FISH and BMB are actual counting of leukemic cells in a sample. PCR is a process to grow DNA/RNA fragments and then compare them to other fragments. When a process is inexact, you need inexact terminology to describe the outcomes. The PCR results are provided as a percentage, but it is not a percentage of leukemic cells in the body. Rather, it is a percentage of BCR-ABL RNA strands to another normal RNA strand found in the white blood cells.
If a PCR were an actual count of leukemic cells (actually BCR-ABL RNA chain sequences) then there would be no such thing as "inconclusive", or "below detectable limits", or "kinda sorta looks like....". But a PCR is an estimation of the amount of BCR-ABL and from the leukemic cells in the sample, and there is a fair amount of uncertainty. It is an estimation since it takes a sample of blood or marrow, puts it through multiple sequences of deconstructing the DNA and growing new DNA/RNA, baking and shaking, denaturing and renaturing, etc. At some point the lab equipment says "time is up" and the process is ended and the BCR-ABL is compared to the amount of ABL (or other control gene) in the sample. If no BCR-ABL is found, it is undetectable. If the test shows a very low level of BCR-ABL, then it may be a false positive, so it is below the cut-off for accurate detection. It is labeled "inconclusive" or "below limits" or whatever.