CAI — Caris Life Sciences Deep-Dive v1.0

Liquid Wheel Research · Deep-Dive Framework · Stages 0–12

🧬

Caris Life Sciences (CAI) — the growth was price, and the price is done

The only public precision-oncology lab sequencing whole exome + whole transcriptome on every sample. Real, FCF-positive, founder-owned — and at $32 the market has paid for the Medicare price step and three unproven franchises on top.

⚠ THESIS: ON WATCH CONVICTION-SPEC · MOAT 6.5 < 9 GATE · CAP ≤5% MACRO 🟡 YELLOW DO NOT INITIATE ABOVE $32.90

6.0

/ 10 overall

prior draft 6.5 ↓

Data as of 2026-09-25 close · $31.98 · 282.6M basic / ~305M fully diluted (treasury method; 10-K: 21.9M options + 4.5M RSUs + 0.15M ESPP = 26.5M overhang) · Mkt cap $9.04B · EV ≈ $8.64B (face debt $400M, cash + securities $793.1M) · FY = calendar · $ in millions unless noted

Team inputs merged: Quant (Stages 2/4/8) · Qual (3/5/6) · Macro (1) · Sentiment (7) · Technicals (9/10) · Red Team (10). Framework: investing-framework.md (16-gate checklist per the IREN v3 template). Supersedes the same-day draft build (2026-09-25 23:35, "v1.0", score 6.5).

⏱Data as of 2026-09-25 close ·computing age…

TL;DR — read in 30 seconds

Profit is real on a cash basis, thin on a run-rate basisTTM FCF $124.0M (not the triage's $63.3M), but quarterly FCF has faded $55.3M → $39.7M → $22.5M → $6.4M, Q3 guided FCF-neutral, and Q2 operating income ex true-ups was ~$3M (1.3% margin).

Growth was price; price is doneFY25 profiling revenue +$417.6M = $311.4M ASP/mix, $52.0M tissue volume, $54.2M Assure. Ex-true-up revenue/case $3,816 → $3,795 → $3,863; 0211U locked flat at $8,455 for 2027+. Forward growth ≈ volume (+18–20%) + new products. H2'26 guide implies +8.9% reported YoY.

Reimbursement risk: unknown → mostly known, mostly goodPreliminary CY2027 CLFS: MI Cancer Seek flat, Assure $3,649 → $5,250. Final ~November. Live risks are MA/commercial realization, DOJ 14-day-rule CID (FCA), and a 39.6% Medicare book on one code.

Valuation fair-to-full, not euphoric9.1x TTM / 8.4x FY26 guide / 7.1x FY27 consensus EV/S. Cheaper than GH (20x) and NTRA (22x) growth-adjusted, in line with TEM, rich vs VCYT/NEO. Probability-weighted 12–24-mo value $30–33 ≈ spot. No asymmetry; the hunting ground was $14.55–18 in May–July 2026.

TradeNo-chase line $32.90. Ladder 25% at $24.5–26.3, 35% at $22.5–23.6, 40% at $20.3–21.5; weekly close <$20.0 invalidates. Cap 5%. Wheel is thin (1–5 contracts, limit at mid); Jan-15-27 $22.5P (basis $20.94) is the best risk/reward strike. Trim bands $36.0–36.5 / $39.3–42.5 / >$45.

Price / Mkt Cap

$31.98

$9.04B · 52wk $14.19–33.65 · EV $8.64B

91% of 52wk range · 5.0% below high

P/S TTM · EV/S FY27

9.1x · 7.1x

"aware/optimistic" boundary · was 4.5x at $14.55

Framework aware band 4–8x = $14–28

TTM FCF (fading)

$124.0M

Q by Q: $55.3 → $39.7 → $22.5 → $6.4M · Q3 guided neutral

Q2 FCF is 12% of the Q3'25 peak

Net cash · Skin in game

+$393M

$793.1M cash vs $400M face debt · CEO 196x / 1,777x

$50M min-cash covenant — trivial

The Bet

Caris is the only public precision-oncology lab that sequences the whole exome and whole transcriptome on every sample, has an FDA-PMA'd WES/WTS companion diagnostic billed at $8,455 on Medicare, runs a 68% gross margin, and has printed five straight FCF-positive quarters with $793M of cash and a founder holding $4.0B of stock. The 2027 Medicare fee schedule (preliminary) left its core code flat and raised its blood test 44%. That is the good news, and at $32 the market has paid for all of it and more: 9.1x TTM sales, ~85x FY27 consensus EPS, RSI 74, 51% above the 200-day, with the President and Vice Chair selling $27M into the September squeeze. The growth that re-rated the stock was ~75% price, not volume, and the price step is done (ex-true-up revenue per case has been flat for three quarters). Ex prior-period true-ups, Q2 operating margin was ~1% and Q3 adjusted EBITDA is guided to $10–16M from $55.7M. The bull needs volume to hold +20%, Detect to ship, MRD to get covered, and pharma to triple in H2. Any of those can happen; none is priced as uncertain. The bet is to own this business at the framework's 4–8x "aware" band on a pullback the technicals agent has mapped ($24.5 → $20.3), not to chase a sector-beta squeeze at the 0.618 retrace.

Thesis status · ON WATCH

The business is real and profitable on a cash basis. The thesis — that comprehensive sequencing compounds into MRD/MCED/pharma franchises before the Medicare price tailwind fades — has not been proven on any of its three new legs, and reported growth halves inside three quarters.

Macro stamp · 🟡 YELLOW

Fed hiked 25bp on 2026-09-16 to 3.75–4.00%; 10y 5.18% (highest since 2007); ISM 54.6/55.4, GDPNow 5.0%, M2 +5.7%. The late-cycle unprofitable-small-cap veto does not fire (mid-cap, operating-profit and FCF positive, net cash). The rate gate is wrong for an ~85x forward P/E asset. Starter/ladder only at add-zones.

Company 101 — what Caris actually is

What it does. Caris runs a CLIA lab in Phoenix that takes a tumor sample (tissue: MI Profile / MI Cancer Seek) or a tube of blood (Caris Assure) from a patient with advanced cancer and sequences the whole exome (~23,000 genes) and whole transcriptome (~61,000 transcripts), then returns a report on which therapies should work. It has done WTS on every sample since 2019 and WES since 2020, building 1.13M profiles, 733k WES, 783k WTS, 845k matched with outcomes (6/30/26). That dataset trains new products: Caris Detect (multi-cancer early detection, cash-pay $3,500, launched 2026-07-01 via Everlywell), ChromoSeq (heme, MolDX $3,228, launched 2026-04-01), MI Clarity (breast-recurrence AI), MRD (launch H2'26), and pharma discovery deals (Genentech, AbbVie, Moderna, Merck KGaA, Xencor). Revenue mix Q2'26: molecular profiling $252.3M (96%), pharma R&D services $11.5M (4%). Cases Q2'26: 59,200 (48,300 tissue / 10,700 blood). Blended ex-true-up revenue per case ≈ $3,863; company "base ASP" >$3,850. Medicare = 39.6% of profiling revenue FY25 (10-K).

The 8-year-old version. When someone has bad cancer, the doctor sends a piece of the tumor to Caris. Caris reads all of the tumor's instructions instead of a short list and tells the doctor which medicine will work. Insurance pays Caris about $3,850 for each answer. Because Caris has read more tumors this way than anyone, it can invent new tests — like a blood test that looks for cancer before anyone knows it's there — and drug companies pay to look at its library.

Dated milestone timeline

Date	Milestone
1987–2004	Halbert founds AdvancePCS (PBM); sold to Caremark 2004 for ~$6–7.5B (proxy bio; the two agent figures conflict — immaterial)
2008	Caris founded (Caris Diagnostics). 2011: anatomic-pathology arm sold to Miraca for $725M; molecular business kept
2015	Precision Oncology Alliance research network formed
Sep-2018	Sixth Street term loan + 13.7M Series C warrants. 2019: WTS on every sample. 2020: WES on every sample. Jul-2020: PLA code 0211U issued
2021	$830M growth round led by Sixth Street
2022	Aug: MI Tumor Seek Hybrid Medicare-covered under NGS NCD. Jun: $2.9M DOJ settlement on Medicare's 14-day rule
Jan-2023 / Mar-2024	$400M 2023 term loan (OrbiMed/Braidwell, SOFR+6.5%) drawn in two $200M tranches
2023-12-08	Caris Assure Medicare coverage for therapy selection
Nov-2024	FDA PMA for MI Cancer Seek (only WES/WTS CDx). CMS gap-fills Assure 0485U at $3,649
Jan-2025	MI Cancer Seek launched at CLFS $8,455 — the ASP step-change. Mar-2025: DOJ Civil Investigative Demand (False Claims Act, 14-day rule)
2025-06-17/18	IPO priced $21, first trade 6/18; 26.6M sh incl. greenshoe; net $528.5M; all warrants exercised, 2025 converts extinguished ($19.9M charge)
2025-08-12	First public print (Q2'25). 2025-08-22: post-IPO high $42.50 intraday / $39.30 close. 2025-11-05: Q3'25, FY25 guide raised. 2025-12-15: lock-up expiry
2025-10-30	Bylaws amended — 3% ownership threshold to bring a derivative suit. Q4'25: Bobby Hill named CCO
2026-01-06	Sales realignment (82 → 146 territories, 250 → 300 reps). 2026-01-12: prelim FY25 (~$81M Q4 prior-period true-ups)
2026-02-26	FY25 results — revenue $812.0M (+97%); Achieve-1 interim MCED data
2026-04-01	ChromoSeq launched; $400M refinance with Blue Owl/Blackstone (SOFR+5.00%, 2031 maturity, +$300M DDTL for acquisitions, +$500M uncommitted); $25.2M extinguishment loss in Q2
2026-05-07	Q1'26 (rev $216.2M, +79%; cases +15%); Achieve-1 blinded validation (60.3% stage I–II sensitivity, 99.2% specificity — no primary citation located). Stock −18.6% next day; low $14.19 on 2026-05-13
2026-05-11/15	Directors Vacirca and Jon Halbert buy at $16.15 / $14.56
2026-06-04	Annual meeting; Poste and Knowles off the board; 33% of non-Halbert votes withheld from Halbert, 45% from Castleman. 2026-06-07/08: $100M buyback authorized; $17.9M executed in Q2 at ~$17.9
2026-07-01	Caris Detect launched (58 cancer types, WGS+WTS+cfRNA, $3,500 self-pay, Everlywell). Dual listing NYSE Texas
2026-08-05	Q2'26 beat/raise (rev $263.7M, +45%; FY26 guide → $1.03–1.04B). Stock +21.6% next day
2026-09-03/04	Spetzler sells 597k sh (~$14.8M) around 1.25M options expiring 9/12. 2026-09-17: Brille sells 400k @ $30.33 ($12.1M). 2026-09-14: Morgan Stanley conference 8-K
2026-09-16	Fed hikes 25bp. 2026-09-15–18: +27% week on 39.2M shares, no Caris filing (sector move with TEM)
2026-09-21	CMS posts preliminary CY2027 CLFS — 0211U $8,455 flat; 0485U $3,649 → $5,250 (+43.9%). Goldman PT $27 → $35

TAM & growth runway (bottom-up, assumptions shown; company/Nephron top-down $150B US is not used)

Therapy selection (US): ~2.04M new cancer diagnoses/yr (ACS 2025) × ~45% advanced/recurrent and guideline-eligible for CGP ≈ 900k patients × 1.2 tests/pt (tissue + ~40% blood attach) ≈ 1.1M tests × $3,900 realized ASP ≈ $4.3B realized TAM (assumption-stacked; unverifiable). Caris FY25 199,300 cases ≈ 18% of tests; management says market ~30–35% penetrated → served market today ~$1.5B. Runway = penetration 35% → 70% over 10 yrs (~7% volume CAGR) plus share.
MRD/monitoring: ~1.5M stage I–III survivors in surveillance × 4 tests/yr × $1,275 (NTRA Medicare ASP) ≈ $7.7B realized. NTRA alone ~1.1M tests/yr. Caris enters H2'26 at 0 share; 5% by 2031 ≈ $400M.
MCED self-pay: ~110M US adults 50+ × $3,500; 1% annual uptake = $3.9B, 0.25% = $960M. Halbert's "$1B capacity, tripling to $3B" brackets the ambition. Assume 0.1–0.3% uptake 2027–29 = $350M–$1.1B. No Medicare coverage path exists for MCED today.
Pharma data/discovery: TEM ~$350M/yr; Caris FY26 guide $75–80M; realistic 2031 $150–250M.

