Should Remdesivir Be Used for the Treatment of Patients With COVID-19? Rapid, Living Practice Points From the American College of Physicians (Version 1).

Key Question 1

What are the effectiveness and harms of remdesivir in patients with coronavirus disease 2019 (COVID-19)?

Key Question 2

Do effectiveness and harms vary by symptom duration, disease severity, and treatment duration?

Background

Remdesivir, a broad-spectrum antiviral agent administered intravenously, was developed and studied as a potential treatment for Ebola virus disease and Marburg virus infection (1–3). In vitro and in vivo preclinical studies found antiviral activity for remdesivir against corona-like viruses, including Middle East respiratory syndrome coronavirus (4–6), severe acute respiratory syndrome coronavirus (SARS-CoV-1) (5), the circulating human coronaviruses HCoV-OC42 and HCoV-229E (7), and SARS-CoV-2 (8). Currently, the effectiveness of remdesivir is being tested as a treatment for patients infected with SARS-CoV-2 (COVID-19) and has been authorized for emergency use for treating COVID-19, by the U.S. Food and Drug Administration (9) in the United States, and in other countries (10–13).

The American College of Physicians (ACP) Scientific Medical Policy Committee (SMPC) based these rapid and living practice points (Table 1) on a systematic evidence review conducted by the U.S. Department of Veterans Affairs (VA) Evidence Synthesis Program in Minneapolis, Minnesota (14) (Appendix). This version of the practice points, based on a search completed on 3 June 2020 and updated through 31 August 2020, was approved by the ACP's Executive Committee of Board of Regents on behalf of the Board of Regents on 14 August 2020 and submitted to Annals of Internal Medicine on 13 August 2020. Because many studies are planned or under way, literature surveillance is ongoing, with updates currently planned for every 2 months through December 2021. The target audience for these practice points includes clinicians and the public. The target patient population includes all nonpregnant patients with COVID-19.

Table 1. Practice Points

Critical and important outcomes were determined by the evidence review team in collaboration with methodological and content experts. The magnitude of the effect (such as little or no, slight, modest, or large) for critical and important outcomes was determined by applying thresholds prespecified by the evidence review team (Table 2). Table 3 presents clinical considerations, the Figure and Tables 4 and 5 summarize current evidence, and Table 6 identifies additional evidence gaps. Appendix Tables 1 and 2 present the data estimates supporting the practice points.

Table 2. Thresholds for Determining Magnitude of Effect*

Table 3. Clinical Considerations

Figure.

Evidence description.

The evidence search and assessment were conducted by the U.S. Department of Veterans Affairs Evidence Synthesis Program, Minneapolis, Minnesota (14). Current search for evidence, completed on 3 June 2020, aimed to identify RCTs evaluating remdesivir for treatment of patients with COVID-19. COVID-19 = coronavirus disease 2019; ECMO = extracorporeal membrane oxygenation; RCT = randomized controlled trial.

* Patients requiring mechanical ventilation or ECMO were excluded from 1 RCT (17); therefore, despite a few patients (3.3%) developing a requirement for invasive mechanical ventilation between screening and the beginning of the treatment, this study is analyzed as being representative of patients with severe disease not requiring mechanical ventilation or ECMO at baseline.

† Within the evidence reviewed, severe COVID-19 is defined as hospitalized patients meeting 1 or more of the following criteria: radiographic infiltrates on imaging, an oxygen saturation level ≤94% on room air, tachypnea (respiratory rate >24 breaths per minute without supplemental oxygen), or need for supplemental oxygen or mechanical ventilation; moderate COVID-19 is defined as hospitalized patients with radiographic infiltrates and oxygen saturation greater than 94% on room air; and mild COVID-19 was not defined (14).

‡ Most (88.7%) of the participants enrolled in 1 RCT (16) had severe disease, so this study is analyzed as being representative of patients with severe disease.

Table 4. Evidence Summary for Patients with Moderate* COVID-19: What Information Does the Evidence Provide?

Table 5. Evidence Summary for Patients with Severe* COVID-19: What Information Does the Evidence Provide?

Table 6. Evidence Gaps

Appendix Table 1. Estimates: Patients With Moderate* COVID-19†

Appendix Table 2. Estimates: Patients With Severe* COVID-19†

Rationale

Use of Remdesivir in Patients With Moderate COVID-19

Table 4 summarizes the current evidence on the use of remdesivir in patients with moderate COVID-19.

Overall, the current evidence points toward a net benefit for remdesivir in patients with moderate COVID-19 and suggests that a shorter treatment period (5 days) is as effective as a longer one (10 days), with no increase in harms (16). Low-certainty evidence shows that the 5-day course may be superior for mortality, recovery, and clinical improvement; however, low-certainty evidence also shows improvement for several outcomes when comparing the 10-day course to placebo. Thus, the SMPC believes that it is reasonable to consider extending treatment to 10 days for patients whose condition does not improve during the initial 5 days. Because the overall certainty of evidence is low across the comparisons, the SMPC has flagged course duration as a particular area of interest for further discussion and close monitoring.

