Skip to contents

Getting Started with dv.spiderplot

dv-spiderplot.Rmd

The DaVinci Spider Plot Module is an interactive Shiny module designed for visualizing individual patient trajectories over time in clinical trials, particularly valuable for oncology efficacy analysis. This module creates spider plots that display patient-level response data, showing changes in tumor size or other clinical measurements across multiple visits.

Data Preparation

Before creating our spider plot module, we need to prepare the clinical trial data. We’ll use example datasets from the pharmaverseadam package, which provides CDISC ADaM-compliant datasets commonly used in clinical trials.

adsl <- pharmaverseadam::adsl
adtr <- pharmaverseadam::adtr_onco |>
  dplyr::mutate(
    ARM = factor(ARM, levels = c("Placebo", "Xanomeline Low Dose", "Xanomeline High Dose")),
    SEX = factor(SEX, levels = c("F", "M"), labels = c("Female", "Male")),
    AVISIT = forcats::fct_reorder(AVISIT, AVISITN)
  )

# add back labels lost during data manipulation
attr(adtr[["ARM"]], "label") <- "Planned Arm"
attr(adtr[["SEX"]], "label") <- "Sex"
attr(adtr[["AVISIT"]], "label") <- "Analysis Visit"

# add timestamp to be displayed in the app
attr(adsl, "meta") <- list(mtime = Sys.time())
attr(adtr, "meta") <- list(mtime = Sys.time())

Module Configuration

Now that our data is prepared, we can configure the spider plot module via the mod_spiderplot() function. This configuration defines how the spider plot will behave and what options will be available to users.

spiderplot_mod <- dv.spiderplot::mod_spiderplot(
  module_id = "mod_spider",
  subject_level_dataset_name = "adsl",
  results_dataset_name = "adtr",
  subjid_var = "USUBJID",
  x_vars = c("ADY", "AVISIT"),
  y_vars = c("PCHG", "CHG"),
  color_vars = c("ARM", "AGEGR1"),
  color_palette = c(
    "Placebo" = "red",
    "Xanomeline Low Dose" = "green",
    "Xanomeline High Dose" = "blue"
  ),
  tooltip = c(
    "Unique Subject Identifier: " = "USUBJID",
    "Planned Arm: " = "ARM",
    "Analysis Visit: " = "AVISIT",
    "Days Since Randomization: " = "ADY",
    "Change from Baseline: " = "CHG",
    "Percent Change from Baseline: " = "PCHG"
  ),
  facet_rows = c("SEX", "ETHNIC"),
  facet_cols = c("ARM", "AGEGR1"),
  title = "Interactive Spider Plot",
  subtitle = "CDISC-PILOT-01"
)

Running the Demo Application

With our module configured, we can now launch the interactive spider plot application using the DaVinci module manager.

dv.manager::run_app(
  data = list("Demo" = list(adsl = adsl, adtr = adtr)), 
  module_list = list("Spider Plot" = spiderplot_mod),
  title = "Spider Plot Demo",
  filter_data = "adsl",
  filter_key = "USUBJID",
  filter_type = "datasets"
)