Bristol Myers Squibb (NYSE: BMY; BMS) delivered some good news to investors in reporting fourth-quarter and full-year 2025 results on Thursday. The pharma giant beat analyst consensus forecasts by reporting higher than expected numbers for total revenue and its “growth” portfolio of drugs on which BMS is counting on for long-term growth to offset revenue declines from its aging blockbusters set to lose patent exclusivity over the next two years.
As a result, BMS shares climbed 3% on Thursday from $57.62 to $59.52 despite a day of declines for key stock market indexes, after an initial 5% surge to $60.56 at the start of trading. Shares rose another 4% to $61.94 on Friday, finishing the week with a nearly 13% gain.
BMS shares have surged 38% over the past six months, on the strength of rising results, positive clinical data, and five business development deals completed last year.
BMS finished the fourth quarter with net income attributable to the company of $1.1 billion or $0.53 earnings per share (EPS) on a GAAP basis, up from $72 million or $0.04 EPS in Q4 2024. Non-GAAP net income stood at $2.6 billion or $1.26 EPS, 4.6% above analysts’ consensus, though down from $3.4 billion, or $1.67 per share in the year-ago quarter.
Q4 revenues increased 1% to $12.5 billion, 4.8% above analyst consensus. More than half of BMS’s revenue came from its growth portfolio (18 products plus royalty revenues), which increased year over year by 16% to $7.4 billion in the fourth quarter and 17% to $26.409 billion in 2025.
The growth portfolio is led by cancer immunotherapy blockbuster Opdivo® (nivolumab), whose revenue rose 9% to $2.693 billion in the fourth quarter and grew 8% to $10.049 billion, making it BMS’s second biggest-selling drug.
“Ongoing strength”
“We believe our results further demonstrate the ongoing strength in our growth portfolio as we advance our multi-year plan to rewire BMS for long-term growth. These efforts enabled us to enter 2026 with good momentum,” BMS board chair and CEO Christopher Boerner, PhD, told analysts on the company’s quarterly earnings call.
At least one analyst agreed with that upbeat assessment.
“Importantly, the quality of the 4Q revenue upside was encouraging, because the growth portfolio was 6% above [analyst] consensus,” David Risinger, a senior managing director and senior research analyst covering diversified biopharmaceuticals with Leerink Partners, wrote in a research note.
By comparison, Risinger added, BMS’s “legacy” portfolio of older drugs approaching the “patent cliff” was below 2% analyst consensus, finishing Q4 at $5.109 billion, down 15% from Q4 2024. The legacy portfolio is led by the factor Xa-inhibiting blood thinner Eliquis® (apixaban), which saw its product revenue rise 8% during the quarter, to $3.453 billion.
But Eliquis had been expected to rack up $3.71 billion, so the drug “meaningfully underperformed consensus,” Matt Phipps, PhD, partner and group head of biotechnology equity research with William Blair, wrote in a research note.
Eliquis’ full-year revenue also rose 8%, to $14.443 billion, making the drug BMS’s top revenue performer—for now.
BMS is projecting 2026 worldwide revenue growth for Eliquis ranging from 10–15%, or between $15.9 billion and $16.6 billion in sales. That compares with a 14% decline being forecast by the analyst consensus after BMS and Pfizer agreed last year to sell Eliquis for $346 a month, 43% below the $606 list price following talks with President Donald Trump’s administration. The lower price is expected to help generate sales for Eliquis while avoiding penalties under the Inflation Reduction Act (IRA), which covers the drug and took effect in January.
“We’ll have to see.”
“We’ll have to see how Eliquis performs amongst the IRA dynamics,” Jefferies equity analyst Akash Tewari wrote.
Steve Scala, managing director, health care and major pharmaceuticals research analyst with TD Cowen, asked Boerner and Adam Lenkowsky, chief commercialization officer, why BMS expects Eliquis revenues to fall between $1.5 billion and $2 billion in 2027: “I don’t think this product [Eliquis] has grown double-digit in several years, but then will in 2026? And it’s still not absolutely clear why there will be a step down in 2027.”
“We expect in November of this year that we will lose exclusivity in Europe,” Lenkowsky replied. “And we would expect a rapid and steep decline like we have seen with other small molecules outside of the U.S.”
Eliquis’ decline is likely to get steeper in 2028, when Eliquis will lose key U.S. patent protection, according to GEN’s recent A-List of Top 20 Drugs Heading for the Patent Cliff. That list also included BMS’s second highest grossing legacy drug Revlimid® (lenalidomide), which lost full U.S. exclusivity on January 31. Revenues for Revlimid—which BMS inherited by acquiring Celgene for $74 billion in 2019—plummeted 55% year-over-year during Q4 to $602 million and finished 2025 down 49% to $2.951 billion.
