MBdelayed: Simulated survival dataset with delayed treatment effect

In keaven/simtrial: Clinical Trial Simulation

Simulated survival dataset with delayed treatment effect

Description

Magirr and Burman (2019) considered several scenarios for their modestly weighted logrank test. One of these had a delayed treatment effect with a hazard ratio of 1 for 6 months followed by a hazard ratio of 1/2 thereafter. The scenario enrolled 200 patients uniformly over 12 months and cut data for analysis 36 months after enrollment was opened. This dataset was generated by the simtrial::sim_pw_surv() function under the above scenario.

Usage

MBdelayed

Format

A tibble with 200 rows and xx columns

tte

time to event

References

Magirr, Dominic, and Carl‐Fredrik Burman. "Modestly weighted logrank tests." Statistics in Medicine 38.20 (2019): 3782-3790.

Examples

library(tidyr)
library(dplyr)
library(survival)
library(mvtnorm)
fit <- survfit(Surv(tte, event) ~ Treatment, data = MBdelayed)

# Plot survival
plot(fit, lty=1:2)
legend("topright", legend = c("Control", "Experimental"), lty = 1:2)

# Set up time, event, number of event dataset for testing
# with arbitrary weights
ten <- MBdelayed %>% counting_process(arm = "Experimental")
head(ten)

# MaxCombo with logrank, FH(0,1), FH(1,1)
ten %>% tenFHcorr(rg = tibble(rho=c(0, 0, 1), gamma = c(0, 1, 1))) %>%
        pvalue_maxcombo()

# Magirr-Burman modestly down-weighted rank test with 6 month delay
# First, add weights
ten <- ten %>% mb_weight(6)
head(ten)

# Now compute test based on these weights
ten %>% summarise(S = sum(o_minus_e*mb_weight),
                  V = sum(var_o_minus_e*mb_weight^2),
                  Z = S/sqrt(V)) %>%
        mutate(p = pnorm(Z))

# Create 0 weights for first 6 months
ten <- ten %>% mutate(w6 = 1 * (tte >= 6))
ten %>% summarise(S = sum(o_minus_e*w6),
                  V = sum(var_o_minus_e*w6^2),
                  Z = S/sqrt(V)) %>%
        mutate(p=pnorm(Z))

# Generate another dataset
ds <- sim_pw_surv(n = 200,
                enroll_rate = tibble(rate = 200 / 12, duration = 12),
                fail_rate = tribble(
                   ~Stratum, ~Period, ~Treatment,     ~duration, ~rate,
                   "All",        1,   "Control",      42,        log(2) / 15,
                   "All",        1,   "Experimental", 6,         log(2) / 15,
                   "All",        2,   "Experimental", 36,        log(2) / 15 * 0.6),
                dropoutRates = tribble(
                   ~Stratum, ~Period, ~Treatment,     ~duration, ~rate,
                   "All",        1,   "Control",      42,        0,
                   "All",        1,   "Experimental", 42,        0)
       )
# Cut data at 24 months after final enrollment
MBdelayed2 <- ds %>% cut_data_by_date(max(ds$enroll_time) + 24)

keaven/simtrial documentation built on April 17, 2023, 4:03 a.m.