April 2013

Triple-negative breast cancer: a particularly recalcitrant foe

This month, thousands of researchers from industry, government and academia will gather at the annual meeting of the American Society for Biochemistry and Molecular Biology in Boston. Among them will be a number of scientists attending a thematic symposium on breast cancer with an emphasis on survival health disparities and the biology of triple-negative breast cancer, one of the most difficult breast cancers to treat and one that affects black women disproportionately. For this report, ASBMB Today contributor Connor Bamford provides an overview of our understanding of TNBCs. 

Tracking down a disease

“If it were all down to genetics, then we would see a lot more cases of TNBC than we do currently,” muses Fatimah Jackson, a professor of biological anthropology in the anthropology department at the University of North Carolina at Chapel Hill, who is looking at the often-understudied and complex genetic history of those hit hardest by triple-negative breast cancers. Jackson believes that environmental triggers, in conjunction with susceptibility genes, kick off the beginnings of this cancer in certain people.
Some populations across the world have lower rates of TNBC than other breast-cancer subtypes; for other populations, TNBC rates are highest. To figure out how TNBCs work, researchers agree they must look closely at the populations most affected. In this case, they turn to those patients of recent African descent. If researchers can more easily find what factors are responsible for patients’ increased risk, then those of recent African descent could provide clues that could unlock therapies for all other populations.
Jackson, who will be one of the featured speakers at a forum on TNBCs orchestrated by the American Society for Biochemistry and Molecular Biology in April in Boston, says she is motivated not just by the often-incurable nature of the disease but also by the history of African-Americans. KiTani Parker, co-organizer of the TNBC forum, explains why Jackson’s approach is so interesting: “Using a multidisciplinary approach to answer a basic science question provides students with a model of how to think outside the box in order to address complex scientific challenges.” Parker’s message is “Don’t limit yourself.”
One way to study the role of the environment in the history of breast cancer and to identify potential toxins, medicines or diets acting as triggers is known as ethnogenetic layering — which uses a combination of geographical, anthropological and genetic data.
Jackson is focused on understanding how past ethnic population structures from before the transatlantic slave-trade period in West and West Central Africa and among enslaved Africans in North America influence susceptibility to TNBC in certain ethnic peoples’ modern descendants. By combining often-scarce historical information, such as slave-ship and family records, Jackson and her co-workers can connect people in various regions of the United States to contemporary micro-ethnic groups in Africa through their shared ancestries. Genetic-association studies combining medical data and DNA-sequence information from those on both continents identify specific genes contributing to high TNBC susceptibility. But we shouldn’t expect a one-gene-fits-all answer, Jackson says: “We don’t yet have a good sense of the distribution of these genes, yet we know that they’re not uniform across African-Americans.”
Considering the far greater genetic diversity of those of recent African descent compared with those of Asian or European ancestry, Jackson and co-workers are forced to take a focused perspective at the level of local populations to understand the role of genetics in this disease. For example, in a 2008 paper, the technique allowed the ancestral mapping of African-Americans in the Chesapeake Bay area with a high risk of TNBC to specific ethnic groups in Nigeria, Gabon, Equatorial Guinea and Cameroon on the west coast of Africa (a British protectorate in the 1800s), some of whose ancestors were brought over to the Americas during the slave trade (1). It is in those specific ethnic groups that you would want to investigate for susceptibility genes, Jackson says, and it is those specific groups on which she is focusing.
Now that traditional interethnic population barriers for marriage and childbearing are no longer so strong, this change in patterns of gene flow makes genetic tracing difficult without a thorough knowledge of the substructure of complex contemporary groups (2). But Jackson insists it’s still a promising line of study. “The chance to decouple ethnic inheritance, which is a cultural idea, from genetic inheritance, which is a biological idea, can be very empowering,” she says. Yet she believes that it is a combination of both anthropological and genetic methods that will allow researchers to truly understand how this disease begins and how it kills.
  1.   1. Jackson, F. L. C. American Journal of Human Biology 20: 165 – 173 (2008).
  2.   2. Jackson, F. L. C. Annals of Human Biology 35 (2): 121 – 144 (2008).

