CYB704

Sandoz is running a Phase 3 head-to-head study of CYB704, its proposed biosimilar version of Roche's Ocrevus (ocrelizumab), in patients with relapsing multiple sclerosis [1]. The trial measures pharmacokinetic equivalence against the reference product rather than running a traditional efficacy readout, which is the standard regulatory path for a biosimilar. Ocrevus is one of Roche's largest franchises (multi-billion CHF annually per Roche financial disclosures), so even a modest biosimilar share after patent expiry would be a multi-billion-dollar market [2]. The timing is set by Roche's patent protection and lifecycle defense (particularly the subcutaneous Ocrevus Zunovo formulation), not by whether CYB704 works. The scientific case is already de-risked because ocrelizumab's clinical profile is close to a decade old. The commercial question, which involves share capture, pricing, and launch timing, is where this program lives or dies. Sandoz spun out of Novartis in October 2023 as a pure-play biosimilars and generics company and has flagged neurology as a strategic focus [3].

CYB704 is a biosimilar in Phase 3, not a new chemical entity. The Phase 3 study (NCT06847724) is listed as active, not recruiting, with 183 patients enrolled; the estimated primary completion date is November 2026 [1]. Biosimilars do not receive breakthrough, fast track, or orphan designations: those pathways are reserved for novel therapies. Instead CYB704 will travel the FDA's 351(k) abbreviated biosimilar pathway, which requires demonstrating the proposed product is 'highly similar' to the reference with no clinically meaningful differences in safety, purity, or potency. A separate FDA category, 'interchangeable biosimilar,' allows pharmacy-level substitution without physician sign-off and typically requires a switching study: this designation is commercially material because it affects uptake curves, and CYB704's current trial design does not include the dedicated switching arm usually needed to support it. Approval timing depends less on data readout than on resolution of Roche's patent position in the US and Europe. Composition-of-matter patents on ocrelizumab are generally understood to run into the late 2020s in the US, with method-of-use and formulation patents potentially extending protection further; the specific expiry dates vary by patent and jurisdiction and should be checked against the FDA Purple Book and the EPO register before any investment decision. Sandoz has not publicly committed to a US launch date, and the trial sponsor has not disclosed a target BLA submission quarter. Europe may open earlier, since the EMA biosimilar pathway is more mature and SPC (supplementary protection certificate) terms differ by jurisdiction. First-wave biosimilar competition in MS is widely expected in the late 2020s window.

CD20 is a protein studded along the surface of B cells, one of the immune system's two main types of adaptive-immunity cells [4]. Ocrelizumab is a monoclonal antibody that latches onto CD20 and flags those B cells for destruction by the immune system's cleanup machinery, mainly through antibody-dependent cellular cytotoxicity (ADCC, where natural killer cells and other immune effectors recognize the antibody-tagged B cell via its Fc region and destroy it) and complement-mediated lysis (where blood proteins called complement assemble on the tagged cell and punch holes in its membrane). In multiple sclerosis, misbehaving B cells drive the inflammation that strips myelin, the fatty insulation on nerve fibers. Without that insulation, electrical signals fail to travel cleanly down nerves, producing the relapses and progressive disability that define the disease. Clearing out the B cells calms the inflammation and slows the damage. The mechanism is heavily validated. Rituximab, an older anti-CD20 antibody first approved for lymphoma in 1997, was used off-label for MS for years before ocrelizumab arrived. The registrational OPERA I and II trials showed ocrelizumab cut the annualized relapse rate (a count of how many MS flare-ups a patient experiences per year on average) by about 46 percent versus interferon beta-1a over 96 weeks [5]. The ORATORIO trial extended the mechanism into primary progressive MS, where ocrelizumab remains the only approved therapy [6]. Since CYB704 is designed to be structurally and functionally equivalent to ocrelizumab within biosimilar tolerances, the biological case is inherited. The critical comparability challenge for an anti-CD20 biosimilar is Fc region glycosylation: the sugar chains attached to the antibody's Fc region directly modulate ADCC potency, and small shifts in the glycosylation profile across cell line, bioreactor, and purification conditions can change effector function enough to break comparability. This is where biosimilar manufacturing variability actually matters for regulators.

