paul_davies's picture
Theoretical physicist; cosmologist; astro-biologist; co-Director of BEYOND, Arizona State University; principle investigator, Center for the Convergence of Physical Sciences and Cancer Biology; Author, The Eerie Silence and The Cosmic Jackpot
Somatic Mutation Theory of Cancer

Cancer is one of the most intensively studied phenomena in biology, yet mortality rates from the disease are little changed in decades. Perhaps that's because we are thinking about the problem in the wrong way.

A major impediment to progress is the deep entrenchment of a 50 year-old paradigm, the so-called somatic mutation theory. It goes like this. A somatic cell serially accumulates genetic damage, eventually reaching a point at which it decouples from the organism's regulatory systems and embarks on its own agenda.

Cancer cells acquire a range of distinctive hallmarks—unfettered proliferation, evasion of apoptosis, motility and migratory powers, genomic rearrangements, epigenetic alterations, and changes in the mode of metabolism, chromatin architecture and elasticity (to mention a few)—that collectively confer remarkable robustness and survivability. In the standard picture, cancer, with all these attendant hallmarks, is considered to be re-invented de novo in each host organism: the result of a dream run of "lucky" genetic accidents. The gain of all these amazing fitness functions, co-located in the same neoplasm (population of new cells), over a period of as little as months or even weeks, is attributed to a sort of ultra-fast-paced Darwinian evolution going on in the body of the host organism. Unfortunately this theory, despite its simplicity and popular appeal, has only one successful prediction: that the administration of chemotherapeutic drugs is very likely to fail on account of the neoplasm's ability to rapidly evolve a resistant sub-population.

Armed with the somatic mutation paradigm, the research community has become fixated on the promise of sequencing technology, which enables genetic and epigenetic changes in cells to be measured on a vast scale. If cancer is caused by mutations, so the reasoning goes, then maybe subtle patterns can be teased out of petabytes of bewildering cancer sequencing data. If so, then the answer to cancer—perhaps even that elusive general-purpose cure—might be found by identifying common defects amid all that stunningly complex malfunctioning genetic machinery. Never has science offered a clearer example of a preoccupation with trees at the expense of the forest.

Stand back and take a hard, skeptical look at that forest. Cancer is widespread among multicellular organisms, afflicting mammals, birds, fish and reptiles. It clearly has deep evolutionary roots, probably stretching back over a billion years to the dawn of multicellularity. Indeed, it represents a breakdown of multi-celled cooperation. Unchecked, cancer follows a very predictable pattern of progression, usually spreading around the body and colonizing remote organs. It seems to be executing an efficient pre-loaded genetic and epigenetic program. Like a genie in a glass bottle, once it gets out it has a well-defined agenda. Many things can shatter the bottle, but the real culprit is the genie. The cancer research community, unfortunately, is preoccupied with seeking mostly irrelevant patterns amid the random shards of glass while ignoring the genie.

Why are our cells harboring such dangerous genies? The answer has been known for a long time, but it is mostly shrugged aside. The same genes that are active in cancer are also active in early embryogenesis (even in gametogenesis), and to some extent in wound-healing and tissue regeneration. These ancient genes are deeply-embedded and well-protected in our genomes. They run the core functionality of cells. Top of the functionality list is the ability to proliferate—the most fundamental modality of living organisms, with nearly 4 billion years of evolutionary refinement behind it. Cancer seems to be the default state of cells that are stressed or insulted in some way, such as by aging tissue architecture or carcinogenic chemicals, with tumors representing a reversion to an ancestral phonotype.

In biology, few things are black or white. The somatic mutation paradigm is undeniably of some relevance to cancer, and sequencing data is certainly not useless. Indeed, it could prove a gold mine if only the research community comes to interpret that data in the right way. But the narrow focus of current cancer research is a serious obstacle to progress. Cancer will be understood properly only by positioning it within the great sweep of evolutionary history.