Protection From SARS-CoV-2 Infection With an Oral Nosode (BiosimCovex): A Randomized, Double-Blind, Placebo-Controlled, Phase III Study

Abstract

Background and Objectives

This study investigated a potentized, oral homeopathic SARS-CoV-2 nosode (BiosimCovex) for safety, efficacy, and immunogenicity against COVID-19 infection.

Methods

A phase III, multicentre, randomised, double-blind, placebo-controlled trial was conducted at 13 sites in India. 2671 unvaccinated, RT-PCR and SARS-CoV-2 antibody-negative participants were randomized 2:1 to receive 6 doses over 3 days of BiosimCovex (1781) or placebo (890). SARS-CoV-2 prevention efficacy was assessed through RT-PCR. Efficacy was defined as 100 x (1-IRR), where IRR is the ratio of positive RT-PCR test rates in the BiosimCovex compared to the placebo group. A two-sided Clopper-Pearson 95% confidence interval was adjusted for surveillance time. Immunogenicity was measured by SARS-CoV-2 antibody conversion at days 15, 30, 45, and 60. Safety was assessed through symptom questionnaires and laboratory studies every 15 days. Illness burden was assessed through symptom diaries.

Results

Maximum efficacy occurred at day 45 and was 50.06% overall, 71.88% in individuals aged 18-40 years; and 61.57% in females. By day 45, there were 23 positive RT-PCR cases per group (BiosimCovex 1.3%; placebo 2.6%, P = 0.007). A total of fifty-nine positive RT-PCR cases occurred, with thirty-three (1.86%) in the BiosimCovex and twenty-six (2.94%) in the placebo group, for a relative risk (RR) of 0.63 at the end of the study over 60 days. Antibody conversion increased over time, with 48.4% becoming antibody positive by day 60 in the treatment group compared to 13.4% in the placebo group (P < 0.001). All 169 (6.33%) mild adverse events resolved without sequelae (6.52% in treatment; 5.96% in placebo). There was a significant reduction in symptom duration (P = 0.0085) in the treatment group for those who became ill.

Conclusions

A potentized oral nosode was safe and appeared to prevent SARS-CoV-2 infection and be immunogenic over 60 days. Effectiveness was highest at day 45 in females and in the 18- to 40-year-old age group.

Introduction

The COVID-19 pandemic sparked a search for novel, safe, and effective prophylactics and treatments for pandemics. While new vaccines and drugs were developed, worldwide access to vaccines, vaccine hesitancy, and rapid adaptation by emerging variants remain a challenge. ¹

Potentization is a method of using serial dilutions of starting material from an infected person and has been used for over 200 years by homeopathic practitioners.^2,3 Material is processed in alcohol or water, accompanied by vigorous agitation (succussion). The Investigational Medicinal Product (IMP) BiosimCovex is a homeopathic nosode, which was prepared using the standardized process of serial succussion, as described in the Homeopathic Pharmacopoeia of the United States (HPUS) and the Homeopathic Pharmacopoeia of India (HPI). ⁴ The starting material was an oropharyngeal exudate from a patient with laboratory-confirmed SARS-CoV-2 infection (Wuhan strain) and underwent 30 cycles of potentization. The resulting 30C potency has been demonstrated to contain nanoparticles (manuscript in preparation), and we hypothesize that these nanoparticles capture immunogenic information from exosomes of the starting material. Spectroscopic studies show that potentization produces various other physicochemical properties, ⁵ affects the liquid’s structural properties, reduces the substrate to nano-dimensions, ⁶ and increases membrane permeability, ⁷ which may produce immunological signalling and responses in biological systems.

Nosodes: Microbial Inhibition, Protection, and Immunomodulation

Homeopathic physicians have used potentized preparations of infectious agents for nearly 200 years and claimed these preparations can reduce infection and illness. However, until the modern era, no rigorous research has evaluated these claims. Nosodes made from infectious organisms (such as Mycobacterium tuberculosis, ⁸ HIV ⁹ ), and tissues ¹⁰ have been shown to produce biological responses.^11,12 Nosodes have demonstrated safety and reported effectiveness during human epidemics. For example, during a meningococcal outbreak in Brazil, 65 826 people aged 0-20 who received a nosode had a death rate of 6.07 compared to a control group of 84.9 per 100 000 (91% effectiveness to prevent death). ¹³ In Cuba, during a leptospirosis epidemic, 2.3 million people given a nosode reportedly had an 84% reduction in incidence, while untreated regions had a 21% increase. ¹⁴ Influenza nosode resulted in changes to T and B cell balance after influenza antigen challenge, indicating subtle changes in anti-viral immune response regulation. ¹⁵