NCT06847724 enrolls 183 adults with relapsing MS and randomizes them between CYB704 and reference Ocrevus [1]. The primary endpoint is area under the concentration-time curve (AUC), a pharmacokinetic measure of total drug exposure over time. Secondary endpoints cover safety, immunogenicity (whether patients develop antibodies against the drug itself, which can neutralize activity or trigger reactions), and clinical efficacy including MRI lesion counts and relapse rate. The design is standard for biosimilar comparability. The sample size is powered for PK equivalence, not for a powered efficacy comparison, which is appropriate given the 351(k) regulatory path. Enrollment is complete, with the status flipping to active, not recruiting earlier in 2025, and estimated primary completion in November 2026 [1]. The study is being run by Sandoz, which spun out of Novartis in October 2023 and is one of the largest standalone biosimilars companies by revenue [3]. Sandoz has prior biosimilar approvals for filgrastim, pegfilgrastim, etanercept, rituximab, and adalimumab, giving it one of the deeper track records in the category.

Our model estimates a 18% chance this drug is eventually approved. It starts from the historical base rate for Phase 3 drugs in this area (about 51%), then adjusts using ten facts about the trial and sponsor. What moves the number most: it is helped by more secondary endpoints than usual and the sponsor's strong record of getting drugs approved; it is held back by heavier-than-usual blinding and weak or limited earlier-phase results. The other facts land near average for this stage, so they leave the estimate roughly where the base rate put it.

Efficacy risk is low. Pharmacokinetic equivalence between a well-manufactured biosimilar and its reference is a known problem, and Sandoz has solved it before across multiple molecules. Immunogenicity is the usual source of surprises: if CYB704 provokes neutralizing antibodies at different rates than Ocrevus, the comparability case collapses. Safety risk tracks the parent molecule. Ocrelizumab's label carries warnings for infusion reactions, serious infections (including hepatitis B reactivation and progressive multifocal leukoencephalopathy, or PML, a rare but potentially fatal brain infection caused by reactivation of the JC virus in immunosuppressed patients), and a watched signal for breast cancer that regulators continue to monitor [8]. Execution risk is minimal, since enrollment is complete and Sandoz is running the trial. The real risks sit downstream. Roche is actively defending Ocrevus with Ocrevus Zunovo, a subcutaneous formulation approved in 2024 that lets patients self-inject at home in ten minutes instead of coming in for a two-hour IV infusion [9]. That is a formidable patient-stickiness tool, because once a neurologist and patient settle into a subcutaneous routine they are unlikely to switch back to IV for a biosimilar discount. US composition patents are generally expected to lapse in the late 2020s, with method-of-use and formulation patents potentially extending protection further; actual dates must be verified in the FDA Purple Book. Biosimilar price erosion for infused anti-CD20 drugs has been slower than for subcutaneous biologics: look at the Rituxan biosimilar launch, where discounts settled around 25 to 35 percent rather than the 70 to 85 percent seen in the Humira class.

Worth watching, but not for the trial itself. The PK data readout is procedural. What matters is Sandoz's launch timing versus Roche's patent position and the Ocrevus Zunovo defense. The signal to watch is the first ocrelizumab biosimilar settlement with Roche. Once one biosimilar sponsor cuts a deal and secures a locked-in launch date, the others typically follow within quarters. Public disclosures and biosimilar trackers indicate multiple sponsors have ocrelizumab biosimilar programs in clinical development, including Celltrion (CT-P53, which received a partial EMA Phase 3 IND clearance in August 2023), Biocad (Phase 3 commenced November 2025), and Samsung Bioepis (program added to pipeline, January 2026), with additional industry commentary on Amgen's biosimilar roadmap [10]. Sandoz is unlikely to be first across the line, but it has the strongest commercial infrastructure to scale once launched. Check back when either (a) Sandoz publicly files its 351(k) application with the FDA, which it has not committed to timing on, or (b) a competitor cuts a patent settlement that establishes the launch window. For Roche, Ocrevus is one of several franchises defining the second half of the decade, sitting alongside Polivy, Vabysmo, and Hemlibra as pillars holding up growth as Perjeta and Tecentriq run off patent [2]. For Sandoz, this is a test of whether a pure-play biosimilars company can win a neurology franchise against an entrenched biologic with a subcutaneous lifecycle extension already deployed.