key: cord-0262396-zspu2y3t
authors: Zubkova, T. G.; te Velthuis, A. J. W.; Shaw, M.; Mehle, A.; Boltz, D.; Gmeinwieser, N.; Stammer, H.; Milde, J.; Mueller, L.; Margitich, V.
title: Enisamium reduces influenza virus shedding and improves patient recovery by inhibiting viral RNA polymerase activity
date: 2020-12-30
journal: nan
DOI: 10.1101/2020.12.23.20247569
sha: 497382e7e54746fd0662a11334c6cd8a9801fdb7
doc_id: 262396
cord_uid: zspu2y3t

Infections with respiratory viruses constitute a huge burden on our health and economy. Antivirals against some respiratory viruses are available, but further options are urgently needed. Enisamium iodide (laboratory code FAV00A, trade name Amizon(R)) is an antiviral marketed in countries of the Commonwealth of Independent States for the treatment of viral respiratory infections, but its clinical efficacy and mode of action are not well understood. Here, we investigated the efficacy of FAV00A in patients aged between 18-60 years with confirmed influenza and other viral respiratory infections. FAV00A treatment resulted in reduced influenza virus shedding (at day 3, 71.2% in FAV00A group tested negative versus 25.0% in placebo group, p < 0.0001), faster patient recovery (at day 14, 93.9% in FAV00A group had recovered versus 32.5 % in placebo group, p < 0.0001), and reduced disease symptoms compared to placebo (from 9.6{+/-}0.7 to 4.6{+/-}0.9 score points in FAV00A group versus 9.7{+/-}1.1 to 5.6{+/-}1.1 score points in placebo group, p < 0.0001). Using mass-spectrometry, and cell-based and cell-free viral RNA synthesis assays, we identified a hydroxylated metabolite of FAV00A, VR17-04. VR17-04 is capable of inhibiting influenza virus RNA synthesis and present in plasma of patients treated with FAV00A. VR-17-04 inhibits the activity of the influenza virus RNA polymerase more potently than its parent compound. Overall, these results suggest that FAV00A is metabolized in humans to an inhibitor of the influenza virus RNA polymerase that reduces viral shedding and improves patient recovery in influenza patients.

shedding and improves patient recovery in influenza patients. 48

49 Clinical data are available on ClincalTrials.gov under NCT04682444. 50 51 . 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 (which was not certified by peer review) preprint

The copyright holder for this this version posted December 30, 2020. ; https://doi.org/10.1101/2020.12.23.20247569 doi: medRxiv preprint

Infections with respiratory viruses, such as influenza A virus (IAV), are the cause of morbidity 55 and mortality in humans, and a source of serious medical and socio-economic problems for 56 human health worldwide. Of additional concern is the simultaneous circulation of several 57 respiratory viruses in the human population, including influenza B virus (IBV), adenoviruses 58 (ADV), parainfluenza viruses (PIV), respiratory syncytial virus (RSV), coronaviruses (CoV), and 59 various IAV subtypes (e.g., H3N2 and H1N1). All these viruses have the potential to spread 60 rapidly among the human population, become resistant to antivirals, cause severe disease, and be 61 associated with secondary complications (1). Moreover, novel respiratory viruses may emerge 62 from animal reservoirs, such as wild birds or bats, and cause a pandemic when no or little pre-63 existing immunity exists in the human population. Examples of recent outbreaks are the SARS-64

CoV-2 pandemic virus in 2019-2020 and the H1N1 pandemic IAV strain in 2009-2010. 65 Together, these characteristics make respiratory viruses a major threat to human health and 66 extraordinarily difficult target for the development of preventive and therapeutic mitigation 67 measures. 68

Seasonal IAV and IBV infections can be prevented by inactivated or live attenuated 69 vaccines. Vaccines have also been approved for emerging highly pathogenic IAV H5N1 or H7N9 70 strains, and pandemic SARS-CoV-2. However, the efficacy of these vaccines is dependent upon 71 antigenic similarity between the vaccine strain and the circulating viruses, and potentially 72 diminished in high-risk groups (2, 3). Furthermore, the lead-time to produce new vaccines is 73 long. Currently, no vaccines exist against human seasonal CoV or RSV infections, although 74 vaccines are in development (4-6). 75 . 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) preprint

