DIVIDE AND CONQUER

 

By Steve Mason, Ph.D.
 

The future of cancer therapy is steadily moving toward improved targeted therapies, enabling doctors to maximize efficacy by tailoring treatment to a patient’s specific cancer subtype and mutational landscape.  Several recent studies have examined the genome-wide prevalence of mutations in different tumor types (1) and their resulting metastases (2), but application of these findings has been hampered by the limited ability to correlate the mutational analysis with changes in gene expression.  In particular, one major difficulty is separating tumor-promoting ‘driver’ mutations from inconsequential ‘passenger’ mutations (3).  As driver mutations may be therapeutic targets, it is important to determine how specific mutations affect tumor behavior.  Recently, in Nature, Curtis et al. reported a correlated analysis of DNA and RNA profiles from breast cancer tumors, revealing new tumor subtypes likely driven by specific mutations, providing much-needed insights for targeted therapies (4).



Perhaps the greatest challenge for the successful application of targeted therapies is matching patients with the right treatment (5), which hopefully can be accomplished by determining the tumor subtype.  Breast cancer has traditionally been categorized into four subtypes (basal-like (aka triple-negative breast cancer), luminal A/B or HER2+) by their levels of the estrogen, HER2/neu, and progesterone receptors.  Microarray analyses have enabled the identification of 6 molecular subtypes: luminal-A, luminal-B, HER2, basal-like, normal-like, and claudin-low (6-8).  However, these classifications are not wholly sufficient as there can be wide variation in prognosis within a subtype (9).  There have been some notable targeted therapy successes (the HER2/neu-targeting Herceptin, for one), but being able to pinpoint the exact driver mutations and their effects in a patient’s tumor is an ideal that is still largely unrealized.  Furthermore, tumors in some patients have evolved Herceptin resistance (10), necessitating the further development of new targeted therapies.



Utilizing a dataset of nearly 2,000 tumors from breast cancer patients, Curtis et al. analyzed germline copy number variations (CNVs), acquired copy number aberrations (CNAs) and single nucleotide polymorphisms (SNPs) (4).  They then correlated their genomic analysis with gene expression data, enabling them to differentiate cis-acting variants (mutations that affect their own expression) from trans-acting variants (mutations that affect the expression of other genes).  While trans-acting loci affected a greater number of mRNAs, cis-acting loci tended to have more pronounced effects on their targets, leading the authors to use the strongest cis-associated genes to classify the tumors in their dataset. This resulted in 10 integrative clusters with similar copy number profiles and hazard ratios, indicating relative biological homogeneity within the clusters.



Interestingly, these novel subgroups provide valuable insight into breast cancer progression and prognosis.  The subgroups range from having very little genomic variation and a more favorable outcome to having variations in regions containing possible and confirmed driver mutations and a much worse prognosis.  In the latter case, one of the subgroups had two separate amplicons, one at chromosome 11q13.3 and one at 11q13.5-11q14.1.  This implicates multiple genes as potential drivers, and the authors propose that several genes may act together rather than one oncogene dominating; a situation that may greatly complicate the application of targeted therapies.  Another subgroup, with the worst prognosis, was characterized by amplification of HER2, validating the therapeutic importance of agents like Herceptin.



In contrast, one of the groups with minimal genomic variation and favorable prognosis was enriched for the expression of immune response genes associate with increased lymphocyte infiltration.  The authors note that this fits well with recent findings associating increased tumor-associated CD8+ T cells with a positive outcome (11); it also provides further clinical relevance for data derived from animal models demonstrating a role for activated CD8+ T cells in promoting tumor regression (12-14).  Thus, this new clinical analysis provides further support for research into tumor-targeted immunotherapies.



While the increase from 4 (or 6) to the 10 subtypes may seem intimidating to the uninitiated eye, this combined analysis of genomic alterations and gene expression provides a much-needed window into breast cancer pathogenesis.  By continually parsing out the molecular determinants of cancer, researchers can focus on tackling smaller, more specific problems.  The genes implicated in this work will undoubtedly guide laboratory and drug discovery research toward understanding when, how, and whether to target the implicated genes.  Thus ‘dividing’ the tumors may ultimately lead to their being ‘conquered’.



 

Steve Mason is Senior Editor of Biological Sciences at Cancer InCytes Magazine.



​

References



1. Sjoblom, T., et al. The consensus coding sequences of human breast and colorectal cancers. Science 314, 268-274 (2006).

2. Kabbarah, O., et al. Integrative genome comparison of primary and metastatic melanomas. PloS one 5, e10770 (2010).

3. Carter, H., et al. Cancer-specific high-throughput annotation of somatic mutations: computational prediction of driver missense mutations. Cancer Res 69, 6660-6667 (2009).

4. Curtis, C., et al. The genomic and transcriptomic architecture of 2,000 breast tumours reveals novel subgroups. Nature (2012).

5. Higgins, M.J. & Baselga, J. Targeted therapies for breast cancer. J Clin Invest 121, 3797-3803 (2011).

6. Sorlie, T., et al. Repeated observation of breast tumor subtypes in independent gene expression data sets. Proc Natl Acad Sci U S A 100, 8418-8423 (2003).

7. Prat, A., et al. Phenotypic and molecular characterization of the claudin-low intrinsic subtype of breast cancer. Breast Cancer Res 12, R68 (2010).

8. Perou, C.M., et al. Molecular portraits of human breast tumours. Nature 406, 747-752 (2000).

9. Houdebine, L.-M., et al. Prolactin and casein gene expression in the mammary cell. in Regulation of gene expression by hormones (ed. McKerns, K.W.) 71-92 (Plenum Press Publishing Corp., New York, 1983).

10. Nahta, R. Pharmacological strategies to overcome HER2 cross-talk and Trastuzumab resistance. Current medicinal chemistry 19, 1065-1075 (2012).

11. Mahmoud, S.M., et al. Tumor-infiltrating CD8+ lymphocytes predict clinical outcome in breast cancer. J Clin Oncol 29, 1949-1955 (2011).

12. Gonzalez-Martin, A., Gomez, L., Lustgarten, J., Mira, E. & Manes, S. Maximal T cell-mediated antitumor responses rely upon CCR5 expression in both CD4(+) and CD8(+) T cells. Cancer Res 71, 5455-5466 (2011).

13. Hahn, T., Jagadish, B., Mash, E.A., Garrison, K. & Akporiaye, E.T. alpha-Tocopheryloxyacetic acid: a novel chemotherapeutic that stimulates the antitumor immune response. Breast Cancer Res 13, R4 (2011).

14. Stagg, J., et al. Anti-ErbB-2 mAb therapy requires type I and II interferons and synergizes with anti-PD-1 or anti-CD137 mAb therapy. Proc Natl Acad Sci U S A 108, 7142-7147 (2011).