·Week of June 29 - July 3, 2026

Clinical Trials Brief | July 6, 2026

This week: WHO opening enrolment in PARTNERS, the first randomised treatment trial for Bundibugyo Ebola, FDA extending Casgevy to children aged 2 and over partly by extrapolation, the European Commission approving intrathecal onasemnogene abeparvovec for older SMA patients, FDA adding a class-wide hypoglycaemia warning to JAK inhibitors for patients with diabetes, and Takeda signing an AI discovery collaboration with Insilico.

WHO's PARTNERS platform trial opens enrolment in the DRC, the first randomised test of treatments for Bundibugyo Ebola

What changed

On July 2, 2026, enrolment opened in the Democratic Republic of the Congo for PARTNERS (Platform Adaptive Randomised Trial for New and Repurposed Filovirus TreatmentS). The trial is sponsored by WHO and coordinated by the Institut National de Recherche Biomédicale (INRB), the Institute of Tropical Medicine in Antwerp and the University of Oxford, with support from Africa CDC. It will test whether the monoclonal antibody MBP134 and the antiviral remdesivir, alone or in combination, improve survival in Bundibugyo virus disease, with participants followed for at least 28 days. Reuters, via SABC, reported that the trial could enrol more than 1,000 patients. As of July 2, Oxford reported more than 1,400 diagnosed cases, nearly 210 recoveries and more than 440 deaths in the DRC. WHO declared the outbreak a public health emergency of international concern on May 17, 2026.

Why it matters

There is no approved vaccine or treatment for Bundibugyo virus. The products that work for Zaire ebolavirus have not been shown to work across Ebola species. That makes this outbreak a live test of whether the research community has learned from 2014-2016, when trials started late and many ended underpowered as case numbers fell. PARTNERS uses a platform-adaptive design, which allows arms to be added or dropped without restarting the trial, and it was set up with national authorities from the outset. The operating conditions are difficult: Reuters reported an attack on an Ebola treatment centre in Ituri the same week.

Takeaway

Outbreak trials succeed or fail on speed of activation, so the main metric to watch is how quickly PARTNERS enrols relative to case counts. For sponsors and CROs, it is a working example of the emergency-trial model now being written into EU and global guidance: pre-agreed master protocols, local co-sponsorship, and adaptive arms.

Sources

This item is grounded in the University of Oxford and WHO announcements of July 2, 2026 and Reuters reporting via SABC.

FDA extends Casgevy to children aged 2 and over, using extrapolation for the youngest group

What changed

On July 1, 2026, FDA approved a supplement extending Casgevy (exagamglogene autotemcel, Vertex) to patients aged 2 years and older with sickle cell disease with recurrent vaso-occlusive crises or with transfusion-dependent β-thalassaemia. The indication was previously limited to patients aged 12 and older. FDA described it as the first gene therapy approved for patients as young as 2 with sickle cell disease. According to Consultant360, safety and effectiveness in sickle cell disease were evaluated in a trial of 11 patients aged 5 to under 12. Contemporary Pediatrics reported that the 2-to-under-5 age group was covered by extrapolation from product characteristics and the 5-to-11 data rather than direct enrolment. Treatment still requires full myeloablative conditioning before infusion.

Why it matters

This is a notable use of extrapolation for a one-time, CRISPR-edited cell therapy with myeloablative conditioning, where the benefit-risk balance in very young children depends heavily on the conditioning regimen and not only on the edited cells. The scientific rationale is to intervene before irreversible organ damage accumulates. The regulatory signal is that FDA is willing to extend a gene therapy label downward on a small paediatric dataset plus mechanistic reasoning, which is consistent with its broader 2026 push toward flexible evidence for rare diseases.

Takeaway

For paediatric gene therapy programmes, an extrapolation plan agreed early with FDA can save years of age-de-escalation studies. The trade-off is that long-term follow-up and registry data carry more weight after approval, so sponsors should plan postmarketing data collection in the youngest group as carefully as the pivotal study.

The EU approves intrathecal onasemnogene abeparvovec for SMA patients aged 2 and over, including adults

What changed

On July 2, 2026, Novartis announced that the European Commission had approved Itvisma (onasemnogene abeparvovec), an intrathecal formulation of the active substance in Zolgensma, for children aged 2 and older, teenagers and adults with 5q spinal muscular atrophy and a bi-allelic SMN1 mutation. It is given as a single, fixed-dose injection into the spinal canal, without adjustment for age or weight. The approval rests on the registrational Phase 3 STEER study, with supportive data from the Phase 3b STRENGTH and Phase 1/2 STRONG studies. In STEER, Itvisma produced a statistically significant 2.39-point improvement on the Hammersmith Functional Motor Scale Expanded, sustained over 52 weeks. STRENGTH enrolled patients previously treated with other SMA therapies. FDA approved Itvisma in November 2025.