It was with the National Cancer Act of 1971 that President Nixon declared what became known as the War on Cancer and what then looked like the beginning of the disease’s end. This modern — and often optimistic — struggle against the disease grew out of the work of William Halstead and his contemporaries, such as Wilhelm Rontgen and Sidney Farber, who discovered the three pillars of our anticancer strategy: surgery, radiotherapy and chemotherapy.
By the time the National Cancer Act came around, we craved new and improved therapies. Toward the end of the 20th century, as molecular biology emerged, we found ourselves attempting to generate more targeted cancer drugs — small molecules and, more recently, monoclonal antibodies directed specifically at key proteins expressed by cancer cells (1).
At their core, cancer cells behave like an altered version of normal cells that have acquired novel characteristics but still show substantial similarities with unaltered cells. This makes it exceptionally hard to find molecular targets that only cancer cells have. It’s hard, but it’s not impossible. This, basically, has been the story of cancer research for the past four decades.
Breast cancer under the microscope
As Paul Mullan explained to me in his second-floor office on a typically rainy Belfast morning late last year, “I think breast cancer in general is an excellent paradigm for how advancing cancer research really pays off in terms of patient survival.”
Mullan, a principal investigator at Queen’s University Belfast’s Centre for Cancer Research and Cell Biology, has been working on breast cancer for the past 15 years. Each successive revolution in biology has been rapidly applied to cancer research, he explains, and each has illuminated new aspects of the disease. Despite this, there are many unanswered questions about certain groups of patients, and these are the questions that interest Mullan and his colleagues most.
For example, researchers now know that there are many kinds of breast tumors, and some can be distinguished easily because they either express or lack the expression of certain marker proteins. For three types (there are at least five distinct types), researchers have created therapies targeted toward three molecules: the Her2 protein, a growth-factor receptor tyrosine kinase that can be targeted by a humanized monoclonal antibody called Herceptin (trastuzumab), and the estrogen or progesterone hormone receptor proteins, which are tackled with endocrine therapy. But what about the breast cancers that fail to express those targets?
Triple-negative breast cancer, as it is now known, is a particularly recalcitrant foe. KiTani Parker, co-organizer of the ASBMB program on TNBC and a member of the society’s committee focused on minority affairs, says bluntly, “The prognosis for this subset of breast cancer patients is poor.”
There is very little that can be offered on top of traditional surgery and toxic chemotherapy for TNBC patients, because their tumors lack the three molecules for which there are drugs to act against and also because they are very heterogeneous and have unpredictable clinical and biological behaviors.
What’s worse, while these cancers respond well to chemotherapy initially, they are aggressive, and there is a high risk of them returning and becoming resistant to common chemotherapies.
“It is critical that a multifaceted approach be applied to address this health issue, which is also health disparity in the African-American community,” says Parker. TNBC is more common in premenopausal and minority women, and it is particularly deadly in premenopausal black women. While more white women develop breast cancer per capita, more black women die from the disease. Parker adds, “Application of the technology and methodology is, in my opinion, best appreciated when it comes full circle to benefit human health.”
Enter the genome
Chuck Perou, distinguished professor of genetics and molecular oncology at the University of North Carolina at Chapel Hill, has been investigating the basic biology of breast cancer since he was a postdoctoral fellow in the laboratory of geneticist David Botstein at Stanford University in the late 1990s. “Over the last decade, the community has come to appreciate that breast cancer is multiple diseases — not a single disease,” says Perou.

clustering of normal breast tissue and tumors 
A basal tumor stained with CK14, which can be used as a marker to identify this subtype. Click on the image in order to see it more closely. | Credit: Courtesy of Chuck Perou  

