the appointed contract research organization (CRO). The clinical trial was also audited by an independent third party.

3. Results:

3.1. Characteristics of participants

The participants in both study arms had similar baseline characteristics with no significant difference for gender, age, weight, height and Body mass index (BMI) as shown in Table 2 and Appendix 2. The mean age of the participants was 36.85 years and 34.92 years in WS and HCQ arm, respectively.

Table 2.

Demographic characteristics of both treatment arms.

Variables	WS (N = 200)	HCQ (N = 200)	p valuea
Gender Male [n (%)] Female [n (%)]	111 (55.50%) 89 (44.50%)	114 (57.00%) 86 (43.00%)	0.7624b
Age (Years) [Mean (SD)]	36.85 (11.89)	34.92 (11.37)	0.0979c
Weight (kg) [Mean (SD)]	64.25 (11.57)	63.68 (12.16)	0.6308c
Height (cm) [Mean (SD)]	161.48 (9.45)	161.22 (9.89)	0.7863c
BMI (kg/cm2) [Mean (SD)]	24.66 (4.24)	24.42 (3.56)	0.5366c
Comorbidities [n (%)]	19 (9.5%)	14 (7.0%)	0.0979b

WS: Withania Somnifera; HCQ: hydroxychloroquine; SD: standard deviation; kg: kilogram; cm: centimeter; BMI: Body mass index.

^ap > 0.05 indicates no significant difference between two treatments.

^bp-value calculated using chi-square test.

^cp-value calculated using t-test.

As shown in Fig. 1, nine participants were lost to follow-up (7 in WS and 2 in HCQ) while 8 participants were withdrawn from the study (4 in WS and 4 in HCQ) due to switching of the treatment (3), consent withdrawal (2), Investigational product (IP) non-compliance (1), PI decision (1) and due to AE (1). Deviations related to visits outside of the defined window period, scheduled laboratory assessment and randomization allocation was observed, which were primarily due to the prevailing COVID-19 pandemic situation. There were 3 protocol violations related to switching over from HCQ to WS after 4 weeks of treatment and these 3 participants were excluded from the efficacy analysis.

Fig. 1.

CONSORT Flow Diagram for participant disposition.

3.2. Comparison of the primary efficacy measure between groups

3.2.1. Proportion of participants contracted SARS-CoV-2 infection

Total 5 (2.7%) participants in WS arm and 2 (1.0%) participants in HCQ arm contracted SARS-CoV-2 infection as confirmed by RT-PCR test on nose/mouth swab during the study by PP analysis. Further, 5 (2.5%) participants in WS arm and 2 (1.0%) participants in HCQ arm contracting SARS-CoV-2 infection in ITT analysis. The 95% confidence interval (CI) for the proportion difference in participants contracting SARS-CoV-2 infection between WS and HCQ by both ITT and PP analyses were within the pre-specified equivalence margin of 15% (7.5% two sided) as shown in Table 3 and Fig. 2. Thus, the primary efficacy measure was met for PP and ITT population.

Table 3.

Comparison of proportion of participants contracting COVID-19 between two treatment arms.

Data sets	WS Arm	HCQ Arm	Proportion difference	95% CI
PP population (N = 382)	2.7% (5/188)	1.0% (2/194)	1.6%	−1.08%–4.33%#
Non-vaccinated (PP) (N = 279)	2.9% (4/136)	1.4% (2/143)	1.5%	−1.89%–4.97%#
Seronegative (PP) (N = 351)	2.9% (5/173)	0.6% (1/178)	2.3%	−0.40%–5.06%#
ITT∗∗ (N = 393)	2.5% (5/198)	1.0% (2/195)	1.5%	−1.10%–4.10%#
mITT∗∗∗ (N = 396)	2.5% (5/200)	1.0% (2/196)	1.5%	−1.10%–4.06%#

WS: Withania Somnifera; HCQ: hydroxychloroquine; Data presented as % as proportion (Number of incidence/total no. of participants). ∗PP: Per protocol; ∗∗ITT: Intention to treat (all the randomized participants completed at least 4 weeks of intervention and completed the follow up visit at 4 weeks post enrollment); ∗∗∗mITT: Modified Intention-to-treat (Participants who received at least 1 dose of study medications). ^#95% CI for equivalence analysis using Two one sided tests (TOST).

