Since the 1940s, blocking the body’s production of androgen has been the only systemic therapy against prostate cancer. Today, while we’ve gotten better at targeting androgen signaling, treatments targeting the genetic causes of prostate cancer have lagged behind those for other forms of the disease.
A University of Colorado Cancer Center study published in the journal Molecular Cancer Research shows that one of the essential genetic features of an aggressive form of prostate cancer may also be its Achille’s heel. Prostate cancers that lose the gene MAP3K7 are genetically unstable and especially dangerous. But this genetic instability comes at a cost — the current study shows that these MAP3K7-negative cells gain genetic instability at the cost of a dismal capacity for DNA repair. In other words, their instability makes them delicate.
The study identifies genetic differences between prostate cancers that retain MAP3K7 and those that lose it. A main difference is the over-activity of a few other genes, including CDK1 and CDK2.
The study goes even further: If these dangerous prostate cancer cells magnify CDK1 and CDK2 and are poor at repairing DNA, what about turning off CDK1/2 and further nixing DNA repair?
When researchers used the drug dinaciclib to silence CDK1/2 along with the drug olaparib to further inhibit DNA repair, prostate cancer cells died in droves. The finding could lead the way to the first successful, genetically targeted treatment for MAP3K7-negative prostate cancer.
The story starts a decade ago, when the laboratory of CU Cancer Center investigator, Scott Cramer, PhD, showed that prostate cancers with the combined loss of genes CDH1 and MAP3K7 were more likely than other prostate cancers to grow and invade other tissues. But the question of what to do about this dangerous prostate cancer subtype proved tricky. Unfortunately, while doctors can turn off some over-active genes that cause cancer (e.g. ALK that causes some lung cancers and BRAF that drives some melanomas), they can’t replace missing genes. Treating prostate cancers that were missing MAP3K7 wasn’t as easy as replacing MAP3K7.
While the Cramer lab went on to successfully develop a mouse model of CDH1/MAP3K7-negative prostate cancer that could someday allow them to test new treatments against this subtype, what exactly these new treatments might be remained elusive. Until they turned to big data.
The lab of CU Cancer Center collaborator James Costello, PhD, deals in numbers. Lots of numbers. In this case, Costello writes, “Arrays were batch normalized using RMAexpress with background adjustment, quantile normalization, and median polish summarization. Probesets were mapped to HGNC gene symbols… Multiple probesets mapping to the same gene were averaged. This final gene expression compendium matrix, M, is the input to define the Transcriptional Regulatory Associations in Pathways (TRAP) network.”
It all sounds so simple!
The gist is that Costello’s lab was able to use mathematical modeling to compare cancers without MAP3K7 to cancers with MAP3K7 to see differences in underlying cellular functions. These results led to the identification of CDK1 and CDK2 as druggable gene targets. These genes are central in helping cancer cells rush through the process of replication. And the drug dinaciclib that targets CDK1/2 is already in human clinical trials.
Even alone, dinaciclib killed prostate cancer cells. In fact, supporting the storyline, it killed far more cells in prostate cancers lacking MAP3K7. And the team was able to show that one of the ways dinaciclib killed these cells was through further limiting DNA repair. (Technically, dinaciclib kept prostate cancer cells from repairing “double strand breaks.”)
The team reasoned that if reducing DNA repair killed some cells, then a dramatically reducing DNA repair could kill even more. And dinaciclib isn’t the only drug that stops DNA repair. Olaparib is another of these drugs. When the team treated MAP3K7-negative prostate cancer cells with the combination of dinaciclib with olaparib, more cancer cells died than the sum of the cells that were killed by each drug alone.
In fact, olaparib is one in a class of drugs known as PARP inhibitors that, while they have shown promise in laboratory and mouse model studies, have also been challenging to use with patients due to high toxicity. However, with dinaciclib sensitizing these cancer cells, it took a relatively low concentration of olaparib to kill them, implying that the combination may be a way to reduce the side effects currently associated with olaparib use.
“We looked at the pathology, and then used computational approaches to identify genetic pathways and genes within these pathways, and then tested the drugs identified to target these cells. And they did what they were supposed to do,” Cramer says.
“We’re excited,” Costello says. “There’s still a lot more work to do, but this approach could really change the way we think about targeting prostate cancer.”
With the success of these studies in prostate cancer cells, the team is excited to explore the use of this combination in mouse models of MAP3K7-negative prostate cancer.
This article was originally published on September 16, 2019, by the University of Colorado Cancer Center. It is republished with permission.