Safety and Effectiveness of Molnupiravir (LAGEVRIO^®) Capsules in Japanese Patients with COVID-19: Interim Report of Post-marketing Surveillance in Japan

This PMS was initiated on December 27 2021 and it is planned to enroll 3000 patients with “infection caused by SARS-CoV-2” that are treated with molnupiravir, in principle within 5 days of COVID-19 diagnosis, in the 3-year period to December 2024. This interim analysis includes data collected from December 27 2021 to June 15 2022 at 151 enrolled medical institutions. Of the enrolled medical institutions, survey forms were collected from 120 institutions. This paper only contains data on patients from whom informed consent was obtained.

In this survey, it was planned to enroll a population of Japanese patients with “infection caused by SARS-CoV-2,” with risk factors for illness progression. These risk factors included age ≥ 65 years, malignant tumor, chronic obstructive pulmonary disease (COPD), chronic kidney disease, type 2 diabetes mellitus, hypertension, dyslipidemia, immunodeficiency after solid-organ transplantation, obesity [body mass index (BMI) ≥ 30 kg/m²], smoking, and others. Patients were enrolled regardless of outpatient or inpatient status, vaccination status, or history of medical complications. The inpatient population included cases of onset during hospitalization for reasons other than COVID-19 infection; and cases hospitalized for COVID-19 isolation, treatment, etc. Some patients were classified as others and included nursing home residents and patients receiving home-visit medical care, etc. In principle, patients were administered molnupiravir within 5 days of SARS-CoV-2 infection. Data was collected in a single-arm observational design by the continuous registration method. To reduce selection bias, patients who received molnupiravir were consecutively enrolled without selecting cases, and as such did include patients without risk factors for disease progression. The observation period was through 29 days from the start of molnupiravir administration.

This PMS was conducted in accordance with the Ministry of Health, Labour, and Welfare (MHLW) ordinance for Good Post-marketing Study Practice (GPSP). The survey protocol was approved by the Pharmaceuticals and Medical Devices Agency (PMDA). Additionally, this survey was conducted in accordance with the protocol; the Helsinki Declaration of 1964, and its later amendments; and all other applicable regulations. Institutional review board (IRB) approval was not mandatory for this survey according to the GPSP.

The physician in charge of the survey explained to the patient or his/her legally acceptable representative the objective of the survey, the information to be collected through the survey, and the uses of the survey results, and obtained informed consent from the patient or his/her legally acceptable representative.

The safety analysis set included all patients for whom the survey form was collected, and excluded patients that were not administrated molnupiravir.

The effectiveness analysis set included all patients in the safety analysis set and excluded the following patients: outside the dosage and administration instructions, and effectiveness not evaluable due to loss to follow-up, etc.

As shown in Fig. 1, of 1236 enrolled patients, a total of 1031 patients were included in the safety analysis set [case report forms (CRFs) were not collected from 202 patients; and three patients were excluded from the safety analysis set as molnupiravir was not administered]. Whilst in principle, patients were administered molnupiravir within 5 days of SARS-CoV-2 infection, 41 patients (3.98%) were enrolled 6 days or more after infection.

Of the 1031 patients included in the safety analysis, 884 patients were included in the effectiveness analysis set, and 147 patients were excluded for the following reasons: outside the dosage and administration instructions in 18 patients, effectiveness not evaluable in 133 patients, and four patients were excluded as they met both exclusion criteria.

As shown in Fig. 2 and Table 1 the safety analysis set comprised predominantly older aged patients. The median age of patients was 68.0 years (range: 18–107 years) [Table 1]: 57.23% of the population was ≥ 65 years and 45.30% were aged 70 years or older (Fig. 2). As presented in Table 1, 53.06% of patients were male.

In the safety analysis set, 689 (66.83%) were outpatients, 294 (28.52%) were inpatients, and 48 (4.66%) were other patients.

The reason for the use of the molnupiravir, in all 1031 patients, was infection due to SARS-CoV-2 as judged by a physician, and most patients had COVID-19 symptoms at the start of treatment (97.28%).

