Samsung Bioepis Co., Ltd. has presented detailed Phase 3 evidence showing that its proposed pembrolizumab biosimilar SB27 met the prespecified efficacy-equivalence criterion against Merck & Co.’s Keytruda in first-line metastatic non-squamous non-small cell lung cancer. The November 6 SITC presentation comes just days after Samsung Bioepis submitted SB27 for U.S. approval across 18 proposed cancer indications, moving another serious competitor toward a market currently dominated by one of the world’s highest-selling medicines.
The 555-patient study produced a Week 24 objective response rate of 34.6% with SB27 and 37.9% with reference pembrolizumab. The 90% confidence interval for the response-rate ratio was 0.737 to 1.071, remaining entirely within the predefined equivalence range of 0.712 to 1.405. Secondary efficacy, safety, pharmacokinetic and immunogenicity results were also broadly comparable, giving Samsung Bioepis a substantial clinical package to support its regulatory application.
Yet the commercial story is already more complicated than a simple race to copy Keytruda. Formycon AG and Zydus Lifesciences Limited announced in October that the FDA had accepted FYB206, their proposed pembrolizumab biosimilar, for review, while Samsung Bioepis has so far announced submission rather than FDA acceptance of its own application. At the same time, Merck is rapidly commercialising subcutaneous Keytruda Qlex and has explicitly warned investors that conventional Keytruda sales could face material biosimilar pressure from 2028 to 2029.
How closely did SB27 actually match Keytruda beyond the headline response rate?
The Phase 3 study randomized 555 adults across 14 countries, with 283 receiving SB27 and 272 receiving reference pembrolizumab. Both treatments were administered intravenously at 200 mg every three weeks alongside pemetrexed and carboplatin, with carboplatin given during the first four cycles. The primary endpoint was objective response rate at Week 24 using blinded independent central review under RECIST criteria.
The detailed SITC dataset suggests similarity extended beyond that primary endpoint. Median progression-free survival was 8.25 months with SB27 and 7.10 months with reference pembrolizumab, with a hazard ratio of 0.91 and a 95% confidence interval of 0.71 to 1.16. Median duration of response was 11.27 months versus 9.00 months, with a hazard ratio of 0.71 and a confidence interval of 0.46 to 1.09.
Median overall survival was 12.68 months with SB27 and 13.11 months with reference pembrolizumab. The overall-survival hazard ratio was 1.06 with a 95% confidence interval from 0.78 to 1.46, meaning the available analysis does not suggest a meaningful survival separation between the treatments but is not designed to establish that the two products have identical survival outcomes. Final survival follow-up remains important because the Phase 3 programme is still listed as active but not recruiting, with full study completion estimated for August 2027.
The interpretation is therefore appropriately narrow. Samsung Bioepis was not attempting to show SB27 is superior to Keytruda, nor does a biosimilar programme require that result. The purpose was to demonstrate that observed differences fall within scientifically justified comparability boundaries, supporting the broader totality-of-evidence case that SB27 is highly similar to reference pembrolizumab without clinically meaningful differences.
Does the safety dataset support Samsung Bioepis’ biosimilarity argument?
Grade 3 or higher treatment-emergent adverse events occurred in 55.4% of SB27-treated patients and 59.9% of patients receiving reference pembrolizumab in the reported interim analysis. Immune-related adverse events were reported in 27.9% and 31.3%, respectively, while infusion-related reactions occurred in 0.7% of each group.
Samsung Bioepis also reported comparable pharmacokinetic and immunogenicity profiles. Those observations matter because biosimilarity is not established through tumour response alone. Regulators evaluate the complete evidence package, including analytical characteristics, exposure, immunogenicity and safety, to determine whether residual differences between the proposed biosimilar and reference product are clinically meaningful.
SB27 also has a separate Phase 1 programme that compared pharmacokinetics with U.S.- and European-sourced Keytruda in patients with resected Stage II to IIIA non-small cell lung cancer. Samsung Bioepis reported that the interim Phase 1 analysis met predefined pharmacokinetic bioequivalence criteria and showed comparable efficacy and safety, adding another component to the regulatory package beyond the metastatic-lung-cancer Phase 3 study.
The available data therefore support Samsung Bioepis’ claim of comparability, but SB27 remains a proposed biosimilar. Biosimilarity has not yet been established by the FDA, and submission of an application does not guarantee acceptance for substantive review or eventual approval.
