Back of the Napkin Bios

AbCellera: The Bubble Proxy

Back of the Napkin Bios — Wed, 22 Apr 2026 12:31:17 GMT

I‘m going to keep this one short, because the thesis is relatively simple and the key data isn’t here yet. I will be adding this one to the watchlist (which is a list that continues to grow — yes, QURE, TWST, CLPT, CMPS, GHRS etc are keeping me busy these days) and I plan to update the list monthly. However, think of this as a name to have on the radar that works on its own merits, but works considerably better as the XBI gets bubbly again. If you’re looking for a clean proxy for the next biotech cycle with a real catalyst embedded inside it, AbCellera (NASDAQ: ABCL) belongs on your list.

There’s a version of this trade that works without a biotech bubble and there’s a better version that works with one. AbCellera is currently mispriced against both scenarios and the calendar is about to start clarifying which world we’re living in.

Legend Biotech: The Kelonia Read-Through

Back of the Napkin Bios — Mon, 20 Apr 2026 16:25:26 GMT

This will be a “best ideas only” bio substack providing actionable, highly digestible and high conviction value to investors focused on quality not quantity. Disclaimer: Nothing in this article (or substack) is considered investment advice and is solely the opinion of the author.

*PT Below*

What Lilly Did This Morning

Eli Lilly announced it is acquiring Kelonia Therapeutics for $3.25 billion in cash upfront, with up to $3.75 billion in additional milestones, a potential $7 billion deal. Kelonia’s lead program, KLN-1010, is an in-vivo BCMA CAR-T for multiple myeloma. It is in Phase 1, with first clinical data presented at ASH in December 2025.

That is the factual anchor. Whatever conclusions one draws from it, the reference point for the rest of this post is that a large-cap pharma was willing to write a $3.25B upfront cheque for a pre-commercial, Phase 1 BCMA CAR-T program earlier today.

For context, Legend Biotech (NASDAQ: LEGN), which owns 50% of the profits of the only approved BCMA CAR-T on the market (a product doing approximately $1.9 billion in 2025 global sales) closed last Thursday at a market cap of roughly $3.9 billion. LEGN is up roughly 11% today on the Kelonia news, which brings market cap to somewhere in the $4.3 billion range.

Whether that is a pricing error or a rational discount for real risks is the question worth working through.

The Platform Compounders: Biotech’s Best Buy-and-Hold Names

Back of the Napkin Bios — Thu, 16 Apr 2026 23:56:21 GMT

The Test I’m Running On Each Name

I’m not giving you every “platform-ish” biotech. I’m filtering for companies where the technology generates candidates more efficiently than the industry baseline, and this efficiency is demonstrated, not claimed.

The marker of a real platform is that each successive program costs less time and money per IND than the last, and targeting a new disease is a matter of redirecting the same primitive rather than starting discovery from scratch.

A genuine platform-expansion compounder needs all four of these, not three:

An Important Nugget From NORD

Back of the Napkin Bios — Wed, 15 Apr 2026 16:23:00 GMT

In 2012, the FDA established its Patient-Focused Drug Development (PFDD) program under the Prescription Drug User Fee Act. The premise was straightforward: patients and caregivers have direct knowled…

argenx: The TAM Is About to Double

Back of the Napkin Bios — Fri, 10 Apr 2026 12:30:27 GMT

*PT Below*

Outside of NVO (which I wrote about a couple of days ago) argenx (ARGX) may be the most compelling (aka sleep-easy compounding) large-cap biotech setup I can identify right now. Not because the stock is cheap (it isn’t on conventional metrics) but because the denominator is about to change dramatically, and the market is still pricing this as though Vyvgart’s growth story is mature.

Let me walk through three things: the TAM inflection, the pipeline architecture, and what I think this company is actually worth in five years.

NRIX Update: Bexdeg Stays on Track

Back of the Napkin Bios — Thu, 09 Apr 2026 20:08:23 GMT

I covered NRIX in depth last fall, so this is a quick catch-up rather than a full re-introduction. The short version: the thesis is largely intact, execution is proceeding on the timelines we had, and the more interesting incremental development is that the inflammation/autoimmune angle is becoming harder to ignore.

What’s New on the Clinical Side

The registrational Phase II in CLL continues enrolling, with the goal of completing enrollment by year-end 2026 and data in 2027. Patient finding remains a known friction point (this is a heavily pre-treated population) so that’s worth watching. The confirmatory Phase III in 2L+ CLL is expected to initiate by mid-2026, which is on track with prior guidance.

Novo is a Buy Right Now

Back of the Napkin Bios — Tue, 07 Apr 2026 12:31:32 GMT

This will be a "best ideas only" bio substack providing actionable, highly digestible and high conviction value to investors focused on quality not quantity. Disclaimer: Nothing in this article (or substack) is considered investment advice and is solely the opinion of the author.

This will be a quick note as I recently read Left’s post on the new Wegovy label and then I looked at the stock chart. Everyone decided Lilly won the GLP-1 wars sometime around mid-2024 (hence why the stock is down ~60%) but I'll keep it short on why that conclusion is premature, and why the inflection is happening right now.

Proteostatic Valuation

Back of the Napkin Bios — Mon, 06 Apr 2026 17:39:50 GMT

The standard AMT-130 bull case is a huntingtin-lowering drug in a rare disease. Approve it, price it like Zolgensma, collect the orphan drug premium, done. That model produces a number somewhere around ~$100 per share and requires you to believe the FDA eventually does its job.

That’s probably the wrong model entirely.

Another Invitation to the Pain Trade

Back of the Napkin Bios — Thu, 02 Apr 2026 16:21:33 GMT

PAX – Group | Portfolio Update (Chart: Cumulative Return Oct ‘23 – Mar ‘26)

Cumulative = +571.38% | YTD = +8.1% | 52-Week = +125.46%

Thirty to Sixty Days

Back of the Napkin Bios — Wed, 25 Mar 2026 01:45:44 GMT

A new draft FDA guidance was sent to the White House for review on March 23rd. The title: “Leveraging Prior Knowledge in the Development of Human Gene Therapy Products Incorporating Genome Editing.”

Read that again.

Leveraging Prior Knowledge. In gene therapy development. Incorporating genome editing.

That is the evidentiary architecture of the entire AMT-130 program (Enroll-HD natural history data, external controls, prior clinical knowledge from an already-enrolled patient cohort) being potentially codified into formal FDA guidance. Coincidentally, it appears to be moving now, in the transitional window between Prasad’s departure and the normalization of CBER leadership, with career staff back at the wheel.

Here is what I know, and what it means for the thesis for AMT-130 and other related gene therapies.

The Paper No One Is Paying Attention To

Back of the Napkin Bios — Fri, 20 Mar 2026 12:03:38 GMT

If you read my HD primer, you came away understanding that Huntington’s disease is not a simple story of a bad protein causing bad things. It’s a cascade of a molecular domino sequence that begins with a CAG repeat expansion and ends, years or decades later, in the death of medium spiny neurons in the striatum. The question the field has spent forty years failing to answer cleanly is: which domino is the critical one?

A couple of days ago, two papers published simultaneously in Science Translational Medicine (one of the most prestigious translational research journals in existence) give the clearest answer to date. If you’ve been following the AMT-130 thesis, the answer should stop you in your tracks.

The Sensitivity Standard

Back of the Napkin Bios — Thu, 19 Mar 2026 12:02:37 GMT

*BOTN NTM Price Target at the end of the article*

The liquid biopsy market is about to bifurcate. On one side: the high-volume, broad-panel, targeted-gene-count assays that have dominated commercial oncology testing for the past five years. Think Guardant360, FoundationOne Liquid, the panel players that read a few hundred genes and report back actionable mutations for therapy selection. These are useful tests but they are also, increasingly, commoditizing. Price competition is intensifying and differentiation is narrowing as reimbursement rates are established but not expanding. On the other side: a smaller set of platforms attacking a completely different clinical problem with a fundamentally different technical approach. Not broad-panel therapy selection but ultra-sensitive, tumor-informed molecular residual disease detection.

The 2026 Genomics Stack: Follow the DNA

Back of the Napkin Bios — Sat, 14 Mar 2026 12:30:40 GMT

There’s a version of this post that is just five separate company write-ups stacked end-to-end. You have seen that version from every sell-side desk covering the space. It gives you a bull case, a bear case, and a price target for each name in isolation and then it stops, leaving you to figure out how they interact. That last part is, inconveniently, where most of the actual investment insight lives. The names are table stakes but the architecture is the alpha.

What I want to do here is something harder and considerably more useful: map the load-bearing dependencies between these companies — the specific, concrete ways in which one company’s success is structurally necessary for, or additive to, another’s. How a dollar of AI-driven drug discovery demand entering the system at the synthesis layer propagates forward through sequencing infrastructure and eventually reaches clinical diagnostics revenue with a lag of 18 to 24 months that creates systematic mispricings if you know where to look. How the competitive skirmish between ILMN and TXG in spatial genomics is being analytically misread by the market in ways that are simultaneously hurting both valuations when the correct interpretation actually strengthens both theses. How BillionToOne’s sensitivity advantage (which the market is treating as a standalone technology story) is structurally downstream of the same platform investments that 10x Genomics and Illumina have been making for the better part of a decade, and how that matters for thinking about the defensibility of BillionToOne’s position over time. How GRAIL’s FDA outcome (whatever it is) does not exist in isolation from the broader liquid biopsy reimbursement landscape and why a GRAIL approval is actually a commercial accelerant for BillionToOne rather than a competitive threat.