Our last author did a series of mouse studies at Walter Reed Army Institute of Research (WRAIR) and found that a nosode produced from Francisella Tularensis reduced mortality to subsequent exposure to the same pathogen by 22% and delayed death by 5 days compared to water-treated groups and also produced specific anti-tularensis antibodies. ¹⁶ This study also found hydrogen bonding changes in the nosode preparation compared to the succussed water control samples. A study of an E. Coli nosode reported reduced piglet diarrheal severity comparable to an antibiotic. ¹⁷ In vitro studies have shown growth inhibition of E. coli, K. pneumoniae, S. typhi, and C. albicans using respective nosodes. ¹⁸ Observational studies have shown a decrease in HIV ¹⁹ and Hepatitis C ²⁰ viral load after subjects’ exposure to their respective nosodes. Jesmine et al found that Hepatitis C nosode reduced liver cancer cell growth in vitro by modulating telomerase and topoisomerase II activities, promoting apoptosis via a mitochondrial pathway. ²¹ Khuda-Bukhsh et al reported that HIV 30C nosode induced cytotoxic effects on A549 cells, by changing nuclear condensation, DNA fragmentation, ROS (reactive oxygen species) generation, and matrix metalloproteinases, and had an inhibitory action on cell proliferation, cell migration, expression of telomerase reverse transcriptase and Top II genes, and increased expression of pro-apoptotic genes. ²² While these are a heterogeneous mix of studies and models, they do indicate that nosodes can and do produce biological and immunological effects consistent with historical claims of protection.

We initiated the present study against the backdrop of this data in the early stages of the COVID-19 pandemic, recognizing that in the absence of widely available vaccines and treatments, there was an urgent need for safe and easily developed, easily administered prophylaxis. BiosimCovex was selected because it was postulated to contain a nanoform of biologically active material that could stimulate an immune response and possibly prevent infection in individuals subsequently exposed to SARS-CoV-2 and because preliminary studies signaled that it produced such effects (see below). Thus, it might provide a complement to vaccines or an option for those who could not get vaccines or other prophylactic treatments. It could also serve as a specific test of the general homeopathic claims.

SARS-CoV-2 Nosode (BiosimCovex)

BiosimCovex is a SARS-CoV-2 (Wuhan strain) nosode, ²³ made at Mumbai’s Haffkine Institute on 22nd May 2020 from an infected individual’s oropharyngeal swab, as per HPI guidelines. ²⁴ The 30 C potency was prepared by potentizing SARS-CoV-2 viral material using alcohol as a vehicle and then applied onto globules, as per the standard manufacturing process. It was tested for toxicity in Swiss albino mice, repeat dose toxicity in Wistar Rats and rabbits as per OECD guidelines. ²⁵ A phase I human trial showed safety and upregulation of biomarkers for SARS-CoV-2 prophylaxis. ²⁶

A phase II study was done in 2233 high-risk individuals exposed to SARS-CoV-2 infection who were sequestered at Mumbai’s municipal corporation-run quarantine facility and found that BiosimCovex was safe and reduced the risk of infection up to 62% compared to placebo. ²⁷

The current study sought 2800 unvaccinated, healthy volunteers to further test the nosode’s efficacy, safety, and immunogenicity against SARS-CoV-2 infection.

Materials and Methods

Study Design and Participants

A phase III randomized, placebo-controlled, double-blind study was conducted at 13 sites across India. Inclusion criteria included absence of SARS-CoV-2 virus infection (negative RT-PCR test) and absence of COVID-19 antibodies (IgG and IgM) at baseline. Participants, aged 18-65, of any gender, provided written consent before enrolling.

Exclusion criteria included: recent illness, mild to moderate COVID-19 symptoms, prior SARS-CoV-2 vaccine, or concurrent COVID-19 study participation. Individuals were also excluded if they were pregnant or breastfeeding, or had a history of severe chronic cardiorespiratory disease, end-stage renal disease, progressive liver dysfunction, or active psychosis.

