One of the major problems in cancer pharmacology is the prediction of
the outcome of treatment in terms of both toxicity and tumor response.
Indeed, why is a patient presenting a high toxicity at a standard dos
e? Also, why is another patient presenting a response to chemotherapy,
while another does not respond, even if the same doses and drugs are
used? The causes of variability in the response to chemotherapy can be
classified in 2 classes, these related to the host, and those related
to the tumor. Concerning the host, physiopathological parameters can
influence the outcome of treatment e.g. age, sex, kidney and liver fun
ctions, plasma protein binding, concomitant treatments, and pharmacoge
netics. With regard to the tumor, many factors can also influence chem
otherapy e.g tumor type, localization, volume, aggressiveness, dissemi
nation stage, prior treatments (resistance), biological marker levels,
and pharmacogenetic phenotype. Although some progress has been made i
n the past years concerning the clinical use of some factors responsib
le for the variability of pharmacological response in oncology, much r
emains to be done in order to adapt the treatment to a particular pati
ent. Although pharmacogenetic phenotyping has been feasible for some a
nticancer drugs, this area deserves more effort in the future. Indeed,
pharmacogenetic phenotyping of both the patient and the tumor will pr
obably allow to tailor chemotherapy to an individual patient, in order
to optimize the patient's chances to obtain a tumor response with min
imum systemic toxicity.