Head and Neck Adenoid Cystic Carcinoma Prognostic Model
by Xingyuan Chen·Updated 3mo ago
286.2 KB1files
Available on 1 platform
Sign in to view source links and access this dataset
Description
A prognostic model for cancer-specific survival in non-metastatic head and neck adenoid cystic carcinoma, derived from 2,688 patient records in the SEER database. The study applies a Fine-Gray competing risks model, identifying age, T-stage, N-stage, and treatment status as key factors. It provides a validated nomogram for predicting 3-, 5-, and 10-year survival.
Use Cases
Validate the prognostic nomogram's performance for 3-, 5-, and 10-year cancer-specific survival predictions in independent patient cohorts.
Analyze the impact of independent prognostic factors like age, T-stage, and N-stage on head and neck adenoid cystic carcinoma mortality.
Compare survival predictions from the Fine-Gray competing risks model against traditional Cox regression models using the same patient cohort.
Assess the clinical utility of variables such as perineural invasion (PNI) and POCRT/PORT treatment status for individualized survival estimates.
Strengths
Model validation on 2,688 patient records from the SEER database.
Nomogram demonstrated high concordance indices, up to 0.965 in the validation cohort.
Analysis includes 673 cancer-specific deaths, providing a clear endpoint for the competing risks model.
Limitations
Data is presented only in a PDF report format, requiring manual extraction for computational analysis.
The dataset scope is limited to 286.2 KB, containing summary model results rather than the raw patient-level data.
Patient data is sourced from a specific time range (2004-2015) and may not reflect current treatment protocols.
Provenance
Source
Surveillance, Epidemiology, and End Results (SEER) database.
Collection Method
Retrospective cohort study applying Fine-Gray competing risks and Cox regression models.
Time Range
Patients diagnosed between 2004 and 2015.
Freshness
Data covers patients diagnosed between 2004 and 2015; the model publication was last updated in 2026.
Geography
United States (based on SEER database coverage).
The primary data asset is a 286.2 KB PDF containing the research paper and model; raw tabular patient data is not provided. Users must extract numerical data or model coefficients from the document for reuse.