Hi Kat,
Regarding your Onc believing the Sprycel "umbrella" is not big enough to catch all of the CML if the dose is too low. He is correct and not correct in his analogy. He is correct that dose matters, but incorrect regarding resistance. Cancer is not bacteria. Cancer is about population dynamics.
Without going into too much "math" detail on population dynamics (https://en.wikipedia.org/wiki/Population_dynamics), drugs that are designed to kill cells (apoptosis) are modeled using population dynamic calculations. Most "doctors" are not schooled in these models so they use words like resistance which is incorrect. Cancer cells in the way they grow and multiply are not that different mathematically to the way roaches, fleas or bunny rabbit populations grow. These populations - barely noticeable - can suddenly explode in numbers.
Drugs like Sprcyel and Gleevec are population controllers - like flea traps, flea sprays or wolves that like eating rabbits. So in the scheme of things there are two competing forces in CML treatment that effect outcome. The rate of CML cell division and expansion and the rate of CML cell death caused by our TKI killing cml cells. If the rate of cell death is greater than the rate of cell expansion, our CML starts to fade away and ultimately go into remission. The fade away can be gradual or it can be sudden. Just like flea populations can also collapse after using a fogger. It doesn't mean all of the fleas are dead, just that enough of them are wiped out that re-establishing the "colony" is difficult - especially if a residual trap is left around.
TKI's work that way. They effect the population dynamics of CML.
When our CML is first diagnosed, the cell counts are explosive and out of control (for the most part - in reality our bodies are trying like crazy to keep it under control which is why the spleen enlarges among other things). When we take Gleevec (for some people), it binds imperfectly to the energy site of the CML cells and kills them. It does not kill all of the cells, but as long as it kills more cells than are being replaced, the overall CML population will go down and as it goes down, the CML population ultimately collapses to a new steady state condition. This is why some people remain at various levels of PCR.
Gleevec is a dose dependent drug. The more you take the better it works. Precisely because it doesn't kill every CML cell it comes into contact with. More drug is necessary to hit more cells more of the time. But of course more drug leads to toxicity and if a patient would need the equivalent of 2000 mg of Gleevec to be effective at controlling CML - the drug is impractical - 800 mg is pretty much the upper limit.
Sprycel on the other hand is a threshold drug. What makes Sprycel different is that it is more potent than Gleevec. When ever Sprycel comes into contact with CML cells they have no chance, they die - and they die at higher levels in the cell division hierarchy than Imatinib does (i.e. kill the fire ant queen, not just fire ants). So the amount of drug needed to cause a population collapse of CML cells is far lower and once crossed - more drug makes no difference - the CML population is already under collapse.
It is not quite correct to say that more Sprycel doesn't increase the rate of collapse, just that the increase is so small it's not important. So taking more drug just leads to toxicity. The key for patients therefore is to find the threshold that triggers CML collapse.
In mathematical terms, Gleevec impact on CML is mostly linear, where Sprycels impact on CML is logarithmic.
The key - is - if it works. If Sprycel works, it tends to work very well. When it doesn't work - it doesn't work, more drug usually (there are always exceptions) makes no difference. Gleevec, when it works, usually works even better with more drug - so its threshold of effectiveness is not as sharp as Sprycels.
As we are all slightly different, finding the right drug, the right dose and even the right other things (like Curcumin, or vitamin D) requires individual testing.
Hope this helps answer the 'umbrella' question.