Time Horizon: 12-18 months. Phase 1b interim data (2027) is the decisive resolution event. AACR (April 17-22, 2026) is the near-term catalyst. Q1 2026 earnings (May 7) is the first clean burn-rate confirmation.

Base Rate: Clinical-stage biotech, Phase 1 oncology with biomarker signal.

Base rate: Phase 1 oncology -> approval = 3.3-6.2%
With validated biomarker: ≈10.3%
With ORR > 40%: ≈25-30%
Conditional estimate for PLN-101095: 15-20%
Prior odds: 0.18 (midpoint 15% base -> 0.15/0.85)

Epistemic State: Doorway. Two patterns fit the evidence. 60% Pattern A (cash discount + working mechanism), 40% Pattern B (n=10 is noise). Cannot collapse without Phase 1b data.

B -- Business Model

Pliant Therapeutics is a pre-revenue clinical-stage biotech. Zero revenue. 49 employees. The entire company is a single option: PLN-101095, an oral small molecule dual integrin inhibitor (avb8/avb1), combined with pembrolizumab for checkpoint-inhibitor-refractory solid tumors.

The former lead program, bexotegrast (avb6/avb1 for idiopathic pulmonary fibrosis), was terminated in March 2025 after a DSMB stopped the BEACON-IPF trial. The stock lost approximately 80%. What remains is the cash treasury, the oncology pivot, and a proprietary library of 15,000+ integrin binding molecules developed over 10 years.

The company does not hold quarterly earnings calls. It issues press releases and presents at 1-2 healthcare conferences per year. Two analysts actively cover the name (Piper Sandler, Oppenheimer). This is unusually low transparency for a public company -- management narrative is untested under adversarial questioning.

Cash flow structure: CF_t = 0 for all foreseeable periods. Value = Cash + Option_value(PLN-101095) - Lease_liability. There is no path to revenue without drug approval or a partnership/licensing deal.

The Science -- Why the Product IS the Edge

What PLN-101095 Actually Does

Two-sentence version: it turns the water off at the tap. Every other TGF-beta approach tried to mop the floor while the tap was running. PLN-101095 blocks the integrin (avb8) that activates TGF-beta in the first place.

The Biological Logic

The immune system in checkpoint-inhibitor-refractory patients is not broken -- it is suppressed. T cells are present but physically excluded from tumors by a TGF-beta-created collagen fortress. This was established by Mariathasan et al. in Nature 2018 (n=348 urothelial cancer patients treated with atezolizumab): non-response was associated with a TGF-beta signaling signature in peritumoral fibroblasts. In mouse models, co-blockade of TGF-beta plus anti-PD-L1 facilitated T cell penetration into tumor centers and provoked vigorous anti-tumor immunity.

The suppression chain in tumors:

1. Cancer cells produce latent TGF-beta (inactive, wrapped in a "straightjacket" protein)
2. Regulatory T cells (Tregs) express avb8 integrin on their surface
3. avb8 recruits MMP-14 protease -> cleaves the straightjacket -> releases active TGF-beta
4. Active TGF-beta: excludes CD8+ T cells, suppresses cytotoxicity, promotes more Tregs
5. Checkpoint inhibitors cannot help because T cells never reach the tumor

PLN-101095 intervenes at step 3 -- blocks avb8 from activating latent TGF-beta. Already-active TGF-beta (needed for homeostatic functions in normal tissue) is unaffected. The activation mechanism is unique to avb8: unlike other integrins that use mechanical force (reversible), avb8 recruits the protease MMP-14 for irreversible enzymatic cleavage. This makes avb8 the rate-limiting step for TGF-beta-mediated immune suppression in the tumor microenvironment.

Critical finding from Laine et al. (Nature Communications 2021): in the tumor microenvironment, approximately 80% of avb8-expressing host cells are Foxp3+ regulatory T cells. About 40-45% of intratumoral Tregs express avb8. Only avb8-positive Tregs can efficiently activate TGF-beta. Ablating avb8 on Tregs in mouse models halved TGF-beta signaling in tumor-infiltrating T cells, dramatically increased CD8+ T cell cytotoxicity, and achieved 25-50% complete tumor control -- with no autoimmune side effects and no effect on Treg function in lymph nodes.

A second relevant finding: high avb8 expression and high PD-L1 expression rarely co-occur in human tumors (Takasaka et al., JCI Insight 2018). This means avb8 blockade may be most effective in PD-L1-low tumors where checkpoint inhibitors alone fail -- precisely the patient population with the greatest unmet need. The mechanisms are complementary, not redundant.

Why Every Other TGF-beta Approach Failed

No TGF-beta-targeting agent has ever been approved for cancer. The failure list:

Bintrafusp alfa (Merck KGaA/GSK): Bifunctional anti-PD-L1 antibody fused to TGF-beta trap. Neutralized active TGF-beta (all three isoforms) plus blocked PD-L1. Three Phase 2/3 failures in under a year: NSCLC, BTC (twice). $4.2 billion GSK partnership terminated September 2021. Why: pan-isoform neutralization removed tumor-suppressive TGF-beta. eBioMedicine/Lancet 2022 analysis showed patients treated with bintrafusp alfa had increased new metastases and greater tumor volumes compared to historical controls. The drug was mopping the floor while the tap ran, AND it was removing the safety mechanism at the same time.

Galunisertib (Eli Lilly): Small molecule TGFbRI (ALK5) kinase inhibitor. Blocked ALL TGF-beta signaling in ALL tissues. Caused dose-dependent heart valve lesions in animal models. Required intermittent dosing (2 weeks on, 2 weeks off) to manage cardiac risk, which likely limited efficacy. Development halted January 2020.

