全面提供抗病毒藥物治療預期成效：數理模式研究

Abstract

背景與研究目標：由於 SARS-CoV-2 Omicron BA.5 variants 與後續新 Omicron 變異株「免疫逃脫」的特性，導致 COVID-19 疫情至今仍難以被清除。我國防疫政策目前已全面調整為輕症免通報免隔離，將防疫工作著重於減少額外的重症數及死亡發生。我國過去主要推行疫苗接種政策以預防確診者發展為重症，然經過數波疫情後，過去接種疫苗的保護力已經逐漸衰退，民眾接種疫苗的意願也不如從前。口服抗病毒藥物是另一有效降低重症風險的方式，然而，目前公費口服抗病毒藥物的開立嚴格限制在具重症危險因子者，認證過程繁複，過去許多確診者因不清楚自己是否符合領用條件而延誤就醫，錯過黃金治療時間。如能提供所有通報之確診者抗病毒藥物，應能提升民眾通報意願，使原先符合領用條件的病人均能拿到口服抗病毒藥物，也降低其餘通報確診者的重症風險，有望降低整體重症及死亡率，但目前尚缺乏具有實證數據佐證的數理模式研究。本研究擬依據臺灣 2022 年至 2023 年的四波 COVID-19疫情流行數據，建立 COVID-19 傳播數理模型，以分析對所有 COVID-19 確診者全面提供抗病毒藥物治療的預期成效，提供我國防治政策參考。

方法：本研究分為系統性文獻回顧及數理建模兩部分。系統性文獻回顧以特定關鍵字探索相關文獻，透過統整實證研究資料，設定合乎2023年新Omicron變異株傳播特性的數理模型。數理建模部分擬建立 SARS-CoV-2 Omicron BA.5 變異株及後續新Omicron 變異株之 SEIRS 傳播數理模型。模型考慮自然免疫、疫苗免疫、及綜合免疫的免疫衰退，以四個單純混合模型（Homogeneous-mixing model）分別擬合臺灣自 2022/4/1 - 2023/6/25 的四波（BA.2、BA.5、BQ.1、XBB） COVID-19 疫情。以此為基礎，預測臺灣 2023 年下半年疫情走勢，並評估全面提供 COVID- 19 通報確診者抗病毒藥物相較現今給藥策略之效益，主分析評估可直接或間接預防的 COVID-19 重症病例數及死亡人數，次分析則評估可預防的新發 Long-COVID 個案數，敏感度分析考慮可能新出現的 SARS-CoV-2 變異株，模擬疫苗保護力存在變異的情境，另外也考慮新變異株可能具備較強的免疫逃脫能力的情境，以評估不同情境下全面提供抗病毒藥物的效益。

結果：本研究預測在現今給藥政策下，2023 年全年將有 61,518 例 COVID-19 確診者發展為重症，於 2023 年下半年平均每日新發約 190 例重症。全面提供口服抗病毒藥物的介入後，可望大幅降低 2023 年重症數至 23,406 例，2023 年下半年平均每日重症數降至約 70 例。全面提供口服抗病毒藥物相較於對照情境，可望降低 61.9 % 重症及 62.4 % 死亡。此外，本研究估計 2023 年全年將有 12,607,188 例確診者於康復後仍有 COVID-19 相關症狀。在全面提供口服抗病毒藥物的介入情境下，可降低約 29.5 % 的新發 Long-COVID 個案，使 2023 年的預期新發 Long-COVID 數降至 8,886,132 例。敏感度分析顯示，若疫苗對於新變異株之保護力存在變異（± 10 %），全面提供抗病毒藥物仍可提供與原先相似的保護力。此外，若新變異株具有強大免疫逃脫能力，則全面提供抗病毒藥物對重症的保護力將顯著提升，能大幅下降因大量新發感染引起的重症、死亡、及 Long-COVID 。成本效益分析顯示，若全面提供抗病毒藥物的政策下 Paxlovid 的單位藥價降至 5.3 美元，則每減少一例重症所需花費為 3,983 美元。