Recruitment management across 13 sites was monitored in alignment with the biweekly published COVID-19 positivity and infection rates. When reported infection rates in the general population fell below 3% in any of the cities where the study sites were located, recruitment there was paused and not resumed unless the overall infection rate ticked above 3%. During the study, the illness and infection rates fluctuated, although they generally trended downward as the study progressed.

The sample size estimate was based on the infection proportions in the BiosimCovex group (0.016) and the placebo group (0.039), in the prior Mumbai study. ²² Power analysis was done by comparing infectivity rates for placebo vs nosode with a two-sample test for proportions. The sampling distribution of the difference in sample proportions ${\widehat{p}}_1-{\widehat{p}}_2$ is approximately normal, with: Mean = 0.0394−0.0160 = 0.0234. Applying the null hypothesis, the sample size estimates were n1 = 933 (placebo) and n2 = 1867 (nosode), for a total of 2800. Trial recruitment was terminated in March 2022, after we had screened 2805 participants.

Randomization and Blinding

An Interactive Web Response System (IWRS) performed centralized randomization and treatment assignment. Blinded block randomization was done by pre-generated randomization schedules using SAS-V-9.4. To enhance the speed of recruitment, and because of the severity of the COVID-19 pandemic in India, the lack of a vaccine and very limited treatment for it at the time, and the positive results in the earlier study of BiosimCovex in a quarantined population, we chose a 2:1 randomization ratio in favor of the nosode. Our primary motivation was humanitarian and was designed to facilitate recruitment and prophylactic exposure. The statistical bias was addressed by assigning participants randomly, and the study and statistical analysis were fully blinded. 2800 participants were assigned in a 2:1 ratio to receive BiosimCovex or a placebo. The placebo pills (identical in appearance and taste and impregnated with 90% potentized ethanol) or BiosimCovex in identical vials were coded, with the subject’s identification number and corresponding IMP code assignment from the IWRS. Investigators, clinical staff, participants, consultants, analysts, and all others involved in the study were blinded to treatment assignment. All investigational medicinal products (IMPs) were managed by a centralized drug depot, which was responsible for generating and labelling the vials with the random codes. IMP’s were dispatched directly to the respective sites from the drug depot; the latter maintained and had exclusive access to the code book during the entire study period, maintaining blinding of the randomization and group assignment process to participants, clinicians, and investigators throughout the study and initial data analysis.

Procedures

Study Medication and Dose Regimen

In each group, BiosimCovex or placebo, 6 pills were taken orally, twice a day for 3 days, constituting one course. The first dose was given at the clinic, and subsequent doses were self-administered at home. Unused medication was returned.

Study visits occurred every 2 weeks. Safety was assessed through in-person interviews and baseline and end-of-study laboratory investigations. Symptom duration, intensity, immune status questionnaires, and laboratory tests were assessed biweekly. Antibody and RT-PCR tests were done at baseline, and biweekly RT-PCR tests on days 15, 30, 45, and 60 were added in January 2022 and were performed on a subset of participants, per the protocol amendment, as stated in the outcome section. Throughout the trial, participants who presented with symptoms at unscheduled visits were also tested for SARS-CoV-2.

Outcomes

At the beginning of the study, the primary endpoint was the number of symptomatic cases of RT-PCR-confirmed COVID during the 60 days of the study. Because of the devastating spread of COVID in the early months of the pandemic in India, strict quarantine restrictions were implemented. This likely acted as a disincentive to participants to report symptoms. Considering fluctuating COVID-19 rates, and to ensure that this trial would yield important information and justified human use, we queried the data in December 2021 (while maintaining absolute blinding), asking how many total symptomatic cases we had identified in our enrolled population. We learned that we had no cases, meaning that we could not generate any informative, between-group differences in symptom outcomes. Concurrently, ongoing data collection on vaccine efficacy, COVID community transmission, and the emergence of long-COVID suggested that prevention of actual infection (based on RT-PCR, to identify the COVID-positive cases) might be a more important endpoint for the study than symptoms. Also, given the emergence of the Omicron variant, which generally produced milder, more easily ignored symptoms, it was advisable to conduct more frequent RT-PCR tests. So, with the approval of the DSMB and the EC (Ethics Committee), the protocol was amended in January 2022 to add RT-PCR testing every 15 days and make between-group differences in RT-PCR positive rates the primary outcome. The protocol amendment was notified to the ethics committees, and informed consent was amended, informing the participants about this change in the RT-PCR test schedule. This modification allowed the examination of actual infection rates, including mild and asymptomatic cases, which might otherwise be missed. Because there were no self-reported cases of COVID during the study, the original primary endpoint (between-group comparison of self-reported RT-PCR-confirmed COVID cases) provided no meaningful data; thus, RT-PCR positivity alone became the primary determinant of the efficacy of BiosimCovex in reducing SARS-CoV-2 infection in the ITT (intention to treat) population.