SRK-181 (Scholar Rock): Monoclonal antibody selectively targeting latent TGFbeta1 (not beta2 or beta3). Better selectivity than bintrafusp. DRAGON Phase 1 results: 23% ORR in clear cell RCC (n=30), 27% in melanoma (n=11), 0% in NSCLC (n=11). The NSCLC failure is important because NSCLC is one of PLN-101095's three Phase 1b cohorts. Shelved; Scholar Rock is seeking a partner.

Livmoniplimab (AbbVie): Anti-GARP mAb targeting TGF-beta1 bioavailability on Tregs. Failed Phase 2/3 in March 2026.

PF-06940434 (Pfizer) and CRB-601 (Corbus): Both monoclonal antibodies targeting avb8 integrin -- the same target as PLN-101095. Neither showed meaningful monotherapy activity. Pfizer terminated January 2025. Corbus discontinued March 2026 after SITC 2025 poster showed no monotherapy efficacy signal across 25 patients (8 grade 3 TEAEs, 1 drug-related). The ApexOnco analysis (Jacob Plieth, March 2026) explicitly states PLN-101095 is "the last player standing with a clinical-stage project that inhibits avb8."

Why Oral Small Molecule Beat Two mAbs on the Same Target

Two antibodies targeted the same integrin (avb8) and showed zero efficacy. PLN-101095, an oral small molecule, showed 40% ORR. Three mechanistic explanations:

Continuous exposure. Oral BID dosing provides 24/7 receptor occupancy. Monoclonal antibodies (IV every 2-3 weeks) follow peak-trough pharmacokinetics. If TGF-beta activation is a continuous process -- Tregs constantly activating latent TGF-beta -- then even brief windows of inadequate avb8 blockade may allow enough TGF-beta activation to maintain immune suppression. The small molecule eliminates these gaps.

Tissue penetration. Small molecules cross the fibrotic stroma that defines immune-excluded tumors. The Mariathasan paper showed TGF-beta-mediated exclusion involves fibroblast-rich, collagen-dense peritumoral stroma. A 150kDa antibody cannot efficiently penetrate dense collagen barriers -- precisely where TGF-beta exclusion operates. A small molecule can.

Dual avb8 + avb1 inhibition. PLN-101095 blocks both integrins. avb1 is expressed on cancer-associated fibroblasts (CAFs) and contributes to TGF-beta activation in the tumor stroma through mechanical force. The mAbs only targeted avb8. The dual inhibition blocks TGF-beta activation at two cellular sources simultaneously: avb8 on Tregs and tumor cells, plus avb1 on CAFs.

If PLN-101095 works because oral continuous exposure + tissue penetration + dual inhibition is mechanistically necessary, the competitive moat is real and structural. If it works for some other reason, or does not work at n=50, the thesis is dead.

The IFN-gamma Monotherapy Signal

The Phase 1 trial uses a distinctive 14-day monotherapy run-in: PLN-101095 alone for 14 days, then pembrolizumab added from day 15. This design isolates the drug's effect on the tumor microenvironment before any checkpoint inhibitor confounds the signal.

During the monotherapy window:

  • All 4 responders showed 4-13x increase in plasma IFN-gamma
  • Zero of 6 non-responders showed meaningful IFN-gamma increase
  • Binary separation -- step function, not a continuum

This is not just a biomarker. This is proof the mechanism works. The causal chain: PLN-101095 blocks avb8 -> TGF-beta stops being activated -> T cells in the tumor microenvironment reawaken -> they produce IFN-gamma -> IFN-gamma appears in blood. All of this happens BEFORE checkpoint inhibitor is added. The drug opened Lock 1 (TGF-beta suppression) on its own. Pembrolizumab then opens Lock 2 (PD-1 exhaustion).

Binary separation in a biomarker is extraordinary. Most oncology biomarkers (PD-L1, MSI-H, TMB) show overlapping distributions between responders and non-responders. A step function suggests the mechanism is either ON or OFF in a given patient -- either TGF-beta is their dominant immune suppression mechanism, or it is not. If validated, this becomes a companion diagnostic: measure IFN-gamma at day 14, treat only responders.

Normal plasma IFN-gamma levels are typically below 10 pg/mL. A 4-13x increase represents robust systemic immune activation, consistent with a significant release from TGF-beta suppression in the tumor microenvironment. IFN-gamma's role is well-characterized: directly kills tumor cells, upregulates MHC-I (making tumors more visible), activates macrophages toward anti-tumor M1 phenotype, inhibits angiogenesis, and recruits additional immune cells. The Merck KEYNOTE biomarker paper (Ayers et al., JCI 2017) established IFN-gamma-related mRNA profiles as predictive of PD-1 blockade response.

Caveat: P(perfect binary separation persists at n=50+) is low, likely below 20%. Some overlap will appear. But even 80% sensitivity / 90% specificity would be unprecedented for a TGF-beta pathway approach.

Phase 1 Data: Granular Detail

Phase 1 dose escalation enrolled 16 total patients across 5 dose cohorts. This dataset was compiled from 10-K, press releases, and company slides via OncLive -- the full picture is not available in any single source.

CohortDosenResponsesNotes
1250mg BID10Sub-therapeutic
2500mg BID20Sub-therapeutic
31000mg BID63 (50% ORR)Selected for Phase 1b expansion
41000mg TID40Same dose level, TID schedule failed
52000mg BID31

Among 10 patients at therapeutic doses (>=1000mg BID): 4 responses (1 complete response, 3 partial responses). ORR = 40%. 95% CI: [12%, 74%].

Responding tumor types: cholangiocarcinoma, melanoma, HNSCC, NSCLC -- four different cancers. Cross-tumor activity is a mechanism signal.

Average tumor reduction among responders: 71% (Oppenheimer conference, February 2026, q=0.70 -- not in SEC filings). These are deep responses, not marginal shrinkage.

Median duration of response: 15 months.

