Genomics
Real, accelerating transplant-dx franchise that just turned GAAP-profitable and is reshaping itself (sell low-margin kits, buy oncology MRD) — but ~46% of its core segment still rides on a single Medicare LCD that CMS has confirmed it will re-issue; a quality compounder priced for the good case with a binary policy fuse attached. WATCHING, leaning constructive, gated on the new LCD.
Research
The verdict
"Real, accelerating transplant-dx franchise that just turned GAAP-profitable and is reshaping itself (sell low-margin kits, buy oncology MRD) — but ~46% of its core segment still rides on a single Medicare LCD that CMS has confirmed it will re-issue; a quality compounder priced for the good case with a binary policy fuse attached. WATCHING, leaning constructive, gated on the new LCD."
Primary sources
Source documents — open to read in full
CareDx sells transplant patients a way to know if their new organ is being rejected without a biopsy. That is the whole business in one sentence, and after a year of CEO reshaping it is finally close to being literally true — the lower-margin instrument/kit business is on its way out the door.
The model. CareDx is a Brisbane, California molecular-diagnostics company (Nasdaq: CDNA, incorporated Delaware, CIK 0001217234) ``. It reports three segments:
. The franchise (AlloSure Kidney/Heart/Lung + AlloMap Heart + AlloSure Heart) was **78% of total revenue in Q1 2026** (up from 73% a year earlier) — i.e. the company is more concentrated in its core, not less.. **This segment is being sold** (see Lens 9) — divested to EuroBio Scientific for **$170M cash** .Customers & payers. The "customer" is the transplant center / nephrologist / cardiologist who orders the test; the payer is the insurer. The defining fact of this business: Medicare = 46% of testing-services revenue in FY2025 (down from 50% and 53% in 2024/2023) . Medicare-covered patients are only ~25% of *tests* but ~46% of testing-services *dollars* because Medicare reimburses dd-cfDNA at a high fixed rate. No other payer is >10% of revenue . Average revenue per test ~$1,480 ``.
Suppliers. Key inputs are sequencing instruments/reagents from Illumina, cell-prep tubes from Becton Dickinson and Streck, lab equipment from Beckman Coulter, and reagents from Qiagen . Illumina is also a *commercialization partner* — CareDx holds worldwide rights to Illumina's NGS line for transplant diagnostics under a license-and-commercialization agreement .
Contract structure. Testing is largely fee-for-service per test (no take-or-pay; revenue recognized on accrual at expected collectible amount). PDS is recurring SaaS. So the revenue quality is mixed: SaaS is sticky, but the cash cow (testing) is exposed to per-test reimbursement decisions made by a government payer — the structural tension that runs through this entire dossier.
Map the chain end to end, names attached:
Upstream inputs → CareDx:
CareDx (the conversion step): runs the assay (dd-cfDNA quantification or gene-expression) in its own labs, returns a rejection-risk result through its software to the ordering clinician.
Downstream → end customer → who pays:
Chokepoints / single-source dependencies (the ones that matter):
This lens passes the "names or it didn't happen" test: Illumina, Qiagen, Beckman Coulter, BD, Streck, Palmetto/MolDX, Noridian — the chain is concrete, and its weakest link is regulatory, not industrial.
What CareDx actually has:
Bargaining power. Over suppliers: limited — Illumina/Qiagen are larger and CareDx is platform-dependent. Over customers (clinicians): moderate-to-strong, via evidence and embedded software. Over the payer: weak — Medicare sets the price, full stop. The single most important sentence about CareDx's moat is that it does not control the price of its best product.
Durability verdict: a real but narrow moat — evidence + workflow + scale in one disease area — sitting on a reimbursement foundation the company doesn't own. Wide enough to defend the #1 share position; not wide enough to make the Medicare risk go away.
