Estimating TP53 Mutation Carrier Probability in Families with Li-Fraumeni Syndrome Using LFSPRO

Abstract

Background: Li-Fraumeni syndrome (LFS) is associated with germline TP53 mutations and a very high lifetime cancer risk. Algorithms that assess a patient's risk of inherited cancer predisposition are often used in clinical counseling. The existing LFS criteria have limitations, suggesting the need for an advanced prediction tool to support clinical decision making for TP53 mutation testing and LFS management.Methods: Based on a Mendelian model, LFSPRO estimates TP53 mutation probability through the Elston-Stewart algorithm and consequently estimates future risk of cancer. With independent datasets of 1,353 tested individuals from 867 families, we evaluated the prediction performance of LFSPRO.Results: LFSPRO accurately predicted TP53 mutation carriers in a pediatric sarcoma cohort from MD Anderson Cancer Center in the United States, the observed to expected ratio (OE) = 1.35 (95% confidence interval, 0.99-1.80); area under the receiver operating characteristic curve (AUC) = 0.85 (0.75-0.93); a population-based sarcoma cohort from the International Sarcoma Kindred Study in Australia, OE = 1.62 (1.03-2.55); AUC = 0.67 (0.54-0.79); and the NCI LFS study cohort, OE = 1.28 (1.17-1.39); AUC = 0.82 (0.78-0.86). LFSPRO also showed higher sensitivity and specificity than the classic LFS and Chompret criteria. LFSPRO is freely available through the R packages LFSPRO and BayesMendel.Conclusions: LFSPRO shows good performance in predicting TP53 mutations in individuals and families in varied situations.Impact: LFSPRO is more broadly applicable than the current clinical criteria and may improve clinical management for individuals and families with LFS. Cancer Epidemiol Biomarkers Prev; 26(6); 837-44. ©2017 AACR.

Figure 1

A hypothetical family pedigree to illustrate the utility of LFSPRO for genetic counseling. An arrow indicates the counselee, for whom the TP53 mutation probability is calculated by LFSPRO on the basis of the family history of cancer. Cancer type and age at diagnosis (years) in affected family members are indicated. (A-D) Four variations in clinical scenarios.

Figure 2

Validation results. (A) ROC curves of LFSPRO for three datasets. Also shown are true positive and false positive rates of classic LFS and Chompret criteria for the three datasets. (B) ROC curves of the MDACC dataset at different times of follow-up to evaluate the effect of ascertainment of family history in real clinical settings. The corresponding result from the Chrompret criteria is also shown. RF: roll forward. LCD: last contact date. In this analysis, the prevalence of germline mutation in TP53 is 0.0006 and the percentage of de novo mutations among all TP53 germline mutations is 20%.