If (1), your results are obscure or lack applicability, then they are not useful anyway and so it doesn't matter if they are ever confirmed. A surprisingly large amount of scientific research fits into this category.
While (2) few researchers set out solely to verify another's result, if the result is noteworthy, it will be incorporated as a component of future research and therefore tested indirectly. This other comment (not mine) explains how:
Comments
Neither of these are actually problems.
If (1), your results are obscure or lack applicability, then they are not useful anyway and so it doesn't matter if they are ever confirmed. A surprisingly large amount of scientific research fits into this category.
While (2) few researchers set out solely to verify another's result, if the result is noteworthy, it will be incorporated as a component of future research and therefore tested indirectly. This other comment (not mine) explains how:
https://news.ycombinator.com/item?id=6662124