At the start of administration of the drug, 862 (83.61%) patients had received at least one vaccine dose against SARS-CoV-2, and 657 (63.72%) patients were vaccinated twice.

The severity of COVID-19, before molnupiravir treatment based on the clinical management of patients with COVID-19 (Supplementary material: Table S1) was mild in 87.88%, moderate I in 9.60%, and moderate II in 2.23% of the patients.

At the start of molnupiravir treatment (Table 2), 96.80% patients had risk factors for severe COVID-19 illness. The most frequent risk factor for progression to severe COVID-19 (Table 2) was “elderly ≥ 65 years,” which was recorded in 590 patients (57.23%). The second most observed risk factor was hypertension, followed by smoking. Three hundred twenty-three (31.33%) patients had one risk factor; 330 (32.01%) patients had two risk factors, 210 (20.37%) patients had three risk factors, 93 (9.02%) patients had four risk factors, and 42 (4.07%) patients had five risk factors or more. In the analysis populations, however, 33 (3.20%) patients without such risk factors were also included.

At the start of molnupiravir treatment (Table 1), 863 patients (83.71%) had comorbidities (including diseases with risk factor for severe COVID-19 illness) and 153 patients (14.84%) had no underlying comorbid diseases. The most common comorbidities were hypertension (45.00%), type 2 diabetes mellitus/diabetes mellitus (22.31%), and dyslipidemia (20.17%).

Regarding treatment practice, molnupiravir was started at a median of 1–2 days from COVID-19 symptom onset. For most patients (93.11%), molnupiravir was started within 5 days. For approximately half of the patients (50.53%), molnupiravir was started 1–2 days from the symptom onset. The median time from positive SARS-CoV-2 virus test results to molnupiravir administration was 1 day (range: 1–9 days). After molnupiravir treatment started, 96.61% of patients completed treatment.

Of 1031 patients in the safety analysis population, the frequency of ADRs was 78 events in 68 patients (6.60%). A list of all ADRs observed in this survey is presented in Table S2 (Supplementary Material). Among the ADRs that occurred, four events in four patients (0.39%) were serious ADRs. Furthermore, the ADR rate observed in patients aged 65 years and above was 5.25% (31/590).

ADRs that occurred in three or more patients were diarrhoea in 26 patients (2.52%), rash in six patients (0.58%), dizziness in five patients (0.48%), faeces soft in four patients (0.39%), and vomiting and headache in three patients each (0.29%) (Table 3).

Most ADRs occurred soon after molnupiravir administration, and 34 ADRs were observed in 32 patients, 1–2 days after the start of administration.

A total of 35 patients (3.39%) in the safety analysis set discontinued molnupiravir administration. Twenty-one patients discontinued molnupiravir due to the onset of AEs, two patients discontinued administration due to lack of effectiveness, and 12 patients for other reasons (patient request, patient lost to follow-up, etc.). Among the 21 patients that discontinued the administration due to AEs, a total of 16 patients experienced 19 ADRs. The ADRs that led to discontinuations were as follows: diarrhea (four cases), rash (three cases), vomiting (two cases), drug eruption (two cases), and other events (one case each).

ADRs were assessed by physicians and serious ADRs were observed in four patients (four events) including rash, hepatic function abnormality, COVID-19 (aggravation), and interstitial pneumonia (aggravation) observed in one patient each.

There were seven deaths (Table 4) [no outpatients, three inpatients (one with and two without oxygen or mechanical ventilation at the start of molnupiravir administration), and four other patients in nursing homes, etc.]. The causal relationships of all deaths were judged by physicians not to be related to molnupiravir. Six of the recorded deaths were observed in patients aged 80 years or greater, and the age of one patient that died was unknown. All cases of death had risk factors for severe COVID-19 illness. The causes of death were as follows: COVID-19 (including complications related to COVID-19 and death due to associated symptom) in three patients; and urinary tract infection/sepsis, diabetic ketoacidosis, senescence, and unknown (with comorbidities of chronic kidney disease and high blood pressure, and a recorded AE term of respiratory disorder) in one patient each.

