SUPERNUS PHARMACEUTICALS, INC. (SUPN) stock intrinsic value fair value estimate & owner earnings analysis

Supernus Pharmaceuticals sells branded central-nervous-system medicines in the United States. Revenue was $719.0 million in 2025. The largest product is Qelbree, a non-stimulant ADHD treatment the company developed itself: $304.7 million in 2025, up from $241.3 million, and $89.2 million in the second quarter of 2026, 15% higher than a year earlier. Next come GOCOVRI for Parkinson's dyskinesia ($146.8 million), the Parkinson's products APOKYN and ONAPGO (ONAPGO launched in April 2025 and sold $13.5 million in the second quarter of 2026), and the fading epilepsy brands Oxtellar XR and Trokendi XR, which now face generic competition. In July 2025 Supernus bought Sage Therapeutics, which brought a 50/50 US partnership with Biogen on ZURZUVAE, a postpartum-depression pill. That partnership contributed $53.0 million in the five months Supernus owned it in 2025 and $35.4 million in the second quarter of 2026 alone. The company has no debt, having repaid its $402.5 million of convertible notes in 2023, and held about $372 million of cash and short-term securities at June 30, 2026. How we measure earnings. We start from operating cash flow and remove working-capital swings, stock compensation and the small amount of capital spending needed to run the business. Over 2016-2025 this owner-earnings figure ranged from $0.64 to $2.92 per share. In the last five years it was $1.92, $2.30, $1.67, $2.61 and $1.66. The 2025 figure is held down by about $48 million of one-time cash costs of the Sage deal ($10.0 million of transaction costs and $38.3 million of payments to former Sage employees). Without them 2025 would have been about $2.50, only slightly below 2024's $2.61. The charges for amortizing acquired product rights (about $89 million in 2025) and the $54.9 million write-down of APOKYN in the second quarter of 2026 are non-cash and are added back. The fact that those rights wear out is instead built into each case as dated patent and generic-entry assumptions. Small milestone payments on past acquisitions that ran through operating cash flow were treated as part of the purchase price. We use the five-year median of $1.92 per share (about $112 million) as the starting level. Results so far in 2026 support it: first-half cash flow and the quarterly income statements suggest a current run-rate of roughly $1.60-2.15 per share. What drives value. This is a business of dated exclusivity cliffs, and each one is in the filings. Qelbree: eight or more generic makers have challenged its patents. The cases are consolidated in New Jersey, and a legal stay blocks FDA approval of their copies until October 2, 2028. Discovery closes November 17, 2027, and no trial date has been set. The listed patents expire September 4, 2029, February 7, 2033 and April 2, 2035. GOCOVRI: settlements already allow generics from June 4, 2029. ZURZUVAE: patents run from 2034 to 2037. XADAGO generics may launch in December 2027, and the Orenitram royalty ends in 2027. Supernus's own record after generic entry shows sales fall about 55-60% in the first year and then decline more slowly: Oxtellar XR went from $99.5 million in 2024 to $40.7 million in 2025. We apply that pattern to every future loss of exclusivity. Base case ($17.47 per share). Qelbree grows 10% and then 8% before slowing, and generics arrive in mid-2031. That is our judgement of the midpoint between generics arriving when the stay expires and the 2033 patents holding. GOCOVRI grows modestly and then erodes from mid-2029. ZURZUVAE grows 10% a year for two years, then more slowly, and faces generics from mid-2035. ONAPGO grows to about $95 million. The older brands and royalties shrink about 22% a year. Last year's owner earnings fell 36% as reported, or about 4% excluding the Sage deal costs. Before the research treatment described below, the first forecast year is about 8% above our starting level, for three reasons in the filings: the deal costs do not recur, ZURZUVAE will be included for a full year instead of five months, and Qelbree's recent 15% growth. Overhead shrinks with revenue once the cliffs hit. We treat about half of research spending as paying for itself. That rests on a record: internal programs produced Oxtellar XR, Trokendi XR and Qelbree, each of which became a major product (Oxtellar XR alone sold $113.4 million in 2023, and Qelbree is now above $300 million a year). The rest of research spending is treated as a pure cost. No value is given to the current pipeline (SPN-817 and SPN-820 are in mid-stage trials, and SPN-443 is due to begin its first human study in the second half of 2026). Earnings from today's products fade through the late 2030s; the cash on hand is counted through the interest it earns and its eventual return to owners. Discounted at 6%, the base case is worth $17.47 per share. Bear case ($5.08). Generic Qelbree launches when the stay expires in October 2028. ZURZUVAE grows only about 3% a year and loses exclusivity in 2034. ONAPGO stays small. The cost base does not shrink quickly enough, so owner earnings turn negative around 2029-2030 before management cuts. Half of today's cash is spent on acquisitions that do not earn their cost. That risk is on the record: Supernus wrote down acquired intangibles by $20.2 million in 2023, withdrew Osmolex ER in 2024, and wrote down APOKYN by $54.9 million in 2026. Bull case ($40.60). Qelbree's patents hold to 2035 and it keeps growing at 13% then 10%. ZURZUVAE grows 15% then 12% and keeps exclusivity to 2037. ONAPGO reaches about $130 million. The company repeats its Qelbree record once more with a new product that launches around 2031 and reaches about $250 million in sales. What could make this wrong. The Qelbree patent case matters more than anything else: the gap between the bear and base cases is mostly the difference between an October 2028 and a mid-2031 generic launch. The ZURZUVAE economics are reported as a single collaboration revenue line, so the profit margin on it is our estimate. The company has also spent roughly $900 million on three acquisitions since 2020 (USWM, Adamas and Sage) to replace fading products, and owner earnings per share over 2021-2025 averaged about $2.03, below the $2.51 of 2018-2020. If that pattern continues, cash reinvested in deals may not add value. Margin-of-safety price. This is a patent-cliff business with a wide gap between cases and one large binary legal event within about two years. The estimate calls for a margin of safety of about 45% below the base value, which gives a margin-of-safety price of $9.50. At that price the cash on hand alone covers about two-thirds of the price, and if the shares reached the base value within about three years, the base case implies more than 20% a year.