Biomedical Engineering Reference
In-Depth Information
in individual cells of defence mechanisms as an exaggeration of physiological
phenomena, such as are ABC transporters (the P-gp, or P-glycoprotein, being its
most known representant), but they may also result, at least as likely, in proliferative
populations encompassing mitoses, from mutations yielding more fit, i.e., resistant
in the presence of drug, subpopulations.
A classical solution to this problem is to forbid too low drug concentrations
that are supposed to create environmental conditions favourable to the development
of more fit drug resistant cell populations without killing them, as is also the
case, for instance, in antibiotherapy with bacteria. Nevertheless, other, more recent,
arguments to support an opposite view, have been put forth: assuming that there
exists a resistant cell population at the beginning of the treatment, or that it may
emerge during the treatment, then delivering high drug doses often produces the
effect to kill all sensitive cells, giving a comparative fitness advantage to resistant
cells, that subsequently become very hard to eradicate. Thus a paradoxical solution
has been proposed, at least in slowly developing cancers: killing just enough cancer
cells to limit tumour growth, but letting enough of these drug sensitive cancer cells
to oppose by competition for space the thriving of resistant cells, that are supposed
to be less fit, but just the same, usually slowly, will invade all the tumour territory if
no opponents are present [ 58 , 61 ]. Indeed, such free space left for resistant tumour
cells to thrive, when high drug doses have been administered with the naive hope
to eradicate all cancer cells, may result in the rise of tumours that escape all known
therapeutics, a nightmare for physicians which is unfortunately too often a clinical
reality. Hence the proposed strategy to avoid high doses, that are able to kill all
sensitive cells, and to only contain tumour growth by keeping alive a minimal
population of drug-sensitive tumour cells.
Both those constraints, toxicity and resistance, can be considered as part of
the objective function by setting the objective to be a balance between two
objectives. For instance, Kimmel and Swierniak in [ 75 ] proposed to minimise a
linear combination of the number of cancer cells and of the total drug dose. This
yields an unconstrained optimisation problem, that has a simpler resolution, while
still taking into account the diverging goals of minimising the number of cancer
cells and keeping the number of healthy cells high enough.
But whereas cancer and healthy cells are two quite distinct populations, with
growth models that may easily be distinguished and experimentally identified
by their parameters, it is more difficult to take into account the evolutionary
lability (i.e., the genomic instability) and heterogeneity of cancer cell populations
with respect to mutation-selection towards drug resistance, according to evolution
mechanisms that are not completely elicited. Note that acquired (as opposed to
intrinsic , i.e., genetically constitutive) drug resistance may result as well from
individual cell adaptation (enhancement of physiological mechanisms) as from
genetic mutations, both under the pressure of a drug-enriched environment, as
discussed in [ 42 ]. In this respect, acquired resistance may be reversible, if no
mutation has initiated the mechanism, or irreversible, and it is likely irreversible
in the case of intrinsic resistance.
Search WWH ::




Custom Search