Currently there are around 17 million bitcoins in circulation, with another 4 million still to be mined. You can see more detail here:
http://www.bitcoinblockhalf.com/ y
https://blockchain.info/es/charts/total-bitcoins?timespan=allEstimates place a stop to Bitcoin mining around 2.140, with current mining techniques and progressive difficulty in the mining algorithm. Quantum computers could place that date way before, but they are not ready yet and still need some years to be ready to resolve the mining process in a heartbeat.
The estimates I've seen place a 25% quota of bitcoin lost due to forgetting passwords, formatting hard drives, broken computers etc. This estimate seems rather high, rendering 4 million bitcoins orphan. It's true that people are probably more wary now days of storing them securely than before, due to their high value in comparison to when value was low and bitcoin a thing for the unconventional.
Miners make a large percentage of their income from validation transactions, regardless of mining new bitcoins. One would expect that over time, a progressive boost in transaction will take place due to bitcoin becoming an operative asset, in detriment of a speculative asset. If this happens, transaction price should be reduced, but benefit for validators increased overall.
What really matters in relation to bitcoin value is its potential and operative use (it's rather a speculative product too, but that should give way to a broader operative usage with time). If these factors remain and enhance over time, bitcoin price should rise in the long run, oblivious to the fact that there may be 4 million bitcoins that are unretrievable. The key is obviously the irretrievability.
If, hypothetically, they were to be recovered at some point and put into circulation at the same time, there would obviously be a price crash, but that scenario has basically 0 chances of becoming real.