Method
How matching works
The matcher reads public registry records and the dates you type. It does not decide whether you can enroll. The study team does.
What gets searched
A search asks ClinicalTrials.gov, EU-CTR, WHO ICTRP, and ISRCTN for studies that are recruiting or not yet recruiting. The condition query joins the canonical name and longer synonyms with OR. Biomarkers are a separate query, merged afterward, so a marker such as HER2 does not delete breast cancer studies that never mention that word. Prior drugs stay out of the retrieval filter. They are compared later with the eligibility text.
Location is not used to drop the rest of the country. A study with no listed sites stays unknown, not distant. Phase II and later is the default view because early-phase studies are a smaller set. Counts on the condition pages are the ClinicalTrials.gov slice of that query, cached for about a day, with the retrieval time printed on the page.
How a line is labeled
Eligibility text is split into inclusion and exclusion. Each line is marked met, not met, or unknown, and the registry sentence is kept next to the label. Age and sex use the structured fields when the registry sends them. An exclusion sentence that names a drug already on the profile is not scored as a point in favor of the study. A washout date is the registry interval plus the end date you entered. If the interval is not tied to that drug name, the forecast does not invent a date.
The number on a result is a summary of those labels. It is an estimate from public criteria and the information you provided. A model may help turn a patient narrative into those fields. It does not overrule a hard mismatch on age, sex, or an opposite biomarker, and it does not change the default order by itself.
What stays off the index
Search results live in this browser until you leave. Notes are not published, and a personalized result page is not indexed. Public pages list registry records only. Confirm any study with the team running it before you try to join.