The State of Rare Disease Clinical Trials With 75% of 2026 Behind Us

Somewhere in the United States, a family got a Goodpasture syndrome diagnosis this week. At some point, probably late at night, someone in that family will search for clinical trials. They will find zero. Not zero nearby. Zero anywhere on Earth. Goodpasture syndrome is 1 of 6 conditions we track with no recruiting trial, and that was true on September 9th, 2026, when we first ran this census, and still true when we ran it again on October 4th.

Hold that thought, because the rest of this story points the other way. With 75% of 2026 behind us, we did something we had never done: a full census of every study listed on ClinicalTrials.gov for every rare disease on Trial Friend, pulled on a single day using the same queries that power our disease pages. Then we lined the results up against the year's approvals, rejections, and the FDA decisions still pending. The picture that emerged is bigger, busier, and more unequal than we expected. This post walks through all of it, the good and the bad. The numbers below come from our October 4th re-run across 204 diseases, and a short section near the end shows what moved since September 9th.

How many rare disease clinical trials are running in 2026?

Across the 204 conditions we track, the October 4th census found 9,910 unique active studies. Of those, 5,754 are recruiting right now. Through the end of September, 1,768 of the active studies carry 2026 start dates, which works out to roughly 45 new rare disease studies opening every single week of this year. For a field that patients often experience as agonizingly slow, the research pipeline itself is anything but.

Momentum is one thing. Where that momentum lands is another, and that is where this year gets interesting.

The approvals tracked on our FDA calendar tell you what kind of year this has been. Through October 4th the calendar counts 19 rare disease approvals and label expansions in 2026, 16 of them between August 19th and September 30th. Between August 19th and September 3rd alone, a stretch of just 16 days, the FDA approved the 6 therapies below plus a label expansion for Imaavy into warm autoimmune hemolytic anemia. Several were firsts in the full sense of the word, for diseases that had never had an approved treatment.

Trutakna (atacicept), Vera Therapeutics

Approved July 7th for IgA nephropathy under the accelerated pathway. A self-administered weekly injection in a kidney disease that had zero targeted therapies a few years ago.

Pasatru (garetosmab), Regeneron

Approved August 19th for adults with fibrodysplasia ossificans progressiva, the ultra-rare condition where soft tissue turns to bone. It is the first FOP treatment shown in a placebo-controlled trial to reduce new bone lesions, joining Sohonos, the flare-management drug cleared in 2023.

Genglycos (pariglasgene brecaparvovec), Ultragenyx

Approved August 19th, an AAV8 gene therapy for glycogen storage disease type 1a granted accelerated approval.

Lisraya (brepocitinib), Priovant Therapeutics

Approved August 27th as the first oral targeted therapy for dermatomyositis.

Mimrylo (rusfertide), Takeda and Protagonist Therapeutics

Approved August 28th for polycythemia vera, the first hepcidin mimetic ever cleared for the disease.

Besremi (ropeginterferon alfa-2b), PharmaEssentia

Expanded August 30th into essential thrombocythemia.

Zanvastro (zilganersen), Ionis Pharmaceuticals

Approved September 3rd for Alexander disease, a first for this devastating leukodystrophy.

Read that list again and notice the pattern. Two of those 7, Genglycos and Zanvastro, are first-ever treatments for their disease, and most of the rest are firsts of their kind. September then added the 7 approvals listed in the decision section below, including first-ever treatments for Sanfilippo syndrome type A, ataxia-telangiectasia, and MCT8 deficiency, plus a pediatric expansion for Joenja in APDS and Gazyva for childhood-onset nephrotic syndrome. A decade ago, a run like that would have been a decade's worth of news for these communities. This year it happened over one summer.

The setbacks nobody puts in a press release

An honest census includes the failures, and 2026 has had real ones. Tabelecleucel, Pierre Fabre's cell therapy for EBV-positive post-transplant lymphoproliferative disease, took its second complete response letter on January 9th and has no new decision date yet. In May, Pierre Fabre said it had agreed with the FDA on a path to resubmit, using more patients and longer follow-up from its ALLELE trial. Camurus got a June rejection for Oclaiz, its self-injectable octreotide depot for acromegaly, and returns for a December 18th decision. UX111, the Sanfilippo gene therapy, was rejected outright in 2025 on manufacturing grounds unrelated to whether it works, and its second application was approved as Fayuvi on September 17th, 2026.

