Lifetime Cancer Risks of PTEN Mutation Carriers—Response

We thank Daniels and colleagues for their comments on our report (1). Given that the Cleveland Clinic is the main referral center of patients with germline PTEN mutations in the United States, it is certainly possible that ascertainment bias may exist, an issue we acknowledged in our report. Referral bias is an inherent challenge for the study of rare genetic disorders, but clearly a population-based approach for evaluating such a rare syndrome would be impractical. We also agree that this bias would be ideally addressed by prospective follow-up of affected individuals, an effort we are currently undertaking. The Kaplan–Meier method was selected because of its widespread use in other studies evaluating lifetime cancer risks in genetic syndromes. The other methods referenced by Daniels and colleagues require either long-term follow-up or recruitment of several members within each family, making these models incompatible with this study's design and patient population.

As noted in the original article, our cohort has been prospectively ascertained. Thus, not all the cancers reported are prevalent. For example, renal cell carcinoma is a minor criterion and colon cancer and melanoma are not criteria at all. Yet, increased standardized incidence ratio (SIR) and age-related penetrances are noted for these cancers. These findings are also comparable with previous literature (2). We used standard criteria for calling pathogenic mutations, ranging from bioinformatic algorithms (SIFT, PolyPhen, Mutation Taster) to functional studies. We only used for this study mutations and promoter variants that have been shown to be pathogenic. There are many more variants in the promoter that our laboratory is struggling to functionally characterize. To be conservative, these have not been included in this study.

Daniels and colleagues suggest that it would be premature to recommend changes in clinical practice based on data from our series. Existing clinical practice and NCCN recommendations derive primarily from previous retrospective observations, many of which were contributed by our group. Our approach has identified colorectal carcinoma, melanoma, pediatric thyroid cancer, and papillary renal cell carcinoma as features of germline PTEN mutations, observations which are beginning to be confirmed by other investigators (3, 4). The adult and pediatric testing criteria we have developed are also superior to existing NCCN criteria (5). Together with the fact that this data set is derived from the largest known series of patients with PTEN mutations, we believe the recommendations outlined in our article represent current best practice for counseling and follow-up of these patients.

See the original Letter to the Editor, p. 4213