Skip to contents

Kaplan-Meier (KM) plot function for anchored and unanchored cases using ggplot

Source: R/plot_km2.R
kmplot2.Rd

This is wrapper function of basic_kmplot2. The argument setting is similar to maic_anchored and maic_unanchored, and it is used in those two functions.

Usage

kmplot2(
  weights_object,
  tte_ipd,
  tte_pseudo_ipd,
  trt_ipd,
  trt_agd,
  trt_common = NULL,
  trt_var_ipd = "ARM",
  trt_var_agd = "ARM",
  km_conf_type = "log-log",
  km_layout = c("all", "by_trial", "by_arm"),
  time_scale,
  ...
)

Arguments

weights_object

an object returned by estimate_weight

tte_ipd

a data frame of individual patient data (IPD) of internal trial, contain at least "USUBJID", "EVENT", "TIME" columns and a column indicating treatment assignment

tte_pseudo_ipd

a data frame of pseudo IPD by digitized KM curves of external trial (for time-to-event endpoint), contain at least "EVENT", "TIME"

trt_ipd

a string, name of the interested investigation arm in internal trial dat_igd (real IPD)

trt_agd

a string, name of the interested investigation arm in external trial dat_pseudo (pseudo IPD)

trt_common

a string, name of the common comparator in internal and external trial, by default is NULL, indicating unanchored case

trt_var_ipd

a string, column name in tte_ipd that contains the treatment assignment

trt_var_agd

a string, column name in tte_pseudo_ipd that contains the treatment assignment

km_conf_type

a string, pass to conf.type of survfit

km_layout

a string, only applicable for unanchored case (trt_common = NULL), indicated the desired layout of output KM curve.

time_scale

a string, time unit of median survival time, taking a value of 'years', 'months', weeks' or 'days'

...

other arguments in basic_kmplot2

Value

In unanchored case, a KM plot with risk set table. In anchored case, depending on km_layout,

  • if "by_trial", 2 by 1 plot, first all KM curves (incl. weighted) in IPD trial, and then KM curves in AgD trial, with risk set table.

  • if "by_arm", 2 by 1 plot, first KM curves of trt_agd and trt_ipd (with and without weights), and then KM curves of trt_common in AgD trial and IPD trial (with and without weights). Risk set table is appended.

  • if "all", 2 by 2 plot, all plots in "by_trial" and "by_arm" without risk set table appended.

Examples

# unanchored example using kmplot2
data(weighted_sat)
data(adtte_sat)
data(pseudo_ipd_sat)

kmplot2(
  weights_object = weighted_sat,
  tte_ipd = adtte_sat,
  tte_pseudo_ipd = pseudo_ipd_sat,
  trt_ipd = "A",
  trt_agd = "B",
  trt_common = NULL,
  trt_var_ipd = "ARM",
  trt_var_agd = "ARM",
  endpoint_name = "Overall Survival",
  km_conf_type = "log-log",
  time_scale = "month",
  break_x_by = 2,
  xlim = c(0, 20)
)

# anchored example using kmplot2
data(weighted_twt)
data(adtte_twt)
data(pseudo_ipd_twt)

# plot by trial
kmplot2(
  weights_object = weighted_twt,
  tte_ipd = adtte_twt,
  tte_pseudo_ipd = pseudo_ipd_twt,
  trt_ipd = "A",
  trt_agd = "B",
  trt_common = "C",
  trt_var_ipd = "ARM",
  trt_var_agd = "ARM",
  endpoint_name = "Overall Survival",
  km_conf_type = "log-log",
  km_layout = "by_trial",
  time_scale = "month",
  break_x_by = 2
)


# plot by arm
kmplot2(
  weights_object = weighted_twt,
  tte_ipd = adtte_twt,
  tte_pseudo_ipd = pseudo_ipd_twt,
  trt_ipd = "A",
  trt_agd = "B",
  trt_common = "C",
  trt_var_ipd = "ARM",
  trt_var_agd = "ARM",
  endpoint_name = "Overall Survival",
  km_conf_type = "log-log",
  km_layout = "by_arm",
  time_scale = "month",
  break_x_by = 2
)


# plot all
kmplot2(
  weights_object = weighted_twt,
  tte_ipd = adtte_twt,
  tte_pseudo_ipd = pseudo_ipd_twt,
  trt_ipd = "A",
  trt_agd = "B",
  trt_common = "C",
  trt_var_ipd = "ARM",
  trt_var_agd = "ARM",
  endpoint_name = "Overall Survival",
  km_conf_type = "log-log",
  km_layout = "all",
  time_scale = "month",
  break_x_by = 2,
  xlim = c(0, 20),
  show_risk_set = FALSE
)