News Release

Endometrial and acute myeloid leukemia cancer genomes characterized

Peer-Reviewed Publication

NIH/National Cancer Institute

Two studies from The Cancer Genome Atlas (TCGA) program reveal details about the genomic landscapes of acute myeloid leukemia (AML) and endometrial cancer. Both provide new insights into the molecular underpinnings of these cancers with the potential to improve treatment. These studies represent the sixth and seventh in a series of genomes of at least 20 major cancers.

The first study is on endometrial cancer:

Study establishes basis for genomic classification of endometrial cancers; proper categorization is important for choosing the best treatment

A comprehensive genomic analysis of nearly 400 endometrial tumors suggests that certain molecular characteristics – such as the frequency of mutations – could complement current pathology methods and help distinguish between principal types of endometrial tumors, as well as provide insights into potential treatment strategies. In addition, the study, led by investigators in The Cancer Genome Atlas (TCGA) Research Network, revealed four novel tumor subtypes, while also identifying genomic similarities between endometrial and other types of cancers, including breast, ovarian, and colorectal cancers.

These findings represent the most comprehensive characterization of the molecular alterations in endometrial cancers available to date. They were published May 2, 2013, in the journal Nature. TCGA is funded and managed by the National Cancer Institute (NCI) and the National Human Genome Research Institute (NHGRI), both part of the National Institutes of Health.

"With this latest study in a series of 20 planned TCGA tumor type characterizations, more genomic similarities are emerging between disparate tumor types," said NIH Director Francis S. Collins, M.D., Ph.D. "Teasing out heretofore unknown genomic markers or mutations in various cancers is again proving the value of TCGA."

Clinically, endometrial cancers fall into two categories: endometrioid (type I) and serous (type II) tumors. Type I is correlated with excess estrogen, obesity, and a favorable prognosis, while type II is more common in older women and generally has a less favorable outcome. Type I tumors are often treated with radiation therapy, which helps stop or slow cancer growth, given in addition to or after the primary treatment. Type II tumors are generally treated with chemotherapy, in which drugs are used to kill the cancer cells or stop them from growing.

Distinguishing between different types of endometrial cancers is currently based on histology, an examination of a thin slice of tissue under a microscope. But categorizing endometrial cancer tissues is often difficult, and specialists frequently disagree on the classification of individual cases.

In this study, investigators showed that approximately 25 percent of tumors that pathologists classified as high-grade endometrioid showed frequent mutations in TP53, a tumor suppressor gene, as well as extensive copy number alterations, a term for when a cell has too many or too few copies of a genomic segment. Both are key molecular characteristics associated with serous tumors, along with a small number of DNA methylation changes, which are additions of a basic chemical unit to pieces of DNA. Most endometrioid tumors, by contrast, have few copy number alterations or mutations in TP53, though there are frequent mutations in other well known cancer-associated genes, including PTEN, another tumor suppressor gene, and KRAS, a gene involved in regulating cell division.

These data suggest that some high grade endometrioid tumors have developed a strikingly similar pattern of alterations to serous tumors, and may benefit from a similar course of treatment.

"This study highlights the fact that some tumors with the same characterization by pathologists may have very different molecular features. That's where these findings will be directly implemented in additional research, and also in the context of clinical trials," said Douglas A. Levine, M.D., head of the Gynecology Research Laboratory at Memorial Sloan-Kettering Cancer Center, New York, and a co-leader in the study.

According to the authors, the new findings provide a roadmap for future clinical trials for endometrial cancer. "Each tumor subtype might warrant dedicated clinical trials because of the marked genomic differences between them that are indicative of different drivers of cancer," said study co-leader Elaine Mardis, Ph.D., co-director of the Genome Institute at Washington University School of Medicine, St. Louis. "Developing therapies for each subtype independent of the other may improve outcomes, as has been shown in breast cancer."

Investigators also found genomic similarities between endometrial cancers and other tumor types. Previous TCGA research showed that a form of ovarian cancer (high-grade serous ovarian carcinoma) and a subtype of breast cancer (basal-like breast cancer) share many genomic features. In this study, the scientists found that endometrial serous carcinoma also has some of these same genomic characteristics. The cancers share a high frequency of mutations in TP53 (between 84 and 96 percent) and a low frequency in PTEN, with only 1 to 2 percent mutated. Surprisingly, the researchers also found many shared characteristics between endometrioid tumors and colorectal tumors. Both cancer types demonstrate a high frequency of microsatellite instability, where the repair mechanism for DNA is broken, and mutations in POLE, a gene responsible for producing a protein involved in DNA replication and repair. These genomic changes led to high mutation rates in both tumor types.

