Process configuration into complete GSD analysis

Transforms raw study configuration into comprehensive group sequential design analysis with graphical multiple testing procedures. Orchestrates sequential processing steps to generate analysis schedules, hypothesis testing boundaries, operating characteristics, and comprehensive reporting outputs.

Usage

process_config(config)

Arguments

config

Study configuration list with 7 required components:

**Study Metadata:** - `study_name`: Character string identifying the study - `study_description`: Brief description of study design - `alpha`: Overall Type I error rate (typically 0.025 for one-sided)

**analyses:** Data frame defining analysis schedule. Each row specifies one planned analysis with trigger conditions: - `endpoint`: Endpoint triggering the analysis (e.g., "OS", "CR") - `strata`: List of patient strata contributing to trigger - `treatments`: List of treatment arms contributing to trigger - `sample_size`: Target sample size for binary endpoints (NA for TTE) - `events`: Target events for time-to-event endpoints (NA for binary) - `power_subsets_any`: Named list of hypothesis subsets for "at least one rejection" power - `power_subsets_all`: Named list of hypothesis subsets for "all rejections" power

**hypotheses:** Data frame defining statistical hypotheses. Each row specifies one hypothesis: - `type`: "Primary" or "Secondary" - `endpoint`: Endpoint being tested - `strata`: List of strata for this hypothesis - `control`: Control treatment arm name - `test`: Test treatment arm name - `analyses_analysed`: Vector of analysis indices (from analyses data frame) testing this hypothesis - `sf`: Spending function ("none", "asHSD", "asOF", "asP", "asKD", "asUser") - `sfpar`: Spending function parameter (e.g., gamma for HSD; NULL if not applicable) - `nominal`: Nominal alpha spending at interims (optional; NULL for spending function) - `test_method`: Statistical test ("logrank", "stratified_logrank", "unpooled_proportions", "pooled_proportions", "cmh")

**enroll_rate:** Data frame specifying enrollment assumptions by stratum: - `stratum`: Patient stratum identifier - `treatments`: List of treatment arms for this stratum - `rate`: Enrollment rate (patients per month) - `duration`: Enrollment duration (months) - `ratio`: Randomization ratio vector (e.g., c(1,1) for 1:1)

**distribution_tte:** Data frame with time-to-event distribution parameters: - `endpoint`: TTE endpoint name - `stratum`: Patient stratum - `treatment`: Treatment arm - `duration`: Duration of constant hazard period (Inf = constant throughout) - `fail_rate`: Hazard rate (events per month) - `dropout_rate`: Dropout hazard rate (per month)

**distribution_bin:** Data frame with binary endpoint parameters: - `endpoint`: Binary endpoint name - `stratum`: Patient stratum - `treatment`: Treatment arm - `rate`: Response rate (probability of success) - `maturity_time`: Time (months) when endpoint can be evaluated

**graph:** Graphical multiple testing procedure specification: - `g`: Transition matrix (k×k for k hypotheses). Element g[i,j] = fraction of alpha from hypothesis i transferred to j when i is rejected. Diagonal = 0, row sums ≤ 1 - `w`: Initial weight vector (length k). Element w[i] = initial alpha allocation to hypothesis i. Sum must equal 1.0

Value

List with 22 components: - `analyses`: Enriched analyses data frame (original + 11 computed columns including index, dist_type, hypotheses_analysed, time, description_trigger_short/long, hypotheses_information_fractions, hypotheses_information) - `hypotheses`: Enriched hypotheses data frame (original + 18+ computed columns, expanded by weight scenarios, including information_fractions, specs, power, hurdles, nominal_p, description_effect_size) - `tables`: List of 7 summary tables (table1-5: hypothesis/analysis summaries; table6a-b: operating characteristics by analysis and overall) - `config`: Original configuration with processed distribution and enroll_rate - `graph_figure`: ggplot object visualizing graphical testing procedure - `information_figure`: ggplot showing information availability over time - `alpha_spend_figure`: ggplot of alpha spending functions by hypothesis - `timeline_type1_figure`: ggplot of study timeline (hypothesis-centric view) - `timeline_type2_figure`: ggplot of study timeline (analysis-centric view) - `bin_figure`: ggplot of binary endpoint distributions - `bin_rd_figure`: ggplot of binary endpoint risk differences - `tte_figure`: ggplot of time-to-event survival curves - `tte_ahr_figure`: ggplot of average hazard ratio over time - `tte_cumhaz_figure`: ggplot of cumulative hazard functions - `tte_dropout_figure`: ggplot of dropout hazard rates - `tte_dropout_probability_figure`: ggplot of dropout probabilities over time - `tte_hazard_figure`: ggplot of hazard rate functions - `tte_hr_figure`: ggplot of hazard ratios over time - `tte_median_figure`: ggplot of median survival times - `tte_quantiles_figure`: ggplot of survival quantiles - `tte_weighted_figure`: ggplot of weighted survival curves - `er_figure`: ggplot of enrollment rates by stratum - `er_cum_figure`: ggplot of cumulative enrollment over time

Examples

if (FALSE) { # \dontrun{
# Load example configuration
config <- load_config_from_repository("dualEP_study")

# Process configuration
results <- process_config(config)

# Access outputs
print(results$tables$table1)  # Hypothesis summary
print(results$graph_figure)   # Graphical procedure
} # }

Usage

Arguments

Value

See also

Examples