By product segment (FY2025, all ``):
| Segment | FY2025 rev | YoY | Trend / cause |
|---|---|---|---|
| Testing Services | $275M | +10% | Volume +14% (~200k tests) partly offset by mix; re-accelerating hard in 2026 (Q1 testing rev $91M, +48% YoY, volume ~54,900 +17% `` → ARPT up ~26%, a price/mix tailwind) |
| Patient & Digital Solutions | $57M | +31% | Fastest leg; SaaS adoption across transplant centers |
| Product | $48M | +19% | HLA-typing kits — being divested to EuroBio for $170M `` |
| Total | $379.8M | +14% | vs $333.8M FY2024, $280.3M FY2023 `` |
By geography (FY2025 ``): US $359.9M (94.8%), Rest of World $19.9M (5.2%). Essentially a US business; the ex-US monitoring assets sit largely in the Product segment that is being sold — so post-divestiture CareDx becomes even more a pure-play US testing-services + SaaS company.
The trend that matters: testing services decelerated to +10% in FY2025 then re-accelerated to +48% in Q1 2026 ``. That swing is the whole bull case — it says the 2023 reimbursement wound has healed, volume is compounding, and pricing/mix (AlloSure Lung, iBox/AlloSure Plus, broader coverage) is now a tailwind rather than a headwind.
The most recent print is Q1 2026 (reported 2026-04-28) and it is the inflection quarter.
; PDS $16M .Guidance: management raised FY2026 to $447–465M revenue (+20% YoY at $456M midpoint) and adjusted EBITDA $43–57M (+58% at $50M midpoint) ``. Tone shifted decisively positive.
Market reaction / what was priced in: the print drove the stock toward the ~$26 area (mkt cap ~$1.35B by late June) . The reaction was constructive but not euphoric — the market is rewarding the profitability turn while still discounting the LCD overhang (analyst average PT ~$25.80–27.33, i.e. roughly *at* the price ).
Anything unusual vs its own history: yes — the FY2025 vs FY2024 annual comparison looks like a collapse (net income $52.5M → net loss $21.4M) but that is an accounting illusion (Lens 10): FY2024 carried a one-time $96.3M litigation-expense reversal. The Q1 2026 result is the clean, like-for-like signal, and it is unambiguously good.
No transcripts on the research shelf (transcripts/ empty), so this is ``-grounded.
The tonal arc across the last several quarters is a clear fear → stabilization → confidence progression:
Phrases now recurring: "precision medicine," "higher-growth, higher-margin," "simplify the operating model," "AEBITDA margin expansion." Phrases that disappeared: the guidance-withdrawal / reimbursement-crisis language of 2023. The thing to watch on the next call is whether they keep talking about the LCD pro-actively (good — managing it) or go quiet (a tell).
Peer set = diagnostics names with overlapping franchises (transplant + MRD/molecular). Multiples are
| Company | Ticker | Mkt cap | FY2026E revenue | EV/Sales (FY26E) | P/E | Note |
|---|---|---|---|---|---|---|
| CareDx | CDNA | ~$1.35B `` | ~$456M mid `` | ~2.6x `` | ~n/a (GAAP near-breakeven) | Adj-EBITDA-positive; net cash |
| Natera | NTRA | n/a | $2.62–2.70B `` | n/a | n/a (Q1 net loss $(85)M ``) | Direct transplant rival (Prospera) + huge oncology/women's-health base |
| Veracyte | VCYT | ~$2.83B `` | ~$582–592M `` | ~4.5x `` | n/a | Molecular-dx peer (oncology/thyroid); profitable-er |
| Guardant Health | GH | n/a | 2025 rev ~$981M (+33%) `` | n/a | n/a (loss-making) | MRD/liquid-biopsy scale comp |
Read: on EV/Sales, CareDx (~2.6x) trades at a discount to Veracyte (~4.5x) `` despite faster current top-line growth (CDNA +39% Q1 vs VCYT ~13-14% guided). The discount is the LCD risk premium the market attaches — the question the whole thesis turns on is whether that discount is too large (bull) or correctly sized for a binary policy event (bear). Natera and Guardant are larger and loss-making, so P/E comparison is meaningless; on a growth-adjusted EV/Sales basis CDNA is the cheapest profitable name in the cluster. I did not source live EV/Sales for NTRA/GH and will not fabricate them.
