Cellular-scale boron distribution in tumour and normal tissue
In plain words
The capture fragments travel only about a cell diameter, so the dose depends on whether the boron is inside each cell or outside it. Measurements of this in patients are lacking.
Precise statement
Measure the B-10 concentration distribution at 1 to 10 micron resolution in tumour, tumour margin and normal tissue for the clinical carriers (boronophenylalanine and sodium borocaptate), including the fraction of tumour cells with low uptake, and convert it into a microdosimetric specific-energy distribution for the $\alpha$ and Li-7 fragments of ranges about 9 and 5 micron. An answer is a measured distribution and the resulting ratio of cellular to bulk-averaged dose.
What would settle it
Neutron autoradiography or secondary-ion mass spectrometry on biopsy samples taken at treatment time, combined with track-level dose calculation.