For all of 2025, BMS reported net income attributable to the company of $7.1 billion or EPS of $3.46 on a GAAP basis, improved from a net loss of $8.9 billion, or $(4.41) per share a year earlier. On a no-GAAP basis, net income attributable to BMS stood at $12.5 billion or $6.15 last year, up from $2.3 billion, or $1.15 per share in 2024.
BMS also offered investor guidance for 2026 of $6.05 to $6.35 in non-GAAP EPS, on revenue expected to range from ~$46.0 billion to $47.5 billion, 4–7.5% above the analyst consensus of $44.2 billion.
Potential new therapy readouts
During the earnings call, Boerner also said BMS expects to report topline clinical data for six potential new therapies in 2026:
- Admilparant (BMS-986278), an oral second-generation lysophosphatidic acid receptor 1 (LPA1) antagonist being evaluated for efficacy, safety, and tolerability in idiopathic pulmonary fibrosis in the Phase III ALOFT-IPF trial (NCT06003426).
- Arlocabtagene autoleucel (“arlo-cel”; BMS-986393), a GPRC5D-directed chimeric antigen receptor (CAR) T-cell therapy candidate being assessed as a fourth- and subsequent-line treatment for adults with RRMM in the Phase II registrational QUINTESSENTIAL trial (NCT06297226).
- Iberdomide, which is expected to report data in progression-free survival from the Phase III EXCALIBER-RRMM trial (NCT04975997). The trial is assessing the first-in-class cereblon E3 ligase modulator (CELMoD™) in combination with dexamethasone and daratumumab (IberDd) vs. daratumumab, bortezomib, and dexamethasone (DVd) in relapsed or refractory multiple myeloma (RRMM). In September, BMS announced positive data for IberDd in the trial’s other endpoint, showing a statistically significant improvement in minimal residual disease (MRD) negativity rates.
- Mezigdomide (formerly CC-92480), an oral CELMoD being studied for RRMM in combination with bortezomib and dexamethasone (MEZIVd) compared with the combo of pomalidomide, bortezomib, and dexamethasone (PVd) in the Phase III SUCCESSOR-1 trial (NCT05519085).
- Milvexian, an oral factor XIa (FXIa) inhibitor being studied vs. Eliquis for atrial fibrillation in the Phase III Librexia AF trial (NCT05757869) and vs. placebo in recurrent ischemic stroke in the Phase III Librexia STROKE trial (NCT05702034). The Librexia trials are being carried out by Johnson & Johnson in collaboration with BMS.
- RYZ101, a first-in-class targeted alpha-particle radiopharmaceutical developed by RayzeBio, which BMS acquired for approximately $4.1 billion in 2024. RYZ101 is being studied vs. standard of care as a treatment for second- and subsequent-line GEP-NETs in the ACTION-1 trial (NCT05477576).
“Together, these represent an attractive set of near-term catalysts that can meaningfully enhance the long-term growth profile of our current growth portfolio,” Boerner said.
BMS also anticipates what Boerner called “meaningful” pivotal line extension readouts for two marketed drugs pursuing additional indications. One is Sotyktu® (deucravacitinib), an oral, selective TYK2 inhibitor being studied for systemic lupus erythematosus in two Phase III trials, POETYK SLE-1 (NCT05617677) and POETYK SLE-2 (NCT05620407). The other is Cobenfy® (xanomeline and trospium chloride), a non-dopaminergic muscarinic agonist targeting the M1 and M4 receptors, which is approved for schizophrenia in adults and is now under study in Alzheimer’s disease psychosis in three Phase III trials: ADEPT-1 (NCT05511363), ADEPT-2 (NCT06126224), and ADEPT-4 (NCT06585787).
“This is a data-rich period for BMS, which could drive the introduction of more than 10 new medicines and over 30 meaningful launch opportunities by 2030,” Boerner projected. “The increasing pace of pivotal readouts later this year will serve to better define the potential of our pipeline candidates. We are confident in our ability to deliver an attractive and durable growth profile heading into the next decade.”
“In the long game” for AI
Robert Plenge, MD, PhD, BMS executive vice president, chief research officer, elaborated on BMS’s approach to AI in an interview with GEN during J.P. Morgan. Last year, Plenge and Greg Meyers, BMS’s executive vice president and chief digital & technology officer, articulated the company’s “predict first” approach to applying AI and machine learning to drug research, discovery, and development.
“We continue to be in the long game for AI. By that I mean, it is a transformational technology that’s going to take time,” Plenge told GEN. “There are a few specific examples where we’re seeing impact today in how we make molecules, how we propose new targets and therapeutic hypotheses, and how we analyze our clinical data. We’re doing all of that. But I think just the whole process of research and development is so complex that I think it’s going to take many years for all of these pieces to come together to really change the trajectory of R&D. So, we’re making substantial investments today.”