During his years at the Botstein lab, Perou and colleagues exploited then-new microarray technology, offering the scientific community a much more sound understanding of how these cancers work. Perou was integral in the investigation that first uncovered the sheer diversity of breast cancer subtypes (2).
While researchers had known that some cancers responded differently to therapies than others, and while they could infer that the cancers differed on fundamental levels, the biology of these cancers had never been interrogated before on such a large scale. The Botstein team and others took advantage of the data and reagents generated from the Human Genome Project (just coming to an end in the late 1990s) to see how differently the cancers expressed their genes. “The genome project set the stage by identifying all the genes and actually gave us the boost in the early phase by providing us with a resource of cDNA clones for each of the genes. And that’s what we spotted onto the microarrays,” Perou says. Advances in both computational clustering approaches and microarray generation allowed Perou and colleagues to observe objectively the expression of thousands of genes across the genome.
Studies like Perou’s and more recent whole-genome sequencing efforts have uncovered the existence of not one or two kinds of breast cancer but of at least five biologically distinct disease subtypes characterized by both unique gene expression patterns and their marked differences in acquired somatic mutations. One of these subtypes, termed basal-like breast cancer, showed large similarities with TNBC. Although TNBCs and basal-like breast cancers are not necessarily the same disease, a majority of TNBCs have the characteristics of basal-like breast cancers.
When Perou’s group looked at how patient groups responded to existing treatments, it was those patients with tumors clustering within the basal-like subtype who fared worst. Many basal-like patients relapsed within two to three years of diagnosis, while those with the luminal A subtype had an 80 percent chance of three-year survival (34). “We’re like, ‘God, we’ve got to give them more than chemo!” says Perou. “If we’re ever going to develop therapeutics against the triple-negative diseases, it has to work on the basal-like subtype.”
Despite the evidence that basal-like breast cancers, as the most common of the TNBCs, are aggressive, it has taken clinicians a while to appreciate fully the existence of the basal-like subtype, Perou says. But he remains confident that the clinical benefits of his and others’ work will emerge: “It doesn’t matter whether you’re triple-negative or (specifically the basal-like subtype TNBC) — you still get the same treatment, which is disappointing, and I’m going to take it as a failure of our research. But I think we’ll change that in a few years.”
Stefan Ambs, a senior investigator at the National Cancer Institute’s Laboratory of Human Carcinogenesis, notes that there is also evidence from whole-genome analysis that the type and number of acquired mutations in a breast tumor varies substantially between
individual tumors (5).
Somatic mutations in only few genes (TP53, PIK3CA and GATA3) occur commonly across all breast cancers. In contrast, there are numerous subtype-associated mutations, with certain mutations being common in estrogen receptor-positive tumors (e.g., PIK3CA) while others, like those in the p53 tumor suppressor gene, are most frequently observed in basal-like tumors and TNBC. Furthermore, the number and type of mutations may also vary greatly between TNBC tumors, leading to large disease heterogeneity, with some tumors having only a few mutations whereas other tumors may have acquired many mutations at the time of diagnosis (6).
“To make matters worse,” Ambs explains, “only some parts of a tumor may contain certain mutations, while other parts of the same tumor may carry different mutations. This intertumor and intratumor heterogeneity shows us that understanding the biology and therapeutic responses of patients with TNBC may require in-depth sequence data for these tumors.”
Ambs emphasizes that others have confirmed the landmark findings by Perou and co-workers and have shown that TNBC and specifically the basal-like breast cancer subtype cause an excessive mortality within the first five years after diagnosis. But the disease, when established, may not behave more aggressively in black patients than in white ones (7, 8, 9).
The over-representation of TBNC in women of African descent has been further confirmed in a study of breast cancer patients from Nigeria and Senegal (9) showing that a large proportion of breast cancer patients in these countries present TBNC. “Future studies are needed that seek an increased understanding of the causes of TBNC in women of African ancestry, while other research should dissect the disease heterogeneity of TBNC and how this increased understanding of disease heterogeneity can be translated into improved therapeutic approaches for TBNC,” Ambs says.

