Bristol Myers has myeloma covered, but its biggest replacement bets wait for 2027

The technology

Bristol Myers Squibb is a large-cap drugmaker in a hurry to replace its two biggest products. Eliquis sold $4.48 billion and Opdivo $2.49 billion in the second quarter of 2026, and both face patent expiries later this decade. The replacement plan has three main legs: CELMoD protein degraders in myeloma, the muscarinic drug Cobenfy in new psychiatric uses, and the factor XIa anticoagulant milvexian.

The CELMoD leg is working. Zenbexus (iberdomide) won accelerated approval on August 13 on MRD data, and on October 8 BMS said the same trial showed median progression-free survival of 42 months against 20 months. Mezigdomide has an FDA decision date of May 13, 2027. The other legs are less certain.

The odds

Zenbexus has cleared the hard part. The progression-free survival result gives the accelerated approval a direct path to confirmation, though the comparator was an older bortezomib regimen.

Milvexian has already failed once. LIBREXIA ACS showed no reduction in cardiovascular events against placebo, with equal serious bleeding. The atrial fibrillation trial, a noninferiority test against Eliquis, now reads out in the first quarter of 2027. Industry data put the odds of a phase 3 drug reaching a filing at 57.8%, per BIO, Informa and QLS for 2011 to 2020. We treat AF as close to a coin flip with upside if bleeding is clearly lower.

Cobenfy in Alzheimer's psychosis is the hardest call. ADEPT-2 was delayed after site irregularities, and all three ADEPT readouts now begin in early 2027. Cobenfy's launch in schizophrenia is also slow, at $63 million in second-quarter sales.

The money

The base business is stronger than expected. BMS raised 2026 revenue guidance to about $49.0 billion to $50.0 billion and non-GAAP earnings to $6.75 to $7.00 a share. The Growth Portfolio grew 15% to $7.56 billion in the quarter. Camzyos rose 60% to $416 million, Breyanzi 41% to $484 million and Reblozyl 29% to $735 million. Revlimid, the older myeloma drug, still sold $425 million, half its level a year earlier, and that decline will continue.

Debt is the constraint. At June 30, BMS had $11.46 billion in cash and marketable debt securities and $43.12 billion of debt, a net debt position of $31.66 billion. That favors licensing over big acquisitions, which matches the $600 million upfront Hengrui deal and the CEO's remark that BMS does not "feel any compulsion to chase deals".

The risks

The main risk is timing. Eliquis growth this year is a gift that will reverse when generics arrive, and the two largest pipeline bets that could offset it now land in 2027. A miss in LIBREXIA-AF would remove the most direct Eliquis successor. A miss in ADEPT would cap Cobenfy at a slow-growing schizophrenia brand.

Smaller risks: CELMoD safety (Zenbexus carries boxed warnings for clots and embryo-fetal harm), and an open IRS transfer pricing dispute covering 2008 to 2012.

Our read

BMS is a cash-generating company running a replacement race against a known clock. Myeloma is the bright spot and should carry growth in hematology into the 2030s. The swing factors are two 2027 readouts. Until LIBREXIA-AF and the first ADEPT data arrive, the story rests on how long Eliquis holds up and how quickly Zenbexus, Camzyos and Breyanzi scale. The Hengrui pact and the AI deals with Anthropic and NVIDIA are long-dated and will not move the numbers soon. The next checkpoint is third-quarter results on October 29.