Trial Primary and Secondary Endpoints – Full Analysis Set (FAS)
| Full Analysis Set (n=183) | Statistic | Estimate (95% CI) | Treatment Effect | Estimate (95% CI) | p value | |
| AZITHRO | SOC | |||||
| Primary outcome | ||||||
| Incidence of clinical improvement or live discharge | Subdistribution HR* | 1.023 (0.758; 1.379) | 0.8839 | |||
| - at 15 days | CIF# (%) | 78.2% (69.5%; 84.6%) | 81.3% (69.0%; 89.0%) | |||
| - at 29 days | CIF# (%) | 86.6% (78.9%; 91.6%) | 89.1% (77.8%; 94.8%) | |||
| Time to sustained clinical improvement or live discharge | Median (Days) | 6 (6; 8) | 8 (6; 10) | |||
| Secondary Outcomes (compared to FAS, total n-values may be lower than n=119 for AZITHRO and n=64 for SOC due to missing values**) | ||||||
| Clinical status at day 15 | n/N (%) | Common OR of having lower clinical status at day 15$ | 0.83 (0.47; 1.53) | 0.5776 | ||
| 1. Not hosp., no limitations | 14/97 (14.58%) | 11/53 (20.75%) | ||||
| 2. Not hosp., limitations | 50/97 (51.55%) | 28/53 (52.83%) | ||||
| 3. Hosp., no suppl. oxygen | 7/97 (7.22%) | 2/53 (3.77%) | ||||
| 4. Hosp., requiring oxygen | 10/97 (10.31%) | 4/53 (7.55%) | ||||
| 5. Hosp., NIV | 5/97 (5.15%) | 3/53 (5.66%) | ||||
| 6. Hospitalised, MV or ECMO | 7/97 (7.22%) | 4/53 (7.55%) | ||||
| 7. Death | 4/97 (4.12%) | 1/53 (1.89%) | ||||
| Cumulative clinical status up to day 15 | Geometric meanΣ | 42.61 (39.66; 45.77) | 42.60 (38.64; 46.98) | Treatment ratio∫ | 1.00 (0.90; 1.12) | 0.9508 |
| All-cause mortality | ||||||
| - 15-days | KM (%) | 3.4% (1.3%; 8.7%) | 1.6% (0.2%; 10.6%) | Hazard Ratio◊ | 1.799 (0.201; 16.09) | 0.5996 |
| - 29-days | KM (%) | 7.6% (4.0%; 14.0%) | 6.5% (2.5%; 16.3%) | Hazard Ratio◊ | 1.109 (0.339; 3.628) | 0.8666 |
| ICU (29 days) | ||||||
| - Incidence (all patients) | CIF# (%) | 26.1% (18.5%; 34.2%) | 23.4% (13.9%; 34.4%) | Subdistribution HR* | 1.066 (0.572; 1.985) | 0.8412 |
| - Duration of ICU stay (ICU admitted patients) | Median (days) | 11 (7; 18) | 17 (3; 28) | Subdistribution HR for live discharge from ICU* | 1.293 (0.661; 2.529) | 0.4534 |
| Mechanical ventilation (29 days) | ||||||
| - Incidence (all patients) | CIF# (%) | 12.0% (6.9%; 18.6%) | 15.6% (8.0%; 25.6%) | Subdistribution HR* | 0.738 (0.313; 1.741) | 0.4838 |
| - Duration of MV (MV patients) | Median (Days) | 13 (NC; NC) | 19 (NC; NC) | Subdistribution HR for live weaning from MV* | 0.897 (0.263; 3.034) | 0.8609 |
| Supplemental oxygen (29 days) | ||||||
| - Incidence (all patients) | CIF# (%) | 83.2% (75.1%; 88.9%) | 79.7% (67.4%; 87.8%) | Subdistribution HR* | 0.998 (0.837; 1.191) | 0.7811 |
| - Duration of supplemental oxygen (patients with supplemental oxygen) | Median (Days) | 6 (5; 8) | 7 (5;11) | Subdistribution HR for live weaning from oxygen* | 1.042 (0.737; 1.479) | 0.9465 |
| Hospital stay (29 days) | Subdistribution HR for live hospital discharge* | |||||
| - Occurrence of live hospital discharge | CIF# (%) | 85.8% (78.0%; 91.0%) | 84.4% (72.5%; 91.4%) | 1.064 (0.780; 1.451) | 0.6954 | |
| - Duration of hospital stay (Days) | Median (Days) | 7 (6; 8) | 8 (6; 9) | |||
| Safety Outcome | ||||||
| Combined cardiac endpoint (hs-Troponin >0.5 ng/ml and/or ventricular arrhythmia requiring intervention and/or sudden cardiac death) | n/N (%) | 24/119 (20.17%) | 12/64 (18.75%) | Odds Ratio¥ | 1.40 (0.61; 3.19) | 0.4265 |
| Exploratory Outcomes | ||||||
| Normal CT at 5–7 weeks | n/N (%) | 25/75 (33.33%) | 12/40 (30.00%) | Odds Ratio‡ | 1.56 (0.64;3.79) | 0.3233 |
| DLCO (% predicted) | Estimated meanΩ (%) | 66.49 (59.96; 73.02) | 67.00 (57.91; 76.09) | Treatment difference### | 0.22 (−11.1; 11.51) | 0.9695 |
| 6-min walking test | Estimated meanΩ (meters) | 496.5 (465.0; 528.0) | 456.3 (413.2; 499.3) | Treatment difference### | 32.51 (−21.2; 86.25) | 0.2383 |
Abbreviations: NIV: non-invasive ventilation, MV: mechanical ventilation, ECMO: extracorporeal membrane oxygenation.
#Event rates were estimated as cumulative incidence functions (CIF), taking into account the competing risk of death.
*Score>1 favours azithromycin. Hazard ratio was obtained using a Fine & Gray model for competing risk data including treatment, study site and study period as factors.
$Score>1 favours azithromycin. Odds ratio was obtained from a multinomial logistic regression with factors for treatment, disease severity and Clinical Status on Day 0.
ΣGeometric mean was obtained using a general linear model including treatment as factor.
∫Score>1 favours SOC. Treatment ratio was calculated as the ratio of geometric means, obtained using a general linear model including treatment, study site, study period and clinical status on Day 0 as factors. (Notes: Data were log-transformed prior to statistical analysis).
†Score>1 favours SOC. Hazard ratio was obtained using a Fine & Gray model for competing risk data including treatment, study site and study period as factors.
◊Score>1 favours SOC. Hazard ratio was obtained using Log-Rank test after event rates were estimated using Kaplan-Meier methodology.
‡Score>1 favours azithromycin. Odds ratios were obtained using logistic regression including treatment, study site and study period as factors in the model.
¥Score>1 favours SOC. Odds ratios were obtained using logistic regression including treatment, study site and study period as factors in the model.
ΩObtained using a general linear model including treatment as factor.
ΔScore>1 favours SOC. Hazard ratio was obtained using a Fine & Gray model for competing risk data including treatment, study site and study period as factors
###Higher score (> 0) favours azithromycin. Obtained using a general linear model including treatment, study site and period as factors.
**Missing data are accounted for by multiple imputation.