breast cancer chart 

A needle in the haystack
One of the important clues as to the origins of some basal-like tumors came when their genome-expression patterns were compared with those from other breast cancers from patients with the defective form of the BRCA1 gene, which normally encodes a protein involved in genome repair. These inherited recessive mutations, which increase the risk of developing breast cancer and ovarian cancer, force many carriers to undergo elective removal of both breasts and ovaries.
Strikingly, in one study by Perou and colleges, every BRCA1-mutant carrier who had a breast tumor was clustered among the basal-like subtype (10). The finding afforded a poor prognosis for patients with BRCA1-defective tumors but promisingly offered up a potential treatment: chemotherapy that would enhance DNA damage that BRCA1-negative cancers couldn’t fight (11). Fortunately, a number of companies had developed a diverse array of such molecules in the 1960s: poly (ADP-ribose) polymerase inhibitors. A couple of these PARP inhibitors have progressed into phase III studies for the treatment of BRCA1-associated ovarian cancers and even some lung cancers. However, over the past few years, they have yielded disappointing results in terms of directly treating breast cancers, according to a recent phase III report (12), and there remains a need for a more sophisticated approach.
The development of PARP-inhibitor resistance and the potential for toxicity in patients lacking complete genome-repair pathways have contributed to this dissatisfaction. Also, with PARP inhibitors’ use in basal-like cancers, only 10 percent to 15 percent of such tumors involved BRCA1 dysfunction (5). Perou, for one, believes there are other more general targets in well-characterized cancer-promoting mutations, such as loss of key tumor-suppressor proteins like p53 or even the amplification of tumor-promoting genes like MYC. The problem is that researchers have been trying to develop drugs against those for years with no luck. Yet Mullan believes that we shouldn’t forget about these agents: “accurate stratification may help uncover other driver oncogenic pathways and assist in the development of novel therapies for the remaining DNA damage repair proficient TNBC tumors.” So for now, the search for more treatments for the rest of these cancers continues.
To the future
The challenge for this generation of cancer biologists and their successors is to find and validate the biomarkers that define these categories and test the effectiveness of new drugs and drug combinations on each of them.
“By aligning treatments to markers such as the estrogen and HER2 receptors, we have dramatically increased the survival rates,” Mullan says. “Now successful treatment of TNBC is the next big hurdle.”
While some form of personalized medicine will be required to diagnose accurately and provide a correct prognosis for each patient, Perou suspects that much of the data extracted from personal genomic tests in the clinic will provide little immediate clinical benefit. “With breast cancer today, you have three treatments. We’re not going to jump to 100 different treatments, which is what is suggested with personalized medicine,” Perou says. “I think we’ll go from three to five to 10, and we’ll put people into these categories — and that will be a significant improvement over what we have now. The thought of having 100 different treatments is beyond the curve of the current medical system.”
Parker emphasizes that “the complexity of breast disease requires that a multidisciplinary approach be incorporated to better understand this subtype of breast disease.”

  1.   1. ASBMB Today’s “Right on Target,” August 2012.
  2.   2. Perou, C. M. et al. Nature, 406: 747 – 752 (2000).
  3.   3. Sørlie, T. et al. Proc. Natl. Acad. Sci. U.S.A. 98: 10869 – 10874 (2001).
  4.   4. Nielsen, T. O. et al. Clin. Cancer Res. 10: 5367 – 5374 (2004).
  5.   5. The Cancer Genome Atlas Network, Comprehensive molecular portraits of human breast tumours, Nature, 490: 61 – 70, 2012.
  6.   6. Shah, S. P. et al. Nature, 486: 395 – 399 (2012).
  7.   7. O’Brien, K. M. et al. Clin. Cancer Res., 16: 6100 – 6110 (2010).
  8.   8. Dawood, S. et al. J. Clin. Oncol., 27: 220 – 226 (2009).
  9.   9. Huo, D. et al, J. Clin. Oncol., 27: 4515 – 4521 (2009).
  10. 10. Sørlie, T. et al. Proc. Natl. Acad. Sci. U.S.A., 100: 8418 – 8423 (2003).
  11. 11. Fong, P.C. et al. N. Engl. J. Med., 361: 123 – 134 (2009).
  12. 12. O’Shaughnessy, J. et al. N. Engl. J. Med., 364: 205 – 214 (2011).

Connor BamfordConnor Bamford (connorggbamford@gmail.com) is a Ph.D. student at Queen’s University in Belfast, U.K.

found= true2273