Horizon	Base revenue	Bull revenue	What has to be true (base)
FY26 (guide)	$1.035B	$1.04B	+20% volume, flat ASP, pharma $75–80M
3-yr FY29	$1.75B	$2.3B	Core +15%/yr incl. Assure 2027 price; Detect $150M; MRD $75M; pharma $110M
5-yr FY31	$2.4B	$3.6B	~25% of therapy-selection tests, ~5% MRD share, ~0.4% MCED uptake
10-yr FY36	$4.0B	$7B+	Bull requires MCED Medicare coverage — no statutory path today

Leading indicator to track is clinical case volume and sequential net adds (record +6,400 in Q2'26), not revenue.

The Flywheel — status: HALF-SPINNING

Sequence WES+WTS on every clinical sample SPINNING HARD

100% of samples since 2020

Largest matched clinico-genomic dataset SPINNING HARD

1.13M profiles / 845k with outcomes; compounds ~200k+ cases/yr

Train AI into new assays + discovery targets SPINNING AS OF 2026

Detect, MRD, ChromoSeq, MI Clarity — 3 launches; MRD pending; Achieve-2 unread

Broader menu + FDA/MolDX coverage + pharma deals WEAK

Pharma revenue $63.1M FY24 → $45.3M FY25 → $16.8M 1H26; new products are self-pay (Detect) or coverage-pending (MRD); ChromoSeq covered

More oncologists / sites / channels order Caris SPINNING SLOWLY

6,000+ MDs, 300 reps, Everlywell, OneOncology channel (sourcing unverified): +18% volume vs +31–63% at TEM/NTRA/GH

Samples + FCF reinvested in capacity and cheaper sequencing supply SPINNING

Capex $22.1M Q2, inventory +$47M, capacity $1B → $3B, supplier diversification → back to 1

Verdict: the data → product half works. The product → demand → pharma half is unproven, and it is the half the multiple is paying for. A flywheel exists on paper; the framework's "no flywheel = not a long-term hold" test passes narrowly.

What Matters — make-or-break questions, most important first

What is the sustainable ex-true-up operating margin? Current answer: Q2'26 op income $26.9M − $23.6M true-ups ≈ $3.3M (1.3%); Q3 adj EBITDA guided $10–16M. Cash interest ~$37M/yr exceeds ex-true-up operating income. What changes it: Q3/Q4 prints with true-ups disclosed and adj EBITDA ≥$30M on flat ASP; GM holding >62% through the Detect build.
Does volume hold +20% once the sales realignment laps? Current: Q2 59,200 (+18% YoY, +12% QoQ, record +6,400 sequential); Q3 guide 61–62k. 50 new reps in 146 territories productive ~1H27. Changes it: Q3 ≥62k and Q4 ≥65k = bull; Q3 <60k = the "growth halves" math is right on both legs.
Is the price step really done, or is it going the other way? Current: company base ASP $4,121 (Q1) → ">$3,850" (Q2) → $3,800–3,900 (Q3 guide) = −6% in two quarters, more than the blood mix shift (17.4% → 18.1%) explains; tissue/blood ASP disclosure discontinued in Q2. Changes it: tissue ASP ≥$4,300 confirmed; commercial covered lives 132M → 200M with realization rising; true-ups stay positive.
Is the CY2027 Assure rate ($5,250) final, and is 0211U safe under MolDX/NCD policy? Current: preliminary; final ~November 2026. Worth ~$26–38M FY27 revenue at 40% Medicare mix on 43–60k Assure cases. Changes it: final CLFS; RESULTS Act (would freeze PAMA to 2029); any MolDX coverage narrowing on comprehensive panels.
DOJ False Claims Act CID — scope and reserve. Current: March 2025 CID on the 14-day rule, "too early to assess," no reserve, prior $2.9M settlement 2022. Changes it: closure, a settlement <$25M (benign), or >$100M with a Corporate Integrity Agreement (scenario 3).
Does Detect ship volume, and does Achieve-2 hold up prospectively? Current: zero disclosed units; "demand outstrips $1B capacity"; DTC campaign ~Oct-2026; Achieve-2 (n≈25k) late-2026/2027. Changes it: ≥5k units/month by the Q4 call; stage I–II sensitivity ≥50% prospectively.
Can MRD get MolDX coverage as the 4th–5th entrant? Current: tumor-naive CRC first, launch H2'26; MolDX submission "compiling relapse data." Changes it: coverage decision date; validation data vs Signatera.
Does the pharma line recover ($58M+ needed in H2'26 on $17.2M RPO)? Current: $16.8M 1H26; guide trimmed to $75–80M. Changes it: Q4 deliverables; a named multi-year deal.
Who is the marginal seller? Current: Whitney cut 6.5% → 4.0% in H1; Spetzler/Brille $27M in September; 26.5M options/RSUs; Halbert 25.0M shares pledged. Changes it: 13F/13G refresh post-9/30; Form 4 footnotes (10b5-1 or not); pledge disclosure.
Will management keep the 30% adj-EBITDA reinvestment cap, and where does the $300M DDTL go? Current: "nothing in the pipeline" for M&A; capex ~$60M FY26. Changes it: an acquisition into MRD/MCED (dilutes the organic story either way).

What Changed Since the Prior Version (Stage 12)

No TRACKER file exists for CAI. The reference point is the same-day draft build (2026-09-25 23:35). This rerun incorporates the red team's verification pass and the technicals agent's computed levels verbatim.

Score

6.5 → 6.0

Red team's ex-true-up margin math + Q3 guide-down

Thesis

INTACT → ON WATCH

Moat 6 → 6.5 (qual's vector-by-vector score adopted)

Gate tally

9/5/2 → 8/6/2

PASS / WATCH / FAIL · Gate 16 moved PASS → WATCH

Diff panel — every item that moved

Date	Change	Effect
2026-09-26	Red team verified 4/1/26 8-K: term loan = Term SOFR + 5.00% (base + 4.00%), $50M minimum qualified-cash covenant, $300M DDTL, $500M incremental. Quant's "SOFR + 4.00% with 2.00% floor" was the base-rate margin.	Interest ~$37M/yr confirmed; covenant trivially met.
2026-09-26	Fully diluted share count corrected 298M → ~305M (10-K: 21.9M options + 4.5M RSUs + 0.15M ESPP; treasury method).	All per-share scenario math in Valuation recomputed; quant's figures were ~3% too high.
2026-09-26	Medicare share: 39.6% of profiling revenue FY25 (10-K) is the only verified figure; qual's 38.9%/49.7%/51.6% could not be sourced to the Q2'26 10-Q.	Concentration section uses 39.6%.
2026-09-26	H2'26 gross-margin guide corrected to "in that 60%" range (Q2 transcript). Qual's "high-60s" was wrong.	Gate 16 (operational efficiency) downgraded PASS → WATCH.
2026-09-26	J.H. Whitney stake: 6.5% (proxy record date) → 4.0% / 11.26M sh at 6/30/26 (13G/A). Sold ~7M shares in H1'26.	Qual's "not a distribution" read revisited; supply overhang is real.
2026-09-26	CY2027 CLFS treated as preliminary (posted Sept 2026; final ~Nov). Macro's "imminent" framing was stale; qual/quant were current.	Validator/breaker dated to the November final.
2026-09-26	Fed action 9/16/26 = +25bp hike (macro, CNBC/Fox), not a cut (sentiment's aggregator source).	Macro stamp stays 🟡; the 9/17 +9.6% day was a Fed-relief/sector day.
2026-09-26	Ex-true-up TTM net income: quant's "strip ~$153M" double-counted intra-year true-ups on in-window cases. Defensible haircut is ~$60–90M, giving TTM GAAP NI ex-true-ups ≈ +$15–45M, not −$50M.	Gate 14 stays WATCH; the direction of quant's point stands, magnitude softened.
2026-09-26	Interest coverage restated: ex-true-up operating income run-rate ≈ $13M vs ~$37M cash interest (macro's "a third of op income" understated the problem 3x).	Feeds the operating-leverage argument (Financials) and Red Team scenario 5.
2026-09-26	Entry ladder replaced with the technicals agent's computed levels (no-chase $32.90; zones $24.5–26.3 / $22.5–23.6 / $20.3–21.5). Draft build's $30 no-chase and $25.0–26.5 / $20.5–22.5 / $17.0–18.5 zones are superseded.	Trade Plan.
2026-09-26	Score 6.5 → 6.0; thesis INTACT → ON WATCH; moat 6 → 6.5 (qual's vector-by-vector score adopted).	Verdict, Checklist.

Prior claims KEPT

Conviction-spec classification (moat < 9)
No-chase at current price
Macro 🟡
Skin-in-the-game 196x / 1,777x
TTM FCF $124.0M correction of the triage
9/15–9/18 rally was sector beta with no Caris filing
Net cash positive under either debt framing (draft used $452.4M gross debt incl. leases → $341M net; this version uses $400M face → $393M net; both cited)

Prior claims MISSED / corrected

Draft's "moat 6" and "FCF-inflection window closed in 2025" retained — but the draft's ladder and no-chase line are replaced
Draft's Gate 16 PASS is now WATCH
Draft's probability-weighted $35.9 is replaced by $30–33 on 305M shares and the red team's distribution

The 16-Gate Checklist — prior draft → v1.0

Tally: prior 9 PASS / 5 WATCH / 2 FAIL → v1.0 8 PASS / 6 WATCH / 2 FAIL. Gate 16 moved down. The two FAILs decide sizing (moat) and timing (P/S band).

8/16gates pass

PASS 8 (prior 9) — 2, 5, 6, 8, 10, 12, 13, 15

WATCH 6 (prior 5) — 1, 3, 7, 9, 14, 16

FAIL 2 (prior 2) — 4 moat (sizing), 11 P/S band (timing)

Pass

Watch

Fail

WATCH

01Growth 30–50%+

Q2'26 rev +45.4%; FY26 guide $1.03–1.04B (+27–28%); H2'26 implied +8.9% vs 2H25 ($555M vs $509.7M, the latter carrying ~$120M of true-ups); FY27 consensus $1,215M (+18%). Volume is the honest number: +18% Q2, ~20% guided.

prior WATCH → WATCH

PASS

02FCF positive (fading)

TTM $124.0M (OCF $168.6M − capex $44.6M); 5 straight positive quarters since Q2'25; but $55.3 → $39.7 → $22.5 → $6.4M and Q3 guided neutral.

prior PASS → PASS (fading)

WATCH

03Visible flywheel

Data → new assays half spins (3 launches in 2026); assays → demand → pharma half does not: pharma R&D $11.5M Q2 (−38% YoY), volume +18% vs peers +31–63%.

prior WATCH → WATCH

FAIL

04Moat ≥9/10 (Tier-1 gate)

Process power/data scale 7/10 is real (1.13M profiles, 845k with outcomes); switching costs low, network effects absent, price regulator-set, lab commoditized, 4th–5th into MRD. Conviction-spec sizing.

prior FAIL (6) → FAIL (6.5)

PASS

05Founder-led

David Halbert founded 2008; Chairman + CEO; 43.9% beneficial (DEF 14A 4/23/26). Survived a −67% post-IPO drawdown (the "rite of passage," not quite −80%).

prior PASS → PASS

PASS

06Skin in the game ≥5x (ideal tier)

125.0M sh × $31.98 = $3.998B ÷ $20.35M 2025 comp = 196x; ÷ $2.25M normalized 2024 comp = 1,777x.

prior PASS → PASS (ideal tier)

WATCH

07Insider net activity 12mo

Open-market: ~$32.5M sold (Spetzler $14.8M, Brille $12.1M, Poste $3.25M, Power $1.7M, Denton $0.24M) vs ~$1.6M bought (J. Halbert $0.99M @ $14.56, Vacirca $0.50M @ $16.15). Net −$31M = 0.7% of insider holdings; CEO zero sales; 25.0M CEO shares pledged. Option-expiry housekeeping, but all selling above $24 and all buying below $17.

prior WATCH → WATCH

PASS

08Institutional adds/trims

12-mo 13F: 116 buyers / 19 sellers; latest quarter $842M in vs $216M out. New: PointState 6.83M sh, BlackRock, CPPIB. Register is getting faster-money (Coatue, Braidwell). Whitney sold ~7M sh H1'26.

prior PASS → PASS

WATCH

09Earnings beat history (8)

