Antibiotics that target cancer cells, such as doxorubicin and bleomycin, have made an important impact on the treatment of cancer patients since the 1960s.

Knowledge of these and other natural products, such as distamycin, has led to the design of a new generation of DNA-targeted agents, DNA code-reading molecules.

In contrast to non-specific alkylating agents, anti-tumour antibiotics, such as doxorubicin, seem to have greater selectivity for cancer cells than normal cells, based on their improved clinical efficacy.

As doxorubicin is a non-specific DNA-intercalating agent, it was surprising that its enhanced selectivity for cancer cells could be obtained.

Some insight into the basis of this selectivity was gained when it was discovered that doxorubicin induces protein-associated strand breaks by trapping topoisomerase II that is covalently bound to DNA.

This important discovery encouraged researchers to consider processes that occur on DNA - such as transcription, replication and repair - as molecular events that might be more susceptible to DNA-interactive drugs in cancer cells than in normal cells.

Because these processes require protein binding to DNA, the selectivity of agents that target these processes might be dependent on the level of the associated target protein (for example, topoisomerase I or II), so cells with elevated levels of topoisomerase would be more susceptible to doxorubicin, and a basis for enhanced cancer-cell selectivity emerges.

Despite the general acceptance that the topoisomerase-II-DNA complex is an important molecular target for doxorubicin, it is likely that other mechanisms, such as direct oxidative damage to DNA, might contribute to the overall efficacy of doxorubicin, and the relative importance of molecular targets will vary from one tumour to another.