This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Parmigiani, G. Articles by Berry, D. A. Search for Related Content PubMed Articles by Parmigiani, G. Articles by Berry, D. A. Social Bookmarking                            What's this?

In Reply

Departments of Oncology, Biostatistics, and Pathology, Johns Hopkins University, Baltimore, MD

Donald A. Berry

Department of Biostatistics, University of Texas M.D. Anderson Cancer Center, Houston, TX

We appreciate the opportunity to comment on the important observation made by Capalbo and colleagues. Their study focuses on a subset of individuals whose carrier probabilities, as assessed by the BRCAPRO model, are in excess of 80%. These individuals generally present with a family history showing obvious Mendelian patterns of inheritance. Yet when genetic testing for BRCA1 and BRCA2 is performed, a small fraction of patients have received a negative, or equivocal, report.¹

There are several groups of individuals that fall in this category. One, as pointed out by Capalbo and by others,² are those who harbor large genomic rearrangements in one of their genes. In the United States, BRCA1 and BRCA2 testing for clinical purposes is done by Myriad Genetics (Salt Lake City, UT), who report recent progress in considering some of the large genomic rearrangements that occur in relatively high prevalence in European populations.³ However, this remains an area where progress is needed. A second group are individuals who harbor a deleterious missense variant that is still categorized as a variant of uncertain clinical significance. Classification of missense mutations as deleterious is a slow process because of the large variety of these mutations and the small number of carriers of each individual one. This is an area where significant additional progress in both testing and counseling is possible, building on recent in-silico techniques for accelerated classification of variants.^4,5 A third group of individuals harbor deleterious mutations in genes more rarely involved in inherited breast cancer, such as TP53 or CHEK2.² Finally, a small fraction of family histories with the appearance of Mendelian transmission may occur as a result of shared behavioral and environmental factors, as well as by chance.

The potential for a false-negative genetic test in the presence of a strong family history of cancer poses a challenge to counseling and decision making. It is important to help counselees interpret their testing results. Particular caution is necessary in view of the imperfect sensitivity of testing, as one's carrier probability could remain high despite negative test results. Mendelian risk prediction models such as BRCAPRO can provide critical help at this stage of counseling. When using such a model, the pretest prediction made is for whether the individual being counseled carries a deleterious mutation, not for whether such a mutation will be found.⁶ Also, the Mendelian calculation implicitly includes mutations in autosomal dominant genes other than BRCA1 and BRCA2, assuming that those genes have similar implications for cancer risk. If an estimate of the sensitivity of the genetic test is available, after a negative test result, a post-test probability will evaluate the chance that a mutation of the type not probed by the test is present nonetheless. For example, for an individual with pretest probability of 95%, a negative test leads to a post-test probability of 79% if the sensitivity of the test is 80%, and 65% if the sensitivity of the test is 90%. The BRCAPRO model, as implemented in both BayesMendel⁷ and CancerGene,⁸ allows users to easily input sensitivity estimates and obtain post-test probability. In Berry et al,¹ we suggest an estimate of sensitivity of 85% that can be used for this purpose. This estimate includes all of the categories of missed inherited susceptibility described earlier, and thus may vary over time and across populations. Studies such as that of Capalbo are an important contribution toward refining our understanding of this critical parameter.

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

The author(s) indicated no potential conflicts of interest.

REFERENCES

1. Berry DA, Iversen ES Jr, Gudbjartsson DF, et al: BRCAPRO validation, sensitivity of genetic testing of BRCA1/BRCA2, and prevalence of other breast cancer susceptibility genes. J Clin Oncol 20:2701-2712, 2002[Abstract/Free Full Text]

2. Walsh T, Casadei S, Coats KH, et al: Spectrum of mutations in BRCA1, BRCA2, CHEK2, and TP53 in families at high risk of breast cancer. JAMA 295;1379-1388, 2006[Abstract/Free Full Text]

3. Myriad Genetics Laboratories: BRACAnalysis technical specifications. Salt Lake City, UT, 2006

4. Zhou X, Iversen Jr ES, Parmigiani G: Classification of missense mutations of disease genes. Journal of the American Statistical Association 100:51-60, 2005[Medline]

5. Karchin R, Monteiro ANA, Tavtigian SV, et al: Functional impact of missense variants in BRCA1 predicted by supervised learning. PLoS Comput Biol 3;e26, 2007[CrossRef][Medline]

6. Berry D, Parmigiani G, Sanchez G, et al: Probability of carrying a mutation of breast-ovarian cancer gene BRCA1 based on family history. J Natl Cancer Inst 89; 227-238, 1997[Abstract/Free Full Text]

7. Chen S, Wang W, Broman KW, et al: BayesMendel: An R environment for Mendelian risk prediction. Stat Appl Genet Mol Biol 3:21, 2004

8. Euhus DM: Understanding mathematical models for breast cancer risk assessment and counseling. Breast J 7:224-232, 2001[CrossRef][Medline]

CiteULike    Complore    Connotea    Del.icio.us    Digg    Facebook    Reddit    Technorati    Twitter    What's this?

This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Citing Articles Citing Articles via Google Scholar Google Scholar Articles by Parmigiani, G. Articles by Berry, D. A. Search for Related Content PubMed Articles by Parmigiani, G. Articles by Berry, D. A. Social Bookmarking                            What's this?