Quick note: The Longevity Letter has a new home (Kit → Beehiiv)! If this landed in spam or promotions, drag it to your primary inbox so you don't miss future issues.

Hi {{first_name}} ,

This week, the most talked-about blood test in cancer screening stumbled publicly.

GRAIL reported results from the NHS-Galleri trial: 142,000 people, three years, the largest multi-cancer early detection study ever run. The primary endpoint was not met. Their stock fell 47% overnight.

The financial press ran "cancer test fails." Social media declared MCED dead.

Both takes miss the point. The actual story is richer, more complicated, and ultimately more hopeful than either headline suggests.

Six people in my family have had cancer. Most of them died from it. One of my cousins was a physician in his 30s — he left behind a wife and two kids.

I trained as a hematology/oncology fellow at Columbia before moving into startups and longevity medicine. Cancer is personal for me in a way it is for a lot of families, and that's exactly why I've spent years tracking this field.

I've also developed a longevity-focused cancer screening protocol into practice on myself and for my patients (customized to each individual). I've had the Galleri test done. I've had two whole-body MRIs. I've had whole genome sequencing (WGS) completed — that's a complete readout of your entire DNA sequence, capturing inherited risk that routine consumer genetic tests miss.

The protocol I walk through in this issue isn't a theoretical framework. It's what I actually do.

In this issue:

GRAIL was built on one conviction: cancer deaths are mostly a timing problem. Find the tumor at Stage I instead of Stage IV, and most cancers are curable.

The company was spun out of Illumina in 2015. Jeff Huber, a former Google executive who had just lost his wife to colorectal cancer, became CEO. Bezos Expeditions and Bill Gates both wrote checks in the Series A. Illumina eventually bought the whole company in 2021 for $8 billion, a deal that immediately drew antitrust scrutiny from both U.S. and European regulators because Illumina makes the sequencing machines that power liquid biopsy. (The regulators won and GRAIL was forced to re-spin as an independent public company in June 2024 and now trades on NASDAQ as GRAL.)

The story of GRAIL is also the story of how much money and political capital has been poured into one idea. That idea has held up.

The Galleri test is a liquid biopsy, and that phrase is doing a lot of work.

When any cell in your body divides, dies, or is damaged, it releases small fragments of DNA into the bloodstream. This is cell-free DNA (cfDNA) — present in everyone, healthy or not. The challenge is signal versus noise: cancer cells shed cfDNA too, but in small quantities, mixed in with billions of fragments from normal cells.

GRAIL reads methylation patterns in that cfDNA. Methylation is a chemical modification (a methyl group attached to specific cytosine bases on the DNA strand) that regulates gene expression. Normal cells have characteristic methylation patterns. Cancer disrupts them in ways that are consistent within a cancer type and distinguishable from normal tissue.

The test sequences cfDNA from a blood draw, reads hundreds of thousands of methylation sites, and runs a machine learning model against a trained reference set. If a cancer-like pattern appears, a second algorithm predicts where in the body the signal originated, which tells the clinician where to look first.

Specificity is 99.5%. Fewer than 1 in 200 tests generates a false positive, which matters enormously for a screening test because false positives trigger biopsies, surgeries, and real anxiety. Sensitivity varies by stage. Stage I detection is lower, improving substantially at Stage III and IV. The test is better at catching what's already somewhat advanced. Not ideal, but better than finding nothing.

Here's what the 142,000-person study actually showed.

The trial ran three years across England, enrolling participants aged 50-77 in a randomized controlled design. The pre-specified primary endpoint was a statistically significant reduction in combined Stage III and IV cancer diagnoses in the Galleri arm versus standard of care. That endpoint was not met.

What was observed is more interesting. The Galleri group had a 4-fold higher overall cancer detection rate. In years two and three, there was a greater than 20% reduction in Stage IV cancers among 12 pre-specified deadly cancer types. More Stage I and II detections. Fewer cancers discovered via emergency presentation, which is a proxy for "found it too late." Full data will be presented at ASCO 2026.

