Regression or LSD for discrete quantitative factor?

I have a 2-factor factorial experiment where Factor A is quantitative and continuous with 2 levels (100 and 200). Factor B is quantitative, but discrete (number of times a treatment was applied; 1, 2, 3, or 4 times).

I carried out an ANOVA, but what is the correct way to explore Main effects and interactions, regression or multiple comparison procedures? As the factors are quantitative I was thinking regression, but was unsure if only having 2 levels for Factor A. And moreso, with Factor B being discrete and not continuous, is regression still valid, as the part of the line between 1 and 2, for example, has no real meaning.

Any help would be greatly appreciated.