Safety profile from company slides (OncLive, q=0.70): 50% rash, 19% anemia, 19% diarrhea, 31% serious TEAEs, 12.5% discontinuation (2 of 16 patients). This detail was omitted from press releases. The 31% serious TEAE rate is unreliable at n=16 (one additional event changes the rate by 6 percentage points) but warrants monitoring. Oncology tolerates higher AE rates than fibrosis, and no specific safety signal was flagged by investigators.

Cohort 4 (1000mg TID) showed 0 responses despite the same dose level as Cohort 3 (1000mg BID). TID dosing may cause different PK profile or this may be small-n noise. Phase 1b selected 1000mg BID -- the dose that worked.

Probability of Approval

StageBase RatePLN-101095 Conditional
Phase 1 -> approval (all oncology)3.3-6.2%--
Phase 1 WITH biomarker≈10.3%--
Phase 1 with ORR > 40%≈25-30%--
PLN-101095 overall estimate--15-20% (3-5x base rate)

Positive factors: ORR above 40%, binary biomarker, 1 CR, 15-month DOR, novel mechanism, unmet need, no competitor, cross-tumor activity, academic KOL presenting at AACR.

Negative factors: n=10, 31% serious TEAEs, no randomized comparator, entire TGF-beta class has failed, single-program company, no C-suite FDA approval track record.

For accelerated approval, FDA precedent suggests ORR >= 25-30% in approximately 50-100 patients with durable responses. Median approved ORR across 2010-2019 accelerated approvals was 39%. PLN-101095's Phase 1 ORR of 40% is at that threshold but needs confirmation in Phase 1b.

Patent and IP Position

300+ applications pending, 14 US + 44 foreign patents issued. Small molecule patents expire 2037-2047, covering through potential approval and peak commercial period. Integrin chemistry is non-trivial (structurally similar binding sites across the integrin family); competitor replication would take 3-5 years minimum. The proprietary library of 15,000+ integrin binding molecules represents 10 years of platform development. (10-K, lines 630-664)

China Supply Chain Risk

10-K discloses sole-source vendors in China for clinical drug supply. Tariff and sanctions disruption could delay Phase 1b timeline. "If our supply chain is interrupted...the development of our product candidates could be materially delayed." (10-K, lines 4715-4741)

Phi -- Financial Trajectory

Balance Sheet (December 31, 2025)

ItemAmount
Cash + equivalents$45.4M
Short-term investments$145.5M
Restricted cash$1.5M
Total liquid assets$192.4M
Long-term debt$0
Lease liability (total)$29.1M
Current liabilities$16.4M
Stockholders' equity$181.2M
Accumulated deficit-$859.4M
Shares outstanding61.9M
Net liquid assets$163.3M
Net liquid per share$2.64
Cash per share$3.11

Oxford Finance's $30M term loan was fully repaid October 14, 2025 ($1.828M prepayment penalty). All liens, covenants, and security interests were discharged. The balance sheet is unencumbered. (10-K Note 7, lines 7840-7848)

The cash portfolio is 80.5% investment-grade corporate debt ($139.9M), with 79.7% maturing within one year. Net unrealized losses: -$15K. Blended yield: 5.9%, generating $11.4M in interest income for FY2025. This is not a portfolio hiding losses. (10-K Note 3, lines 7688-7718)

Burn Rate

The critical detail the street has not processed: bexotegrast wind-down consumed $42.4M in FY2025 despite the program being terminated in March. Those costs -- CRO termination, trial close-out, vendor settlements -- go to zero in FY2026. Core PLN-101095 and preclinical external R&D expenses were only $8.9M in FY2025. (10-K, lines 6631-6646)

Q4 2025 total operating expenses: $23.6M, including $2.5M in one-time restructuring charges (10-K Note 17). True steady-state quarterly OpEx: $21.1M. $15M of payables settled in FY2025 (AP fell from $6.0M to $0.5M, accrued R&D from $14.4M to $4.8M) -- bexotegrast vendor close-outs, one-time.

Bottom-up FY2026 cash burn reconstruction:

Gross OpEx (Q4 adj x 4):            $84.4M
Less non-cash items:
  SBC (≈$12-13M, declining):        -$12.5M
  Depreciation (≈$1.5M):            -$1.5M
  Non-cash lease (≈$3.3M):          -$3.3M
Cash operating expenses:             ≈$67M
Less interest income (≈$8-9M):       -$8.5M
Net cash burn FY2026E:               ≈$58M
With Phase 1b ramp (H2 2026):       ≈$62-68M

Street consensus implies approximately $89M annual burn (derived from Q1 2026 EPS estimate of -$0.36). The gap is $27M/year -- entirely attributable to bexotegrast wind-down costs the street has not zeroed out.

ParameterStreet ConsensusPrimary SourcesGap
Annual cash burn≈$89M≈$62M-$27M/yr
Cash runwayH1 2028 (2.0 yr)Q1 2029 (2.9 yr)+0.9 yr
Cash at Phase 1b interim (2027)≈$88M≈$115M+$27M

Source: 10-K FY2025 (q=0.95). This is arithmetic from audited financials.

Quarterly Trajectory

QuarterR&DG&ATotal OpExNotes
Q1 2025≈$37.4M≈$16.8M$54.2MFull bexotegrast
Q2 2025≈$38.0M≈$12.3M$50.3MWind-down begins
Q3 2025$17.9M$10.3M$28.2MRestructuring #1 effective
Q4 2025$15.6M$8.0M$23.6MIncludes $2.5M restructuring #2
Q4 adj$13.4M$7.7M$21.1MTrue steady-state

Burn rate is on a declining trajectory. FY2024: $228M -> FY2025: $156M -> FY2026E: ≈$94M gross ($62M cash). The derivative is strongly negative.

