[BostonCard]

Thanks for finding it.  This was the study referred to in the STAT article linked by the OP.  As noted, the study had to close earlier than planned due to the lack of cases.  I agree with dabigv, that remdesivir is not a miracle cure, but that a negative, underpowered study doesn't exclude a benefit.  Also, there are some hints it might work in patients given the drug early:

Although not statistically significant, patients receiving remdesivir had a numerically faster time to clinical improvement than those receiving placebo among patients with symptom duration of 10 days or less (hazard ratio 1·52 [0·95–2·43]).

(from the abstract)

Although not statistically significant, in patients receiving remdesivir or placebo within 10 days of symptom onset in the ITT population, those receiving remdesivir had a numerically faster time to clinical improvement than those receiving placebo (median 18·0 days [IQR 12·0–28·0] vs 23·0 days [15·0–28·0]; HR 1·52 [0·95–2·43]; appendix p 6).

(from the discussion)

This has some biological plausibility, since an antiviral would expect to have its maximum benefit if instituted early.  And although not statistically significant, if it is replicated in another study, a decrease of time to clinical improvement from 23 days to 18 days would be fairly clinically meaningful.

We should all wait until the NIAID study to be reported.

BC

[BostonCard]

Thanks for finding it.  This was the study referred to in the STAT article linked by the OP.  As noted, the study had to close earlier than planned due to the lack of cases.  I agree with dabigv, that remdesivir is not a miracle cure, but that a negative, underpowered study doesn't exclude a benefit.  Also, there are some hints it might work in patients given the drug early:

Although not statistically significant, patients receiving remdesivir had a numerically faster time to clinical improvement than those receiving placebo among patients with symptom duration of 10 days or less (hazard ratio 1·52 [0·95–2·43]).

(from the abstract)

Although not statistically significant, in patients receiving remdesivir or placebo within 10 days of symptom onset in the ITT population, those receiving remdesivir had a numerically faster time to clinical improvement than those receiving placebo (median 18·0 days [IQR 12·0–28·0] vs 23·0 days [15·0–28·0]; HR 1·52 [0·95–2·43]; appendix p 6).

(from the discussion)

This has some biological plausibility, since an antiviral would expect to have its maximum benefit if instituted early.  And although not statistically significant, if it is replicated in another study, a decrease of time to clinical improvement from 23 days to 18 days would be fairly clinically meaningful.

We should all wait until the NIAID study to be reported.

BC