key: cord-0901407-0y4f1sgn
authors: Tsiakos, K.; Tsakiris, A.; Tsibris, G.; Voutsinas, P.; Panagopoulos, P.; Kosmidou, M.; Petrakis, V.; Gravvani, A.; Gkavogianni, T.; Klouras, E.; Katrini, K.; Koufargyris, P.; Rapti, I.; Karageorgos, A.; Vrentzos, E.; Damoulari, C.; Zarkada, V.; Sidiropoulou, C.; Artemi, S.; Papapostolou, A.; Michelakis, E.; Georgiopoulou, M.; Myrodia, D.-M.; Tsiamalos, P.; Syrigos, K.; Chrysos, G.; Nitsotolis, T.; Milionis, H.; Poulakou, G.; Giamarellos-Bourboulis, E.
title: Oral clarithromycin in COVID-19 of moderate severity: the ACHIEVE open-label trial using concurrent matched comparators
date: 2020-12-26
journal: nan
DOI: 10.1101/2020.12.22.20248753
sha: 5d622a025dee59a898e6a8c55e309d07699894f5
doc_id: 901407
cord_uid: 0y4f1sgn

Background: Clarithromycin clinical efficacy has not been described in COVID-19. Research question: Is oral clarithromycin beneficial for treating patients diagnosed with COVID-19? Study and methods: An open-label non-randomized trial in 90 patients with COVID-19 of moderate severity was conducted at four study sites in Greece between May and October 2020. Ninety participants with respiratory tract infections received clarithromycin 500 mg every 12 hours for 7 days; another 90 standard-of-care (SOC) propensity score-matched concurrent controls received azithromycin plus hydroxychloroquine. The composite primary endpoint was defined for patients with upper respiratory tract infection as: (a) no need for hospital re-admission or (b) lack of progression into lower respiratory tract infection and, for patients with lower respiratory tract infection, as at least 50% decrease of the score of respiratory symptoms at the end-of-treatment (EOT) without progression into severe respiratory failure (SRF). The incidence SRF at the test-of-cure (TOC) on day 14 was a secondary endpoint. For clarithromycin-treated patients, viral load of SARS-CoV-2, biomarkers, the function of mononuclear cells, and safety were assessed; biomarkers were also measured in SOC comparators. Results: The primary endpoint was attained in 86.7% of patients treated with clarithromycin (95% CIs 78.1-92.2%) and 73.3% of concurrent SOC comparators (95%CIs; 63.4-81.4%). The odds ratio for the primary endpoint with clarithromycin treatment in univariate analysis was 2.36 (95%CIs 1.09-5.08; P: 0.039). Results were confirmed after multivariate stepwise logistic regression analysis (odds ratio 3.30; 95% CI 1.10-9.87; P: 0.033). At the TOC visit, the incidence of SRF was 12.2% (n = 11 ; 95%CIs 6.9-20.6%) among patients treated with clarithromycin (odds ratio for SRF 0.38; 95%CIs 0.17-0.84) versus 26.7% (n= 24; 95%CIs 18.6-36.6%) among concurrent SOC comparators (P: 0.023). Clarithromycin use was associated with decreases in circulating levels of C-reactive protein, of tumour necrosis factor-alpha and of interleukin (IL)-6; by an increase of the ratio of Th1 to Th2 mononuclear responses; and by suppression of SARS-CoV-2 relative viral load. No safety concerns were reported. Patients starting clarithromycin with the first five days from symptoms onset achieved better responses. Interpretation: Clarithromycin treatment is associated with early clinical improvement in patients with moderate COVID-19. Modulation of the Th1/Th2 responses is proposed as the mechanism of action.

. CC-BY-ND 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 26, 2020. ; https://doi.org/10.1101/2020.12.22.20248753 doi: medRxiv preprint

The progression of pneumonia by the novel coronavirus SARS-CoV-2 (also known as COVID-19) is unpredictable. Some patients deteriorate into severe respiratory failure (SRF) through a process of complex immune dysregulation which is often accompanied by unfavourable outcome 1 Two double-blind, randomized, placebo-controlled clinical trials of our group have reported that intravenous administration of clarithromycin is associated with substantial reduction of the risk of death among critically ill patients with Gramnegative sepsis 9, 10 . The first trial was conducted among patients with sepsis developing after ventilator-associated pneumonia; survival by day 90 was 40% in the placebo group and 57% in the clarithromycin treatment group 11 . We have also recently shown that 28-day mortality of severe CAP was 20.8% among patients . CC-BY-ND 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 26, 2020. ; https://doi.org/10.1101/2020.12.22.20248753 doi: medRxiv preprint 6 treated with a combination of β -lactam and clarithromycin and 33.8% when treated with a combination of β -lactam and azithromycin 12 .