Fig. 2.

Forest plot for proportion of participants contracting COVID-19.

PP: Per protocol; ITT: Intention to treat; mITT: modified Intention to treat; HCQ: hydroxychloroquine; ESI: Employees' State Insurance; BHU: Banaras Hindu University.

Similarly, in a subgroup analysis of mITT, non-vaccinated participants and seronegative participants, the proportion difference of participants contracting SARS-CoV-2 infection between the WS and HCQ was within the equivalence margin. Only 1 out of 118 COVID-19 vaccinated participants became RT-PCR positive in WS arm after receiving both vaccine doses.

Among the participants in PP population, 31 participants tested seropositive (positive antibody results for either IgG or IgM or both antibodies), 15 in WS and 16 in HCQ arm. Of the 7 COVID-19 positive participants, 6 participants reported negative IgG and IgM during the study and only 1 participant in HCQ arm reported IgM positive since baseline. The details of the 7 participants contracted COVID-19 during the study is presented in Appendix 3.

3.2.2. Proportion of infection free participants

As a second primary efficacy measure of infection free participants by PP analysis, 97.3% and 99.0% participants were infection free with WS arm and HCQ arm, respectively. While, 97.1% and 98.6% participants were infection free in ITT analysis. The proportion difference of infection free participants between two treatments is within the equivalence margin for PP, ITT. Similar trend was observed in subgroup analysis of mITT, non-vaccinated and seronegative participants, who remained infection free during the study as presented in Table 4.

Table 4.

Comparison of Proportion of infection free participants between two treatment arms.

Data sets	WS Arm	HCQ Arm	Proportion difference	95% CI
PP population (N = 382)	97.3% (183/188)	99.0% (192/194)	−1.6%	−4.33%–1.08%#
Non-vaccinated (PP∗) (N = 279)	97.1% (132/136)	98.6% (141/143)	−1.5%	−4.97% −1.89%#
Seronegative (PP) (N = 351)	97.1% (168/173)	99.4% (177/178)	−2.3%	−5.06%–0.40%#
ITT∗∗ (N = 393)	97.5% (193/198)	99.0% (193/195)	−1.5%	−4.10%–1.10%#
mITT∗∗∗ (N = 396)	97.5% (195/200)	99.0% (194/196)	−1.5%	−4.06%–1.10%#

WS: Withania Somnifera; HCQ: hydroxychloroquine; Data presented as % as proportion (Number of incidence/total no. of participants); ∗PP: Per protocol; ∗∗ITT: Intention to treat (all the randomized participants completed at least 4 weeks of intervention and completed the follow up visit at 4 weeks post enrollment); ∗∗∗mITT: Modified Intention-to-treat (Participants who received at least 1 dose of study medications); ^#95% CI for equivalence analysis using Two one sided tests (TOST).

3.3. Comparison of the secondary efficacy measure between arms

3.3.1. Serology for specific anti SARS-CoV-2 IgM and IgG antibodies

Participants’ immune status was evaluated based on serology profile for anti SARS-CoV-2 IgM and IgG antibodies. Participants with negative (<1.40) SARS-CoV-2 nucleoprotein-protein-specific antibodies were 55% (6 out of 11) in WS arm and 50% (2 out of 4) in HCQ arm with no significant difference for mean antibody levels between WS and HCQ for participants in positive (2.43 vs. 2.97, p = 0.5613) and for participants in negative antibodies (0.16 vs. 0.18, p = 1.0000) (refer Appendix 4). Also, the mean SARS-CoV-2 spike protein-specific IgG antibody levels in 15 participants were not significantly different between WS and HCQ (295.86 vs. 311.05, p = 0.5357) (Appendix 5 from supporting information).

3.3.2. Comparison of WHOQOL-BREF and HR-BHF QOL questionnaires between two arms

As this study was conducted during the peak COVID-19 pandemic, certain limitation were enforced in order to comply with GoI directives, with the goal to maintain minimum physical interaction with the participants.

