NKGEN BIOTECH – (NKGN – NASD) LOVE THE SCIENCE, HATE THE BALANCE SHEET

Last Thursday (one week ago today) NKGen Biotech (NKGN – NASDAQ) announced some VERY promising Phase 1 clinical data in a (very) small trial, many patients of late stage Alzheimer Disease (AD) patients.  The headline was:

“90% of patients demonstrated improvement or maintained stable cognitive function as per Alzheimer’s Disease Composite Score (ADCOMS) following 11 weeks”.

A few years ago, this kind of data would have sent the stock TO THE MOON. NKGen would have been able to easily raise funds, would have likely done it the same morning. All would be well.

Today? Not so much.

Yet the data and the indication here are both really good. Targeting AD has HUGE potential.

Alzheimers Disease (AD) is one of the biggest and highest unmet-need markets there is. A winner in the space is easily a $1 billion+ drug.  

Source: NKGen October Presentation

In Sept 2022, Biogen (BIIB-NASD) added $10 billion in market cap –or $70/share—in one day as its Phase 2 data showed its Leqembi drug somewhat effective against Alzheimers.

I would like to own a stock that goes up $70 in a day. So it’s definitely worth trying to understand this company and this stock! NKGN has a market cap of US$78 million and with $20 million (though only $5 million of it is 3^rd party), an Enterprise Value of roughly $100 million.

And the technology that NKGen uses is differentiated from other companies targeting the disease. They are going about this in a unique way, and have shown that it works. The FDA DID approve NKGN for a next Phase 1/2a trial.

It all lines up – I look at the data and I feel almost compelled to buy the stock. Except the market sucks. And they need cash—immediately:

You rarely see a biotech say something this grim in their quarterly financials:

Source: October 19th S-1 Filing

Oh boy.

Is the science good enough to overcome the public company woes? Let’s find out:

 THE SCIENCE LOOKS GOOD

First, understand that this market is a biotech winter.

That means:

1. no one is paying attention.
2. no one cares about good news.
3. stocks won’t move even when they should.

NKGN’s approach is to go after the source of AD using Natural Killer cells (hence the NK in the name.

Source: NK Gen Biotech Investor Presentation

Natural Killer (NK) cells are one of the fighters that your immune system uses to attack pathogens. They are a white blood cell (technically called a lymphocyte) that is always “on patrol” in your blood stream, about 15% of all white blood cells. They are a first line of defense, ready to attach to foreign cells they come across and destroy them.

With AD, there is known connection with a certain type of protein that accumulates as plaque in the brain.

This protein is a foreign cell. As such, the body starts manufacturing antibodies (called T-cells) to get rid of them. The problem comes as the T-cells do as much harm as good by attacking the brain and causing inflammation. That inflammation leads to all the symptoms we see from AD.

You really have two problems: the foreign proteins and the T-cells being created to stop them.  NK cells can deal with both.

NK cells can identify and attack the “rogue” T-cells. They can also knock off the proteins themselves, effectively cleaning house of both the “dirt” and the “janitor”.

The real promise of natural killer cells? They not only attack foreign cells, but they do it without a lot of side effects. Natural killer cells are very good at distinguishing between a foreign cell and a healthy cell, so they don’t cause the immune system to go haywire on itself.

NKGen gets these natural killer cells from a sample of the patients blood.

A drawback of NK cells is that they are hard to multiply outside of the body. If you get too many together, they start to attack each other. This has been a limitation for other companies, but NKGen has solved this by developing a genetically engineered feeder cell that is mixed in the petri dish with the NK cells.

The feeder cell becomes the “prey”. With the NK cells too busy attacking feeder cells to attack one another, they multiply in greater quantities.

The feeder cells also are designed to draw out more activating receptors from the NK cells. This makes the NK cells more “aware”, so that they are better at attaching to a foreign cell when they come across one.

Finally, NKGen adds enzymes and proteins into the petri dish mix. These improve the “machinery” of the NK cells, making them more adept at killing their prey.

Putting it all together, NKGen takes normal NK cells from the patient, expands them into even more cells, makes them better at attacking the foreign cells and more able to find them to boot.

You end up with a souped-up natural-killer cell juiced to do one thing: destroy the enemy.

