 |
| This slide was taken from: Sandhu et al., Microarray-Based Gene Expression Profiling For Molecular Classification of Breast Cancer and Identification of New Targets for Therapy. These are hematoxylin and eosin stained breast cancer sections from different patients displaying distinct types of breast cancer. First pay attention to the histology without the brown stain on the far left. Try to notice differences between Luminal A, Luminal B, Her2, and Basal-like types. Now look at the rest of the figure. Note that the blue is marking the nuclei of individual cells and the brown is marking a specific cellular target: on the second column to the left the antibody is against the estrogen receptor, the next column the antibody is to the progesterone receptor, the next column the antibody is against the HER2 receptor tyrosine kinase. In the article, you will see that these three markers have been the main ones used to distinguish breast cancer types and determine a therapeutic approach. Also, please be careful, because this slide represents an extremely general view of these breast cancers. Imagine, that each subtype has enormous histological heterogeneity, as well as mRNA heterogeneity which you will read about in the articles. For example, Her2 might be expressed in certain luminal types, and might even be devoid in some "Her-2 types". Hmmm...Fathom that! |
We are going to take a skip to Cancer Genetics. So after finishing chapter 10 on Chromosome Structure, please read: Chapter 22 (focus on section 22.4, The Genetic Basis of Cancer). You should also read sections 1 & 2 in Chapter 21 (focus on the DNA Microarray technique, The Proteome, and Protein Microarrays) and revisit Chapter 20 pages 559-565 on Next Generation Sequencing.
Instructions for Discussion on November 2nd:
In the beginning of this month, a landmark paper was published in Nature for Breast Cancer Research. (Thanks Mo for passing it to me) A global consortium of cancer biologists worked together using six high throughput platforms with a common goal: to identify novel targets in breast cancer and to attempt to better define the four emerging breast cancer subtypes. I have posted this article (but not the supplements, please go to Nature directly for those if you have time) and a review article on breast cancer subtypes on Black Board (with link below).
Instructions for the blog:
Please remark on one interesting aspect of breast cancer that you have learned for the first time from reading the Nature article and/or the review. For class: please identify one finding in the paper that seems to be novel for breast cancer research. It may be a missense mutation specific to one of the breast cancer subtypes, or a technique that was used in their study in a very different/comprehensive manner, or whatever you find interesting. Be prepared to share with the class your interesting finding.
For Nature article, Click Here: Breast Cancer Molecular Portraits_Nature Article
For Breast Cancer Review Article (a softer introduction to the topic), Click Here: Review article on Breast Cancer Subtype Array Analysis_Lab Medicine Journal
I learned that different types of breast cancer are generally classified based on immunohistochemical staining of protein biomarkers. What a clever way of getting around doing gene expression analysis!
ReplyDeleteI never thought that I would want to become a pathologist, but after reading this article and realizing how important it is to correctly identify the type of cancer for patient treatment AND for accurate data for further breast cancer research, I realize that maybe, just maybe, pathology might be in my future.
One thing I learned that was interesting was how the researchers found novel mutated genes. These previously unidentified genes included TBX4 and TBX5. TBX5 mutations cause Holt-Oram syndrome.
ReplyDeleteI've met a patient with this syndrome and it has really fascinating phenotypic / physical manifestations. HOS involves underdevelopment of upper body parts (as well as bone deformations) and can cause holes in aortic tissue. It affects far more females than males. There has to be some connection between this autosomal dominant mutation and breast cancer. If the limbs and organs of the torso are affected it makes sense that breast tissue would also demonstrate malfunction (or cancerous mutations). I wonder if other developmental diseases like this also have a higher rate of cancers and what kind of cancers they are.
Very interesting comment Mo! I too looked up TBX5 when reading the article and was curious about any potential relationship between Holt-Oram and breast cancer. It raises an interesting question about tissue specificity both in how gene mutations might generate a range of abnormalities (bone deformity to cancer) and why many (all?) cancers have a tendency to metastasize preferentially to certain tissues.
DeleteI did not know that breast cancer had 4 subtypes, and thus I didn’t know about the “good” kinds (easy to treat) and the “bad” kinds (difficult to treat, poor prognosis); I am referring to the subtypes with the estrogen and progesterone receptors and the subtypes without them. It is interesting and scary that the more lethal basal-like breast cancer can kill patients about 4 years after the diagnosis and can quickly damage the brain and lungs instead of the liver and bones. Though, I am curious to know why hypermethylation of certain genes are common in the basal cancer considering that DNA methylation inhibits transcription and thereby silence genes; or I cannot tell if the cancer persists as a result of the silencing of the genes. This cancer also affects women of “certain age and ethnicity,” (which I assume that women who are not white are more prone to the disease – cannot say anything about age ranges), I am interested to know why does it generally target certain ethnicities.
ReplyDeleteA lot of the review article is focused on the basal-like breast cancer, but I am curious to learn how this cancer compares with the HER2+ breast cancer (the other estrogen/progesterone receptor negative cancer), especially because the receptor-deficient cancers are harder to treat than there receptor-positive counterparts.