Comparison of mean change in individual domain as well as overall WHO QOL-BREF showed no significant difference after 12 weeks of study between the WS and HCQ (p > 0.05) (refer Appendix 6). No substantial inference could be drawn on the HR BHF QoL data (refer Appendix 7). The comparison for mean change in any of the individual component of HR BHF QoL data between WS and HCW showed no significant difference (p > 0.05).

3.3.3. Comparison of cytokine levels between two arms

Cytokine analysis were performed at two study sites for Gamma Interferon, TNF-α, TNF-β, IL-6, IL-2, IL-10, IL-17, MCP 1 in subgroup of 117 participants (refer Appendix 8 and Appendix 9). There was no significant difference between WS and HCQ for mean change in cytokine levels from baseline to 12 weeks (p > 0.05) for any of the cytokines.

Of significance, the study conducted demonstrated and validated the ancient knowledge that WS possessed immunomodulatory properties. Notably, in concordance with the already known Ayurveda knowledge, the immunomodulatory effect of WS stood scientifically validated, by the statistically significant difference in cytokine levels at 12 weeks compared to the baseline levels for TNF-alpha (p < 0.0001), IL-2 (p < 0.0001), IL-10 (p = 0.0004), IL-17 (p < 0.0001) and MCP-1 (p < 0.0001). While, IL-6 level was significantly increased in HCQ arm only (p < 0.0001). TNF-alpha and IL-2 were significantly increase while MCP 1 was significantly decreased within both treatment arm.

3.3.4. Ayurvedic analysis

Prakruti analysis is Ayurveda's unique way of determining a body type for an individual. WS and HCQ were comparable for proportion of participants contracting SARS-CoV-2 infection in terms of Kapha (2.4% vs. 0.8%), Pitta (2.7% vs. 3.1%) and Vata (2.8% vs. 0%) as a dominant prakruti as shown in Table 5.

Table 5.

Ayurvedic prakruti analysis for participant contracting COVID-19 in both treatment arms.

Assessment	WS (N = 200)			HCQ (N = 200)
	Kapha (n = 127)	Pitta (n = 37)	Vata (n = 36)	Kapha (n = 124)	Pitta (n = 32)
No. of COVID-19 positive	3	1	1	1	1
Proportion	2.4% (3/127)	2.7% (1/37)	2.8% (1/36)	0.8% (1/124)	3.1% (1/32)

WS: Withania Somnifera; HCQ: hydroxychloroquine.

3.3.5. Safety

In terms of safety, no serious adverse event (SAE) was observed in this study. Adverse event (AE) profile analysis showed that 98 (24.5%) of participants reported at least one AE. Although numerically fewer AEs were observed with WS compared to HCQ, no significant difference was observed between WS and HCQ for the number of AEs (70 vs. 94, p = 0.0609) as well as number of participants experiencing AE (53 vs. 45, p = 0.3523), respectively, as judged by the investigators during the study. Majority of AEs were mild in nature. No SAE or death was reported. One participant was withdrawn due to AE of tuberculosis (TB) during the study. Among the 165 AEs, 111 incidences (67.3%, 53 in WS and 58 in HCQ) were of GI related AEs and majority of them were abdominal discomfort (n = 33), nausea (n = 26), vomiting (n = 17), hyperchlorhydria (n = 17), constipation (n = 8) and diarrhea (n = 6). This was followed by lab investigations (n = 15, 9.1%, 8 in WS and 7 in HCQ), headache in HCQ only (n = 9, 5.4%) and 8 incidence each of infections (including 7 SARS-CoV-2 infection) and general disorder. Remaining AEs were respiratory related (n = 6), urological (n = 3), metabolic (n = 2 anorexia in HCQ) and 1 incidence each of backache (in HCQ), sleepiness (in HCQ) and anemia (in HCQ). None of the ECGs, performed during the study, showed any abnormality that could be attributed as a drug-related AE or of any clinical significance. The number of AEs and number of participants experiencing those AEs during the study were summarized as per treatment arm and system organ class in (see Appendix 10). Fifteen (15) laboratory abnormalities reported as AEs are summarized in Appendix 10 under investigations. All these AEs were mild, except 1 moderate AE of raised fating blood glucose (FBS) in HCQ arm. Eight participants reported abnormalities in lipid profile (5 in WS and 3 in HCQ) in parameters such as cholesterol, Low Density Lipoprotein (LDL), Very Low Density Lipoprotein (VLDL). All participants with laboratory abnormalities were advised to follow up with their physician. The results of laboratory test results of hematology and biochemical parameters showed no statistically significant differences between WS and HCQ (p ≥ 0.05), except for hematocrit (%, p = 0.0181), Neutrophils (%, p = 0.0368) and serum bilirubin (mg/dl, p = 0.0429).

