key: cord-0691053-bfu0jguu
authors: Kinoshita, T.; Shinoda, M.; Nisizaki, Y.; Shiraki, K.; Hirai, Y.; Kichikawa, Y.; Tsushima, K.; Sinkai, M.; Komura, N.; Yoshida, K.; Kido, Y.; Kakeya, H.; Uemura, N.; Kadota, J.
title: Phase 3, multicentre, double-blind, randomised, parallel-group, placebo-controlled study of camostat mesilate (FOY-305) for the treatment of COVID-19 (CANDLE study)
date: 2022-04-02
journal: nan
DOI: 10.1101/2022.03.27.22271988
sha: c61433cdba60e967f339be9b52c690ae09acfbf1
doc_id: 691053
cord_uid: bfu0jguu

Background: In vitro drug-screening studies have indicated that camostat mesilate (FOY-305) may prevent SARS-CoV-2 infection into human airway epithelial cells. This study was conducted to investigate whether camostat mesilate is an effective treatment for SARS-CoV-2 infection (COVID-19). Methods: This was a phase 3, multicentre, double-blind, randomised, parallel-group, placebo-controlled study. Patients were enrolled if they were admitted to a hospital within 5 days of onset of COVID-19 symptoms or within 5 days of a positive test for asymptomatic patients. Severe cases (e.g., those requiring oxygenation/ventilation) were excluded. Patients were administered camostat mesilate (600 mg qid; four to eight times higher than the clinical doses in Japan) or placebo for up to 14 days. The primary efficacy endpoint was the time to the first two consecutive negative tests for SARS-CoV-2. Findings: One-hundred and fifty-five patients were randomised to receive camostat mesilate (n=78) or placebo (n=77). The median time to the first test was 11 days in both groups, and conversion to negative status was observed in 60.8% and 63.5% of patients in the camostat mesilate and placebo groups, respectively. The primary (Bayesian) and secondary (frequentist) analyses found no significant differences in the primary endpoint between the two groups. No additional safety concerns beyond those already known for camostat mesilate were identified. Interpretation: Camostat mesilate is no more effective, based on upper airway viral clearance, than placebo for treating patients with mild to moderate SARS-CoV-2 infection with or without symptoms. Funding: Ono Pharmaceutical Co., Ltd.

4 were no differences between the study groups in terms of other efficacy endpoints. This study used a dose that was four to eight times higher than the clinical doses of camostat mesilate used in Japan for the acute symptoms of chronic pancreatitis and postoperative reflux oesophagitis. The study identified no additional safety concerns beyond those already known for camostat mesilate.

After starting this study, another randomised, placebo-controlled study reported the efficacy and safety of camostat mesilate for the treatment of patients with COVID-19, albeit at a lower dose of 200 mg three times daily. That study also found no difference between camostat mesilate and placebo for the primary endpoint (the time to discharge or a clinical improvement in clinical severity of at least two points on a seven-point ordinal scale). Along with this evidence, our study did not support the use of camostat mesilate as a treatment option for COVID-19. However, since the administration of camostat mesilate was started after the onset of symptoms and presumably the peak viral load, we cannot exclude the possibility that camostat mesilate may be effective if administration is started earlier in the course of infection, or perhaps as prophylactic use in close contacts.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 2, 2022. ; https://doi.org/10.1101/2022.03.27.22271988 doi: medRxiv preprint INTRODUCTION SARS-CoV-2 is a highly transmissible virus that causes a potentially severe infection , which is continuing to spread worldwide and thus represents a significant global health threat. 1, 2 The symptoms and severity of COVID-19 vary considerably. Some patients develop advanced disease within about 10 days of onset, with life-threatening symptoms, including severe inflammatory reactions, dyspnoea, and severe acute pneumonia. 3, 4 Another challenge is the emergence of novel variants displaying altered transmissibility, infectiveness, disease severity, and mortality risk.

Currently, severe cases are generally treated with remdesivir, dexamethasone and symptomatic therapies.

