The Signal

PAVM's $30M Series D preferred financing contains a warrant call provision explicitly tied to MolDx publishing a draft LCD for EsoGuard Medicare coverage. Upon publication, PAVM can force warrant exercise at $0.001/share—effectively free money if the catalyst hits.

This is revealed preference. Sophisticated investors negotiated a contract term that only makes sense if they expect the draft LCD soon. You don't put a catalyst-specific call provision in a deal unless both sides believe that catalyst is near-term.

The Thesis (LUCD, not PAVM)

LUCD is the clean play. PAVM is a holding company with 23% LUCD ownership, burning cash, now levered to punitive preferred terms (300-500% redemption multiples, 15% interest rate, tight covenants). LUCD is the pure-play on EsoGuard.

Timeline evidence:

  • Sep 4, 2025: MolDx CAC meeting—unanimous expert endorsement for coverage
  • Nov 12, 2025: CEO Aklog on Q3 call: "confident close in final stages," draft LCD "soon," describes process as "formality"
  • Oct 2025: California CLA meeting—MolDx medical director publicly confirmed positive trajectory
  • Feb 4, 2026: Warrant call provision tied to draft LCD (3 months post-"imminent" guidance)

Revenue mechanics: Currently recognizing ≈17% of billed revenue ($1.2M of $7.1M Q3) due to reimbursement uncertainty. Medicare coverage would flip this—plus 12-month lookback on all Medicare claims filed. Management explicitly targeting Medicare patients now to maximize lookback revenue.

The Numbers

LUCD: $1.21, 75% idio variance, 4% short interest, 30% 1Y momentum, no P/E (pre-profit)

  • Cash: $47M post-September offering, runway through 2026
  • Test volume: ≈2,800/quarter, ≈$7M billed, ≈$1.2M recognized
  • Medicare = 50% of 30M target population

PAVM: $11.76 (+23% today), 130% idio variance, 6% short interest

  • Net cash from deal: ≈$7.7M after paying off prior holder
  • Covenants: $5M minimum cash (rising to $8M July 2026), LUCD ownership coverage test (note value cannot exceed 65%/50% of LUCD stake value, floor $20M)
  • Preferred terms: 300% redemption multiplier year 1, rising to 500% by year 3

What I Don't Know

  1. Draft LCD timing is still uncertain. Management said "imminent" in November 2025. It's now February 2026. MolDx doesn't publish schedules.

  2. Commercial payer follow-through. Management claims positive UnitedHealthcare discussions, but no contracts signed. Medicare coverage doesn't guarantee commercial payer cascade.

  3. Dilution math. If Series D converts at $6.50 + warrants exercise = ≈9.2M new PAVM shares. LUCD has 138M shares outstanding with 49.6M preferred convertible to common. Both are dilution-heavy.

  4. Why now? If MolDx coverage was truly "imminent" in November, why did PAVM need to accept such punitive terms in February? Either timing slipped, or PAVM's cash position was more desperate than disclosed.

The Edge Question

This is a binary catalyst trade. The edge question is: Is P(Medicare coverage in 2026) > market-implied probability?

Evidence for higher P:

  • Unanimous CAC endorsement (Sep 2025)
  • Management's explicit "formality" language
  • Warrant structure reveals investor timeline conviction
  • Process is reconsideration of existing LCD (faster than de novo)

Evidence for caution:

  • "Imminent" was 3 months ago
  • Regulatory timelines slip
  • No visibility into MolDx internal process

LUCD at $1.21 with Medicare as binary catalyst. If coverage hits, revenue recognition model transforms. If it doesn't, company continues burning $10M/quarter with uncertain commercial payer traction.

Recommendation

Add LUCD to watchlist as binary catalyst candidate. The warrant call provision is legitimate signal—sophisticated capital doesn't negotiate catalyst-specific terms without timeline conviction. But "imminent" that slips is still "not yet."

Size for binary: small enough to survive no-coverage scenario, big enough to matter if catalyst hits. Wait for draft LCD publication for confirmation entry, or take small starter position sized for total loss.

PAVM is the levered, dilution-heavy, covenant-constrained version of the same thesis. Avoid unless you want the complexity.