SELLAS Life Sciences Group
This is the archived full research. Read the story
A small New York drug developer with no sales, waiting on one survival trial in a deadly blood cancer, priced as though the trial has already worked.
The scores
Growthn/a →
No product is on sale, so there is no revenue to grow; the case is a trial result, not a growth rate.
Business quality2/4 →
No moat until approval; if approved, the first second-remission drug with orphan and biologic exclusivity, but venetoclax and transplant remain.
Financial position3/4 ↑
$134.7M of cash, no debt and about $30M a year of cash used fund years of trials, though not a launch.
Execution record2/4 →
REGAL enrolled and reached its count, but ten months after the registered date, and the China arbitration was lost.
Per share capture1/4 ↓
Shares nearly tripled in under two years through warrants, and a launch will need more; royalties go to MSK.
Potential2/4 →
About 8,750 patients a year by the company's count, worth perhaps $1.1B at Onureg's price, but only if REGAL succeeds.
Risk profile2/4 →
Binary: one trial result decides most of the value, and the cash would survive a failure.
Valuation2/4 ↑
At 11.6 times book value (what it owns minus what it owes), the price pays $1.39B for unproven drugs.
Evidence quality2/4 →
The filings and registry are solid, but GPS's efficacy rests on ten patients and single-arm studies, with no call to question management.
If it works 0.6× · if it fails you keep about 5%.
The Verdict
We rate SELLAS Life Sciences Neutral, with low conviction. The business is weak today: it sells nothing, it has nearly tripled its share count in two years, and its lead drug's evidence rests on ten patients and studies without a fair comparison group. The price is full: the shares cost about $1.39B more than the cash, which already pays for a successful trial. The bet is priced for success: if the trial works, an approved drug at a sensible price adds little to today's value, and if it fails holders keep only about a twentieth. The result is due this month, so the view can change quickly.
The long caseIf REGAL shows patients live longer, GPS becomes the first drug to keep AML in a second remission, protected for years.
How we weighed it
The price against our value
Against it
No fair value is drawn because there are no sales; the price pays well above the cash for drugs not yet shown to work.
The business
Against it
Weak
Nothing is approved and the company has never sold a product; a moat appears only after a positive trial and approval.
Growth against what the price assumes
Cuts both ways
An approved GPS could reach thousands of patients a year with nothing else approved in its setting, but no sales exist today.
What could go wrong
Against it
One open-label trial decides most of the value, the earlier evidence is small, and new shares will be needed to launch.
Against competitors
Cuts both ways
No approved drug keeps a second remission, but doctors already continue venetoclax and azacitidine and send fit patients to transplant.
How its group is behaving
Cuts both ways
Small biotech shares move on their own trial news; the group's mood matters less than the readout.
The latest news and report
Cuts both ways
The final count was reached and the result promised this month, but it arrived late and the shares fell on it.
The evidence under this - 6 claims
SLS009 has its own value apart from GPS, so not all of the $1.39B is a bet on REGAL.
The comparison was not randomised and the GPS group was ten patients, so patients well enough to take a vaccine could simply have been healthier.
The 28 October release promised the result 'this quarter', which runs to the end of December.
The company
SELLAS is a 13-person company in New York that develops cancer drugs and pays others to run its trials and make them. GPS, licensed from Memorial Sloan Kettering Cancer Center, is its lead. The second, SLS009, licensed from China's GenFleet, is a drug given by drip alongside standard leukemia drugs.
Its future customers would be leukemia doctors and the insurers who pay them. Today it has none, because nothing is approved.
The evidence under this - 4 claims
The comparison was not randomised and the GPS group was ten patients, so patients well enough to take a vaccine could simply have been healthier.
The patent covering the WT1-A1 peptide and some method-of-use patents expire in 2026, so exclusivity would rest on regulation more than on patents.
How it makes money
It does not make money yet. It earns only interest on its cash and spends about $30M a year on trials and staff, paid for by selling shares. If GPS is approved, SELLAS would sell it in the US and Europe and pay a royalty of a few percent to Memorial Sloan Kettering; a Chinese partner holds the rights in Greater China.