However, appropriate treatments have not been established for asymptomatic patients or patients with moderate symptoms who do not require oxygen therapy. Convenient oral drugs that can be prescribed to patients recuperating at home or other outpatient settings are also needed.

As one potential therapeutic target, it was discovered that the spike protein (S protein) of SARS-CoV-2 binds to angiotensin converting enzyme II (ACE2) on the host cell membrane as a functional receptor. 5, 6 The S protein is then cleaved into S1 and S2 by host-derived protease activity. The S1 fragment binds to ACE2 and the S2 fragment is cleaved by a type II transmembrane serine protease (TMPRSS2) expressed on the host cell membrane. These steps promote fusion of the viral envelope (outer membrane) with the cell membrane. Therefore, ACE2 and TMPRSS2, which are expressed on airway epithelial cells, are key factors in SARS-CoV-2 infection. In vitro drug-screening studies have indicated that 4-(4guanidinobenzoyloxy)phenylacetic acid (GBPA), the active metabolite of the serine protease inhibitor camostat mesilate (FOY-305), inhibits TMPRSS2 and prevents SARS-CoV-2 infection of a human airway epithelial cell-derived cell line (Calu-3 cells). [6] [7] [8] [9] [10] [11] [12] [13] Repurposing drugs that have already been approved for other indications may help facilitate the drug development process and shorten the development time. 14, 15 In Japan, camostat mesilate is an oral drug . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 2, 2022. ; https://doi.org/10.1101/2022.03.27.22271988 doi: medRxiv preprint 7 that has been used to treat the acute symptoms of chronic pancreatitis and postoperative reflux oesophagitis for more than 30 years, and has shown a good safety profile over this period of time. 16 Based on the preclinical evidence, it has been postulated that camostat mesilate may also be useful for treating . In support of this hypothesis, one retrospective study of critically ill COVID-19 patients with organ failure treated in an intensive care unit revealed a decline in disease activity within 8 days of admission among patients treated with camostat mesilate but not in patients treated with hydroxychloroquine. 17 This clinical improvement was accompanied by a decline in inflammatory markers, such as C-reactive protein and interleukin-6, and increased oxygenation.

This double-blind Phase 3 study was conducted to evaluate the efficacy and safety of camostat mesilate and hence explore the role of TMPRSS2 as a potential treatment target for mild to moderate SARS-CoV-2 infection with or without symptoms.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 2, 2022. 

This study adhered to the Declaration of Helsinki, Good Clinical Practice, and relevant local/international guidelines. The protocol and patient consent forms were approved by the ethics committees or institutional review boards at all participating institutions (appendix p 1).

Patients aged at least 18 years were eligible for this study if they were admitted to the participating hospitals within 5 days of onset of SARS-CoV-2 symptoms or within 5 days of a positive test for asymptomatic patients. SARS-CoV-2 infection must have been tested using a standard method at the time the study was conducted (e.g., reverse transcriptase-polymerase chain reaction [RT-PCR] test, loopmediated isothermal amplification [LAMP] test, or antigen test). Only patients with asymptomatic/mild or moderate infection were eligible. Patients with severe infection, such as those requiring oxygenation, ventilation, or admission to an intensive care unit were excluded. Major exclusion criteria were prior history of SARS-CoV-2 infection, history of vaccination for SARS-CoV-2, and history of treatment with camostat mesilate or nafamostat mesilate. Further eligibility criteria are described in the study protocol (appendix). All patients provided informed consent.

The study comprised a double-blind phase (up to 14 days) in which they were randomised to receive camostat mesilate or placebo, and a 2-week follow-up period after the last dose of the study drug.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 2, 2022. ; https://doi.org/10.1101/2022.03.27.22271988 doi: medRxiv preprint 9 Randomisation was performed using a minimisation method with the following randomisation factors: medical institution, age (at least 65 vs less than 65 years), and absence/presence of underlying diseases (chronic respiratory disease, chronic kidney disease, diabetes mellitus, hypertension, cardiovascular diseases, and obesity [body mass index, BMI, at least 30 kg/m 2 ]). Patients were enrolled, randomised, and allocated to the appropriate treatments by the investigators/subinvestigators using an interactive web response system, which was managed by the sponsor.