Cash Flow Statement Reconciliation (FY2025)

Net loss:                  -$149,344K
+ SBC:                      +$26,234K
+ Depreciation:              +$1,696K
+ Non-cash lease:            +$3,277K
+ Investment accretion:      +$3,956K
+ Debt extinguishment loss:  +$1,828K
- Gain on sale:                  -$44K
Working capital changes:
  Prepaid decrease:          +$2,541K
  AP decrease:               -$5,480K
  Accrued decrease:         -$12,167K
  Lease liability decrease:    -$876K
  Other:                        +$43K
= Operating cash flow:     -$128,336K

Critical observation: AP fell from $5,960K to $480K and accrued R&D from $14,363K to $4,804K. That is $15M of payables settled -- bexotegrast vendor close-outs. One-time; FY2026 working capital should be neutral or positive. Cash paid for operating leases: $5,562K/yr. Cash paid for interest: $2,539K (zero going forward). (10-K, lines 7210-7275)

Dilution Risk

A new $300M universal shelf registration (including $50M ATM with Leerink Partners) was filed March 30, 2026, replacing the prior $150M ATM facility. Zero draws have been made on either facility. Authorized shares: 300M common (61.9M outstanding -- massive headroom).

At $1.35, a $50M ATM draw would create 37M new shares -- 59% dilution. Management is aware of this math. The "use of proceeds" section includes language about potential acquisition of companies or technologies, suggesting management is positioning for in-licensing or acquiring additional programs. Good data at AACR -> raise at higher prices -> dilution modest -> acquire complementary assets. (S-3 filing, March 30, 2026)

Valuation Snapshot

MetricValue
Market cap$84M
Cash$192.4M
Enterprise value-$108M
Cash/share$3.11
Net liquid/share$2.64
Price/cash0.44
Price/book0.46

K -- Competitive Position

Monopoly in avb8 Clinical Oncology

PLN-101095 is the only clinical-stage avb8 inhibitor with demonstrated anti-tumor activity. Every competitor has exited or shelved:

  • Pfizer PF-06940434 (avb8 mAb): Terminated January 2025. No efficacy data published.
  • Corbus CRB-601 (avb8 mAb): Deprioritized March 2026 after SITC 2025 poster showed safety-only data. No monotherapy efficacy across 25 patients, 8 grade 3 TEAEs, 1 drug-related. CRB-601 required monotherapy activity as a gating factor but "no meaningful effect was seen." (ApexOnco, Jacob Plieth, March 2026)
  • Scholar Rock SRK-181 (latent TGFbeta1 mAb): Shelved after DRAGON trial. 23% in RCC, 27% melanoma, 0% NSCLC. Seeking partner.
  • AbbVie livmoniplimab (anti-GARP): Failed Phase 2/3 (March 2026).
  • Incyte INCA33890 (TGFbetaR2 x PD-1 bispecific): Phase 3 in MSS CRC. Different target, different indication. Validates TGF-beta pathway without competing directly. 13% ORR in MSS CRC (n=105).

The competitor dropout validates the modality hypothesis: oral small molecule continuous exposure may be fundamentally superior to intermittent IV mAb for avb8 inhibition.

Best-in-Class ORR in ICI-Refractory Setting

ApproachORRnNotes
PLN-101095 + pembro40%10Secondary ICI-refractory, 1 CR + 3 PR
mRNA-4359 (Moderna)24%varies
ICI rechallenge19.4%meta-analysisStandard of care
Visugromab14-18%varies
Solnerstotug14%varies

TAM

ICI-refractory solid tumors. Approximately 50% of ICI-treated patients develop resistance. ICI market exceeds $50B. Even capturing 1-2% of the refractory market = $500M-$1B peak sales.

Moat Assessment

TypeStrengthDetail
K_dataModerate15,000+ integrin binding molecules. Proprietary screening assays. Phase 1 clinical dataset with biomarker correlation. 10 years of platform development.
K_regPotentialIf PLN-101095 reaches accelerated approval, regulatory exclusivity provides years of protection. Currently zero.
K_scaleWeakOral small molecule = scalable manufacturing. But no manufacturing infrastructure.
DurabilityFragileMonopoly is real but depends entirely on n=10 Phase 1 data confirming in Phase 1b. One safety signal or efficacy miss = zero.

G -- Governance

Board

10 of 11 directors are independent (91%, well above Nasdaq minimum). Staggered three-class structure. Lead Independent Director: Hoyoung Huh, M.D., Ph.D. (McKinsey partner, Cornell MD/PhD). R&D Committee provides unusual clinical oversight.

Key directors:

  • David Pyott (Class I, since 2021, age 71): Former CEO Allergan (1998-2015). Board: Alnylam, BioMarin. Heavyweight pharma operator.
  • Steve Krognes (Class III, since June 2024, age 56): Former CFO Denali Therapeutics and Genentech. M&A at Roche. Board: Denali, Guardant Health, argenx. Harvard MBA, Wharton BS. Financial and M&A expertise.
  • John Curnutte (Class II, since 2017, age 73): Former EVP R&D Portola Pharmaceuticals. President Schering-Plough Biopharma. Harvard MD/PhD. R&D Committee Chair.
  • Katharine Knobil (Class II, since 2022, age 61): Former CMO Agilent Technologies. 20+ years GSK (corporate CMO). Audit and NomGov committees.
  • Thomas McCourt (Class II, since 2023, age 67): CEO Ironwood Pharmaceuticals. Former Amgen, Novartis commercial.
  • Suzanne Bruhn (Class III, since 2016, age 61): CEO Charcot-Marie-Tooth Association. Former CEO Promedior (fibrosis biotech). Directly relevant to integrin platform heritage.

