Gene Editing
The company commercializing prime editing — a "search-and-replace" gene editor that can fix a mutation without cutting both strands of DNA. Three programs are now cleared for or in the clinic, and there's barely enough cash to get them there.
Personal notes published for reference. Figures are sourced from company filings, releases, and press coverage as of the date above — the interpretation is mine and may be wrong.
CEO Allan Reine has spent the last year doing the unglamorous part: cutting headcount, consolidating facilities, narrowing the pipeline to liver-directed in vivo programs plus the CGD asset, and winning the Beam arbitration. The science comes from David Liu's lab at the Broad Institute, where prime editing was invented. It's a disciplined team running a narrow plan, not a visionary one — which is what a company with this balance sheet needs.
Prime editing can in principle correct most known disease-causing mutations — the broadest toolkit in gene editing. In practice the TAM is built one rare disease at a time: ~100,000 PiZZ carriers in the US for AATD, a few thousand Wilson patients with the H1069Q variant, and a very small CGD population. The platform ceiling is huge; each individual program is small.
The weakest of the four, and the one that fails my test. $108.8M of cash at June 30, 2026, down from $191.4M at year-end, against a $42.1M quarterly net loss. Management guides runway "into 2027" — which means a raise is coming before the 2027 data readouts, not after. There's essentially no revenue ($4.6M in 2025, collaboration income).
Better than the stock suggests. PM359 produced the first-ever human prime-editing data (published in NEJM) and holds RMAT, Fast Track, Orphan, and Rare Pediatric designations. Wilson disease (PM577a) cleared its IND and a New Zealand CTA in July, and AATD (PM647) cleared its IND on Sept 24, 2026. Every regulatory milestone management set for 2026 has landed.
Three shots on goal, all with 2027 catalysts. The two liver programs share a single lipid-nanoparticle delivery system, so data from one partly de-risks the other.
An ex vivo prime-edited stem-cell therapy for p47phox-deficient chronic granulomatous disease, an immune disorder that leaves patients defenseless against bacterial and fungal infections. Both treated patients showed rapid engraftment and durable restoration of NADPH oxidase activity with no safety concerns. After FDA conversations, Prime believes those two patients may support accelerated approval — a BLA is targeted for 1H 2027. If it works, it's the first approved prime-editing therapy anywhere. The market is tiny; the precedent is not.
An in vivo, one-time IV infusion that corrects the H1069Q variant of ATP7B — the most common Wilson mutation in the US — in the liver. IND cleared by FDA and CTA approved in New Zealand; a global Phase 1/2 is starting in 2H 2026 with first data in 2027.
The biggest commercial program. It corrects the Pi*Z (E342K) mutation in SERPINA1, restoring normal M-AAT protein, using the same liver LNP as PM577a. The IND was cleared Sept 24, 2026 for a single-arm Phase 1/2 with ascending doses of a one-time infusion. It starts in lung-only AATD patients and adds a liver-disease cohort once tolerability is shown. First data in 2027. Rights were secured in July when an arbitration tribunal ruled Prime "did not breach" its agreement with Beam Therapeutics.
A prime-edited CAR-T collaboration with Bristol Myers Squibb in hematology/oncology, and an in vivo cystic fibrosis program backed by the Cystic Fibrosis Foundation. Neither drives the thesis, but both bring in non-dilutive money and outside validation.
This is the whole bear case, and it's arithmetic rather than opinion.
$108.8M in cash at June 30 and roughly $40M+ a quarter of burn gives about 2.5 quarters before cost cuts, which is why management says "into 2027" and nothing more specific. The BLA filing, the Wilson data, and the AATD data all land in 2027. So the company has to raise before it has the data that would let it raise on good terms.
At a ~$516M market cap and $2.90 a share, a raise large enough to reach those readouts (~$150–250M) means roughly 25–50% dilution unless it comes through a partnership. The AATD IND clearance is exactly the kind of catalyst a company raises into.
A pharma partnership on PM647 — AATD is a large enough market to interest big pharma — with upfront cash in place of equity. The BMS relationship shows Prime can sign these. A licensing deal before a raise would change the per-share math completely.
The size, price, and structure of the next financing, and whether it comes with a partner. Also any FDA alignment update on the two-patient CGD BLA — confirmation there would be the single biggest de-risking event before 2027.
A small-cap biotech's shareholder base tells you who'll write the check at the next raise, so it matters more here than for a self-funding name.
T. Rowe Price, ARK, Sumitomo Mitsui Trust, and Nikko Asset Management lead a fragmented holder list with no dominant anchor. Specialist biotech funds that usually lead insider-friendly financings aren't prominent at the top, which makes a clean raise harder.
Trackers disagree sharply for this name — anywhere from ~44% to ~69% institutional depending on source and date. Treat the headline number as directional.
A binary-ish 2027 with a financing problem in front of it. The science is working; the question is how many shares it costs to find out.
If this plays out: ~$1.50
The company raises near the 52-week low with no partner, diluting existing holders heavily. The FDA then declines to accept a two-patient package for CGD and asks for more patients, which pushes approval out years. The LNP programs take until late 2027 to read out, and gene-editing sentiment stays cold. The stock has already fallen from $6.94 to $2.90 over the past year, and nothing about the cash position stops it from going lower before the data arrive.
If this plays out: ~$5
A 2026–early 2027 raise dilutes holders meaningfully but funds the company through its readouts. The CGD BLA gets filed on schedule and early Wilson and AATD data show editing in the liver with clean safety. Each of those would re-rate the platform, since they prove prime editing works in vivo in humans. That lands near the low end of the ~$6–7 analyst consensus once the extra shares are counted.
If this plays out: ~$11
A pharma partner licenses PM647 for real upfront cash, which removes the financing overhang without heavy dilution. PM359 becomes the first approved prime-editing therapy, validating the platform with regulators. Strong 2027 AATD data then turn the company from a rare-disease platform into one with a large indication. The street-high $11 target is what that looks like; it's more than 3.5x today's price.
$2.90 at the Sept 25, 2026 close (52-week range $2.67–$6.94, market cap ~$516M), near its lows despite the PM647 IND clearance the day before. Consensus is a Buy from 13–14 analysts with an average target around $7.02–7.11 (range $4.25–$11). A gap that wide shows the market is pricing the financing risk and the sell side is pricing the pipeline. The scenario targets above are my own rough, directional estimates, not formal price targets. They're anchored to the sell-side range rather than a full model.
Runway only "into 2027" with every major catalyst in 2027. A dilutive raise is close to certain; the only open questions are its size and terms.
A two-patient accelerated-approval BLA tests how flexible the FDA is willing to be. A request for more patients would push PM359 out by years.
Both liver programs ride on the same LNP. A safety signal in either one would likely hit both.
Beam (base editing), Intellia (CRISPR), and others are chasing the same liver diseases, including AATD, with more cash. Prime editing is more versatile than they are, but that doesn't make it first to market.
CGD is ultra-rare. Even an approval there is mainly a proof point, not a revenue engine. The revenue case runs through AATD and depends on 2027 data.