The strangest story of the year might be tolebrutinib. Sanofi's brain-penetrant BTK inhibitor for non-relapsing secondary progressive multiple sclerosis delayed disability progression by 31% in its Phase 3 trial, yet the FDA rejected it in December 2025 over liver injury risk, including one fatal case. Then in June 2026 the European Commission approved the exact same drug as Cenrifki. Right now an MS patient in Berlin can get a medicine that an identical patient in Boston cannot. Regulators looking at the same data reached opposite conclusions about whether the benefit justifies the risk, and both decisions are defensible. That is how thin the margins are in progressive MS.

Who sponsors rare disease clinical trials? (It is not who you think)

Ask most people who runs clinical trials and they will say drug companies. For rare diseases, the data says otherwise.

Industry leads just 26% of recruiting rare disease trials. Academic medical centers, hospitals, and nonprofits lead 68%. The 2 most prolific sponsors on the list are not pharmaceutical companies at all: Assistance Publique-Hopitaux de Paris, the Paris public hospital system, and the National Cancer Institute are tied at 75 recruiting rare disease trials each. MD Anderson follows at 61, then Mayo Clinic at 56. The top industry name, Novartis, sits at 60, with AstraZeneca at 46 and Takeda at 34. Memorial Sloan Kettering, Hospices Civils de Lyon, Cincinnati Children's, Penn, Mass General, and 3 hospital systems in China all crack the top 15.

How big is a typical rare disease trial? Smaller than you think

The median recruiting trial in our census wants 80 participants. Not 8,000. Eighty. Cardiovascular and diabetes trials routinely enroll thousands, which is why rare disease research runs on a completely different logic: small populations force small studies, and a single family enrolling or not enrolling can genuinely move a trial's timeline.

One number in this census speaks directly to the most common fear patients bring us. Among recruiting trials that report their study design, 73% are open label, meaning every participant receives the active treatment and knows it. The placebo-controlled, double-blind study people picture is the minority in rare disease research. We explain why in our guide to placebos, randomization, and blinding.

Which rare diseases have the most trials, and which have none?

Research attention is not spread evenly. It is not even close to evenly. Chronic graft-versus-host disease has 556 recruiting trials. Hodgkin lymphoma has 545, multiple sclerosis 429, myelodysplastic syndromes 342, pulmonary arterial hypertension 263. These conditions sit near well-funded research infrastructure: transplant programs, oncology networks, neurology departments.

Now the other end. Among our 204 diseases, 27 have 2 or fewer recruiting trials. The bottom 6 have none at all: Goodpasture syndrome, Factor V deficiency, Hajdu-Cheney syndrome, inherited erythromelalgia, linear IgA disease, and Bartonella infection. Real conditions, real patients, and in October 2026 not a single study anywhere is recruiting for any of them. The same 6 were at zero in September.

“The gap between the most-studied and least-studied rare diseases is not a gap. It is a canyon: 556 recruiting trials on one side, zero on the other.”

Trial Friend census, October 2026

Every rare disease FDA decision left in 2026

Here is where the year's plot resolves. September held 7 decisions we track and every one of them ended in an approval, 5 of them ahead of schedule. Between October 4th and New Year's Eve, the FDA is scheduled to rule on 22 more rare disease applications.

September 17th: UX111, Ultragenyx (approved as Fayuvi)

Gene therapy for Sanfilippo syndrome type A. Approved 2 days ahead of its September 19th date as the first treatment in the history of the disease, on a full approval rather than the accelerated one expected. Our Fayuvi coverage has the label details.

September 19th: Aqneursa, IntraBio (approved September 18th)

Label expansion into ataxia-telangiectasia; already approved for Niemann-Pick disease type C. Approved 1 day early as the first treatment for A-T anywhere in the world. Our Aqneursa coverage has the trial results and how to get it.

September 26th: Zilurgisertib, Mirum / Incyte (approved September 25th as Atebrioz)

Oral ALK2 inhibitor for FOP. Approved a day early, weeks after Pasatru's August approval, making it FOP's second approval of 2026 and its third overall.

September 25th: Lirafugratinib, Elevar (approved September 23rd as Lyrfigtu)

FGFR2 inhibitor for FGFR2 fusion-driven cholangiocarcinoma. Approved 2 days early as the first FGFR drug built to block FGFR2 alone. Our Lyrfigtu coverage has the trial results and label warnings.

September 28th: Emcitate, Egetis (approved September 28th)

Thyroid hormone analog for MCT8 deficiency, a thyroid hormone transport disorder. Approved on its action date as the first treatment for the disease.

September 30th: Apitegromab, Scholar Rock (approved September 11th as Isembyld)

Muscle-targeted add-on therapy for spinal muscular atrophy. Approved 19 days ahead of its date. Our Isembyld coverage has the label details.

