Cancer and Your Genes

Mutations in the DIS3L2 Gene Cause Perlman Syndrome: Another Key Clue in Unraveling the Puzzle of Wilms Tumor Development

Wilms tumor (also referred to as nephroblastoma) is a relatively uncommon pediatric cancer (about 1 out of every 9,000 children are affected) - although it is the most common of the childhood cancers involving the kidney.

Sometimes, Wilms tumor occurs "sporadically" - showing up in the kidney on one side and not associated with other tumors or congenital anomalies.  However, Wilms tumor can occur much more frequently in the context of several different syndromes.

For example, it can occur more commonly in individuals with the so-called WAGR syndrome, in which the "W" refers to Wilms tumor, the "A" refers to aniridia (absence of the iris of the eye), the "G" refers to genital anomalies, and the "R" refers to developmental delay (previously referred to as mental retardation).  WAGR syndrome is associated with deletions of a region of chromosome 11 (11p13 to be exact), and we now know that while aniridia in WAGR occurs because of deletion of a gene called PAX6, the risk for Wilms tumor in individuals with 11p13 deletion is due to the deletion encompassing a gene called WT1.  Other individuals born with mutations in WT1 can have other syndromes (including Frasier syndrome and Denys-Drash syndrome) or sometimes develop Wilms tumor without any other features of a syndrome. 

Other genes and syndromes are also known to be associated with Wilms tumor development.  A complete description of all of them is beyond the scope of this post, but I'll just briefly mention one other genetic cause: biallelic BRCA2 gene mutations (i.e., a deleterious mutation in both copies).  Thus, in the relatively small number of people who are known to be unlucky enough to have biallelic BRCA2 mutations, Wilms tumor is relatively frequent.

Perlman Syndrome and Wilms Tumor

Another syndrome in which Wilms tumor occurs is Perlman syndrome.  In fact, for infants with this syndrome who survive past the neonatal period, nearly two-thirds will develop Wilms tumor.  Infants with Perlman syndrome have features of overgrowth disorders.  As in a number of cancer predisposition syndromes, individuals with Perlman syndrome are more likely to develop bilateral (two separate primary tumors on both sides of the body) tumors and also tend to develop Wilms tumor at a younger age than is typical.

A New Study Has Identified a Gene for Perlman Syndrome

Perlman syndrome is extremely rare; however, clinicians and scientists have actively sought a genetic cause for this disease both to facilitate counseling and care for Perlman syndrome families and to learn more about how Wilms tumor develops.  In a new study reported online in the journal Nature Genetics, Dewi Astuti, Mark Morris, Eamonn Maher and their colleagues studied six families in which one or more children were affected with Perlman syndrome. 

After mapping the responsible gene (with a technique called "autozygosity mapping") to chromosome 2q37.1, they were able to identify mutations in a gene called DIS3L2 as the cause of this autosomal recessive condition.

They went on to do quite a bit of work digging into the potential function of this gene - which I won't go into here - and also beginning to assess the role that mutations in this gene might play in Wilms tumors that are not associated with Perlman syndrome.

They found mutations or deletions of DIS3L2 (that probably occurred "somatically" during tumor development rather than being something that the child was born with) in a substantial fraction of the 40 other Wilms tumors that they studied.

Key Implications of This Study

A number of key implications arise from this study.  Here are a few:

1. Identification of a causative gene for Perlman syndrome will help clinical geneticists and genetic counselors provide key information and management recommendations to families with one or more children with Perlman syndrome.
2. The demonstration that the DIS3L2 gene is important in protecting against Wilms tumor development will lead to future studies that potentially look at its role in Wilms tumor prognosis and in the distant future theoretically could lead to new treatments.
3. There are a few hints that DIS3L2 might be involved in the development of some other, more common, cancers.  This study may just be the tip of the iceberg.

Selected References

Astuti D, Morris MR, Cooper WN, et al.  Germline mutations in DIS3L2 cause the Perlman syndrome of overgrowth and Wilms tumor susceptibility.  Nature Genetics 2012 (published online before print Feb 5 2012)

_____

About the Author: Matt Mealiffe, M.D. is a Clinical Cancer Geneticist with dual board certification in Clinical Genetics and Internal Medicine.  He received his undergraduate degree at Stanford University, his M.D. at the Yale University School of Medicine, and did his residency and fellowship training at the University of Washington.  Dr. Mealiffe writes at CancerAndYourGenes.com to cut through the hype surrounding personalized medicine and improve the public's understanding of what we really know - and don't know - about the connection between our genes and cancer risk.

Other Cancers Seen in Families with RAD51D Mutations

Two days ago, we discussed a recent study by Dr. Nazneen Rahman and colleagues which clearly demonstrated that inherited mutations in a gene called RAD51D result in a substantial elevation in risk for ovarian cancer.  Some families with mutations in the study also had some other cancer types in the family tree.  So, does this mean that the other cancer types in these families are also due to the RAD51D mutations?

In short, the answer is that we don't know.  Ovarian cancer risk was the only one that Dr. Rahman and her colleagues could pin down as being associated with RAD51D mutations with any statistical confidence.  For the others, future studies can address the specific question; however, they may be challenging and require extremely large sample sizes in order to have the statistical power to answer the question.

