By John Russell

Sept. 25, 2008 | This month, two papers in Science* described the integrated genomic analysis of pancreatic cancer and a form of brain cancer, glioblastoma multiforme (GBM). They not only revealed important biology, but also suggest that the pharmaceutical and health care industries must fight solid tumors differently—principally by aiming at pathways rather than individual targets and focusing on early detection rather late-stage cures.

The studies, led by researchers from Johns Hopkins, examined roughly 20 tumors of each cancer type. They sequenced all protein-coding genes (~21,000) and examined expression patterns and copy number changes. What they found was tremendous genetic diversity. A typical pancreatic cancer tumor had 63 genetic alterations; on average 49 of those changed the genes and their products. Results were similar for GBM, where a typical tumor contained about 60 genetic alterations.

This picture of solid tumor heterogeneity poses a potential nightmare for selecting individual targets. However, when the genetic changes were put into pathways, a different picture emerged. In pancreatic cancer, investigators identified a core of 12 altered pathways that each individually affected more than two-thirds of the tumors analyzed. Similar pathways were found to be altered in GBM as well as a few others.

These results were considered so striking that a press teleconference was quickly arranged the day before online publication in Science, and was attended by paper authors and cancer researcher heavyweights: Bert Vogelstein, co-director of the Ludwig Center for Cancer Genetics and Therapeutics (LCCGT) and Howard Hughes Medical Institute (HHMI); Victor Velculescu, associate professor of oncology, Johns Hopkins Kimmel Cancer Center, and Ken Kinzler, co-director of LCCGT and HHMI.

Cutting to the chase, Vogelstein said, “It is extremely unlikely that drugs which target a single gene, like Gleevec, will be active against a major fraction of solid tumors. Instead of screening for drugs against single proteins, our work suggests it may be more productive to screen for drugs that act on the core pathways that are dis-regulated in most cancers.

“By targeting the pathways, it’s possible new drugs could be effective against a much greater fraction of tumors. This is a very different perspective from what’s now operative in the drug development community,” said Vogelstein.

Secondly, said Vogelstein, “I think it’s accurate to say that 99 percent of applied cancer research now goes toward developing new therapeutics. It is, I think, apparent from studies like ours that it is going to be even more difficult perhaps than previously expected to derive real cures from such therapies. One interpretation of our work is that the proportion of effort and funding devoted to other ways of managing cancer, such as prevention and early detection, should be greatly increased…as they may have much more success in minimizing cancer deaths.”

It is important to note these kinds have studies discussed here are made possible by tools that are only now starting to prove their value. Pathway analysis tools, for example, were important in this work; indeed, GeneGo not only provided tools for the cancer work discussed here, but also conducted much of the analysis, and landed authors—Tatiana Nikolskaya (president and CSO) and Yuri Nikolsky (CEO)—on both papers.

More tools are needed. Vogelstein, for example, suggests that new imaging technologies to detect activated pathways must be developed. He further predicts that a simple blood test for some cancers is no longer science fiction.

“We now know precisely how many genetic alterations in coding genes there are in typical pancreatic and brain tumors, for example, and with current technologies it’s actually easy to detect many of them in the cancers, and it will be possible soon to detect many of them on other samples from patients, say, in their blood.” It’s worth noting the thought is to detect DNA that gets into the blood from tumor cells dying in situ before metastases occurs. It’s not precisely clear how that happens; only that it does.

GBM Is Two Diseases, Not One
“One of the most surprising discoveries we made was that 12 percent of GBM patients had mutations in a gene called isocitrate dehydrogenase (IDH1). This gene had never been previous implicated in brain cancer or any other tumor type and could not have been identified without the unbiased genome-wide analyses that we performed,” said Velculescu.

“Interestingly, all of the mutation in IDH1 affected the same protein residue, the arginine at position 132, suggesting that alteration is likely to dramatically affect protein function. Patients with mutations in IDH1were different from those that are usually afflicted with GBM. The average age of patients with IDH1 mutations was 20 years younger than those without IDH1, and nearly have half of patients under age 35 had IDH1 alterations.

“Additionally, patients with IDH1 mutations had different clinical outcomes, including longer survival than for those without alterations. GBM used to be thought of as one disease. It is now clear that there are two. IDH1 mutations identify a category of GBMs that are different both biologically and clinically. In the near future, that information regarding IDH1 alterations will be useful for clinical management of GBM patients.

“Patients could be stratified by mutations in this gene and treated with existing therapies that may work better in this subgroup or treated with new compounds designed specifically against IDH1. Recent reports have suggested the targeting of other IDH enzymes may sensitize cells to certain therapies and provides hope the IDH1 may ultimately be useful as a therapeutics target.”

----------------------------------
*Williams Parsons, D. et al. “An Integrated Genomic Analysis of Human Glioblastoma Multiforme.” Science, published online 4 September 2008; Jones, S. et al. “Core Signaling Pathways in Human Pancreatic Cancers Revealed by Global Genomic Analyses.” Science, published online 4 September 2008

----------------------------------

This article first appeared in Bio-IT World’s Predictive Biomedicine newsletter. Click here for a free subscription.