Skip to main content

Advertisement

You are viewing the new article page. Let us know what you think. Return to old version

Statistical modelling of biomarkers incorporating non-proportional effects for survival data

Personalised medicine is replacing the one-drug-fits-all approach with many prognostic models incorporating biomarkers available for risk stratifying patients with breast cancer, such as the Nottingham Prognostic Index and Adjuvant! Online and more recently multiparameter assays, OncotypeDx and Mammaprint.

Evidence of biomarkers having non-proportional effects have been emerging and therefore violating the assumption of proportional hazards when performing Cox regression. A classic example is the risk of recurrence after breast cancer depends on the duration of follow-up for estrogen receptor (ER) and progesterone receptor (PgR) expression status[1]. The gene signature MammaPrint has also been shown to have possible non-proportional effects with better prediction of patients at high risk of early relapse rather than those at risk of later disease progression[2, 3].

A review of existing approaches for the analysis of non-proportional effects with respect to survival data found there to be a number of well-developed approaches for incorporating non-proportional effects but a lack of application of these approaches in practice. Two key approaches are the multivariable fractional polynomial time (MFPT) approach by Sauerbrei et al.[4] and flexible parametric models proposed by Royston & Parmer[5].

There is a need for more widespread use of flexible modelling to move away from standard analysis using a Cox model when the assumption of proportional hazards is violated. Fully determining the effects of markers in prognostic studies will help develop novel models for the selection of patients for appropriate treatments.

References

  1. 1.

    Mulligan AM: Clinical Cancer Research. 2008, 14: 4168-4174. 10.1158/1078-0432.CCR-07-4543.

  2. 2.

    Buyse M: Journal of the National Cancer Institute. 2006, 98: 1183-1192. 10.1093/jnci/djj329.

  3. 3.

    Desmedt C: Clinical Cancer Research. 2007, 13: 3207-3214. 10.1158/1078-0432.CCR-06-2765.

  4. 4.

    Sauerbrei W: Biometrical Journal. 2007, 49: 453-473. 10.1002/bimj.200610328.

  5. 5.

    Royston P: Statistics in Medicine. 2002, 21: 2175-2197. 10.1002/sim.1203.

Download references

Author information

Correspondence to Jacqueline Stephen.

Rights and permissions

Reprints and Permissions

About this article

Keywords

  • Breast Cancer
  • Estrogen Receptor
  • Progesterone
  • Progesterone Receptor
  • Survival Data

Comments

By submitting a comment you agree to abide by our Terms and Community Guidelines. If you find something abusive or that does not comply with our terms or guidelines please flag it as inappropriate. Please note that comments may be removed without notice if they are flagged by another user or do not comply with our community guidelines.