Compounds that disrupt the network of blood vessels within a tumour can be effective as drugs against solid tumours because they deprive those tumours of their oxygen supply. One of the molecular targets for these vascular-disrupting agents is a site within the protein tubulin that binds colchicine, a natural product obtained from crocus plants (genus Colchicum) that is used for the treatment of gout. Drugs that target this site have proved to be extremely effective in killing tumour cells but clinical trials have revealed unacceptable levels of systemic toxicity, with severe cardiac side effects. Research is now focusing on targeting these compounds more specifically to the tumour, thus reducing their toxicity.
Tumours are known to express a family of proteases, the matrix metalloproteases, that have many roles in tumour growth and metastasis. One possible strategy for the design of tumour-targeted drugs is the development of pro-drugs that are converted to their active forms only by cleavage by one or more of these matrix metalloproteases. A group of researchers led by Jason Gill and Paul Loadman at the Institute of Cancer Therapeutics, University of Bradford, UK has now designed and developed a pro-drug of a colchinine-like compound that is broken down to its active form by membrane-type 1 matrix metalloproteinase (MT1-MMP) and that has proven to be extremely effective in mouse models of cancer.
The compound synthesised by Gill, Loadman and their co-workers and named ICT2588 was a conjugate of the colchincine derivative azademethylcolchicine (ICT2552) with a modified peptide designed to be cleaved from the active molecule specifically by MT1-MMP. This compound was first found to be much more active against a human fibrosarcoma cell line (HT1080) that expresses MT1-MMP strongly than against a breast carcinoma cell line (MCF7) that does not express this metalloproteinase at all; in contrast, the unconjugated drug ICT2552 was equally active against both cell lines. The compound was also shown to metabolise rapidly to azademethylcolchicine in xenografts of HT1080 cells, but not in MCF7 xenografts, normal mouse liver cells or mouse blood plasma.
The researchers then implanted a strain of immunodeficient nude mice with HT1080 tumour cells and treated them with ICT2588, ICT2552 or vehicle. All mice treated with ICT2588 showed significant anti-tumour responses and one entered complete remission. The unconjugated compound, ICT2552, showed a slightly less marked antitumour effect with greater toxicity, and the drug did not accumulate in the tumour to as great an extent. In almost all cases, however, a rim of viable tumour cells remained after the treatment.
Finally, Gill, Loadman and their co-workers treated mice bearing HT180-based tumours with different combinations of ICT2588 with doxorubicin. The most marked effects were seen when doxorubicin was administered after vascular collapse had been induced with ICT2588, with four out of a cohort of seven mice entering complete remission with no rim of viable tumour cells observed. This very promising result suggests that doxorubicin has successfully treated the small proportion of tumour cells that are resistant to ITC2558.
Taken together, these results indicate that ITC2558, a peptide-linked pro-drug of the vascular disrupting agent azademethylcolchicine (ICT2552), is cleaved to the active product by MT1-MMP and will therefore be selectively active against the majority of solid tumours that secrete this enzyme. Gill and Loadman predict that this compound is unlikely to show the cardiac toxicity present in the vascular disrupting agents that have so far entered clinical trials, and they strongly support its being tested in the clinic.
Reference
Atkinson, J.M., Falconer, R.A., Edwards, D.R., Pennington, C.J., Siller, C.S., Shnyder, S.D., Bibby, M.C., Patterson, L.H., Loadman, P.M. and Gill, J.H. (2011). Development of a Novel Tumor-Targeted Vascular Disrupting Agent Activated by Membrane-Type Matrix Metalloproteinases. Cancer Res. 70(17), 6902-12. doi: 10.1158/0008-5472.CAN-10-1440