The AH-64 Apache attack helicopter first flew as a prototype in 1975. In 1976, it was selected by the U.S. Army as the successor to the AH-1 Cobra, and began full production in 1982, entering military service in 1986. Almost 50 years after its first flight, it remains in production, with more than 2400 built, and is expected to continue in production until at least 2026. With successive consolidations in the U.S. defense industry, the Apache, originally designed and manufactured by Hughes Helicopters, was subsequently built by McDonnell Douglas and now Boeing.
The U.S. Army operates a fleet of 1891 Apaches, and the design (in some cases locally produced under license) is used by the armed forces of Egypt, Greece, India, Indonesia, Israel, Japan, Kuwait, Morocco, Netherlands, Qatar, Saudi Arabia, Singapore, South Korea, Taiwan, United Arab Emirates, and the United Kingdom.
My understanding (which dates from the late 1980s, when I visited the Comanche crew systems lab at NASA Ames) was that the Comanche was not considered a replacement for the Apache, but as a reconnaissance and light attack helicopter which would designate targets for the Apache. The protracted development program (selection in 1991, first flight in 1996, cancellation in 2004, at which point US$ 7 billion had been spent) was probably reason enough to pull the plug.
After the Comanche was cancelled, the Army started another light helicopter program, the Bell ARH-70, but that too was cancelled in 2008.
Two words: mission creep. In the late 90s and early 2000s it was turning into a semi-stealth Apache with detachable wings for external ordnance.
That created accounting problems by exacerbating apparent cost overruns. Imagine some brilliant guy looked at the F-15A prototype, envisioned the F-15E, and then (instead of waiting for production to go long enough to account for all basic program costs) demanded that they enter service as the F-15E. the delays and apparent overruns would have created apparent reasons to kill the program.