This two-arm study consisted of laboratory experiments to evaluate the antiviral activity and cytotoxicity of the anionic iron tetracarboxyphthalocyanine derivative (APD) (Golden Technology Corp., Brazil) and a triple-blind randomized controlled trial.
All the in vitro experiments were conducted in Biosafety Level (BSL) BSL-2 and BSL-3 facilities at the Institute of Biomedical Sciences, University of São Paulo, Brazil, according to the laboratory biosafety guidance recommended by the WHO for the novel coronavirus (SARS-CoV-2) .
- Antiviral and cytotoxic activity of APD
To determine the antiviral activity and cytotoxicity of APD, a 2.0 mg/mL (I) stock solution prepared in sterile distilled water was serially diluted by 2-fold (1.0 mg/mL to 0.39x10-2 mg/mL, i.e., 1/2 to 1/512) in Dulbecco’s modified essential medium (DMEM) supplemented with 2.5% fetal bovine serum (FBS) in a 96-well cell culture plate to a final volume of 100 mL per well. The dilutions were made in quadruplicate to determine both virus neutralization and cytotoxicity.
After dilution, 100 mL of SARS-CoV-2 (SARS.CoV2/SP02.2020.HIAE. Br) at 103 TCID50/mL (MOI=0,02) was added to the wells and incubated for 30 min at 37 °C. Then, 150 mL of the mixture (APD plus virus) was transferred to a 96-well cell culture plate previously seeded with 1x105 Vero CCL-81 cells/mL and grown to 80-90% confluence. The cells were then incubated at 37 °C in a 5% CO2 atmosphere for 72 h .
The plate was visually evaluated using an optical microscope to determine cell integrity and morphology, and then, samples were collected (in quadruplicate) for RNA extraction and real-time Polymerase Chain Reaction (RT-PCR) for the quantitative detection of the amount of active virus. The cells were fixed with Naphthol Blue Black (Sigma-Aldrich®).
- Nucleic acid extraction and real-time RT-qPCR for SARS-CoV-2 RNA detection
The extraction of total nucleic acids (RNA and DNA) was carried out using the semiautomated NucliSENS® easyMag® platform (BioMerieux, Lyon, France) following the manufacturer's instructions. The detection of viral RNA was carried out using the AgPath-ID One-Step RT-PCR Kit (Applied Biosystems Inc., EUA) on an ABI 7500 SDS real-time PCR machine (Applied Biosystems, Weiterstadt, Germany) using a published protocol and primers and probes specific for the E  RNA copies/mL were quantified by real-time RT-qPCR using a specific in vitro-transcribed RNA quantification standard kindly provided by Christian Drosten, Charité - Universitätsmedizin Berlin, Germany, as described previously .
The antiviral activity was expressed as the percent reduction in the active SARS-CoV-2 virus RNA/mL, calculated according to Eq. (1), after contact with the test specimen compared to the number of virus particles in the positive control.
Reduction (%) = [(B-A)/B] x 100 (Eq. 1)
where A and B are the numbers of RNA copies/mL recovered from the supernatant of APD-treated and APD-untreated cells, respectively.
- Indirect Immunofluorescence (IIF)
The methodology described here was adapted from Sales-Medina et al., 2020 . Briefly, at 72 h.p.i., the plates were fixed for 30 min in 4% paraformaldehyde in 1X PBS (pH 7.4) and subjected to indirect immunofluorescence detection of viral cellular infection. After washing twice with 1X PBS 0.05% Tween 20 (PBST), the plates were blocked with bovine serum albumin (BSA) (3% w/v in 1X PBS; Sigma-Aldrich) for 30 min at room temperature and washed twice with PBST. Convalescent serum from a Brazilian patient with COVID-19 diluted 1:1000 in PBS was used as a primary antibody to detect SARS-CoV-2 in Vero cells. The primary antibodies were incubated for 30 min, and the plates were washed twice with PBST. Subsequently, goat anti-human IgG labeled with Alexa 488 (Thermo Scientific) diluted to 4 μg/mL in PBS was used as the secondary antibody, and the cells were incubated for 30 min with 5 μg/mL 4′,6-diamidine-2′-phenylindole dihydrochloride (DAPI, Sigma-Aldrich) in PBS to stain the nucleic DNA. The plates were washed twice with PBST and imaged in an Operetta High Content Imaging System (Perkin Elmer) using a 20x magnification objective. Five images were acquired per well and analyzed using Harmony software (Perkin Elmer), version 3.5.2. Image analysis consisted of identifying and counting the Vero E6 cells based on the nuclear segmentation, viral infection, and cytoplasmic staining detected by the immunofluorescence assay .