Only 5 public prints. EPS 3/5 beats (Q2'25 miss on deemed dividend; Q2'26 hairline miss). Revenue beats + raises in Q3'25 and Q2'26; consensus for Q1'26/Q2'26 revenue unverified. Cannot grade 8 consecutive.

prior WATCH → WATCH

PASS

10Explainable to an 8-year-old

"They read all of a tumor's DNA and RNA and tell the doctor which medicine will work; insurance pays about $3,850 per answer."

prior PASS → PASS

FAIL

11P/S band (for entry)

9.1x TTM = "aware/optimistic" boundary; hunting ground 2–4x = $7–14; "aware" 4–8x = $14–28. Was 4.5x at $14.55 in May 2026.

prior FAIL → FAIL

PASS

12No shareholder dilution (overhang noted)

Basic 281.2M → 282.6M YoY (+0.5%); zero converts/warrants; $100M buyback, $82.1M left. Overhang: 26.5M options/RSUs/ESPP = 9.4% of basic, many struck ~$2.44.

prior PASS → PASS (overhang noted)

PASS

13Current ratio ≥2:1

5.5x ($1,058M / $191M); quick 4.8x.

prior PASS → PASS

WATCH

14Net margin positive

TTM GAAP NI $105.1M (10.6%) inflated by ~$81M Q4'25 prior-period true-ups; H1'26 GAAP NI −$1.1M; ex-true-up TTM NI ≈ +$15–45M (red-team haircut). Q2 ex-true-up op margin ~1.3%.

prior WATCH → WATCH

PASS

15Cash > debt

$793.1M cash + restricted + ST securities vs $400M face ($393.0M carrying) → net cash $393M; $50M min-cash covenant; maturity 2031-04-01.

prior PASS → PASS

WATCH

16Operational efficiency rising

GM 43.7% (Q3'24) → 68.1% (Q2'26) and op margin −60% → +10%: done. But GM guided to ~60% for H2, OCF/capex 8.8x → 1.3x, Q3 adj EBITDA $10–16M vs $55.7M, opex guide raised to $595–600M. Reinvestment phase is now.

prior PASS → WATCH ↓

The two FAILs decide everything. Gate 4 (moat 6.5 < 9) caps sizing at conviction-spec, ≤5%. Gate 11 (9.1x TTM P/S) says the entry is not here — it was at 4.5x in May. Everything in WATCH is a Q3-print question.

Financials — the honest pictures (Stages 2 + 4)

One chart tells the story: reported revenue is ex-true-up revenue plus catch-up cash on old cases. Strip the catch-up and the line flattens as the $8,455 price step anniversaries. Volume is the honest growth number.

Revenue by quarter: ex true-ups vs prior-period true-ups ($M)

Q4'25 carried ~$81M of prior-period true-ups at ~100% margin · Q4'24 true-up split n/a · "~" values back-solved, ±1%

Clinical cases — the leading indicator

Tissue (MI Profile) + blood (Caris Assure) · Q2'26 record +6,400 sequential adds · Q4'24 split n/a

Profiling revenue per case ($) — reported vs ex true-ups

Ex-true-up ASP flat at ~$3,800–3,860 for three quarters — the price leg is done

Cash generation by quarter ($M) — OCF, CapEx, FCF

FCF $55.3M → $6.4M in four quarters; OCF/CapEx 8.8x → 1.3x; Q3 guided FCF-neutral

7a. Capital-structure six-pack (Stage 2)

Item	Value	Source
Price	$31.98	2026-09-25 close
Shares	282,595,186 basic (10-Q cover, 2026-08-03); ~305M fully diluted (treasury method on 26.5M options/RSUs/ESPP)	Q2'26 10-Q; FY25 10-K
Market cap	$9,037M (sanity: 31.98 × 282.6M ✓)
Cash + ST securities	$793.1M = $690.9M cash + restricted + $102.2M marketable	6/30/26 10-Q
Debt	$400.0M face (2026 Term Loan, Blue Owl/Blackstone, Term SOFR + 5.00% / base + 4.00%, interest-only yr 1, matures 2031-04-01, $50M min qualified-cash covenant); $393.0M carrying; Q2 interest $9.2M (~9.2% all-in). $300M DDTL (acquisitions only, through 2027-08-01) + $500M uncommitted incremental. No converts, no warrants.	8-K 4/1/26 (red-team verified); 10-Q
EV	$8,644M on face debt ($8,697M incl. leases per yfinance)
Short % float	10.15% (13.68M sh, latest) vs 11.58% (13.33M sh, prior print); float 134.9M; days-to-cover 2.4–2.6. CONFLICT: more shares short yet lower % = float denominator changed; unreconciled.	stockanalysis / yfinance
Share count YoY	281.2M (2025-08-07) → 282.6M (2026-08-03): +0.5%	10-Q covers
Buyback	$100M authorized 2026-06-07; ~1.0M sh / $17.9M executed Q2'26 (cash paid $21.4M incl. tax-withholding shares); $82.1M remaining. 1.44M sh / $16.6M repurchased in 2025.	Q2'26 10-Q
Insider ownership	Halbert 43.9% (125.0M sh, DEF 14A); officers + directors 50.0% (145.9M sh); FMR 10.8% (3/31/26); Sixth Street 6.9%; Whitney 4.0% (6/30/26). yfinance "insiders 48.8% + institutions 56.3%" double-counts Halbert-affiliated entities.	DEF 14A; 13G/As

7b. Eight-quarter longitudinal table ($M; volumes from press releases; "~" = back-solved from YoY %, treat ±1%)

Metric	Q3'24	Q4'24	Q1'25	Q2'25	Q3'25	Q4'25	Q1'26	Q2'26
Revenue	101.6	129.9	120.9	181.4	216.8	292.9	216.2	263.7
— Molecular profiling	93.8	94.4	114.1	162.9	207.6	282.1	210.8	252.3
— Pharma R&D services	7.8	35.5	6.8	18.5	9.2	10.8	5.4	11.5
Revenue growth YoY	n/a	n/a	49.9%	81.3%	113.4%	125.4%	78.8%	45.4%
Prior-period true-ups in revenue	6.8	n/a	~3.9	12.0	38.3	~81*	~10.4	23.6
Revenue ex true-ups	94.8	n/a	~117.0	169.4	178.5	~212	~205.8	240.1
Clinical cases (leading indicator)	42,956	~43,900	~45,900	~50,170	50,763	52,700	52,800	59,200
— MI Profile (tissue)	38,409	n/a	~40,100	~43,040	43,226	44,150	43,600	48,300
— Caris Assure (blood)	4,547	n/a	~5,800	~7,130	7,537	8,550	9,200	10,700
Cases YoY	n/a	n/a	~30%	~18%	18.2%	20%	15%	18%
MP revenue / case (reported)	$2,184	n/a	$2,486	$3,247	$4,090	$5,353	$3,992	$4,262
MP revenue / case ex true-ups	$2,025	n/a	~$2,400	~$3,008	~$3,335	~$3,816	~$3,795	~$3,863
Gross margin	43.7%	55.7%	47.2%	62.7%	68.0%	76.1%	65.4%	68.1%
Operating income	−60.8	−37.1	−58.0	−18.0	32.6	88.4	5.3	26.9
Op margin	−59.9%	−28.5%	−47.9%	−9.9%	15.1%	30.2%	2.4%	10.2%
Adj. EBITDA (company)	n/a	−23.0	−36.2	16.7	51.2	106.1	26.2	55.7
GAAP net income	−67.7	−36.9	−102.6	−71.8	24.3	82.0	−0.5	−0.6
EPS diluted (actual / est)	−2.59 / n/a	−1.73 / n/a	−3.57 / n/a	−1.11 / −0.22	0.08 / −0.14	0.28 / 0.10	−0.00 / −0.02	−0.00 / 0.01
SBC	4.7	5.0	14.7	28.3	13.7	13.4	15.9	22.2
OCF	−69.4	−38.9	−31.3	7.3	62.4	44.8	32.9	28.5
CapEx	1.8	2.3	2.7	1.4	7.1	5.1	10.3	22.1
FCF	−71.3	−41.2	−34.0	5.9	55.3	39.7	22.5	6.4
OCF / CapEx	neg	neg	neg	5.2x	8.8x	8.8x	3.2x	1.3x
Diluted wtd shares (M)	35.6	35.5	35.6	64.9	297.2	296.1	282.6	282.9
Cash + ST inv (EoQ)	n/a	66.2	33.4	721.2	755.5	798.6	823.5	793.1
Debt face (EoQ)	400	400	400	400	400	400	400	400

* Q4'25 true-ups ~$81M per prelim 8-K; includes prior-quarter. Q2'26 detail: total costs $236.8M (COGS ~$84.1M; S&M $53.8M; G&A $66.5M; R&D $32.4M → opex ex-COGS $152.7M = 57.9% of revenue). Inventory/supplies +$47M QoQ to $123.8M (Detect capacity). AR $116.9M (~40 days). Contract liabilities $21.1M. Pharma RPO $17.2M. "Other expense" $25.1M = debt-extinguishment loss (non-recurring). Payer concentration Q2'26: Payer 1 37.3% / Payer 2 12.6% / Payer 3 15.6% of revenue; Payers 1+2 = 47% of AR; Payer 2 = 23.6% of AR on 12.6% of revenue (flag).

7c. GAAP NI ↔ OCF ↔ FCF reconciliation (TTM Q3'25–Q2'26)

Revenue $989.6M → GAAP NI +$105.1M → + SBC $65.1M → + D&A $20.8M → + extinguishment loss $25.2M → + other non-cash / working capital net −$47.7M → OCF $168.6M → − CapEx $44.6M → FCF $124.0M (12.5% margin)

Working-capital drag: Q4'25 AR build on true-up revenue −$59.1M; Q2'26 inventory build −$26.4M. Company 8-K quarters sum to $123.9M.

SBC = 6.6% of revenue and 62% of GAAP NI. FCF treating SBC as cash ≈ $59M (6% margin).
Quality flag (a): Q4'25 GAAP NI $82.0M carried ~$81M of prior-period true-ups at ~100% margin; red-team haircut for out-of-window true-ups is ~$60–90M, so TTM NI ex true-ups ≈ +$15–45M.
(b) FY25 carried −$52.3M fair-value charges and −$19.9M extinguishment (IPO-era instruments); none in 2026.
(c) GAAP EBITDA Q2'26 ~$33M vs adj. $55.7M — the gap is SBC.
(d) Rule of 40 on TTM = 45.4 + 12.5 = 57.9; on forward guide ≈ 27.5 + ~3 (FY26 FCF tracking $20–40M) ≈ 30 → PARTIAL.

7d. Operating-leverage argument — and where it fails

The leverage happened: gross margin 43.7% → 68.1% in seven quarters as $8,455 MI Cancer Seek replaced legacy panels on a largely fixed lab cost base; operating margin swung −60% → +10%; opex ex-COGS fell from >100% of revenue to 58%. That is a completed event, not a forward one.

Where it fails: (1) the incremental margin came from price, and ex-true-up price is flat; (2) management is deliberately spending it — opex guide $595–600M (57–58% of FY26 revenue, flat as a %), capex ~$60M+ (from $13M FY25), GM guided down ~8 points to ~60% for H2 on Detect/MRD capacity; (3) Q3 adj EBITDA $10–16M = 4–6% margin vs 21% in Q2; (4) ex-true-up operating income run-rate (~$13M annualized off Q2) does not cover ~$37M of cash interest. The leverage returns only if volume compounds at 20% on the new cost base — 2027, not 2026. Framework leading indicators both point down: OCF/CapEx 8.8x → 1.3x; FCF ÷ op margin 55.3/15.1 → 6.4/10.2.

7e. Balance-sheet gates

Current ratio 5.5x (PASS ≥2:1); quick 4.8x.
Net cash $393M; gross debt / TTM adj EBITDA 1.7x; net debt / EBITDA −1.6x; interest ~$37M covered ~6x by TTM adj EBITDA (but only ~2–3x by Q3-guided run-rate).
Liquidity vs commitments: $793M vs ~$90M annualized capex + $47M inventory + $82M buyback → 5+ years even at zero FCF. Covenant $50M min cash — trivial.
Open flags: unremediated material weakness in ICFR (FY24 audit, still open 12/31/25); DOJ CID, no reserve; AR concentration 47% in two payers.

7f. Debt schedule / converts

Single instrument: $400M 2026 Term Loan, interest-only through ~2027-04, matures 2031-04-01, floating (each +25bp SOFR ≈ +$1.0M/yr; no hedge disclosed — open). Prepayment premium exists, unquantified (open). Refi replaced the 2023 OrbiMed/Braidwell loan (SOFR+6.5%, 10.5% at YE25, 1% exit fee) — ~$6M/yr saving; $25.2M extinguishment loss booked Q2'26. No converts, no warrants, no ATM.