The copyright holder for this this version posted December 30, 2020. ; https://doi.org/10.1101/2020.12.23.20247569 doi: medRxiv preprint 7 Sample size of clinical study 123 A sample size calculation was performed to design the clinical trial. The maximal difference 124 between pre-and post-treatment parameters that was considered not significant was set to 0.25. 125

Thus, a sample size of 30 patients in both the FAV00A-and placebo-treated groups would be 126 needed to demonstrate an efficacy of FAV00A with a statistical power of 90% (bilateral) and a 127 confidence level of 5%. We further estimated that about 15% of the patients would withdraw or 128 be withdrawn (see exclusion criteria below) from the study prior to its completion. Using Fisher's 129 exact test for subsequent analyses, we estimated that a starting cohort of 50 patients per group 130 would be sufficient to achieve the statistical power stated above. conjunctival infection, enlarged lymph nodes, abnormal arterial blood pressure, and auscultation 155 findings (lung and heart). The subjective symptoms included weakness, myalgia, headache, 156 increased body temperature (i.e., subjective severity rating), chills, sore throat and cough. All 157 patients were monitored for adverse events throughout the course of the clinical trial. 158

A score system was used to assess patient health. Objective symptoms were scores as 159 follows: normal or abnormal blood pressure was counted 0 or 4 score points; lung auscultation 160 was counted 0 for vesicular breath sound and wheezing or crepitation were scored 2 or 4 points, 161 respectively; clear and rhythmic heart sounds were each scored 0 points, whereas noisy and 162 arrhythmic heart sounds were scored 2 points each. The subjective symptoms were assessed 163 using a 4-point Likert scale, ranging from 1 (absent) to 4 (severe). After patient assessment, a 164 sum score was calculated. These scores ranged from 4 to 28 (minimum/maximum) for objective 165 symptoms and from 7 to 28 score point for subjective symptoms. 166

In addition to scoring patient's health, information about the number of days since day 0 167 without routine activities due to respiratory infection and overall treatment efficacy (complete, 168 . 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) preprint

The copyright holder for this this version posted December 30, 2020. ; https://doi.org/10.1101/2020.12.23.20247569 doi: medRxiv preprint significant or moderate improve, no significant change, worsened) was recorded by the patient 169 and investigator. Finally, nasal swabs were collected from the patients on days 0, 3 and 7 to 170 determine viral antigen levels by viral antigen immunofluorescence testing. is the author/funder, who has granted medRxiv a license to display the preprint in perpetuity. (which was not certified by peer review) preprint 

A total of 137 patients were recruited and screened for respiratory virus infections, and 100 were 289 randomized for treatment with FAV00A (n = 60) or the placebo control (n = 40) (Fig. 1A) . Age 290 and gender distributions were similar between the two groups (Fig. 1B) . All patients completed 291 the trial as per protocol and no dropouts occurred. At the start of the clinical trial, the patients 292 reported 1.6 ± 0.5 (mean ± SD) days without routine activity due to respiratory infection. Antigen 293 testing showed that the patients treated with FAV00A or placebo were positive for IAV (43.0%), 294 ADV (16.0%), a combination of IAV and ADV (14.0%), or IBV (12.0%). An additional 8.0% of 295 patients tested positive for PIV or RSV, whereas 7% of the patients (5% in the FAV00A group 296 and 10% in the placebo group) were tested negative for IAV, IBV, ADV, PIV or RSV antigens. 297

Overall, 69% of patients were infected with influenza A and/or B viruses. Both the FAV00A-and 298 placebo-treated groups had a similar distribution of respiratory viruses (Fig. 1C ). Throughout the 299 study, there was no deviation from the scheduled medication intake, with a mean treatment 300 duration of 6.8 ± 0.8 days. Topical decongestants were used for a shorter duration of treatment 301 and by a lower proportion of patients treated with FAV00A, when compared to the placebo group 302 (23.3%, n = 14 vs. 75%, n = 30). Expectorants were used by all patients in both treatment groups, 303 but the majority of the FAV00A-treated patients used the drug for 7 days compared to 14 days 304 for the placebo-treated patients. 305 306 . 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 this version posted December 30, 2020. ; https://doi.org/10.1101/2020.12.23.20247569 doi: medRxiv preprint

