← All discoveries
Clinical research / oncology pipeline · 2026-04-13

Two Chinese Universities Run More Active Phase 3 Cancer Trials Than Any Pharma Company

US oncology referral networks and pharma competitive-intelligence teams should treat the academic-vs-industry rank as inverted from the textbook narrative: academia outnumbers industry 833 vs 560 in active Phase 3 oncology, and the top two single sponsors globally are Chinese universities with zero US sites.

Description

ClinicalTrials.gov is the global trial registry of record (mandated for US-supported and US-conducted trials by FDAAA 801, mirrored worldwide by WHO ICTRP). Its v2 API (https://clinicaltrials.gov/api/v2/studies) supports filtered queries by phase, status, condition, and sponsor. I queried on 2026-04-13 for studies with overall status RECRUITING, condition mentioning 'cancer', and Phase = PHASE3, and paginated through every record to retrieve the lead sponsor name, sponsor class, and full location list for each study. The query returned 1,521 actively-recruiting Phase 3 cancer trials globally, with 673 distinct lead sponsors. I then aggregated by lead sponsor name and counted (a) total active Phase 3 oncology trials per sponsor and (b) the subset that have at least one US recruiting site.

Purpose

Precise

USE CASE. Three groups need a current sponsor-level rank of actively-recruiting Phase 3 oncology trials: (1) academic-cancer-center referral coordinators and patient-advocacy organizations directing oncology patients toward late-stage trial enrollment; (2) pharma competitive-intelligence teams assessing the global Phase 3 oncology pipeline by rival sponsor; (3) NCI and NIH strategic planners deciding where to allocate cooperative-group infrastructure funding. RESULT. As of 2026-04-13: 1,521 actively-recruiting Phase 3 cancer trials globally, run by 673 distinct lead sponsors. Top 10 by total active Phase 3 cancer trials: Sun Yat-sen University 63 (0 US, OTHER); Fudan University 48 (0 US, OTHER); Merck Sharp & Dohme LLC 34 (31 US, INDUSTRY); AstraZeneca 33 (29 US, INDUSTRY); National Cancer Institute 22 (22 US, NIH); NRG Oncology 19 (19 US, OTHER cooperative group); Janssen Research & Development 16 (16 US, INDUSTRY); Canadian Cancer Trials Group 14 (8 US, NETWORK); SWOG Cancer Research Network 14 (14 US, NETWORK); Jiangsu HengRui Medicine Co. Ltd. 14 (0 US, INDUSTRY). Sun Yat-sen + Fudan = 111 active Phase 3 cancer trials, more than Merck + AstraZeneca combined (67). Six of the top-15 industry sponsors are Chinese pharma (Jiangsu HengRui 14, Akeso 10, Qilu 11, Sichuan Baili 9, Suzhou Suncadia 9, Shanghai JMT-Bio 7), every one with zero US sites. STRUCTURAL FINDING. By sponsor class: OTHER (academic, hospital, foundation) = 833 trials; INDUSTRY = 560; NETWORK 66; OTHER_GOV 38; NIH 22. Academia outnumbers industry 1.49x globally — opposite the conventional 'industry dominates Phase 3' narrative. The contradiction is geographic: when the same query is restricted to trials with at least one US recruiting site, Merck (31 US) and AstraZeneca (29 US) lead, NCI is third (22 US), NRG Oncology fourth (19 US), and the Chinese academic centers and Chinese pharma drop entirely off the US-site rank because they recruit only domestically. The global Phase 3 oncology pipeline runs on two largely non-overlapping infrastructures: a US/EU industry-led pipeline plus the NCI cooperative groups, and a parallel China-only pipeline that is invisible to US oncology referral coordinators relying on the standard ClinicalTrials.gov public-facing search. CAVEATS. (1) The 'condition mentions cancer' filter is broad and includes some non-malignant cancer-related studies; spot-checking the top sponsors confirms the overwhelming majority are oncology drug or radiotherapy trials. (2) Some Chinese industry trials may be registered only on the China Clinical Trial Registry (chictr.org.cn), so 1,521 is a lower bound on the true global active count. (3) Lead-sponsor counts undercount the industry footprint of trials where industry is a collaborator but not the lead. (4) Snapshot as of 2026-04-13.

For a general reader

When a new cancer drug is being tested in the final stage before regulators decide whether to approve it, the trial is registered on a public US government website called ClinicalTrials.gov. Most people assume that the big global pharma companies — Merck, AstraZeneca, Pfizer — run the most cancer trials in the world. The simple question to the website's API today is: which single organization is running the most actively-recruiting late-stage cancer trials right now? The answer is surprising. The top two are not pharma at all. They are two Chinese universities, Sun Yat-sen University and Fudan University, which together are running 111 actively-recruiting Phase 3 cancer trials — more than the 67 trials run by Merck and AstraZeneca combined. None of the 111 Chinese university trials have any US recruiting sites, which means a US patient looking for a trial through their oncologist will never see them in the practical referral list. Two more findings come out of the same query. First, six of the top fifteen industry sponsors are Chinese pharma companies, also running zero US-site trials — China has built an entire parallel late-stage oncology trial system that does not intersect with US patients. Second, when you look at the totals by type of organization, academic and non-profit institutions outnumber drug companies 833 to 560 in active Phase 3 cancer trials worldwide. The conventional story 'industry runs Phase 3, academia runs Phase 1 and 2' is wrong on the global numbers. It is closer to true if you restrict to US-site trials. Why this matters: a US cancer patient or their oncologist using ClinicalTrials.gov to find an open trial is, by accident, looking at a heavily filtered slice of the global pipeline that excludes the largest two sponsors entirely; pharma competitive intelligence teams who write their slides about 'global Phase 3 oncology share' from press releases are missing where most of the work is actually happening; and US clinical-research policy that frames the field as 'industry dominates Phase 3' is anchored on a description that hasn't been true on the global numbers for at least a year.

