A curious but fortunate characteristic of virus epidemics is their limited life spans. No one knows why, but guesses include herd immunity and mutations of the virus.
The following graph from the Centers for Disease Control and the National Center for Health Statistics shows the time profile of the COVID-19 weekly death counts from February onward. (For an interactive version of the graph, go here.)
In the U.S., the virus got underway in March. For the week ending March 14, the total number of deaths nationwide was 52. During the following month, the number of deaths increased rapidly, peaking in the week ending April 18 at a count of 17,026. Contagion Myth: Why Vi... Buy New $22.49 (as of 04:23 EDT - Details)
From that time onward, the death count declined rapidly to a weekly number of 3,684 in late June. A second “wave” began in July. The peak of that second wave was 6,794 deaths during the week ending July 25. After that, a steeper decline commenced and accelerated.
The peak death count for Americans under age 25 was 28 (for the week ending April 11) and has been under that number since. Only a single death occurred in that age group during the latest reported week, and there were no deaths recorded in the 25–34 age group.
Virus epidemics behave differently from virtually all other diseases. If you graphed timelines of the number of cancer deaths, fatal heart attacks, and fatal strokes, those timelines would be virtually flat.