September 30th: Camzyos expansion, Bristol Myers Squibb (approved September 30th)

Extends the hypertrophic cardiomyopathy drug to adolescents ages 12 to 17, the first HCM drug approved for anyone under 18. Approved on its action date. Our Camzyos coverage has the SCOUT-HCM results.

The fourth quarter is just as loaded. October opens with Genentech's Enspryng for thyroid eye disease on the 15th, a decision we broke down in our Enspryng TED post, then INOVIO's DNA immunotherapy for recurrent respiratory papillomatosis on the 30th. November holds Agios's mitapivat for sickle cell disease on the 1st, Beren's adrabetadex for Niemann-Pick disease type C on the 17th, Savara's Molbreevi for autoimmune pulmonary alveolar proteinosis and Capricor's deramiocel for Duchenne on the 22nd, Sanofi's venglustat for Gaucher disease type 3 on the 25th, then BridgeBio's BBP-418 for limb-girdle muscular dystrophy 2I and Zydus's saroglitazar for primary biliary cholangitis on the 27th, and Vertex's povetacicept for IgA nephropathy plus Cogent's bezuclastinib for GIST on the 30th. Genentech's Gazyva for primary membranous nephropathy and Regeneron's cemdisiran for myasthenia gravis are also due in November without a public date. December closes with Vanda's imsidolimab for generalized pustular psoriasis (12th), Oclaiz's return and Deciphera's tirabrutinib for primary CNS lymphoma (both the 18th), Hansa's imlifidase for kidney transplant desensitization (19th), the Gilead/Arcellx CAR T anito-cel for multiple myeloma (23rd), Praxis's relutrigine for SCN2A and SCN8A epilepsies (27th), and a December 30th doubleheader: Cogent's bezuclastinib for systemic mastocytosis and Saol's SL1009 for pyruvate dehydrogenase complex deficiency. AstraZeneca's Ultomiris expansion for IgA nephropathy is also due sometime in Q4. One application fell off the dated list: Novo Nordisk's denecimig for hemophilia A was expected in the third quarter, and on October 2nd Novo said the FDA extended its review over manufacturing-facility remediation, with no clinical deficiencies and no new date. Every one of these lives on our FDA calendar with primary sources, and it updates the day each decision lands.

What the numbers say about where rare disease research is heading

Put the whole year side by side and you get 2 true stories. In one, 2026 is a landmark: 19 approvals and expansions through October 4th, at least 6 of them history's first treatment for their disease, 45 new studies opening weekly, and a December docket so crowded the FDA is ruling on 2 rare disease drugs on December 30th alone. FOP, a disease with 1 treatment when the year began, now has 3. In the other story, a Goodpasture family is still searching and finding nothing, along with families across 21 other conditions that have 2 trials or fewer. Both stories are in the same dataset. Progress in rare disease is real, and it is nowhere near evenly distributed. We will re-run this census when the year closes and publish what changed.

What changed in rare disease trials between September 9th and October 4th

We re-ran the census 25 days after the first pull, and 2 things moved at once. Trial Friend added 9 disease pages in that window, among them thyroid eye disease, systemic sclerosis, immune thrombocytopenia, and ataxia-telangiectasia, so part of the growth is wider coverage rather than new research. Even so, the picture held. Active trials rose from 9,501 to 9,910 and recruiting trials from 5,547 to 5,754. The sponsor split barely budged, with academic and nonprofit sponsors leading 68% both times. The median recruiting trial still wants 80 people, 73% of recruiting trials are still open label, and the same 6 diseases still have zero recruiting studies. On the regulatory side the change was dramatic: the approval count went from 7 to 19, every September decision we tracked was an approval, and the only setback was a review extension for denecimig with no clinical concerns attached.

Methodology, for anyone who wants to check our math

We queried the ClinicalTrials.gov v2 API on September 9th, 2026, and again on October 4th, 2026, once for each disease Trial Friend covers (195 in September, 204 in October), using the same condition search our live disease pages use. The figures in this post are from the October run unless the text says otherwise. Active means a status of recruiting, not yet recruiting, or active but not recruiting. Trials appearing under multiple diseases were counted once in the totals (9,910 active and 5,754 recruiting are unique studies), while per-disease counts keep the overlap since that reflects what a patient searching each condition actually sees. Sponsor class, phase, enrollment targets, and masking come from each study's registration record. Approval and decision dates come from FDA announcements and company press releases tracked on our calendar. Two honest caveats: ClinicalTrials.gov condition matching occasionally includes studies that mention a disease without targeting it, and these numbers shift daily as trials open and close. Treat them as a snapshot, precise on the day we took it.