"TCGA's multidimensional approach to collecting genomic data, including clinical and pathology information, have made these findings possible," said Harold Varmus, M.D., NCI director. "Without the integrated characterization of so many tumor samples, correlations between histology and genomic data may not have been observed or potential clinical outcomes identified."

With a complete analysis of the study's findings, investigators have identified four novel genomic-based subtypes of endometrial cancer, which may set the stage for new diagnostic and treatment approaches. Each of the four genomic subtypes clustered together and was named for one of its notable characteristics:

The POLE ultramutated group was named for its unusually high mutation rates and hotspot mutations (sequences highly susceptible to mutation) in the POLE gene. The hypermutated microsatellite instability group exhibited a high mutation rate, as well as few copy number alterations, but did not carry mutations in the POLE gene. The copy number low group showed the greatest microsatellite stability but a high frequency of mutations in CTNNB1, a gene critical for maintaining the linings of organs, such as the endometrium. The copy number high subtype was composed of mostly serous tumors, but included some endometrioid samples. This subtype displayed copy number alterations and a mutation landscape that was characteristic of serous tumors.

Endometrial cancer is the fourth most commonly diagnosed cancer among women in the United States. NCI estimates that close to 50,000 women will be diagnosed with endometrial cancer in 2013, with more than an estimated 8,000 deaths from the disease. For a majority of patients diagnosed with aggressive, high grade tumors with metastases, the five-year survival rate is about 16 percent, though chemotherapy has been associated with an improvement in survival, and new targeted agents are being tested.

"Finding genomic similarities among types of breast, ovarian, endometrial and colorectal tumors once again reveals that cancer, although very complex, may have themes extending beyond tissue type that can be exploited for therapeutic benefit," said Eric D. Green, M.D., Ph.D., NHGRI director. "These similar genomic features demonstrate hitherto unknown commonalities among these cancers.

To date, the TCGA Research Network has generated data and published analyses on glioblastoma multiforme, ovarian serous adenocarcinoma, colorectal adenocarcinoma, lung squamous cell carcinoma and invasive breast cancer. Data generated by TCGA are freely available at the TCGA Data Portal and CGHub.

This work was supported by the following grants from the NIH: 5U24CA143799-04, 5U24CA143835-04, 5U24CA143840-04, 5U24CA143843-04, 5U24CA143845-04, 5U24CA143848-04, 5U24CA143858-04, 5U24CA143866-04, 5U24CA143867-04, 5U24CA143882-04, 5U24CA143883-04, 5U24CA144025-04, U54HG003067-11, U54HG003079-10 and U54HG003273-10 and supplemented by the Recovery Act.

More details about The Cancer Genome Atlas, including Quick Facts, Q&A, graphics, glossary, a brief guide to genomics and a media library of available images can be found at http://cancergenome.nih.gov.

Reference: The Cancer Genome Atlas Research Network. Integrated Genomic Characterization of Endometrial Carcinoma. Nature. May 2, 2013. DOI:10.1038/nature12113.

The second study is on acute myeloid leukemia:

TCGA researchers identify potential drug targets, markers for leukemia risk; New study reveals relatively few mutations in AML genomes

Investigators for The Cancer Genome Atlas (TCGA) Research Network have detailed and broadly classified the genomic alterations that frequently underlie the development of acute myeloid leukemia (AML), a deadly cancer of the blood and bone marrow. Their work paints a picture of a cancer marked by relatively few mutations compared to other types of cancer occurring in adults. They also found that AML is powerfully influenced by mutations in genes that cause epigenetic changes (chemical changes to the genome that do not change the DNA nucleotide sequence) that can affect the expression of genes. TCGA is jointly supported and managed by the National Human Genome Research Institute (NHGRI) and the National Cancer Institute (NCI), both part of the National Institutes of Health.

The findings, which appeared online May 1, 2013, in the New England Journal of Medicine set the stage for identifying potential new drug targets and treatment strategies for AML. They may also offer better guidance for predicting the severity of disease for individual patients.

"These results provide important new insights into the genomics of a deadly and difficult-to-treat cancer, and underscore the power and scope of The Cancer Genome Atlas project," said NIH Director Francis S. Collins, M.D., Ph.D.

"Rather than just random snapshots about individual patients, this study provides a more detailed look at the aberrant genomes of AML than we have ever had before," said NHGRI Director Eric D. Green, M.D., Ph.D. "It has the potential to open up new directions in AML research, and perhaps, in the design of new therapeutics, its impact could be felt in the near future."

AML, the most common acute form of adult leukemia, develops when immature white blood cells fail to mature and instead accumulate in the bone marrow. The leukemia cells reduce the production of healthy blood cells, leading to anemia, abnormal bleeding and infections, and, if untreated, death.