Double-trigger change-in-control provisions (shareholder-friendly). No golden parachutes. No tax gross-ups. CEO CIC cash severance approximately $1.5M. Equity acceleration near zero at current prices (all underwater). Clawback policy per Dodd-Frank. Anti-hedging/pledging policy. No related party transactions. No going concern language. (DEF 14A, April 2025)

Compensation

CEO Bernard Coulie total compensation: $6.6M ($641K salary + $384K bonus + $2.2M stock awards + $3.3M option awards + $53K other). This is 8% of the current $84M market cap -- severe mismatch. (DEF 14A, Summary Compensation Table)

Compensation structure: base salary benchmarked to peer group 50th percentile. Annual cash incentive target: CEO 60% of base (max 150% of target). Equity split: 50% stock options (4-year monthly vest), 50% RSUs (3-year annual vest). No PSUs.

Bonus performance metrics: clinical/regulatory milestones 55% weighting, research 15%, finance/operational 30%. All achieved at 100% pool funding. Individual modifiers 100-110%. These metrics are appropriate for pre-revenue clinical-stage biotech.

The peer group was set at $350M-$3.5B market cap (22 companies including Arcus, Arvinas, Blueprint, Cytokinetics, Scholar Rock). This was appropriate when PLRX traded at $15-18 ($1B market cap). At $84M, the peer group is 4-42x larger.

All 7.5M executive stock options carry an average exercise price of $15.28 versus the current stock price of $1.35. Intrinsic value: zero. Unrecognized SBC: $24.6M over approximately 2 years -- dead-weight expense with no retention value. (10-K Note 11)

Say-on-Pay failed in 2024 (45.3% approval, down from 97.2% in 2023). Root cause: equity grants sized at $15.15 average price, then granted at $34.65 after a data catalyst. The board responded with grant timing policy changes, stockholder engagement (43% of shares represented accepted meetings), new compensation consultant (Pearl Meyer), and new Compensation Committee Chair (Darren Cline). (DEF 14A)

CEO direct share ownership: Coulie/Leyman Family Trust 409,317 shares + direct 296,318 shares + exercisable options 1,323,626. Total: 2,029,261 shares (3.31%). Direct shares worth approximately $966K at current prices versus $6.6M annual compensation. Skin in the game has been severely diluted by the stock collapse.

CMO Departure

Eric Lefebvre, M.D., was terminated effective December 15, 2025. The 8-K filed December 8, 2025 states: "On December 2, 2025, Pliant Therapeutics, Inc. notified Eric Lefebvre, M.D., Chief Medical Officer, that Dr. Lefebvre's employment with the Company will terminate effective December 15, 2025."

Company-initiated, not voluntary resignation. Timeline:

  • December 2: Company notifies Lefebvre he is being terminated
  • December 4: Lefebvre presents PLN-101095 data at Piper Sandler (2 days after notification)
  • December 8: 8-K filed
  • December 15: Separation effective
  • December 16: Consulting agreement begins
  • January 9: Formal separation per 10-K Exhibit 10.10

Part of restructuring round #2 ($2.5M charges, 10-K Note 17). Retained as consultant. COO Minnie Kuo (Gilead/Vir clinical operations) absorbed clinical responsibilities. With 49 FTEs and a single program, CMO + COO was redundant.

This is less concerning than a voluntary departure over clinical disagreement. The fact that Lefebvre presented data 2 days after being notified, without apparent disruption, suggests an orderly transition. However, no key person insurance is in place. (10-K Risk Factors)

Insider Activity

January 20, 2026: CEO ($114K), CFO ($31K), COO ($9K), CHRO ($10K) all sold on the same date -- consistent with 10b5-1 automatic plans, not discretionary selling. Dollar amounts are small. Net insider sales over 12 months: -$164K. Zero open-market purchases. Bearish but at small absolute amounts. (Form 4 filings)

Institutional Ownership

Holder%FilingSignal
Tang Capital9.54%13D (active intent)Known biotech activist. One point below 10% poison pill trigger.
Deep Track Capital9.37%13GHealthcare-focused. At maximum below pill.
T. Rowe Price7.88%13GPassive.
BlackRock7.82%13GIndex/passive.
Vanguard4.99%13GIndex/passive.
Paradigm BioCapital4.98%13GBiotech-focused.
Citadel3.4%13GNew position. Market maker + possible directional.
Morgan Stanley0.9%13G/AReduced from >5%. Exiting.
Point72 (Steve Cohen)0.1%13G/AEssentially exited.

Top 6 holders own 44.6% of the company. The counterparty structure is visible: generalist institutions (Morgan Stanley, Point72) have exited. Healthcare-specialist value funds (Tang, Deep Track, Paradigm) hold at or near maximum positions.

Tang's 13D filing (not 13G) signals active intent. Kevin Tang is a known biotech activist. At 9.54% with a 10% non-passive trigger on the poison pill, he is one percentage point from triggering the rights agreement. Possible actions: waiting for AACR data to engage the board, building case for strategic alternatives or return of capital, positioning for negotiated acquisition premium.

Poison Pill

Renewed March 3, 2026 to March 2027 (8-K). Dual-tier trigger: 10% for non-passive investors (activists, strategic acquirers), 20% for passive investors. This is aggressive -- typical pills use 15-20%. Mechanism: crossing threshold allows all other holders to buy stock at 50% discount, massively diluting acquirer. Rights Agreement explicitly states not in response to any specific takeover offer. Goodwin Procter LLP advising.

At negative EV, the pill serves dual purpose: prevents opportunistic accumulation at distressed prices forcing liquidation without premium, and ensures any control transaction goes through the board for a negotiated premium.

Beta -- Factor Profile

MetricValue
Beta (SPX)1.25
Idiosyncratic vol62.5%
Total vol65.1%
%Idio variance≈92%
Short % float6.3%
Days to cover6.0

92% idiosyncratic variance, well above the 75% target. Market and sector explain less than 10% of return variance. Sector correlation is essentially zero -- this stock does not move with XBI or IBB. Pure company-specific bet with minimal factor contamination.

r_PLRX = alpha + 1.25 x r_SPY + ≈0 x r_XBI + epsilon

Var contribution:
  Market: ≈8%
  Sector: ≈0%
  Idio: ≈92%

Momentum: 1-week +14.9%, 1-month +1.5%, 1-year +8.0%. 52-week range: $1.09-$1.95. Currently at 30% of range. Not a momentum-driven name.

RSI(14D): 53.8 -- neutral. 50-day MA $1.26 (above), 200-day MA $1.42 (below). Stock below 200-day.

Options Positioning (as of April 6, 2026)

Four expirations available:

ExpiryDTETotal OIP/C RatioCharacter
April 1710d4,7580.19 (bull)AACR event. 75% of all OI.
May 1538d130.30Empty.
July 17101d1,5293.27 (bear)Insurance. 1,170 puts at $1.
October 16192d811.61Thin.

Options Floor/Ceiling Map

$5.00  ....  Jul 87 calls (negligible)
       |
$2.00  XXXX  Apr 3,949 calls CEILING (expires Apr 17!) + Jul 265 calls (weaker)
       |
$1.35  <-- CURRENT PRICE
       |
$1.00  XXXX  Apr 681 puts + Jul 1,170 puts FLOOR (persists through July)
       |
$0.50  ....  Oct 50 puts (negligible)

The $2 call wall (3,949 contracts = approximately 200K shares of dealer delta-hedging resistance, nearly a full day of average volume) expires April 17 -- the day AACR opens. The oral presentation is April 18. The ceiling lifts mechanically the day before the data drops.

The $1 put floor (681 April + 1,170 July = 1,851 total contracts) provides approximately 200K shares of dealer delta-hedging support through July 17. The floor strengthens after April expiry (1,170 > 681).

Implied probability distribution (Q-measure from deltas, 10-day horizon):

  • P(below $1.00) = 14%
  • P($1.00 - $2.00) = 37%
  • P(above $2.00) = 49%

Adjusted to risk-neutral then real-world (beta=1.25, SR_market≈0.4):

  • P_market(below $1.00) ~ 17%
  • P_market($1.00 - $2.00) ~ 39%
  • P_market(above $2.00) ~ 44%

IV term structure inverts: 228% at 10-day to 112.5% at 101-day. Market expects vol crush after AACR resolves the near-term binary.

Delta -- Expectations Gap

What Price Encodes

At $1.35 with approximately $177M estimated current cash (after one quarter of burn from $192M):

Pipeline value (implied) = Market cap - Cash + Lease = $84M - $177M + $29M = -$64M

The market says PLN-101095 will destroy $64M of shareholder value. Solving for implied P(success):

At V_success = $500M:  P*(success) = 6.8%
At V_success = $300M:  P*(success) = 12.4%

Market-implied P(success): 7-12%. Phase 1 oncology base rate with biomarker: 10.3%. Research estimate: 15-20%.

Alpha vs Beta

Expected 12-18 month total return (to cash floor): +77% (from $1.35 to $2.39)
  Market beta contribution (1.25 x ≈8%):              +10%
  Sector beta contribution (≈0 x XBI):                 ≈0%
  Idiosyncratic alpha (cash floor convergence):        +67%  <- the actual thesis

Pipeline optionality (additive, not included in base):
  At 15% P(success) x $300M: +$0.73/share additional (+54%)
  At 25% P(success) x $500M: +$2.02/share additional (+150%)

92% of variance is idiosyncratic. The thesis stands or falls on two arithmetic claims (burn rate, cash floor) and one judgment call (pipeline probability). Factor exposure is minimal.

Gap 1: Burn Rate (|Delta| = Large, q = 0.95)

Street consensus implies approximately $89M/year cash burn. Bottom-up reconstruction from Q4 2025 audited financials yields approximately $62M/year. The gap is entirely attributable to bexotegrast wind-down costs ($42.4M in FY2025 dropping to approximately $0 in FY2026) that the street has not modeled.

This extends runway by 0.9 years. Phase 1b interim data arrives with approximately $115M remaining (research estimate) versus $88M (street). The company is not about to run out of money.

What would close it: Q1 2026 earnings report (May 7, 2026). The first clean quarter without bexotegrast costs. When the burn prints at $15-17M instead of the consensus $22M, the runway recalculation happens automatically.

Gap 2: Cash Floor (|Delta| = Large, q = 0.95)

The stock trades at $1.35 versus $2.86/share in cash ($3.11/share gross, $2.64 net of lease liability). Price-to-cash: 0.47.

For this discount to be rational, either burn must dramatically exceed estimates AND no value event occurs before cash exhaustion, OR management dilutes massively at current prices, OR forced liquidation occurs below NAV. All three are contradicted by: the declining burn trajectory (Q4 run-rate, bexotegrast elimination), zero ATM draws to date, and activist positioning (Tang/Deep Track at 19% combined, near the poison pill trigger).

What would close it: Burn rate confirmation at Q1 earnings. Tang Capital 13D activity (board engagement, strategic alternatives push). Acquisition inbound interest (at current prices, any acquirer paying up to 129% premium receives the pipeline for free after netting out the cash). Time itself -- forced sellers are exhausted (Morgan Stanley, Point72 have completed exits).

The Acquisition Math

Acquirer buys 100% at 50% premium:
  Purchase price: $84M x 1.50 = $126M
  Cash received: $192M
  Net: Acquirer is PAID $66M to take PLN-101095 + platform + 58 patents

Acquirer buys 100% at 100% premium:
  Purchase price: $84M x 2.00 = $168M
  Cash received: $192M
  Net: Acquirer is PAID $24M to take PLN-101095 + platform + 58 patents

Acquirer buys 100% at cash value:
  Purchase price: $192M
  Cash received: $192M
  Net: PLN-101095 + platform + 58 patents FOR FREE

To pay NET $0 for the pipeline: acquirer pays $192M (129% premium, $3.11/share)

Potential acquirers: Merck (pembrolizumab combo partner, natural synergy), Incyte (TGF-beta pathway adjacent, Phase 3 in MSS CRC), any mid-cap oncology company wanting ICI-resistance mechanism at zero net cost. This is why the poison pill exists and why Tang filed a 13D.

