A large-scale study reveals azvudine’s comparable efficacy to Paxlovid, fewer adverse effects, and potential antitumor benefits in patients with liver cancer.

Study: Real‐world effectiveness and safety of oral azvudine versus nirmatrelvir‒ritonavir (Paxlovid) in hospitalized patients with COVID-19: a multicenter, retrospective, cohort study. Image Credit: Cryptographer/Shutterstock.com

In a recent study published in the Signal Transduction and Targeted Therapy, a group of researchers compared the effectiveness and safety of azvudine versus nirmatrelvir-ritonavir (Paxlovid) in hospitalized coronavirus disease 2019 (COVID-19) patients, with a focus on clinical outcomes, adverse events (AEs), and potential benefits for patients with malignant tumors.

Background

Since the outbreak of COVID-19 in December 2019, caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), it has resulted in over 776 million infections and 7 million deaths globally as of August 2024. Despite vaccination reducing severe cases, its efficacy against immune-evasive variants remains limited.

COVID-19 treatment mainly includes antiviral and immunomodulatory drugs, but the latter is effective mostly in severe cases. Paxlovid and azvudine are widely used antivirals, yet their comparative efficacy and safety remain debated.

Further research is needed to clarify their roles in managing COVID-19 and associated conditions like hepatocellular carcinoma.

About the study

The present multicenter, retrospective cohort study was conducted in Henan and Xinjiang provinces, China, involving hospitalized patients with confirmed SARS-CoV-2 infection between December 5, 2022, and January 31, 2023.

The study population included 37,606 patients from ten hospitals in Henan and 3,270 patients from one hospital in Xinjiang. Eligibility criteria required participants to be aged 18 or older, have a positive reverse transcription polymerase chain reaction (RT-PCR) test for SARS-CoV-2, and have received standard therapy along with either azvudine or Paxlovid.

Patients who did not receive antiviral agents received other antiviral regimens, were pregnant, or had contraindications to either drug were excluded. Data were collected from electronic medical records, including demographics, admissions, outcomes, prescriptions, and laboratory results.

Participants were grouped by drug prescription, and propensity score matching (PSM) at a 1:2 ratio ensured balance in baseline covariates. Outcomes included all-cause death, disease progression, and AEs, categorized by the Common Terminology Criteria for Adverse Events Version 5.0.

Statistical analyses used Kaplan–Meier curves, Cox regression, and subgroup evaluations. Sensitivity analyses addressed missing values, alternative models, and early discharge or mortality. The study adhered to ethical guidelines under the Declaration of Helsinki.

Study results

The study included 7,145 patients with confirmed COVID-19 from ten hospitals in Henan Province, China. After rigorous inclusion and exclusion criteria, 6,943 patients receiving azvudine and 1,202 receiving Paxlovid were eligible for analysis. PSM at a 2:1 ratio balanced baseline characteristic, yielding 2,404 azvudine and 1,202 Paxlovid recipients for the final cohort.

The primary outcome was all-cause death, with 469 deaths observed: 288 in the azvudine group and 181 in the Paxlovid group. Kaplan–Meier analysis showed a significantly lower risk of death in the azvudine group compared with Paxlovid (p = 0.038).

Multivariable Cox regression analysis confirmed this, with a hazard ratio (HR) of 0.82 (95% confidence interval [CI]: 0.676-0.987, p = 0.036). Composite disease progression occurred in 681 patients: 446 in the azvudine group and 235 in the Paxlovid group.

Kaplan–Meier analysis revealed no significant difference between groups (p = 0.95), and Cox analysis yielded an HR of 1.15 (95% CI: 0.975-1.345, p = 0.097).

Sensitivity analyses supported the accuracy of these findings. Results remained consistent across multiple methods of handling missing data, alternative matching models, and the exclusion of early discharges or deaths.

For example, after the imputation of missing values, Cox analysis indicated a 21% lower risk of death with azvudine (HR: 0.79, 95% CI: 0.658–0.959, p = 0.016). A Probit-based matching model also showed a significant reduction in mortality risk for azvudine compared with Paxlovid (HR: 0.73, 95% CI: 0.603-0.884, p = 0.001).

Validation in a cohort from Xinjiang Province, comprising 79 azvudine and 78 Paxlovid recipients, showed no significant differences in composite outcomes (p = 0.27). Still, Cox analysis revealed a lower mortality risk for azvudine (HR: 0.53, 95% CI: 0.283-0.989, p = 0.046).

Subgroup analyses indicated azvudine was particularly beneficial for patients initiating treatment more than five days after diagnosis (HR: 0.56, 95% CI: 0.39-0.78) and those with primary malignant tumors (HR: 0.33, 95% CI: 0.20-0.54).

Safety evaluations indicated fewer AEs with azvudine compared with Paxlovid, particularly for grades 1 and 2 AEs. These findings suggest azvudine may be a safer and more effective alternative to Paxlovid for certain hospitalized COVID-19 patients.

Conclusions

To summarize, this large-scale, multicenter, retrospective cohort study highlighted the effectiveness and safety of azvudine compared with Paxlovid in hospitalized COVID-19 patients. Among 37,606 patients analyzed, those receiving azvudine showed a lower risk of all-cause death and comparable rates of composite disease progression compared to Paxlovid.

Subgroup analyses revealed azvudine’s greater benefits for patients with malignant tumors, moderate disease, or delayed treatment initiation.

Additionally, azvudine demonstrated significant antitumor effects, suppressing hepatocellular carcinoma cell proliferation and enhancing immune responses.