Researchers examined the genomes of tumor specimens from 200 adult cases of spontaneously occurring, newly diagnosed AML. These cases represented all of the known subtypes of AML in approximately the same proportion as the general population. In this way, the study provided a realistic view of the disease, particularly in the number and frequency of genomic alterations. Each AML genome was compared to the normal genome derived from a skin sample of the same patient. Out of the 200 samples, 50 were analyzed by using whole genome sequencing, which is an examination of the complete DNA blueprint of the cells. Researchers analyzed the genome's protein-coding regions in the remaining samples, and used powerful new sequencing methods to look at changes in RNA in each case.

By studying a large number of AML cases, investigators were able to predict that they have identified virtually all of the mutations that occur in at least 5 percent of AML patients. Surprisingly, they found that overall, AML genomes have relatively few mutations, and such tumors are among the least mutated adult cancers. The average number of mutations in genes for each AML genome was 13, in contrast to solid tumors such as breast, lung or pancreatic cancer, which often have hundreds of mutated genes.

Because investigators found more than 1,600 genes that were mutated at least once in the 200 samples, they organized the recurrently mutated genes into nine categories based on their function or the known pathways involved. Some of these categories include tumor suppressor genes, signaling genes and epigenetic modifiers, with the latter being the most frequently mutated class of genes in the study. Epigenetic changes are alterations to DNA that often involve the addition or removal of chemical tags (such as methyl groups), which can affect when genes are turned on and off.

"This data set helps to integrate what was previously fragmented information," said study co-leader Timothy J. Ley, M.D., associate director for cancer genomics at The Genome Institute at Washington University School of Medicine in St. Louis. "We didn't realize how few recurrent mutations there were, and no one was thinking even five years ago that AML was associated with a high frequency of mutations in genes that encode epigenetic modifiers."

By finding comparatively few recurrently mutated genes, yet frequent alterations in genes that help control gene expression, the investigators may have narrowed the search for likely drug targets and disease markers.

Other results were also somewhat surprising. Researchers knew that mutations in signaling genes, which help control cell growth and development, were very common in AML, and thought that all AML samples may have at least one signaling gene mutation. But the TCGA findings showed that these genes are mutated in only 60 percent of cases. These include mutations in the gene FLT3, which occur in about a third of cases, making it one of the most commonly mutated genes in AML. FLT3 is important for normal blood cell development. The researchers also found that many AML patients have concurrent mutations in three commonly mutated genes: FLT3, NPM1 and DNMT3A. Patients with this combination of gene mutations appear to have a unique subtype of AML.

Investigators unexpectedly found recurring mutations in cohesin genes, which are important in cell division.

The study is the first to report a recurrently mutated microRNA gene in AML. MicroRNAs can play an important role in regulating gene expression, particularly in turning off gene activity.

Abnormal chromosome rearrangements and gene fusions (where two genes join to form a new, altered gene) are frequently useful in diagnosing and providing prognostic information for AML patients. The study uncovered many such fusions that had not been described before, and nearly half of the AML samples were found to have gene fusions.

Currently, only a few good markers exist to help guide treatment decisions for the majority of patients with intermediate risk. Some of the recurrently mutated genes identified in this study may allow for better prognostic information that will be relevant for AML patients.

"We've never had such a complete picture of AML, and this data set will be mined by researchers for years," said co-study leader Richard Wilson, Ph.D., director of Washington University's Genome Institute. "These findings have probably identified every pathway in which a modification – and perhaps new drugs – might be beneficial. They also further refine our understanding of the importance of individual mutations for disease classification and prognostication, and will help us build better disease models."

"These results will enable investigators to examine patient samples for mutation patterns and affected pathways, and to begin new studies to try to understand the relationships between these genetic mutations and treatment results," added Ley.

"This study of AML reinforces the value of the approach we are using to study the genomic diversity among tumors of many different cancer types and even within a single kind of cancer such as AML," noted NCI Director Harold Varmus, M.D. "Only such a systematic analysis of cancer types could have uncovered such clear patterns, such as the apparent importance in AML of genetic mutations that lead to changes in gene expression and cell traits."

This work was supported by the following grants: U24CA143845, U24CA143858, U24CA144025, U24CA143882, U24CA143866, U24CA143867, U24CA143848, U24CA143840, U24CA143835, U24CA143799, U24CA143883, U24CA143843, U54HG003067, U54HG003079, U54HG003273, and P01CA101937.

Reference: The Cancer Genome Atlas Network. Genomic and epigenomic landscape of adult de novo acute myeloid leukemia. New England Journal of Medicine. Online May 1, 2013. In print May 30, 2013. DOI: 10.1056/NEJMoa1301689.

###


Disclaimer: AAAS and EurekAlert! are not responsible for the accuracy of news releases posted to EurekAlert! by contributing institutions or for the use of any information through the EurekAlert system.