GRIDINT Research · Medicine

The phase-three gate.

Phase three is where medicine gets expensive: the last trial before a treatment reaches the world. This report reads a window of 1,000 completed phase-3 trials (987 of them carry complete rows into the data hub’s explorer — the missing thirteen are counted, not hidden) from the U.S. registry and finds the gate’s true shape: the median trial enrolled just 224 patients — while the largest enrolled 79,560, a — spread inside a single phase. Every figure on this page is computed from the window at render time, and the window is stated, not hidden.

published 2026-08-27 · data cutoff 2026-08-26 · edition 1 — updated only by a new dated edition · source: ClinicalTrials.gov v2

THE FIELD

A thousand gates, sized.

Every trial in the window: start year across, enrollment up — on a log scale, because the spread demands it. The dense band is ordinary phase three — hundreds of patients. The dots floating above are the mega-trials: vaccines and cardiovascular outcomes studies that enroll like small cities. The slider hides the small end; the color marks the giants.

A1Enrollment × start year, log scale — the mega-trials in amber
WHAT YOU ARE LOOKING AT Each dot is one completed phase-3 trial with a cataloged start date and enrollment (— of the 1,000 carry both). Amber marks the top decile by enrollment; pink the top ten trials outright. Tap or hover any dot for the trial.
ENROLLMENT ≥ 0 (log steps: 0 → 10 → 100 → 1k → 10k)
sourceClinicalTrials.gov v2 · phase-3, completed · window of 1,000 stated methodlog10 enrollment · rows missing a value are absent by honesty
THE SPONSORS

Who runs the gate.

Ranked by completed trials in the window, the sponsor list reads like a pharmaceutical index — the top fifteen names account for — of the window — with academic centers and public agencies threading between them. The dot on each bar is that sponsor’s median enrollment: some run many small gates, some run few giant ones.

B1The fifteen most frequent sponsors — trials and median enrollment
WHAT YOU ARE LOOKING AT Bars count completed trials per lead sponsor; the dot marks the sponsor’s median enrollment against the window median (dashed). Tap or hover for both numbers.
sourcesame window · lead sponsor as registered
THE CONDITIONS

What the gate is for.

The most-trialed conditions in the window. The list is medicine’s money map: chronic disease dominates late-stage research — arthritis, asthma, hypertension, HIV, the big cancers — and the tail of single-trial conditions is exactly where patients wait longest.

C1The twenty most-trialed conditions
WHAT YOU ARE LOOKING AT Conditions ranked by completed trials naming them (a trial can name several; the first three are counted — stated). Tap or hover for counts.
sourcesame window · conditions as registered, first three per trial
THE MEGA-TRIALS

The thirty largest gates.

Sorted by enrollment, the trials that moved tens of thousands of patients through phase three — each with its registry identifier. Deliberately the last data view: a table is the record, not the story.

D1The thirty largest by enrollment — edition 1
sourceClinicalTrials.gov · NCT ids resolve at clinicaltrials.gov/study/<id>
METHOD

How this was made, and its limits.

The data: the U.S. clinical-trials registry’s public API, phase-3 studies, pulled by our warehouse on schedule. Every figure is computed from the window in your browser. The cards are the limits — and this report’s first limit is its own window, stated in its standfirst.

A window, not the registry
1,000 studies per the API page the warehouse holds — completed trials only. Shares are of the window and say so; the next edition widens it.
Registry ≠ results
Completion means the trial ended — not that the treatment worked, published, or reached approval.
Enrollment as registered
The registry’s count field can be actual or anticipated; this payload does not distinguish, so neither do we — stated.
Conditions are self-declared
Sponsors name their own condition terms; we count the first three per trial without recoding.
Lead sponsor only
Collaborators are not counted — a large academic network can hide behind a single lead name.
FINDINGS

Four things the window shows.

1The gate is smaller than its reputation. The median completed phase-3 trial enrolled — patients. The blockbuster image of phase three — tens of thousands of subjects — describes a tiny elite of trials, not the phase. → the field, A1
2The top ten trials carry — of all enrollment. Ten studies — one percent of the window — enrolled as many patients as the bottom — combined. Concentration, not scale, is the phase’s signature. → the mega-trials, D1
3Fifteen names run — of the gate. The most frequent lead sponsors are the industry’s biggest — and their strategies differ visibly: many small gates for some, few giant ones for others. → the sponsors, B1
4The money map is chronic disease. — leads the conditions table — late-stage medicine concentrates where patients are many and treatment is lifelong. → the conditions, C1
A GRIDINT RESEARCH REPORT · MEDICINE · AUTHORED FROM OFFICIAL RECORDS · RENDERED BY THE GRIDINT BUILDER
published 2026-08-27 · data cutoff 2026-08-26 · edition 1 — updated only by a new dated edition

RELATED BY SOURCE

ClinicalTrials.govClinical trials data hubthe hub behind itClinicalTrials.govClinical trials — built with GRIDINTthe same records