Iovance Biotherapeutics

IThe business

Iovance sells one cancer treatment, Amtagvi, and the IL-2 drug that goes with it. Amtagvi is made from a patient’s own tumour. It was approved for advanced melanoma in the U.S. on 16 February 2024. Total revenue was $325M in the twelve months to June 2026, of which Amtagvi was about $273M C (calculated).¹³⁴³⁹ The company has lost money every year since it began, $290M in the same twelve months C (calculated), and is not close to profit on today’s sales.⁴³

Iovance has a real product that sells and a share price that treats it as a bigger one. The shares went from $2.73 at the end of 2025 to $14.82 on 30 September 2026 and to $8.02 now, on news about lung cancer that I think is more mixed than the first headlines.

TIL therapy, from the start

The immune system has T cells, white cells that can kill other cells they recognise as foreign. Many tumours already hold T cells that have found the cancer. These are called tumour-infiltrating lymphocytes, or TIL. They are too few, and the tumour wears them out, so they lose.

TIL therapy tries to give them a better chance. A surgeon removes a tumour. A lab keeps the T cells and grows them, with the help of a signalling protein called IL-2, from thousands to billions. The patient has chemotherapy to clear away other white cells, gets the billions back, and then gets more IL-2 to keep them going. The cells are the patient’s own, so each dose is made for one person. That is also why it is slow, costly and done only in hospitals.

The numbers on the label explain who gets it. Among 73 patients in the study at the recommended dose, 23 responded, which is 31.5% C (calculated), and the median length of response had not been reached.¹¹ A response means the tumours shrank by at least 30%. It does not mean a cure. In the label study 7.5% of patients died of treatment-related causes, 12 of 160 R (reported).¹¹ The treatment is hard, and it is for people with few options.

Longer follow-up is better than the label. The five-year paper pooled 153 patients and reported a 31.4% response rate, a median response of 36.5 months and 19.7% alive at five years R (reported).¹⁰ Real-world use is at least not worse: a study of 41 patients treated at hospitals with the commercial product found a 44% response rate, and 52% in patients who had two or fewer earlier treatments R (reported).¹⁰¹⁴ Treating people earlier looks better, which matters for what comes next.

How much it sells

Product revenue began in 2023 with $1M of Proleukin, and Amtagvi sales began in 2024. Revenue was $164M that year and $264M in 2025, of which Amtagvi was $220M R (reported).⁹ In the second quarter of 2026 it was a record $99M, of which Amtagvi was $91M, up 68% on a year earlier R (reported).³ On 29 September 2026 the company raised 2026 revenue guidance to $410M to $420M from $350M to $370M, after the second quarter beat its own guidance by more than 10% R (reported).⁵¹⁰

The company does not tell us how many patients it treats each quarter. It says about 2,000 patients have been treated with its TIL products, commercial and in trials together.¹⁰ I estimate the second quarter at roughly 190 infusions, taking $90.7M of Amtagvi revenue and a net price of about $470,000 E (estimate). That net price is my guess. The launch list price was about $515,000 and the 10-Q says prices of both products have risen since.¹⁹³ The gap between list and net is rebates and other deductions the company does not break out in a number I can use. If the net price is lower, patients are more. Either way the sale is one-off, so every patient has to be found again.

The network that finds them has grown to about 100 authorised treatment centres, a third of them community hospitals, and the company expects at least 110 by the end of 2026.⁵⁸¹⁰ It says more than 75% of patients live within 200 miles of one.¹⁰ The first community centres began treating in late 2025.¹⁴ That is the change behind the recent quarters.

Two things in the revenue need care. Proleukin was $44M in 2025 and $9M in the second quarter of 2026, and the fourth quarter of 2025 alone had $22M R (reported).⁹¹⁴³ I do not know why and the company has not said, so I do not use Proleukin as a measure of demand. And the first quarter dips in 2025 and 2026 are the annual plant maintenance, so I read the sales trend across a year, not quarter on quarter.

The plant

All commercial Amtagvi is made at the Iovance Cell Therapy Center in Philadelphia, which the 10-K calls the first FDA-approved facility to make a commercial TIL product.⁹ The company has moved all lifileucel making in-house.¹⁴ It says the plant can treat more than 5,000 patients a year from North America, Europe and Asia-Pacific, and it is now at under a fifth of that by my count E (estimate).¹⁰ Gross margin, excluding depreciation, was 56% in the second quarter of 2026 and 50% for the first half, up from about 43% in the third quarter of 2025 R (reported).⁸¹⁰²⁰ A plant sized for 5,000 and run at under 1,000 carries a lot of fixed cost. That is where the margin gain should come from, and where it would stop if sales stalled.