Based on the above evidence, it seems likely that treatment of COVID-19 with oral clarithromycin may attenuate the hyper-inflammatory responses of the host and decrease the risk of progression into SRF. We conducted the ACHIEVE trial (Antiinflammatory Clarithromycin to Improve SARS-CoV 2 Infection Early) to investigate whether early administration of oral clarithromycin in patients with upper or lower respiratory tract infection (RTI) due to SARS-CoV-2, may attenuate the inflammatory burden, modulate the immune response of the host and result in early clinical improvement. We compared the efficacy of clarithromycin with a group of propensityscore matched concurrent comparators receiving standard-of-care treatment (SOC) with azithromycin and hydroxychloroquine. The efficacy of clarithromycin was also studied in relation to the delay on onset since start of symptoms and to the lineage of the infecting strain of SARS-CoV-2.

ACHIEVE is an open-label non-randomized trial conducted in four study sites The initial draft of the manuscript was written by the first and the last author. All authors vouch for the adherence of the trial to the protocol and first and last author vouch for the accuracy and completeness of the data and analysis.

Enrolled patients were adults (age ≥18 years) with confirmed infection by SARS-CoV-2 virus by real-time PCR reaction of nasopharyngeal secretions; and infection of the upper respiratory tract (URTI) or of the lower respiratory tract (LRTI).

The URTI was defined as the acute presentation of at least two of the following signs: a) core temperature ≥ 37.5 0 C; b) new onset of cough; c) chills or rigor; and d) total absolute lymphocyte count less than 1,500/mm 3 . The LRTI was defined as the presence of infiltrates compatible with lower respiratory tract infections in chest X-ray or in chest computed tomography accompanied by at least one of the following: a) new onset of cough or worsening cough; b) dyspnea; c) respiratory rales compatible with lung infection; and d) total absolute lymphocyte count less than 1,500/mm 3 . . CC-BY-ND 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 In order to derive the most appropriate ones among concurrent comparators treated at the same time period with azithromycin and hydroxychloroquine, the inclusion and exclusion criteria of the ACHIEVE trial were applied to all available concurrent comparators. Patients meeting the inclusion criteria and not meeting the exclusion criteria were eligible for inclusion in the comparator SOC group. Patients who were administered biological disease modifying agents (bDMARDs) because of severe COVID-19 were excluded from participation in the SOC group. Written informed consent was provided by the patient or legal representative before screening.

Enrolled patients received one tablet of 500 mg of clarithromycin every 12 hours for 7 days. All other drugs except macrolides and hydroxychloroquine/ chloroquine phosphate were allowed. Study visits were done daily starting from day 1 before start of treatment until day 8 after end of treatment (end-of treatment visit, EOT). The last test-of-cure (TOC) visit was done on day 14. For patients discharged earlier from hospital, EOT and TOC visits were done by phone calls. Recorded baseline information was severity assessed by APACHE II score, SOFA score and pneumonia severity index (PSI), comorbidities and laboratory information. On each visit follow-up, the state of the patient was assessed and the respiratory symptoms score (RSS) was calculated (see Supplementary methods). Fifteen ml of whole blood . CC-BY-ND 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 26, 2020. ; https://doi.org/10.1101/2020.12.22.20248753 doi: medRxiv preprint was collected on day 1 before start of treatment and on EOT into EDTA-coated tubes and sterile and pyrogen-free tubes for the isolation of peripheral blood mononuclear cells (PBMCs), serum and plasma. PBMCs were stimulated for cytokine production. Blood was also sampled from concurrent SOC comparators on day 1 before start of treatment and on day 8 for the measurement of biomarkers.