respiratory disease 308 Antigen testing on day 3, the first visit after initiation of treatment showed that 71.2% (n = 42) of 309 the patients treated with FAV00A and 25.0% (n = 10) of the placebo group were negative for the 310 above-mentioned respiratory virus antigens (p < 0.0001). On day 7, the second visit after 311 initiation of treatment, all patients treated with FAV00A tested negative for respiratory virus 312 antigens, while 82.5% (n = 33) of the placebo control patients tested negative (p = 0.0013) (Fig.  313   2A) . The proportion of patients that tested antigen negative on days 3 and 7 after initiation of 314 treatment was significantly higher (p < 0.0001) in the FAV00A-treated group (70.7%, n = 41) 315 compared to placebo-treated group (25.0%, n = 10). A subgroup analysis indicated a similar 316 positive effect of the FAV00A treatment when compared to the placebo for patients infected with 317 IAV or IBV (75.0%, n = 24 and 22.7%, n = 5, respectively) or ADV (71.4%, n = 5 and 12.5%, n 318 = 1, respectively). Analysis of patient behavior during the clinical study showed that the 319 proportion of patients with no or just one day without routine activities after the start of the 320 treatment was higher in the FAV00A-treated group compared to the placebo group. In the latter 321 group the majority of patients was not able to perform routine activities for two or more days 322 (Fig. 2B) . These group differences were highly significant (p < 0.0001). 323

Analysis of the objective symptoms fever, pharyngeal hyperemia and abnormal lung auscultation 326 contributed mainly to the objective symptom score at the start of the clinical trial on day 0. There 327 was no difference between the objective symptom scores between the two groups on day 0 (9.6 ± 328 0.7 for FAV00A group and 9.7 ± 1.1 for placebo). A low proportion of trial participants 329 . 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 this version posted December 30, 2020. ; https://doi.org/10.1101/2020.12.23.20247569 doi: medRxiv preprint presented conjunctival infection (22.0%) and enlarged lymph nodes (1.0%) and none had 330 abnormal auscultation of the heart or arterial blood pressure at enrollment. The decrease in the 331 objective symptom score from day 0 till day 14 was statistically significant (p < 0.0001) in the 332 FAV00A-treated group (from 9.6 ± 0.7 to 4.6 ± 0.9 score points) when compared to the placebo 333 group (from 9.7 ± 1.1 to 5.6 ± 1.1 score points). Absence of all objective symptoms was observed 334 in both patient groups starting from day 7 (Fig. 3A) and a significant increase in the absence of 335 all 3 symptoms was observed in the FAV00A-treated patients on day 14 ( Fig. 3A; p < 0.0001) . 336

Fever, pharyngeal hyperemia as well as lung auscultation were similar in both treatment groups 337 on day 0, and pharyngeal hyperemia as well as lung auscultation resolved significantly faster in 338 the FAV00A group compared to the placebo group ( Fig. 3B-C) . Together, these observations 339 strongly suggest that FAV00A treatment leads to a faster and highly significant reduction in 340 objective clinical symptoms associated with respiratory virus infections. 341

The subjective symptoms weakness, headache, increased perceived body temperature, sore 344 throat, and cough were present in all patients on day 0 and contributed most to the mean 345 subjective symptom score. Less than half of the patients reported myalgia (46.0%, n = 46) and 346 just a few participants had chills (7.0%, n = 7) at day 0. However, the subjective symptoms were 347 predominantly of mild intensity and almost absent on day 3 in both groups, meaning that they 348 had limited impact on the score. 349