Positive Cases Reported in the Study Continued to be Followed up for Secondary Outcomes

Secondary outcomes included the efficacy of BiosimCovex to prevent severe COVID-19, symptomatic SARS-CoV-2 infection, and death caused by COVID-19. Exploratory analyses of intensity/severity of COVID-19 symptoms, viral kinetics, symptom duration, all-cause mortality, burden of disease (BOD), use of concomitant medication (COVID-19 positive), and influenza-like symptoms for RT-PCR negative participants were also performed. Total antibodies to SARS-CoV-2 in human serum by Enzyme-Linked Immunosorbent Assay or chemiluminescent immunoassay (CLIA) methods (depending on the site location) were measured for all participants at baseline and day 60, and for a subset at days 15, 30, and 45. Antibody levels were measured at multiple visits. Blinding to group assignment was maintained throughout the study.

Statistical Analysis

A statistical analysis plan was developed before the database lock and unblinding. Data underwent blinded quality control and review, addressing anomalies before unblinding. Statistical analyses were performed using SAS®Version-9.4 with a 5% significance level (P ≤ 0.05). Power calculation was based on the total sample.

Continuous data were described with 95% confidence intervals (CI), and categorical variables were summarized using frequency count and percentage. Missing values were imputed using Last Observation Carried Forward (LOCF) for participants with at least 30 days of study completion.

The following hypotheses were evaluated. Null and Alternative Hypothesis: H₀:P1 ≥ P2 vs Ha: P1 < P2, where P1 is the proportion of subjects with positive RT-PCR in the BiosimCovex group and P2 is the proportion of subjects with positive RT-PCR in the placebo group. We used a one-sided test to determine the primary outcome of positive infection using RT-PCR. We also calculated the hazard ratio and two-sided P-value using a stratified Cox proportional regression model to analyse COVID-19 symptom duration (time to recovery).

The statistical analysis plan finalized before the start of the study specified that a one-sided test for relative risk would be performed, in the direction of lower risk for the treatment group than for the placebo group. This plan was justified on the basis of the data from the prior Mumbai quarantine study demonstrating a positive (preventive) effect against SARS-CoV-2 infection. The one-sided P-values address the question of whether the observed results are likely due to chance. However, in order to provide pragmatic information on the possible magnitude of the effect of the treatment in the population at large, and to allow for effects in both directions, two-sided confidence intervals are presented as well. These intervals are given for the efficacy, also known as the relative risk reduction. Efficacy is directly computed from the relative risk (RR) as (1 – RR) ×100%, representing the percentage reduction in risk as a result of the treatment. It can also be interpreted as the difference in risks for the 2 groups, as a percentage of the risk for the placebo group. The associated two-sided CI for efficacy was derived with the Clopper-Pearson method, adjusted for surveillance time (total time in 1000 person-days for the given endpoint across all participants within each group at risk).

A summary of qualitative and quantitative SARS-CoV-2 antibodies was generated. Quantitative P-values for comparison between the treatment groups were calculated using ANCOVA, considering baseline values as a covariate. For a qualitative data analysis, the participants having antibody titre above the reference range (present) were assigned a 1, and negative (absent) for antibodies were assigned 0. The P-values for comparison between the treatment groups were calculated using the two-sided chi-square test. The percentage of participants having an antibody present was reported for visits on days 15, 30, and 60.