Why did the trial miss its primary endpoint? Three years is a short follow-up for a disease that develops over decades. The pre-specified endpoint was ambitious. And the trial showed higher-than-anticipated Stage III incidence, meaning some of what got caught still wasn't early enough. These are real limitations. They are not reasons to abandon the technology.

GRAIL submitted the final module of its FDA Premarket Approval application in January 2026. The FDA evaluates MCED tests differently than a randomized trial primary endpoint: analytical performance, clinical validation, and Stage I-III detection benefit. That application is still pending.

Three weeks ago, on February 3, 2026, President Trump signed the Nancy Gardner Sewell Medicare Multi-Cancer Early Detection Screening Coverage Act. Bipartisan: 341-88 in the House. It creates a Medicare coverage pathway for MCED tests following FDA approval, the first time Congress has directly mandated CMS to cover this category of diagnostic. The NHS trial result doesn't undo that law.

"GRAIL" has become synonymous with multi-cancer detection. That's both a marketing win and a clinical misconception.

At least five other companies are building MCED tests with meaningfully different approaches, and one of them has already beaten GRAIL to FDA approval.

Guardant Health: SHIELD

Guardant's Shield test got FDA approval in July 2024 as the first blood-based test approved as a primary screening option for colorectal cancer. Not an MCED test yet, but it proved the liquid biopsy regulatory pathway actually works for screening indications. Medicare covers it. Shield V2 shows 84% sensitivity for colorectal cancer at 90% specificity. Guardant is now running SHIELD Lung, a 10,000-participant study. Their playbook: single-cancer approvals first, then build toward multi-cancer. Slower, but the regulatory precedent is cleaner.

Exact Sciences: CancerGuard

Exact Sciences acquired Thrive Earlier Detection, a Johns Hopkins spinout commercializing the CancerSEEK test from Bert Vogelstein's lab, for $2.15 billion in January 2021. CancerSEEK analyzed 8 cancer-associated proteins plus 16 cancer gene mutations simultaneously. Exact Sciences folded that technology into CancerGuard, which launched in September 2025 at $689, cheaper than Galleri, distributed through Quest Diagnostics' 7,000 patient access sites. The key differentiator is combining protein markers and DNA mutations rather than methylation alone. Multiple signal types should improve sensitivity. The validation data is still maturing.

Freenome

Freenome is the best-funded private player in the space. Their platform analyzes cell-free DNA, cell-free RNA, and proteins together — a multiomics approach (analyzing multiple types of biological data together, like DNA, RNA, and proteins, to get a fuller picture) — using machine learning to integrate the signals. Roche signed an exclusive international licensing deal worth $200 million or more in November 2025, bringing total funding past $1 billion. Their PREEMPT CRC study enrolled 35,000 participants. They have early data on pancreatic cancer detection, one of the hardest targets in the space. Freenome isn't approved or commercial yet, but Roche doesn't write $200 million checks on a hunch.

Harbinger Health

Harbinger Health is the competitor most people in longevity medicine aren't watching yet. The Cambridge, MA company raised $140 million in 2023 and presented at both ASCO and AACR in 2025. Their "reflex" MCED test is designed for high-risk populations, including people with obesity who are underserved by risk stratification tools built for average-population studies. Their data shows meaningful per-cancer positive predictive value and early-stage sensitivity across multiple high-incidence cancers. No FDA submission timeline has been announced publicly, but the data presentations suggest they're building toward one.

Delfi Diagnostics

Delfi raised a $225 million Series B and takes a different technical angle: cfDNA fragmentation patterns rather than methylation. The argument is that how DNA fragments in the blood reflects the chromatin structure (the tightly packed organization of DNA inside cells) of the cell it came from, and cancer changes that structure. Computationally intensive, potentially complementary to methylation-based tests. Earlier stage, focused on lung cancer for now.

What makes the competitive picture matter: the Nancy Gardner Sewell Act doesn't specify which company's test gets covered. It covers the category. Every company in this race is competing for the same Medicare reimbursement slot.

The current generation of MCED tests, Galleri included, are first-generation tools. They work. They're not the ceiling.

The limitation everyone is racing to solve is Stage I sensitivity. Small early tumors shed very little DNA, and that signal-to-noise problem doesn't disappear with incremental refinement. The next approaches are attacking it differently.