Where it can be sold

Outside the U.S. the story is slow. The European application was withdrawn in 2025 and will not come back until 2027, so there is no European approval. The company says the U.K. could approve this year. I treat ex-U.S. sales as small until I see them.

The confirmatory trial, and what could go wrong with accelerated approval

Amtagvi was approved on a response rate in a single-arm study, not on survival in a randomised one. The FDA’s label says continued approval may depend on a confirmatory trial.¹¹ That trial is TILVANCE-301. It randomises about 670 people with melanoma who have not yet been treated, to lifileucel plus the drug pembrolizumab, or pembrolizumab alone, with dual main goals of response rate and progression-free survival.¹⁰²¹ The registry lists primary completion in March 2028 R (reported), an estimate.²¹ A first look at response rate may come earlier and could support a first-line filing.⁸

If the trial misses on progression-free survival, the FDA can ask the company to withdraw the second-line approval, and since 2022 it has had faster tools to do so. I have not got a firm figure for how often that happens, so I treat it as a fork. In my bear case the trial misses in 2029 and the U.S. label is lost or narrowed in 2030. I put that at about one chance in four E (estimate), from roughly a one-in-two chance that a combination that looked very good in a small early study misses on survival in a large one, and a bit under a one-in-two chance that the FDA then acts.

Lung cancer, and what happened on 29 October

Lung is the larger prize. The company says the U.S. market in previously treated non-squamous lung cancer is about seven times that of second-line melanoma.⁸¹⁰ In October 2025, 39 patients showed a 25.6% response rate against 12.8% for the standard chemotherapy docetaxel, and the company called the result best in class.¹⁰ The FDA gave Fast Track status.¹⁴ The company then pointed to a possible launch in the second half of 2027.¹⁴¹⁰

On 29 October 2026 it updated the data to 69 patients. The response rate was 24.6%, 17 patients, in line with the early read, with 30 more patients added.⁷ The company said the broad programme will give a final read of about 80 patients, with an sBLA planned for 2027, and that talks with the FDA are ongoing.⁷ The release does not repeat the 2027 launch. It then put the main weight on a subgroup: patients whose tumour had a normal KRAS gene, a normal HER2 gene and who smoke or used to. Of 22 such patients, 13 responded, 59.1% R (reported).⁷ The subgroup was found by looking back at the data, and the company says a new study, POSITILITY, will test it beforehand in a prospective cohort starting in November, with registrational intent subject to FDA feedback.⁷ The stock fell 35% that day R (reported).¹

I read the table this way. The broad result did not improve, and docetaxel is a low bar. The subgroup is real data, but 13 of 22 has a wide margin. A simple 95% interval for 13 of 22 runs from about 36% to 79% C (calculated). Picking the best-looking slice after the fact usually gives a number that comes back toward the average in a fresh group. The company’s own forward-looking statement warns that retrospective subgroup results may not be confirmed.⁷ So the evidence supports hope that lung will work in some patients. It does not yet support the approval date, the label or the size of market that the old story used. I build my lung odds from this, in the valuation.

Other cancers

The soft-tissue sarcoma trial, SARATOGA, is under way after three responses in the first six patients. It has Fast Track status and aims at two rare, hard sarcomas.⁸²² The endometrial cancer study showed a 40% response in five patients with serous disease, all of whom had already had chemotherapy and a PD-1 drug, and the company is talking to the FDA about a fast path.¹⁰⁸ Cervical cancer was an early Iovance trial, with a Breakthrough designation and a manuscript still to come, but I see no registrational plan for it in the 2026 documents.⁹⁸ Head and neck cancer appears only as one cohort of the IOV-5001 basket.¹⁰ Each of these is a small first step, and I give them little value, $80M of revenue in 2035 in the base case after weighting.

The next generation, and the IL-2 problem

Today’s regimen leans on high-dose IL-2, which is what puts patients in the hospital with fluid leaks and low blood pressure, and it needs an intensive-care unit on call.¹¹ Iovance has one answer in development, IOV-3001, a modified IL-2 that is meant to grow useful T cells while sparing the ones that dampen the immune system, so that a lower dose might do.⁸¹⁰ It is a better IL-2, not a TIL treatment that does without IL-2, and I found no IL-2-independent Iovance product in the documents I read. It is in a Phase 1 safety stage.