Descriptive summaries with 95% CIs were provided. We coded and classified all reported adverse events (ADRs) using MedDRA version 23.0 standards, grouping them by preferred term (PT) and system organ class (SOC). We presented frequency counts (n) and percentages (%) of adverse events, including serious adverse events, along with a 2-sided 95% exact CI. Subjects with multiple ADRs were counted once in SOC/PT. All other reported P-values are one-sided unless otherwise indicated.

Ethics Approval and Consent to Participate

This study was performed in line with the principles of the Declaration of Helsinki. Ethics committees for each site approved this study. The protocol was approved by ethical committees and registered at the Clinical Trial Registry of India (https://ctri.nic.in, CTRI/2021/05/033834).

Consent to Participate

Informed consent was obtained from all participants included in the study.

Results

We screened 2805 participants from October 4, 2021, to March 8, 2022. Last visit occurred on May 11, 2022. Demographics appear in Table 1 and the flow chart (Figure 1).

Table 1.

Demographic Characteristics – Randomized Population

Characteristics	BiosimCovex (N = 1781)	Placebo (N = 890)	Total (N = 2671)
Age Mean (SD)	38.38 (12.38)	39.03 (12.47)	38.59 (12.41)
Male n (%)	936 (52.6)	484 (54.4)	1420 (53.2)
Female n (%)	845 (47.4)	406 (45.6)	1251 (46.8)
Height (m) Mean (SD)	1.64 (0.08)	1.654 (0.09)	1.64 (0.08)
Weight (kg) Mean (SD)	61.16 (10.32)	61.46 (10.07)	61.26 (10.23)
BMI (kg/m2) Mean (SD)	22.63 (3.83)	22.499 (3.74)	22.59 (3.80)

Figure 1.

Flow of Participants Through Study. The Per Protocol Population Included Patients From the Modified Intention-to-treat Population, Except Those With One or More Major *protocol Deviations

After exclusions, 2671 participants aged 18-65 were randomly assigned to receive BiosimCovex (1781) or placebo (890) in 6 doses over 3 days. There were no significant differences in the characteristics or dropout rates between groups, resulting in 2657 participants for ITT analysis. 98.6% of participants (2633) completed the 60-day follow-up. Demographic characteristics are shown in Table 1 and Figure 1. The 2 reasons for dropout were loss to follow-up and protocol noncompliance.

Relative Risk and Efficacy

The primary efficacy measure for BiosimCovex in preventing SARS-CoV-2 infection was the difference in positive RT-PCR case rates between the 2 groups. A total of fifty-nine positive RT-PCR cases occurred, with thirty-three (1.86%) in the BiosimCovex and twenty-six (2.94%) in the placebo group, for a relative risk (RR) of 0.63 (95% CI: 0.38 to 1.05) at the end of the study over 60 days. The RT-PCR case rate by visit was as follows: no positive cases were reported at day 15. On day 30 visit, there were ten cases (1.97%) in the BiosimCovex group and 6 cases (2.29%) in the placebo group, corresponding to a relative risk (RR) of 0.86. On day 45 visit, thirteen cases (1.38%) were observed in the BiosimCovex group compared to seventeen cases (3.53%) in the placebo group, with a relative risk (RR) of 0.39. On day 60 visit, ten cases (0.6%) were reported in the BiosimCovex group and 3 cases (0.3%) in the placebo group, resulting in a relative risk (RR) of 1.65. Overall prevention effectiveness peaked at day 45 [Efficacy (E) = 50.06 %; 95% CI (11.48, 71.82)]. Exploratory analyses on day 45 by age and gender showed the most pronounced case rate difference occurred in participants aged 18-40 (efficacy (E) = 71.88%; 95% CI (33.42, 88.12)) and in females (E = 61.57%; 95% CI (18.65, 81.84) with corresponding P-values of P = 0.001 and P = 0.0048, respectively. Although there was a statistically significant difference in case-rate (P = 0.038) on day 60, the efficacy was lower [E = 36.6%; 95% CI (−5.31, 61.84) than on day 45 (E = 50.06 %; 95% CI (11.48, 71.82)]. Additional exploratory analyses showed that on day 60, efficacy among females was 51.96%; 95% CI (6.89, 75.21). No significant differences in males and ages 41-65 were observed. The summary of SARS-CoV-2-positive cases at each visit showed no cases on day 15, sixteen cases on day 30, thirty cases on day 45, and thirteen cases on day 60 (Refer to Table 2).

