Concept Architecture
Concept
Theoretically, the Mixture Cure Model is a parametric survival model that assumes the study population comprises two latent subpopulations: individuals who are susceptible to the event of interest and individuals who are permanently cured. It is founded on cure modelling theory and finite mixture modelling, allowing long-term survivors to be explicitly represented within survival analysis. The model exists because conventional survival models assume every individual will eventually experience the event if followed indefinitely, an assumption that is often inappropriate in oncology and other diseases where a proportion of patients may be effectively cured.
Mathematically, the overall survival function is expressed as a weighted mixture of the cured and uncured populations. The cure fraction represents the probability of belonging to the cured group, while the susceptible group follows a conventional parametric survival distribution such as the Weibull, Log-Normal or Generalised Gamma distribution. Model parameters are estimated simultaneously using maximum likelihood estimation.
In practice, mixture cure models are fitted to individual patient survival data using specialised survival modelling software. The cure fraction and survival distribution are estimated jointly and validated using likelihood-based criteria, graphical assessment and external clinical evidence. In health economics, mixture cure models are widely used to extrapolate long-term survival for health technology assessments when durable treatment responses suggest that a proportion of patients may remain event-free indefinitely.
Purpose
Used to estimate long-term survival when a proportion of patients are considered permanently cured, supporting realistic survival extrapolation and lifetime cost-effectiveness modelling.
Mathematical Formulae
Primary Formula
S(t) = � + (1 ? �)S?(t)
where:
� = cure fraction
S?(t) = survival function of the susceptible population
Supporting Formulae
P(Cured) = �
P(Susceptible) = 1 ? �
L = ?[� + (1 ? �)S?(t?)]???? ? [(1 ? �)f?(t?)]??
where:
L = likelihood function
�? = event indicator
f?(t) = density function for susceptible individuals
Related Mathematical Methods
- Maximum Likelihood Estimation
- Finite Mixture Model
- Cure Fraction Model
- Non-Mixture Cure Model
- Weibull Model
- Log-Normal Model
- Generalised Gamma Model
- Parametric Survival Analysis
Example
A clinical trial evaluates an immunotherapy for advanced melanoma. A mixture cure model estimates that 24% of patients are permanently cured (� = 0.24), while the remaining 76% follow a Weibull survival distribution.
At 10 years, the susceptible survival probability is estimated as:
S?(10) = 0.18
Overall survival is therefore:
S(10) = 0.24 + (0.76 ? 0.18)
S(10) = 0.377
The model predicts a 37.7% probability of surviving 10 years, reflecting both the cured population and the remaining susceptible patients.
Excel Implementation
| Function | Example Formula | Health Economics Application |
|---|---|---|
| EXP | =EXP(-((A2/$B$1)^$B$2)) | Calculate susceptible survival using a Weibull model. |
| SUMPRODUCT | =$C$1+(1-$C$1)*D2 | Calculate overall mixture survival using the estimated cure fraction. |
| Solver | Maximum likelihood optimisation | Estimate cure fraction and survival model parameters simultaneously. |
| IF | =IF(A2>120,$C$1,D2) | Explore long-term survival behaviour in extrapolation analyses. |
VBA (Optional)
Automate estimation of mixture cure models, compare alternative survival distributions and generate long-term survival extrapolations.
Sources
- Boag JW. Maximum Likelihood Estimates of the Proportion of Patients Cured by Cancer Therapy. Journal of the Royal Statistical Society: Series B. 1949.
- Berkson J, Gage RP. Survival Curve for Cancer Patients Following Treatment. Journal of the American Statistical Association. 1952.
- Maller RA, Zhou X. Survival Analysis with Long-Term Survivors.
- Othus M, Barlogie B, LeBlanc ML, Crowley JJ. Cure Models as a Useful Statistical Tool for Analyzing Survival. Clinical Cancer Research. 2012.
- NICE. Health Technology Evaluation Manual.
- Briggs A, Claxton K, Sculpher M. Decision Modelling for Health Economic Evaluation.
Related Concepts (2)
Library
Publications
1
NICE DSU Technical Support Document 14: Survival analysis for economic evaluations alongside clinical trials – extrapolation with patient-level data — Nicholas R. Latimer, TSD 14 ed., 2013 (NICE Decision Support Unit (University of Sheffield))
The reference guidance on survival analysis for economic evaluation: fitting standard parametric models (exponential, Weibull, Gompertz, log-logistic, log-normal) to censored trial data and extrapolating to estimate lifetime survival benefit, with a process guide for model selection and justification.
Frequently Asked Questions (6)
What is a mixture cure model?
A cure model treating the population as a mixture of a cured fraction facing only background mortality and an uncured fraction remaining at risk.
Source: Boag 1949
What does the mixture in a mixture cure model refer to?
The mixture in the name refers to the population being modelled as a blend of two distinct groups, a cured fraction and an uncured fraction, each with its own survival experience. The overall survival curve is a weighted average of the two, the cured contributing only background mortality and the uncured a disease-specific distribution on top of it. This explicit split into two subpopulations is what distinguishes the mixture form. The weight is the estimated cure fraction. Lambert and colleagues (2007) describe it.
Source: Lambert et al. 2007
How does a mixture cure model work?
A mixture cure model works by specifying overall survival as a weighted combination of the two subgroups: the cured fraction, whose survival reflects only background mortality, and the uncured fraction, whose disease-specific survival follows a chosen distribution. The model estimates the cure fraction and the parameters of the uncured survival. As the uncured experience events, overall survival declines toward the cured fraction, producing a plateau. This explicit mixture structure directly represents the cured and uncured subgroups and their differing survival.
Source: Boag 1949
How does a mixture cure model differ from a non-mixture cure model?
A mixture cure model explicitly splits the population into cured and uncured subgroups, modelling overall survival as their mixture, whereas a non-mixture cure model formulates survival directly in terms of the cure fraction without separating the population into two groups, using a different mathematical structure that also yields a plateau. Both represent a cured fraction and produce similar survival plateaus, but they differ in formulation. The mixture form is more intuitive, directly representing the two subgroups, while the non-mixture form arises from a different derivation.
Source: Latimer 2013
When is a mixture cure model appropriate?
A mixture cure model is appropriate when evidence suggests a distinct subgroup of patients are effectively cured after treatment, so that disease-specific survival plateaus and long-term survivors are seen, and when it is meaningful to separate cured and uncured patients explicitly. A visible plateau and long follow-up support its use. Where no cure occurs and all remain at risk, a standard survival model is used, so the mixture cure model is reserved for genuinely curable conditions with a distinguishable cured fraction.
Source: Latimer 2013
What are the challenges of mixture cure models?
Mixture cure models require long follow-up to observe a plateau and estimate the cure fraction reliably, since short data may confound cure with slow progression, leaving the cured fraction uncertain. The estimate depends on the distribution assumed for the uncured and on background mortality, and whether cure genuinely occurs may be contested. Because the cure fraction strongly affects long-term and mean survival, its uncertainty matters, so evidence of cure and sensitivity to the assumptions are examined carefully when using the model.
Source: Boag 1949
Trust Record
Verified by Dr Darrin Baines
British health economist
Professional identity: darrinbaines.org
Verification date: 22 Oct 2025
Content version: 1.0.0
Canonical Identity
- Persistent URI
- https://healtheconomics.wiki/concept/mixture-cure-model
- Term code
- HE-EM-SM-052
Stable URI · Machine-readable · Resolvable · CC BY 4.0