Related Experiment Video
Updated: Jul 12, 2026
04:57
Establishing a Competing Risk Regression Nomogram Model for Survival Data
Published on: October 23, 2020
MELPREDICT: a logistic regression model to estimate CDKN2A carrier probability.
K B Niendorf1, W Goggins, G Yang
1Center for Cancer Risk Analysis, Massachusetts General Hospital, Boston, MA 02114, USA.
Journal of Medical Genetics
|September 20, 2005
Summary
This study developed MELPREDICT, a model to estimate the probability of carrying CDKN2A mutations in familial melanoma patients. The model shows robust performance in identifying mutation carriers, aiding in risk assessment for hereditary melanoma.
Area of Science:
- Genetics
- Oncology
- Medical Informatics
Background:
- Germline alterations in the CDKN2A gene are associated with a portion of familial melanoma cases.
- A reliable method for identifying individuals at risk of being CDKN2A mutation carriers is currently lacking.
Purpose of the Study:
- To develop and validate a predictive model for estimating CDKN2A mutation carrier probability in familial cutaneous melanoma.
- To identify clinical features associated with CDKN2A germline mutations.
Main Methods:
- A logistic regression model (MELPREDICT) was constructed using data from 116 familial cutaneous melanoma patients.
- Clinical features including proband age at diagnosis, number of proband primaries, and number of additional family primaries were analyzed.
- The model's performance was evaluated using an external validation cohort.
Main Results:
- The MELPREDICT model incorporates age, proband primaries, and family primaries to estimate carrier probability.
- The model achieved an area under the curve of 0.881 in the Boston dataset and 0.803 in an external Toronto cohort.
- Mean estimated probabilities were 55.4% for carriers and 5.1% for non-carriers in the Boston cohort.
Conclusions:
- The developed logistic regression model (MELPREDICT) provides a first-iteration approximation of CDKN2A carrier probability.
- External validation demonstrated relatively robust performance of the MELPREDICT model.
- This model can aid in identifying individuals at higher risk for CDKN2A mutations in familial melanoma.
