Reading a ClinicalTrials.gov Record for Diligence

← Back to All Insights

A clinical trial registry entry is one of the few places in early-stage biomedical diligence where a company has made dated, structured, public statements about its own program — repeatedly, over years, and for applicable clinical trials under a regulatory obligation. It is easy to open one, check the recruitment status and close it.

The status line is the least informative field on the page. Two other things on the same record carry more: the revision history, and the results clock.

The obligation, stated precisely

Registration and results reporting for applicable clinical trials run on Section 801 of the Food and Drug Administration Amendments Act of 2007, implemented by the 2016 final rule at 42 CFR Part 11. The provision that matters most for reading a record is the deadline: results information for an applicable clinical trial subject to § 11.42 is due no later than one year after the study's primary completion date, per 42 CFR 11.44(a).

Two lawful exits from that deadline exist, and both are worth knowing because either can look like delinquency from the outside. A responsible party can certify that it is seeking initial approval, licensure or clearance of the product (11.44(c)) or approval of a new use (11.44(b)); either certification delays the results deadline, for no more than two years after the certification is submitted. A good-cause extension may be requested under 11.44(e). NIH maintains a plain-language summary of FDAAA 801 and the final rule that is the right place to start before drawing conclusions.

The revision history is the document

ClinicalTrials.gov keeps every submitted version of a record and will show them arranged by submitted date, with changes color-coded — deletions struck through, additions highlighted. Two versions can be selected and compared directly. On any study page this sits under the Record History tab (?tab=history on a study URL).

This is the part of the record a summary cannot give you, because the current version is by definition the version that survived. What the history shows is the sequence of decisions:

• Primary endpoint changes, and when they happened relative to the primary completion date. An endpoint revised before enrollment closed reads very differently from one revised afterwards.
• Enrollment revisions. An anticipated enrollment adjusted downward over successive versions is the registry's way of recording a recruitment problem that no field is named for.
• Primary completion date movement. One slip is a program. Six slips is a pattern, and the pattern is visible only in the history.
• Status transitions — particularly into and out of suspended or withdrawn states, and the sponsor's own stated reason where one was given.

None of these is disqualifying and none predicts anything. Trials are often revised for good reasons. What the history gives you is a dated, sponsor-authored account of how the program actually unfolded, which you can then put next to the narrative in the deck and see whether the two describe the same study.

Fields that are routinely over-read

Recruitment status. This is a self-reported field. For applicable clinical trials initiated since January 18, 2017, the rule requires it to be updated within 30 calendar days of a change (42 CFR 11.64), but “Recruiting” still means only that the record said recruiting as of its last update. Check the update date before treating it as current; the field is easy to quote as though it were live.

Enrollment. Anticipated and actual enrollment are different values and are easy to conflate when a figure is lifted into a memo. Which one is displayed depends on where the study is in its life.

Phase. ClinicalTrials.gov records trials without FDA-defined phases, including device trials, as “Not Applicable”; device programs are not organized the way drug programs are. A device trial without a phase is not a gap in the record.

Sponsor. The sponsor and the responsible party (the sponsor, or a principal investigator the sponsor has designated) are separate fields, and either may be an academic center, a cooperative group or a predecessor entity rather than the company pitching you. A trial is not automatically the target's asset because the target's technology appears in it.

What the registry cannot tell you

The registry is a registry. It records what sponsors submit, when they submit it. It does not know whether a site is actually enrolling this month, whether an investigator has concerns, whether an interim look has occurred, or whether the data will support the indication being sought. A clean record is not evidence of a clean program, and — the direction that matters more — an absent record is not evidence of an absent trial. Not every study is an applicable clinical trial subject to the reporting requirements; the definition of an applicable device clinical trial excludes small feasibility studies and prototype studies whose primary outcome relates to feasibility rather than health outcomes (42 CFR 11.10).

That asymmetry deserves to be stated flatly, because it is where confident-sounding diligence goes wrong: you can prove presence from this system, and you largely cannot prove absence.

What we do with it

In a Vantage assessment the registry record is read as a document with a history rather than a status line, and every trial fact that reaches the report carries its NCT identifier so the reader can open the same record and check the same versions. Where the record is silent — no results posted, no applicable-trial obligation, a sponsor that is not the target — the report says that, in those words, rather than smoothing the gap over with prose. Surfacing what the record does and does not contain is the deliverable; concluding what it means for the outcome is the reader's call, and we do not take it for them.