> ## Documentation Index
> Fetch the complete documentation index at: https://docs.valyu.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# KRAS G12C phase 3 trial landscape

> Map clinical trial records to dated result disclosures, such as KRAS G12C phase 3 trials, in a cited landscape and CSV with Valyu DeepResearch.

**Example question:** What did CodeBreaK 200 and KRYSTAL-12 disclose about previously treated KRAS G12C-mutated NSCLC by the end of 2024?

**You get:** A two-trial landscape and CSV separating registered design from dated result disclosures, including abstract versus full-paper evidence.

**Start with:** Interventions; population and line of therapy; phase; design; disclosure cutoff; required trial identifiers.

## Run with DeepResearch

Set `VALYU_API_KEY` in your environment, then run this illustrative request. Change the question and CSV fields for your own task. The CSV fields below describe the requested output, not a saved result.

<Accordion title="Run the research request">
  ```bash cURL theme={null}
  curl --fail-with-body https://api.valyu.ai/v1/deepresearch/tasks \
    -H "x-api-key: $VALYU_API_KEY" \
    -H "Content-Type: application/json" \
    --data-binary @- <<'JSON'
  {
    "query": "Map CodeBreaK 200 and KRYSTAL-12 as randomized phase 3 studies in previously treated KRAS G12C-mutated advanced non-small-cell lung cancer. Use a disclosure cutoff of 31 December 2024. Separate current registry design metadata from original result disclosures available by that date. Distinguish conference abstracts from full journal reports. Do not use later papers to fill missing cutoff-era results. Cite each row and mark inaccessible or absent fields. Do not rank sotorasib against adagrasib from separate trials.",
    "mode": "standard",
    "research_strategy": "Prioritize ClinicalTrials.gov for design and original conference abstracts or journal papers for results. Use secondary coverage only as a discovery lead, not a replacement citation. Cite every row; mark missing or inaccessible evidence.",
    "report_format": "A concise evidence table, comparison limits and unresolved questions.",
    "deliverables": [
      {
        "type": "csv",
        "description": "One row per evidence item with the requested fields and supporting source URL.",
        "columns": [
          "nct_id",
          "trial_name",
          "population",
          "phase",
          "intervention",
          "comparator",
          "primary_endpoint",
          "record_accessed_date",
          "disclosure_type",
          "disclosure_date",
          "reported_result",
          "source_url",
          "cutoff_limit"
        ]
      }
    ]
  }
  JSON
  ```
</Accordion>

Save the returned `deepresearch_id`. Poll [Get status](/api-reference/endpoint/deepresearch-status) until the task completes, then read the report and download the CSV from `deliverables`.

## How it works

1. Find registry records for CodeBreaK 200 and KRYSTAL-12. Extract design and identifiers with record access dates.

2. Locate original result disclosures available by the cutoff. Label conference abstracts separately from full papers and post-cutoff publications.

3. Join disclosures to trials using identifiers. Return separate rows for different disclosure dates and distinguish design metadata from reported results.

To build the workflow inside your own agent, use [Search](/search/quickstart) to find sources and [Contents](/guides/content-extraction) to read selected pages. Keep the CSV fields above as your extraction schema.

## Check the result

Check trial identifiers and publication dates independently. An earlier conference disclosure and a later full paper are different evidence records. Review population, denominator and endpoint definitions before comparing any figures.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.