Dear Editor,

We appreciate Dr. Hagiya’s detailed observation on our original article [1, 2]. We have responded to his Letter to the Editor as follows.

He pointed out that our defined evaluation period of Day 2–28 was too long for assessing hospitalization caused by COVID-19, because the clinical course of progressive pneumonia in COVID-19 patients is usually observed within several days to approximately one week from SARS-CoV-2 infection. We agree with his comment that early suppression of COVID-19 symptoms is important. However, for the following reasons, we have concluded that our evaluation period of Day 2–28 after COVID-19 diagnosis was appropriate:

  • While Dr. Hagiya focused on progressive pneumonia as an indicator of severe COVID-19, our primary endpoint was the all-cause hospitalization ratio in patients at high-risk for severe COVID-19, regardless of pneumonia. Our definition included not only progressive pneumonia, but also worsening of underlying disease due to COVID-19.

  • The period of 28 days after antiviral treatment is commonly used for hospitalization rate evaluation, as seen in other studies [3,4,5] cited in the Introduction of our manuscript.

To address Dr. Hagiya’s concern, we performed a sub-analysis by stratifying the data into different periods after COVID-19 diagnosis and treatment as ad-hoc sensitivity analyses.

Our analysis confirmed that the inverse probability of treatment weighting (IPTW) adjusted hospitalization frequency was lower in the ensitrelvir treatment group across all time periods. The risk ratios for up to 7 and Day 10 were 0.340 and 0.459, respectively, indicating a significantly lower frequency of hospitalization in the ensitrelvir group (Table 1). These findings support the early effect in reducing hospitalization by ensitrelvir described in the original manuscript. For the periods up to Day 14 and Day 21, the differences were not significant due to reduced detection power, but a similar risk reduction was observed as with the primary analysis at Day 28. These additional analyses have demonstrated robustness of the original results and support our conclusion on the effectiveness of ensitrelvir in reducing hospitalization.

Table 1 Frequency of hospitalization and risk ratio in ensitrelvir treatment group and no antiviral treatment group
Full size table

Furthermore, Dr. Hagiya mentioned the diverse clinical courses of COVID-19 patients based on their vaccination history, immune status, and virus variants [6]. To support part of this comment, we conducted additional subgroup analyses for each high-risk factor for severe COVID-19. Using the same method as in the main analysis, we evaluated the effect of ensitrelvir in reducing all-cause hospitalizations in each subgroup of high-risk factors. As shown in Table 2, the unadjusted hospitalization rate in the no antiviral group ranged from 0.727 to 2.470%, with no hospitalizations observed in the HIV/AIDS subgroup (0/179). The adjusted risk ratio ranged from 0.396 to 0.923, indicating a lower incidence of hospitalization in the ensitrelvir group compared with no antiviral group. In some subgroups, risk ratios could not be calculated due to the absence of hospitalizations in the ensitrelvir group. These results suggest that patients with high-risk factors form a heterogeneous population and include specific subpopulations that may benefit from more aggressive antiviral treatment, taking into account their specific backgrounds.

Table 2 Incidence and hospitalization risk in patients with high-risk factors in ensitrelvir treatment group and no antiviral treatment group
Full size table

We believe that further studies are needed to explore the COVID-19 patient backgrounds that are considered at higher risk of hospitalization and to identify the group for whom treatment with antiviral drugs is recommended.