The evidence under this - 3 claims
Where the money comes from, stream by stream
| Stream | Share of sales (nine months to September 2026) | How it earns it |
|---|---|---|
| None | 0% | SELLAS has no product on sale; its only income is interest on its cash. |
Who runs it
Angelos Stergiou, a doctor, founded the company in 2012 and has run it since; he was paid a $676,000 salary and a $390,390 bonus in 2025, the bonus tied to clinical and funding goals. Officers and directors together own about 1.3% of the shares.
The evidence under this - 2 claims
Why it wins, and who could beat it
Why customers come back
Nothing protects it yet. If the trial succeeds, GPS would be the first drug approved to keep this leukemia away after a second treatment, and US law would protect it from copies for years as a rare-disease drug made from proteins. Some patents end in 2026, so the rules, not patents, would protect it.
The need is real: only about one patient in three is alive five years after diagnosis, and the one approved drug for keeping the disease away is labelled only for the first remission.
The evidence under this - 4 claims
Doctors already give venetoclax and azacitidine after a second remission off-label, which is part of REGAL's control arm.
The patent covering the WT1-A1 peptide and some method-of-use patents expire in 2026, so exclusivity would rest on regulation more than on patents.
Part D figures are gross of rebates and cover only Medicare patients on an oral drug; GPS is an injection billed differently.
The threat
Doctors already keep many patients on the drugs that put them into remission, AbbVie's venetoclax and azacitidine, and send fit patients to a stem cell transplant, the only proven cure. Newer drugs such as Syndax's Revuforj treat relapse in patients with certain gene changes.
The evidence under this - 3 claims
It treats relapse in two genetic subtypes; it is not a maintenance drug for every patient in second remission.
Doctors already give venetoclax and azacitidine after a second remission off-label, which is part of REGAL's control arm.
The rivals: ABBV, BMY, SNDX, and 1 more that is private, in-house or listed abroad.
Who else does this, rival by rival
| Ticker | Rival | Where it competes | How far along | Priced against it |
|---|---|---|---|---|
| ABBV | AbbVie's venetoclax with azacitidine, continued after remissionPriced against it: One segment only | Two cancer drug candidates, GPS and SLS009 | On sale | One segment only |
| BMY | BMS's oral azacitidine, OnuregPriced against it: One segment only | Two cancer drug candidates, GPS and SLS009 | On sale | One segment only |
| SNDX | Syndax's menin inhibitor RevuforjPriced against it: Different product | Two cancer drug candidates, GPS and SLS009 | On sale | Different product |
| Not listed | Allogeneic stem cell transplantPriced against it: Not listed | Two cancer drug candidates, GPS and SLS009 | On sale | Not listed |
How big it could get
How big this could get
SELLAS counts about 8,750 patients a year in the US, the largest European markets and Japan who reach a second remission. That figure is the company's own arithmetic from older reports, and not all of those patients would be eligible.
For scale, Medicare paid about $129,000 a patient for Bristol Myers Squibb's Onureg in 2024, across 665 patients. At that price, 8,750 patients would be about $1.1B a year before any share of the market is assumed.
The evidence under this - 3 claims
The counts are from 2022 reports and the 50% rate is from older salvage regimens; not all those patients are ineligible for transplant, which REGAL requires.
Part D figures are gross of rebates and cover only Medicare patients on an oral drug; GPS is an injection billed differently.
The results
What growth costs
With no sales there is nothing to compare costs against. Spending fell in 2025 after the trial finished enrolling, then rose 56% in the first nine months of 2026 as the company prepares to apply for approval.
The evidence under this - 2 claims
The last report (12 Nov)
In the quarter to September 2026 SELLAS lost $10.1M, as research spending rose to $7.0M while it prepares to apply for approval. It ended the quarter with $134.7M of cash after warrant holders paid in $85.1M this year, and no debt.
It said the trial result would come in November and that the first SLS009 data in newly diagnosed patients would come by the end of the year.