7g. Organic vs acquired; anniversary

100% organic — no acquisitions FY25/1H26. But FY24→FY25 profiling revenue +$417.6M = $52.0M MI Profile volume + $311.4M MI Profile ASP/mix + $54.2M Assure (10-K bridge). The ASP step began Q1'25 and fully anniversaried by Q1'26; Q3'25 ($38.3M) and Q4'25 (~$81M) true-up comps make 2H26 reported growth ~+9%. Medicare pays $8,455 for MI Cancer Seek; blended realized ASP ~$3,850 → commercial/MA realization is a fraction of list: each point of commercial coverage is upside, MA denials are the risk. Assure MolDX $3,649 (→ $5,250 prelim 2027); ChromoSeq $3,228; Detect $3,500 cash. Both core codes are CDLT (PAMA triennial), not ADLT.

7h. Dilution history

Pre-IPO 36.5M basic (12/31/24) → IPO (2025-06-18): 26.6M new shares + preferred conversion + all warrants exercised (Sixth Street's $50M debt-to-Series C at $1.61 among them) → ~281M basic (8/7/25); diluted wtd 297.2M in Q3'25 (includes ~15M in-the-money options). Buybacks: 1.44M sh 2025, 1.0M sh Q2'26 → 282.6M basic (8/3/26). Outstanding overhang (10-K): 21,885,508 options + 4,494,342 RSUs + 146,450 ESPP = 26.5M (9.4%). Q2'25 "EPS −$7.97" on aggregator sites = net loss to common incl. ~$445M IPO preferred deemed dividend; GAAP net loss was −$71.8M.

Management (Stage 5)

Skin-in-the-game score

Score = insider equity value ÷ annual total comp · gate ≥5x · log scale 5x weak → 50x strong → 1,000x ideal

Halbert: 125,009,756 sh × $31.98 = $3.998B ÷ $20.35M (2025) = 196x · ÷ $2.25M (normalized 2024) = 1,777x → PASS, ideal tier

Insider	Shares	Equity @ $31.98	Comp	Score	Grade
David Halbert (Chair + CEO)	125,009,756 (43.9%)	$3.998B	$20.35M 2025 (IPO-year: $15.4M RSU + $1.5M bonuses) / $2.25M normalized 2024	196x / 1,777x	PASS ideal tier; zero sales; 25.0M sh pledged
David Spetzler (President)	~1.18M direct post-exercise (+ remaining options n/a)	~$37.8M	$11.6M 2025 / ~$1.2M normalized	3.3x / 31x	MODERATE net seller Sept-2026
Officers + directors (group)	145.9M (50.0%)	$4.67B	—	—	—

Net 12-month insider activity (open-market only, 9/25/25–9/23/26)

Sells ≈ $32.5M

Spetzler 597,212 sh ≈ $14.83M on 9/3–9/4, sell-to-cover on 1.25M options @ $2.44 expiring 9/12, kept ~653k net new shares
Brille 400,000 @ $30.33 = $12.13M on 9/17, exercised-and-sold all, holds 230,303 after
Power 62,250 ≈ $1.69M on 12/11/25
Denton 7,500 @ $31.82 on 9/22
Poste 130,000 ≈ $3.25M Aug-2026 Form 144s
Unidentified 20,000-sh Form 144 8/3/26

Buys ≈ $1.60M

Jon Halbert 68,000 @ $14.56 on 5/15/26
Vacirca 31,050 @ $16.15 on 5/11/26
Denton 4,184 @ $24.09
Spetzler 400 @ $23.85

Net ≈ −$31M (agents ranged $28.9–32.5M sold; difference is Poste + the unidentified 144). Whether Spetzler/Brille were under 10b5-1 plans: Form 4 footnotes unread (open). Pledges (DEF 14A): Halbert 25.0M via Halbert Family Capital LLC; Jon Halbert 325k; Spetzler 150k. Loan sizes undisclosed.

Scores (0–10, evidence)

Trait	Score	Evidence
Management	6/10	Capital allocation +: conservative guides (FY25 raised twice, FY26 once, never cut); refinanced ~$6M/yr cheaper; $17.9M buyback near the lows; explicit ~30% adj-EBITDA reinvestment cap; "nothing in the pipeline" on M&A. Ethics/governance −: pledged shares, brother on board, 3% derivative-suit bylaw (Oct-2025), EGC exemptions (no say-on-pay), $925,773 company-paid personal security, stopped disclosing tissue/blood ASP in Q2'26, material weakness two years running. Long-term +: 40-year record; self-funded the WES/WTS pivot through a $245M OCF burn year.
Operations	7/10	Customer obsession: 6,000+ ordering physicians, EHR integration at ~3,100 sites (~75% e-orders) — sourcing unverified; only FDA-PMA'd WES/WTS CDx. Innovation: 3 launches in 2026 + MRD H2. Failure tolerance: took a visible Q1'26 volume hit to re-district sales and said so. Minus: slowest volume grower of four public peers; pharma line shrinking.
Structure	4/10	Talent density real at the top (Sledge ex-ASCO president CMO; Radovich CSO; Spetzler). Meritocracy: Glassdoor 2.9/5, 35% recommend, recurring "nepotism/favoritism" themes (326 reviews, Sept-2026 — weak evidence, URL not captured). Decentralization: none — 43.9% owner is CEO + Chair; 4 of 10 directors are ex-AdvancePCS.
Leadership-page triage (CTO > CPO > CEO)	🟡	CTO Valeriy Domenyuk PhD exists (internal builder, no external footprint found). No CPO, no Chief AI Officer; Chief Innovation Officer Jim Abraham PhD. Real technical leader is President Spetzler. CCO Bobby Hill came from reimbursement — telling: the growth engine is payer contracting. Yellow, not red.

Board read (10 seats)

AdvancePCS bloc 4/10: D. Halbert; Jon Halbert (brother); Laura Johansen (ex-AdvancePCS GC); Danny Phillips (ex-AdvancePCS CFO, Audit Chair).
Capital 2: Peter Castleman (lead independent, ex-J.H. Whitney chair, 3.7% personally; 45% of non-Halbert votes withheld from him 6/4/26); Brian Brille (Vice Chair/EVP, ex-BofA banker, insider, just sold 63% of direct stake).
Operators who matter: David Fredrickson (AstraZeneca EVP Oncology — pharma channel); Lloyd Minor MD (Dean, Stanford Medicine); Jeffrey Vacirca MD (CEO NY Cancer & Blood, OneOncology co-founder — community-oncology channel; bought the low); Joseph Gilliam (Glaukos COO).
Sixth Street (6.9%) and Fidelity (10.8%): no seat. No dual-class; 43.9% is de facto control without controlled-company disclosure. 33% of non-Halbert shares withheld from Halbert on 6/4/26.

Moat & Competition (Stage 6)

Moat score

6.5 / 10

Fails the ≥9 Tier-1 gate → conviction-spec sizing

Contestability

HIGH

Lab is a commodity, price is regulator-set, data lead is finite (~3–5 yrs)

Who is winning

Volume: GH > NTRA > TEM > CAI

ASP/margin/FCF: CAI · Data monetization: TEM (5x Caris's pharma line). Caris wins profit per test and is losing tests per quarter.

Moat vector	Score	Evidence
Process power / data scale	7	WES+WTS on every sample since 2019/20 → 1.13M profiles, 845k with outcomes; no competitor sequences that deep at that volume (TEM ~400k deep multimodal; FMI/GH panel-based). Competitors converging (TEM xT tumor-normal; NovaSeq X cost curve). Lead ~3–5 yrs.
Scale economies shared with customers	2	ASP went $2,300s → $3,850; gains kept by Caris, paid by Medicare. Fails J-Griff's "shared" test.
Switching costs	4	Low for oncologists (multi-vendor ordering normal; EHR integration mild friction). Moderate for pharma partners — 4% of revenue.
Network effects	3	Data-scale, not user-network. POA is a research network.
Brand	5	Strong in community oncology; weaker vs FMI in academic centers; FDA PMA adds credibility.
Regulatory / reimbursement	7	0211U at $8,455 (flat for 2027) + NGS NCD coverage via PMA is a real asset — and a single point of failure (39.6% Medicare, one code, PAMA triennial, DOJ CID).
Vertical integration	6	Owns Phoenix lab; supplier diversification underway (Illumina agreement through 2033, no minimums; revalidation risk on a PMA assay).

Competitor comparison (Q2'26 prints; multiples yfinance TTM 2026-09-25)

	CAI	Guardant (GH)	Natera (NTRA)	Tempus (TEM)	Foundation Medicine (Roche)	Exact → Abbott
Q2'26 revenue / YoY	$263.7M / +45%	$335.0M / +44%	$753M / +38%	$382.5M / +22%	private	acquired 3/23/26
Volume growth	+18% (tissue +13%, blood +50%)	oncology +63%; Shield 66k (+300%)	Signatera 283k/qtr (+56%)	oncology +31%	n/a	n/a
Gross margin	68% (guided ~60% H2)	64–65% non-GAAP	n/a	n/a	n/a	n/a
FCF	+$6.4M Q2; +$124M TTM	negative; breakeven target end-2027	positive	−$245M FY25	n/a	n/a
Data / pharma revenue	$11.5M/qtr (−38%)	n/a	n/a	~$87M/qtr (~25% of rev; NRR 126%; TCV >$1.1B)	deepest pharma CDx	n/a
FY26 guide	$1.03–1.04B (+27–28%)	$1.34–1.36B (+36–38%)	n/a	~$1.6B	n/a	n/a
Market cap / P/S TTM	$9.0B / 9.1x	$23.8B / 20.1x	$59.5B / 22.0x	$15.3B / 10.7x	n/a	EV ~$23B ≈ 7–8x sales at takeout
Position	Tissue-comprehensive leader; #1 profit/test; 4th–5th into MRD; MCED cash-pay	Blood therapy-selection + CRC screening leader (Shield FDA + Medicare)	Owns MRD (1.1M+ tests/yr, Medicare)	Data monetization leader; 50% of US oncologists connected; buying Personalis (MRD)	Tissue CDx incumbent; bought SAGA (MRD) Apr-2026	Cologuard/Cancerguard with Abbott's balance sheet

Other peers for multiple context: VCYT $3.7B / 6.5x / +16% / 20% EBITDA margin; NEO $2.5B / 3.3x; ILMN 9.1x.

Earnings Tracker (Stage 12)

Prints (5 public; 3 pre-IPO quarters had no estimates)

Quarter	Date	Revenue (YoY)	EPS actual / est	Reaction	Guidance action
Q3'24	pre-IPO	$101.6M	−2.59 / n/a	—	—
Q4'24	pre-IPO	$129.9M	−1.73 / n/a	—	—
Q1'25	pre-IPO	$120.9M (+50%)	−3.57 / n/a	—	—
Q2'25	2025-08-12	$181.4M (+81%)	−1.11 / −0.22 MISS	n/a (IPO charges; −$7.97 to common incl. deemed dividend)	FY25 guide set
Q3'25	2025-11-05	$216.8M (+113%)	0.08 / −0.14 BEAT	n/a	FY25 RAISED ($720–730M)
Q4'25	2026-02-26	$292.9M (+125%)	0.28 / 0.10 BEAT	n/a	FY26 set $1.00–1.02B (prelim 1/12/26 ahead)
Q1'26	2026-05-07	$216.2M (+79%)	−0.00 / −0.02 BEAT (yfinance; Investing.com est −0.12 — CONFLICT)	−18.6% next day ($19.84 → $16.15); −26.7% at 5 days	FY26 reiterated; cases +15% vs ~20% target
Q2'26	2026-08-05	$263.7M (+45%)	−0.00 / 0.01 hairline MISS	+21.6% next day ($16.45 → $20.00); +53.9% at 20 days	FY26 RAISED to $1.03–1.04B; pharma trimmed $75–85M → $75–80M; opex raised $590–595M → $595–600M

Revenue vs consensus for Q1'26/Q2'26: unverified (prior draft cited "5/5 revenue beats, avg +18.4%" from tvremix — not reproduced this pass). Streak: 4 consecutive revenue beats with two raises; EPS 3/5.

