The “Viagra effect”: how known drugs can be repurposed to target cancer stem cells

by Paul Krzyzanowski

Repurposing known drugs for new applications is a strategy with fascinating potential, with two of the most notable examples being Thalidomide and Viagra. Thalidomide was commonly used in the late 1950s as a sedative in pregnant women, later being associated with serious birth defects. Today, it is used to treat multiple myeloma. Viagra was being developed by Pfizer to treat high blood pressure when its ability to ‘treat’ erectile dysfunction was identified as a side-effect, resulting in a complete shift in marketing strategy.

The major allure of finding novel uses for existing drugs is that the long process of early Phase I clinical trials can be sidestepped, as the drugs are already known to be safely delivered. This approach decreases the overall cost of developing drug candidates and brings the development of treatments for rare and neglected diseases closer to reality.

The principle of identifying a drug candidate is straightforward: take a large number of different chemical compounds and test each one for some desired activity. Small molecule screening has long been used by pharmaceutical companies to identify potential drug candidates for probably as many disease conditions as there’s a market for.

Recent years have seen small molecule screening projects being published by academic groups as the required technology becomes more available and public funding is used to promote research with more immediate economic benefits. Some of the most exciting chemical screening studies manage to incorporate knowledge of cancer stem cell biology to yield results that may have immense potential for anti-cancer research.

At The Hospital for Sick Children in Toronto, one such small molecule screening study tested a new kinase inhibitor drug library for agents that inhibit the survival of neuroblastoma cells from patients. Neuroblastoma is one of the most common tumours found in children, with a likely stem cell origin. This work, led by Dr. David Kaplan, was published in Cancer Research and identified several compounds that preferentially inhibit neuroblastoma cell growth. 

Kaplan’s team was able to narrow their interest down to BI 2536, a Polo-like kinase 1 (PLK1) inhibitor currently in clinical trials for used against lymphoma and several other adult cancers, and showed that BI 2536 and a new PLK1 inhibitor called BI 6727, significantly reduced neuroblastoma tumour growth in mice.  Since their publication, they have repeated their studies using cells from additional patients. As these drugs are already in Phase II trials for other cancers, it seems that they might be brought up to speed for use in pediatric neuroblastoma in the near future. One interesting aspect of Kaplan’s drug screening program is the use of normal pediatric stem cells to ensure that the drugs are not toxic to normal non-cancerous cells.

In another similar study, recent work published in Blood by Dr. John Dick used a 4,000 chemical library to identify kinetin riboside as a compound that kills leukemia initiating stem cells while also preventing their engraftment. By assessing the ability of each compound to inhibit leukemia cell growth, this group was able to quickly narrow 4,000 potential drugs down to 80 (19 of which were already known anti-leukemia drugs).

Using a clever approach to identify the most specific cancer cell killing compounds, John Dick’s team determined the potential toxicity to normal cells for a few dozen of the 80 drugs mentioned above. Kinetin riboside was identified as being three times more toxic to leukemia cells as compared with cells representing the normal hematopoietic system. The results of this study add to other work from the Mayo Clinic showing that kinetin riboside also has anti-myeloma activity by halting the cell cycle, adding bits of evidence that this drug might become a versatile anti-cancer compound in the future.

Unfortunately, kinetin riboside doesn’t seem to be currently in use or in any clinical trials. However, as an aside, the Mayo Clinic’s press release regarding the myeloma study was quick to point out, perhaps a little irresponsibly, that kinetin riboside is found in coconut milk. I don’t know if this press release was the trigger, but there’s a proliferation of web sites touting coconut products as anti-cancer remedies.

For the record, another earlier study did in fact report that kinetin riboside is found in coconut water, but at a very low concentration of under 2 ppb. To provide some context on just how small this amount is, the Ontario limit for lead in drinking water is 10 ppb (for those in the US, the EPA limit is 15 ppb). I want to emphasize that one can’t simply make the leap from carefully controlled studies regarding a potential drug to ingesting it without first showing that it is effective in people and more importantly doesn’t cause any harm. So, while one may decide to drink coconut water, it should not be in place of a proven and possibly life-saving therapy such as chemotherapy.

These insights into new anti-neuroblastoma and anti-leukemia drugs are just two examples of how combining industrial drug screening technology with knowledge of cancer stem cells can yield significant advances in developing new disease treatments. Sometimes the most insightful findings come from work that combines multiple areas of expertise to answer one very incisive question.