Bristol-Myers Squibb Company earnings call
Full-year 2026 guidance raised on strong first-half momentum
BMS reported a strong Q2 2026 beat, raised guidance, but management's narrative centered on both the strength of its portfolio and the broader competitive dynamics in its key therapeutic areas. The call had notable cross-company signals around AI partnerships and the competitive landscape in IPF/PPF. BMS raised FY2026 revenue and EPS guidance, mainly driven by the outsized performance of Eliquis (up 21% in Q2) and the growth portfolio.
Buzzberg read Full-year 2026 guidance raised on strong first-half momentum BMS reported a strong Q2 2026 beat, raised guidance, but management's narrative centered on both the strength of its portfolio and the broader competitive dynamics in its key therapeutic areas. The call had notable cross-company signals around AI partnerships and the competitive landscape in IPF/PPF. BMS raised FY2026 revenue and EPS guidance, mainly driven by the outsized performance of Eliquis (up 21% in Q2) and the growth portfolio. Read full analysisCollapse analysis
BMS reported a strong Q2 2026 beat, raised guidance, but management's narrative centered on both the strength of its portfolio and the broader competitive dynamics in its key therapeutic areas. The call had notable cross-company signals around AI partnerships and the competitive landscape in IPF/PPF. BMS raised FY2026 revenue and EPS guidance, mainly driven by the outsized performance of Eliquis (up 21% in Q2) and the growth portfolio.
- Management confirmed partnership agreements with NVIDIA and Anthropic to embed AI into drug discovery, a significant enterprise-level partnership signal.
- The company provided key pipeline updates, with delays for cobenfi (ADEPT program) and milvexian (AFib trial), but framed these as positive signals (slower event accrual, DMC endorsements).
- In IPF/PPF, management highlighted their differentiation on tolerability vs. standard of care (like Boehringer's nintedanib) and the emerging second-line/add-on market dynamic.
What matters now
The highest-signal changes from the call.
Milvexian AFib readout slipped to Q1 2027
Management increasingly encouraged on Milvexian despite delay
Show 3 more callouts
Cobenfy ADEPT program data now expected early 2027
Company targets more than 10 new medicines by decade end
Eliquis loss of exclusivity expected to feed Milvexian launch
Actuals
| Metric | Reported | Change |
|---|---|---|
| Revenue | $12.973B | +13% QoQ |
| EPS | $2.04 | +29% QoQ |
| Gross margin | 71.28% | Reported |
| Operating margin | 31.5% | Reported |
| Free cash flow | $3.088B | Reported |
| Capex | $0.305B | Reported |
Forward guidance
| Metric | Period | Range | Midpoint | Status |
|---|---|---|---|---|
| EPS | FY2026 | $7.00–$7.30Above consensus | $7.15 | Raised |
| Revenue | FY2026 | $51.5B–$52.5BAbove consensus | $52B | Raised |
Management read
Upbeat
Management emphasized strong execution, raised full-year guidance, and repeatedly expressed confidence in the pipeline and long-term growth trajectory despite pushing some readouts into 2027.
Management AI read
Management said BMS has embedded AI across research and enterprise operations, with demand for compute growing alongside the maturity of its AI capabilities. They announced new partnerships with Anthropic and NVIDIA to help scientists understand disease biology, design and test candidate molecules faster, and make better-informed decisions earlier.
Companiesreturns since call
Partners
BMS is partnering with NVIDIA to embed AI across its research operations, likely driving demand for NVIDIA's compute platforms (DGX, etc.) in the life sciences vertical.
Evidence
“we've recently announced partnership agreements with both Anthropic and NVIDIA. Concretely, this will help our scientists do three things, understand disease biology more deeply, design and test candidate molecules faster...”
Supply chain
Milvexian's AFib trial (versus Eliquis) readout delayed to Q1 2027 due to slower-than-expected event accrual; DMC 'continues to endorse the study.' This is a classic case of ‘patience is a virtue’—the slower event rate partly reflects the comparator (Eliquis) working well, but also hints at potential favorable unblinded dynamics. — A delay here pushes back the competitive threat to Eliquis (and Pfizer) but also the upside optionality for BMS; the DMC endorsement is a positive signal on the primary endpoint (non-inferiority on strokes).
Evidence
“having DMC continues to assess the efficacy and safety data, as you say, and giving us in very much closer to the final readout, the reassurance that the study can progress is a good thing.”
Supply-chain alpha · 3returns since call
Milvexian's AFib trial (versus Eliquis) readout delayed to Q1 2027 due to slower-than-expected event accrual; DMC 'continues to endorse the study.' This is a classic case of ‘patience is a virtue’—the slower event rate partly reflects the comparator (Eliquis) working well, but also hints at potential favorable unblinded dynamics.
Eliquis demand is surging so much in 2026 that it’s actually masking the underlying transition of the BMS portfolio; management reiterated the $1.5-$2B step down in Eliquis revenue for 2027 as generic entry looms, meaning ‘headline growth’ in 2026 is flattered by Eliquis timing (WAC cut and CPI penalty elimination).
Evidence
“we still expect the $1.5 to $2 billion step down in 2027 that we referenced on the Q1 call.”
BMS's IPF/PPF program (amilpirant) is targeting a market where the recent launch of a competitor (likely BI's Giscade) is already validating the 'switch/add-on' market opportunity created by generic entry; BMS sees this as a tailwind for its eventual launch.
Evidence
“what we're seeing today with some of the most recently approved products has an overall rate of diarrhea of 40%, over 60% in combination. And what we've seen with admilpirant in our phase two studies is a side effect profile that is compar…”
Methodology & coverage
Management-only analysis. All 3 validated company mentions are shown. Reported actuals and forward guidance are kept separate. Public evidence is limited to eight short attributed quotes. AI-generated analysis can be incomplete or wrong; verify important claims against the original source.