The five-year tape has one dominant lesson: CDNA trades on Medicare reimbursement headlines first, earnings second. ``
. Craig-Hallum cut its target from **$26 → $9** . Analysts estimated ≥$15M of HeartCare revenue going "to $0" . Testing revenue fell **20% in FY2023** ($263.7M → $209.7M) . This single regulatory event re-rated the entire equity.Pattern verdict: the market reacts to (1) Medicare coverage/LCD news (the big mover), (2) guidance changes, (3) capital-allocation/M&A. It does not react much to quarterly volume beats absent a reimbursement read-through. That is precisely why the pending new LCD is the single most important catalyst on the calendar — it is the variable with the largest historical price elasticity.
CEO: John W. Hanna (President & CEO since 2024-04-15) ``.
insider-transactions.csv absent); note a recent CEO sale of ~17K shares (June 2026) `` — small and likely routine/10b5-1 given the size, but worth flagging as a data point, not a thesis., (b) **redeployed into Naveris** (viral-mediated cancer MRD) for **up to $260M ($160M upfront + $100M earnout)** , and (c) authorized a new $100M buyback (24mo) on top of the prior $50M = $150M total ``. That is textbook "sell what you don't want to own, buy growth, return the rest" — a real capital-allocator's playbook, executed while the core turned profitable.predates/overlaps his arrival and relates to the 2023 reimbursement period. Selling Product *to a named competitor* (EuroBio appears in the 10-K competitor list) is mildly notable but commercially sensible. Watch comp/dilution: SBC ran **$34.9M in FY2025** on $379.8M revenue (~9% of sales) — high, typical for the sector, and the main thing keeping GAAP profit thin.Acting as a forensic analyst on the FY2025 10-K + Q1 2026 10-Q ``:
. So **FY2024 income from operations was only $40.8M** and FY2025's swing to a **$(30.8)M operating loss / $(21.4)M net loss** is *not* a deterioration of the underlying business — it's the absence of a one-off benefit plus reinvestment. Anyone citing "net income collapsed 141%" is reading an artifact. and is $44.6M at Q1 2026 — collections improving, not stretching. Healthy sign.. FY2023 carried large impairments (the $(203)M operating loss year) ; the Naveris acquisition will add goodwill/intangibles to watch going forward, plus earnout-liability accounting on the $100M contingent consideration.Regulatory findings (required sub-section):
regulatory/regulatory-findings.md (generated 2026-06-23 from SEC EDGAR EFTS, LR + AAER, period 2021–2026) returned zero findings ``.. A restrictive **Proposed LCD (DL40063/DL39787/DL40060, billing articles DA60146/DA60152)** with utilization limits and a bundled-payment concept was pending. **Crucial Q1 2026 update: the MACs decided NOT to finalize the proposed Aug-10-2023 LCD, but CMS stated the MACs intend to issue a NEW LCD "in the coming months"** . So the worst-case proposal is dead — but a replacement policy is explicitly coming, and its terms are unknown.Forensic verdict: clean books, no enforcement history, conservative-trending working capital. The only "red flag" is interpretive (don't be fooled by the FY2024 one-off) and the only real risk is exogenous (the next LCD). This is an honest set of financials.
Built bottom-up from Q1 2026 actuals + company guidance. Important structural note: the Product segment ($48M FY2025) is being divested (~Q3 2026 close) and Naveris (~$34M 2025 revenue, 30–40% growth) is being acquired (~Q3 2026 close) ``. Net, these roughly offset on revenue in the near term but improve mix (sell low-margin kits, buy high-growth MRD). Company FY2026 guidance already reflects the expected timing.
); adj EBITDA **$50M** mid . GAAP EPS thin-positive, ~$0.20–0.35 . Non-GAAP EPS ~**$1.10–1.30** .. If hit, that's ~+10% revenue but a **doubling of EBITDA** — i.e. the story is *margin*, not just growth. GAAP EPS ~**$0.55–0.85** . Non-GAAP EPS ~$1.50–1.80 ``.; adj EBITDA ~**$120M** ; GAAP EPS ~$0.90–1.20 ``.Scenario fan (FY2027 GAAP EPS):
No forecast.ts create in this unattended watchlist run (per SKILL). If logged later, the scoreable base call would be: "CDNA FY2027 adjusted EBITDA ≥ $90M, p≈0.55, resolves 2028-02-28" — chosen because adj EBITDA (not GAAP EPS) is management's own yardstick and the cleanest measure of the leverage thesis, and 0.55 reflects a genuine-but-not-dominant probability given the LCD overhang.