Promises vs delivered (last 4 calls)

Promise / target	Made	Status · Sep 2026	Grade
FY25 rev $720–730M	Q3'25 call (2025-11-05)	$812.0M incl. true-ups	KEPT
Volume +21–22%	Q3'25 call	22%	KEPT
FY25 GM ~62%	Q3'25 call	66%	KEPT
Tissue ASP >$4,000 in 1H26	Q3'25 call	Company base $4,121 Q1; qual's "$4,091" unverifiable	KEPT
Achieve-1 readout 1H26	Q3'25 call	Delivered	KEPT
MRD-CRC MolDX submission "shortly"	Q3'25 call	Still compiling data Aug-2026	SLIPPED
Adj-EBITDA capped ~30% for reinvestment	Q3'25 call	Q3'26 guided $10–16M	KEPT
FY26 rev $1.00–1.02B	Q4'25 call (2026-02-26)	Raised to $1.03–1.04B	RAISED
Pharma $75–85M	Q4'25 call	1H $16.8M; trimmed to $75–80M	AT RISK
Opex $590–595M	Q4'25 call	Raised to $595–600M	RAISED
Capex ~$60M	Q4'25 call	Tracking ($22.1M Q2, $15–20M Q3)	TRACKING
Positive adj EBITDA + FCF FY26	Q4'25 call	On track (1H FCF $28.9M)	ON TRACK
Tissue ASP $4,000 Q1 → $4,200 FY avg	Q4'25 call	Disclosure discontinued Q2	WATCH
Blood ASP $2,400–2,500	Q4'25 call	Q1 "just under $2,500"	KEPT
Detect launch Q2	Q4'25 call	~1 month late, 7/1	SLIPPED
ChromoSeq pending MolDX	Q4'25 call	4/1, $3,228	KEPT
300 reps	Q4'25 call	July	KEPT
No downward PAMA	Q4'25 call	0211U flat, 0485U +44% prelim	KEPT
Q1 growth 70–74%	Q4'25 call	79%	BEAT
Q2 cases >58,000 (~47,500 tissue / >10,000 blood)	Q1'26 call (2026-05-07)	59,200 / 48,300 / 10,700	BEAT
Q2 capex ~$30M	Q1'26 call	$22.1M	UNDER
Detect "later in Q2"	Q1'26 call	7/1	SLIPPED
NY State Assure approval	Q1'26 call	Pending	PENDING
PAMA data submitted 5/1, no downward adjustments	Q1'26 call	Confirmed	KEPT
Detect revenue contribution assessed at Q2 call	Q1'26 call	Not addressed	PUNTED
PAMA price increase in Sept	Q2'26 call (2026-08-05)	9/21 prelim	KEPT
FY26 $1.03–1.04B; Q3 volume ~20%, cases 61–62k; Q3 blended ASP $3,800–3,900; GM "in that 60%" range H2; Q3 adj EBITDA $10–16M; FCF neutral Q3 / positive Q4; MRD launch 2H26 (tumor-naive CRC first); MI Clarity V2 2H26; Detect capacity $1B → $3B, DTC campaign ~Oct; Assure covered lives 131.9M → 200M by YE26; pharma Q4-weighted; supplier diversification; no M&A	Q2'26 call	Graded at the Q3 print	PENDING

Scorecard: 14 kept / 3 slipped (Detect timing, MRD MolDX, NY State) / 1 at-risk (pharma FY26) / rest pending. Pattern: revenue and ASP promises are sandbagged and beaten; pipeline/regulatory dates slip 1–2 quarters.

Sentiment & Positioning (Stage 7)

Contrarian gate

CAUTION

DO NOT INITIATE HERE — "institutionally discovered, retail-forgotten, insider-sold," crowded on the sector trade

Street mean PT

$28.15 / $26.50

stockanalysis (14) / MarketBeat (13) · −12% / −17% vs spot · 10/14 Buys with mean PT under spot

Short interest

10.15%

of float, falling (covered ~8.5% into the rally); float only ~48% of shares; DTC 2.4–2.6

Narrative map — three stories, what price discounts

N1 · "AI TechBio data platform / Tempus-lite at half the multiple"

Discounted: YES — this drove $14 → $32 and the Sept rally was TEM-sympathy (9/15 +9%, 9/17 +9.6% on a Morgan Stanley reimbursement note, no Caris filing). Problem: the data business is $11.5M/qtr and down 38% YoY.

N2 · "Reimbursement-driven margin machine"

Discounted: MOSTLY — 68% GM, 5 straight FCF+ quarters, ~85x forward P/E already assumes it. Tail risk both ways sits in the CLFS final (Nov) and MA/commercial realization.

N3 · "Detect (MCED) is the free option"

Discounted: PARTIALLY by long-horizon bulls (TIKR $95–119 by 2030); the Street explicitly is not paying for it (Wolfe, UBS, JPM). Zero revenue disclosed — the one narrative that can surprise either way.

Analyst view

stockanalysis: 14 analysts, 9 Strong Buy / 1 Buy / 4 Hold / 0 Sell; PT low $22 / mean $28.15 / high $36 (mean −12% vs spot). MarketBeat: 13 analysts, 7 Buy / 5 Hold / 1 Sell (Weiss, algorithmic); consensus Hold; mean $26.50 (−17%). Mean PT trend: $39 (Sep-25) → $38 → $33 → $28 (Jun-26) → ~$28; Holds 0 → 4; coverage 8 → 14.

Goldman Buy, PT $27 → $35 (9/21 — the only PT above spot; rationale unread)
UBS init Neutral $28 (9/2–9/9)
BTIG Buy $32 → $33 (8/6)
Evercore Outperform $26
JPMorgan Overweight, PT cut $30 → $25 after the beat-and-raise (8/6; reason unpublished)
Wolfe → Peer Perform (6/1, "wants to see if the salesforce realignment bears fruit")
Citi $35 → $28 (5/8; a later "$34 → $42" TipRanks headline unverified)

Read: "buy the business, price is past our targets" — stale-target setup that resolves either with PT-chasing upgrades on a clean Q3 or valuation downgrades.

Positioning

Who holds it

Retail absent: StockTwits 1,063 lifetime messages, ~37 poll watchers; no Reddit threads surfaced (search blocked from this host — inferred from absence). Substack/SA voices turning cautious on price, not business (SA contributor Buy → Hold at $28.70 FV; DeltaSheets "premium for platform expansion before proof").
Institutions: 116 buyers / 19 sellers over 12 months; latest quarter +$842M / 38.0M sh in vs −$216M / 8.5M out; top holders Coatue $265M, T. Rowe $248M + $130M, Braidwell $147M, PointState $122M (new), BlackRock $99M (new), FMR 30.6M sh / 10.8% (3/31/26).
Short interest 10.15% of float and falling (covered ~8.5% into the rally); float only ~48% of shares.
Insider tape: buys only at $14.56–16.15 (May); sells $27M at $24.76–30.33 (Sept). Whitney 18.3M → 11.3M sh in H1'26.

Marginal buyer / seller

Marginal buyer: hedge funds / crossover (PointState, Coatue, Braidwell) and index/quant (BlackRock, Amundi) chasing the August print and the Sept sector squeeze, plus short covering.

Marginal seller: insiders exercising options, Whitney, and the company's own float structure (26.5M options/RSUs). Company buyback $82.1M remaining is a signal, not a floor.

What the market gets wrong

(a) Bearish — growth-rate illusion: 45% Q2 growth is ~2/3 price; ASP anniversaries in Q4'26/Q1'27; consensus already has Q1'27 rev $263M vs Q4'26 $294M and FY27 +14–18%. A 9x EV/S name whose growth halves in three quarters is mispriced unless Assure/Detect/MRD fill the gap — the sector-sympathy buyers are not modeling this.
(b) Bearish — insider tape vs narrative: nobody inside bought above $16.15; the first open-market sells since IPO landed at $24.76–30.33; a founding PE holder cut its stake 38% in six months; the market treats a 48% float as fixed.
(c) Bullish — Street anchoring: 10/14 Buys with mean PT under spot is a lag, not a bear signal. Estimates only moved after the Aug 5 print (+53.9% 20-day drift, biggest in the name's history). A Q3 print ≥$265M with ≥62k cases forces the $25–28 cohort to chase. This is the only clean bull argument in the sentiment data — and it is a trade into an event, not an entry.

Contrarian verdict: neither hated nor retail-crowded; "institutionally discovered, retail-forgotten, insider-sold," crowded on the sector trade. Framework: "high P/S after a crash = still crowded" — P/S ~9 after +125% off the low with fundamental growth about to halve is CROWDED. Gate: CAUTION / DO NOT INITIATE HERE. The "despair + executing" setup existed at $15–19 in May–June with insider buys; it closed. Wait for post-Q3 re-anchoring (Street's fair value $24–28), a PAMA/MolDX headline, or a sector unwind.

Valuation (Stage 8) — lenses at $31.98

Lens	CAI	Framework read
P/S TTM ($989.6M)	9.1x	"aware" (4–8) / "optimistic" (>10) boundary
EV/S TTM	8.7x
EV/S FY26 guide ($1,035M)	8.4x
EV/S FY27 consensus ($1,215M)	7.1x
P/FCF TTM ($124M)	73x	FY26 FCF tracking $20–40M → 230–450x
EV / adj. EBITDA TTM ($239M)	36x	EV/GAAP EBITDA ~50x
Trailing P/E (real: TTM NI $105.1M ÷ ~290M)	~89x	aggregator "3.34x / EPS $9.58" is garbage — do not cite
Forward P/E FY27 ($0.376)	85x
PEG	3.1	on 27% revenue growth; EPS-based PEG meaningless → FAIL

Own history (P/S on then-TTM revenue)

IPO $21 → ~11x; 2025-08-22 high $39.30 → 20.7x ($534M TTM); 2026-05-15 low $14.55 → 4.5x TTM / 4.1x FY26 guide (the framework's "attractive" zone, four months ago); now 9.1x. Monthly closes: Jun-25 26.72 · Aug 38.39 · Nov 25.53 · Feb-26 20.14 · May 16.72 · Jul 15.67 · Aug 25.49 · Sep 31.98.

What re-rated $14.55 → $31.98 (+120% in 4 months)

8/5 Q2 beat + raise + fifth FCF-positive quarter
$100M buyback with $17.9M executed near the lows
Director buys at $14.56/$16.15
Detect launch + capacity-tripling narrative
Preliminary CY2027 CLFS left 0211U flat and raised 0485U 44% (~$26–38M FY27 revenue — the market added ~$2.3B of cap 9/13 → 9/25)
11.6% short float squeeze fuel
Blue Owl/Blackstone refi removed refi risk

Offsetting: $27M of insider sales into the rip.

Peer multiples

GH 20.1x P/S (+44%, EBITDA-negative); NTRA 22.0x (+38%); TEM 10.7x (+22%, EBITDA-negative); VCYT 6.5x (+16%, 20% EBITDA margin, 23x fwd P/E); NEO 3.3x; ILMN 9.1x. M&A comp: Abbott/EXAS $105/sh, EV ~$23B ≈ 7–8x sales at ~+10% growth. Verdict: CAI at 8.7x EV/S with positive FCF is cheaper than GH/NTRA growth-adjusted, in line with TEM, rich vs VCYT/NEO, and roughly at the strategic-takeout multiple. Fair-to-full; not euphoric.

Bull / Base / Bear — 12–24 months

305M fully diluted; P/S band × FY27 revenue ÷ shares; buyback ignored.

Scenario range vs spot

Bear $13.8 · Base $31.5 · Bull $53 · spot $31.98 · probability-weighted 25/50/25 = $32.5 · red team's distribution $28–29

BEAR $13.8−57%

BASE $31.5−1%

BULL $53+66%

PW $32.5

SPOT $31.98

$10$20$30$40$50$60

	Bear	Base	Bull
FY27 revenue	$1,050M — true-ups vanish, ASP flat, MA/commercial pushback, DOJ overhang, volume +15%, Detect/MRD immaterial	$1,200M (≈ consensus $1,215M) — volume +20%, ASP +2%, pharma $85M, Detect/MRD ~$40M	$1,350M — volume +25%, Assure attach >50%, MRD MolDX, Detect DTC works, pharma $100M+
FY27 FCF	~$40M	~$150M	~$250M
P/S band (framework)	4x — "cautious" / top of "left for dead"	8x — top of "aware"	12x — "optimistic"
Equity value	$4.2B	$9.6B	$16.2B
Price	$13.8 (May-low retest)	$31.5 (flat)	$53
vs spot	−57%	−1%	+66%

Probability-weighted 25/50/25: 0.25 × 13.8 + 0.50 × 31.5 + 0.25 × 53 = $32.5

Red team overlap-collapsed: P<$22 ≈ 40% · $22–32 ≈ 30% · $32–40 ≈ 15% · >$40 ≈ 15% → $28–29

PM view: expected value $30–33 ≈ spot. Asymmetry is absent at $32. Entry math works at ≤6x FY27 base ($1,200M × 6 ÷ 305M ≈ $23.6 — inside the technicals agent's Zone 2) or if bull-case KPIs (MRD coverage, Detect volume, initial FY27 guide >$1.25B) print first. Macro cross-check: de-rate to the 4–8x band on FY26 revenue = $16–27/sh; $32 sits above the band. Stage-9 note: the FCF-inflection window (first FCF-positive print Q2'25) closed in 2025 — this is a Stage-8 valuation call, not an inflection trade.