Gap 3: Pipeline Probability (|Delta| = Very Large, q = 0.50)

Market prices P(success) at 7-12%. Research suggests 15-20%, driven by: (a) IFN-gamma binary biomarker separation not broadly disseminated (conference/slides only, not in press releases), (b) competitive monopoly in avb8 not priced (competitor dropout), (c) stale sell-side coverage from the bexotegrast era (JP Morgan $0 target set before December 2025 Phase 1 data; only 2 of 6 analysts updated post-data), (d) bexotegrast PTSD bleeding into PLN-101095 valuation despite being a different drug targeting a different integrin for a different disease.

This is the largest gap by dollar value ($109-164M, or $1.76-$2.65/share) but lowest by credibility. P(success) is judgment, not arithmetic. The 95% CI on 40% ORR at n=10 is [12%, 74%].

What would close it: AACR oral presentation (April 17-22) if it includes data beyond the original 10 patients. Phase 1b interim data (2027) is the decisive resolution. Partnership or licensing agreement would partially validate mechanism.

Gap 4: Institutional Dynamics (|Delta| = Large, q = 0.95)

Generalist forced sellers (Morgan Stanley from >5% to 0.9%, Point72 to 0.1%) have created selling pressure disproportionate to fundamentals. In an $84M market cap, a 5M+ share liquidation by Morgan Stanley mechanically depresses the price regardless of value. This selling is exhausted or nearly exhausted. The natural buyer base (Tang, Deep Track, Paradigm, Citadel) is already positioned. Any incremental buying faces a thin float with limited sellers.

What would close it: Time. The sellers are done.

Disconnect Map

DimensionStreet/Price SaysPrimary Sources ShowDelta Direction
Burn rate≈$89M/yr≈$62M/yrStreet too bearish
Runway2.0 years2.9 yearsStreet too bearish
Pipeline value-$64M+$45M to +$100MStreet too bearish
CMO departureNegative signalFired in restructuring (less bearish)Street slightly too bearish
Dilution riskHighReal but not imminent (zero draws)Approximately correct
Comp/retentionNot pricedCEO comp 8% of market cap, all equity underwaterStreet too bullish
Safety profileNot widely known50% rash, 31% serious TEAEsStreet too bullish
Institutional dynamicsBroad sellingForced selling exhausted, specialists at maxStreet too bearish

Five dimensions net bullish, one approximately correct, two bearish. The bullish gaps are larger in magnitude and higher in source credibility than the bearish gaps.

Steelman Bear Case

The strongest argument against this thesis is not that n=10 is noise. It is that the cash discount is rational because management will destroy the cash before any value event occurs.

The argument:

  1. CEO compensation ($6.6M) is 8% of market cap with no meaningful retention mechanism (all equity underwater). Management has no incentive to return capital or sell the company -- their pay is cash-denominated, not equity-linked. They are better off continuing to operate and collect salary than liquidating.

  2. The S-3 shelf ($300M universal + $50M ATM) is loaded. At some point, management needs to raise capital for Phase 1b expansion. Even a "responsible" $30M raise at $1.50 dilutes 32%.

  3. The Phase 1b IND specifies three tumor cohorts (NSCLC, ccRCC, high TMB). Enrollment across three cohorts is expensive and slow. The $62M/year burn estimate assumes current run-rate; Phase 1b ramp could push FY2027 burn to $80-90M, shortening runway.

  4. Scholar Rock's SRK-181 showed 0% ORR in NSCLC despite activity in other tumors. If NSCLC is biologically unresponsive to TGF-beta blockade regardless of modality, one of three Phase 1b cohorts may fail, narrowing the addressable market and reducing partnership value.

  5. No one on the C-suite has taken a drug through FDA approval. CEO's prior exit was ActoGeniX sold for $60M. The skill set to navigate accelerated approval -- FDA engagement, manufacturing scale-up, commercial preparation -- is unproven.

  6. 31% serious TEAEs at n=16. If this rate persists or worsens in Phase 1b, the therapeutic window narrows. The FDA precedent for accelerated approval requires favorable benefit-risk, not just ORR.

  7. No earnings calls. No analyst Q&A. Management controls the narrative through press releases and fireside chats with a single moderating analyst. When the only voice is management's, bullish signals are overrepresented.

The rebuttal: Tang Capital's 13D filing directly addresses points 1-3. An activist with 9.54% near the pill trigger has the leverage to force strategic alternatives -- return of capital, asset sale, or merger -- if management burns cash without delivering value. The board (Pyott, Krognes) has M&A experience. Double-trigger CIC protections are shareholder-friendly. Tang's presence is the governance mechanism the thesis depends on.

Points 4-7 are real. They cannot be refuted with current evidence. They are the reason the epistemic state is 60/40 and not 80/20.

Kill Criteria

Thesis dies if:
- AACR data shows ORR regression to <20% in expanded dataset        -> exit to cash floor
- Phase 1b enrollment fails to start by Q3 2026                     -> exit (operational failure)
- Q1 2026 burn exceeds $22M/quarter ($88M annualized)               -> reassess burn model
- Tang Capital files 13G (downgrading from 13D = abandoning activism) -> exit
- S-3 ATM draw at <$1.50/share (dilutive raise at floor)            -> exit
- Safety signal: any patient death attributed to drug                -> exit immediately
- Stock below $0.90 (breach of $1 options floor)                     -> exit

Thesis strengthened if:
- AACR includes new patients beyond original 10 with ORR >= 30%     -> re-rate toward cash floor
- Q1 burn prints <$17M (confirming our $62M model)                  -> add confidence on floor
- Partnership announced (validates mechanism)                        -> full re-rate
- Tang Capital increases position or files 13D amendment             -> activist catalyst imminent
- IFN-gamma binary separation confirmed in expanded dataset          -> pipeline P(success) increase