What it did to the thesisIt changed nothing about the case. The cash covers years of trials at this pace, so the company will not be forced to sell shares before the result, but the quarter held no new evidence on whether GPS works.
The quarter in detail
Earnings per share $-0.05 against $-0.05 expected; $-0.05 as filed, with no adjustments.
The evidence under this - 4 claims
The registry lists primary completion at the end of 2026, so data on half the patients may be early.
The 28 October release promised the result 'this quarter', which runs to the end of December.
What management did with the money
SELLAS has funded itself by selling shares and warrants, the right to buy shares later at a set price. The share count went from 74 million at the end of 2024 to 204 million in November 2026. In July 2026 it lost an arbitration against its Chinese partner. A $150M programme to sell more shares is in place and unused.
Insiders in 2026Q2: 1 purchases and 0 sales, from their Form 4 filings.
The evidence under this - 4 claims
A launch after approval would need a sales force and stock, which the current plan does not include.
The price, and where we disagree
In short. Full: the price pays about nine years of a plausible GPS's sales for a drug whose trial result is not yet out.
The 80th death in the trial was announced on 28 October 2026, which starts the final count, and the result is promised for November. The shares fell 15% that day and are 53% below their June high, after law firms questioned the timing of the company's statements. More than a quarter of the shares that trade are sold short.
The next test is the result itself: whether patients given GPS lived meaningfully longer than those on standard care.
The evidence under this - 5 claims
The registry had estimated primary completion in December 2025, so the final count came about ten months later than registered.
The 28 October release promised the result 'this quarter', which runs to the end of December.
Such investigations are solicitations for clients; none has become a filed suit.
How we value it
No fair value: SELLAS has no sales, has never made a profit, and its value turns on one trial result not yet public, so a value built on earnings would rest on a guess; the bet prices it instead.
How we read the multiple
At $7.47 the 204 million shares are worth about $1.5B. Take away the $135M of cash and the price pays about $1.39B for two drugs that are not yet approved.
If GPS is approved and reaches about 1,160 patients a year at Onureg's price, sales would be about $150M a year, so the price is about nine years of those sales. That already assumes success.
Against its own history. It trades far above its usual price against its net worth, because the money raised this year is small beside what buyers now pay for the trial.
The evidence under this - 4 claims
SLS009 has its own value apart from GPS, so not all of the $1.39B is a bet on REGAL.
A maintenance drug that extends survival keeps each patient on treatment longer, which would raise the revenue per patient above one year's price.
Priced against its rivals
None of the listed rivals above sells the same thing to the same buyers as a large part of what it does. A price comparison needs at least two, so the price is judged against its own history instead.
Also checked and set aside: Curis (a substitute, not the same product).
Private or foreign rivals that cannot be priced: Delta-Fly Pharma, Inc., AROG Pharmaceuticals, Inc., Cothera Bioscience.
The bet: what it pays if it works
If the trial works and GPS is approved, a reasonable result is about 1,160 patients a year, nearly twice Onureg's Medicare patients, and about $150M of sales. At six times sales that is worth about $900M, less than today's market value, so even success pays little from here.
If the trial fails, holders keep the cash left after a couple of years of SLS009 trials, about a twentieth of today's value. The bet is priced for success.
| Revenue in two to three years | Patients treated for a year needed | At the stated multiple |
|---|---|---|
| $50.00M | about 387 at 100% kept | 0.2× |
| $150.00M (the big win) | about 1.2K at 100% kept | 0.6× |
| $357.60M | about 2.8K at 100% kept | 1.4× |
To pay 2× (a Fair bet) it would need about $507.61M of revenue (about 3.9K patients treated for a year); to pay 5× (a Cheap bet), about $1.27B of revenue (about 9.8K patients treated for a year), at the same multiple.
The steps it has to take
- Done REGAL enrolment completed
- Done Committee lets REGAL continue at the 60-death interim
- Done 80th death reached, final analysis triggered
- In progress Positive REGAL topline survival result due 2026-12
- Not started Licence application filed with the FDA due 2027-12
- Not started FDA approval and first sales due 2028-12
The detail behind the label
The evidence under this - 4 claims
A maintenance drug that extends survival keeps each patient on treatment longer, which would raise the revenue per patient above one year's price.