結論：對每位 COVID-19 確診者全面提供口服抗病毒藥物治療可望大幅度降低染疫後重症、死亡、及罹患 Long-COVID 的人數。

Background and Research Objective: Due to the "immune escape" characteristics of the SARS-CoV-2 Omicron BA.5 variants and subsequent new Omicron variants, the COVID-19 pandemic remains difficult to eradicate. COVID-19 prevention policy in Taiwan has currently been adjusted to focus on reducing the number of severe cases and deaths. In the past, Taiwan mainly implemented vaccination policies to prevent COVID-19 cases from developing into severe cases. However, after several waves of outbreaks, the protection provided by past vaccinations has gradually waned, and the public's willingness to get vaccinated has decreased. Oral antiviral drugs represent another effective way to reduce the risk of severe cases. However, currently, the publicly funded prescription of oral antiviral drugs is strictly limited to individuals with severe risk factors, and the certification process is complex. Many confirmed cases in the past have delayed seeking medical attention because they were unsure if they met the eligibility criteria for receiving oral antiviral drugs, missing the optimal treatment time. If all reported confirmed cases could receive antiviral drugs, it should increase the willingness of the public to notify and ensure that patients who meet the criteria can receive oral antiviral drugs, and further reducing the risk of severe cases in other reported confirmed cases. However, currently, there is a lack of evidence-based mathematical models to support this hypothesis. This study aims to establish a mathematical model for COVID-19 transmission based on the epidemic data from Taiwan from 2022 to 2023. The model will be used to analyze the expected effectiveness of providing antiviral drug treatment to all confirmed COVID-19 cases, and providing valuable insights for our country's epidemic prevention and treatment policies.

Methods: This study consist of two parts: a systematic literature review and mathematical modeling. The systematic literature review will use specific keywords to explore relevant literature and integrate empirical research data to establish a mathematical model that aligns with the transmission characteristics of the new Omicron variant in 2023. In the mathematical modeling part, a SEIRS transmission model will be established for both the SARS-CoV-2 Omicron BA.5 variant and subsequent new Omicron variants. The model will consider the decay of immunity from natural infection, vaccination, and hybrid immunity. Four simple homogeneous-mixing models will be used to fit the four waves (BA.2, BA.5, BQ.1, XBB) of COVID-19 epidemics in Taiwan from April 1, 2022, to June 25, 2023. Based on this, the study will predict the trend of the epidemic in the latter half of 2023 and evaluate the benefits of universal antiviral treatment compared to the current treatment strategy. The primary analysis will assess the number of severe COVID-19 cases and deaths that can be directly or indirectly prevented. The secondary analysis will evaluate the number of new Long-COVID cases that can be prevented. Sensitivity analysis will consider the possibility of new SARS-CoV-2 variants emerging, simulate scenarios with varying vaccine efficacy, and also consider scenarios where new variants may possess stronger immune escape capabilities.

Results: This study predicts that under the current medication policy, there will be 61,518 COVID-19 cases developing into severe cases throughout the year 2023, with an average of approximately 190 new severe cases per day in the second half of 2023. However, it is expected that the number of severe cases in 2023 will significantly decrease to 23,406 under the scenario of universal antiviral treatment, with an average of around 70 new severe cases per day in the second half of 2023. Sensitivity analysis indicates that even if there are variations (± 10%) in the effectiveness of vaccines against new variants, the universal antiviral treatment would still provide similar protection as originally estimated. Furthermore, if new variants exhibit strong immune evasion capabilities, the effectiveness of universal antiviral treatment against severe cases would significantly increase, leading to substantial reductions in severe cases, deaths, and Long-COVID cases caused by a large number of new infections. Cost-benefit analysis shows that if the unit price of Paxlovid is reduced to 5.3 USD under the policy of comprehensive antiviral drug provision, the cost required to prevent one severe case would be 3,983 USD.

Conclusion: Providing oral antiviral drug treatment to each confirmed COVID-19 case is expected to significantly reduce the number of severe cases, deaths, and Long-COVID cases after infection.