Pharma Strategy Blog

Breast cancer isn’t a topic I cover very often on this blog, mainly because there is so much written about it elsewhere, but I confess to being fascinated by the ongoing work on cancer cell seeding and metastases from Joan Massague and Larry Norton at Memorial Sloan Kettering.   The reason for this is that the origins of metastases might have wider applicability to other cancers, so the studies as a body of work are important in the field.   For those of you looking for some background on this important topic, check out this previous post on cancer cell seeding first.

Unfortunately, while at AACR the other week I lost track of time in the huge poster session one morning and missed Joan Massague’s plenary talk, but hopefully it will be available on the webcasts later this month, and I will post another update then.

Where are we now?

Science Translational Medicine

In the latest research, published last month in Science Translational Medicine, the researchers used Genome-Wide Analysis Studies (GWAS) to look at methylome signatures in breast cancers to see if there was any correlation between different patterns and survival.

We know that widespread changes in DNA methylation patterns occur during oncogenesis and tumour progression, so detailed research using GWAS may help us uncover patterns that ultimately help us unravel the heterogeneity of the disease.   The more homogenous the subgroups, the greater the chances of successful treatment and improved outcomes for patients down the road.

To this end, samples from breast cancer tumours (discovery set, n=39; validation set, n=132) were gathered from patients at MSKCC and analysed.

What did they find?

In the current study, Fang et al., (2011) noticed that one pattern in particular was associated with low metastatic risk and hence improved survival, ie the breast CpG island methylator phenotype (B-CIMP).  The paper is well worth checking out for the survival curves of B-CIMP+ vs. B-CIMP- alone.  Beautiful.

Interestingly, this trend was independent of other known breast cancer markers including estrogen receptor/progesterone receptor (ER/PR) and human epidermal growth factor receptor 2 (HER2) status.  They also found similar results i.e. presence of B-CIMP genes in other human tumour types, including glioma and colon cancer, which are widely available in the Cancer Genome Atlas.

What do these findings mean?

There are several important findings that emerge from this work.

1. Discovery of a global B-CIMP gene that is associated with improved survival helps identify a group of patients who are likely to do better
2. B-CIMP phenotype may play a mechanistic role in metastatic risk
3. Presence of B-CIMP across multiple tumour types suggests that it targets the same genes across different cancers

The authors concluded that:

“Our findings may enable the development of new molecular diagnostics that more accurately reflect the epigenomic underpinnings of breast cancer prognosis.  These diagnostics may help further refine our ability to implement personalized medicine for breast cancer patients.”

If that scenario happens, it will be an amazing discovery in the development of breast cancer research and treatment and may spur further innovation through improved targeted agents, diagnostics and patient selection.

References:

Fang, F., Turcan, S., Rimner, A., Kaufman, A., Giri, D., Morris, L., Shen, R., Seshan, V., Mo, Q., Heguy, A., Baylin, S., Ahuja, N., Viale, A., Massague, J., Norton, L., Vahdat, L., Moynahan, M., & Chan, T. (2011). Breast Cancer Methylomes Establish an Epigenomic Foundation for Metastasis Science Translational Medicine, 3 (75), 75-75 DOI: 10.1126/scitranslmed.3001875