Red Team (Stage 10) — unsoftened

The bear case

This is a tissue lab that got a one-time Medicare price reset, booked the catch-up as profit, and is now priced as a data platform.

(a) Growth is price and the price is done: 75% of FY25 profiling growth was 0211U replacing legacy panels; ex-true-up revenue/case is flat at ~$3,800–3,860; 2027 CLFS locks 0211U flat.
(b) The profit is catch-up cash on old cases: $33.6M FY25, $34.0M 1H26, $23.6M in Q2 alone at ~100% margin; Q2 op income ex true-ups ≈ $3.3M (1.3%); annual cash interest ~$37M; FY23 true-ups were negative (−$1.9M) and the 10-K says payers "have sought to recover" overpayments.
(c) FCF is rolling over: $55.3M → $6.4M in four quarters; Q3 guided neutral (≈ −$29M implied); Q3 adj EBITDA $10–16M; capex $22.1M vs $1.4M a year ago; inventory +$47M; GM to ~60%; FY26 FCF tracking $20–40M → P/FCF 230–450x; the $17.9M buyback bought back less than half of one half-year's SBC ($38.1M).
(d) 2H26 optics: guide implies 2H26 $555M vs $509.7M = +8.9%; Street Q3 $256M implies Q4 ≈ $299M vs $292.9M = +2% against a comp with $81M of true-ups.
(e) The market paid ~$2.3B of cap for the Assure rate, worth ~$26–28M of FY27 revenue, and it's preliminary.
(f) Single payer, single code, under FCA investigation: Medicare 39.6%, Payer 1 37.3% of revenue, Payers 1+2 47% of AR, Payer 2's AR share ~2x its revenue share; DOJ CID with no reserve after a 2022 settlement on the same rule.
(g) Pipeline is late, uncovered, or 4th to market: pharma $63.1M → $45.3M → $16.8M 1H; Detect cash-pay with no coverage path and zero disclosed volume; MRD behind NTRA/GH/Roche+SAGA/TEM+Personalis; tissue volume +13% is slowest of the peers.
(h) Governance and supply: 43.9% CEO/Chair with 25.0M shares pledged, brother on the board, 4/10 ex-AdvancePCS directors, 3% derivative-suit bylaw, material weakness two years, ASP disclosure discontinued, $27M September insider sales at $24.76–30.33, Whitney −38%, 26.5M options/RSUs waiting for strength.
(i) Valuation: 9.1x TTM / ~85x FY27 EPS for a ~1% ex-true-up operating margin and growth halving; May printed 4.1x FY26 guide.

Ten downside scenarios (12–24 mo; price = EV/S × FY27 revenue + net cash ÷ 305M) and the PM's response

1 · Anniversary cliff

$22−31% · P 35%

Q4'26 +2–5% YoY; Street cuts FY27 to ~$1.17B; 5.5x

ACCEPT. This is the base path for reported numbers; it is why the ladder starts at $24.5–26.3 and not here. Mitigant: judge on volume and ex-true-up ASP, not headline growth.

2 · Reimbursement reversal

$17.5−45% · P 15%

Final CLFS trims Assure, RESULTS Act freezes PAMA, MolDX narrows 0211U coverage; 4.5x

MITIGATE. Prelim is posted; final in Nov is the dated check. RESULTS Act would freeze, not cut. MolDX coverage narrowing is the real tail — thesis breaker if it happens.

3 · DOJ FCA settlement $150–250M + CIA

$21–23−30% · P 20%

Treble damages + Corporate Integrity Agreement

ACCEPT the risk, reject the base-rate. Prior settlement was $2.9M; FCA cases on the 14-day rule have mostly settled in the tens of millions. But no reserve, no scope disclosed — it stays #3 in the ledger and caps sizing at spec.

4 · True-ups flip negative

$17.7−45% · P 15%

Payer 2 AR written down; GM 60%; 4.5x

MITIGATE. Trend is +$3.9M → +$33.6M → +$34.0M — rising realization, not reversal. But Payer 2's AR/revenue mismatch is unexplained. KPI #3 tracks it quarterly; a negative true-up quarter = cut to starter.

5 · Reinvestment trap

$24.8−22% · P 40%

FY27 capex $120M, opex $700M, FCF ≈ 0; 6.0x

ACCEPT. Management told us: 30% EBITDA cap, capacity to $3B. This is the most likely path and it is why P/FCF is not the entry lens. Own it only in the 4–8x band where FCF optionality is free.

6 · Detect flops

$26.5−17% · P 45%

Achieve-2 <45% stage I/II, DTC $40M yields <$25M

ACCEPT. Detect is in nobody's guide and is not in our base case ($40M combined Detect+MRD). The narrative premium is the part of $32 we refuse to pay for.

7 · Competitive volume stall

$17.3−46% · P 20%

Total volume +10%, ASP flat; 4.5x

REJECT at 20%, accept at ~10%. Q2 record sequential adds and 20% Q3 guide argue against a stall inside 12 months; 2027 is the risk as TEM WES and FMI CDx mature. Validator: Q3 ≥62k cases.

8 · Supply air-pocket

$21−35% · P 30%

Pledge margin loop, PE/Fidelity blocks, option exercise-and-sell, momentum exit on an in-line Q3

ACCEPT — this is the ladder's Zone 3. A −35% move is ~0.5σ at HV60 73%. Never initiate above the no-chase line; buy the air pocket, do not be in it.

9 · Macro derate

$25−21% · P 30%

10y >5.5%, second hike; growth basket EV/S −30%

ACCEPT. Macro stamp is 🟡 for exactly this; add only on a macro-driven de-rate into the band or after a yield reversal (<5.0%).

10 · Accounting / governance shock

$19−40% · P 10%

Restatement from the material weakness, or key-man event on a 43.9% CEO with pledged stock

ACCEPT, cannot mitigate. Size cap 5% is the mitigation. Pledge disclosure ≤5M shares would lower this probability.

Upside case (the 1 in 10:1)

Volume +25%, Assure attach >50%, Detect $150M FY27, MRD MolDX, pharma $100M → FY27 $1.35B at 11x ≈ $50 (+56%), P ≈ 15%. Expected 12–24-mo price on the red team's distribution ≈ $28–29.

PM accepts the conclusion: asymmetry is negative at $32

The bull's best rebuttal (volume re-accelerating off a self-inflicted trough; true-ups as a leading indicator of structurally higher commercial ASP; Assure 2027 locked; FCF dip is a capex choice; relative value vs GH/NTRA and the EXAS takeout; stale Street; insider sales as option housekeeping) is why the thesis is ON WATCH rather than BROKEN, and why the ladder exists at all.

Known-Unknowns Ledger — deduped, ranked by thesis impact

Sustainable ex-true-up operating margin — "base ASP" undefined, true-ups unbudgeted; Q3 adj EBITDA $10–16M. Moves everything.

CY2027 CLFS: preliminary ($5,250 Assure, $8,455 0211U) vs final (~Nov); whether any commercial contracts index to CLFS; MI Tumor Seek Hybrid (81479) treatment.

DOJ CID scope: providers, claims, period, sealed qui tam, reserve discussions with the auditor. Exposure unbounded from public data.

Tissue vs blood ASP after Q1'26 (disclosure discontinued); whether the −6% company "base ASP" move Q1 → Q3 guide is mix or MA erosion.

Halbert pledge loan size, lender, margin-call price (25.0M shares; policy cap ~20% of holdings ≈ $800M without approval). Plus Jon Halbert 325k / Spetzler 150k pledged.

Detect units, backlog, Everlywell rev-share, DTC spend; reorder rate. Zero disclosed.

Payer 2 identity and why AR share (23.6%) is ~2x revenue share (12.6%); Payer 1/3 identities (Medicare FFS presumed for Payer 1).

True-up run-rate and cases-in-period vs cash-in-period revenue; the Q4'25 ~$81M "prior period" vs 10-K $33.6M "prior year" and the 9M'25 $23.4M vs Q3 $38.3M + H1 $15.9M inconsistencies; prelim Q4 MP revenue ~$270M vs actual $282.1M.

MRD assay validation data and MolDX timeline (tumor-naive CRC first, vs NTRA's covered 1M+ tests/yr).

Achieve-2 prospective results (n≈25k) and primary citations for Achieve-1 (60.3% / 99.2%) and the "Galleri ~17–20% stage I" comparison.

Sixth Street (19.4M sh), Fidelity (30.6M at 3/31/26) intentions post-lock-up; any 10b5-1 plans (Halbert, Spetzler, Brille); Form 4 footnotes unread; unidentified 8/3/26 Form 144 (20k sh).

Term-loan prepayment premium, SOFR floor, any rate hedge.

Material-weakness specifics, remediation status, auditor identity.

Pharma R&D pipeline: which contracts deliver $58M+ in H2'26 on $17.2M RPO; one biopharma group = 17.9% of AR.

Consensus revenue/EPS at print for Q1'26 (−$0.02 vs −$0.12 conflict) and Q2'26; cause of the 5/8 −18.6% (likely volume/opex).

Sales-rep cohort productivity post-realignment; whether +6,400 sequential adds was pull-forward from re-districting.

NY State approval for Assure (pending since ~Apr-2026; ~6% of US incidence).

Headcount/attrition since 1,846 at YE25; Glassdoor signal unverified (URL only).

Illumina dependence (agreement through Aug-2033, no minimums) vs unnamed higher-throughput vendors; revalidation risk on a PMA assay.

Q4'24, Q1'25, Q2'25 case splits derived from YoY %, not printed; Assure ASP contradiction ("$4,121" vs "just under $2,500" in Q1 transcript).

Short-interest series beyond two prints; float-denominator change between them.

Analyst rationales: Goldman $35 (9/21), UBS $28 (9/2), JPM cut to $25 after a beat-and-raise (8/6), Citi "$34 → $42" headline.

Peer forward EV/S (TTM only pulled); 0211U ADLT route (GH360 CDx at $8,455 unverified); Section 232 medical-consumables outcome; RESULTS Act status; Core PCE Aug-2026.

Catalyst (if any) for the 9/15–9/18 +27% week beyond the Fed/TEM sector move; whether the 908-lot Jan $30P and 3,282-lot Apr $40C are a large holder's collar.

Next earnings date (assumed ~2026-11-05/10; unverified) — gates every option expiry choice in the Trade Plan.

Trade Plan (Stages 9 + 10) — technicals agent's computed levels, verbatim

Structure · 2026-09-25 close

RSI(14)

73.8

Overbought · MACD 2.452 / sig 2.089 / hist +0.363

vs SMAs (daily)

+16 / +39 / +51%

SMA20 27.50 · SMA50 23.02 · SMA200 21.13

Volatility

ATR $1.47 · 4.59%

HV20 58% · HV60 73% · HV252 67% · BB(20,2) 33.52 / 21.48

Fib battle

AT 0.618 · 31.69

Retrace of 42.50 → 14.19. Hold on weekly close → 36.44; fail → 28.3 / 25.7

Price 31.98. 52w range 14.19 (2026-05-13) → 33.65 (2025-10-07); 5.0% below 52w high. Post-IPO ATH 42.50 intraday 2025-08-22 (close 39.30). Major swing 42.50 → 14.19 (−66.6%, a completed "secular correction"); current up-leg 14.19 → 32.89 (+132% in 4.5 months). Price sits exactly AT the 0.618 retrace of the decline (31.69): hold above 31.69 on a weekly close = path to 36.44; failure = mean-reversion toward 28.3/25.7. 200-WEEK MA = n/a (needs ~2029); substitute "big dog" = daily SMA200.

