Precision medicine.
Paid before proven.

Congress has agreed to pay for a test that doesn't officially exist yet.

A law passed in February puts multi-cancer blood tests, which look for signs of dozens of cancers in a single blood draw, under Medicare starting in 2028, as long as the FDA has approved them. None has been. The furthest along is GRAIL's Galleri, and in September an FDA advisory panel agreed its benefits outweigh its risks, then split 6–4 on whether it actually works. The money has been promised. The evidence is still being argued over.

That gap between payment and proof runs through all of precision medicine. The idea is to match care to a patient's biology instead of treating everyone with the same diagnosis the same way, and most of the investable part is testing. Some tests find cancer early from a blood sample (a liquid biopsy). Others read a tumor's mutations to pick the right drug, or track tiny traces of tumor DNA after surgery to catch a relapse months before a scan would, which the industry calls minimal residual disease testing, or MRD. The rest is drugs and gene edits aimed at a specific genetic cause.

So far, investors have been willing to pay ahead of the verdict. The testing-focused precision medicine stocks have had a strong 2026. Natera is up nearly 80%, and GRAIL is up about two-thirds even after losing half its value in a single day in February, when a big NHS trial of Galleri missed its main goal.

In 2027, the verdicts start landing. Precision medicine stocks will trade on proof: trial data a regulator accepts, guideline listings, and insurer coverage. The list is grouped by how much of it each company has: the ones already getting paid, the one waiting on the FDA, the companies building end-to-end testing businesses, the toolmakers, and two drug makers facing binary FDA decisions.

These three have the evidence, the guideline listings or the coverage, and the revenue that comes with them.

Natera NASDAQ: NTRA

Natera dominates MRD testing with Signatera, which looks for a patient's own tumor DNA in the blood after treatment. It ran about 283,000 clinical MRD tests last quarter, up from 181,000 a year earlier, and total revenue grew 38%.

The proof is piling up. In May the FDA approved Signatera as a companion diagnostic for a bladder cancer immunotherapy, the first blood-based MRD test approved that way. That means the test is written into how a drug is prescribed, which is about as sticky as diagnostics gets.

Guardant Health NASDAQ: GH

Guardant Health sells Shield, a blood test for colon cancer for people who won't do a colonoscopy or a stool test. Volume roughly quadrupled to about 66,000 tests last quarter, and total revenue grew 44%.

What changed this year is payment. The American Cancer Society added Shield to its colon cancer screening guideline, and UnitedHealthcare started covering it on August 1, the first major private insurer to do so. A test that insurers pay for is a real business. Guardant's next fight is a direct competitor, Freenome's test, which Abbott launches this fall.

Caris Life Sciences NASDAQ: CAI

Caris Life Sciences reads every gene and every active gene message in a patient's tumor, far more than most tumor tests, to help pick the right treatment. That's expensive to do and valuable to own: it builds one of the deepest clinical datasets in oncology.

The business is getting close to self-funding. Revenue grew 45% last quarter at a 68% gross margin, it nearly broke even, and it raised guidance. It also launched its own multi-cancer screening test this year.

GRAIL NASDAQ: GRAL

GRAIL's Galleri test looks for more than 50 types of cancer in one blood draw. It is the test the Medicare law was written for, and it's the clearest binary bet on this list.

The evidence is mixed. The NHS trial in England missed its main goal in February, though it did show fewer late-stage diagnoses in later rounds. The FDA panel in September backed the benefits but split on effectiveness. If the FDA approves Galleri, GRAIL has a Medicare market waiting in 2028. If it doesn't, the law pays for a test nobody has yet. GRAIL has about $860 million in cash and brought in Samsung as an investor this year, so it can afford to wait.

The biggest moves among precision medicine stocks this year were companies buying their way into owning a patient's whole testing journey.

Abbott NYSE: ABT

Abbott closed its $20.6 billion purchase of Exact Sciences in March. That gave it Cologuard, the best-known at-home colon cancer test, and Oncotype DX, the test that guides breast cancer treatment. It also took over Exact's deal to sell Freenome's new blood test in the U.S.

