Ye Qiu ¹ Hao Wen ² Haoru Wang ¹ Wenjun Sun ¹ Guangchao Li ¹ Shaoqiang Li ¹ Yan Wang ¹ Jingnan Zhai ¹ Yangqing Zhan ¹ Yutian Su ¹ Zhiwei Long ² Zhengtu Li ¹ Feng Ye ^3*

• ¹ First Affiliated Hospital of Guangzhou Medical University, Guangzhou, Guangdong Province, China
• ² Guangzhou Medical University, Guangzhou, Guangdong Province, China
• ³ State Key Laboratory of Respiratory Disease, Guangzhou Institute of Respiratory Health, The First Affiliated Hospital of Guangzhou Medical University, Guangzhou, China

Background: Nirmatrelvir-ritonavir (Paxlovid) has received emergency use authorization from the US Food and Drug Administration owing to its effectiveness and safety. However, data on the effectiveness and safety of Paxlovid use in COVID-19 patients with onset of more than five days are lacking. Methods: A real-world retrospective study was performed during the outbreak involving the SARS-CoV-2 BA.5.2 subvariant. Hospitalized COVID-19 patients (including mild, moderate, severe and critical cases) were divided into three groups: Paxlovid treatment within (Group A) or more than (Group B) five days of COVID-19 onset and no Paxlovid treatment during more than five days of COVID-19 onset with only basic symptomatic treatment (Group C). Endpoints were all-cause 28-day mortality, improvement in clinical classification, and a composite endpoint of disease progression, viral load and virus elimination time. Safety was assessed by comparing adverse events reported during treatment in each group. Results: During the period, 248 hospitalized COVID-19 patients, including 55 in Group A, 170 in Group B, and 23 in Group C, were enrolled. There were no significant differences in the clinical classification improvement rate [80.0% (16/20) vs. 81.3% (52/64), p=1.000; 60.0% (21/35) vs. 55.7% (59/106), p=0.653, respectively] or all-cause 28-day mortality [0% (0/20) vs. 1.6% (1/64), p=1.000; 11.4% (4/35) vs. 6.6% (7/106), p=0.576, respectively] between Groups A and B for nonsevere and severe cases. However, the clinical classification improvement rate in Group B was markedly higher than that in Group C [81.3% (52/64) vs. 50.0% (6/12), p=0.049] among nonsevere cases. Cycle threshold values of the N and ORF genes in Group B were significantly increased after Paxlovid treatment [31.14 (IQR 26.81-33.93) vs. 38.14 (IQR 36.92-40.00), p<0.001; 31.33 (IQR 26.00-33.47) vs. 38.62 (IQR 35.62-40.00), p<0.001, respectively]. No significant differences in reported adverse events of neurological disease (p=0.571), liver injury (p=0.960) or kidney injury (p=0.193) between Group A and Group B were found. Conclusions: Paxlovid treatment within ten days of onset can shorten the disease course of COVID-19 by reducing the viral load. Paxlovid is effective and safe in treating COVID-19 with onset of more than five or even ten days when patients have a high viral load.