Just how substantial, BMS isn’t quantifying. But Plenge did shed light on where the company’s AI investment has focused: “Within research, it’s how to pick targets, how to make molecules, how to translate those findings into early clinical development in particular, which includes finding the right patient subsets. We’re continuing to invest in all of those areas. And over the last year, we continue to make progress in each of those areas.”
“We now have 100% of our small molecules that are AI-enabled, and more than 50% of our large molecule programs are AI-enabled. That allows us to reduce cycle times by at least about 10 to 20%,” Plenge said. “We believe that we can continue to reduce those cycle times, potentially even as much as 50% in the coming years.”
How much time does the 10–20% reduction cut from the drug discovery process?
“If you consider it to be about 2 1/2 years to go from an early-stage therapeutic hypothesis where you have chemical matter to the point where you have an IND that’s approved by regulatory authorities, it’s basically 10–20% off of that,” Plenge replied. “We hope that that’s several months in the near term, but we believe that will be multiple months and potentially even a year off of that timeline in the coming years.”
Leaders and laggards
- Ginkgo Bioworks Holdings (NYSE: DNA) shares soared 21% from $8.32 to $10.10 Friday after the provider of autonomous laboratories said it had demonstrated, in collaboration with OpenAI, an AI system that autonomously designed, executed, and learned from biological experiments with little human involvement. In a preprint posted to bioRxiv, Ginkgo reported that the AI system lowered cell-free protein synthesis reaction costs by 40% compared with the current state of the art, while running 36,000 experimental conditions across six iterative cycles. Ginkgo said the companies combined OpenAI’s GPT-5 reasoning model with Ginkgo’s cloud laboratory infrastructure, built from its reconfigurable automation carts (RAC) technology and Catalyst automation software, in order to design, execute, and analyze experiments in an iterative, closed-loop workflow. “By integrating LLMs [large language models] with programmatic control of a cloud lab, we demonstrate that an LLM-driven autonomous lab can successfully perform a real-world scientific task, highlighting the potential of AI-driven autonomous labs for scientific advancement,” the Ginkgo/OpenAI research team summarized.
- Novo Nordisk (NASDAQ Copenhagen: NOVO-B) shares skidded 25% over three days after it forecasted sales and net income declines for 2026, to levels below analyst expectations. Novo Nordisk expects to achieve adjusted sales growth of 5–13% below the 2025 level at constant exchange rates (CER), excluding $4.2 billion in U.S. sales it now expects to gain following the reversal of 340B Drug Pricing Program sales rebate provisions. Novo Nordisk finished 2025 with a net profit of DKK 102.434 billion ($16.2 billion), up 1% from DKK 100.988 billion (about $16 billion) in 2024, on net sales that rose 6% to DKK 309.064 billion ($48.9 billion) from DKK 290.403 billion ($45.9 billion) in 2024. Adjusted operating profit growth is also expected to lag 5% to 13% below 2025 at CER, the company said. Shares dipped 1.4% Tuesday from DKK 373.05 ($59.00) to DKK 367.80 ($58.17), then slid 17% Wednesday to DKK 304.65 ($48.18), and dropped another 8% Thursday to DKK 280.65 ($44.39), before rebounding 5% Friday to DKK 295.50 ($46.74)
- Roivant Sciences (NASDAQ: ROIV) shares jumped 22% from $21.14 to $25.82 Friday after the drug developer announced positive results for its small molecule TYK2 and JAK1 inhibitor candidate brepocitinib in cutaneous sarcoidosis (CS) in the Phase II BEACON trial (NCT06978725). Roivant said the 13 patients dosed with brepocitinib 45 mg achieved a 22.3-point improvement in mean Cutaneous Sarcoidosis Activity and Morphology Instrument–Activity (CSAMI-A) at Week 16 versus a 0.7-point improvement among the seven placebo patients. Brepocitinib also showed what Roivant termed rapid, deep, and sustained improvements across all other efficacy endpoints measured with a consistent safety profile. The 31-patient BEACON trial randomized patients 3:2:2 to once daily brepocitinib 45 mg, 15 mg, or placebo with a 16-week treatment period. In addition to CS, brepocitinib is also in development for dermatomyositis and non-infectious uveitis.
- Sangamo Therapeutics (NASDAQ: SGMO) shares nosedived 41% over three days after the genomic medicine developer announced the pricing of an underwritten $25 million offering that investors considered too dilutive to their existing shares. The offer consisted of 35,190,292 shares of its common stock and pre-funded warrants to purchase 17,787,033 shares of its common shares of its common stock, together with accompanying warrants to purchase 52,977,325 shares of its common stock. Shares tumbled 32% from 57 cents to 39 cents on Tuesday, fell another 8% to 36 cents on Wednesday, and dipped 3% to 35 cents on Thursday before bouncing back 13% on Friday to finish the week at 40 cents and a 29% decline. Sangamo said it intended to use the net proceeds from the offering for working capital and general corporate purposes.