The other two are about making the cells stronger. IOV-4001 switches off the PD-1 gene in the cells, so the tumour cannot use that signal to turn them off. IOV-5001 carries an immune signal, IL-12, on the cell, meant to rouse tumours that the immune system leaves alone.⁸¹⁰ Both are early, with small numbers of patients, and neither is in my valuation beyond a nominal amount.

IIThe moat

What protects Iovance today is a plant, a rule book and a set of hospitals that have learned to run the treatment. It is narrow. Nothing here stops a rival from entering melanoma with a better product, and nothing yet shows that Iovance can raise its price without losing patients.

Start with the legal protection. Iovance says it owns or licenses more than 400 granted or allowed patents and has filed more than 1,000 more, and it expects exclusivity through at least 2042.⁸ Those are the company’s words, and I have not tested them. The U.S. also gives a new biologic 12 years before a copy can be approved, which for a February 2024 approval runs to February 2036 E (estimate), and the FDA’s orphan designation for melanoma, granted in 2015, adds seven years from approval for that use, to February 2031 E (estimate).⁹¹³ These protect the product. They do not protect against a different treatment.

Then the knowledge. The 10-K calls the Philadelphia plant the first FDA-approved facility to make a commercial TIL product, and the company believes it is the only one with a centralised, scalable process.⁹ Growing a patient’s own tumour cells in large numbers, in 31 days, with few failures, is hard. In the label study, 8 of 189 patients who had surgery could not get a product made, about 4% C (calculated).¹¹ Competitors have to learn this, and the cost of learning is years and capital.

Then the network. About 100 treatment centres have been trained and have workflows for tumour harvest, shipping, inpatient care and IL-2.⁵ Their staff are trained once and can serve every indication, the company says, so the same centres can treat lung or sarcoma patients later.¹⁰ This is the most valuable part of the protection if more indications come. It is nothing if there are none.

What does not protect Iovance:

It has no price power that I can see in the data. The company raised list prices, and net revenue per patient is not disclosed. The Medicare and Medicaid rules, state laws and proposals that would cap list prices at foreign levels are named as risks in the 10-K.⁹

It is a one-time treatment. A patient who is treated does not come back. In a drug taken every month, the company keeps the patient and the price keeps rising. Here, the company has to find a new patient to earn the same revenue.

Hospitals can make TIL themselves. The Netherlands Cancer Institute, with Danish and British partners, ran a Phase 3 trial of its own TIL against ipilimumab in melanoma, reported in the New England Journal of Medicine in December 2022, and the headline result was a doubling of progression-free survival.⁹²⁶ A hospital with the right licence, plant and staff could do the same, at a cost the company may not match. In the U.S. the FDA approval and the patents matter more. I give it a weight, not a name.

It rests on one raw material, IL-2, whose supply Iovance bought in 2023 by acquiring Proleukin from Clinigen.⁹ That is a strength today and a dependency: the regimen needs it, and nobody has a cheaper replacement yet.

IIIThe competitive advantage

The advantage is that Iovance has the only FDA-approved treatment for melanoma after a PD-1 drug, and it works in about three patients in ten, for years. The risk is that "only" is a statement about today.

The patients are the ones who failed a checkpoint inhibitor, the PD-1 drugs such as pembrolizumab and nivolumab that are first-line treatment. For them there is little: the company says more than half of first-line patients progress within twelve months.¹⁰ Before Amtagvi the answers were a second immune drug, chemotherapy, or a trial. A 31% response rate that is still going in a third of responders at five years has no match in that group in the documents I read.¹⁰ The company’s own estimate of the U.S. second-line-and-later pool is 8,500 patients a year.¹⁰

Competitors fall in four groups. Large drug makers are working on melanoma and lung treatments after PD-1 drugs, and the 10-K names BioNTech, Bristol-Myers Squibb, Daiichi Sankyo, Eisai, Genmab, Immunocore, IO Biotech, Merck, Moderna, Pfizer, Regeneron and Replimune.⁹ Other cell therapy companies, among them Immatics, Obsidian and AbelZeta, are developing engineered T cell and TIL products.⁹ Other IL-2 developers include Alkermes, Merck, Nektar and Sanofi.⁹ And hospitals can run their own TIL. The 10-K does not name Adaptimmune or Lyell, so I read their own filings. Adaptimmune sold its approved cell therapy Tecelra and its other cell therapies to US WorldMeds for $55M in cash plus up to $30M of milestones, and then left the Nasdaq in October 2025.²⁷²⁸ Lyell stopped its TIL programme, LYL845, in 2024 after buying a different cell therapy.²⁹ Neither is a threat to Iovance today. Both show how hard it is to make a living from cell therapy for solid tumours.

