Since many “soft” metal ions bind to the same DNA co-ordination sites as the widely used anti-cancer drug, cisplatin, it seems likely that some of these should also yield effective chemotherapeutic agents. New results in this area were presented at a symposium held in Trieste, Italy, on 30th June–1st July 1988 which focused on the design and modes of action of non-platinum metal ions in cancer chemotherapy.

Ruthenium appears to be a likely candidate, even though its chemistry differs substantially from that of platinum. The most significant differences are its octahedral geometry and greater propensity to undergo redox reactions. The hypoxic environments of many tumours may favour the reduction of Ru(III) compounds (which are relatively slow to bind to most biological substrates) to Ru(II) species, which bind rapidly. Once co-ordinated to a DNA target, the metal may interfere with DNA metabolism by a variety of mechanisms. Migration between sites can also occur in a manner which is controlled by redox potential and pH.

According to Dr. Jacqueline K. Barton of Columbia University, advantage can be taken of the two additional binding sites provided by octahedral geometries to design chiral compounds that are selective for the particular shapes of the various DNA conformers. Outer-sphere binding may occur electrostatically, hydrophobically or intercalatively. If open coordination sites are available on the metal ion, subsequent bond formation may ensue. Scission of the DNA induced by the metal ion may occur by photochemical or redox routes and can be highly selective.

An interdisciplinary group in northern Italy headed by Professors G. Mestroni in Trieste and F. Quadrifoglio in Udine has been investigating the solution chemistry and DNA interactions of a family of compounds involving cis and trans isomers of X₂(DMSO)₄ Ru, where X=Cl or Br, and DMSO is dimethyl sulphoxide. While only mildly active, these compounds are fairly non-toxic and the trans isomer strongly inhibits metastases of Lewis lung carcinoma. In aqueous solution, the cis complex rapidly loses the O-bound DMSO ligand, while the trans isomer quickly aquates on losing two cis-DMSO ligands. The trans isomer binds to DNA much more rapidly and to a higher degree than the cis isomer. DNA co-ordination by the cis species almost ceases in chloride concentrations approximating that of plasma, and the site of binding of both is suggested to be on the N7 of guanine with a possible involvement by the phosphate.

Dr. Bernard K. Keppler from the University of Heidelberg, Germany, reported very promising results on a number of compounds with the general formulae: trans-HB[RuB₂Cl₄] and (HB)₂[RuBCl₅], where B=nitrogen heterocycle. These compounds are effective against colorectal tumours, which commonly occur and, once metastasised, are presently incurable. Keppler, in collaboration with M. E. Heim, also reported on the first stage of clinical trials of Budotitane, the titanium compound diethoxybis(1-phenylbutane-1,3-dionato)titanium(IV), which is also active against autochthonous colorectal carcinomas. This compound is well tolerated at a level of 150 mg/m², it exhibits liver toxicity but no myelosuppression or nephrotoxicity.

Dr. Petra Köpf-Maier of the Free University of Berlin presented results on three different types of metallocene complexes: a) neutral cisdiacido complexes, Cp₂MX₂, M=Ti(IV) or V(IV); b) cationic metallocenes, such as ferrocinium, and c) decasubstituted neutral metallocenes, (C₅R₅)₂M, where R=phenyl or benzyl and M=Sn(II) or Ge(II). Examples of each type were active in markedly inhibiting tumour growth and induced severe cytological and histological changes in the tumours treated, which suggested that the complexes interfere with nucleic acid metabolism. Reaction of the titanium complex resulted in [Cp₂Ti]⁺ coordinated to purine nucleosides through both monodentate N7 and N7-O6 chelation. The vanadocene moiety appears to bind in a labile outer-sphere fashion to phosphate groups. Doses of Cp₂TiCl₂ are limited by liver toxicity.

Ruthenium and platinum compounds may be used to direct radiosensitising molecules to DNA, according to Professor Nicholas Farrell of the University of Vermont. This approach is particularly useful in radiotherapy applied to hypoxic or anoxic areas of tumours, where radiation is less effective. Studies on the radiation killing of cells indicate that metal complexes ligated with nitroimidazoles are more active than either the metal or ligand precursor molecules alone and so point a way to the design of new agents.

The ruthenium isotope, ⁹⁷Ru, has excellent radiophysical properties for use in diagnostic imaging agents. Its γ -ray is easily collimated by existing radioscintigraphic cameras and its 3-day half-life provides adequate time for synthesis and quality control. Dr. Suresh Srivastava of Brookhaven National Laboratory presented the first clinical studies of a liver-imaging agent using a complex with diisopropyl carbamoylmethyl iminodiacetate, which proved very effective in imaging the livers of neonates suspected of liver dysfunction. Radiolabelling studies also showed that some Ru(III) complexes can be bound to transferrin and carried to receptor sites on tumours. The metal ion is fixed inside the cell, possibly by a redox mechanism, while the transferrin is released.

The symposium discussions concerning the various approaches to ruthenium-containing anti-cancer drugs reflected the versatility of this element in synthesis, electron transfer, and even photochemistry. These properties, coupled with the affinity of ruthenium’s intermediate oxidation states for imine nitrogens, facilitate DNA targeting for both chemotherapeutic and radiosensitising agents. Finally, the existence of isotopes with desirable properties for diagnostic imaging indicates that exploration of the medical applications of ruthenium is likely to produce useful pharmaceuticals.