The length of the double-blind period (up to 14 days) was chosen because of the typical clinical course. 18 There were no changes to the study that were implemented after commencing enrolment.

Eligible patients were allocated to either camostat mesilate or placebo film-coated tablets, which were visually indistinguishable in appearance and packaging, to be administered at a dose of 600 mg four times daily (qid; before breakfast, before lunch, before dinner, and bedtime) for up to 14 days. The administration status was confirmed by clinical study staff, such as the principal investigator. The dose of camostat mesilate was chosen based on (1) preclinical EC50 values, which determined the clinical target exposures; (2) modelling and simulation to predict high dose exposure in the clinic; and (3) a results of a Phase 1 study of the safety and pharmacokinetics of camostat mesilate at 600 mg qid in healthy Japanese volunteers. 19 During the study, it was prohibited to administer drugs with antiviral effects (e.g., remdesivir, favipiravir, ciclesonide, nafamostat mesilate, hydroxychloroquine, ivermectin, combination drug of lopinavir and ritonavir, povidone-iodine) and drugs with anticytokine effects (e.g., tocilizumab, Janus kinase inhibitors) from the day of onset of symptoms until completion of the study. However, these drugs could be continued at the same dose in patients already using these to treat a pre-existing comorbidity. The use of . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 2, 2022. ; https://doi.org/10.1101/2022.03.27.22271988 doi: medRxiv preprint other unapproved drugs or camostat mesilate as a commercial product was prohibited.

In the randomised period, the allocated treatment was to be discontinued in accordance with the study criteria listed in appendix p 4 (table S1), which included patient request, emergence of an adverse event that made it difficult to continue the study, negative test for SARS-CoV-2 on two consecutive occasions, and increasing disease severity (exacerbation of pneumonia and SpO2 of 93% or less despite oxygen therapy). Efficacy evaluations were not conducted after confirmation of SARS-CoV-2 negativity.

Treatments beyond day 14 were at the attending physician's discretion or institutional policies and were not recorded.

The primary efficacy endpoint was the time to the first two consecutive negative SARS-CoV-2 tests . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. 

The objective of this study was to establish superiority of camostat mesilate over placebo in patients with COVID-19 using the time to negative SARS-CoV-2 test as the primary endpoint.

The criteria for effectiveness were a Bayesian posterior probability of at least 92% with a hazard ratio exceeding 1.0. The criteria for ineffectiveness were a Bayesian posterior probability of 8% or less with a hazard ratio exceeding 1.0.

The time to a negative SARS-CoV-2 test was assumed to follow an exponential distribution, with a median time of 14 days in the placebo group and a median time of 7 to 8 days in the camostat mesilate group. The probabilities of meeting the assessment of effectiveness or ineffectiveness with 50 patients per group (100 in total) were calculated at various analysis time-points by applying numerical simulations in SAS version 9.4. Because the timing of the interim analysis was dependent on the status of the COVID-19 outbreak, numerical simulations were performed, assuming the time-points when 40 to 80 subjects . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 2, 2022. ; https://doi.org/10.1101/2022.03.27.22271988 doi: medRxiv preprint would have been randomised. The prior distribution of the regression coefficient was assumed to be uniform. Based on the numerical experiment, irrespective of which time-point the interim analysis was performed, the probability that treatment would be effective was 75% to 90% if the median duration was 7 or 8 days in the camostat mesilate group. Moreover, if the median duration was 14 days, the probability that treatment was effective was controlled within 10%. (appendix p 2 [sample size calculation]). From these data, we therefore considered it was possible to demonstrate superiority of camostat mesilate over placebo with a sample size of 100 patients (50 patients per group).

A modified intention-to-treat analysis set was used for efficacy analyses by excluding any patients who tested negative for SARS-CoV-2 on day 1 (local laboratory tests). All analyses were performed on an asrandomised basis. The safety analysis set comprised all patients who received at least one dose of the allocated drug.

