The"Newcastle Nomogram"-Statistical modelling predicts malignant transformation in potentially malignant disorders

Area of Science:

• Oral oncology
• Clinical prediction modeling
• Cancer risk assessment

Background:

• Nomograms are established tools for predicting cancer risk and treatment response.
• No nomogram currently exists for predicting clinical outcomes in oral potentially malignant disorder (PMD) management.
• Accurate risk assessment is crucial for objective decision-making in PMD patient care.

Purpose of the Study:

• To develop a predictive nomogram for malignant transformation (MT) in oral potentially malignant disorders (PMD).
• To provide clinicians with a tool for objective risk assessment and decision-making in PMD management.
• To facilitate personalized patient management strategies based on predicted MT probability.

Main Methods:

• Retrospective analysis of clinico-pathological data from 590 newly presenting PMD patients.
• Multiple logistic regression modeling to predict MT probability based on age, gender, lesion type, site, and biopsy diagnosis.
• Internal validation and calibration using bootstrap resampling (n=1000) to ensure model reliability.

Main Results:

• Oral potentially malignant disorders (PMDs) were primarily leukoplakias (79%), frequently located on the floor of the mouth and lateral tongue (51%).
• 17% of patients (99/590) progressed to oral squamous cell carcinoma during the study.
• The developed nomogram demonstrated good predictive performance with bias-corrected discrimination (Dxy=0.58) and calibration (C=0.790), showing 87% sensitivity and 96% negative predictive value.

Conclusions:

• The
• Newcastle Nomogram
• was developed using a validated statistical model to predict MT probability in PMD.
• The nomogram utilizes readily available patient-specific clinico-pathological data.
• It serves as a pragmatic graphical aid for clinicians in diagnosing and managing PMD, improving decision-making.