Of the 1031 patients included in the safety analysis, 884 patients were included in the effectiveness analysis set. Among these, 612 (69.23%) were outpatients, 227 (25.68%) were inpatients, and 45 (5.09%) were classified as other. In the effectiveness analysis set, outpatients (612 patients, 69.23%) and inpatients without oxygen or mechanical ventilation at the start of molnupiravir administration (199 patients, 22.51%) were assessed for the frequency of the composite effectiveness endpoint (hospitalization or death, and initiation of oxygen/mechanical ventilation or death, respectively).

Among outpatients, 16/612 patients (2.61%) were hospitalized, for all causes, through 29 days from the start of administration (Fig. 3a). As presented in Table 5, among outpatients 12/16 were over 60 years old; 14/16 patients had presence of comorbidities including risk factors for severe COVID-19 illness; 10/13 (missing data for three patients) patients had received concomitant medications for COVID-19; and 16/16 had risk factors for severe COVID-19 illness. Twelve of the 16 patients (1.96%) had COVID-19-related hospitalization, and no deaths were reported in outpatients.

Among inpatients (without oxygen at start of molnupiravir administration), 9/199 patients (4.52%) started oxygen administration through 29 days from the start of administration (Fig. 3b). No patient was placed on mechanical ventilation (including ECMO). Two inpatients died [2/199 (1.01%)] within 29 days from the start of administration. For inpatients the composite endpoint [death, start of oxygen, or mechanical ventilation (including ECMO)] frequency was 5.03%.

The use of oxygen administration in the effectiveness analysis set is presented in Table S3 (Supplementary Material). Of the 884 patients in the effectiveness analysis set, 12 patients (1.36%) initiated oxygen therapy after molnupiravir administration, including nine inpatients, one outpatient, and two other categorized patients. The median volume of oxygen supply was 2 L/min (range: 1–15 L/min) in ten patients, which excludes two patients who had unknown oxygen supply volumes. A total of 15 L/min of oxygen mask with reservoir was supplied to the fatal case with urinary tract infection/sepsis.

The frequency of all-cause death was 0.79% (7/884) in the effectiveness analysis population, and the frequency of death caused by COVID-19 was 0.34% (3/884).

In the effectiveness analysis set, 84.05% had comorbidities and the vaccination rate was 83.37%. Additionally, as characteristic of the aging Japanese population, 65.16% of the analysis set were aged 60 years or older, and a further 43.44% of patients were aged 70 years or older.

This interim analysis is the first report of the safety and effectiveness of molnupiravir in over 1000 patients, reflecting a real-world setting in Japan. It is significant in that it provides detailed large-scale information on the background of patients prescribed molnupiravir in Japan, as well as the actual administration status, safety, and to a limited extent, effectiveness.

This PMS included approximately 84% of patients who had comorbidities, 84% who were vaccinated, and 45% aged 70 years or older, reflecting the aging society of Japan. As defined in large-scale surveys, age and comorbidities were reported to increase the probability of developing severe COVID-19 illness [14]. Older people generally have multiple age-related comorbidities, and such patients are expected to be at a high risk of developing severe COVID-19 illness.

This survey was conducted while Omicron was the dominant SARS-CoV-2 variant, and as vaccine availability was widespread, the majority of the survey population had been vaccinated. In the period when Omicron was the dominant SARS-CoV-2 variant compared with the period when Delta was dominant, the Government Advisory Board data [15] suggested a decrease in deaths due to respiratory failure caused by severe COVID-19 pneumonia. Additionally in the periods January–February 2022 and March–April 2022, during the time period that data was collected in this PMS, the MHLW reported [16, 17] in certain prefectures of Japan that the death rate was higher in older age patients. In these data, regardless of inpatient or outpatient status, the death rate was 0.01% in patients aged 50 years or younger and 1.99% in patients aged over 60 years (January–February 2022); and 0.01% in patients aged 50 years or younger, and 1.13% in those aged over 60 years (March–April 2022). Additionally, by age group, the death rates (%) were 0.29/0.10, 1.23/0.94, 3.67/2.67, and 6.21/4.05 (January–February/March–April) in patients aged in their 60s, 70s, 80s, or 90s or over, respectively. This demonstrates the importance of focusing on the care of older age patients and patients with comorbidities.