THE DATA LOOKS GREAT

The data that NKGen presented was in very sick patients with poor mental scores. Five had mild Alzheimer’s, three moderate and two severe.

Almost all the patients were stable many improved:

Source: NKGen October 27th Press Release

The 22-week data was obtained for 9 of the 10 patients. After 11 weeks of dosing 9 of the 10 patients were either improved or stable. At the 22 week, point (12 weeks the final dose) 4 of the 9 patients were better than at the end of dosing and 8 were as good or better.

Source: NKGen Cantor Global Health Conference

Given how quickly AD conditions deteriorate this data suggests that the natural killer cells are having an impact.

Even more encouraging, a few of the patients improved even after the end of dosing. How? Because the NK cells are so good at eliminating amyloid proteins that the T-cell population continued to decline in the weeks after treatment, leading to a further reduction in symptoms.

NKGen gave a presentation at the Cantor Global Health Conference back in September where they showed visible evidence of how their treatment has worked on a couple patients. After seeing this, I wanted to buy stock.

CEO Dr. Paul Song showed two remarkable videos of very debilitated AD patients that could not perform the simplest of tasks.

But within weeks of the treatment, both had regained enough cognitive function to do things like eat with chopsticks and dance!

It was a very promising presentation. But understand: this is only 10 patients.

AD trials usually have HUNDREDS of patients. Of course, most trials are targeting the disease at a much earlier stage, so they need a large patient population. Because NKGen is targeting a much later stage (sicker) population, they can get a signal much more easily. The bar is lower.

NKGen’s results were good enough for the FDA to give the thumbs up for a second trial.  The FDA approved a Phase I/Ia study in moderate AD patients.

Whereas the Phase I trial was a dose escalation trial that only gave four total doses over 11 weeks, this new trial will give a higher dose and a prolonged dosing regimen – opening up the possibility for an even better response.

WATCH AND WAIT

A LESSON IN HOW NOT TO DO A SPAC

How did NKGen got into this mess? You guessed it: they came to market as a SPAC—a Special Purpose Acquisition Company that was so popular 2-3 years ago. It’s a blind pool of capital where investors buy shares at US$10—and then get to choose to get a refund on their investment once they find out what private, operating company is being vended into the SPAC.

In March, Graf Acquisition IV and NKGen announced their business combination.

If all had gone according to plan, NKGen would have had an Enterprise Value of $160 million and a minimum of $50 million of cash to fund operations. With no redemptions, NKGen would have had $173 million of cash, enough to fund through 2025.

Instead, the deal succumbed to the fate of many SPACs nowadays – holders chose to redeem their shares at par instead of participating in the de-SPAC transaction.

It all began to unravel in May. At a Special Meeting for the SPAC to extend the time to complete its deal, 11.1 million of the 17 million publicly held shares elected to redeem, and NOT hold on through the transaction. That left only 6.2 million shares, and only $62 million in Graf’s trust account.

The merger was contingent on a minimum $50 million cash requirement for the deal at closing. After fees and whatever other payments, there was not going to be enough cash.

Over the summer the market for biotechs just got worse. Graf likely realized that more holders would flee before the deal was done. Graf was desperate to raise enough cash to get past the finish line.

To do that they made three deals (two of which may be deals with the devil).

First, NKGen’s parent company, NKMax in South Korea, committed $10 million in the form of a four year convertible note with warrants on–relatively favorable terms–with 5% cash pay interest or 8% PIK interest (at company discretion) and the note converting to shares at $10.00 per share.

Second, Graf went to the market and came up with a plan to sell 10 million warrants at a buck a piece, at a time when Graf’s public warrants, with an $11.50 strike price, were trading at about 40 cents per warrant.

These warrants are some of the more complicated pieces of paper I have come across. Here are how their terms are disclosed (I don’t expect you to read this):

Source: NKGen 8K Disclosure

Yikes!

This is an extremely complicated bit of legalese. I would need a page or more to describe it in full. These are the highlights:

• NKGen is issuing 10.2 million 5-year warrants in 3 equal-sized tiers, with strike prices at $10, $12.50 (up 25%) and $15.00 (up 50%).
• The warrants are cash pay – if all warrants are exercised NKGen gets $125 million cash.
• The strike price for the bottom tier of the ladder can be adjusted downward every six months if the average price falls below strike, to a floor of $5.00
• Below the floor things get complicated

The under $5.00 scenario is too complicated to get into here. It essentially limits the upside to the warrant holders from a falling share price (ie. avoid the death-spiral trap) while still ensuring the warrant holders don’t take a loss.