From a more positive outlook, considering the amount of detailed research done for breast cancer (particular for the more dangerous basal-like breast cancer), I am very optimistic for medicine of the next generation in terms of therapy/treatment evolutions and revolutions. Maybe it can be one of us to successfully decrease the incidence of the cancer victimizing 1.2 million women.
One of the things I learned was that the BRCA1 gene is implicated in many subtypes of breast cancer, not just one. We learned in class that BRCA1 mutations can lead to cancer since BRCA1 expression allows for DNA repair and degradation of DNA that can't be repaired.
ReplyDeleteIt is interesting how one gene can be associated with several different mechanisms that lead to several different phenotypic expressions, including completely different diseases. The fact that many breast cancers involve BRCA1 could suggest similar treatments for these cancers that target the specific BRCA1 gene mutation involved. It would be interesting to compare cancers resulting from different BRCA1 mutations and the therapies that seem to best treat them.
Reading this article made me realize just how little I know about breast cancer! I didn't even know about the three basic therapeutic groups, let alone the four mRNA-expression subtypes. I also naively thought that since the BRCA1 and BRCA2 had been identified we would be closer to having targeted treatments for cancers caused by mutations on these genes. I learned instead that these tumors fall into the triple-negative category and are actually among the hardest to treat. It's a good reminder that identifying a genetic cause of cancer doesn't guarantee an easily identified treatment.
ReplyDeleteAnother aspect I found interesting was how many of the novel mutated genes researchers identified were implicated in other diseases. It makes me wonder what else is contributing to seemingly unique diseases if they share common genetic origins. For example, if the driving events for basal-like and serous ovarian tumors are so similar, what determines whether an individual gets breast cancer or ovarian cancer?
I think I'm just as astounded as you guys are at the sheer number of possible variations of breast cancer there are. The similarities found between basal-like tumors and serous ovarian cancer remind me of the first article posted on this blog, where Dr Wartman's leukemia was treated with drugs usually prescribed for advanced kidney cancer.
ReplyDeleteThis study and many others like it seem to suggest that we need to radically revolutionize the way we classify cancers. It appears that it is no longer sufficent to claim someone has 'skin cancer' or 'bowel cancer' or 'leukemia' simply because their cancer is in the skin or the bowel or the blood. The most interesting thing that I took away about breast cancer from this article is probably that breast cancer may no longer necessarily be breast cancer at all. As future physicians, there is no doubt in my mind that amendments to how we classify cancer will have significant impacts on how we choose to diagnose it.
This article showed me how little I know about breast cancer, and the research and technology revolving around therapy for it. Learning about the different sub types was definitely new to me, but I'd rather mention the cell receptor-marking method. Awesome idea, how did they even know that cancer cells differ in that respect? I'm very happy to have learned a little bit about targeted cancer therapies, as well as the struggles faced by those with triple negative breast cancer. The science that unfolded as the researchers probed further into the identification of various basal-like breast cancers was inspiring. I also thought it was nice how they organized all of the variables being looked at (i.e. morphological features, clinical behavior of basal-like breast cancers, etc)into their own sub sections of the article. Thanks for posting.
ReplyDeleteThis comment has been removed by the author.
ReplyDeleteThis article seems to be an outstanding contribution to the field of cancer research. Not only have the authors successfully classified and characterized the three types of breast cancer, but have identified over 30,000 mutations present in just over 500 breast tumors. (It must have been incredibly difficult to decide which of these mutations was relevant.) This documentation is not only very relevant for the purposes of designing breast cancer specific treatments, but also for knowing what mutations to look for in analyses of other cancers. I was surprised by the number of overlaps in mutations found in different cancers. Mutations associated with more than one "type" of cancer include TP53 (breast, brain, leukemia, etc.), CDH1 (breast, gastric), BRCA1 / BRCA2 (ovarian, breast), CHEK2 (colon, prostate, breast), and PIK3K1 (glioma, endometrium, breast?). I am sure these are only a few examples of a pool of many. Luke is right, it seems we will have to start classifying cancers on their genetic basis rather than the body part they affect.
ReplyDeleteAs we discussed in lecture today, the ever changing state and composition of breast tissue makes me curious about how these results varied for different age groups. If therapies were designed based on the findings in this paper, they would no doubt have to be tailored to women based on their age, whether or not they've had a child, and whether they've gone through menopause. I wish I knew more about how breast tissue changed over time and why it seems that some carcinogens (tobacco smoke for example) seem to disproportionately affect young women before they've had a child. Smoking is so insidious. Studies now are pointing to a link between breast cancer and second-hand smoke.
Also, I didn't realize that estrogen and progesterone (levels and receptors) played such a large role in breast cancer. This has me thinking about potential dangers/benefits/unknowns of hormonal birth control (which has been associated with some increases in breast cancer but decreases in other cancers including cervical).