Clinical Trial Design and Oversight
This triple-blind randomized controlled trial was conducted in accordance with the principles of the Declaration of Helsinki and ethical standards of human experimentation with the approval of the Human Research Ethics Committee of Bauru School of Dentistry of the University of Sao Paulo, Brazil (CAAE 34070620.6.0000.5417). This clinical study was also registered at REBEC - Brazilian Clinical Trial Register (RBR-58ftdj). The study complied with the Consort 2010 checklist of information to include when reporting a randomized trial. This study was carried out as a controlled trial from 10th August to 4th November 2020 at Bauru State Hospital, Bauru, with hospitalized patients who tested positive for SARS-CoV-2. All the participants received the World Health Organization standard care hospital treatment (world medical protocol - antibiotics, corticoids and anticoagulants) plus one of the two mouthwash interventions (active and nonactive mouthwashes). Based on previous studies, APD antimicrobial compound-containing mouthwash was chosen as the active mouthwash (AM) [14,15, 26] for comparison with a nonactive mouthwash (NAM) negative control. Both mouthwashes were produced with exactly the same formula (color, flavor, other ingredients) except for the presence or absence of the active ingredient. Once the mouthwash intervention was given to patients receiving medical treatment for COVID-19, the use of a negative control mouthwash was necessary. The active compound concentration used was between the 1:8 and 1:16 titer according to the in vitro arm of the study. All the oral care materials used during the study (mouthwashes, toothbrushes, toothpaste and dental floss) were produced by Rabbit Corp, Brazil.
Patient Inclusion and Exclusion Criteria
The inclusion criteria were patients who were 18 years old or older, who were admitted fewer than 7 days from the onset of severe acute respiratory syndrome (SARS), who were suspected of being positive for SARS-CoV-2 and who were admitted to the hospital with a mild or moderate clinical condition  with no need for intensive care unit (ICU). To be enrolled in the study, each participant read and signed the informed consent form after understanding the risks and objectives of the study. The exclusion criteria included patients who had contraindications to using mouthwash due to medical reasons or the inability to gargle and spit.
Randomization and masking
Sample randomization was performed as follows: the mouthwash bottles (AM and NAM) were placed in a closed package with consecutive numbers according to the patients selected. An EXCEL® database was created from these numbered packages and used for randomization. After randomization, packages with mouthwash bottles and oral care kits were delivered to the hospitalized patients. This study was considered triple-blind because the patients, the examiner and the statistician were blinded to the treatment groups.
The eligible participants were randomly assigned to one of the following groups: AM or NAM. The patients were instructed to use 5 mL of the mouthwash and to switch between gargling/rinsing for 1 min up to 5 times a day: upon awakening, after breakfast, after lunch, after dinner and before bedtime [14,15]. Each patient followed this adjunctive therapeutic protocol until the outcome associated with COVID-19 medicine treatment was recorded.
The primary outcome was the time to clinical improvement, defined as the length of stay in the hospital (patients did not need oxygen therapy support for more than 24 h and no longer showed any symptoms). Secondary outcomes were clinical evolution, need for care in the ICU and death. The criteria for the transfer of a patient to the ICU was the presence of respiratory effort requiring the use of O2 above 8.0 L/min, which is the initial graduation of nonremovable masks. The mouthwash protocol was also verified regarding the conditions of use and side effects [14,15].
Statistical analysis was conducted using R (R Core Team) and SAS® software version 3.8 (SAS Institute Inc). Descriptive and exploratory data analyses were performed. A Mann-Whitney U nonparametric test was used to compare the groups regarding age, number of comorbidities, and duration of symptoms prior to hospitalization. The frequency of admission to intensive care and deaths were compared between groups using Fisher’s exact test. The time to clinical improvement is presented by a Kaplan–Meier plot and was compared using a Cox regression model, with an estimate of the hazard ratio association measure and 95% confidence interval. Since a significant difference was observed between groups regarding the median age, the survival analysis was adjusted for the patient's age.
The sample of 41 patients included in this investigation provided a test power of 0.80 for a minimum detectable hazard ratio of 2.5, with α=0.05, considering the follow-up time of 22 days and median hospitalization time in the control group of 7 days.