The development of clinical practice recommendations or guidelines for the clinical use of pharmacogenomics data is an essential issue for improving drug therapy, particularly for drugs with high toxicity and/or narrow therapeutic index such as anticancer drugs. Although pharmacogenomic-based recommendations have been formulated for over 40 anticancer drugs, the number of clinical practice guidelines available is very low. The guidelines already published indicate that pharmacogenomic testing is useful for patient selection, but final dosing adjustment should be carried out on the basis of clinical or analytical parameters rather than on pharmacogenomic information. Patient selection may seem a modest objective, but it constitutes a crucial improvement with regard to the pre-pharmacogenomics situation and it saves patients' lives. However, we should not overstate the current power of pharmacogenomics. At present the pharmacogenomics of anticancer drugs is not sufficiently developed for dose adjustments based on pharmacogenomics only, and no current guidelines recommend such adjustments without considering clinical and/or analytical parameters. This objective, if ever attained, would require the use of available guidelines, further implementation with clinical feedback, plus a combination of genomics and phenomics knowledge.