The question for the advantage is how long the second-line niche holds. Three things could shrink it. First-line treatment may get better, so fewer patients fail and arrive at Amtagvi. Another treatment may reach the same patients without surgery, a hospital stay and a wait of five weeks, and with a much lower price. And Iovance’s own first-line trial, if it works, would move the product to an earlier line with more patients, which would help the company and also change what second-line even means.

Next to its own competitors, the lung programme is the test of the advantage. In lung cancer the usual treatment, docetaxel, produced 12.8% response, 5.6 months of response and 12.3 months of overall survival.¹⁰ Amtagvi has better response and durability in a small group. But lung patients are older and sicker than melanoma patients, a tumour has to be harvested from the lung, the lymphodepletion is hard on lungs, and the patients wait five weeks. I do not know how many of the 35,000 or so second-line patients I assume each year could be treated, which is why my eligibility figure of 35% is only a guess E (estimate).

So I say the advantage is real and narrow. It is a lead in one disease, built on a plant and a network, that gets valuable if it spreads to a second and third disease. Whether it spreads is what the price is paying for.

IVManagement

The company has been run by an interim chief executive for at least five years, and it has paid for ten years of losses by selling shares. The people are experienced. The incentives reward growing sales and a rising share price, and they do not reward the thing I care about, which is a small share count.

Frederick Vogt is the interim chief executive and president as well as general counsel. The company’s pay tables show him as the principal executive since 10 June 2021, when Maria Fardis left that role R (reported).¹² He has been at the company since 2016.¹² The 2026 proxy and every 2026 filing still say "interim". I do not know why a five-year interim has not become permanent, and the filings do not say. It could be a board that has not found anyone better, or one that does not want to pay for one. Either way, the chief executive does not hold the title that signals a long plan.

How they are paid, and what that pays them to do

Dr. Vogt’s salary is $800,000 and his target bonus 65% of it. His 2025 bonus was $312,000, 60% of target, and his total pay was $1,888,175 R (reported), 12.3 times the median employee, who was paid $153,409.¹² That is modest for a $3.8B company. The bonus is set on financial results for the approved products, the Amtagvi launch, and regulatory, clinical and operational milestones.¹² The long-term pay is time-vesting options and restricted shares, with no performance test.¹²

What that rewards is clear. It rewards revenue, launch milestones, trial and regulatory progress, and a share price that rises. It does not reward return on the capital the company burns, or the number of shares used to burn it. Time-vesting options and shares are worth more if the price climbs, however it climbs. Many of the older options are far out of the money, with strikes up to $46, so some officers have little to lose on those and much to gain from a new grant at a low price. Vogt’s unvested restricted shares were worth $1.1M at the year-end price, before the run.¹²

The board has approved selling shares freely. In June 2026 stockholders voted to raise the authorised share count from 500M to 650M, a 30% increase in room to dilute.³¹ The at-the-market agreement with Jefferies allowed $350M of sales from August 2025, with a fee of up to 3%.³² The 18 June 2026 filing shows $89.7M of that capacity still unused.⁶

Their own money

The only large insider holding I found is Wayne Rothbaum’s 30.0M shares through Quogue Capital, 6.7% of the class, valued at about $241M at the close C (calculated).¹² He is a director. The 29 October release quotes him on the scientific rationale for the KRAS strategy.⁷ That is a named holder with a stated view, and it is the company’s release, not his own statement. All other directors and executives held under 1% each at the record date.¹² I did not read the Form 4 filings for open-market buying or selling this year, so I cannot say what insiders did with their shares during the run to $14.82.

What they did with the capital

Over ten years the company sold $2,675M of shares and lost $2,671M C (calculated).⁴ It built a plant, bought Proleukin in 2023 from Clinigen, won an approval and has a product. That is a real result. It is also true that what the company got for each added share has varied by a factor of nine, from about $26 in 2020 to $2.87 in 2025. In 2025 the share count rose by 107M against $306M raised, about $2.87 for each added share C (calculated).⁴ The February 2024 offering went out at $9.15 on 23.0M shares four days after approval.³³ The best-timed raises were in 2020 and 2021, at about $26 and $22 for each added share, and the worst was 2025. The raise after approval, at $9.15, was in between.

