Model predicts which drugs affect cancer cell metabolism

By inhibiting their unique metabolic signatures, cancer cells can be killed off in a specific and selective manner, said Prof Eytan Ruppin of Tel Aviv University.

The efficacy of this method is said to have been demonstrated in computer and laboratory models pertaining to kidney cancer.

Because the researchers’ new approach is generic, it also holds promise for future investigations aimed at effective drug therapies for other types of cancer.

The researchers’ computer model is a reconstruction of the thousands of metabolic reactions that characterise cancer cells.

By comparing it to a pre-existing model of a normal human cell’s metabolism, they could distinguish the differences between the two and then identify drug targets with the potential to affect the specific, special characteristics of cancer metabolism.

To test their predictions, the researchers chose to target cells from a specific type of renal cancer.

‘In this type of renal cancer, we predicted that using a drug that would specifically inhibit the enzyme HMOX, involved in Heme metabolism, would selectively and efficiently kill cancer cells, leaving normal cells intact,’ said Ruppin. Their computer model led them to hypothesise that the Heme pathway was essential for the cancer cell’s metabolism.

In an experimental study led by Prof Eyal Gottlieb’s lab at the Beatson Institute for Cancer Research in Glasgow, the researchers, including Dr Tomer Shlomi of the Technion in Haifa, were able to verify this prediction in mouse and human cell models, and to study these metabolic alterations in depth.

While the first model was built to characterise a specific type of cancer, this approach could be applied in the future for creating models for other types of cancer.

‘This is the next big challenge for us,’ said Ruppin. ‘We are going to continue to build models for other types of cancer and to seek selective drug therapies to defeat them.’

Their multidisciplinary approach requires the predictions of a computer model and the findings of experimental clinical trials, and may lead to the faster development of more selective and effective cancer treatments.

The results were recently published in the journal Nature.