4. Discussion

Many clinical studies in human have investigated the pharmacological properties and clinical benefit of WS in an attempt to authenticate its use, including the studies in children. The effect of WS was studied in breast cancer, moderately severe rheumatoid arthritis (RA), schizophrenia, oligospermia, female sexual dysfunction, insomnia and anxiety, as an anxiolytic, body weight management, hypothyroidism, increasing muscle mass and strength, improving generalized weakness and muscular strength, enhancing cognition and memory and improving quality of sleep, among others [[15], [16], [17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29]].

In animals, immunostimulatory activity of WS was observed in mice as well as immunomodulatory activity against IgE mediated and cell-mediated hyper-reactivity [13,30]. WS significantly decreased the serum levels of TNF-α, IL-1β, IL-6 and matrix metalloproteinase-8 (MMP-8), while, IL-10 levels were increased, suggesting the anti-inflammatory action in rats [31]. This anti-inflammatory properties of WS was also shown in human to be clinically effective and safe as a polyherbal formulation in controlling moderately severe RA [17].

In this prospective, randomized, controlled, open label, two-arm, parallel, multicenter drug trial of a 12 weeks duration, authors have attempted to demonstrate the efficacy of WS as a prophylactic agent against SARS-CoV-2 infection as well as equivalence with that of HCQ in high-risk HCWs who are involved in the management of and providing services to SARS-CoV-2 infected patients. Total 400 consenting volunteers were enrolled in this two-arm study across 8 study sites. The current study was designed with an element of urgency during the peak period of the first pandemic wave in June 2020, when there were no medical interventions yet announced or approved for the infliction, and also HCQ was the only medical intervention approved for prophylaxis against SARS-CoV-2, especially for HCWs.

Few of the study specific activities (such as randomization procedure, Zelen model implementation, response to QoL questionnaire, HCQ dispensing) were adapted to maintain the integrity of the study. As per Zelen Model, participants were permitted to exercise their choice of the study drug. At two study sites (ESIC Mumbai and BHU, Varanasi), participants were allowed to select the blinded envelop prior to randomization, which neither influenced the treatment allocation nor create any bias and thus did not affect the outcome of the study.

The participants response for the WHO QOL and HR-BHF score were documented by the designated study team member based on the answers given by the participants in person, instead of documenting the QoL by the participants themselves. Although it is perceived as a limitation, this method was implemented so as to avoid transmission/surface contact of SARS-COV-2 virus during COVID-19 pandemic and to keep minimum contact (i.e. exchange of documents).

Also, variations observed in HCQ dosing, with some participants receiving 8 doses as compared to 9, did not impact the study results, as further subgroup analysis showed equivalence between the WS and HCQ. Drug compliance criteria as percentage (%) of drug consumption was not pre-defined and participant with at least 80% compliance to the study medications were included in the analysis.

The strength of this is study is that RT-PCR test was performed at every 4 weeks for all the randomized participants, regardless of any symptoms of COVID-19 infection. This study holds more significance in comparison to a recently published study on the prophylaxis of SARS-CoV-2 infection by concluding that WS is non-inferior to HCQ [32]. The study under reference, besides demonstrating equivalence between WS and HCQ, and also scientifically validating the immunoboosting properties of WS based on the evidences from eight different sites, from trials managed by Ayurvedic as well as modern medicine experts as study investigators.

The study has systematically not only addressed clinical and biochemical parameters as has been presented in the preceding sections, but also ensured the quality of the test material, which is independently verified by the sponsoring institution.

Further, the study has also involved a third-party audit of the clinical trial, and the required corrective and preventive action (CAPA) report has been filed.

The conduct of this multicentric trial through well-defined protocol and standards set, despite the challenges faced due to COVID-19 pandemic, and its critical monitoring by modern medicine & sciences and Ayurveda experts shall serve as an exemplary model for future study of AYUSH interventions through modern science and medicine parameters.

