OS Therapies, a New York and Rockville-based clinical stage oncology company, announced on 17 September 2026 that it had accepted a request from the UK Medicines and Healthcare products Regulatory Agency relating to Project Orbis for OST-HER2, its candidate for the prevention or delay of recurrence in fully resected, pulmonary metastatic osteosarcoma. Two days earlier, on 15 September 2026, the company announced what it described as full alignment with MHRA and the European Medicines Agency on a pending Conditional Marketing Authorisation Application. Both statements are company announcements, not regulatory documents, and OST-HER2 is not authorised for human use in the UK, the EU, the US or Australia. What makes the sequence worth industry attention is not the candidate but the procedural pattern behind it: an orphan indication, a historical control comparison accepted at scientific advice stage, an adaptive confirmatory trial agreed across three agencies, and a regulator offering to carry the file internationally.
What the Project Orbis acceptance actually commits
MHRA asked OS Therapies to bring OST-HER2 into Project Orbis, the FDA-founded initiative under which several oncology regulators review a single submission in parallel rather than one after another. The company said on 17 September 2026 that it accepted. Orbis coordinates timing and shares assessment insight between agencies. It does not create an approval in any territory.
The company's own wording on the mechanics is not consistent. The 17 September announcement headline describes a Project Orbis invitation from MHRA, while a summary bullet in the same release describes MHRA becoming the company's sponsor into Project Orbis, and the body text describes MHRA becoming the company's representative to Project Orbis. In chief executive Paul Romness's quoted framing, MHRA "has stepped up and volunteered to take the international lead with Project Orbis". For anyone modelling review timelines, the distinction between an inviting regulator, a sponsoring regulator and a lead reviewing regulator is not cosmetic, and the announcements leave it open.
The same release states that FDA's CBER, its biologics evaluation and research arm, has recently re-opened Project Orbis, which the company characterises as applying to foreign-sponsored US companies only. That is reported by OS Therapies rather than confirmed in an agency notice, and it is the sort of scope condition that determines whether a biologics file is eligible at all. Anyone building a filing strategy on it should read the agency position, not the press release.
What MHRA agreed on historical controls
The 15 September announcement is precise about the concession that matters. MHRA, the company said, "agreed that the use of comparable historical control data - derived from a systematic evaluation of all suitable available peer-reviewed literature by independent biostatistics advisors presented in its most recent Scientific Advice Meeting - was appropriate in the orphan indication Prevention or Delay of Recurrence in Fully Resected, Pulmonary Metastatic Osteosarcoma." The company also said it reached full alignment on the design of its confirmatory Phase 3 trial, including the proportion of patients to be dosed with remaining Phase 2 drug product rather than forthcoming Phase 3 material, and that the Phase 3 must have commenced before a conditional marketing authorisation can be granted.
OS Therapies states that this advice aligns with MHRA's recently published draft Rare Disease Regulatory Framework, published for consultation by the UK government. Accepting a literature-derived external control arm in a resected metastatic paediatric and young adult sarcoma population is a methodological choice with consequences: it removes the need to randomise patients to an arm the sponsor considers unchanged, and it transfers the analytical risk onto the comparability of the historical cohorts. Rare disease developers, from platform companies to established orphan specialists such as AOP Orphan, will read the precedent closely, because the same argument is available in any indication where recruitment is the binding constraint.
Context: Romness, in the 15 September announcement, framed the agencies' position as "the recognition from international regulators that the standard of care in Osteosarcoma has not meaningfully changed in the last forty years". That is his characterisation of regulatory reasoning, not a published agency finding.
Parallel review measured against sequential filing
Set the described route against the conventional one. The conventional orphan oncology path files with FDA, then follows with an EMA centralised procedure and a separate UK submission, absorbing three assessment cycles, three sets of questions and three clock stops in series. The route OS Therapies describes runs a UK and EU conditional authorisation submission and a US Biologics License Application on overlapping timelines, with Project Orbis as the coordinating layer and a single adaptive Phase 3 design that FDA, MHRA and EMA have each reviewed.
