Alkermes is betting its sleep purchase and orexin drugs can outrun Vivitrol generics

The technology

Alkermes sells four proprietary medicines in neuroscience: Vivitrol for alcohol and opioid dependence, Aristada and Lybalvi for schizophrenia and bipolar I disorder, and LUMRYZ for narcolepsy. It also earns manufacturing and royalty revenue on Vumerity and on long-acting Invega products. These are the cash engines.

The growth story is the orexin 2 receptor, a target for wakefulness. Alixorexton is the lead drug, now in Phase 3 Brilliance studies in narcolepsy types 1 and 2 and in a Phase 2 study in idiopathic hypersomnia. A second molecule, ALKS 7290, produced early ADHD data in September that the company itself calls exploratory.

The odds

Alkermes has two kinds of risk. The commercial portfolio has proven demand, so the questions there are about generics and growth. The pipeline carries ordinary development risk.

Use the published base rates. Across indications, 57.8% of Phase III programs reach an NDA or BLA submission, and 52.4% of programs that enter Phase III are eventually approved. Only 28.9% of Phase II programs move to Phase III. For ALKS 7290, a drug that has completed a Phase 1b, the all-indication chance of approval from Phase I is 7.9%. The Phase 2 ADHD study with about 312 patients is the real test.

LUMRYZ in idiopathic hypersomnia is a safer bet on paper. Its Phase 3 study met the primary endpoint, and 90.6% of NDA and BLA submissions are approved, though a supplemental filing differs from an original one.

The money

At June 30, Alkermes held $691.6 million in cash and investments, up from $538.2 million at March 31. That compares with $1.32 billion at year-end 2025 before the Avadel deal, which was paid for with cash and $1.525 billion of term loans due in 2031. The company signed Amendment No. 1 to its credit agreement on Aug. 12.

Revenue guidance for 2026 is $1,730 million to $1,840 million, and adjusted EBITDA guidance is $370 million to $410 million. GAAP is weaker: Alkermes now guides to a net loss of $95 million to $115 million because of acquisition accounting and a contingent-consideration charge. Second-quarter revenue of $496.0 million was up 27%.

The risks

Generic competition is the first. Alkermes licensed Teva to market generic Vivitrol from Jan. 15, 2027, and in July it ended its Amneal authorized-generic agreement. Vivitrol is guided to $460 million to $480 million of 2026 sales.

Second, LUMRYZ cannot be promoted for idiopathic hypersomnia before March 1, 2028. Third, the orexin class is young, and a safety problem in one drug could hurt the others. Fourth, leadership is changing: Blair Jackson took over as CEO on Aug. 1, and the board added Anne Whitaker in September. Debt from the Avadel deal also limits flexibility.

Our read

Alkermes has positive adjusted EBITDA and four proprietary products that together brought in $411.7 million in the second quarter. It is now trying to turn a sleep acquisition and an orexin pipeline into a second act. So far the company has done what it said it would: it closed Avadel, started Phase 3, and delivered ADHD data when promised.

The next tests are not close together. Vibrance-3 data come toward year-end, the LUMRYZ filing for idiopathic hypersomnia is due by the end of 2026, and a generic Vivitrol is possible from January. The longer-term picture depends on Phase 3 narcolepsy results that have not been reported. This is analysis, not advice.