Simple Tumor Marker and Cancer Progression Data — simple_cancer

A simplified longitudinal dataset containing tumor marker measurements and cancer progression for 100 cancer patients. Ideal for teaching and initial exploration of joint models.

Usage

simple_cancer_data

Format

A data frame with 588 observations and 6 variables:

patient_id: Character. Unique patient identifier (CA_001 to CA_100)
age: Numeric. Patient age at baseline (years)
treatment: Factor. Treatment group (Standard, Experimental)
visit_time: Numeric. Time of tumor marker measurement (months from baseline)
tumor_marker: Numeric. Tumor marker level (units/mL)
survival_time: Numeric. Time to progression/death or last follow-up (months)
progression_status: Numeric. Event indicator (0 = censored, 1 = progression/death)

Source

Simulated data for educational purposes

Details

This simplified dataset is perfect for:

Learning joint modeling concepts
Quick algorithm testing
Demonstrating treatment effects
Smaller sample size for faster computation
Clear biomarker-survival relationship

Features:

Tumor marker levels generally increase over time
Experimental treatment slows marker increase
Higher marker levels increase progression hazard
10% event rate over 24 months follow-up

Examples

data(simple_cancer_data)

# Basic joint modeling analysis
library(ggplot2)

# Marker trajectories by treatment
ggplot(simple_cancer_data, aes(x = visit_time, y = tumor_marker, color = treatment)) +
  geom_smooth(method = "loess") +
  labs(title = "Tumor Marker by Treatment", 
       x = "Time (months)", y = "Tumor Marker")

# Individual trajectories for first 20 patients
first_20 <- subset(simple_cancer_data, patient_id %in% unique(patient_id)[1:20])
ggplot(first_20, aes(x = visit_time, y = tumor_marker, group = patient_id)) +
  geom_line(alpha = 0.7) + geom_point(alpha = 0.5) +
  facet_wrap(~treatment) +
  labs(title = "Individual Tumor Marker Trajectories")