Evidence from 1 randomized controlled trial (RCT) (16) compared a 5- or 10-day course of remdesivir with standard care, although “standard care” was not defined. Among outcomes rated as critical, remdesivir (5- or 10-day course) may reduce mortality slightly and result in slightly fewer serious adverse events compared with standard care (low certainty). Evidence also showed a modest increase in recovery and clinical improvement with a 5-day course, and slight increases in recovery and clinical improvement with a 10-day course, compared with standard care (low certainty). A 5-day course may also reduce time to recovery slightly (low certainty), but evidence is insufficient to make any conclusions about a 10-day course. Both courses of remdesivir (5- and 10-day) may slightly reduce the need for invasive mechanical ventilation or extracorporeal membrane oxygenation (ECMO) (low certainty). However, the occurrence of any adverse events may increase with a 5-day course (slight effect) and with a 10-day course (modest effect) compared with standard care (low certainty).

Evidence comparing a 5- versus 10-day course of remdesivir (16) showed that a 5-day course may reduce mortality slightly and may increase recovery (modest effect) and clinical improvement (slight effect) compared with a 10-day course (low certainty). However, evidence showed little to no difference between the 2 courses in reducing the need for invasive mechanical ventilation or ECMO (low certainty), and evidence is insufficient to show a difference in time to recovery. Evidence for potential harms showed that a 5-day course may result in fewer adverse events (any) compared with a 10-day course (modest effect), although the shorter course may not result in fewer serious adverse events (low certainty).

No evidence was found for any effect on other critical outcomes (hospital length of stay) or important outcomes (time to clinical improvement, nonserious adverse events) with either course in patients with moderate COVID-19. No evidence was identified as to whether outcomes vary by symptom duration in patients with moderate COVID-19.

Use of Remdesivir in Patients With Severe COVID-19

Table 5 summarizes the current evidence on the use of remdesivir in patients with severe COVID-19 (15, 17, 18).

Overall, the current evidence points toward a net benefit for a 10-day course of remdesivir in patients with severe COVID-19 (including those requiring mechanical ventilation or ECMO at baseline) compared with placebo (15, 18). No evidence was found comparing a 5-day course of remdesivir with placebo or standard care in patients with severe COVID-19. In the absence of this direct evidence, the SMPC looked at the indirect evidence that a 5-day course is as effective as a 10-day course of remdesivir with the same, or probably fewer, potential harms in patients with severe COVID-19 not requiring mechanical ventilation or ECMO at baseline (17). In addition, the compliance data showed that a 10-day course (10 doses) was used in 40.8% of patients with severe COVID-19 (including those requiring mechanical ventilation or ECMO at baseline), and 38.1% received fewer than 10 doses because they recovered and were discharged from the hospital (15). However, for a subgroup of patients with severe COVID-19 receiving mechanical ventilation or ECMO at day 5, extending treatment to 10 days may be beneficial compared with discontinuing treatment on day 5 (17).

Evidence from 2 RCTs (15, 18) showed that among outcomes rated as critical, a 10-day course of remdesivir compared with placebo may slightly reduce mortality (low certainty) and probably increases recovery by a large effect (moderate certainty), and that there are probably fewer serious adverse events (modest effect, moderate certainty). Evidence from 1 RCT showed that a 10-day course may not reduce hospital length of stay (low certainty) (18). Low-certainty evidence also showed improvement with a 10-day course compared with placebo for the following important outcomes: time to recovery (large effect), clinical improvement (modest effect), time to clinical improvement (slight effect), the need for mechanical ventilation or ECMO (slight effect), and nonserious adverse events (slight effect). Evidence was insufficient regarding differences in any adverse events.

For a 10-day course of remdesivir compared with placebo, the outcomes of mortality (18), time to recovery (15), and time to clinical improvement (18) did not vary by symptom duration (≤10 days vs. >10 days), and time to recovery also did not vary by baseline oxygenation or ventilation requirements (15). No evidence was found on whether other outcomes vary by symptom duration.

Evidence from 1 RCT (17) that compared a 5-day course with a 10-day course of remdesivir showed that the 5-day course may reduce mortality slightly versus the 10-day course in patients with severe COVID-19 who did not require mechanical ventilation or ECMO at baseline (17). However, a post hoc analysis suggested that a 5- versus a 10-day course might result in a large increase in mortality among the most critical patients of those with severe COVID-19 (those receiving mechanical ventilation or ECMO at day 5) (17). Treatment beyond 5 days did not improve mortality among patients who were receiving noninvasive positive pressure ventilation or high-flow oxygen, those receiving low-flow oxygen, or those breathing ambient air. This finding suggests that extending treatment to 10 days for patients receiving mechanical ventilation or ECMO at day 5 may be beneficial (17). Compared with a 10-day course, a 5-day course shows a modest increase in recovery, a slight decrease in the time to recovery, and a modest reduction in the need for mechanical ventilation or ECMO (low certainty). Evidence for potential harms showed that a 5-day course of remdesivir results in fewer serious adverse events (large effect, low certainty) and a fewer number of any adverse events (slight effect, low certainty) compared with a 10-day course.