Analysis of the progression of subjective symptoms over time revealed a more 350 pronounced decrease in subjective symptoms in the FAV00A-treated group compared to the 351 placebo group at days 3, 7 and 14 after initiation of treatment (Fig. 4) . Specifically, we found that 352 . 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 this version posted December 30, 2020. ; https://doi.org/10.1101/2020.12.23.20247569 doi: medRxiv preprint by day 14, the score had decreased from 15.7 ± 2.2 to 7.1 ± 0.5 in the FAV00A group compared 353 to 15.4 ± 1.8 to 8.0 ± 1.3 in the placebo group. The mean group differences were -1.2 score points 354 on day 7 (p<0.0001) and -0.9 score points each on day 3 (p=0.0065) and day 14 (p<0.0001). In 355 addition, we found that the absence of all above mentioned subjective symptoms was more 356 pronounced in the FAV00A-treated group on both day 7 and day 14 compared to the placebo 357 control (Fig. 4A) . We also found that the subjective symptoms that contributed most to the sum 358 score (i.e., weakness, headache, increased perceived body temperature, sore throat, and cough) 359 abated faster in the FAV00A-treated group compared to the placebo group (Fig. 4B-F) . These 360 results are also true when treated and non-treated patients suffering from different virus infections 361 were compared on day 14 (data not shown). These findings suggest that FAV00A treatment has a 362 positive outcome on clinically relevant subjective symptoms, such as feelings of weakness, 363 headache, sore throat and coughing. 364 365 FAV00A treatment leads to faster patient recovery 366 Next, we analyzed the recovery time of FAV004-treated and placebo-treated patients in terms of 367 resolution of symptoms and return to premorbid health status. In support of the above described 368 observations, a complete recovery from symptoms associated with the viral respiratory infections 369 was reported by 18.3% (n = 11) of the patients treated with FAV00A on day 7, while none in the 370 placebo group had fully recovered on that day. At day 14, 93.3% (n = 56) of the FAV00A-treated 371 patients had fully recovered, whereas only 32.5% (n = 13) of patients in the placebo group 372 returned to normal health status. A significant improvement in the patient's health status was 373 reported by 60.0% (n = 36) of the FAV00A-treated patients compared to 15.0% (n = 6) of the 374 placebo-treated patients on day 3 (p < 0.0001). FAV00A treatment led to a significant 375 . 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. optimized for each cell line. We found that FAV00A significantly affected WSN titers in A549 391 cells with an IC 50 of 322 µM, in line with previous observations (Fig. 5A and 5E ). Higher IC 50 392 values were observed for the RD, Caco-2 and HepG2 cells (Fig. 5B , C, D and E). Selectivity 393 indexes for FAV00A were >11 for A549, RD, and Caco-2 cells (Fig. 5E ), but lower for HepG2 394 cells due FAV00A's higher cytotoxicity in these cells (Fig. 5D and E) . These data confirm that 395 FAV00A can inhibit influenza A virus infections in a variety of human cell lines derived from 396 lung, liver, colon and skeletal muscle tissues; however, its antiviral effects are strongly cell-type 397 . 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) preprint

The copyright holder for this this version posted December 30, 2020. ; https://doi.org/10.1101/2020.12.23.20247569 doi: medRxiv preprint dependent, with respiratory tract derived cells such as NHBE (12) and A549 cells (Fig. 5A) The weak inhibition of the IAV RNA polymerase activity by FAV00A is unlikely to be 432 responsible for the obvious antiviral efficacy of FAV00A observed in cell culture and in patients. 433

Based on the data obtained with FAV00A treatment in different cell types (Fig. 5) , we 434 hypothesized that a metabolite of FAV00A could be the actual inhibitor of the IAV RNA 435 polymerase in cell culture. To test this hypothesis, we treated A549 cells with FAV00A for 24 h, 436 lysed the cells and then performed IAV RNA polymerase activity assays in the presence of the 437 cellular extract. We observed a strong, concentration-dependent inhibition of IAV RNA 438 polymerase activity, which is consistent with the hypothesis that a metabolite of FAV00A 439 inhibits influenza virus RNA synthesis during IAV infection (Fig. 6C) . 440

To confirm that FAV00A is metabolized in humans, we analyzed human plasma samples 441 from a phase I pharmacokinetic study using HPLC and MS/MS. This analysis revealed the 442 presence of 4 phase I metabolites (Fig. 6D) , of which two were hydroxylated (VR17-02 and 443 . 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) preprint

The copyright holder for this this version posted December 30, 2020. ; https://doi.org/10.1101/2020.12.23.20247569 doi: medRxiv preprint 21 VR17-04), one was demethylated (VR18-06), and one was both hydroxylated and demethylated 444 (VR18-05). We also detected phase II metabolites that had been formed by glucuronidation or 445 sulfate conjugation of phase I metabolites. Since glucuronidation or sulfate conjugation typically 446 result in drug inactivation, we did not study the phase II metabolites further. In order to test if 447 infections, when compared to placebo. In addition, objective symptoms (fever, pharyngeal 465 hyperemia) and subjective symptoms (weakness, headache, increased perceived body 466 . 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) preprint

The copyright holder for this this version posted December 30, 2020. ; https://doi.org/10.1101/2020.12.23.20247569 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) preprint