Baseline characteristics were analysed using descriptive statistics, including numbers (proportions) of patients and summary statistics, as appropriate.

For the primary endpoint, the time to SARS-CoV-2 negativity (in days) was calculated as the date of the first (of two consecutive) SARS-CoV-2 negativity test minus the date of randomisation plus one. The events and reasons for censoring patients in this analysis are defined in appendix p 6 (table S3). Cox proportional hazards model stratified by the randomisation factors (age and underlying disease) was used to determine the posterior mean hazard ratio with two-sided 95% credible intervals for the camostat mesilate group relative to the placebo group. The distribution of the regression coefficients was assumed to be uniform. In a secondary analysis, we applied the log-rank test stratified by age and underlying disease and plotted Kaplan-Meier curves for both groups to calculate the median time to SARS-CoV-2 negativity and 95% confidence intervals were calculated using the Brookmeyer-Crowley method with double log transformation. As a sensitivity analysis, we investigated the influence of patients who tested . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 2, 2022. ; https://doi.org/10.1101/2022.03.27.22271988 doi: medRxiv preprint 13 positive for SARS-CoV-2 after conversion to a SARS-CoV-2 negative status. The events and reasons for censoring patients in this analysis are defined in appendix p 7 (table S4).

The proportions of patients negative for SARS-CoV-2, and the distribution of disease severity were compared between the two groups using the Mantel-Haenszel test stratified by the randomisation factors. The actual values for the ordinal scale of severity were compared between the treatment groups using the proportional odds model, which included treatment group and randomisation factors as factors.

The median time to the resolution of clinical symptoms was estimated using the Kaplan-Meier method and 95% confidence intervals were calculated using the Brookmeyer-Crowley method with double log transformation. Changes in viral load, antibody responses (IgG and IgM), and safety outcomes were analysed descriptively in terms of the number and percentage of patients or summary statistics, as appropriate.

All tests were two-sided with a significance level of 5%. Because the primary analysis of the primary efficacy endpoint was based on Bayesian interim monitoring, a significance level was not applied. No adjustment for multiplicity between other endpoints or time-points was made.

Some changes in the statistical analyses were implemented before unblinding of the data. These additional analyses were performed to further evaluate efficacy, and are summarized in the appendix p 8 (table S5) .

SAS version 9.4 (SAS Institute, Cary, NC, USA) was used for all statistical analyses.

The study was registered on ClinicalTrials.gov (NCT04657497) and the Japan Registry for Clinical Trials (jRCT2031200198).

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 2, 2022. ; https://doi.org/10.1101/2022.03.27.22271988 doi: medRxiv preprint

The study was funded by Ono Pharmaceutical Co., Ltd. The sponsor provided funding for the study and publication of the manuscript. Employees of the sponsor were involved in study design; collection, analysis, and interpretation of the data; and reviewed the manuscript. The corresponding author had full access to all the data and had final responsibility for the decision to submit for publication.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Among 155 patients, 78 (50·3%) were male and 77 (49·7%) were female, 59 (38·1%) were at least 65 years old, and 71 (45·8%) had at least one underlying disease, the most common being hypertension in 44 patients (28·4%) (table 1). The median interval between the onset of symptoms (date of positive test for asymptomatic patients) to date of registration was 4 days (range 0-5 days). RT-PCR was the predominant testing methods, being used for 142 patients (91·6%). Nasopharyngeal swabs were used in 108 patients (69·7%), nasal swabs in 16 patents (10·3%), and saliva samples in 30 patients (19·4%). The median viral load was 6·91 log10 copies/mL (range 3·40-9·40 log10 copies/mL). All of the patients were hospitalised without requiring oxygen therapy (i.e. ordinal scale of 3). One hundred and eight patients had symptoms at registration. Both groups were very similar, demonstrating the robustness of the randomisation scheme.