In August 2025 the company announced a restructuring with a workforce cut of about 19%, a charge of $5.1M, and later gross margin gains.⁹ It had 975 employees at the end of 2025.⁹ The chief financial officer resigned in June 2025 and a new one arrived in August.³⁴³⁰ I treat that as a company that cut back when it had to. It is better than not cutting.

VFinancials

Iovance has never made money. Its loss is falling because sales have grown, and the company says its cash lasts into 2028. The ten-year record says be careful with that.

The loss was $53M in 2016 and peaked at $444M in 2023. In the twelve months to June 2026 it was $290M C (calculated).⁴³ The reason is sales. The costs did not move much: research spending is down from $333M in 2023 to $268M in the twelve months, and selling and administrative costs are $149M, flat for two years C (calculated).⁴³

The quarter that matters is the second of 2026. Revenue was $99.3M. Cost of sales was $43.6M, so gross profit was $55.7M, a 56% margin that leaves out depreciation C (calculated).³⁸ Research was $58.9M, selling and administrative $39.3M and depreciation and amortisation $9.5M. The operating loss was $51.9M, less than half of $113.8M a year earlier R (reported).³ Stock pay in the quarter was $11.7M, against $14.9M a year earlier R (reported).⁸ If revenue lands in the guidance range of $410M to $420M, the second half would average about $122M a quarter C (calculated).⁵ That would cut the quarterly operating loss to about $35M on my margins E (estimate). It is a long way from profit.

What an owner could take out

The reported loss understates the problem. An owner cares about cash after the plant and after stock pay. The company pays about $47M a year in shares, because it cannot pay everyone in cash and wants to keep staff. Those shares are a claim on the business.

Paying for it

Iovance has no debt and no convertible notes. The Q2 balance sheet shows no borrowings, only $46M of lease liabilities, and a $1M note outstanding at the end of 2025 is gone.³ The only convertible securities are preferred shares that turn into about 2.0M common shares, plus 12.7M restricted shares and 17.6M options that were mostly out of the money at the June price.³ The accumulated deficit is $2.90B R (reported).³ That is clean. The cost of that cleanliness is the share count.

The share count was 62M at the end of 2016 and 453M at 30 June 2026 R (reported).⁴³ Part of that is sensible financing of a company with no revenue for ten years. Part of it is the market’s price. In the first half of 2026 the company raised $148M, in all of 2025 $306M, and in 2024 $397M R (reported).³⁴ Because the price in 2025 was low, each dollar bought many shares. The shelf filed on 18 June 2026 left $89.7M of at-the-market capacity.⁶ The stock tripled in the third quarter, so I expect much of it was used. The Q3 report is due on 5 November, after this page, and I have to estimate.¹

Cash, and how long it lasts

At 30 June 2026 the company had $303.7M of cash, investments and restricted cash, with $110.8M in cash, $186.9M in short-term investments and $6.0M restricted R (reported).³ The charts and my model leave out the restricted cash, so they start from $297.7M. It says that funds operations into the second half of 2028.⁸¹⁰ In February, with $303M at the year-end and less revenue, the company had said the third quarter of 2027.¹⁴ Revenue and a smaller loss moved the date out by about a year in six months.

For the September cash I assume the company used the $89.7M of at-the-market capacity at about $11.50 a share, kept 97% after fees, and burned $45M in the quarter, which gives about $340M and about 462M shares E (estimate).⁶³ The first half burned about $74M a quarter once I net out the shares sold, and the second half should be lower as sales rise, so a range of about $325M to $355M is fair. If it sold nothing, cash would be about $253M. The right answer will be in the Q3 10-Q. I treat the sale as the likelier case because the company sold shares through 2025 and in both quarters of 2026, and the third-quarter price was the highest since early 2022.

My own runway is different from the company’s because I count stock pay as a cost and assume a bigger R&D bill than it plans. In my base case the cash trough is $161M, about 2028, and free cash flow turns positive in 2029 at $60M E (estimate). That works without new money, but I would not count on it: a lung trial that needs extra patients, a delayed filing or a slower launch would each need more. In my bear case cash is enough only because I cut costs hard in 2029 and 2030.