That's a stool test, a blood test and a tumor test under one sales force that already calls on doctors' offices across America. Abbott is a giant, and diagnostics is one piece of it. The stock is down about 22% this year, so the market isn't paying much for the bet yet.

Freenome NASDAQ: FRNM

Freenome went public through a SPAC merger in July. A week later, the FDA approved its SimpleScreen blood test for colon cancer, making it the second blood-based colon screening test on the market after Guardant's.

Freenome doesn't need to build a sales force; Abbott is selling the test in the U.S. starting this fall. Roche has licensed Freenome's technology outside the U.S. for a future multi-cancer test. It's a small company plugged into two of the biggest diagnostics distributors in the world.

Tempus AI NASDAQ: TEM

Tempus AI sells tumor tests, but its real asset is the clinical and genomic data those tests generate, which it licenses to drugmakers. It reported its first GAAP profit last quarter, though only thanks to gains on investments. Operations still lost money.

In July it agreed to buy Personalis, an MRD testing company, in a deal worth about $1.5 billion, expected to close around the turn of the year. That gives Tempus the follow-up test after the first diagnosis: more tests per patient, and more data per patient.

Illumina NASDAQ: ILMN

Illumina makes the gene sequencers that most clinical tests run on. Its clinical consumables, the chemicals each test uses up, grew about 15% outside China last quarter, and that's the line that tracks precision medicine's growth.

China has barred it from selling there, and research budgets have been soft. But every Signatera, Shield or Galleri result is a sale for whoever makes the sequencer.

Roche OTCQX: RHHBY

Roche is the only giant building the whole stack. It owns Foundation Medicine, a major tumor-profiling lab, and in June it launched Axelios, a new sequencer built on its own technology, for research use for now. It agreed to buy PathAI, which uses AI to read pathology slides, for $750 million up front, and bought SAGA Diagnostics, an MRD company.

A sequencer, the tests and the slide-reading software, all from one company, is a direct challenge to Illumina. Roche has also shown investors a cost target of about $150 per genome. It's a slow-moving pharma giant, but it's the one with the most pieces.

Both of these are binary bets on FDA decisions in the next six months.

Intellia Therapeutics NASDAQ: NTLA

Intellia Therapeutics edits genes inside the body with CRISPR, delivered by an infusion. Its treatment for hereditary angioedema, a rare swelling disorder, is under priority FDA review with a decision due March 10, 2027. If approved, it'll be the first in-body CRISPR drug on the market.

Its other big program, for a heart disease called ATTR, was put on hold last year after a patient developed severe liver injury and later died. The holds were lifted this year and enrollment has resumed. Intellia is a single-decision story for the next six months.

Praxis Precision Medicines NASDAQ: PRAX

Praxis Precision Medicines has two drugs with FDA decisions coming. Relutrigine, for rare epilepsies caused by specific gene mutations, is due December 27, after the FDA pushed it back three months. Ulixacaltamide, for essential tremor, is due January 29. Neither has an advisory committee planned. A third drug missed its main trial goal in June. Two approvals would turn Praxis into a commercial company almost overnight; two rejections would leave little.

The fight in precision medicine is moving from the single test to the whole patient.

A cancer screening test is the top of a funnel. A positive blood test leads to a scan, then a biopsy, then a tumor profile to pick a drug, then years of MRD monitoring to catch a relapse. One screening result can trigger thousands of dollars of follow-up testing, most of it repeating for years.

This year's deals make sense in that light. Abbott bought Exact to own screening and tumor testing, then took on Freenome's blood test. Tempus is buying an MRD lab. Roche bought an MRD company and a pathology AI company while launching its own sequencer. Guardant already sells a screening test, a tumor profile and an MRD test.

When Medicare starts paying for multi-cancer screening in 2028, the big new flow of patients will enter at the top of those funnels, and precision medicine stocks will be repriced by who owns the rest. The company that gets paid most won't necessarily be the one with the best screening test. It'll be the one that owns the most steps after it.