Table 2.

Efficacy Analysis

Prevention of COVID-19 cases	BiosimCovex (N = 1772)		Placebo (N = 885)		RR (relative Risk)	P-value *One-sided	E (CI) **Two-sided
	Positive n (%)	Negative n (%)	Positive n (%)	Negative n (%)
Outcome (Day 45)
Overall	23 (1.3)	1749 (98.7)	23 (2.6)	862 (97.4)	0.49	0.0077	50.06 (11.48, 71.82)
Females	12 (1.4)	831 (98.6)	15 (3.7)	390 (96.3)	0.39	0.005	61.57 (18.65, 81.84)
Males	11 (1.2)	918 (98.8)	8 (1.7)	472 (98.3)	0.69	0.2283	28.96 (−75.45, 71.23)
Age 18-40	8 (0.8)	1002 (99.2)	14 (2.8)	483 (97.2)	0.29	0.001	71.88 (33.42, 88.12)
Age 41-65	15 (2)	747 (98)	9 (2.3)	379 (97.7)	0.83	0.3469	15.14 (−92.17, 62.52)
Outcome (Day 60)
Overall	33 (1.9)	1739 (98.1)	26 (2.9)	859 (97.1)	0.63	0.0381	36.61 (−5.31, 61.84)
Females	17 (2)	826 (98)	17 (4.2)	388 (95.8)	0.53	0.0134	51.96 (6.89, 75.21)
Males	16 (1.7)	913 (98.3)	9 (1.9)	471 (98.1)	0.89	0.4185	8.14 (−106.31, 59.10)
Age 18-40	17 (1.7)	993 (98.3)	16 (3.2)	481 (96.8)	0.53	0.0277	47.72 (−2.60, 73.36)
Age 41-65	16 (2.1)	746 (97.9)	10 (2.6)	378 (97.4)		0.3032	18.53 (−77.83, 62.68)
	Summary of COVID-19 antibodies
Presence of COVID antibodies n (%)	833 (48.4)	-	113 (13.4)	-		0.0001	-

The first positive RT-PCR tests occurred at visit 3 (day 30). Out of 769 participants tested at that visit, 16 were positive (10 in BiosimCovex group, 6 in the placebo group), with rates of 1.97% and 2.29%, respectively (P = 0.385). At the day-45 visit, a total of 30 positive RT-PCR tests occurred among the 1418 tested participants (13 in the BiosimCovex group, 17 in the placebo group, representing rates of 1.39% and 3.53%, respectively; P = 0.0039). By day 45, there were 46 cumulative positive RT-PCR cases: 23 in each group, representing a case rate of 1.29 in % BiosimCovex group and 2.59% in the placebo group (Figures 2 and 3).

Figure 2.

Cumulative Cases per 100

Figure 3.

Incidence Rate per 100 at Each Visit

The calculated surveillance person-time estimate of efficacy in preventing SARS-CoV-2 infection was 102 cases per 1000 person-years for BiosimCovex and 206 cases per 1000 person-years for placebo over 60 days. The highest efficacy was observed in the 18-40 age group, with 63 cases per 1000 person-years for BiosimCovex and 223 cases per 1000 person-years for placebo.

There were no self-reported symptomatic COVID-19 cases, and no deaths. Of the 59 cases of RT-PCR-confirmed infection, 32 were reported as symptomatic upon probing (19 BiosimCovex and 13 placebo), and all symptoms were mild. Because of the small number of symptomatic cases, the efficacy of BiosimCovex to prevent symptomatic illness and death was not reportable. Among RT-PCR positive symptomatic cases identified during the study, symptom duration in the BiosimCovex group was lower than in the placebo group (−0.3 days; P = 0.0085).

All participants were SARS-CoV-2 antibody negative at baseline (except for IgG positive at baseline n = 37, BiosimCovex n = 20, and placebo n = 17). Sensitivity analysis excluding these participants did not alter the statistical significance. Rates of antibody conversion from negative to positive climbed steadily throughout the study, with 48.4% in the treatment group becoming antibody positive by day 60 compared to 13.4% in the placebo group (Figure 4). Qualitative and quantitative analysis of antibodies showed statistically significant results at P < 0.001 over 60 days.

