I am not suggesting that you are lying or are being unreasonable, but I will suggest that what you are describing, while certainly possible, would be very rare. In fact, it would be so rare that people would be right to have questions and concerns about those performances, but not so unlikely that people should publicly suggest doping without hard evidence.
We all like to use ourselves as examples, and I can do the same thing you did. I didn't run at all in high school. I walked on in college and ran 40 mpw most weeks, eventually getting up to 50 mpw. I didn't run consistently in the summers, and often missed time during my 4 years of college due to injury. I was as undertrained as anyone when I set my PRs in college and in my first year of grad school. Those PRs were at about 90% age graded for an open athlete.
Fast forward to my masters career. I wanted to try to break 16:00 one more time as a runner, so I started getting serious about running. I was also eating better and drinking less than ever because my wife and I were starting our family. So I finally dropped down back under 150 pounds, was eating better, drinking less and, because I decided I would run a marathon before hanging up the flats, I was also running much more mileage than ever before.
I went on to one of the better stretches of masters road racing that anyone on these boards has had in recent years, with masters wins at some big races and posting some pretty good times. As it turns out, do you know where those times fit in on an age graded basis? Yep, you guessed it - right around the 90th percentile.
We all have our anecdotes, and mine is no more valid than yours, but I would say that my outcome (performing in line with prior times), is the more common outcome than the outcome you describe. It doesn't mean guys running PRs in their 40s are cheating, but it does mean that if they aren't cheating, they are outliers among outliers.