photo walkheader_zpsd5bunpmi.png

Cancer Diversity: Enemy of Personalized Medicine ?

In what appears to be a season of blockbuster conceptual research results in the cancer medical literature, comes a recent study involving leukemia cells but which has implications for the entire spectrum of cutting edge cancer treatment. The general current prevailing thought on the origin of cancer is that a pattern of gene mutations or epigenetic changes emerge – that initiates signaling miscues to the normal orderly growth pathways of a cell – thus launching subsequent rapid and erratic cellular division and growth: cancer.

The present study takes the above as a starting point, but in a heretofore unknown (to this author) minor strain of research over the last few years – asks the questions: are these genetic mutations physically uniform throughout a tumor, and does the mutation profile of the tumor remain uniform over time?   Of course, their answers are no and no.  They cite research primarily from the last two years, but going back until 2008 demonstrating that there exists genetic diversity within a single tumor – and they indicate that “intraclonal genetic and phenotypic diversity is an inherent feature of [cancer].”

The implications of this concept are large, in that much of the idea of personalized medicine and targeted therapy is aimed at specific genetic patterns.  If such a pattern is not uniform throughout a tumor then the results of the therapy may not be effective, or could possibly be ameliorated.

Greaves and colleagues primarily from The Institute of Cancer Research in London E-published the results of their research on November 6th 2013 in the journal Genomics Research that offers an in-depth view of genetic diversity in leukemia cells. In a multi-stage procedure the authors used state-of-the-art technologies to undertake single cell whole genome DNA sequencing, repeated in approximately three hundred cells. This was done for an individual with leukemia, and later for two other leukemia patients. The results showed not merely the expected mutations, but a large range of genetic mutations, and which were seen to be distributed in a manner which suggested sub-clonal populations (as identified by mutation patterns).  These data were sorted and placed into a tree-and-branch clonal phylogeny by algorithm which physically demonstrated the relationships and temporal evolution of the DNA mutation patterns.  In effect, these results showed that the individual patients did not have “leukemia” but in fact (rather astoundingly) had between two and ten genetically distinct leukemias.

Interestingly, the authors refer to their research process in such terms as “interrogating” clonal genetic complexity, in the sense that the DNA profile might perhaps be coaxed (through similar observation) to tell the story of what has happened to the cell over time. The researchers make the point that the beginning of disorderly cell division and growth due to DNA mutations in cancer increases the likelihood of subsequent DNA mutations in the same tumor / cell line. They indicate that leukemia is less likely to be complex in these matters than carcinoma (one might insert here “pancreatic cancer”).  Finally, with all of the potential problems that these results confer to the idea of personalized medicine (treatments aimed at a patient’s tumor profile) the authors’ offer some promise in the sense that future doctors on understanding a more complex picture of the underpinnings of a given tumor, may be more likely to treat the “founder lesion” rather than a clonal branch or even a more minor clone – thus conferring more likely success in therapy.

This is large concept work – shown in meticulous, brilliant detail. It is actually rather breathtaking science that reveals again the unexpected at the cutting edge of medicine; plaudits to these researchers.           

 More here


Dale O’Brien, MD