Why can Samsung seek 18 cancer indications after running Phase 3 in only one lung cancer setting?
This is one of the most important concepts behind the economics of biosimilar development. Samsung Bioepis’ U.S. application reportedly seeks approval across 18 pembrolizumab indications, even though its comparative Phase 3 efficacy study was conducted in metastatic non-squamous non-small cell lung cancer.
That is possible because biosimilar regulation does not generally require sponsors to repeat the reference medicine’s entire clinical-development programme indication by indication. Where scientifically justified, a regulator can extrapolate biosimilarity across additional approved indications using the totality of analytical, functional, pharmacokinetic, immunogenicity and clinical evidence together with understanding of the reference product’s mechanisms.
The economic implications are substantial. Keytruda has accumulated an unusually broad oncology label spanning multiple tumour types, biomarkers, treatment lines and combination regimens. Requiring a biosimilar developer to reproduce every pivotal Keytruda study would make development prohibitively expensive and undermine much of the cost advantage the biosimilar model is intended to create.
Samsung Bioepis must still convince regulators that extrapolation is justified for the indications it seeks. The application across 18 cancers should therefore be understood as the proposed regulatory scope, not 18 independently proven indications for SB27.
Why could Formycon and Zydus reach the market ahead despite skipping a comparative Phase 3 efficacy trial?
The most revealing part of the pembrolizumab biosimilar race may be the difference between development strategies.
Formycon and Zydus announced on October 13 that the FDA had accepted the BLA for FYB206, which the companies described as the first publicly announced accepted U.S. application for a proposed Keytruda biosimilar. They developed FYB206 without a comparative Phase 3 efficacy trial after agreeing a streamlined pathway with the FDA, relying heavily on analytical comparability and pharmacokinetic evidence.
That means Samsung Bioepis’ large Phase 3 trial may strengthen confidence in its evidence package without necessarily conferring a regulatory timing advantage. FYB206’s developers expect an FDA decision around the third quarter of 2027 based on the standard Biosimilar User Fee Amendments review timeline. Samsung Bioepis submitted its U.S. application on October 30, but no FDA acceptance announcement has yet established its corresponding review timeline.
The contrast also reflects the evolution of biosimilar regulatory science. As analytical methods become increasingly capable of characterising complex biologics, regulators have shown greater willingness in appropriate cases to reduce reliance on large comparative efficacy trials that may add limited information beyond robust analytical and pharmacokinetic evidence.
For developers, that can materially lower cost and shorten timelines. For Samsung Bioepis, it means the commercial value of its 555-patient Phase 3 programme lies more in regulatory robustness and market confidence than in proving that every future Keytruda biosimilar must conduct a similar trial.
Why does Keytruda’s enormous revenue make this a very different biosimilar opportunity?
Merck reported combined Keytruda and Keytruda Qlex sales of $8.60 billion in the third quarter of 2026, up 6% year over year. Combined sales reached approximately $25.0 billion during the first nine months of the year, while Keytruda generated about $31.7 billion globally in 2025.
Those numbers place pembrolizumab in a different commercial category from most biologics that have previously faced biosimilar competition. Even a relatively modest share of the reference product’s intravenous market could represent substantial revenue for successful biosimilar entrants, while relatively small percentage reductions in average treatment cost could translate into large savings for healthcare systems and oncology payers.
The opportunity will nevertheless be contested aggressively. Samsung Bioepis is not alone. Formycon and Zydus are already under FDA review, while Celltrion has filed in South Korea and Amgen has completed clinical development of its own proposed pembrolizumab biosimilar. Several additional developers have programmes at different stages globally.
This competitive density could ultimately matter as much as which company obtains the first approval. Biosimilar markets often become more price competitive as additional entrants arrive, and hospital formularies, oncology networks and payers may negotiate across multiple highly similar options rather than simply switching from an originator to a single substitute.
How is Merck using Keytruda Qlex before intravenous biosimilars arrive?
Merck has already changed the competitive landscape by launching Keytruda Qlex, a subcutaneous formulation combining pembrolizumab with berahyaluronidase alfa. The FDA approved Keytruda Qlex in September 2025 for appropriate adult and paediatric solid-tumour indications covered by intravenous pembrolizumab.
The product has gained commercial traction quickly. Keytruda Qlex generated $830 million in the third quarter of 2026 and $1.42 billion during the first nine months of the year, according to Merck’s latest results. Combined Keytruda and Keytruda Qlex revenue continued growing even as standalone intravenous Keytruda revenue came under increasing pressure from patients moving to the newer formulation.
