Skip to contents

Cancer Biomarker Time-Dependent ROC Test Dataset

Source: R/data_timeroc_docs.R
timeroc_cancer_biomarker.Rd

Simulated dataset representing a cancer biomarker study with tumor marker measurements and survival outcomes. Designed to test basic time-dependent ROC functionality with realistic cancer progression patterns.

Usage

timeroc_cancer_biomarker

Format

A data frame with 300 observations and 9 variables:

patient_id

Character. Unique patient identifier (CA_001 to CA_300)

age

Integer. Patient age at diagnosis (30-90 years)

sex

Character. Patient sex ("Male", "Female")

cancer_stage

Character. Cancer stage at diagnosis ("I", "II", "III", "IV")

tumor_biomarker

Numeric. Continuous tumor biomarker level (higher = worse prognosis)

follow_up_months

Numeric. Follow-up time in months (0-60)

death_event

Integer. Death indicator (1 = death, 0 = censored)

treatment_type

Character. Treatment received ("Surgery", "Surgery+Chemo", "Surgery+Radio", "Palliative")

hospital_center

Character. Treatment center ("Center_A" to "Center_E")

Source

Simulated data generated using create_timeroc_test_data.R

Details

This dataset simulates a cohort study following cancer patients for up to 60 months. The tumor biomarker shows realistic stage-dependent distributions with higher values in advanced stages. Survival times are generated based on biomarker level, age, and cancer stage using exponential survival models.

Key Features:

  • Realistic biomarker-outcome associations

  • Stage-stratified survival patterns

  • Administrative censoring at 60 months

  • Random dropout patterns

  • 245/300 events (81.7% event rate)

Recommended TimeROC Parameters:

  • Timepoints: 12, 36, 60 months

  • Marker: tumor_biomarker

  • Event: death_event

  • Time: follow_up_months

See also

timeroc, timeroc_datasets

Examples

if (FALSE) { # \dontrun{
# Load the dataset
data(timeroc_cancer_biomarker)

# Basic time-dependent ROC analysis
result <- timeroc(
  data = timeroc_cancer_biomarker,
  elapsedtime = "follow_up_months",
  outcome = "death_event", 
  marker = "tumor_biomarker",
  timepoints = "12, 36, 60"
)

# View AUC results
result$aucTable$asDF
} # }