CAI weekly — candles · daily Bollinger(20,2) & SMA 20/50/200 · entry zones · trim bands

67 weekly bars, 2025-06-20 → 2026-09-25 · indicators computed on DAILY data, sampled weekly · daily RSI(14): 73.8 · hover / touch-drag for OHLC

SMA20 (d)SMA50 (d)SMA200 (d)BBentry zones 1–3trim bands A/Bno-chase 32.90invalidation 20.0

Key levels (computed)

Fibs & confluence

Pullback fibs of 14.19→32.89: 0.382 = 25.75 · 0.5 = 23.54 · 0.618 = 21.33 · 0.786 = 18.19
Retrace of 42.50→14.19 (overhead): 0.5 = 28.34 (reclaimed 9/16) · 0.618 = 31.69 (current battle) · 0.786 = 36.44 · 1.0 = 42.50
Confluence cluster #1: 21.1–21.5 = SMA200 21.13 + 0.618 fib 21.33 + BB-lower 21.48 + P/S 6x ≈ 21.0 — strongest support on the chart
Confluence cluster #2: 23.0–23.5 = SMA50 23.02 + 0.5 fib 23.54 + 20-day low 23.20 (2026-09-10)
Q2 earnings gap 16.54 (8/5 high) → 17.94 (8/6 low) unfilled — the "line in the sand." Overhead supply 34.4–39.3, then 42.50

Volume profile ($1 bins, 24mo) & basing read

26–27: 78.8M (largest node) · 18–19: 75.4M · 16–17: 71.9M · 15–16: 57.8M · 17–18: 57.1M · 19–20: 52.6M · 27–28: 51.2M · 25–26: 47.7M. 29–33 is thin air.

May–Jul 2026 11-week base 14.19–18.8 on rising volume = accumulation after the 67% drawdown. Breakout Aug 6 on 2.1× avg vol, 5-week shelf 23.6–28.3 (Aug 14–Sep 11) on declining volume (12.2M wk of 9/11, lowest since IPO), expansion 9/15–9/18 on 1.5–2.8× avg. Textbook stage-2 mark-up. Avg daily vol 20d 5.35M; 9/25 vol 2.9M — last 3 up-days on fading volume; short-term exhaustion risk with RSI 74.

Entry Ladder — no position today

Price is 16% above SMA20, RSI 74, at 0.618 major retrace; framework says never chase. P/S band math (TTM rev $989.6M, 282.6M sh): 5x = $17.5 · 6x = $21.0 · 7x = $24.5 · 8x = $28.0 · 9x = $31.5 · 10x = $35.0 · 12x = $42.0. Current 9.1x.

$32.90NO-CHASE LINE (20-day high). Above 33.65 (52w high) = breakout, but ride only with shares already owned; do not initiate above 32.90. Extended target of that breakout is 36.44 (0.786), thin air in between.0%

$24.5–26.3Zone 1 — starter. 0.382 fib 25.75, largest volume node 26–27, weekly close 26.27 (8/21), P/S 7x 24.5. Normal shakeout, −18 to −23% from here.25% of intended $

$22.5–23.6Zone 2 — core add. SMA50 23.02, 0.5 fib 23.54, 9/10 low 23.20, Oct-2026 chain has $22.5 strike. −26 to −30%.35%

$20.3–21.5Zone 3 — the "big dog" zone. SMA200 21.13, 0.618 fib 21.33, BB-lower 21.48, 0.786 of major swing 20.25, P/S 6x. −33 to −37%. Weekly close < 20.0 invalidates the ladder → wait for gap-fill 17.9–16.5 / P/S 5x 17.5 and re-base.40%

Alternative if it never pulls back: only a weekly close > 33.65 followed by a retest that holds 31.7 ± 1 ATR earns a starter (10%); otherwise sit out. Stage-9 note: FCF-inflection rerate already happened (Q2 print + Aug gap) — we are late to that trade, not early.

Sizing rule (framework, this tier)

Conviction-spec (moat 6.5 < 9) → ≤5% of portfolio at full ladder (Tier-1 cap is ≤10%; size down outside circle of competence; 10–15 positions max; cash 10% → 25–30% in a dicey tape — and the macro stamp says dicey). Zone allocations above are percentages of that ≤5%. Pair the ladder with the November CLFS final and the Q3 print; add on a macro-driven de-rate toward the 4–8x band or after a confirmed 10y reversal under ~5.0%, not on an index high.

Exit / Trim Framework — set before sizing (≤10% position cap per Stage 10)

Trigger	Action	Why
Trim band A · $36.0–36.5	Sell 20–25% / write calls	0.786 major retrace 36.44, P/S ~10.3x
Trim band B · $39.3–42.5	Sell to half	Aug-2025 close high 39.30 → intraday ATH 42.50, P/S 11–12x
Trim band C · > $45	Core only	P/S >12.8x, extension >2σ
Valuation rule	Trim to core	If P/S(TTM) > 12x OR EV/S(fwd) > 10x while revenue growth decelerates below 30% y/y; add back only inside Zones 1–3 with growth ≥35%
Trend/stop rule	Weekly close below daily SMA50 (23.0) after a full position = cut to half; weekly close < 20.0 = out	Structure broken, gap-fill mode. Trailing: never let a >30% winner become a loser (move stop to cost once 36 tags)
Time rule	Reduce to starter regardless of price	If 2 quarterly prints pass without a new 52w high and without clinical-volume or biopharma-bookings acceleration. Reassess at each earnings (next ~early Nov 2026, unverified)
Thesis breakers (technical hand-off)	Exit on a red high-volume day (>2× avg) — don't wait for levels	MolDX coverage/price cut on MI Profile or Caris Assure; biopharma revenue concentration loss; insider (Halbert/Sixth Street) Form-4 selling on strength; any ATM/convert announcement into this rally (float 48% — dilution risk is real)

Wheel Note — chains dated 2026-09-25 session

LAST-TRADE prices, post-close bids dead — RE-QUOTE LIVE. Liquidity: THIN. Strikes in $2.50/$5 steps only. Total OI: Oct-16 puts 1,356 / calls 2,929; Jan-15-27 puts 1,741 / calls 1,632; Apr-16-27 calls 3,314 (of which 3,282 is the $40 strike — a single large position); Dec-17-27 puts 756 / calls 1,119. Daily option volume mostly <100 per strike except block prints (908 × Jan $30 put on 9/25; 946 × Apr $40 call). Verdict: wheel-able only in 1–5 contract size with limit orders at mid; not a core wheel name. IV rank: n/a (no IV history); proxy — ATM IV ~62–70% vs HV20 58% / HV252 67% → IV roughly at/above realized, mid-range, premium is fair not rich. ATR% 4.6 = strike-to-strike ($2.50) is ~1.7 ATR.

CSP ladder (yield = premium/strike; basis = strike − premium)

Zone	Contract	Last (IV, OI)	Yield / DTE (ann.)	Basis / note
Zone 1 proxy	Jan-15-27 $30P	4.02 (IV 62%, OI 909)	13.4% / 112 DTE (~44% ann)	Basis 25.98 — only if you WANT assignment near Zone 1
Zone 1 alt	Dec-18 $25P	2.00 (IV 81%)	8.0% / 84 DTE (~35% ann)	Basis 23.00
Zone 2	Jan-15-27 $25P	2.29 (IV 75%)	9.2% / 112 DTE (~30% ann)	Basis 22.71
Zone 2 → 3	Jan-15-27 $22.5P	1.56 (IV 83%)	6.9% (~23% ann)	Basis 20.94 — lands in Zone 3; best risk/reward
Zone 3	Jan-15-27 $20P	0.61 (IV 96%)	3.1% (~10% ann)	Basis 19.39
Zone 3 LEAP	Dec-17-27 $17.5P	2.02 (IV 58%, OI 590)	11.5% / 448 DTE (~9.4% ann)	Basis 15.48 = below the whole 2026 base — the "get paid to wait for a disaster" strike
Oct-16 (21 DTE)	$25P / $30P	0.20 / 1.29	0.8% / 4.3%	Dead for CSPs: $30 is only 6% OTM with RSI 74

Covered calls at trim bands (own shares first)

Band	Contract	Last (OI, IV)	Yield / DTE	Note
Band A (36)	Oct-16 $35C	0.80 (OI 1,146, IV 52%)	2.5% / 21 DTE
Band A (36)	Dec-18 $35C	3.20 (OI 469)	10% / 84 DTE
Band A (36)	Jan-15-27 $35C	3.90 (OI 163)	12.2% / 112 DTE
Band B (40)	Dec-18 $40C	1.75 (OI 162)	5.5% / 84 DTE
Band B (40)	Apr-16-27 $40C	3.30 (OI 3,282, IV 70%)	10.3% / 203 DTE
Band B (40)	Dec-17-27 $40C	7.85 (OI 63)	24.5% / 448 DTE	LEAP call-write = synthetic exit at 47.85

Do NOT write calls below 35 into a fresh stage-2 breakout with 36.4/42.5 targets open. PM mechanics: earnings gate is mandatory — every expiry above the Oct-16 carries the Q3 print (~Nov 5–10, unverified); size CSPs so total assigned notional ≤ the Zone allocation within the ≤5% cap. Chart data: reports/cai/px.js (67 weekly bars, 2025-06-20 → 2026-09-25).

Risks — ranked by what kills the thesis

#	Risk	Mechanism
1	Price leg reverses instead of plateauing	MolDX/NCD narrows 0211U coverage, MA prior-auth denials rise, true-ups turn negative (10-K: payers have recouped before). Kills the margin structure. Probability low-moderate; severity −45%.
2	DOJ False Claims Act outcome	Treble damages + CIA + constrained tissue capture. Unbounded from public data; no reserve.
3	Volume decelerates below 15%	As re-districting laps and TEM/FMI take tissue share — turns Caris into a mid-teens grower at 9x sales.
4	New products fail to ship volume	Detect cash-pay with no coverage into Grail/Exact/Guardant; MRD 4th–5th into NTRA's fortress; capacity tripled ahead of demand, GM to ~60%.
5	Reinvestment consumes FCF through 2027	The cash story dies even if revenue is fine; P/FCF becomes meaningless.
6	Supply overhang	25.0M pledged CEO shares, 26.5M options/RSUs, PE/Fidelity blocks, momentum holders, into a 48% float at HV60 73%.
7	Macro de-rate	10y >5.5% or a second hike; floating-rate interest +$1M per 25bp.
8	Governance/accounting shock	Unremediated material weakness, controlled-company behavior without the label, key-man on a 70-year-old founder.
9	Pharma line does not recover	Highest-margin revenue; $58M needed in H2 on $17.2M RPO.
10	Concentration	39.6% Medicare on one code; one biopharma group 17.9% of AR; single sequencing vendor.

Validators (thesis strengthening) — check-by dates

Q3'26 print (~2026-11-05/10, unverified): cases ≥62k (tissue ≥+15%), ex-true-up ASP ≥$3,850, adj EBITDA ≥$16M, revenue ≥$265M, true-ups disclosed and shrinking as a %.
CY2027 CLFS final (~2026-11-15): Assure ≥$5,250 confirmed; 0211U $8,455 unchanged; no RESULTS Act enactment.
13F/13G refresh (2026-11-14): Sixth Street and FMR stakes flat or higher; no new Whitney 13G/A reduction.
Q4'26 call (~2026-02-25): Detect units disclosed ≥5k/month; MRD launched with a MolDX submission date; initial FY27 guide ≥$1.25B; pharma FY26 ≥$75M delivered.
H1'27: Achieve-2 prospective stage I–II sensitivity ≥50%; Assure covered lives ≥200M; NY State approval; DOJ CID closed or settled <$25M; Halbert pledge disclosed ≤5M shares.

Breakers (thesis broken → exit or stop adding)

Q3 cases <60k or ex-true-up ASP <$3,700 (2026-11-10)
Final CLFS cuts either code, or a MolDX/NCD policy narrows comprehensive-panel coverage (2026-11-30)
A negative prior-period true-up quarter, or Payer 2 AR write-down (any print)
DOJ settlement >$100M or a Corporate Integrity Agreement (any date)
Any Halbert or Sixth Street open-market sale on strength; any ATM/convert into the rally (Form 4 / 8-K watch, continuous)
Weekly close <$20.0 on >30M volume after a print (technical confirmation of a fundamental break)
FY27 initial guide <$1.15B (2027-02)

First-principles KPIs (not Wall Street's) — target vs actual, check-by

Therapy-selection cases and sequential net adds

Now59,200; +6,400 QoQ (record)

TargetQ3 ≥62k (guide 61–62k); Q4 ≥65k; seq adds ≥3,000/qtr

CheckQ3 print ~2026-11-05/10

Ex-true-up profiling revenue per case

Now~$3,863

Target≥$3,850 holding (Q3 guide $3,800–3,900); <$3,700 = breaker

CheckQ3 print

Prior-period true-ups as % of revenue

Now9.0% ($23.6M)

TargetDeclining toward <5% while staying positive

CheckQ3 print

Caris Assure cases and blood attach rate

Now10,700 (+50% YoY); ~40% attach

Target≥12,000 Q3; attach ≥45% by Q4

CheckQ3 print

OCF / CapEx and quarterly FCF

Now1.3x; $6.4M

TargetQ3 neutral (guided), Q4 positive; OCF/CapEx >2x by Q1'27

CheckQ3, Q4 prints

Pharma RPO + contract liabilities

Now$17.2M + $21.1M

TargetRPO ≥$40M by YE26; FY26 pharma ≥$75M

CheckQ3 print; Q4 call

Payer 2 AR share vs revenue share

Now23.6% AR / 12.6% revenue

TargetConverging (<1.5x)

CheckQ3 10-Q

Detect units shipped / Achieve-2

Nowundisclosed / unread

Target≥5k units/month by Q4 call; ≥50% stage I–II prospective

CheckQ4 call ~2027-02; H1'27

Adj EBITDA ex true-ups

Now~$32M (55.7 − 23.6)

Target≥$25M in Q3 despite $10–16M reported guide incl. true-ups

CheckQ3 print

Changelog — a living document

v1.0 — 2026-09-26 (Liquid Wheel Research · deep-dive team)
Full deep-dive rebuild superseding the same-day draft. Red-team verification pass merged: term-loan terms verified from the 4/1/26 8-K (Term SOFR + 5.00%, $50M min-cash covenant, $300M DDTL, $500M incremental); fully diluted share count 298M → ~305M; Medicare share pinned to the verified 39.6% (FY25 10-K); H2'26 GM guide corrected to "in that 60%" range → Gate 16 PASS → WATCH; Whitney stake 6.5% → 4.0% (13G/A 6/30/26); CY2027 CLFS treated as preliminary (final ~Nov); Fed 9/16/26 = +25bp hike; ex-true-up TTM NI haircut softened to ~$60–90M (≈ +$15–45M ex-true-up); interest coverage restated (ex-true-up op income ~$13M vs ~$37M cash interest). Entry ladder replaced with the technicals agent's computed levels (no-chase $32.90; zones $24.5–26.3 / $22.5–23.6 / $20.3–21.5; weekly close <$20.0 invalidates). Scenarios recomputed on 305M shares ($13.8 / $31.5 / $53; PW $32.5; red-team distribution $28–29). Score 6.5 → 6.0; thesis INTACT → ON WATCH; moat 6 → 6.5. Tally 8 PASS / 6 WATCH / 2 FAIL. Page: TL;DR strip, gate-score ring, diff panel, scenario range bar, downside-scenarios grid, known-unknowns ledger, KPI scorecards, sources drawer, staleness banner. Data as of close 2026-09-25 ($31.98).

v1.0 draft — 2026-09-25 (Liquid Wheel Research · 5-agent deep-dive team · SUPERSEDED — carried forward verbatim)
Initial deep-dive: 16-gate checklist (9 PASS / 5 WATCH / 2 FAIL), half-wheel flywheel map (pharma/data spoke shrinking), 8-quarter financial longitudinal with out-of-period revenue series, GAAP-NI → OCF → FCF bridge (TTM FCF $124.0M; triage $63.3M corrected), skin-in-the-game score (196× / 1,777×), earnings tracker, FY2027 scenarios ($19.1 / $33.0 / $58.6, probability-weighted $35.9), entry ladder (Zone A $25.0–26.5 / Zone B $20.5–22.5 / Zone C $17.0–18.5) with $30.00 no-chase line, wheel/CSP strikes (Jan-15-27 30P preferred), first-principles KPIs, 18 resolved agent conflicts, 20-item known-unknowns ledger. Verdict: 6.5/10 — thesis INTACT, DO NOT CHASE. Classification: CONVICTION-SPEC (moat 6/10 < 9 gate); ≤5% max. Data as of close 2026-09-25 ($31.98).

Next scheduled review: Q3'26 print (~2026-11-05/10, unverified) + CY2027 CLFS final (~2026-11-15) + 13F/13G refresh (2026-11-14). Gates: cases ≥62k, ex-true-up ASP ≥$3,850, adj EBITDA ≥$16M, Assure $5,250 confirmed, 0211U $8,455 unchanged.

Sources — consolidated, deduped (documents the team actually used)

Framework / internal

/Users/jamesgilland/vault/Areas/Trading/Research/investing-framework.md
/Users/jamesgilland/vault/Areas/Trading/Research/tem-luc-deep-dive-2026-07-20.md (peer context)
/Users/jamesgilland/clawd/reports/cai/index.html (prior draft build 2026-09-25); /Users/jamesgilland/clawd/reports/cai/px.js (weekly bars)
/Users/jamesgilland/clawd/reports/iren-v3/report-run1.md (16-gate template)

SEC filings (Caris, CIK 2019410)

FY25 10-K: sec.gov/…/cai-20251231.htm
Q2'26 10-Q: sec.gov/…/cai-20260630.htm
Q3'25 10-Q: sec.gov/…/cai-20250930.htm
DEF 14A 2026-04-23: sec.gov/…/cai-20260423.htm
8-K Q2'26 ex99.1 (2026-08-05): sec.gov/…/000201941026000053/cai_ex99-1.htm
8-K Q1'26 ex99.1: sec.gov/…/000201941026000040/cai_ex99-1.htm
8-K Q4/FY25 ex99.1 (2026-02-26): sec.gov/…/000201941026000011/cai_ex99-1.htm
8-K prelim Q4'25 (2026-01-12): sec.gov/…/000201941026000005/cai_ex99-1.htm
8-K 2026-04-01 refinance (term-loan terms): sec.gov/…/cai-20260401.htm
8-K 2026-06-04 annual meeting votes: sec.gov/…/cai-20260604.htm
8-K 2025-10-30 bylaws: sec.gov/…/cai-20251030.htm
8-K 2026-09-14 (Morgan Stanley conference): sec.gov/…/cai-20260914.htm
IPO prospectus 424B4: sec.gov/…/tm2415719-25_424b4.htm ; S-1 (Nephron TAM): sec.gov/…/tm2415719-14_s1.htm
XBRL company facts: data.sec.gov/api/xbrl/companyfacts/CIK0002019410.json ; submissions + Form 4/144 XMLs: data.sec.gov/submissions/CIK0002019410.json ; full-text search: efts.sec.gov/LATEST/search-index
13G/A FMR: sec.gov/…/000119312526051538/primary_doc.xml ; 13G/A Whitney (6/30/26): sec.gov/…/000090514826003576/primary_doc.xml ; Whitney prior 13G/A: stocktitan.net/…/schedule-13g-a-…-c8c3cbaae8e0.html
Form 144 Brille: stocktitan.net/…/144-…-7b3c42a6232d.html ; term-loan 8-K summary: stocktitan.net/…/8-k-…-0534241f0739.html

Earnings calls / company

Q2'26 call: stockanalysis.com/stocks/cai/transcripts/662771-q2-2026/ ; investing.com (Q2'26 transcript) ; finance.yahoo.com/…/caris-life-sciences-inc-cai-050537833.html
Q1'26 call: fool.com (Q1'26 transcript) ; theglobeandmail.com (Q1'26 transcript) ; investing.com (Q1'26 transcript)
Q4'25 call: investing.com (Q4'25 transcript)
Q3'25 call: investing.com (Q3'25 transcript) ; Q3'25 PR: prnewswire.com (Q3'25 results)
Leadership: carislifesciences.com/about/leadership/ ; Detect launch: carislifesciences.com/…/caris-life-sciences-launches-caris-detect/ ; buyback: carislifesciences.com/…/share-repurchase-program/
Sixth Street $830M round: sixthstreet.com/…/caris-life-sciences-raises-830-million…
Blue Owl/Blackstone facility: tradingview.com/news (Blue Owl/Blackstone $1.2B facilities)

Reimbursement / policy

CY2027 CLFS (0211U $8,455; 0485U $5,250): discoveriesinhealthpolicy.com/2026/09/cms-posts-pama-lab-fee-schedule-for.html ; PAMA/RESULTS Act context: discoveriesinhealthpolicy.com/2026/09/can-ai-get-pama-news-from-investor-calls.html
Tariffs: crowell.com (Section 232 pharma tariffs) ; whitecase.com (Section 232 medical supplies)

Peers / M&A

GH Q2'26: sec.gov/…/gh-06302026xexhibit991.htm ; NTRA Q2'26: sec.gov/…/ntra-20260630xex992.htm ; TEM Q2'26: investors.tempus.com (Q2'26 results)
Abbott/Exact: biopharmadive.com/news/abbott-acquire-exact-sciences/806058/ ; abbott.mediaroom.com (close 3/23/26) ; Roche/SAGA: roche.com/investors/updates/inv-update-2026-04-16b

Macro

ISM Mfg Aug-2026: prnewswire.com (ISM Mfg 54.6) ; ISM Services: prnewswire.com (ISM Services 55.4) ; tradingeconomics.com/united-states/manufacturing-pmi
Fed 9/16/26: cnbc.com/2026/09/16/fed-rate-decision-september-2026.html ; cnbc.com/2026/09/16/fed-meeting-today-live-updates.html ; foxbusiness.com (Fed decision 9/16/26) ; raisin.com (Fed breakdown Sept 2026) ; kiplinger.com (Fed meeting Sept 2026) ; bondsavvy.com (dot plot)
GDPNow: atlantafed.org/…/gdpnow ; oil/yields: cnbc.com/2026/09/15/oil-us-treasurys-stocks-pressure.html ; federatedhermes.com (oil/yields) ; CPI: kiplinger.com (CPI Aug 2026) ; jobs: cnbc.com/2026/09/04/jobs-report-august-2026.html ; M2: macroradar.io/m2-money-supply ; fx.co/en/forex-news/3179802

Sentiment / analysts / ownership

stockanalysis.com/stocks/cai/forecast/ ; /statistics/ ; /financials/?p=quarterly ; /cash-flow-statement/?p=quarterly ; /balance-sheet/?p=quarterly
marketbeat.com/stocks/NASDAQ/CAI/forecast/ ; /insider-trades/ ; /institutional-ownership/
TIKR: tikr.com (stock jumped 10%) ; tikr.com (president sold $14.75M) ; tikr.com (up 65%, multiple cheaper) ; finance.yahoo.com (stock jumped)
Wolfe: tipranks.com (Wolfe Peer Perform) ; UBS: marketscreener.com (UBS Neutral $28)
Substack/SA: deltasheets.substack.com/p/cai-fy2026-q1 ; spystsignals.substack.com/p/caris-update-cai ; seekingalpha.com/article/4948490 ; seekingalpha.com/article/4824450 (titles/summaries only) ; thestockobserver.com (9/23) ; stocktwits.com/symbol/CAI
Glassdoor: glassdoor.com/Reviews/Caris-Life-Sciences-Reviews-E426957.htm (URL only; page not captured)

Market data

yfinance (pulled 2026-09-26): CAI daily OHLCV 2025-06-18 → 2026-09-25, Ticker.info, option chains (2026-10-16, 11-20, 12-18, 2027-01-15, 03-19, 04-16, 12-17); ^VIX ^TNX ^FVX ^IRX SPY IWM XBI GH NTRA TEM CL=F DX-Y.NYB; peer multiples and consensus estimates
tvremix MCP get_earnings_history NASDAQ:CAI (beat/miss history, post-print reactions, forward consensus)
Technicals script: /private/tmp/claude-501/-Users-jamesgilland/707b5e1e-66cd-46aa-b301-a36cae3e2f5d/scratchpad/cai_ta.py ; 10-K text: /private/tmp/claude-501/-Users-jamesgilland/707b5e1e-66cd-46aa-b301-a36cae3e2f5d/scratchpad/cai10k.txt

Disclosure & disclaimer: This report is education and personal research, not financial advice. The author may hold positions in the securities discussed, including shares and options. Numbers from company filings/releases through the Q2'26 10-Q (2026-08-05) and market data through 2026-09-25 close — verify before acting; option prices are last-trade, post-close, and must be re-quoted live. Past performance doesn't guarantee future results. Do your own research — that's rather the point.