Part D figures are gross of rebates and cover only Medicare patients on an oral drug; GPS is an injection billed differently.
Where we disagree with the market
The trial will probably work, and the slow deaths show it.
It may work, but the evidence is thin and success is already in the price.
Every analyst covering it rates it a buy, and many holders read the slow pace of deaths in the trial as a sign that GPS patients are living longer.
We cannot read it that way: the deaths are counted across both groups together, and the earlier evidence is ten patients compared with others who were not chosen at random. The trial could well succeed, but the price already pays for that.
What analysts say, and the evidence
The evidence under this - 5 claims
One analyst is no consensus, and FMP's dated series counts only two analysts against five in the snapshot.
A recommendation to continue only means the trial did not cross the futility or efficacy boundaries; it is not evidence that GPS works.
The comparison was not randomised and the GPS group was ten patients, so patients well enough to take a vaccine could simply have been healthier.
The registry had estimated primary completion in December 2025, so the final count came about ten months later than registered.
SLS009 has its own value apart from GPS, so not all of the $1.39B is a bet on REGAL.
What could go wrong, and how we would know
The risks
- One result decides it If the trial misses, most of the value goes, leaving cash and the second drug.
- Thin earlier evidence The case rests on ten patients and studies with no fair comparison group.
- Open-label trial Doctors and patients knew who got GPS, which regulators may weigh against a small benefit.
- More shares Launching a drug costs far more than the cash, so new shares are likely even after success.
- Delays The final count came about ten months later than the trial registry had estimated.
The evidence under this - 5 claims
The comparison was not randomised and the GPS group was ten patients, so patients well enough to take a vaccine could simply have been healthier.
A launch after approval would need a sales force and stock, which the current plan does not include.
The registry had estimated primary completion in December 2025, so the final count came about ten months later than registered.
What we still do not know
- What survival gain the trial was designed to show, and how likely it was to show it.
- Whether insurers would pay for GPS like a cancer drug or like a cheaper vaccine.
- Whether SLS009 works well enough in newly diagnosed patients to carry the company if GPS fails.
The evidence under this - 3 claims
Part D figures are gross of rebates and cover only Medicare patients on an oral drug; GPS is an injection billed differently.
The registry lists primary completion at the end of 2026, so data on half the patients may be early.
What would change our view
- Neutral→AttractiveThe trial works
It needs all of:
- ○The trial shows patients on GPS live longer, by a margin unlikely to be chance.watched as it happens
- ○SELLAS applies for approval with the FDA's agreement.watched as it happens
- Neutral→UnattractiveThe trial fails
It needs:
- ○The trial misses its survival goal.watched as it happens
- Neutral→AttractiveA price that pays little for the drugs
It needs:
- ○Before the result, market value falls to five times book value, what it owns minus what it owes.about $3.22 today, 57% below the last close
We would also be wrong if
- The FDA will not accept an application based on an open-label trial.
- SELLAS sells more than 40 million new shares before any approval.
- SLS009 beats standard drugs in its randomised trial in newly diagnosed patients.
What is coming
- 30 November 2026 REGAL topline primary endpoint result: Whether GPS extends survival in second remission with statistical significance: the whole case.
Every dated event ahead
- 30 November 2026 REGAL topline primary endpoint result: Whether GPS extends survival in second remission with statistical significance: the whole case.
- 31 December 2026 SLS009 relapsed AML data at ASH, and first-line preliminary data: Whether the second drug works where venetoclax fails, the fallback if REGAL disappoints.
- 25 March 2027 FY2026 annual report: Cash left, the plan for a licence application and how much new stock was sold.
Since we wrote this
- 17 NovThesis writtenNeutral, low convictionWritten days before the REGAL topline result, with the shares at $7.47 after a 53% fall from June.