The company says FDA aligned on the adaptive design previously agreed with MHRA and EMA, and that the Phase 3 is expected to commence in the fourth quarter of 2026. It also says EMA informed it that a previously requested Scientific Advice Working Party meeting was no longer required before the EU submission. Where that holds, the saving is in sequencing and in avoided redesign, not in evidentiary standard. A harmonised design means one protocol survives three reviews; it does not mean one reviewer's satisfaction binds another. Commercially, the difference shows up in launch planning and in how long a single manufacturing campaign has to serve trial supply and early access at once, a question that lands on partners across development and supply, from consultancies to contract manufacturers.
What the announcements claim and what they establish
The company's description of its Phase 2b result is that it demonstrated clinically significant benefit in the 12-month event free survival primary endpoint and in the overall survival secondary endpoint. Tested against what has been disclosed, that statement carries less than it appears to. Neither announcement reports the absolute event free survival difference, the relative difference, the sample size, the trial duration or the dropout rate, and no absolute figure was published in either release. Overall survival is named in these announcements as a secondary endpoint, so the survival claim is not a primary endpoint result.
Event free survival at 12 months is a composite intermediate measure, and whether it predicts survival in resected metastatic osteosarcoma is exactly the question a confirmatory trial exists to answer. The company says it will submit a clinical section including 3-year overall survival data before an invited Type B pre-BLA meeting with FDA in the fourth quarter of 2026, completing a BLA filing initiated in January 2026, and that it intends to request rolling review following a mid-September 2026 FDA Type C statistical methods meeting. Until those data are published and assessed, the efficacy position rests on a company characterisation of an uncontrolled comparison against literature controls.
On mechanism, OST-HER2 is described by OS Therapies as a gene-edited, Listeria-based immunotherapy that uses the immune-stimulatory effect of Listeria bacteria to provoke a response against the HER2 protein, targeting two mutated extracellular epitopes and one mutated intracellular epitope, with only one epitope needed in a tumour or micro-metastasis to trigger a response. It holds Orphan Drug, Fast Track and Rare Paediatric Disease designations from FDA, and Orphan Drug, Fast Track and Advanced Therapy Medicinal Product status from EMA and MHRA. Designations describe regulatory handling. None of them is evidence of clinical benefit, and no comparison with existing osteosarcoma treatment has been established.
The case for reading this cautiously
The strongest argument against treating the Orbis acceptance as validation is that every load-bearing fact in the sequence comes from the sponsor. Alignment at a scientific advice meeting records what an agency considers an acceptable approach to generating evidence, not a view on whether the evidence will pass. MHRA's advice on historical controls was given before the application was filed; the company says the complete Conditional Marketing Authorisation Application was still expected in the coming weeks as of 15 September 2026.
There is also a financing dimension that shapes incentives. OS Therapies notes that a granted BLA would make it eligible for a Priority Review Voucher it intends to sell, that the most recent such sale occurred in August 2026 for 220 million dollars, and that no comparable value can be assured. The company's own risk disclosure names cash runway, the timing of VAT refunds and research and development tax credits, and its ability to obtain further financing. OST-HER2 was previously conditionally approved by the US Department of Agriculture for canine osteosarcoma, which is a veterinary authorisation and carries no weight for human licensing. Readers tracking this through outlets such as BioPharma Dive or pharmaphorum should expect the substantive test to arrive with the assessment reports, not with the filings.
Where the regulatory timeline stands now
OS Therapies expects to complete its UK and EU conditional authorisation submissions within weeks of 15 September 2026, to begin the confirmatory Phase 3 trial in the fourth quarter of 2026 with site outreach already under way, and to hold an invited Type B pre-BLA meeting with FDA in the same quarter after submitting a clinical section including 3-year overall survival data.
What remains unknown is the size of the effect. No absolute event free survival difference, sample size or dropout rate has been published in these announcements, and no regulatory assessment report exists yet in any territory. The documents that would resolve it are the Phase 3 registry entry with its pre-registered endpoints, MHRA's and EMA's published assessments if authorisations are granted, and FDA's review documents. The registry entry is the near-term one, and it will show whether the adaptive design agreed across the three agencies matches what is eventually analysed. For developers weighing an external control strategy of their own, specialist advisers such as Alacrita Life Science Consulting are the practical starting point. Prescribing questions belong with a clinician or pharmacist.