DeleteHi, guys! I can't help but mention first the new-ish book "Breasts," by Florence Williams. The scientific journalism left me, among other things, astounded at the utter lack of big chem regulation in the US. Even if environmental endocrine disruption isn't the primary cause of breast cancer and other illnesses, the ubiquity of these effects from consumer products and our general environment in the US surely contributes to bodies gone awry. Estrogen and progesterone receptors do play a role, as Allison points out, and they are surely affected by proven endocrine disruptors that show up in everything from couch cushions to cosmetics - not to mention hormonal birth control!
ReplyDeletePhew. Glad to get that off my chest, so to speak.
As for new knowledge from the article, approximately one fact was novel to me in every sentence. (Thanks, Mo!) I'm most pleased to recognize the different varieties of mutation per different mRNA subtypes (page 62). Class work had implied some mRNA-expression subtypes, and I had subconsciously assumed that they varied only by mutation type, not by number of mutations. When such distinctions are integrated for classification in Figure 2 data (page 64), I think we're justified in soothsaying new approaches to "where" cancer really is in the body.
I am curious about what subtypes the study elected not to include, and why - poor sample size? inconsistent results, implying a lack of causation? I also admit that the further along I read, the less well-equipped I felt for commenting. I'll keep re-reading until Friday. :)
Colleen, thank you for mentioning this new-ish book in your response. I am going to have to check it out.
DeleteThe idea that consumer products can result in hormonal imbalances has always interested me and has been heightened after reading this article. As the supplemental review article mentioned, early-onset menarche is a high-risk factor for the development of basal-like breast cancers. With new information coming out about consumer products that are thought to have an impact on the age of first menarche, as well as other hormonal imbalances, this is very scary when considering the impact it may have on the development of cancer.
Diet and its impact on overall health is an area I am very interested in. I have currently been reading about how the standard American diet is very high in estrogen. This makes me wonder if following a low-estrogen diet could have an impact on the development of breast cancers since estrogen receptors play a role in many of the subtypes.
First and foremost let me be clear, I don’t know much about cancer at all, let alone breast cancer. However, after reading this article I admit that I can recognize certain components of cancer that are hallmark indicators of tumor-suppressor genes and maintenance genes.
ReplyDeleteOf the information found in the study, I discovered that breast cancer is second only to lung cancer in mortality, which is something that surprised me when compared to all of the other deadly types of cancer. I was also interested to find out that the distinct risk factors of breast cancer include early-onset menarche, younger age at first full-term pregnancy, high parity combined with lack of breast feeding and abdominal adiposity. Overrepresentation among premenopausal, African-American women is also surprising, although breast cancer is present in all ages and ethnicities. Lastly, I was intrigued by the fact that basal-like breast cancers have a tendency to disseminate through hematogenous routes, involving the brain and lung, and are less likely to spread to the lymph nodes, liver or bones. I believe that this last finding will have the most impact (of all of the interesting things I listed) in discovery pathways and treatments that cure or increase the survival level of patients suffering from basal-like breast cancer.
Knowing next to nothing about breast cancer going into this article, I'm finding it difficult to choose just one interesting aspect I've learned. The subdivision of breast cancer types based on receptor expression is amazing, especially in its implication for further subdivision within those groups. If I understand correctly, the particular difficulty in treating triple-negative breast cancer (like the basal-like type focused on in the article) is an ostensible lack of overexpressed products to trace the malignancy to and target in treatment. But I imagine an endlessly differentiable tree of cancer gene product expression profiles, each combination representing a unique pathology. Microarray-based gene expression profiling gives us the opportunity to descend further than ever into that tree and identify more molecular targets, with the potential for individually-tailoring treatments to use the cancer cells' defining features against them while sparing as much healthy tissue as possible. I believe the "patient stratification" mentioned in the article addresses this idea. Though practical constraints may preclude such extreme specificity, finer-grained grouping of cancer subtypes is an exciting road ahead.
ReplyDeleteAs everyone has already mentioned, I was truly amazed at how many sub-types of breast cancer exist and the differences in their response to treatment. However, as I learn more about genetics in this class, the differences in the genetic basis of a specific cancer should not surprise me with how many different things can go awry in our genome.
ReplyDeleteI think the most interesting thing that I learned from this article was how many gene mutations that were found in breast cancer cells have also been linked to other diseases. As Luke mentioned, this idea of tissue specific cancer may be an idea of the past as we continue to unravel the genetic basis of cancer. This could have a profound impact not only how physicians will approach treatment, but also impact how funds will be raised for cancer research.
An additional point that I found to be interesting is that basal-like breast cancers are not linked to the clinical parameters that other cancers utilize for prognosis, including tumor size, tumor grade, lymph node status, and the presence of distant metastasis.
I wish that I had more time to do some research on the genes that were mentioned in order to learn more about the roles and pathways that their products have in normal function. The photograph collage at the top of the page kind of reminded me of a quilt and that led me down the thought pathway that the investigators of the study seemed to be gathering many small pieces of a big puzzle and by doing such a huge study, they've been able to sort a lot of the pieces (the pieces being the variety of genes and their behaviors) into a clearer outline that allows for more tangible hope for being able to identify and treat the real roots of the disease.
ReplyDelete