Bull case. CareDx is the category-defining brand in a structurally growing niche (transplant dx ~$4.94B in 2024 → ~$9.21B by 2033, ~7.3% CAGR ), it just **turned GAAP-profitable while growing 39%** , and it is being run by a capital-allocator who is actively improving the business mix — selling $48M of low-margin kits for $170M cash, buying into the much larger oncology-MRD market via Naveris, and buying back $150M of stock. The worst-case 2023 LCD proposal is dead . The data flywheel (200k tests/yr → AlloSure Plus/iBox) widens the evidence moat, PDS deepens workflow lock-in, and at **~2.6x EV/Sales** the stock is the cheapest profitable name in its diagnostics cluster, trading roughly at the analyst average target with a clean balance sheet and optionality (oncology MRD) the market isn't paying for.
Bear case (permanent-impairment risks).
, governed by one MolDX/Noridian policy, and CMS has *confirmed a new LCD is coming* . A restrictive version (bundled payment, utilization caps) would do to 2026 what the March-2023 article did — a ~30% same-day re-rate and a 20% testing-revenue decline ``. This is the one risk that can permanently re-base the franchise's earning power.; ~9%-of-sales SBC . The "profitable now" narrative is real but fragile to any revenue wobble.Pre-mortem (18 months out, thesis broke): the new LCD landed restrictive in late 2026 / 2027 — bundled surveillance payment + tighter utilization. Testing growth fell from +40% to single digits, the $100M-EBITDA-2027 target was abandoned, Naveris integration absorbed management bandwidth without offsetting the core decline, and CDNA round-tripped toward the low-teens. The tell we'd have wanted: the terms of the new LCD draft the moment it posts for comment.
Multiples too high? No — ~2.6x EV/Sales `` on a +20%-guided, EBITDA-positive, net-cash name is not demanding. The multiple is low precisely because of risk #1. You're not paying for optimism; you're being paid (a discount) to underwrite policy risk.
Contrarian view (what the market refuses to see): the market is still trading CDNA as "the 2023 reimbursement victim" and pricing a permanent LCD overhang — but the worst proposal already died, the business re-accelerated to +39%, and management has de-risked the model (more US testing/SaaS, less commodity kit, fresh cash, oncology optionality). If the new LCD lands merely neutral, the risk discount collapses and a ~2.6x name re-rates toward its 4–5x peer — that's the asymmetric, non-consensus payoff.
Dismantling the bull case:
. That's not a normal customer-concentration risk you can diversify away by selling harder — it's an administrative-pricing risk, and the administrator has *announced it's rewriting the rule* . The bull's "worst case already died" is naïve: CMS explicitly said a new LCD is coming, and "new" is not "favorable."; ~46% of that exposed to Medicare. One unfavorable bundling decision and the model's best economics evaporate — *exactly* the 2023 playbook, which took testing revenue −20% .Not a stock anymore — a closed M&A. Lilly bought the whole company for $10.50/share cash (closed Jul 2025); the only live "position" is the $3.00 CVR, which pays only if VERVE-102 reaches a US Phase 3 dosing — market priced ~21% odds, a coin-flip dressed as a lottery ticket.
A rare profitable, debt-free genomic-dx compounder (FY25 16% rev growth, $126M FCF) — but the stock has doubled into a 6.5x-sales / ~30x-FCF valuation just as Natera's FDA-approved Signatera CDx occupies the exact MIBC beachhead TrueMRD is launching into. Quality business, priced for flawless MRD execution it has not yet proven. WATCHING; would buy a reimbursement/launch-driven pullback under ~$40.
A founder-led rare-disease engine with real ($673M) revenue and a pioneer at the helm — but it just lost its biggest pipeline bet (setrusumab) and is burning ~$466M/yr against ~$534M cash, so the entire equity now rides on two H2-2026 FDA approvals (UX111 Sep 19, DTX401 Aug 23) closing the gap to a promised 2027 profit. Binary, not compounding.