Palliative care (PC) is being used by more patients early in their illness progression. The Barretos Prognostic Nomogram (BPN) was created to bridge the gap between survival prognosis and months of life expectancy for patients with advanced cancer. However, the ubiquitous requirement for a ruler and calculator limits its everyday use. Furthermore, the BPN necessitates blood testing. For a study, researchers sought to improve the BPN and develop a prognostic application (App) for usage on cell phones.
It was a reanalysis of two advanced cancer patient cohorts (development, n=215; validation, n=276). The variable ‘metastasis’ has been changed (volume-site combinations), and ‘KPS’ has been replaced with ‘ECOG-PS.’ Backward variable elimination was used to identify the most accurate final models after selecting prognostic variables for multivariable Cox and Log-logistic parametric regression analyses. The validation sample was used to assess calibration and discrimination properties.
The ‘complete version’ model has six parameters: sex, locoregional illness, metastatic locations, ECOG-PS, WBC, and albumin. The variable ‘antineoplastic therapy’ was included in the ‘clinical version’ model (5 variables), whereas the laboratory variables were removed. Both models were correctly calibrated and had sufficient c-Index values throughout validation (0.778 and 0.739). HAprog is a free offline app that physicians use to determine prognosis in less than a minute.
The new HAprog-integrated models were upgraded prognostic tools with acceptable calibration and discriminating features. It might be useful for oncologists working with outpatients with advanced cancer throughout the decision-making process.