The copyright holder for this this version posted December 30, 2020. ; https://doi.org/10.1101/2020.12.23.20247569 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) preprint

The copyright holder for this this version posted December 30, 2020. ; https://doi.org/10.1101/2020.12.23.20247569 doi: medRxiv preprint

The second birth 576 of the drug Amizon: the results of international scientific studies confirm the antiviral 577 effect

enisamium, an isonicotinic acid derivative, against influenza viruses in differentiated 580 normal human bronchial epithelial cells

Influenza virus RNA polymerase: insights into the 582 mechanisms of viral RNA synthesis

The role of the priming loop in 584 influenza A virus RNA synthesis

Mini viral RNAs act as innate immune agonists during influenza virus infection

Assays to Measure the Activity of 590

Influenza Virus Polymerase

temperature, sore throat, and cough) were reduced and disappeared more quickly in patients 467 receiving FAV00A treatment. These results suggest (i) a direct clinical effect of FAV00A on the 468 outcome of viral respiratory infections in humans and (ii) that FAV00A treatment leads to faster 469 patient recovery and a reduction of symptoms. 470The use of concomitant symptomatic therapy (e.g., topical decongestants and 471 expectorants) as well as the impact of the concomitant medication on the evaluation of the 472 symptoms could affect clinical trial results. However, in the FAV00A-treated group, 473 symptomatic drugs were taken less frequently and for shorter durations than in the placebo group, 474suggesting that an overestimation of the FAV00A effects due to concomitant symptomatic 475 therapy can be excluded. In fact, it is tempting to suggest that FAV00A treatment led to reduced 476 symptomatic therapy in the FAV00A-treated group. Moreover, in recent clinical trials with 477 influenza virus neuraminidase inhibitors (e.g., oseltamivir), the evaluated symptoms such as 478 fever, chills, headache, muscle ache, and cough as well as other clinical parameters return to 479 routine activities, were similar to the symptoms and parameters evaluated in this clinical trial (17, 480 18). Therefore, the methods used to evaluate the efficacy of FAV00A for human respiratory virus 481 infections and the results presented here should be reliable. 482We also evaluated symptoms that are not characteristic for respiratory tract infections 483 such as abnormal arterial blood pressure, abnormal auscultation findings of the heart, enlarged 484 lymph nodes, and conjunctival infection. Most of these symptoms were indeed absent in the trial 485 population or only present in a low percentage of the patients and of mild intensity. These non-486 characteristic symptoms were revealed in all or almost all patients on day 3 after initiation of 487 treatment in both treatment groups, suggesting that these atypical symptoms did not contribute 488considerably to the outcome of the clinical trial. The patients included in this study had primarily 489 mild to moderate respiratory disease. Thus, the efficacy of FAV00A against severe forms of 490 . 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) preprint (Fig. 6) . Moreover, 501 using in vitro influenza virus RNA polymerase assays, we demonstrated that FAV00A can 502 directly inhibit influenza virus replication and transcription and that one of its metabolites, VR17-503 04, is a more potent inhibitor of influenza virus RNA synthesis than FAV00A (Fig. 6) . These 504 findings are in line with the inhibitory effect of extracts from cultured human lung cells treated 505 with FAV00A (Fig. 5) ; these observations might imply that FAV00A acts on the influenza virus 506 RNA polymerase after hydroxylation to VR17-04. 507In summary, we have performed a single-blinded clinical trial that shows that FAV00A 508 treatment of patients with viral respiratory infections leads to faster patient recovery and reduced 509 virus shedding when compared to the placebo control. Our results are supported by virus 510infections and viral replication in vitro studies on cells infected with IAV and with influenza viral 511 RNA replication assays that suggest that FAV00A can directly inhibit influenza viral replication 512 through a hydroxylated metabolite, VR17-04. This study thus advances our understanding of 513 . 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. 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) preprintThe copyright holder for this this version posted December 30, 2020. ; https://doi.org/10.1101/2020.12.23.20247569 doi: medRxiv preprint on IAV RNA polymerase activity in vitro. Five hundred or 2500 µg of FAV00A was added to 658 A549 cells for 24 h. Next, cells were lysed and 1/10 of the lysate was added to in vitro 659 . 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. . 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) preprintThe copyright holder for this this version posted December 30, 2020. ; https://doi.org/10.1101/2020.12.23.20247569 doi: medRxiv preprint