Mark Burkard and colleagues knew that when cells — from yeast to humans — have more DNA than normal, their biology changes. They also knew that cancer cells often have extra DNA. What they did not know was whether this feature, called polyploidy, could be exploited in cancer treatments.
“No one has previously characterized the polyploid subset of cancers,” says Burkard, professor of medicine at the UW Carbone Cancer Center and senior author of a recent study published in Molecular Cancer Therapeutics.
Burkard and his research team wanted to know how often cancers are polyploid, what biological characteristics those cancers have, and whether they could find lower toxicity chemotherapy agents that would target cancerous polyploid cells but not the normal, non-polyploid cells.
“Polyploid tumors provide a strong clinical basis for improving treatment, and a biological basis for therapy,” Burkard said.
For the study, the researchers used a primary tumor database and found, for example, that 14 percent of all breast cancers and 34 percent of pancreatic cancers are polyploid. In breast cancers, which Burkard studies in the lab and treats in the clinic, they also found that polyploidy was a predictor of cancer recurrence or death.
From there, Burkard and his team did two things.
First, they searched for polyploid-targeting chemicals from a set of over 45,000 potential anti-cancer compounds. Screening that many compounds would be overly laborious and costly for one lab to tackle.
“The National Cancer Institute for years has done screens where people sent them chemicals and they measured death of roughly 60 cancer cell lines, some of which are polyploid,” Burkard said. “We took that data, re-analyzed it to calculate the correlation between cell sensitivity and ploidy, and found dozens of potential compounds that may be truly specific to polyploidy.”
But cancer cell lines grow in the lab because of all kinds of biological changes, of which polyploidy can be one. Was polyploidy the only target of the drugs? The researches needed a way to identify the true polyploid-targeting compounds.
[In the scientific publication, the researchers called these unknown biological differences “confounding variables.” In conversation, Burkard calls it “some crazy unknown factor.”]
So, they next took two non-polyploid cancer cell lines, manipulated them so that some cells had exactly twice as much DNA, and tested 30 promising compounds for their ability to kill the polyploid cancer cells but not the otherwise identical non-polyploid ones. Between the two sets of experiments, they identified an unknown compound that fit their criteria: specific to polyploid cells, but applicable to many types of cancer.
The compound, DPBQ, somehow tricked the polyploid cells into thinking they were being deprived of oxygen, causing them to initiate a cell suicide program, though Burkard added they do not yet know exactly how it works. In addition to the potential for clinical use, the researchers expect that learning more of DPBQ’s cellular functions will help explain the biology of polyploid cells.