Where the price has been

The stock has been a bad owner’s experience. It closed at $46.40 at the end of 2020, $19.09 in 2021, $6.39 in 2022, and $2.73 at the end of 2025 R (reported).¹ Then it rose fivefold in nine months, with the big jumps on the second-quarter results (+43% on 6 August) and the guidance raise (+31% on 29 September), and gave back 46% in three sessions after the lung update R (reported).¹ A price that moves like that is set by how people feel about the next announcement. I do not use it to say what the company is worth.

VIValuation

At $8.02 the price is 40% above my base value of $5.73 and 64% below my bull of $22.29. My margin-of-safety price is $3.72. The price needs lung cancer to work, and I put the chance at 40%.

What would most likely go wrong

Before any number, here is how I think this goes badly. There are three ways.

First, the confirmatory trial misses. TILVANCE-301 is the proof that Amtagvi helps patients live longer, not only shrink tumours. If lifileucel plus pembrolizumab does not beat pembrolizumab alone on progression-free survival, the second-line approval is on weak ground, the first-line plan is gone, and the company is a small second-line business with a large plant. I give that about one chance in four of a lost or narrowed label E (estimate), and my bear case is built on it.²¹¹¹

Second, lung does not work, and the company has already spent on it. The 29 October release moved the evidence from a broad result to a retrospective subgroup. If the new cohort does not repeat it, or the FDA wants a randomised trial, the pipeline that supports the price is not worth much, and the company will have spent $240M a year on research for nothing.⁷

Third, the money. The company has sold shares for ten years, in good times and bad, and has raised from about $2.87 to $26 for each added share C (calculated).⁴ If it needs more before profit and the price is low, each dollar costs a lot of shares. My model sells new shares at 60% of value in any year the cash would run out. In my three cases it never does, so the model shows no cost for this, and that understates the risk.

Working back from those three, the page checks one thing most closely: how many patients get treated in each indication, and at what price, because everything else follows from it.

What I count, and what I leave out

I value the cash an owner could take out after stock pay and after capital spending, which today is −$343M a year C (calculated), about $53M worse than the reported net loss.⁴³ I value it year by year to 2040, then add a terminal value, then add cash. Shares sold at fair value do not change value per share, so I do not model them as such. I do model the case where the company has to sell shares cheaply, because that is what has happened.

What Amtagvi could be worth

Revenue is patients times price times share of the patients who could be treated. In the U.S. second-line melanoma pool of 8,500 a year the company cites, my base case reaches 1,500 patients by 2030 and 1,600 by 2035, 18% and 19% of the pool E (estimate).¹⁰ My bull case reaches 2,300, 27%. The bear case holds at 1,000 in 2028 and 2029 and then falls as the trial fails. For comparison, Amtagvi revenue of $91M in the second quarter is about 190 patients a quarter, or 770 a year, E (estimate), so my base is a doubling of volume in four years. That is fast but the network is growing.

The weighted lines are the ones to watch. In the base case first-line melanoma contributes 40% of 1,000 patients, which is 400. Lung contributes 33% of 1,160 plus 7% of 2,000, which is 523. No patient will be treated at those weighted numbers. A real patient is treated in a world where the label exists or it does not. In 2035 the base case total is $1,496M, and I am not saying that number happens. I am saying it is the average of worlds that do and do not work.

Read the lung table from the top. The first step is 55%, and the second is 60%. Together they give 33%. A bull-case believer would put the first step at 80% and the second at 80%, and get 64%. A sceptic would remember that most subgroup findings shrink and would stay near 20%. The data to decide between them will be the first POSITILITY patients, and they have not started yet. I cannot judge the second step well, because I cannot read what the FDA wants. The honest answer is that the range of my lung odds, 20% to 65%, is wide, and the value moves a lot over that range.

The three cases

Bear, $0.38. TILVANCE-301 misses in 2029. The FDA withdraws or narrows the second-line label in 2030. Lung never gets a label. Revenue peaks at about $546M in 2029, falls to $350M in 2030 and to $249M in 2035. The company then cuts research to $45M and selling to $60M, and runs the plant at break-even. The value is the cash left after two more years of losses, plus a small business, which is less than today’s cash per share. I give it about one chance in four.

Base, $5.73. The second-line melanoma business keeps growing to about 1,500 patients by 2030, first-line melanoma and lung are weighted by their odds, other cancers add a little, and gross margin reaches 75% by 2035. Revenue is $1,010M in 2030 and $1,496M in 2035. Operating profit after stock pay turns positive in 2029 E (estimate). The lung and first-line risks are averaged in. The risk of losing the second-line label is not, and sits only in the bear case.