Five participants on WS and two participants on HCQ prophylaxis contracted RT-PCR confirmed SARS-CoV-2 infection and the proportion difference by PP and ITT analysis were within the predefine equivalence margin of 15% (7.5% two sided). This outcome of equivalence between WS and HCQ remained consistent when evaluated for non-vaccinated participants as well as seronegative participants. Previous studies have shown effectiveness of HCQ prophylaxis against SARS-CoV-2 infection in HCWs until vaccines were made available and accessible [33].

Of the participants receiving both vaccine doses, only 0.8% became RT-PCR positive in WS arm.

On the Ayurveda component, while the immunomodulatory activity of WS is proved validated through this study, the observation whether participants with Kapha as dominant ayurvedic prakruti are more prone to develop SARS-CoV-2 infection requires further supporting evidence.

Safety of the daily oral administration of WS was comparable with that of the weekly regimen of HCQ. There was no distinct advantage of WS in the current study population for the safety and tolerability profile as compared to HCQ. Maximum AE were related to the gastrointestinal system and were comparable between WS and HCQ. No previously unknown safety concerns was observed with either WS or HCQ as the safety of WS and HCQ is well established in multiple literature [[34], [35], [36], [37]]. This study further corroborates the reported safety and tolerability of WS in healthy volunteers [36].

The use of HCQ as comparator was based on the ICMR recommendation for judicious use of HCQ as a prophylactic agent against SARS-COV-2 infection, which is drawn from the evidence of pre-clinical in vitro studies and a few retrospective observational studies showing lower risk of contracting COVID-19 [36,38].

Further, an observational study in over 12,000 HCWs (4257 using HCQ vs. 7826 not using HCQ) evaluated HCQ's potential prophylactic effect against COVID-19, and reported that HCQ reduced the risk of COVID-19 in well tolerable manner, with further improvement in protection as duration of HCQ intake increased [33].

Thus, in retrospect, selection of HCQ as active comparator and control prophylactic strategy can be considered as a prudent approach under controlled clinical trial setting after rigorous assessment of the risks and benefits based on the available evidence to address the dire need for safe alternatives to combat the COVID-19 pandemic, especially in the absence of any vaccine or other proven prophylactic alternatives against COVID-19.

The result of the present study also supports the interim analysis of the published study in high risk HCW, which indicated that WS was not inferior to HCQ as a prophylaxis against COVID-19 [32].

Together with the result of this study and the long history of WS use in the community with available data on its efficacy and safety in human, WS can, unequivocally, be considered as a safer option to HCQ and an effective Ayurvedic option in COVID-19 management.

5. Conclusion

The objective was to evaluate the prophylactic potential of WS against SARS-CoV-2 infection with a research agenda to bridge the gap between the need and the potential of AYUSH systems in combating COVID-19 pandemic. In the absence of any proven chemoprophylaxis against SARS-CoV-2 infection including vaccines at the time of study initiation, the clinical study results demonstrate that WS is equivalent to HCQ in terms of prophylaxis against SARS-CoV-2 infection as well as safe. The study conducted thus ascertains and scientifically validates the claims on WS as per Ayurveda on immunomodulation. In today's context, while vaccination will continue as a primary measure against protection from SARS-CoV-2 infection, in view of the proven immunomodulatory effect of WS, the Ayurveda intervention may further help to boost and sustain the immunity of even the vaccinated population.

Data availability

The data that support the findings of this study are available on request from the corresponding author.

Author contributions

All authors participated in the manuscript review and writing.

Funding sources

This study (AYUSH-CSIR-HCP-01) was sponsored by Council of Scientific and Industrial Research (CSIR) under the New Millennium Indian Technology Leadership Initiative (NMITLI) (Project no. TLP110001) and is one of the four clinical studies pursued under the collaboration of Ministry of AYUSH and CSIR for COVID-19. Institutional publication number. CSIR-IIIM/IPR/00467.

Conflict of interest

All the authors confirm that there is no conflict of interest to declare. Cadila Pharmaceuticals Limited was the Contract Research Organization for this study.

Appendix A. Supplementary data

The following are the Supplementary data to this article: