Beam Therapeutics, is this for real?

If you follow biotech at all, you probably saw the headline in February. A mid-cap company that's been quietly working on base editing for years just reported clinical trial results that sent investors and KOLs reaching for the same word: transformative. In 29 patients with a genetic disease that has no good treatment options, a single intravenous infusion corrected the underlying mutation, raised functional protein above the protective threshold, and reduced the toxic version of the protein by 84 percent.

So in this edition, I want to take a closer look at Beam Therapeutics and ask the question: is this for real?

Founded in 2017 by David Liu, Feng Zhang, and J. Keith Joung, three of the most prominent figures in modern molecular biology, Beam was built with a straightforward but ambitious vision: to provide lifelong cures for patients suffering from serious diseases. The company has spent the better part of a decade building a fully integrated platform for precision genetic medicines, with internal capabilities spanning gene editing, delivery, and manufacturing. At the center of it all is a suite of gene editing technologies anchored by base editing, a proprietary next-generation CRISPR approach designed to make precise single-base changes in the genome without cutting the DNA.

The reason base editing matters comes down to mechanism. Traditional CRISPR-Cas9 works by introducing a double-stranded break in the DNA at a specific location, then relying on the cell's own repair machinery to stitch it back together. That repair process is messy. The cell joins the two broken ends but doesn't particularly care about the exact sequence, often inserting or deleting bases along the way. The result is a heterogeneous mixture of editing outcomes, and the entire process activates DNA damage response pathways that stress the cell.

Base editing solves this by keeping the precision of CRISPR while removing the cutting. Beam fuses a modified Cas9 enzyme to a deaminase that performs precise chemistry on the genome, converting one base into another, an A to a G or a C to a T. No double-stranded breaks. No reliance on the cell's repair machinery. Just predictable, consistent edits across the population of cells being treated, with less stress on those cells and higher chances they regain functional status after the edit. For most of Beam's history this technology was a story stock, trading on the promise of what base editing could become. That changed when Beam decided to take BEAM-302, an in vivo base editor for alpha-1 antitrypsin deficiency, all the way through clinical development itself.

BEAM-302 is the company's lead in vivo program and the first real clinical test of whether base editing can deliver on its scientific promise. The drug is a lipid nanoparticle carrying two cargoes: an mRNA encoding a base editor protein, and a guide RNA that directs the editor to the exact site of the disease-causing mutation. A single intravenous infusion delivers it to the liver, which is the most accessible organ in the body for systemic drug delivery because hepatocytes naturally take up LNPs via the LDL receptor pathway. Inside the cell, the editor chemically converts the wrong base back to the correct one. The mRNA degrades within days. The DNA edit is permanent.

The disease BEAM-302 is targeting is one of the cleanest test cases imaginable for an in vivo gene editor. Which brings us to AATD.

Alpha-1 antitrypsin deficiency, or AATD, affects roughly 100,000 Americans and is caused by a single misspelled letter in the SERPINA1 gene. That single G-to-A mutation produces a protein, AAT, that misfolds as it's being made in the liver. The misfolded version gets stuck inside hepatocytes instead of being secreted into the bloodstream.

This produces two organ failures from the same mutation. The liver fills up with toxic polymers of the misfolded protein, driving progressive fibrosis and cirrhosis, sometimes ending in transplant. The lungs go unprotected from neutrophil elastase, the protease that AAT is supposed to inhibit, leading to early-onset emphysema. About half of severe AATD patients develop significant lung disease. A meaningful fraction develop liver disease serious enough to require transplant.

The standard of care for the lung is augmentation therapy: weekly intravenous infusions of plasma-derived AAT for life. It runs roughly $100,000 a year per patient. It slows lung function decline but doesn't stop it. It does absolutely nothing for the liver, which has no approved therapy at all. And it can't mount what's called an acute phase response, the natural three- to fivefold increase in AAT levels that healthy people experience during infection or inflammation, exactly when the lungs need the most protection.

A typical AATD patient on augmentation therapy sits in an infusion center every week of their adult life, often with a port surgically implanted in their chest to spare their veins from collapse, while their liver disease progresses untreated. This is the standard of care.

Which brings us back to that data readout. In Beam's Phase 1/2 study, 29 patients received a single intravenous infusion of BEAM-302 across multiple dose levels. At the optimal 60 milligram dose, mean total AAT reached 16.1 micromolar, well above the 11 micromolar protective threshold. Toxic Z-AAT was reduced by 84 percent. Newly produced functional M-AAT comprised 94 percent of circulating protein, a composition that actually exceeds carrier individuals who don't develop disease. Across 21 single-dose patients in the lung disease cohort, there were zero grade three or higher adverse events, and no dose-limiting toxicities.

But the data point that stops you in your tracks is the inducibility patient. One man in the 60 milligram cohort developed a respiratory infection roughly eight months after his single infusion. His total AAT levels spiked from a steady-state of about 16 micromolar to nearly 30 micromolar during the infection, then settled back down as he recovered. Through the entire spike, his AAT composition remained 95 percent corrected M-AAT.

What that single patient demonstrated is that the edited gene is operating under normal physiologic regulation. When his body needed more protein, his body made more protein, and the protein it made was functional. Augmentation therapy fundamentally cannot do this because it delivers a static dose. Beam edited the source code. The cell does the rest.

