Newsletter · · Ashutosh Agarwal

In Vivo Base Editing Cuts LDL 62% in First Human Data - Biotech Pipeline: Gene/Cell, Neuro & Tools - Week of July 5, 2026

Biotech Pipeline gene/cell, neuro and tools newsletter for the week of July 5, 2026. A single-shot in-vivo base editor cut LDL up to 62% in the first human print for the PCSK9 approach, the FDA cleared a pediatric sickle-cell gene therapy in 53 days, and hedge-fund money rotated back into healthcare at a five-year high.

Biotech Pipeline: Gene/Cell, Neuro & Tools

Week of July 5, 2026: In Vivo Base Editing Cuts LDL 62% in First Human Data

For a decade the in-vivo editing pitch has been a promise: one injection, permanent effect, no daily pill. This week it stopped being a promise and became a number in the New England Journal of Medicine, a single dose that knocked out most of a cholesterol gene and dropped LDL by up to 62%. Meanwhile the FDA quietly pushed a gene-therapy cure down to two-year-olds in 53 days, and on the neuro side the tape was all pipeline and plumbing (tau, blood tests, prevention trials) with the commercial launch numbers still nowhere to be found. Here's the week.

TL;DR

  • In-vivo base editing put up hard human data. A single injection of Verve's PCSK9 editor cut the target gene's activity 88% at the top dose and lowered LDL cholesterol by up to 62% across 35 patients, published in NEJM, sponsored by Eli Lilly. The read for the whole editing complex: the "one-and-done" thesis now has a clinical print. (CNBC/CTSNet, Jul 2)
  • The FDA keeps moving fast on cell/gene therapy. A gene therapy for sickle cell disease won supplemental approval for children as young as two, in 53 days, and Orca Bio's Treg cell therapy became the first regulatory-T-cell product approved. (Health:Further, Jul 4)
  • Money is rotating back into healthcare. Hedge funds are the most bullish on the sector in five years, the group just booked its best month since November, and a Mizuho specialist says 2026 is already locked in as a record year for biotech M&A. (CNBC Power Lunch, Jun 30)

What's New

In-vivo editing finally has a headline number. On the July 2 episode of The Beat, host Joel Dunning walked through the marquee paper of the week: "In vivo base editing of PCSK9 with Verve 102 in hypercholesterolemia," in the New England Journal of Medicine. The setup is the elegant one editing bulls have pitched for years, some people carry a loss-of-function PCSK9 variant and simply "don't die of heart disease," with "really low atherosclerotic plaque... and really low LDL." Verve's editor is designed to manufacture that phenotype on demand. The result, in his words: across "35 participants" and "lots of different doses," it "reduced 88% at the highest dose," and "your LDL cholesterol went down up to 62% with one injection." His verdict, "this is super exciting and could be something that literally could go widespread in the coming years," is the read-through: a single-shot, durable edit with cardiovascular-scale addressable population is no longer theoretical. Note the sponsor: the study was run by Eli Lilly (which now owns the program), a reminder that big pharma is underwriting the in-vivo editing frontier itself. (CNBC/CTSNet)

The FDA's fast lane for cell and gene therapy stayed open. On the July 4 Health:Further, the hosts flagged that "the FDA approves the first gene therapy for children with sickle cell disease," extending a one-time functional cure to kids "as young as two," with the agency granting the supplemental approval "in 53 days." They called it "a flat out cure" and "an incredible win for science." In the same breath: Orca Bio's Tregsy won approval as "the first cell therapy based on regulatory T cells," cutting graft-versus-host disease in allogeneic transplant. The candid aside worth keeping, CAR-T "was one of the casualties of the biotech funding fallout," with the cold-chain, per-patient logistics still the bottleneck. It's a tidy encapsulation of the sector: the science keeps clearing gates faster than the manufacturing base can scale. (Health:Further)

Voyager's Al Sandrock made the cleanest case for tau, and for gene therapy as the delivery vehicle. On the June 29 Business of Biotech, the Voyager CEO (and the ex-Biogen R&D chief who shepherded Aduhelm) laid out why tau is the target that matters: amyloid is the "match," but "the tau is the forest fire," and to treat Alzheimer's more effectively "you got to go after tau." Voyager's swing is VY-1706, an IV gene therapy that makes brain cells produce a tau-lowering siRNA, a potential "once and done" versus rival intrathecal approaches that need "a lumbar puncture... every six months." He confirmed the clock: "we just got IND clearance a couple of weeks ago. We expect to enroll the first patient in our phase one trial in the second half of this year," with a tau-PET readout on the antibody program also due in 2H. He name-checked Biogen's tau antisense drug BIIB080 as having "just read out," with data due "at the conference in July." (Business of Biotech)

The anti-amyloid story is quietly moving upstream, to people who aren't sick yet. On the July 1 Brain Talk, UC Irvine's Dr. Joshua Grill detailed the two prevention trials that could redefine the market. He runs AHEAD, the lecanemab-in-preclinical-AD study "funded by the National Institute on Aging and ESI [Eisai]," which "had to screen 20,000 people to find those 1,700" randomized, completing in 2028. Lilly's donanemab equivalent, "Trailblazer ALS-3," could "come even sooner"; both are "now fully accrued and no longer recruiting." The stakes, in his words: a positive result "could result in FDA approval for the first time ever for a treatment in preclinical Alzheimer's disease. That would be monumental." (Brain Talk)

