Science in Society Archive

Cancer Cure & Prevention

Personalized Medicine for Cancer Fact or Fiction?

Companies are marketing genetic profiling to provide personalized cancer therapy, but cancers show numerous mutations that differ not only between individual patients but also from one region to another in a single tumour Dr. Mae-Wan Ho

Cancer as genetic disease dominate approach to therapy

For many years, cancer therapy concentrated on attacking DNA replication, as cancer cells proliferate and replicate their DNA rapidly. But these generally cytotoxic drugs also harmed cells that divide rapidly under normal circumstances such as bone marrow cells, cells in the digestive tract and hair follicles, with inevitable side-effects: decrease in blood cells and immune suppression, inflammation of the gut, and hair loss [1].

More recently, newer therapies target the abnormal biology of cancer cells based on the belief that cancer is a genetic disease involving mutations in key ‘gate-keeper’ cancer genes (oncogenes). These include signal transduction and protein turnover pathways, apoptosis (programmed cell suicide) and signalling receptors. Some of these agents exhibited antitumour activity and have been approved for cancer therapy [2], and new candidates are popping up all the time [3].  

Still, there have been no cures in advanced cancers, though it is hoped that some combinations of agents may do the job.  

Nevertheless, the field of cancer therapy has been gripped by an “overoptimism” that soon, patients with a tumour will undergo a needle biopsy, and a personalized treatment will be devised on the basis of the distinctive genetic characteristics of the tumour. Already, several companies are marketing tests for the genetic signature of a tumour, with the expectation that the genetic signature will determine the treatment and predict treatment outcome.

But a serious flaw in that imagined future of cancer therapy based on personalized medicine is the underestimation of tumour genetic heterogeneity; not just between tumours, but heterogeneity within an individual tumour. This was highlighted in an Editorial in the 8 March 2012 issue of the New England Journal of Medicine [4].

Profuse genetic heterogeneity between tumours and within tumours

In the same issue of the journal, a team of 30 researchers led by Marco Gerlinger from the Cancer Research UK London Research Institute mapped out in detail how heterogeneous a single tumour can be [5].  Tumour samples were obtained from four patients with renal-cell cancer before and after treatment, with multiple samples taken from each patient’s primary and metastatic tumour sites. The team carried out exome sequencing (sequencing of all regions that code for proteins, roughly 1 % of the entire human genome), chromosome aberration analysis and ploidy profiling (to determine how many sets of chromosomes are present instead of the usual two). They also characterized the consequences of genetic heterogeneity within a single tumour using immunohistochemical analysis, mutation functional analysis and profile of messenger RNA expression.

Over a hundred mutations are typically found in each patient (just in the coding regions of the genome; over the entire genome it would typically be thousands), and a branching phylogenetic (evolutionary) tree can be drawn based on shared mutations in different regions. About two thirds of the mutations found in single biopsies were not uniformly detectable throughout all the sampled regions of the same patient’s tumour. Different regions of the same tumour gave a “favourable prognosis” and an “unfavourable prognosis” gene profile. There is no way a single tumour biopsy – the standard of tumour diagnosis and the cornerstone of personalized-medicine – can be considered to represent the genetic profile of the tumour, much less so, the cancer patient.

To make things worse, there are widespread alterations in the total number of chromosomes in the tumour cells (aneuploidy), and many allelic imbalances are found in which one allele of a gene pair is lost, either due to chromosome loss, or difference in gene imprinting that alter gene expression.

Another finding is that different regions of the tumour have different mutations in the very same genes (convergent evolution), suggesting that parallel alterations in epigenetic mechanisms (not immediately involving gene mutations) and signal transduction have taken place to ensure the tumour’s survival.

All that is part and parcel of the fluid genome of cells responding to their microenvironment within the body (see [6] Living with the Fluid Genome, ISIS publication). But most cancer researchers have not faced up to the possibility that most, if not all the genetic mutations and genomic instability are effects, rather than causes of cancer (see later).

We shall look at cancer prevention and cure in depth in this special series of articles.

Personalized medicine in jeopardy?

