Non–Mixture Cure Fraction Model Based on Fréchet- Weibull Distribution With Application To Cancer Survival Data
Communication in Physical Sciences · 2026 · 13 (11) · 1819-1828
DOI: 10.4314/cps.v13i11.9 ↗Kurz gefasst
Fachartikel in der Zeitschrift Communication in Physical Sciences, 2026. Verfasst von Mohammed Saad und 3 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Conventional survival models generally assume that all individuals in a study population remain susceptible to the event of interest, an assumption that may be inappropriate for cancer survival data containing a proportion of long-term survivors. This study develops a Fréchet-Weibull NonMixture Cure Model (FWDNMCM) to account explicitly for the presence of a cured fraction while retaining the flexibility of the FréchetWeibull distribution. The proposed model was formulated by incorporating a cure parameter into the Fréchet-Weibull survival function under the promotion-time framework of Yakovlev et al. (1993). The cumulative distribution function, probability density function, survival function, and hazard function of the proposed model were derived, and the unknown parameters were estimated using the maximum likelihood method. The model was subsequently applied to an acute lymphoblastic leukemia (ALL) survival dataset to evaluate its practical performance and estimate the long-term cured fraction. Kaplan–Meier survival estimates were compared with the fitted survival functions of the conventional Fréchet-Weibull model and the proposed FWDNMCM. The results demonstrate thatincorporating a cure parameter provides a positive asymptotic survival probability, thereby allowing the model to distinguish between susceptible patients and long-term survivors. The fitted survival curves further indicate differences in the cured fractions between the treatment groups, with Treatment Group I exhibiting a higher estimated longterm survival proportion than Treatment Group II. The FWDNMCM also provides a better representation of the observed survival pattern than the corresponding baseline FréchetWeibull model, demonstrating the importance of explicitly modelling population heterogeneity. The proposed framework therefore provides a flexible parametric approach for survival data in which a nonnegligible proportion of subjects may remain event-free over extended follow-up periods. The model offers potential applications in cancer prognosis and treatment evaluation and provides a basis for future extensions involving covariates and multivariate survival outcomes.
Zitieren
APA
Saad, M., Bishir, A., Abbas, U. F., & Lasisi, K. E. (2026). Non–Mixture Cure Fraction Model Based on Fréchet- Weibull Distribution With Application To Cancer Survival Data. Communication in Physical Sciences, 13(11), 1819-1828. https://doi.org/10.4314/cps.v13i11.9
Vancouver
Saad M, Bishir A, Abbas UF, Lasisi KE. Non–Mixture Cure Fraction Model Based on Fréchet- Weibull Distribution With Application To Cancer Survival Data. Communication in Physical Sciences. 2026;13(11):1819-1828. doi:10.4314/cps.v13i11.9
BibTeX
@article{Saad2026,
author = {Saad, Mohammed and Bishir, Abubakar and Abbas, Umar Farouq and Lasisi, Kazeem Eyitayo},
title = {Non–Mixture Cure Fraction Model Based on Fréchet- Weibull Distribution With Application To Cancer Survival Data},
journal = {Communication in Physical Sciences},
year = {2026},
volume = {13},
number = {11},
pages = {1819--1828},
doi = {10.4314/cps.v13i11.9}
}
EUVIMED