During the treatment period, 134 of 155 patients discontinued treatment or dropped out from the study: 68 of 78 (87·2%) in the camostat mesilate group and 66 of 77 (85·7%) in the placebo group (appendix p 9

[table S6]). The most frequent reason for discontinuation of treatment was two consecutive negative SARS-CoV-2 tests in accordance with the study protocol in the camostat mesilate (45 patients, 57·7%)

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 2, 2022. 

The median time to the first two consecutive SARS-CoV-2 negative tests (local laboratory) was 11 days in both groups ( 

The viral load was monitored daily in all patients. As illustrated in figure 3 , the changes in viral load over time were comparable in both groups and there were no apparent differences at any time-point.

The distribution of the ordinal scale of severity was comparable in both groups, with no clear differences at any time ( figure 4) . The ordinal scale was grade 3 (hospitalised, no oxygen therapy) in most patients during the study period because all patients were hospitalised for SARS-CoV-2 testing; outpatients were not enrolled due to the risk of transmission. Nevertheless, none of the patients in either group required intubation/mechanical ventilation or ventilation plus additional organ support, and there were no deaths.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 2, 2022. ; https://doi.org/10.1101/2022.03.27.22271988 doi: medRxiv preprint

The most severe case was a patient in the placebo group whose severity was classified as grade 5 (requiring non-invasive ventilation or high-flow oxygen therapy) on day 9. Other than the patient classified as grade 5 on day 9, none of the other patients in either group experienced a worsening in the ordinal scale by at least two categories at any time during the study. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 2, 2022. ; https://doi.org/10.1101/2022.03.27.22271988 doi: medRxiv preprint DISCUSSION SARS-CoV-2 remains a clinically significant global health crisis and there remains an urgent, ongoing need to identify effective treatments. A number of preclinical studies have been conducted in the search for therapies for COVID-19. It was discovered that GBPA, the active metabolite of camostat mesilate, inhibits TMPRSS2 and prevents SARS-CoV-2 infection of human airway cells. [6] [7] [8] [9] [10] [11] [12] [13] Furthermore, a smallscale retrospective study suggested a potential therapeutic effect of camostat mesilate in patients admitted to an intensive care unit. 21 Despite promising results of preclinical studies, as well as a small retrospective study, the results of our study indicate that camostat mesilate is no more effective than placebo for treating patients with mild to moderate SARS-CoV-2 infection with or without symptoms. The lack of antiviral effect was demonstrated based on the median time to the first two consecutive SARS-CoV-2 negative tests (the primary endpoint) and the other clinical outcomes, with no statistically significant or clinically relevant differences between the two groups. Although this study investigated a high dose of camostat mesilate (600 mg qid; four to eight times higher than the clinical doses in Japan), no new safety concerns were identified.

Another large-scale study conducted in Denmark and Sweden also reported no benefit of administering camostat mesilate at a dose of 200 mg three times daily (lower than the dose used in our study, 600 mg qid) or placebo for 5 days in hospitalized patients. 22 The primary endpoint in that study was a composite of the time to discharge or a clinical improvement in clinical severity of at least two points on a sevenpoint ordinal scale. The median time to clinical improvement was 5 days in both groups.

From the data available to date, it is unclear why the findings of preclinical studies did not translate into a clinical effect. However, several factors related to the design of the study and the mechanism of action of camostat mesilate should be considered:

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 2, 2022. ; https://doi.org/10.1101/2022.03.27.22271988 doi: medRxiv preprint Viral entry pathway-Although TMPRSS2 is one of the primary routes of viral entry, the viral particles might exploit other pathways, such as endocytosis, to compensate for reduced entry via TMPRSS2. Thus, effective treatments may require inhibition of multiple viral entry pathways and a combination of drugs with various mechanisms of action, including camostat mesilate, may be useful for treating Inappropriate timing of administration-The peak viral load is typically reached within 2-3 days after the onset of symptoms and the administration of camostat mesilate was started approximately 3 days after the onset of symptoms in this study. The selected timing of administration of camostat mesilate might not have been best optimised to suppress viral activity. It has been suggested that therapies aimed at blocking infection or viral reproduction may be more effective if they were initiated before the peak viral load. 24 Therefore, some efficacy may be observed if administration is started as early as possible in the course of infection, perhaps as prophylactic administration to close contacts of patients, such as household members, or immediately after a positive test result. In fact, antibody drugs such as casirivimab/imdeviimab have been shown to reduce the risk of onset in uninfected patients and to prevent aggravation. 25 Dosing-Prior to this study, we conducted a Phase 1 study to set the dosage and treatment regimen in this study. An important PK feature of camostat mesilate is that when orally administered, it is rapidly metabolized into an active metabolite (GBPA) by esterases. [26] [27] [28] Those studies demonstrated that camostat mesilate is not detectable in the human plasma, and that GBPA is rapidly eliminated with a half-life of less than 2 h. Therefore, frequent dosing is required to maintain target plasma concentrations. Considering the adherence of the target patient population, we assumed that qid administration of camostat mesilate (morning, midday, evening, and before bedtime) would be acceptable. Specifically, in PK/PD simulations, the times above the EC 50 of camostat mesilate at doses of 800 mg three times daily and 600 mg qid were 9·8 h and 11·5 h, respectively. 19 The results of the PK/PD simulations also suggested an advantage of increasing the dosing frequency rather than increasing the dose per administration. Multiple administrations of camostat mesilate at 600 mg qid were well tolerated in a Phase 1 study. However, in a repeated-dose toxicity study in dogs, camostat 300 mg/kg decreased body weight and food intake, . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 2, 2022. ; https://doi.org/10.1101/2022.03.27.22271988 doi: medRxiv preprint 20 induced vomiting and effects on the gastrointestinal tract, including gastrointestinal injury, and caused death, with a no-observed adverse effect level (NOAEL) of 100 mg/kg. 19, 29 Converting the NOAEL in dogs to humans yielded an equivalent dose of 3333 mg. 19, 30 Thus, the dose used in this study had a safety margin of 1·4-fold. Based on the overall balance between the expected time above EC50 and safety risks, 600 mg qid was determined to be an appropriate dose for this study. The plasma concentration of GBPA was predicted to exceed the EC50 for at least 11·5 h at the dose used (2400 mg/day). 19 However, the targeted time above EC 50 might have been insufficient to inhibit TMPRSS2 and hence prevent viral entry, although the exact relationship between the exposure and antiviral activity is not clear in the clinic.

A limitation of this study is that the improvement of the ordinal scale of severity could not be evaluated correctly because most patients were hospitalised for daily viral testing regardless of the presence or absence of symptoms and were hence classified as grade 3. Another possible limitation is that the effects of camostat mesilate against SARS-CoV-2 were evaluated using nasopharyngeal and nasal swab samples in the majority of patients. However, the appropriateness of an index of upper airway viral load in asymptomatic to moderate cases remains questionable. It is considered that the epidemic strain at the time was a D614G strain, but no data on the type of strain were collected for this study. Efficacy against currently circulating variants is unknown.

There are some strengths of this study that should be mentioned. In particular, this was a double-blind, randomised, placebo-controlled study with robust randomisation as demonstrated by the high similarity of both groups. In addition, this study used a dose that was four to eight times higher than the clinical doses in Japan used for the acute symptoms of chronic pancreatitis and postoperative reflux oesophagitis based on the preclinical and early clinical evidence. Furthermore, the efficacy of camostat mesilate was assessed using multiple clinically relevant endpoints, including local and central laboratory tests for SARS-CoV-2 infection and viral load.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 2, 2022. ; https://doi.org/10.1101/2022.03.27.22271988 doi: medRxiv preprint Although the study results were negative, there were several lessons and the study generated important new evidence. There are still some questions related to the development of clinical trials for emerging infectious diseases, including study design and patient segmentation. Even in a state of emergency, the doses in clinical trials should be carefully selected with consideration of clinical pharmacology, including PK/PD modelling and simulation, when planning clinical trials for a new drug candidate in settings such as this, in order to provide clear evidence supporting or halting ongoing development of the drug.

