Estimating and modelling cure in population-based cancer studies within the framework of flexible parametric survival models
2011

Estimating Cure in Cancer Studies Using Flexible Parametric Models

Sample size: 34664 publication 10 minutes Evidence: high

Author Information

Author(s): Andersson Therese ML, Dickman Paul W, Eloranta Sandra, Lambert Paul C

Primary Institution: Karolinska Institutet

Hypothesis

Can flexible parametric survival models effectively estimate the cure proportion and survival of uncured cancer patients?

Conclusion

Flexible parametric cure models provide better estimates for cancer patient survival, especially for older age groups, compared to standard models.

Supporting Evidence

  • The flexible parametric cure model allows for the inclusion of older age groups in cancer studies.
  • This model provides similar results to standard cure models when they are reliable.
  • The method shows robustness to the number and location of knots used in the analysis.
  • Results indicate that the flexible parametric model fits better for older patients compared to non-mixture models.

Takeaway

This study shows a new way to understand how many cancer patients are cured and how long those who aren't cured might live, especially for older patients.

Methodology

The study used flexible parametric survival models with splines to estimate cure proportions and survival functions for colon cancer patients in Finland.

Potential Biases

Potential biases arise from the assumptions made in the model regarding the cure proportion and the distribution of survival times.

Limitations

The model's estimates can be sensitive to the placement of knots and may not be reliable if the cure point is not reached within the follow-up time.

Participant Demographics

Patients diagnosed with colon adenocarcinoma in Finland from 1953 to 2003, including those aged 80 and above.

Statistical Information

P-Value

p<0.0001

Confidence Interval

30.5-37.1

Statistical Significance

p<0.0001

Digital Object Identifier (DOI)

10.1186/1471-2288-11-96

Want to read the original?

Access the complete publication on the publisher's website

View Original Publication