Multiomics integration (analyzing multiple types of biological data together, like DNA, RNA, and proteins) is the dominant near-term bet. Adding protein biomarkers from proteomics (the study of all proteins in a biological sample), RNA expression patterns, and metabolites to the cfDNA signal gives you more independent biological signals pointing at the same cancer, which should improve sensitivity without sacrificing specificity. Freenome and Exact Sciences are already building this way.

AI-designed peptide sensors are the longer-range bet. The MIT/ARPA-H work I cover below takes a different approach entirely: instead of reading what the cancer sheds, you inject sensors designed by AI to interact with cancer-specific enzymes called proteases (enzymes that cut and break down proteins, often overactive in cancer), then read what comes out in urine. No blood draw. Scaling this to 30 cancer types via an at-home kit is still years away, but the underlying biology is elegant in a way that incremental cfDNA improvements aren't.

Risk-stratified portfolios are the clinical evolution that doesn't require new technology. Right now MCED tests are marketed as universal add-ons. The smarter deployment is personalized panels. If your germline shows BRCA2 and PALB2 variants, your MCED protocol should weight toward breast and ovarian detection. If you're in a high-risk metabolic category, Harbinger's approach may outperform the general-population tools. The field hasn't built the clinical infrastructure to do this systematically yet, and that gap concerns me more than the missed NHS endpoint.

The question for MCED is no longer whether it works. It's who gets it, when, and at what price. The Nancy Gardner Sewell Act answered the price question for Medicare beneficiaries. The FDA will answer timing. The clinical community still needs to answer who.

Cancer screening is a strategy, not a single test. The GRAIL news reinforces why I never put all the weight on any one tool, and why the right protocol depends on your biology, not on which test got good press this week.

This is what I actually do for myself, and what I recommend for patients who want a serious early-detection plan.

Here's what the full protocol looks like:

Layer 1: Family history and genetic risk

Start here. Always. Six relatives in my family have had cancer. That history is the first thing that shapes every decision downstream. A first-degree relative with cancer before age 60, or any BRCA1/2, Lynch syndrome, or PALB2 history changes everything that follows. Get a genetic risk assessment before spending money on advanced screening. This determines which tests matter most for you specifically, and which MCED platforms are better calibrated for your risk category.

Layer 2: Germline genomics

Get a clinical-grade germline genetic panel. Not a consumer ancestry kit. You want a panel covering hereditary cancer syndromes. Invitae and Color both offer these at accessible price points. If you want the most complete picture, whole genome sequencing (WGS) goes further: it's a full readout of your entire DNA, not just a targeted panel of known variants. I've had mine done. The results inform cancer risk, cardiovascular risk, and neurological risk. You do this once and it informs every decision that follows.

Layer 3: Traditional screening, done right and on time

Colonoscopy starting at 45 (or earlier with family history). Mammogram annually starting at 40. Low-dose CT lung cancer screening for current or former heavy smokers. PSA with shared decision-making for men over 50. Cervical cancer screening per the updated HHS guidelines. These aren't glamorous. They're the backbone. Most Americans skip or delay them, and that's where the preventable deaths happen.

Layer 4: Whole-body MRI, for the right person

High false-positive rates mean this isn't for everyone. I've had two done. In appropriately selected patients with strong family history, germline mutations, or significant risk factors, it adds anatomical resolution that blood tests can't provide. Not first-line for an average-risk 40-year-old, but valuable in the right context.

Layer 5: MCED tests

I still recommend Galleri for appropriate patients despite this week's news. The PATHFINDER 2 data, presented at ESMO in October 2025, showed a 7-fold increase in cancer detection when Galleri was added to standard USPSTF screenings. A test stuck at $1,000 out-of-pocket waiting for an FDA decision, yes. But not a failed test.

For patients who want an alternative now, CancerGuard at $689 through Quest Diagnostics is a real option with a different signal profile. The next 18 months of Stage I sensitivity data will tell us a lot about which approach is better.

Don't wait for the perfect test. Build the best available strategy from what exists today and update it every year. That's what medicine has always been.