Furthermore, collaboration between government, industry and academia is essential for the development of therapeutic agents in a pandemic.

In conclusion, the results of this study found clear evidence for not using camostat mesilate to treat mild to moderate SARS-CoV-2 infection with or without symptoms. Of note, no new safety concerns were identified at the high dose used in this study, which exceeds the standard dose used in other indications.

Overall, these findings highlight the continuing need to identify and develop alternative therapies for COVID-19, and the necessity of conducting well-designed studies to confirm whether preclinical findings translate into meaningful clinical efficacy.

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 2, 2022. ; https://doi.org/10.1101/2022.03.27.22271988 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 2, 2022. Junichi Kadota reports consultancy fees from Ono Pharmaceutical Co., Ltd. in relation to this work; and consultancy fees from FUJIFILM Toyama Chemical Co., Ltd., KOBAYASHI Pharma Co., Ltd., and Kyorin Pharma Co., Ltd. unrelated to this work; institutional research funding from MSD Co., Ltd., . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 2, 2022. ; https://doi.org/10.1101/2022.03.27.22271988 doi: medRxiv preprint Taisho Pharma Co., Ltd., Nippon Boehringer Ingelheim Co., Ltd., Daiichi Sankyo Co., Ltd., Pfizer Japan Inc., Kyorin Pharma Co., Ltd., Astellas Pharma Inc., Chugai Pharmaceutical Co., Ltd., Shionogi & Co., Ltd., and Teijin Pharma Ltd. unrelated to this work; and lecture fees from Ono Pharmaceutical Co., Ltd., MSD Co., Ltd., AstraZeneca K.K., Nippon Boehringer Ingelheim Co., Ltd., Pfizer Japan Inc., Shionogi & Co., Ltd., Taisho Toyama Pharma Co., Ltd., Meiji Seika Pharma Co., Ltd., Sanofi K.K., Kyorin Pharma Co., Ltd., Astellas Pharma Inc., Sumitomo Dainippon Pharma Co., Ltd., Bristol-Myers Squibb Company, Daiichi Sankyo Co., Ltd., Chugai Pharmaceutical Co., Ltd., Novartis Pharma K.K., Taisho Pharma Co., Ltd., FUJIFILM Medical Co., Ltd., GlaxoSmithKline K.K., and DENKA SEIKEN Co., Ltd. unrelated to this work.

Conceptualisation: Naoyuki Komura, Junichi Kadota . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 2, 2022. ; https://doi.org/10.1101/2022.03.27.22271988 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 2, 2022. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 2, 2022. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 2, 2022. Values are n (%)

. CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. Did not receive study drug . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 2, 2022. ; https://doi.org/10.1101/2022.03.27.22271988 doi: medRxiv preprint Camostat mesilate Placebo . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 2, 2022. . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 2, 2022. ; https://doi.org/10.1101/2022.03.27.22271988 doi: medRxiv preprint . CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review)

The copyright holder for this preprint this version posted April 2, 2022. ; https://doi.org/10.1101/2022.03.27.22271988 doi: medRxiv preprint

Government responses and COVID-19 deaths: global evidence across multiple pandemic waves

Controlling the pandemic during the SARS-CoV-2 vaccination rollout

Is severe COVID-19 a cytokine storm syndrome: a hyperinflammatory debate

Distinctive features of severe SARS-CoV-2 pneumonia

Camostat mesylate against SARS-CoV-2 and COVID-19-rationale, dosing and safety

Targeting androgen regulation of TMPRSS2 and ACE2 as a therapeutic strategy to combat COVID-19

Camostat mesylate inhibits SARS-CoV-2 activation by TMPRSS2-related proteases and its metabolite GBPA exerts antiviral activity

SARS-CoV-2 cell entry depends on ACE2 and TMPRSS2 and is blocked by a clinically proven protease inhibitor

Structural modeling and analysis of the SARS-CoV-2 cell entry inhibitor camostat bound to the trypsin-like protease TMPRSS2

