ClinicalTrials.gov Module

Overview

The ClinicalTrials module queries the ClinicalTrials.gov v2 REST API to retrieve registered clinical trials for each candidate drug in the specified cancer context.

How It Works

1. All candidate drugs from PubMed and TxGNN modules are collected
2. For each drug, the API is queried: "<drug name> AND <cancer type>"
3. Up to 200 records per drug are retrieved
4. A per-drug trial summary is computed (total trials, highest-phase trial)

Key Output Fields

Raw output (clinicaltrials_output.tsv)

Field	Description
drug	Drug name
nct_id	National Clinical Trial identifier
title	Brief study title
condition	Reported disease conditions
phase	Trial phase (Phase 1--4)
status	Recruitment status

Summary (merged into final report)

Field	Description
n_clinical_trials	Total number of matched trials
top_nct_id	NCT ID of the highest-phase trial
top_phase	Phase of the highest-phase trial
top_title	Title of the highest-phase trial

Phase Priority

Trials are ranked by phase for the summary:

Phase 4 > Phase 3 > Phase 2 > Phase 1

Usage

from clinicaltrials_module import run_clinical_trials, build_trial_summary_table

clinical_df = run_clinical_trials(
    drug_list=["FOSTAMATINIB", "SORAFENIB"],
    cancer_type="lung cancer",
    output_path="clinicaltrials_output.tsv"
)

summary_df = build_trial_summary_table(clinical_df)