What to Watch

EventWhenWhat Matters
AACR oral presentationApril 17-22, 2026New patients beyond original 10. Presenter: Timothy Yap MD PhD, MD Anderson (academic KOL, independent). Session CTMS01, April 18 10:21 AM PT. Recycled data = no new information.
April 17 options expiryApril 17, 20263,949 call OI at $2 expires. Ceiling lifts. 681 put OI at $1 expires but July floor (1,170) replaces it. Structural shift favors upside.
Q1 2026 earningsMay 7, 2026Burn rate -- first clean quarter. Cash balance confirms or denies runway estimate. Consensus expects -$0.36 EPS; our model implies -$0.25 to -$0.30.
Phase 1b enrollmentQ2 2026Operational proof that clinical machine works post-restructuring, post-CMO departure.
July 17 options expiryJuly 17, 2026$1 put floor (1,170 OI) expires. Mechanical support vanishes. If stock still sub-$2, risk increases.
Tang Capital activityOngoingAny 13D amendment, board seat request, or public letter.
Phase 1b interim data2027Resolves the doorway. ORR >= 25% -> full re-rate. ORR < 15% -> cash shell.

Evidence

EvidenceSourceCredibilityLR
Cash $192.4M, zero debt, 5.9% yield on IG corporate debt portfolio10-K FY2025, Note 3, Note 70.951.8
Bexotegrast wind-down $42.4M in FY2025 drops to ≈$0 FY2026; core PLN-101095 R&D only $8.9M external10-K FY2025, lines 6631-66460.951.5
Q4 2025 steady-state OpEx $21.1M/quarter (adjusted for $2.5M restructuring); AP settled $15M one-time10-K FY2025, Note 17, CF statement0.951.5
Phase 1b initiated: NSCLC, ccRCC, high TMB cohorts; 1000mg BID x 14d mono then pembro10-K FY2025; 8-K March 11, 20260.952.0
Only clinical-stage avb8 inhibitor with anti-tumor activity; Pfizer and Corbus both terminated10-K FY2025; ApexOnco March 20260.923.0
CMO Lefebvre terminated Dec 2, presented data Dec 4; company-initiated, retained as consultant8-K Dec 8, 20250.950.8
Tang Capital 9.54% via 13D (active intent); Deep Track 9.37%; both near 10% pill triggerDEF 14A Apr 2025; 13D filing0.951.8
Morgan Stanley reduced from >5% to 0.9%; Point72 to 0.1% -- generalist exits complete13G/A filings, Jan-Feb 20260.951.2
Poison pill renewed March 2026, 10% non-passive / 20% passive trigger, expires March 20278-K March 3, 20260.951.3
CEO comp $6.6M vs $84M market cap (8%); peer group $350M-$3.5B; Say-on-Pay failed 45.3%DEF 14A Apr 20250.950.7
All 7.5M executive options at $15.28 avg exercise; zero intrinsic value; $24.6M unrecognized SBC10-K Note 110.950.7
$300M S-3 shelf + $50M ATM filed March 30, 2026; zero draws on prior or current facilityS-3 March 30, 20260.950.7
Insider selling Jan 20, 2026: CEO $114K, all same date (10b5-1 plans); net -$164K 12moForm 4 filings0.950.7
No key person insurance; 49 FTEs; sole-source Chinese vendors for drug supply10-K Risk Factors, lines 4715-47410.950.8
Patent portfolio: 300+ pending, 14 US + 44 foreign; SM expiry 2037-204710-K lines 630-6640.951.2
No quarterly earnings calls; press releases only; 2 active analystsIR site, coverage records0.850.7
TGF-beta/avb8 immune evasion axis: Mariathasan Nature 2018 (n=348); Laine Nat Commun 2021; Dodagatta-Marri Nat Commun 2021; Takasaka JCI Insight 2018Peer-reviewed literature, multiple independent labs0.952.0
IFN-gamma binary biomarker separation: all 4 responders 4-13x increase during monotherapy; 0/6 non-respondersPress release Dec 4, 2025; company slides0.852.5
Phase 1 ORR 40% at therapeutic dose (n=10); 1 CR, 3 PR; cross-tumor: CCA, melanoma, HNSCC, NSCLC10-K; press release Dec 4, 20250.902.0
71% average tumor reduction among responders; median DOR 15 monthsOppenheimer conf, Feb 20260.701.8
Safety: 50% rash, 31% serious TEAEs, 12.5% discontinuation (2/16); omitted from press releasesOncLive citing company slides0.700.8
SRK-181: 0% ORR in NSCLC (n=11) despite 23% RCC, 27% melanomaScholar Rock ASCO 20240.850.8
CRB-601 (Corbus) discontinued: no monotherapy efficacy across 25 patients; avb8 mAb modality failedApexOnco (J. Plieth), March 20260.702.5
Incyte INCA33890 Phase 3 in MSS CRC validates TGF-beta pathway; different target/indicationPress + NCT, April 20260.851.5
Accelerated development plan: management chose to accelerate Phase 1b, not wait8-K March 11, 20260.851.5
AACR presenter: Timothy A. Yap, MD PhD, UT MD Anderson -- academic KOL, independent validationAACR program0.851.5
CEO: "data surpassed our expectations"Press release Dec 4, 20250.851.3
Bexotegrast safety was disease-specific (IPF), not avb8 class effect; low placebo event rate10-K + analysis0.901.8
Cohort 4 (1000mg TID): 0/4 responses despite same dose level as Cohort 3 (BID: 3/6)10-K; company slides0.850.9
Board quality: Pyott (Allergan CEO), Krognes (Genentech CFO), Curnutte (Portola R&D), Knobil (GSK CMO)DEF 14A Apr 20250.951.3