Figure 4.

Rate of Antibodies (in percent) at Each Visit

There were no significant between-group differences in the use of concomitant medication for SARS-CoV-2-positive participants or change in the immune status questionnaires.

Safety

A total of 2669 participants comprised the safety population. There were 169 (6.33%) adverse events (AE) with no major difference in AEs between the treatment (6.52%) and placebo (5.96%) groups (Table 3, Supplemental material S1).

Table 3.

Adverse Events (AE)

Overall adverse events – Safety population	BiosimCovex (N = 1781)	Placebo (N = 890)	Total (N = 2669)
Participants with AEs n (%)	86 (4.8)	38 (4.3)	124 (4.6)
Participants with study drug related AEs n (%)	6 (0.3)	1 (0.1)	7 (0.3)
Overall AEs by severity (mild) n (%)	116 (6.52)	53 (5.96)	169 (6.33)

Discussion

BiosimCovex helped prevent SARS-CoV-2 infection in an unvaccinated adult population, with an overall efficacy of 50.06% (95% CI: 11.48, 71.82). Efficacy compared to placebo peaked at day 45 (BiosimCovex rate 1.29% vs placebo rate 2.59%) and was highest among those aged 18-40 (71.88%) and females (61.57%). Efficacy appeared to wane after day 45. SARS-CoV-2 antibody conversion rate from negative at baseline to positive at day 60 was 48.4% in the treatment group and 13.4% in the placebo group, demonstrating the likely effects of undetected and asymptomatic infection. In the subgroup exploratory analysis of the RT-PCR negative population (n = 449), where RT-PCR was performed biweekly, the antibody conversion rate at day 60 was 55.67% in the treatment group and 10.07% in the placebo group, suggesting an immunogenic benefit of the nosode.

The current study represents the first large, randomized, placebo-controlled, double-blinded trial of a homeopathic nosode for the prevention of infection. Placebo-controlled trials of COVID-19 vaccines, including mRNA vaccines, showed strong efficacy in preventing severe illness and hospitalization, prompting FDA expert panel recommendations and CDC emergency use authorization for 3 vaccines. In addition to preventing severe COVID-19 illness, several vaccines (Moderna and Pfizer-BioNtech) were shown to prevent SARS-CoV-2 infection up to 120 days (95.2% and 91.6%, respectively. ²⁸ BiosimCovex’s peak efficacy in preventing SARS-CoV-2 infection, while less robust, at 71.88% (ages 18-40), up to 45 days, is still notable and is the only one taken orally.

One significant limitation of the study was a change in the primary endpoint. We did this because the change in pandemic patterns and the absence of any reported illness among study participants posed the risk that the study would fail to provide any meaningful data on symptom mitigation. Because of the municipal quarantine regulations instituted in India during the height of the pandemic, there was a strong disincentive to report symptoms that would trigger testing and quarantining, resulting in loss of livelihood and separation from family and community. The decision to make this change was not taken lightly and was done under fully blinded conditions and with the advice and consent of both our DSMB and the EC. Primary endpoint change in studies is not uncommon, with literature reporting around 30% of trials in COVID-19 pandemic situations suggesting such changes.^29-31

Also prompted by the absence of any reported illness among participants in the first 3 months of the trial was the introduction of repeat RT-PCR tests at each visit. That change resulted in a necessary division of the study participants at the time of analysis into 2 groups: those who had biweekly RT-PCRs and those who did not, limiting our statistical power.

Other limitations include insufficient power for robust exploratory analyses and low infection rates, which may have impacted the results. While it is possible that infections were missed in subjects who did not test every 15 days, these rates would be expected to be balanced between groups in proportion to the 2:1 randomization. The same could be said for antibody rates. Biweekly RT-PCR tests for all participants would have improved infection tracking throughout the study.

Uniform recruitment across locations was complicated by fluctuating regional infection rates. All cases were asymptomatic or mild, so the impact of BiosimCovex on symptom severity could not be determined. The low infection rates across the study population make it impossible to interpret the nosode’s effects, if any, on males and participants above the age of 45. We are also limited by the absence of a proven mechanism of action of nosodes, which constitutes a major gap and a target of future research. The antibody findings in this study alone do not confirm protective immunity without in vitro neutralizing antibodies and cellular immune response studies, which are in progress. (Manuscript in preparation.)