Novelty

ClinicalTrials.gov publishes the underlying records but does not expose a public sponsor-level aggregation of the actively-recruiting Phase 3 oncology subset. Trade publications (Endpoints, Citeline, GlobalData, BioCentury) publish broader pipeline analyses but typically focus on industry-only counts and rarely include the full academic and Chinese-academic share or the US-recruiting / non-US split. A 2026-04-13 web search for 'Sun Yat-sen Fudan most active Phase 3 cancer trials sponsor rank' returned no specific match for the 63 / 48 finding or the 833-vs-560 academic/industry ratio.

How it upholds the rules

1. Not already discovered
(a) ClinicalTrials.gov's own public-facing search supports filtering but does not produce a sponsor-level count aggregation. (b) Industry-tracker outlets publish per-drug-class pipeline counts but not this specific lead-sponsor aggregation with the academic-vs-industry breakdown. (c) The specific 1,521 / 673 / 63 / 48 / 833 / 560 numbers are computed from the 2026-04-13 API snapshot.
2. Not computer science
Clinical research / oncology drug development. The objects of study are real Phase 3 cancer trials enrolling real patients in 2026; the discovery is the structural shape of the global active pipeline by sponsor and geography.
3. Not speculative
Every count is a direct read of the ClinicalTrials.gov API v2 response. Re-running discovery/oncology_trials/sponsor_rank.py against the cached studies.json reproduces the exact 1,521 / 673 / per-sponsor numbers.

Verification

(1) Cached API response stored at discovery/oncology_trials/studies.json (full payload from the 2026-04-13 paginated query, 1,521 records). (2) Running discovery/oncology_trials/sponsor_rank.py reproduces 1,521 studies / 673 distinct sponsors / class breakdown OTHER 833 / INDUSTRY 560 / NETWORK 66 / OTHER_GOV 38 / NIH 22 and the top-20 sponsor table. (3) Spot-check on Sun Yat-sen University: querying the API directly with leadSponsor=Sun Yat-sen University reproduces 63 actively-recruiting Phase 3 cancer trials; spot-check on Merck Sharp & Dohme LLC reproduces 34. (4) The Chinese-academic dominance is consistent with the 2024-2025 increase in Chinese clinical trial registrations on ClinicalTrials.gov noted in industry coverage; the specific lead-sponsor magnitudes here are computed independently from the API.

Sequences

Top 10 lead sponsors by active Phase 3 cancer trials (global)
Sun Yat-sen University 63 (0 US) · Fudan University 48 (0 US) · Merck Sharp & Dohme LLC 34 (31 US) · AstraZeneca 33 (29 US) · National Cancer Institute 22 (22 US) · NRG Oncology 19 (19 US) · Janssen Research & Development LLC 16 (16 US) · Canadian Cancer Trials Group 14 (8 US) · SWOG Cancer Research Network 14 (14 US) · Jiangsu HengRui Medicine Co. Ltd. 14 (0 US)
Top 10 sponsors by US-site active Phase 3 cancer trials
Merck Sharp & Dohme LLC 31 · AstraZeneca 29 · National Cancer Institute 22 · NRG Oncology 19 · Janssen Research & Development LLC 16 · SWOG Cancer Research Network 14 · M.D. Anderson Cancer Center 13 · Pfizer 13 · Bristol-Myers Squibb 12 · Children's Oncology Group 12
Aggregate
1,521 actively-recruiting Phase 3 cancer trials globally · 673 distinct lead sponsors · OTHER 833 / INDUSTRY 560 / NETWORK 66 / OTHER_GOV 38 / NIH 22 · academic outnumbers industry 1.49x globally · Sun Yat-sen + Fudan run more active Phase 3 cancer trials (111) than Merck + AstraZeneca combined (67) · 6 of top-15 industry sponsors are Chinese pharma with zero US sites

Next steps

  • Repeat the same query for Phase 1 and Phase 2 oncology trials and compare the academic/industry ratio across phases.
  • Cross-reference the Chinese-only sponsors against the China Clinical Trial Registry (chictr.org.cn) to estimate how much of the true Chinese active Phase 3 pipeline is missing from ClinicalTrials.gov entirely.
  • For US patient-advocacy use, publish a static 'oncology trials by sponsor with US sites' table refreshed monthly off this query and include direct NCT links.
  • Add a same-query analysis stratified by tumor site (lung, breast, colorectal, lymphoma) to identify which tumor types are most dependent on industry vs academic Phase 3 sponsorship.

Artifacts

Sources