88 Patients to $100 Million: The First Drug for Alexander Disease Costs $285,000 a Dose
On September 3, the FDA approved Zanvastro, the first treatment for Alexander disease, a fatal childhood brain disorder. Ionis priced it at $285,000 per quarterly spinal injection. The company's own peak-sales guidance divides out to about 88 patients on therapy, and a price-per-gram comparison with the only other spinal ASOs ever sold shows the orphan-drug pricing curve at its purest.
$5.7 million. That is the list price of one gram of Zanvastro, approved September 3 as the first treatment for Alexander disease. Each quarterly dose holds 50 milligrams of antisense oligonucleotide and costs $285,000: $1.14 million per patient per year for a disease affecting fewer than 1,000 Americans, which means the entire US market for this drug is smaller than a minor-league baseball crowd. Reuters reported the price the next day; Ionis guided peak sales above $100 million. Do the division: it comes out to 88.
Alexander disease is a rare, progressive, often fatal leukodystrophy with no approved treatment until last month. A GFAP mutation causes toxic overproduction of glial fibrillary acidic protein, which accumulates in astrocytes. Children lose motor control, speech, and the ability to swallow or protect their airways, often starting in early childhood; until last month, treatment meant managing symptoms while the disease advanced.
Zilganersen, sold as Zanvastro, is an antisense oligonucleotide that binds GFAP RNA to cut production of the abnormal protein. It is injected into the spinal canal every 12 weeks, with FDA Breakthrough Therapy, Orphan Drug, and Rare Pediatric Disease designations plus orphan status in Europe.
Fifty-three people ages 2 to 53 enrolled across 13 sites in eight countries for the pivotal trial, presented at the American Academy of Neurology meeting in April and randomized 2:1 against control for a 60-week double-blind period, a trial so small that a single family's decision to enroll could swing the statistics. In patients 5 and older, the 50-milligram dose produced a least-squares mean difference of 33.3% in gait-speed change on the 10-meter walk test at week 61 (p=0.041), while plasma GFAP fell 33.6% (p=0.003). In children 2 to 4, gross motor scores improved 22.9 points on the GMFM-88 scale (nominal p=0.034), the trial's closest brush with genuine reversal, in small numbers but a real direction.
Read the fine print on what "worked" means. This drug does not reverse the disease: walking speed declined in controls while the treated group's held flat, a 33% difference in change, not a 33% improvement. Amy Waldman, who led the trial at Children's Hospital of Philadelphia: "stability is what we're trying to do and slow down disease progression. We don't expect to reverse permanent neurologic damage." Ionis's Holly Kordasiewicz: "A flat line in a degenerative disease and not have that progression is really remarkable." Both statements are true at once; hold both.
88-patient business plan
Ionis guides peak sales above $100 million, which at $1.14 million per patient-year divides out to about 88 patient-years of treatment, and that is the peak: the entire commercial forecast, finished, implying treatment of 9% to 29% of every diagnosed US patient, while a child diagnosed at 5 and treated for 25 years is $28.5 million in lifetime list-price revenue, which is why one pediatric neurologist's caseload can be worth nine figures. Per child, for decades: eighty-eight such children are roughly $2.5 billion in cumulative revenue.
Oppenheimer's Jay Olson told Reuters the approval "meaningfully de-risks Ionis' broader neurology pipeline," including obudanersen due in late 2027.
Price per gram: the orphan curve, quantified
Zanvastro is only the third spinal antisense oligonucleotide ever sold; here is the price per milligram: three drugs, no exceptions.
| Drug | Dose | List price per dose | Price per gram | Annual cost | US patients |
|---|---|---|---|---|---|
| Spinraza (nusinersen, Biogen) | 12 mg | $125,000 | $10.4M | $375K (maintenance) | ~8,000-10,000 |
| Zanvastro (zilganersen, Ionis) | 50 mg | $285,000 | $5.7M | $1.14M | <1,000 |
| Qalsody (tofersen, Biogen) | 100 mg | $14,230 | $142K | $199K | ~500 (SOD1-ALS) |
Comparator prices: Spinraza via Medscape; Qalsody via BioSpace. Patient estimates via NIH and manufacturer disclosures; annual costs assume standard maintenance dosing.
Three facts jump out. First, Zanvastro is the most expensive antisense drug launched annually, nearly 3x Spinraza's maintenance year and almost 6x Qalsody's. Second, Spinraza still charges nearly twice as much per gram, so price tracks population size, not molecule cost: fixed development costs meet shrinking denominators. Third, the ordering is exact: the smaller the pool, the larger the annual check, with spinal muscular atrophy at roughly ten thousand US patients costing $375,000 a year and Alexander disease at under a thousand costing $1.14 million. That is the orphan-drug pricing curve with three data points and no exceptions, as close to a law as pharmaceutical economics gets.
The strongest case against
Steel-manning the industry: this price is arithmetic, not greed, and the arithmetic is brutal, because developing any drug costs hundreds of millions, with a 13-site, 8-country trial dosing toddlers intrathecally ranking among the priciest clinical research per patient enrolled. Amortize that over 300 Americans and no price looks reasonable on a poster; cut the price to $28,500 a dose and peak revenue falls to $10 million, at which point no company develops drugs for 300-patient diseases, which is why no drug is the honest alternative. Without the Rare Pediatric Disease voucher, orphan exclusivity, and breakthrough designation, the math never closes, and the bill lands on insurers and Medicaid: everyone else's premiums. As for the p=0.041 endpoint: with 53 patients in the global trial, a larger confirmatory study is impossible because the patients do not exist, which means one small study plus a biomarker moving 33.6% is the maximum evidence this disease will ever produce, so demanding more certainty is demanding the drug never exist.
Limitations
Every price here is a US list price; net prices after rebates are lower and undisclosed, so $1.14 million overstates what payers transfer. Ionis's roughly 300 US patients and the NIH ceiling under 1,000 are different denominators, so the 9% to 29% range is illustrative. Price per gram is a rhetorical normalization, not a manufacturing-cost claim; the actual cost of goods is proprietary, and durability beyond 61 weeks is unknown. That this de-risks Ionis's Angelman program is one analyst's opinion; the obudanersen readout in late 2027 is the test.
What to watch
Prior-authorization fights at commercial insurers and Medicaid coverage decisions state by state, plus whether EMA approval follows and roughly triples the addressable population. And the Angelman readout in the second half of 2027: the same spinal ASO platform aimed at a population an order of magnitude larger, where the orphan curve predicts a price an order of magnitude smaller, and if obudanersen launches near a million, the curve was never about population size at all.
If this story touches your life, the actionable version is short. Families with a new Alexander diagnosis should get prior authorization moving immediately, because a $1.14 million annual claim will be contested line by line, and every price here is a list price: the number the insurer pays gets negotiated after approval. Investors get a simpler playbook: the Angelman readout in late 2027 is the binary event for the spinal-ASO platform, and the obudanersen launch price will reveal whether the orphan curve is a law or a coincidence.
The Bottom Line
A fatal childhood brain disease got its first treatment last month, one that holds the line rather than pushing it forward, a genuine achievement in a degenerative disease. At $285,000 a dose for 88 patients' worth of peak sales, the price looks obscene until you divide by the denominator; then it becomes the only price at which the drug exists. Three spinal ASOs now define the curve exactly, ten thousand patients at $375,000 a year and one thousand patients at $1.14 million, because the system has decided, without ever voting on it, that the rarer your disease the more your treatment costs, and Zanvastro is the purest data point yet. Watch the Angelman readout next year to see whether the curve bends or just continues.