Bull, $22.29. First-line melanoma is approved in 2029 and reaches 1,500 patients. Lung is approved in the enriched group in 2029 and reaches 1,500 patients, with a first-line lung use adding another 1,500 later. Revenue is $1,809M in 2030 and $3,992M in 2035. That would make Amtagvi bigger than any cell therapy sold today. I give it about one chance in five.

The price against the range

Cash at the end of 2026 is about $305M E (estimate). On 479M diluted shares, that is $0.64 a share, 8% of the price. The rest of the $8.02 is a claim on the business, valued by the market at an enterprise value of about $3.5B, which is 10.8 times trailing revenue and 8.4 times the 2026 guidance midpoint C (calculated).⁵

The base leaves out one risk, losing the second-line label, which sits only in the bear. If I give the bear 25% and the base 75%, the value is $4.39 E (estimate), well below the price. I do not add the bull on top, because the base already weights the lung and first-line outcomes. The price sits near my base only if lung is treated as certain. It does not sit near a mix that counts the label risk.

What you must believe at today’s price

The reverse calculation is simple. To get from $5.73 to $8.02 the business needs 1.19 times my revenue in every year, or lung has to work for certain on my sizes. Neither is absurd. I call both unproven. The melanoma business alone, with no lung and no first-line, is worth $2.93 on my numbers E (estimate), so about 37% of the price comes from what exists today and the rest is a bet.

Multiply it out. The bull case needs three things at once: the second-line label survives (about 75%), the enriched lung label arrives (33%) and first-line melanoma is approved (40%). 0.75 × 0.33 × 0.40 is 10%. One chance in ten is a small chance at a price that is not small.

Discount rate and sensitivity

The 10-year Treasury yields 4.94% R (reported).³⁵ Iovance’s weekly returns have moved 1.3 times the S&P 500 over five years and 1.65 times over two C (calculated).³⁶ At an equity premium of 5% that gives 11.4% to 13.2%, and I use 12%. The rate matters less than the revenue: moving it from 12% to 14% takes the base from $5.73 to $4.70, and moving revenue 30% up takes it to $9.32.

The margin-of-safety price

My margin-of-safety price is $3.72, 35% below my base. I use a wide margin because the range is wide: the bear is $0.38 and the bull $22.29, and I cannot narrow it. At $3.72 an owner pays about the value of the existing melanoma business ($2.93) plus a modest amount for the pipeline. It is a cushion against the pipeline failing, not against everything: in my bear case the shares are worth $0.38. The June quarter-end price was $4.16, not far from this. It is the price I would compare with what I would pay for the business if I could not sell it, not a forecast of where the shares will trade.

Quality and price are two questions

Is it a good business? Not yet. It has a product people want and a narrow lead, but no profit after ten years, a gross margin of 56% on a plant that is mostly empty, a one-time treatment that must find each patient anew, and a share count that has grown 7.3 times. Is it a good price? At $8.02 the price is above my base value, and on 28 October it was 85% higher. A good product can be a poor purchase at a price that pays for the pipeline in advance. This looks like that to me, at today’s price. I am not saying the shares will fall.

The alternative use of the dollar

A dollar in a 10-year Treasury earns 4.94% with no risk of loss to maturity.³⁵ My 12% hurdle is the Treasury plus 7 points for this risk, and at $8.02 the base case does not meet it: the expected return to my base value is negative. The same question on this site’s Eli Lilly page is about a business with large profits: the price needs 22% cash growth a year and I model 13%, so the price is above my value there too. Iovance’s case is smaller in dollars and larger in doubt. I cannot find a reason to prefer the doubt to the Treasury at today’s price.

What I cannot judge

I cannot judge the FDA. Whether a retrospective subgroup in a single-arm lung study gets an accelerated approval is a regulatory call I can only guess, and my 60% is a guess. I cannot judge how many lung patients are fit for a five-week wait, a tumour harvest and an intensive-care stay, so the 35% could be 20% or 50%. I cannot see net price per patient, the Q3 cash, the share count after the summer, or the timing and threshold of the first-line trial’s interim look. Where the range is wide I show the range and do not choose a precise value. If I had to choose which of my inputs is least reliable, it is the one that moves the value most, the lung odds.