The Prasad Shield Is Gone and Makary Is Next in the Crosshairs

Back of the Napkin Bios — Tue, 10 Mar 2026 17:34:43 GMT

For months, FDA Commissioner Marty Makary operated with a kind of political insulation. When things went sideways in the biologics world — Huntington’s patients left without a therapy, rare disease companies blindsided by reversed guidance — there was always someone to absorb the heat. That someone was Vinay Prasad.

That shield is now gone.

With Prasad’s departure from CBER announced last Friday, the deflection strategy has run out of runway and Congress is just getting started. Senate Republicans who spent the fall politely writing letters are now talking subpoenas. The White House, which already forced one Prasad reversal on camera, has watched the political fallout from the QURE situation land squarely on the commissioner’s desk. And on the same day all of this is coming to a head, the FDA announced an approval that perfectly captures everything that is broken about Makary’s leadership: leucovorin (a drug he and Trump publicly promised would help “hundreds of thousands” of kids with autism) was approved Tuesday for a condition affecting fewer than one person per million. The gap between the promise and the reality could not be wider. Makary is now the last man standing and the walls are closing in.

The Poster Child of Prasad's War on Rare Disease

Back of the Napkin Bios — Tue, 10 Mar 2026 02:35:26 GMT

Last Friday, the rare disease biotech world caught a breath of fresh air.

FDA Commissioner Marty Makary sent a note to staff confirming that Vinay Prasad — the controversial CBER director whose tenure became synonymous with goalpost-moving, surprise reversals, and a overt hostility towards rare disease programs — would be departing the agency at the end of April. By Monday morning, ATRA was up roughly 29%, QURE surged 26%, RGNX climbed nearly 20%. The market was voting, loudly, on what it thought Prasad’s presence had been worth to this space.

CLPT: The Pendulum Swung Too Far

Back of the Napkin Bios — Sun, 08 Mar 2026 01:29:42 GMT

Many astute investors know markets overshoot — that's not a novel observation. Occasionally, however, the overshoot is precise enough, and the underlying business durable enough, that the entry point becomes obvious in retrospect. I think that's where Clearpoint Neuro is today. I have written (and pounded the table) about ClearPoint Neuro previously, and on Friday the business traded at the lowest valuation it has seen since the bottom of the biotech trough.

The View From the Trenches

Back of the Napkin Bios — Fri, 06 Mar 2026 18:18:20 GMT

STVN: The Compounder Quietly Doing What Compounders Do

Back of the Napkin Bios — Wed, 04 Mar 2026 13:09:27 GMT

The original mandate of my original post was simple: find businesses that compound quietly (particularly during an inflection), expand margins predictably, and reward patience. Stevanato Group just reported fiscal year 2025 results, and it’s a reminder of why STVN belongs in that category.

Full-year revenue grew 7% (9% at constant currency) to €1.186 billion. Adjusted EBITDA margin expanded 160 basis points to 25.1%. Adjusted diluted EPS grew 13% to €0.54. None of those numbers are going to break the internet but it’s important to look underneath them.

"The Setback Isn't What You Think It Is"

Back of the Napkin Bios — Mon, 02 Mar 2026 19:43:18 GMT

Updated PT Below

In the span of ten weeks, the FDA published four documents that collectively dismantled every argument for requiring a randomized controlled trial in AMT-130’s approval pathway. A December RWE guidance celebrating large natural history databases like Enroll-HD and explicitly removing barriers to their use in regulatory submissions. A January Bayesian methodology framework published in the Federal Register, specifically designed for diseases where traditional RCTs are difficult or impossible to run, endorsing external information borrowing, propensity score methodology, and posterior probability success criteria. A February 19 NEJM article co-authored by Commissioner Makary and CBER Director Prasad declaring the two-trial dogma dead, explicitly naming accelerated approval and rare disease as the primary beneficiaries of the new policy. In addition, a February 23 plausible mechanism pathway guidance jointly issued by CBER and CDER describing in precise detail the evidence package required when RCTs aren’t feasible — a description that maps onto AMT-130’s data package with uncomfortable precision.

Huntington’s Disease: A Primer

Back of the Napkin Bios — Tue, 24 Feb 2026 04:14:17 GMT

The Disease

In 1872, a 22-year-old physician named George Huntington published a paper describing a hereditary condition he had observed in families on Long Island — involuntary writhing movements, psychiatric deterioration, inevitable death, passed from parent to child with merciless consistency. William Osler later called it one of the most accurate portraits of a disease ever written. Huntington correctly identified the inheritance pattern two decades before Mendelian genetics was even rediscovered.

The gene responsible wasn’t found until 1993, located on chromosome 4 by a team working with over 18,000 individuals from a single extended family in the fishing villages around Lake Maracaibo, Venezuela — communities where the disease was so prevalent it was simply called “el mal” — the evil — and accepted as an inescapable part of life. That discovery became the first autosomal disease gene found through genetic linkage analysis, and the methods developed to find it became foundational tools for the Human Genome Project.