Why it matters

Intravenous Zolgensma is dosed by body weight, which has limited its use to young children. A fixed intrathecal dose opens gene replacement to a much larger, older population. The effect size is modest on a motor scale in a population that, in STEER, could sit but had never walked independently (per pharmaphorum). That makes durability and real-world functional outcomes the real test. The approval also brings a one-time therapy into a population where many patients are already on chronic treatment, which raises practical questions about sequencing and whether to continue existing therapy.

Takeaway

Pricing and national reimbursement will determine actual access, and those negotiations will run country by country. For trial designers in neuromuscular disease, the evidence package (one registrational trial plus supportive studies, with previously treated patients included) is a useful reference point for what EU regulators accepted for a broad age range.

Sources

This item is grounded in the Novartis announcement of July 2, 2026 and reporting by Pharmaceutical Technology and pharmaphorum.

FDA adds a class-wide warning on hypoglycaemia in patients with diabetes to JAK inhibitors for non-haematological conditions

What changed

FDA has required updated labelling for JAK/STAT inhibitors indicated for non-haematological conditions to address the serious risk of hypoglycaemia, including severe hypoglycaemia, in patients with diabetes after starting treatment. The process began with FDA safety labelling change letters dated March 26, 2026. The approval letter for the Xeljanz (tofacitinib) supplement, electronically signed on June 30, 2026, implements this change. The revised Xeljanz label adds a new warning, "Hypoglycemia in Patients with Diabetes" (section 5.8), advising clinicians to consider increased blood glucose monitoring and to tell patients with diabetes to report symptoms of hypoglycaemia. According to MedShadow's summary, the change does not apply to JAK inhibitors used for blood disorders, such as ruxolitinib or fedratinib.

Why it matters

Hypoglycaemia is not part of the established JAK inhibitor risk profile, which has centred on cardiovascular events, thrombosis, malignancy and infection. A class-level hypoglycaemia warning means the mechanism is considered relevant across products, and that patients on insulin or sulfonylureas may need dose review when a JAK inhibitor is started. Because these drugs are used in rheumatology, dermatology and gastroenterology, where diabetes is a common comorbidity, the practical exposure is substantial.

Takeaway

In trials of JAK inhibitors in immune-mediated disease, protocols should add glucose monitoring for participants with diabetes and clarify how hypoglycaemia events are captured and graded. It is also a reminder that class labelling changes can arrive mid-trial and require investigator brochure and informed consent updates.

Sources

This item is grounded in the FDA approval letter for the Xeljanz labelling supplement, the revised Xeljanz label, and MedShadow's summary of the class change.

Takeda signs an AI discovery collaboration with Insilico as part of a shift to an AI-native discovery model

What changed

On July 2, 2026, Insilico Medicine announced a strategic collaboration with Takeda to use its Pharma.AI platform to advance drug candidates across Takeda's therapeutic areas. Under the agreement, Insilico leads AI-driven discovery to identify molecules that meet predefined scientific and early-development criteria, and Takeda applies its global development capabilities to take selected candidates into clinical validation. Takeda framed the partnership as supporting its transition to an AI-native discovery model that integrates automation, robotics and generative AI. Financial terms were not disclosed in the announcement.

Why it matters

The deal fits the pattern this brief has tracked all year, with pharma-AI partnerships moving from single-molecule bets toward platform relationships. The division of labour is the notable part. The AI company takes responsibility up to defined candidate criteria, and the pharma partner owns clinical development, which is where AI-discovered assets still have to prove themselves. Insilico also announced a neuroimmune alliance with SK Biopharmaceuticals at BIO 2026, which it valued at up to $2.5 billion, so its partnering pace is accelerating.

Takeaway

The handover criteria in deals like this, meaning what a molecule must show before the partner takes it into the clinic, are becoming the real contractual currency. Clinical development teams receiving AI-originated candidates should expect the same IND evidence bar as any other molecule; the discovery method does not change what regulators require.

Clinical Trials Brief is an educational commentary archive, not a real-time news service. Confirmed facts are based on publicly cited sources where possible. Analytical comments, comparisons, and forward-looking views are personal interpretation unless otherwise stated.