The Debate

Are the Alzheimer's blood tests ready to become the on-ramp to a launch, or still a liability? The bull case is that diagnosis is being commoditized right as the drugs arrive. On the June 29 Medscape CME panel (grant-funded by Lilly), Wake Forest's Michelle Milkey noted two blood biomarkers now have FDA clearance, a p-tau 181 "triaging" test (negative rules AD out) and a Lumipulse amyloid-ratio test, with p-tau 217 "widely clinically available." Cheaper, scalable, rural-friendly: the "great equalizers" for finding treatable patients. But the same panel steel-manned the bear case itself. Most validation was done "in specialized clinics," not primary care, so real-world accuracy is unproven; renal dysfunction can throw false positives (a live concern given "most of their patients have at least mild renal insufficiency"); and a positive result in a cognitively-normal person carries hard downside, "refusal for long-term care insurance, life insurance, disability insurance, and also some senior housing." Grill was blunter on Brain Talk: direct-to-consumer online tests are "not quite ready for prime time," with "variability from provider to provider." The honest netting: the diagnostic funnel that a real Leqembi/Kisunla launch needs is being built, but it isn't trustworthy enough yet to run at population scale, and nobody this week produced actual patient-start numbers to tell us the launch is inflecting. (Keeping Current CME; Brain Talk)

Read-throughs & Names in Play

  • The in-vivo editing complex (Verve/Lilly, Intellia, Beam). The PCSK9 print is the field's proof-of-concept moment on a cardiovascular-sized indication, not a rare disease. Bull: durable, one-shot efficacy validates the platform thesis broadly. Bear: it's Lilly's asset now, and a 62% LDL drop still has to clear a long safety and durability bar before "widespread." (CNBC/CTSNet, Jul 2)
  • Vertex/CRISPR (Casgevy read-through). The only approved sickle-cell gene therapies are the obvious beneficiaries of a pediatric label expansion cleared in 53 days, though the hosts never named the manufacturer, so treat the specific attribution as inference, not tape. The gating factor remains delivery infrastructure, not the science. (Health:Further, Jul 4)
  • Eli Lilly (LLY). Everywhere this week and mostly in the driver's seat: sponsor of the Verve PCSK9 data, owner of the donanemab prevention trial that could read out before AHEAD, grant-funder of the biomarker education, and, per Health:Further, a template for the Takeda/Insilico "up to 600 million" AI drug-discovery deal that followed Lilly's own. (CNBC/CTSNet; Brain Talk; Health:Further)
  • Biogen/Eisai (BIIB). Two catalysts on the tape: the tau antisense drug BIIB080 "just read out," with data at a July conference, and the AHEAD prevention trial (with Eisai) completing 2028. Bull: a preclinical approval would blow the addressable population wide open. Bear: 2028 is a long wait, and the launch metrics for the drug they already sell were, again, absent. (Business of Biotech)
  • Diagnostics picks-and-shovels. Two separate episodes leaned on the same idea, cheap, scalable, non-invasive detection. Mayo/ASU researchers reported "90 plus percent accuracy" using a routine retinal fundus photo plus AI to flag Alzheimer's, a potential adjunct (or competitor) to blood panels and amyloid PET. The whole early-detection stack is where the near-term commercial action is, ahead of the drugs. (Tomorrow's Cure, Jul 1; Keeping Current CME)

What Changed

The sentiment tape flipped hard. On the June 30 Power Lunch, Mizuho's Jared Holtz noted healthcare "gained 10 percent in basically a month," its "best month since November," after being "apoplectic"-level unloved a month ago, when it lagged the S&P by 17%. S3 Partners' Bob Sloan found "hedge funds are now the most bullish they have been on health care in five years." And Holtz's coda on deals: "the M&A tailwind has been unprecedented... I could guarantee you today this will be a record year for biotech M&A just six months through the year." Set that risk-appetite backdrop against a week of concrete science, a landmark editing readout and a 53-day pediatric approval, and the reflexive read is that fundamentals are, for once, showing up to meet the flows. (CNBC Power Lunch)

What Didn't Show Up

The commercial Alzheimer's launch tape was, once again, a no-show. For all the tau, biomarker and prevention-trial talk, no episode this week put out a single Leqembi or Kisunla patient-start number, nothing on SubQ dosing uptake, ARIA-monitoring throughput, or fresh CMS coverage mechanics from Biogen/Eisai or Lilly. On life-science tools, the silence was near-total: no podcast gave voice to Thermo Fisher, Danaher, Agilent, Revvity, Sartorius, Repligen, Bruker, Waters or Illumina, and there was no single-use book-to-bill, destock-to-restock, or NGS-pricing tape, only the sector-level rotation call above. And outside the Verve/PCSK9 paper, there was no company-specific data this week from CRISPR Therapeutics, Intellia, Beam, Editas or Prime on ATTR, ANGPTL3 or durability. Where there was no tape, we left the line blank rather than pad it.