Here are the other cancer types that were seen in the eight families with inactivating RAD51D mutations in the study:

Breast Cancer

•  Seen in 8 out of 8 families (not surprising given that the families were included in the study because they had both ovarian and breast cancer)
• There were 13 breast cancer cases total in these 8 families
• Two of the breast cancer cases were bilateral
• Seven of the 13 cases were found to have inactivating RAD51D mutations (the other 6 were not tested)
• In contrast to ovarian cancer, the association of RAD51D mutations with breast cancer was not statistically significant (Relative Risk = 1.32 with 95% confidence intervals from 0.59 - 2.96)

Colorectal Cancer

•  Seen in 5 out of 8 families (keep in mind that some of these extended families are quite large and that colorectal cancer is common)
• There were 8 total colorectal cancer cases in the 5 families
• None of the affected individuals were tested for the presence of mutations in this study

Lung Cancer

•  Seen in 1 individual in 1 out of 8 families (keep in mind that some of these extended families are quite large and that lung cancer is common)
• The individual with lung cancer was not tested for the presence of the familial RAD51D mutation in this study

Non-Hodgkin Lymphoma

•  Seen in 1 individual in 1 out of 8 families
• The individual with NHL was not tested for the presence of the familial RAD51D mutation

Pancreatic Cancer

•  Seen in 2 individuals total from 2 out of 8 families
• The 2 individuals with pancreatic cancer were not tested for the familial RAD51D mutation

Prostate Cancer

• A total of 2 cases were seen in 1 out of the 8 families
• The 2 individuals with prostate cancer were not tested for the familial RAD51D mutation

It's very important to keep in mind that seeing these cancers in these families could just be due to chance or to the fact that the families were more likely to come to the researchers' attention because of the other cancers (particularly breast cancer).  Thus, we can't say anything about whether RAD51D mutations may lead to elevated risk for any cancer other than ovarian cancer based on this study.  Nevertheless, it will be important in the future to keep an eye on this. 

Bottom Line Implications

1. Mutations in the RAD51D gene lead to ovarian cancer risk in some families in which the familial ovarian cancer risk is not explained by mutations in BRCA1/2

2. Although several other cancer types (including breast cancer) were seen in the families with mutations, there was no statistically significant evidence from this study implicating RAD51D mutations in risk for any of the other cancer types. 

_______

Was this helpful?  Would you like to be notified when we post other similar breast and ovarian cancer risk-related information?  Click here for our free newsletter updates!

_______

Selected References

Loveday C, Turnbull C, Ramsay E, et al.  Germline mutations in RAD51D confer susceptibility to ovarian cancer.  Nature Genetics 2011 (Published online August 7 2011)

Family History of Cancer Changes Over Time: Why You Need to Keep Your Primary Care Doctor Updated

If you are interested in cancer prevention, you may know that for common cancers family history is one of the most important and useful tools that we have to identify people at elevated risk (who are likely to benefit most from more intensive screening and/or preventative measures).

Despite this fact, discussion of family history often gets lost in the shuffle in primary care - which is not surprising given how much stuff today's healthcare providers need to pack into a visit that often may only last 10 minutes or so.   That said, many healthcare providers are doing a better job of asking about a family history of cancer, but it is usually a static, one-time assessment that may happen most at the time of initiation of primary care with a new doctor.

A newly published research article by Argyrios Ziogas, Nora Horick, Sharon Plon, Dianne Finkelstein and colleagues from the Cancer Genetics Network (CGN) in JAMA provides evidence and analysis that suggests that doctors and patients should also pay more attention to changes in family history of cancer over time.

Ziogas and colleagues used a couple of different analytical approaches to try to find answers to this question:

How often do family cancer histories change enough over time that they would merit changes in the intensity of screening based on American Cancer Society guidelines?

They focused on 3 of the most common cancer types: breast cancer, colorectal cancer, and prostate cancer.  Although there are a few caveats, in general, their results suggest that it is well worth it for people to make sure that their doctors are aware of any changes in their cancer history.

In the case of breast cancer, their results suggest that about 4 percent of women will have a change in their family history of cancer between the ages of 30 and 50 sufficient to suggest that they are in a somewhat higher risk group that may benefit from breast MRI screening (per American Cancer Society Guidelines). 

For colorectal cancer, their results suggest that about 5 percent of people will have a change in their family history of cancer between ages 30 and 50 that suggests that a more aggressive colonoscopy screening strategy may be warranted (per American Cancer Society Guidelines).

The impact of prostate cancer family history changes was less impressive. 

Overall, it seems clear that in the clinic we need to think of cancer family history as the often changing entity that it is, rather than the static, never changing, thing that may get entered in the clinic chart at the first visit and then never re-visited.

 What You Can Do: If someone in your family is newly diagnosed with cancer (parents, brothers, sisters, children, aunts, uncles, grandparents, cousins), let your physicians and other healthcare providers know about it the next time you see them.  This way, they can update the family history that they have on file for you and consider whether the information might suggest that you should have more intensive cancer screening.

Cited Reference

Ziogas A, Horick NK, Kinney AY, et al.  Clinically relevant changes in family history of cancer over time.  JAMA 2011; 306:172-8. 

Further Reading

Acheson LS.  Recording, interpreting, and updating the family history of cancer: implications for cancer prevention.  JAMA 2011; 306:208-10.

Qureshi N, Wilson B, Santaguida P, et al.  Collection and use of cancer family history in primary care: Evidence Report/Technology Assessment No. 159.  Rockville, MD: Agency for Healthcare Research and Quality, 2007.  AHRQ Publication 08-E001.

Scheuner MT, McNeel TS, Freedman AN.  Population prevalence of familial cancer and common hereditary cancer syndromes: the 2005 California Health Interview Survey.  Genet Med 2010; 12:726-35.

_____

Know someone with Hereditary Breast and Ovarian Cancer risk due to a BRCA1 or BRCA2 mutation?  Encourage them to check out BRCAscoop.com - a new high quality health information resource focused on newly published evidence related to cancer prevention, screening and treatment for people with BRCA1 or BRCA2 mutations