Predicting Recurrence and Survival in Laryngeal and Hypopharyngeal Cancer
Author Information
Author(s): Abou-Foul Ahmad K., Dretzke Janine, Albon Esther, Kristunas Caroline, Moore David J., Karwath Andreas, Gkoutos Georgios, Mehanna Hisham, Nankivell Paul
Primary Institution: University of Birmingham
Hypothesis
Can clinical predictive models improve prognostication for recurrence and survival in treated laryngeal and hypopharyngeal cancers?
Conclusion
Most predictive models for laryngeal and hypopharyngeal cancers have significant methodological limitations and cannot be confidently recommended for routine clinical use.
Supporting Evidence
- Eleven models were identified, with only two meeting criteria for meta-analysis.
- Most models had poor generalizability, performing worse than the traditional TNM system.
- Only one model achieved excellent discrimination (C-index >0.9).
- High risk of bias was noted for all models, particularly in the analysis domain.
Takeaway
Doctors are trying to create better ways to predict how laryngeal and hypopharyngeal cancer will behave after treatment, but many of the methods they have now aren't reliable enough.
Methodology
The study conducted a systematic review and meta-analysis of clinical predictive models for recurrence and survival in laryngeal and hypopharyngeal cancers, assessing model characteristics, performance, and risk of bias.
Potential Biases
High risk of bias was noted across all models, particularly in the analysis domain.
Limitations
Most models showed poor generalizability and high risk of bias, limiting their clinical applicability.
Participant Demographics
The majority of participants were male (77%-98.3%) and included adults with laryngeal and hypopharyngeal cancers who completed treatment.
Statistical Information
Confidence Interval
0.71-0.76
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website