The primary study endpoint was described on EOT day 8. This was a composite endpoint defined differently for patients with URTI and for patients with LRTI. For patients with URTI the primary endpoint was defined as either no need for hospital re-admission in case of earlier discharge or as lack of progression into lower RTI. For patients with LRTI the primary endpoint was defined as any least 50% decrease of the RSS from the baseline score of day 1 provided that the patient has . CC-BY-ND 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 study endpoints were defined for concurrent SOC comparators using the same definitions as for patients treated with oral clarithromycin in the ACHIEVE study.

Adverse Events (AE) (Common Terminology Criteria for Adverse Events, 

Qualitative data were presented as percentages with confidence intervals (CI) and quantitative data as means and SD; serum biomarkers and cytokines were expressed as means and SD. After applying the inclusion and exclusion criteria of the ACHIEVE study, the available concurrent SOC comparators treated with azithromycin and hydroxychloroquine were selected. Propensity score 1:1 matching was applied to finally select the best matched comparators to the 90 patients treated with clarithromycin. Matching variables were: age, Charlson's comorbidity index . CC-BY-ND 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 1 1 (CCI), admission PSI; the frequency of URTI and LRTI; and admission C-reactive protein. These matching variables were selected because they were all covariates measured before treatment and they are related to the treatment and outcome.

Comparisons between groups for the primary endpoint were performed by the Fisher's exact test using confirmatory forward stepwise logistic regression analysis (IBM SPSS Statistics v. 25.0). Paired comparisons were done by the Wilcoxon's signed rank test. Using the median of the time interval between the start of COVID-19 symptoms and start of clarithromycin, patients were divided into those with an early start and into those with a late start of clarithromycin. The achievement of the primary endpoint was compared between patients with early and late start by the binomial test. Any two-sided P value <0.05 was statistically significant. 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 26, 2020. ; https://doi.org/10.1101/2020.12.22.20248753 doi: medRxiv preprint 1 2 between the two groups, although there was a trend for patients enrolled in the ACHIEVE trial to present with a higher CCI.

The primary endpoint was met in 78 patients treated with clarithromycin 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 26, 2020. ; 1 3 independent variable associated with protection from SRF at TOC was the intake of clarithromycin (odds ratio 0.22; 95% CIs 0.06-0.79; P: 0.022; supplementary Table   1 ). The same analysis indicated that baseline SOFA score, PSI and CRP were independent variables associated with progression into SRF at TOC visit. The proposed mechanism of action of clarithromycin is associated with the increase of the Th1 response from circulating PBMCs compared to an attenuation of the Th2 response. More precisely, we were able to show using stimulation with heatkilled C.albicans for 5 days that the targeted production of IFNγ, which reflects the Th1 response, was increased (eFigure 4A). Under the same conditions, the targeted production of IL-6 that reflects the Th2 response was decreased (eFigure 4B). The ratio of Th1/Th2 response from PBMCs at the EOT was greater among patients with . CC-BY-ND 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. Table 2 ).

. CC-BY-ND 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 26, 2020. ; https://doi.org/10.1101/2020.12.22.20248753 doi: medRxiv preprint 1 5

The adverse events (AEs) and serious adverse events (SAEs) that were captured during the study period of 14 days are listed in eTable 3. Reported events probably or possibly-related to the study drug with incidence greater than 1% were: increase of aminotransferases (Grade I 18.8%; Grade II 1.1%); diarrhea (mild 12.1%; moderate 1.1%); mild vomiting in 1.1%, Grade I increase of bilirubin in 2.22%; and mild allergic reaction in 1.1%) (eTable 3). No drug-related SAEs were reported.

Based on previous experience with clarithromycin 9-12 , we sought to investigate its potential in reducing the hyper-inflammatory response in moderate COVID-19 and contribute to better clinical outcomes than currently available treatment approaches.

The ACHIEVE open-label trial enrolled patients with moderate severity of COVID-19, compared with propensity-matched concurrent SOC comparators hospitalized during the same time-period. Analysis revealed that clarithromycin treatment was associated with a significantly greater clinical benefit at the EOT visit of 86.7%, as compared with 73.3% achieved by the concurrent SOC comparators. This early benefit shown at the end of the 7-day course with clarithromycin was further reflected in the incidence of SRF at the TOC visit of day 14, when treatment with clarithromycin provided 78% relative decrease of the risk for SRF compared to the SOC regimen. The benefit was pronounced for patients starting clarithromycin within the first 5 days from the onset of Covid-19-related symptoms. The benefit of clarithromycin was also linked with decrease of CRP and IL-6 at the EOT.