Considering the high proportion of patients aged 65 years or greater in the survey population, the low number of deaths is particularly noteworthy. Seven deaths [0.79% (7/884)] were observed, which were all unrelated to molnupiravir, as presented in Table 4.

In the safety analysis set, the frequency of reported ADRs was low. Additionally, molnupiravir was well tolerated by patients aged 65 years and older. While the types of ADRs to molnupiravir observed in this survey are similar to those in the phase III MOVe-OUT randomized clinical trial (RCT) (diarrhoea, nausea, dizziness, etc.) the incidences of these ADRs were lower (apart from diarrhea) than in the MOVe-OUT study [12].

Regarding treatment adherence, after molnupiravir treatment started, 96.61% of patients completed treatment, suggesting that there are minimal barriers for patients to complete administration of the treatment formulation and that molnupiravir was well tolerated.

Among outpatients in the effectiveness analysis set, a small percentage of patients (2.61%) were hospitalized, and no deaths were reported. These results support the effectiveness of molnupiravir in this survey. Whilst not directly comparable, the efficacy of molnupiravir was reported in the MOVe-OUT RCT (molnupiravir administered to outpatients with mild or moderate COVID-19 symptoms), in which 7.3% of patients were hospitalized or died through 29 days.

Additionally, recent data has been reported from several real-world clinical studies conducted in various countries including Japan [18][Suzuki et al.], Hong Kong [19], Israel [20] [Arbel et al. preprint], Italy [21], and a clinical trial in the UK [22] [Butler et al.]. In common with our PMS, many of these studies were retrospective cohort studies that included vaccinated patients with mild and moderate disease, and were conducted when the predominant SARS-CoV-2 viral strain was Omicron. Wong et al. reported that in an outpatient population during the Omicron wave, the crude incidence rate of all-cause mortality was 4.2 per 100,000 person-days lower in the molnupiravir group (17.9) compared with matched controls (22.1) [19]. Arbel et al. reported that in high-risk outpatients infected when the Omicron SARS-CoV-2 viral strain was predominant, individuals aged 65 years and older treated with molnupiravir experienced a significant reduction in hospitalization and mortality [20] [Arbel et al. preprint]. The present results cannot be directly compared with these previous studies, as the characteristics of patients included in our PMS and the conditions of molnupiravir use differ from the clinical trial and overseas real-world clinical studies; however, evidence supporting the safety and effectiveness of molnupiravir is increasing.

The UK-based PANORAMIC study was a national, multigroup, prospective, platform adaptive randomized clinical study. In the two arms of this study, usual care and 800 mg molnupiravir was administered twice daily for 5 days, or usual care only was provided. In the preliminary report of the PANORAMIC study, Butler et al. [22] presented secondary endpoint results of time-to-first-recovery (patient-reported outcomes), in addition to the primary endpoint results of death and hospitalization. The patient-reported outcomes and symptom data will be important in assessing the clinical effectiveness of molnupiravir in the future.

There are some limitations to be considered. This survey is not a RCT and has no control group, and therefore, the results cannot compare the safety or effectiveness outcomes with those who did not receive any treatment.

In general, conducting long-term examinations in outpatients is difficult due to the high rate of loss to follow-up. In this context, investigating patient follow-up using the continuous registration method implemented in this survey provides extremely important safety and effectiveness data on patients that have been prescribed therapeutics, despite the fact that there is no control group as in randomized clinical trials. In the effectiveness analysis set, approximately 15% of patients were lost to follow up; however, in this survey, follow-up was limited because this was a noninterventional observational survey in an outpatient setting under conditions of real-world clinical use.

The number of sites that participated in this survey is limited compared with the number of sites in Japan, and there are, therefore, limitations in the evaluation of these results.

Finally, as SARS-CoV-2 strains transition sequentially from prevalent strains at the time of the survey, the results of this survey should be interpreted from the viewpoint that during this survey Omicron was the main SARS-CoV-2 variant.