Druggable targets of SARS-CoV-2 and treatment opportunities for COVID-19

Prevention of SARS-CoV-2 cell entry: insight from in silico interaction of drug-like alkaloids with spike glycoprotein, human ACE2, and TMPRSS2

cell-cell fusion differ in their requirements for receptor expression and proteolytic activation

Molecular insights into human transmembrane protease serine-2 (TMPS2) inhibitors against SARS-CoV2: homology modelling, molecular dynamics, and docking studies

Repurposing existing drugs for COVID-19: an endocrinology perspective

Repurposing therapeutics for potential treatment of SARS-CoV-2: a review

Camostat mesilate therapy for COVID-19

Camostat mesylate may reduce severity of coronavirus disease 2019 sepsis: a first observation

Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study

A phase I study of high dose camostat mesylate in healthy adults provides a rationale to repurpose the TMPRSS2 inhibitor for the treatment of COVID-19

WHO R&D Blueprint: Novel Coronavirus -COVID-19 Therapeutic Trial Synopsis

Camostat mesylate therapy in critically ill patients with COVID-19 pneumonia

Efficacy of the TMPRSS2 inhibitor camostat mesilate in patients hospitalized with Covid-19-a double-blind randomized controlled trial

A quantitative model used to compare within-host SARS

CoV-2, MERS-CoV, and SARS-CoV dynamics provides insights into the pathogenesis and treatment of SARS-CoV-2

Subcutaneous REGEN-COV antibody combination to prevent Covid-19

Elimination of the low-molecular weight proteinase inhibitor camostate (FOY 305) and its degradation products by the rat liver

Metabolic fate of 14C-camostat mesylate in man, rat and dog after intravenous administration

Absorption and excretion of camostat (FOY-305) orally administered to male rabbit and healthy subject

Toxicity study of FOY-305(I): Acute toxicity study in mice and rats, subacute toxicity study in dogs

Guidance for Industry: Estimating the maximum safe starting dose in initial clinical trials for therapeutics in adult healthy volunteers

This study was funded by Ono Pharmaceutical Co., Ltd. The authors express their gratitude to all of the patients, investigators, and healthcare professionals involved in this study, and Yu Nakagama MD, PhD (Specially Appointed Senior Lecturer, Department of Parasitology, Graduate School of Medicine, Osaka City University) for research assistance and advice. The authors also thank Nicholas D. Smith (EMC K.K.) for medical writing support, which was funded by Ono Pharmaceutical Co., Ltd.

Taku Kinoshita, Masahiro Shinoda, Katsuya Shiraki, Yuji Hirai, and Kenji Tsushima report institution research funding (in relation to this work) from Ono Pharmaceutical Co., Ltd. in relation to this work.Yasuhiro Nishizaki reports institution research funding from Ono Pharmaceutical Co., Ltd. in relation to this work; and institution research funding from ITO EN Co., Ltd., AstaReal Co., Ltd., Mizkan Holdings Co., Ltd., and Kanagawa Institute of Industrial Science and Technology unrelated to this work.Yoshiko Kichikawa reports institution research funding from Ono Pharmaceutical Co., Ltd. in relation to this work; and institution research funding from Nobelpharma Co., Ltd., Chugai Pharmaceutical Co., Ltd., Japan Tobacco Inc., and Pfizer Japan Inc. unrelated to this work.Masaharu Sinkai reports institution research funding from Ono Pharmaceutical Co., Ltd. in relation to this work; and institutional research funding from FUJIFILM Toyama Chemical Co., Ltd., AstraZeneca K.K, Chugai Pharmaceutical Co., Ltd., Pfizer Japan Inc., and Genova Inc. unrelated this work.Naoyuki Komura and Kazuo Yoshida are employees of Ono Pharmaceutical Co., Ltd.

CC-BY 4.0 International license It is made available under a is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity.

The copyright holder for this preprint this version posted April 2, 2022. ; https://doi.org/10.1101/2022.03.27.22271988 doi: medRxiv preprint