PRcriteria: Piecewise regression criteria

In cyhsuTN/AdjCOXPH: A Novel Adjustment Method for Cox Proportional Hazards Model in Data with Long-term Survival

Piecewise regression criteria

Description

A function to determine if a KM curve is matured.

Usage

PRcriteria(
  smix.chosen,
  timepoints = NULL,
  slope.relative.change.threshold = 0.7,
  ratio.long.short.threshold = 1/3
)

Arguments

smix.chosen	Input. Extracted survival probabilities.
timepoints	Input. Time points of the extracted survival probabilities.
slope.relative.change.threshold	Input. A threshold for the relative change of the slopes of the last segment and the first segment.
ratio.long.short.threshold	Input. A threshold for the ratio of the lengths of the last segment to the first two segments.

Value

knots	knots.
slop.flatline.with.95CI	the slop of the KM tail with 95% CI.
slope.relative.change	the relative change of the slopes of the last segment and the first segment.
ratio.long.short	the ratio of the lengths of the last segment to the first two segments.
sig	TRUE if matured
PReg.fit	the fitted results of piecewise regression.

References

E.P.Y. Lin, C.Y. Hsu, J.F. Chiou et al. (2022). Cox Proportional Hazard Ratios Overestimate Survival Benefit of Immune Checkpoint Inhibitors (ICI): Cox-TEL Adjustment and Meta-analyses of PD-L1 Expression and ICI Survival Benefit. J Thorac Oncol. 17, 1365-1374. doi:10.1016/j.jtho.2022.08.010.

E.P.Y. Lin, C.Y. Hsu, L. Berry, P. Bunn, Y. Shyr (2022). Analysis of Cancer Survival associated with Immune Checkpoint Inhibitors after Statistical Adjustment: A Systematic Review and Meta-analyses. JAMA Network Open. 5(8):e2227211. doi:10.1001/jamanetworkopen.2022.27211.

Examples

library(CoxTEL)
s1mix.chosen <- c(39, 25, 18.2, 16, 14, 12.4, 11, 11)/100
s0mix.chosen <- c(32, 18,  9.9,  8,  4,    3,  1,  1)/100
PRC1 <- PRcriteria(smix.chosen=s1mix.chosen)
PRC0 <- PRcriteria(smix.chosen=s0mix.chosen)
PRC1$sig
PRC0$sig

cyhsuTN/AdjCOXPH documentation built on July 4, 2024, 8:05 p.m.