The Six Characteristics of Cancer

I've recently come across a seminal oncology paper by Hanahan and Weinberg, entitled, "The Hallmarks of Cancer".  It's part review article and part conceptual, and deals with the "rules that govern the transformation of normal human cells into malignant cancers."  It is these two author's thesis that there are precisely six types of capabilities that are acquired by most (if not all) cancers.

These are:

• Self-sufficiency in growth signals
• Insensitivity to growth-inhibitory signals
• Evasion of apoptosis
• Limitless replicative potential
• Sustained angiogenesis, and
• Tissue invasion and metastasis

The first point, "self-sufficiency in growth signals" refers to the fact that cancers can't rely on the usual growth signals if they are to be successful.  All normal cells require a signal to divide - this is one of the body's defense mechanisms against carcinogenesis - and so cancer cells must overcome this limitation.  Some cancers, for instance, have been shown to have the ability to synthesise their own growth factors.

Secondly, cancers must avoid the usual cellular receptivity to anti-growth signals (again, this receptivity is one of the body's defenses against cancer formation).  As an example, cancers might disable the "don't proliferate" intracellular signalling pathway.

Apoptosis, referred to in the third point, is programmed cell death.  Abnormally functioning cells (e.g. severely damaged cells or wannabe cancer cells) are noticed by the body and ordered to selfdestruct in an orderly fashion: their cell membranes and nuclei are broken down, the chromosomes are cut up and the cytosol is extruded.  The resulting debris is phagocytosed thereafter.  At least, this is what should happen.  However, many cancers seem to be insensitive to the normal apoptotic signals.  The most famous example here is loss of a functioning p53 gene.  This archetypal "tumour suppressor gene" normally codes for a protein that senses DNA damage (including carcinogenic mutations) and either stops replication for long enough to repair this damage, or else simply induces apoptosis.  The loss of a functioning p53 protein is seen in at least 50% of human cancers. 

The fourth characteristic a cancer cell needs is "limitless replicative potential", or immortality.  The characteristic stumbling block to overcome here is telomeres.  We've dealt with this subject before, but, briefly, they are repetitive elements on the ends of chromosomes that get shorter (by about 50-100 base pairs) each time a cell divides.  When the telomere length is eventually lost, the cell can't divide any further and may even die.  More than 90% of cancers show reactivation of telomerase, an enzyme that adds length back to the telomeres, allowing the cancer cells to divide indefinitely.

All solid tumours (as opposed to, say, haematological malignancies) require the capability to induce new blood vessel formation in their vicinity.  In fact virtually no cell in the body can live further than 0.1 milimetres away from a capillary.  Usually this process is tightly regulated, but cancer cells seem to elaborate higher-than-usual amounts of pro-angiogenic factors, and/or lower-than-usual amounts of anti-angiogenic factors.

Lastly, cancers require the ability to invade tissue and then (in most, but not all, cases - see basal cell carcinomas) to metastasise to distant lands.  This characteristic in particular is likely to involve numerous discreet steps, but an example common to many cancers is the upregulating of the production of proteases capable of degrading the extracellular matrix/blood vessel walls/etc.. 

The jury's still out on how whether all cancers require all six characteristics, or whether this is simple the case in many or most cancers.  Of course, the theory doesn't demand that these six hallmarks are attained in the same way.  One type of tumour may achieve self-sufficiency from growth signals by one method, whereas the next tumour goes about the process in a completely different way.  The above points are abilities that cancers are hypothesized to require, and don't refer to any particular biochemical mechanisms.

I don't know about you, but I Hanahan and Weinberg's overview to be enormously useful -I'm going to have to alter an earlier post I made on carcinogenesis to incorporate it!  You can download their original paper by clicking here.