That migration could become strategically important when intravenous pembrolizumab biosimilars enter the United States. A hospital comparing intravenous Keytruda with a lower-cost intravenous biosimilar will face a relatively familiar procurement decision. Keytruda Qlex changes the proposition by offering subcutaneous administration, potentially creating a convenience distinction that cannot be replicated simply by matching the intravenous reference product.
This should not be interpreted as protection from biosimilar competition. Merck itself expects Keytruda sales to be materially negatively affected by biosimilars between 2028 and 2029. The strategic value of Qlex is that it may shift part of the franchise into a differentiated formulation before that pressure becomes significant.
When could U.S. pembrolizumab biosimilar competition actually begin?
Regulatory approval and commercial launch are separate milestones. Merck currently identifies December 2028 as the expiration of Keytruda’s primary U.S. compound patent and has said biosimilar competition could begin around that point, even though additional composition-of-matter patents expire during 2029 and may be litigated.
That creates an unusual interval in which several biosimilar developers could potentially obtain FDA approval before they are commercially free to launch. A 2027 regulatory decision for FYB206, for example, would establish an approved product but not automatically determine its launch date because patent resolution and other market-entry considerations remain separate.
For Samsung Bioepis, the important near-term milestones are therefore regulatory rather than commercial. The company needs FDA acceptance of the submitted application, successful review of the analytical, manufacturing and clinical package and eventually an approval decision. Patent timing will then become increasingly important as 2028 approaches.
Manufacturing quality also deserves attention. Biosimilars are complex biologic medicines produced through living-cell systems, and approval requires regulators to be satisfied not merely with clinical similarity but with manufacturing controls capable of consistently reproducing the intended product at commercial scale.
Will oncologists care which company conducted the larger clinical programme once biosimilars reach hospitals?
Probably less than the current trial headlines suggest.
Once a biosimilar has been approved by the FDA, the regulatory conclusion is that it is highly similar to the reference product and has no clinically meaningful differences in safety, purity and potency under the authorised conditions. Clinicians should therefore not interpret a biosimilar with a larger pre-approval clinical trial as inherently more effective than a competitor approved through a more streamlined evidence pathway.
Hospital purchasing decisions may instead centre on contracting, supply reliability, reimbursement, formulary policy and the ability of manufacturers to maintain uninterrupted access. Oncology treatments are often administered through institutional systems rather than conventional retail pharmacy channels, making health-system procurement particularly important.
Samsung Bioepis has meaningful experience in this environment through an established portfolio of biosimilars across immunology, ophthalmology, haematology and supportive oncology. That infrastructure can help with launch execution, but the company will be entering a crowded market where other large biosimilar manufacturers also have substantial commercial capabilities.
The real competitive differentiator may therefore emerge after approval rather than in the Phase 3 dataset.
What does the SB27 result reveal about the coming Keytruda patent cliff?
The November SITC presentation eliminates one important uncertainty around Samsung Bioepis’ programme. SB27 produced a response rate inside the prespecified equivalence boundary, with similar progression-free survival, response durability, overall survival, severe adverse events, immune-related toxicity, pharmacokinetics and immunogenicity to reference pembrolizumab in the reported analysis.
The remaining uncertainties are increasingly regulatory and commercial. Samsung has filed for U.S. approval but has not yet publicly announced FDA acceptance. Formycon and Zydus are already under review. Merck is expanding Keytruda Qlex rapidly. Several other biosimilar developers are advancing behind them, while the reference product continues generating more than $8 billion in quarterly revenue.
That combination makes pembrolizumab one of the most consequential biosimilar transitions the pharmaceutical industry has faced. The clinical question is becoming easier to answer as multiple developers demonstrate comparability. The harder question is what happens when several approved versions of one of oncology’s most widely used antibodies compete simultaneously with a newer subcutaneous formulation controlled by the originator.
Samsung Bioepis has now shown that SB27 can satisfy the core comparative efficacy test in a large randomized Phase 3 study. Whether that scientific reassurance converts into meaningful U.S. market share will depend on events outside that trial: FDA timing, patent resolution, manufacturing execution, pricing, hospital contracting and how much of Merck’s franchise has already migrated to Keytruda Qlex by the time intravenous biosimilar competition truly begins.