📋 Takeaways:

The technology works. The unsolved problem is building personalized protocols that match patients to the right tools.

If you found this useful, share it with someone who should be thinking about cancer screening.

🏛️ ARPA-H commits $144M to aging science — Seven teams, one big bet. A Brown University team gets $22M to test whether an HIV antiretroviral called Censavudine can suppress retrotransposon activity (retrotransposons are "jumping genes" that become more active with age and are thought to drive inflammation). Endpoints News — full breakdown | ARPA-H PROSPR program page (Feb 24, 2026)

🇯🇵 Japan authorized the world's first iPSC cell therapies — Neurons for Parkinson's and cardiac muscle patches for severe heart failure: conditional approvals, small trials, but iPSC (induced pluripotent stem cells — adult cells reprogrammed back to an embryo-like state, then guided into specific cell types) therapies are no longer theoretical. Japan Times (Feb 19-20, 2026)

💊 Wegovy drops to $675/month in 2027 — Novo Nordisk is cutting U.S. list prices for Wegovy by up to 50% and Ozempic by up to 35% starting January 2027. Self-pay channels are reportedly unaffected, so this mainly changes the math for insured patients and employer health plans. Reuters (Feb 24, 2026)

⚗️ AI-designed proteins may detect 30+ cancers in a urine test — MIT and Microsoft Research used AI to design peptide sensors selectively cleaved by cancer-specific proteases (enzymes that cut proteins, often overactive in tumors), with reporter molecules excreted in urine. ARPA-H is funding a push toward an at-home kit targeting 30+ cancer types. MIT News | MIT Technology Review (Feb 24 coverage)

🧬 The first human klotho trial just started — Klothea Bio initiated a Phase 1b trial of AKL003, an mRNA-based klotho therapy, this week. Klotho is a protein that declines sharply with age and correlates with cognitive function, kidney health, and longevity in animal and population studies. Human trial data has been essentially nonexistent until now. (Feb 19, 2026)

🫀 FDA tells GLP-1 makers to drop the suicide warning — The FDA officially requested manufacturers remove suicidal ideation warnings from GLP-1 receptor agonists; subsequent large analyses failed to confirm the signal that triggered the original label language. The drugs' mental health safety profile is better than the label implied. (Jan 13, 2026)

The MCED field is having its adolescent moment. The technology works, the regulatory framework exists, the competitors are live, and the first big trial result is ambiguous enough to hand ammunition to skeptics and believers alike. That's what progress looks like in medicine — messier than the press release version.

GRAIL stumbled. The field advanced. And three weeks ago, Medicare locked in coverage for whatever FDA-approved test comes next.

If you do one thing after reading this: ask your doctor which standard screenings you're actually due for. That's still where most of the preventable deaths hide — not in the $1,000 blood tests, but in the colonoscopies people keep rescheduling and the mammograms that slip to "next year."

If you want a starting point, use the 5-layer framework above. Start at Layer 1. Know your family history. That's free, and it changes everything.

To your healthspan,

Hillary Lin, MD

Co-Founder & CEO, CareCore

Follow me for more longevity insights: YouTube | LinkedIn | Instagram | TikTok

Want to turn your wellness brand into a full-service health destination? Learn about CareCore's platform or Get Started Here

Where to find me:

👱🏻‍♀️👩🏻‍🦰👩🏻👧🏽👧🏾 Livelong Women's Health Summit - April 17-18, 2026, SF, CA. Delighted to join 50 other thought leaders in speaking on women's longevity! (Code HILLARY50 for $50 off)

Support us with your longevity purchases:

NeuroAgeTx - the most comprehensive brain aging package at up to 61% off for Longevity Letter readers

Timeline - patented Urolithin A for scientifically proven mitochondrial support via mitophagy (Code CARECORE for 10% off)

GlycanAge - measures your immune system's biological age and recommends how to fix it (Code CARECORE for 10% off — 30% off until March 7th!)

ZBiotics - pro-longevity probiotics for when you do have a little fun (Code CARECORE for 10% off)

Have a question or topic you want covered? Reply to this email.

Was this forwarded to you? Subscribe here to get The Longevity Letter every week.