Sources 13
Sources - 13
Every claim above cites one of these, or states the reasoning that derived it. The quote is frozen at the date it was read: a link can rot, and an agent fetch returns a reading of a page rather than the page. Backs counts the whole claim ledger, so it can name a claim this page does not surface.
- S01primarycompany filing2026-03-19the two drug candidates, the REGAL design and event history, the licences with MSK, GenFleet and 3D Medicines, the patents, the named competitors, the patient counts, the 2025 and 2024 statements and the share countbacksC01C02C03C04C06C10C17C18C21C22C23C29C32
- S02primarycompany filing2026-11-12cash, costs, cash used, the warrant exercises and the unused at-the-market programme, warrants and options outstanding, the lost arbitration, the SLS009 data and enrolment, and the share count on 11 November 2026backsC01C07C08C09C10C11C24C25C26C30
- S03primarycompany filing2026-11-12the quarter's costs and loss, cash of $134.7M, and the promise of REGAL topline results in November 2026 and first-line SLS009 data in the fourth quarterbacksC05C07C08C25C26
- S04primarycompany filing2026-03-19the full-year 2025 results and the plan for 2026 announced beside the annual reportread, but no claim rests on it
- S05primarycompany filing2026-04-23the officers and directors, what they own, what the chief executive is paid and what the bonus is paid onbacksC27C28
- S06primarycompany release2026-10-28the 80th death in REGAL on 28 October 2026, which starts the final analysis
has reached the pre-specified 80 th event (number of deaths), triggering the final analysis of the study
backsC03C05 - S07secondaryvendor data2026-11-16Research pack, frozen beside this thesiscontent/SLS/research-pack.jsonthe close, the price history, short interest, the analyst panel and the one analyst's estimates, and the insider rosterbacksC12C14C15C28C30
- S08primaryregulator2025-10-01ClinicalTrials.gov record NCT04229979, the REGAL study U.S. National Library of MedicineREGAL's design as registered: open-label, 127 patients actually enrolled, survival as the primary endpoint, and a control arm of investigators' choice that includes observation
Open-label, multicenter, randomized, parallel group study
backsC02C03 - S09primarygovernment data2026-04-15National Cancer Institute SEER, Cancer Stat Facts: Acute Myeloid Leukemia National Cancer Institutenew US AML cases and deaths expected in 2026, and five-year relative survival of 33.4% for 2016 to 2022
In 2026, it is estimated that there will be 22,720 new cases of acute myeloid leukemia ... an estimated 11,500 people will die of this disease ... 5-year relative survival 33.4% ... SEER 21 (Excluding IL) 2016-2022
backsC16 - S10primaryregulator2026-06-12FDA prescribing information for ONUREG (oral azacitidine), via openFDA U.S. Food and Drug Administrationthat the one approved drug for keeping AML in remission is labelled for a first remission only
indicated for continued treatment of adult patients with acute myeloid leukemia who achieved first complete remission
backsC18 - S11primaryregulator2026-02-04FDA prescribing information for REVUFORJ (revumenib), via openFDA U.S. Food and Drug Administrationthat Syndax's menin inhibitor is approved for relapsed AML with an NPM1 mutation or a KMT2A rearrangement
the treatment of relapsed or refractory acute myeloid leukemia (AML) with a susceptible nucleophosmin 1
backsC20 - S12primarygovernment data2026-06-01CMS, Medicare Part D Spending by Drug, data year 2024: Onureg Centers for Medicare & Medicaid ServicesMedicare Part D spending on Onureg in 2024: 665 patients and about $129,238 per patient
Brnd_Name Onureg Tot_Spndng_2024 85943002.48 Tot_Benes_2024 665 Avg_Spnd_Per_Bene_2024 129237.59771
backsC19 - S13primaryregulator2026-09-03ClinicalTrials.gov record NCT04588922, the SLS009 study U.S. National Library of Medicinethe SLS009 study's registered enrolment of 160 and an estimated primary completion at the end of 2026
Study of SLS009 (Formerly GFH009) a Potent Highly Selective CDK9 Inhibitor in Patients With Hematologic Malignancies
backsC26