Clearly, the lab findings create practical problems for personalised medicine in cancer therapy, as pointed out by both commentator and researchers [4, 5]. Sampling bias in biopsies could fail to identify key cancer markers and contribute to selection of drug resistant clones, or else fail to predict drug resistance to therapy. 

Despite that, neither the Editorial nor the researchers give up hope on personalized medicine. The identification of common mutations in the trunk of the tumour’s phylogenetic tree confirm that the genetic lesions in the original tumour cells are consistently expressed, such as the von Hippel-Lindau gene in renal-cell cancer, and may be a more robust target for therapy. In addition, the genes affected by convergent evolution may be suitable targets for functional inhibition or restoration.

“However”, the Editorial concludes [4], “the simple view of directing therapy on the basis of genetic tumour markers is probably too simplistic.”

Cancer is not a genetic disease

There is, of course, the possibility that the genetic approach is misplaced. The gene mutations, even those in common ‘gate-keeper’ genes could be effects of a more fundamental cause. This is entirely likely given the fluidity of the genome, the ease with which genes can be silenced or activated, and both RNA and DNA sequence changes can occur in response to the environment as described in detail in my book [6]. It would also be consistent with the evidence that the causes of cancers are overwhelmingly environmental. An increase in somatic mutation rate provoked as the result of a stress response, for example, could explain why numerous different mutational changes are typically found from one individual cancer patient to the next, and even within a single tumour. Personalized medicine in cancer therapy may well be extremely time-consuming and costly, if not downright misdirected. Cancer cells under attack in one pathway can switch to another pathway, or else develop drug resistance that enable them to survive and multiply, as bitter experience in cancer therapy has revealed [7] .

There is evidence in support of the view that cells become cancerous as the result of epigenetic ‘adaptive’ mutations in response to chronic stress or environmental stimuli that promote cell proliferation ([8] Cancer an Epigenetic Disease, SiS 54).

Furthermore, by far the most general manifestation of cancer is an abnormality in energy metabolism ([9] Cancer a Redox Disease, SiS 54), which may lend itself to affordable and safer therapies for all (see [10] Does DCA Cure Cancer? SiS 54).

Article first published 02/04/12


  1. Chemotherapy, Wikipedia, 20 March 2012,
  2. FDA-Approved Drugs
  3. “New cancer drug sabotages tumour’s escape route”, Andy Coghlan, NewScientist 24 February 2012,
  4. Longo DL. Tumor heterogeneity and personalized medicine. NEJM (Editorial) 2012, 366, 956-7.
  5. Gerllinger M, Rowan AJ, Horswell S, et al. Intratumor heterogeneity and branched evolution revealed by multiregion sequencing. NEJM 2012, 366, 883-92.
  6. Ho MW. Living with the Fluid Genome, ISIS/TWN, London/Penang, 2003.
  7. Garraway LA and Jänne PA. Circumventing cancer drug resistance in the era of personalized medicine. Cancer Discovery 2012, published online 28 February 2012; doi: 10.1158/2159-8290.CD-12-0012
  8. Ho MW. Adaptive mutations & carcinogenesis, Science in Society 54 (to appear) 2012.
  9. Ho MW. Cancer an electronic disease? Science in Society 54 (to appear) 2012.
  10. Ho MW. Does DCA cure cancer? Science in Society 54 (to appear) 2012.

Got something to say about this page? Comment

Comment on this article

Comments may be published. All comments are moderated. Name and email details are required.

Email address:
Your comments:
Anti spam question:
How many legs on a tripod?

There are 2 comments on this article so far. Add your comment above.

Rory Short Comment left 3rd April 2012 01:01:06
A very interesting idea indeed and it certainly links in my mind with the increasing level of environmental pollutants.

Dr. Michael Godfrey Comment left 3rd April 2012 05:05:21
The continued search for the patentable "quick fix" and the implied cause being the patient's so-called genetic defects will continue to fail as the accumulating evidence confirms that at least 90% of cancers have environmental causes. As one example - high levels of (dental)transitional metals and xenoestrogens (parabens)have both been found in breast cancers.