In January, Beam announced alignment with the FDA on an accelerated approval pathway for BEAM-302 based on biomarker endpoints, specifically AAT levels and the composition of corrected M-AAT versus mutant Z-AAT. That alignment matters for two reasons.

The first is practical. The pivotal trial design is now clear, the endpoint is agreed, and Beam can move from Phase 1/2 directly into the registrational study without regulatory ambiguity hanging over the program. A successful pivotal sets up a BLA submission rather than another round of negotiation with the agency about what the bar should be.

The second reason is philosophical, and it speaks to where the FDA is heading. For therapies that directly address the underlying genetic cause of a disease, measuring the proximal molecular impact of the treatment, how much functional protein is being produced, how much mutant protein has been eliminated, is increasingly being accepted as the right framework for approval. The FDA's recent plausible mechanism pathway guidance reflects exactly this line of thought. For gene editing platforms targeting monogenic diseases, this is the regulatory infrastructure being built for them.

The most credible competitor in AATD is Wave Life Sciences and their drug WVE-006, which targets the same disease on a similar regulatory timeline. Wave's approach is fundamentally different from Beam's in a way that's worth understanding.

WVE-006 is a GalNAc-conjugated oligonucleotide injected subcutaneously every month. It enters hepatocytes the same way Beam's LNP does, then binds to the SERPINA1 messenger RNA at the mutation site and recruits a naturally occurring cellular enzyme called ADAR to chemically convert the mutant base on the mRNA. The corrected mRNA gets translated into functional protein. The advantage Wave is positioning is that RNA editing is reversible. If something goes wrong, the patient can stop dosing and the editing effect wears off. Permanent DNA modification doesn't offer that.

There are real benefits to this approach. Subcutaneous self-administration at home is genuinely easier than IV infusion. GalNAc is a more validated delivery platform than LNPs, with a longer safety track record. And the chronic dosing model spreads cost over time in a way that's easier for payers to absorb than a multi-million dollar upfront payment.

But the data favors Beam on every metric that matters clinically. Wave's corrected M-AAT composition is 64 percent versus Beam's 94 percent. Their total AAT is 11.9 micromolar versus Beam's 16. Their Z-AAT reduction is 71 percent versus 84. The structural reason these numbers are lower is that Wave edits mRNA, not DNA. The hepatocyte's genome remains PiZZ, so it keeps producing mutant mRNA forever. Wave's drug catches a fraction of that mRNA before the ribosome reads it, but ADAR is a finite cellular resource. You can never catch all of it.

The reversibility argument is also less compelling than it first sounds. For a chronic genetic disease, reversibility means lifelong dependence on monthly subcutaneous injections. Miss doses and the disease comes back. Most patients facing a lifetime of disease would prefer a single treatment to a monthly injection forever, all else equal.

Beam plans to initiate the pivotal trial for BEAM-302 in the second half of 2026, with the FDA aligned on the biomarker-based pathway and the trial design defined. The pivotal expansion enrolls roughly 50 additional patients at the 60 milligram dose with a 12-month primary endpoint. If the data replicates the Phase 1/2 profile, the path to BLA submission opens. Sell-side peak sales estimates for BEAM-302 in AATD alone run around $2-4 billion, with significant upside if diagnosis rates increase post-approval the way they did when augmentation launched in the 1980s.

But the reason I'm interested in this company isn't BEAM-302 by itself. BEAM-302 is the proof-of-principle for an in vivo liver editing platform, not a single asset. The same LNP delivery system, the same base editor architecture, and the same manufacturing process extend to other monogenic liver diseases by simply changing the guide RNA. Beam is already doing this. BEAM-301 in glycogen storage disease type 1A is in Phase 1/2. BEAM-304 in phenylketonuria is targeting an IND filing this year, and Beam advanced that program from inception to IND-enabling activities in under two years. The cadence of the pipeline is the platform thesis made concrete.

What makes this interesting from an investor's perspective is that BEAM is largely being valued on the AATD program alone, meaning the broader platform pipeline and the regulatory tailwind are essentially a free option embedded in the current price.

The real question now is whether BEAM-302 is just the best AATD therapy ever developed, or the beginning of something much larger.

Here is what I am into:

What I am reading a lot about
The World Cup kicks off next week, and I'm finding it strange to hold two feelings at once — excited that the biggest sporting event in the world is being co-hosted here for the first time in 30 years, and frustrated that the lead-up has been such a mess. Two pieces that captured both sides for me: Will Leitch in New York Magazine on staying excited about it, and Jonathan Lemire in The Atlantic on why everything feels off.

A podcast I would recommend to anybody
Olivia Rodrigo sat down with the New York Times Popcast crew for an hour and a half this week, and it is rare that a pop star at her level of fame opens up that freely about not just a new album but the criticism that comes with the job. Worth a listen even if you do not follow her music.

What is alarming to me
Cars are now one of the leading causes of death globally, killing roughly as many people each year as tuberculosis, and they are the single biggest killer of people in the prime of their lives. The thing that struck me most reading this Vox piece is that the burden falls overwhelmingly on low- and middle-income countries, and unlike most public health crises we have actually made progress on, this one has stayed roughly flat for the last two decades.

Thanks for joining me for this edition. As always, I would love to hear your thoughts, comments and feedback. You can reach me at [email protected]

Yours truly,
Mark Gad