VIIThe case for the price

What most people miss

The first fact is cost. Selling and administrative spending was $152M in 2024 and $152M in 2025, while revenue rose 61% R (reported).⁹ The sales force grew, and the bill did not. That is how a margin comes. The second is that real-world results are better than the trial. In 41 patients treated with the commercial product at hospitals, the response rate was 44%, against 31.5% in the label study R (reported).¹⁰¹¹ The third is that the company owes no one. It has no debt and no convertible notes, which is rare for a company that has lost money for ten years.³

The bull case

The company’s case is in its own releases. Interim chief executive Frederick Vogt said on 6 August 2026 that record second-quarter revenue and a 56% gross margin came from U.S. demand, and that manufacturing and operating efficiencies are speeding progress toward profit.⁸ On 29 October he said the lung findings point to a more precise way to pick patients, and that they could widen lifileucel’s use as a one-time treatment with a best-in-class profile.⁷ Director Wayne Rothbaum said the response pattern supports testing tumour biology to guide who is treated.⁷ The company sizes the enriched lung group at about 45% of the non-squamous cases found each year in the U.S., which it says cause more than 90,000 deaths a year.⁷ I did not find a statement about Iovance from a named fund manager or analyst in a document I could check, so I cite none.

Against reality

If lung works and the FDA accepts the enriched group, and the melanoma business keeps growing, the price is possible: with the lung label certain, my base value is $8.06, about the $8.02 price. If first-line melanoma is also approved it is $9.85. Today the company has 22 retrospective patients for that argument, a cohort that has not enrolled its first patient, and a regulator that has not said what it will accept. Revenue is on track: $415M this year against $264M last. The profit is not: the owner cash is −$343M a year, and in my base case it stays negative through 2028.

What a 10x would take. A tenfold price is $80.20. On 494M diluted shares that is a market value of $40B, about 10 times today’s. At 20 times earnings it needs about $2.0B of net profit, and at a 40% net margin about $4.9B of revenue. If that took six years, to 2032, it is about 12 times the 2026 guidance, and 1.8 times my bull case for 2032 of $2.7B. It would also need the stock to rise to about 1.5 times its all-time closing high of $52.59 in 2021 C (calculated).¹ I do not think that is plausible from here. A company with one product in one cancer, losing money and diluting, would have to become one of the largest cell therapy businesses ever built. I did not count how often companies of this size have done it, and I would put it at a few percent at most E (estimate).

Where I come down

The revenue moved me: guidance was raised in September, the margin is rising and the loss is falling. The fivefold rise in the share price did not, and the 29 October lung data made my lung story more specific and less certain. Prospective lung data near 40% response with a clear FDA path, or an early first-line win, would raise my values. They stay at $0.38, $5.73 and $22.29, and the price is above my base.

VIIIRisks, and what would change my view

The risks I weigh most are in the valuation: the confirmatory trial missing, lung not working, and the company paying for both by selling shares. Others matter too. All commercial product comes from one plant in Philadelphia. A contamination event, a fire or a quality failure there would stop sales, and the 10-K and the company’s releases both name manufacturing at that plant as a risk.⁹⁷ The treatment can kill: 7.5% of patients in the label study died of treatment-related causes, and the boxed warning applies to every patient.¹¹ A change in the way cells are made, or a safety signal in real use, could narrow the label. Payment is a risk: the 10-K describes proposals to cap U.S. list prices, and the IRA provisions on drug price negotiation.⁹ Ex-U.S. growth has been slower than hoped: the European application was withdrawn.⁹ The chief executive is interim. And the company depends on one raw material, IL-2, through its own Proleukin.⁹

My view would change in either direction. A weak lung cohort, a miss in the first-line trial or another low-priced share sale would lower it. A strong prospective lung result with FDA alignment, an interim first-line win and a permanent chief executive with a plan to cap dilution would raise it.

Scenarios, and what they do to the base

The table in the valuation section reruns the model for single changes. It does not add them up. The price already carries a good outcome on the pipeline. It does not carry any of these.

What I could not find

The Q3 2026 results are due on 5 November, after this page, so my September cash and share count are estimates.¹⁵ I did not find net revenue per patient or patients treated per quarter in a primary source, so both are my estimates. I did not read insiders’ Form 4 filings. I did not find named fund-manager or analyst views I could check, and I did not check the current plans of other TIL developers such as Adaptimmune or Lyell. I do not know what the FDA will accept for lung, or why the interim chief executive has served so long. These are the points where my numbers are weakest.