All in, the warrants raise cash but otherwise are not a great deal for NKGen.

The cash raised from the warrants was still not enough to make up for redemptions. So Graf entered into a couple other deals to get past the finish line.

Graf entered into a “Forward Purchase Agreement” with three investment firms.

Two of the firms purchased shares from redeeming shareholders and on the open market before closing and left that cash in the trust. The other firm purchased one million shares in a PIPE (Canadians–PIPE=Private Placement) at $10.44 per share (the Redemption Price) at closing.

All the cash was also placed into escrow – approximately $33 million all-in.

That money going into the escrow is not necessarily NKGen’s to keep. If the shares can be sold above $10.44, NKGen gets the escrow money. If they can’t, the cash reverts back to the investors after a year as if the shares were redeemed in the first place.

This is also not a great deal. It is the kind of deal you make when you have no other deals available. It is almost assuredly the only way Graf could get the de-SPAC through the gate.

There is one positive about the deal: it was done in part to allow NKGen to benefit in the event shares popped quickly at closing, as the non-redeemed shares are registered and available for resale, and NKGen would get $10.44 released for every share sold. That would be the ideal outcome for NKGen, as they will be able to access the entire $33 million in escrow.

(Almost certainly, none of these shares will get sold down here at $3 for a loss when they get made whole at $10.44 in 11 months from now.)

But, at a Special Meeting on September 29th, 3.4 million publicly held shares chose to redeem rather than proceed. And with the SPAC’s chart looking like this, no wonder:

That basically left Graf’s own shares (2.523 million), a paltry 87,162 of public stock, the Forward Purchase Agreement shares and the legacy owners of NKGen (which put in $10 million themselves):

Source: October 19th S-1 Prospectus

The total shares diluted, including the 10 million warrants that adjust with the share price, is about 32 million.

The float is much lower. NKGen is a VERY low-float stock. Almost all the stock is held by Graf, NKMAX, the warrant buyers and the FPA investors. I don’t know exactly how much stock is free-trading, but it’s likely less than a million shares.

Which is another big wildcard here – stock with very little float can move wildly for no reason.

In the end, Graf was able to get the merger done but it came at a big cost–one that still needs to be paid in full. Because NKGen had to pay so much of their proceeds into the escrow account, they don’t have much cash left to run the business.

How much cash?

Source: S-1 Prospectus

According to the disclosure, NKGen has $6 million of available, unrestricted cash. Which means that unless we get a big pop in the stock that allows them to access those escrow funds, NKGen will have to raise cash.

CONCLUSION—WE WAIT TO SEE WHO WINS,

THE SCIENCE OR THE BALANCE SHEET

I talked with Dr. Song on an hour-long Zoom call, and he was confident that they can raise the cash they need without much trouble. I hope he’s right, because this science looks like–at an early stage–that it makes a real difference for patients, and it could make a real big difference for investors.

The good news is that NKGen doesn’t need a lot of cash to get through the next AD trial. The trial size is small, expected to be 30 patients and it will dose for 12-months.

By the way, the fact that the FDA says they only need 30 patients is a huge vote of confidence for NKGN, as well as a signal the FDA really wants to see SOME kind of treatment for this late-stage Alzheimers. These are very sick patients that are not the target of most AD trials which means not a lot of competition.

Dr. Song figures that at most the trial will cost $10 million and more likely it will be in the $7-$8 million range.

But I’m not going to lie, I am worried about what terms NKGen will have to agree to in order to raise cash. This environment is hard. That Graf had to make the deal they made to get the merger done just highlights how few buyers there are.

On the other hand, we have since seen the 22-week data and we have had the FDA approve a second Phase 1/2a trial. There are new reasons to attract buyers.

Whether that is enough to avoid material dilution – I just don’t know. Which means I have to wait. Once I can make a bet here that is on the science and not the balance sheet I will most assuredly take it.

Disclosure: I am NOT long. Yet.