A longer follow-up would have been useful to determine the trajectory of waning protection and might have allowed for assessing the potential value of multiple courses of the nosode to maintain the protection.

The BiosimCovex group, which experienced symptom absence and a lower infection rate, suggests BiosimCovex’s potential role in curbing SARS-CoV-2 transmission, thereby reducing the risk of long COVID. Long-COVID is a pressing public health concern, affecting 15%-20% ³² of unvaccinated U.S. adults who contract SARS-CoV-2. It appears that the best chance to avoid Long COVID is to avoid becoming infected. Being female increases Long COVID risk by 50%, with the age group 18-30 experiencing the highest impact. ³³

The nosode may offer a convenient option: it is easy to administer, requires no medical personnel or cold storage, has minimal toxicity, features a simplified, rapid, and cost-effective production process, and may provide a novel way to stimulate the immune response. The potentization process has been shown to aggregate nanoparticles of the source materials, ² which may create a biologically detectable signal,^34,35 sufficient to trigger an immune response, including antibodies and reduced infection rate.

By the time the CDC granted EUAs for 3 vaccines in the US, ³⁶ more than 1.5 million deaths had occurred worldwide, ³⁷ suggesting a need for faster development and deployment of a prophylactic. Moreover, over 90% of the world’s population did not receive COVID-19 vaccines because of logistical, political, or cost reasons. In the early stages of such a pandemic, a non-invasive, low-risk preventive, with a 50% effectiveness rate in preventing the spread of infection could have major value. The use of nosodes could potentially provide a faster, cost-efficient, easily distributed protection approach against pandemics and epidemics, allowing even low and middle-income countries (LMIC) to get meaningful protection as vaccines take longer to develop and deploy. Additional research may enable next-generation prophylactic measures for other infections as well. Nosodes and vaccines should be studied for mutually complementary roles in controlling epidemics.

Given our findings that maximum protection occurred at day 45 and waned by day 60, it is necessary to explore booster doses and various potencies. A clear understanding of the biomechanisms underlying the observed efficacy is needed. We are exploring the nanoparticle nature of the potentized BiosimCovex and its capability of triggering an immune response in laboratory studies, and are currently collating those findings, which will be submitted for peer-reviewed publication.

Future research should be sufficiently powered for confirmation and subgroup exploratory analyses. Outcomes should include uniform, quantitative tests for neutralizing antibodies. Although we observed limited adverse events and the duration of effect seemed to peak at day 45, additional studies should include deeper exploration of these findings. Additional research is necessary to clarify BiosimCovex’s potency and immunogenic impact. More studies examining the immunogenic and protective response to BiosimCovex in previously vaccinated individuals should be considered. BiosimCovex was prepared from the Wuhan strain, yet it showed prophylactic benefit when Omicron was more dominant in circulation,^38,39 suggesting possible broad action against emerging viral variants and strains. Further studies will be needed to assess this and whether BiosimCovex plus vaccines might be a more effective prophylactic regimen than either one alone. Identification and characterization of the active agent(s) are necessary to improve and optimize the manufacturing and booster dosing, and to determine the precise biological effects resulting in BiosimCovex’s long-term prophylactic efficacy, including in pre-vaccinated populations.

Conclusions

In summary, we demonstrated the protective benefit of a potentized nosode against SARS-CoV-2 infection in a large, multi-centered, randomized, double-blind, placebo-controlled trial. The degree of protection in specific populations resembled that provided by the current vaccine ²⁸ but declined by day 60. This novel, easily produced, and scalable technology may offer an inexpensive approach for rapidly mitigating infections, epidemics, and pandemics for potentially any infection, opening new avenues for prophylaxis, complementing vaccines and drugs.

ORCID iDs

Rajesh Shah https://orcid.org/0000-0002-1263-9077

Gitanjali Talele https://orcid.org/0009-0002-2215-6973

Joan Walter https://orcid.org/0000-0002-5158-9893

John A. Ives https://orcid.org/0009-0008-6851-205X

Jessica Utts https://orcid.org/0000-0002-4248-3878

Wayne B. Jonas https://orcid.org/0000-0001-8498-3849

Data Availability Statement

All data generated or analysed during this study are included in this published article.