The current study provides insight into the mechanism of action of clarithromycin. Emerging data suggest that active virus replication and persistence in . CC-BY-ND 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. in an attempt to include (and benefit) as many patients as possible since no SOC was framed at that time period. The azithromycin plus hydroxychloroquine concurrent comparators were optimally matched without differences in baseline severity and coadministered treatment. When the study was started, most of the patients with COVID-19 hospitalized in the Greek Hospitals were co-administered azithromycin and hydroxychloroquine as off-label treatment option. In the following months, emerging data showed that hydroxychloroquine does not affect the natural course of . CC-BY-ND 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 26, 2020. ; 1 7 COVID-19 17, 18 . As such, the comparison between the clarithromycin group and the azithromycin plus hydroxychloroquine group should be conceived as comparison between clarithromycin and azithromycin.

In the randomized COALITION trial from Brazil, patients with moderate to severe COVID-19 were randomized to treatment with placebo (n=227), hydroxychloroquine (n=221) and a combination of hydroxychloroquine and azithromycin (n=217). The primary study endpoint was the distribution of the 7-point ordinal scale after 15 days. The achievement of the primary endpoint was similar between the three groups of treatment 19 . However, the design of the COALITION trial was different than the ACHIEVE trial where participants had less severe disease.

Furthermore, the median time from symptoms onset to start of treatment was seven days in the COALITION trial contrary to five days in the ACHIEVE trial. The same group of investigators performed the COALITION II trial were patients were allocated to treatment with standard-of-care with (n=214) or without azithromycin (n=183).

Although the two groups did not differ in the achievement of the primary endpoint after 15 days, patients allocated to azithromycin treatment had a greater chance of being at a greater point of the scale after 7 days of treatment 20 . Both COALITION trials evidenced the safety of azithromycin.

In conclusion, the results of the ACHIEVE trial clearly indicate that clarithromycin treatment in patients with moderate COVID-19 is associated with early clinical improvement and containment of viral load. This is associated with an increase of the ratio of Th1/Th2 response. To the best of our knowledge this is the first study to show the anti-inflammatory impact of clarithromycin on COVID-19.

Further studies are needed to better define its future role in mild and moderate COVID-19.

. CC-BY-ND 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 26, 2020. ;

The study is supported by Abbott Operation Products AG. P. Panagopoulos has received honoraria from GILEAD Sciences, Janssen, and MSD.

G. Poulakou has received independent educational grants from Pfizer, MSD, Angelini, and Biorad. . CC-BY-ND 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 

. CC-BY-ND 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 26, 2020. ; https://doi.org/10.1101/2020.12.22.20248753 doi: medRxiv preprint 2 5 Total white blood cells (/mm 3 ) 6114. 6 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 26, 2020. ; https://doi.org/10.1101/2020.12.22.20248753 doi: medRxiv preprint

Complex immune dysregulation in COVID-19 patients with severe respiratory failure

Hydroxychloroquine and azithromycin as a treatment of COVID-19: results of an open-label non-randomized clinical trial

Diagnosis and treatment of adults with community-acquired pneumonia

Lower mortality among patients with community-acquired pneumonia treated with a macrolide plus a beta-lactam agent versus a beta-lactam agent alone

Antibiotics for bacteremic pneumonia: improved outcomes with macrolides but not fluoroquinolones

Impact of macrolide therapy on mortality for patients with severe sepsis due to pneumonia

Combination antibiotic therapy with macrolides improves survival in intubated patients with community-acquired pneumonia

Clarithromycin treatment

Admission SOFA score, mean (SD) 2.60 (1.55)

Lymphocytes/mm 3 , mean (SD)

CRP, mg/l, mean (SD)

*variable did not enter the equation after three steps of the multivariate model

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 26, 2020. ; https://doi.org/10.1101/2020.12.22.20248753 doi: medRxiv preprint . CC-BY-ND 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 26, 2020. ; https://doi.org/10.1101/2020.12.22.20248753 doi: medRxiv preprint