INBRX-106

INBRX-106 is Inhibrx Biosciences' hexavalent OX40 agonist antibody - six binding arms instead of the usual two - designed to do what every prior OX40 drug has failed to do: actually wake up T cells in solid tumors. On May 11, 2026, Inhibrx reported positive interim Phase 2 data in first-line head and neck cancer: 44.0% objective response rate (ORR; 11 of 25 patients, including 3 complete responses) for INBRX-106 plus pembrolizumab versus 21.4% (6 of 28, zero complete responses) for pembrolizumab alone in the randomized HexAgon-HN study (NCT06295731) [1][2]. This is the first credible signal in a class that has consumed Genentech, AstraZeneca, BMS, and Pfizer programs. The Phase 1/2 backbone (NCT04198766) is active, not recruiting, n=296 [3]. The asset now carries most of the post-spinoff Inhibrx Biosciences (INBX) narrative - and the company has $161.7M cash against $175M long-term debt at a 17.5% effective rate [4][5].

Novel compound, never approved. Lead study NCT04198766 is a Phase 1/2 open-label trial in locally advanced or metastatic solid tumors - non-small cell lung cancer (NSCLC), head and neck, melanoma, gastric, renal, urothelial - testing monotherapy and pembrolizumab combination. Status: active, not recruiting, n=296 [3]. The HexAgon-HN Phase 2/3 in first-line (1L) HNSCC patients with PD-L1 combined positive score (CPS) ≥20 (NCT06295731) is active, not recruiting, enrollment target 410, and reported positive interim data on May 11, 2026 (see executive summary) [1][2]. A small investigator-sponsored neoadjuvant Phase 2 in stage II/III triple-negative breast cancer (NCT06353997, n=12) is recruiting at Providence Health [6]. No FDA designations - no breakthrough, no fast track, no orphan, as of the May 2026 disclosure. Sponsor is Inhibrx Biosciences (NASDAQ: INBX), the spin-out that retained the residual pipeline after Sanofi acquired the parent Inhibrx in May 2024 for INBRX-101 [7]. Financial position: $161.7M cash at March 31, 2026, ~$30M quarterly burn, runway into 2027, but encumbered by $175M long-term debt drawn at a 17.5% effective rate with a $15.75M final fee - total minimum future payments of $190.75M [4][5]. INBX has not formally guided to additional readouts but the Phase 2 follow-up and Phase 3 portion of HexAgon-HN are the next inflection points.

OX40 (TNFRSF4, CD134) is a costimulatory receptor that shows up on T cells after they've been activated - think of it as the gas pedal that comes online once the ignition is already turning. When OX40 gets engaged, effector T cells expand harder, survive longer, and resist exhaustion. The pitch for OX40 agonists was: pair them with a PD-1 blocker (which lifts the brake), and you get a much harder push on antitumor immunity than either alone. Genentech (MOXR0916), MedImmune/AstraZeneca (MEDI0562), Pfizer (PF-04518600), and BMS (BMS-986178) all chased this thesis with bivalent IgG antibodies. All produced disappointing single-agent activity (ORRs in the 4-6% range) and at best modest combination signals. Most programs were shelved [8][9]. INBRX-106's argument is that the failure wasn't the biology - it was the engineering. Bivalent antibodies only cluster two OX40 receptors at a time, which the company argues is insufficient to trigger strong downstream signaling. INBRX-106 has six binding arms, and the preclinical work published in J Immunother Cancer in 2025 shows tighter receptor clustering and superior T cell activation versus a bivalent comparator [10]. A second mechanistic wrinkle the original framing missed: OX40 is also expressed on tumor-infiltrating regulatory T cells (Tregs), which suppress antitumor immunity. The Treg-OX40 axis is double-edged - agonism could deplete Tregs via antibody-dependent cellular cytotoxicity (ADCC) if the antibody Fc is engineered for it, or it could potentiate Tregs and worsen suppression. Residual Treg activity is among the leading explanations for why bivalent OX40 programs underperformed [9]. The published Phase 2 data stresss peripheral CD8+ and CD4+ T cell proliferation (up to 15-fold) but does not explicitly address Treg modulation or ADCC engineering in INBRX-106's Fc [1] - a gap worth pressing the company on. The target itself (TNFRSF4) is well-characterized [11], but no agonist has ever cleared Phase 3.

Three trials matter. NCT04198766 is the Phase 1/2 backbone: dose escalation of INBRX-106 monotherapy plus pembrolizumab combination cohorts across solid tumors, with safety and maximum tolerated dose (MTD) as primary endpoints and objective response rate (ORR - the percentage of patients whose tumors shrink by at least 30%) as the readthrough signal [3]. Active, not recruiting, n=296. Notably, no detailed Phase 1 efficacy data has been formally published in a peer-reviewed venue despite enrollment completion; the May 2026 interim disclosure focused on the randomized Phase 2 portion of HexAgon-HN, not the Phase 1 backbone [1]. NCT06295731 (HexAgon-HN) is the registrational asset: a Phase 2/3 in first-line (1L) HNSCC patients with PD-L1 CPS ≥20, combining INBRX-106 with pembrolizumab versus pembrolizumab alone, n=410, primary endpoint ORR [2]. CPS (combined positive score) is a PD-L1 immunohistochemistry metric counting PD-L1-positive tumor and immune cells per 100 viable tumor cells; CPS ≥20 selects roughly the top ~25-30% of HNSCC patients by PD-L1 expression - the subgroup where pembrolizumab monotherapy is already approved as 1L treatment and produces historical ORRs around 23-30%. The bar to demonstrate added value is therefore real, and the May 2026 interim (44% vs 21.4%) sits well above it. NCT06353997 is a small neoadjuvant TNBC trial (n=12) run by Providence - too small to register, useful for translational signal [6]. Competitive context: Elpiscience's ES102 (China) is in NCT04991506 (completed) and NCT06623136 (recruiting, NSCLC) [12][13]. PD-L1/OX40 bispecifics are an adjacent format under preclinical and early clinical development by Innovent and others [14]; note that GS-1811 (Gilead/Arcus, Denikitug) is sometimes mislabeled as OX40-targeted but is in fact an anti-CCR8 antibody and is not a class competitor.

Our model estimates a 7% chance this drug is eventually approved. It starts from the historical base rate for Phase 2 drugs in this area (about 13%), then adjusts using ten facts about the trial and sponsor. What moves the number most: it is helped by an unusually multi-arm design (14 arms) and larger-than-typical enrollment for this phase; it is held back by the sponsor's thin or weak approval record and weak or limited earlier-phase results. The other facts land near average for this stage, so they leave the estimate roughly where the base rate put it.

Efficacy risk has been partially de-risked by the May 2026 interim but not eliminated. The OX40 class has produced more failed programs than nearly any other costimulatory target. Specific prior failures: MOXR0916 (Genentech) returned a 4% ORR and was halted in November 2019 after grade 3 rash/diarrhea blocked dose escalation [9]; MEDI0562 (AstraZeneca) showed no objective efficacy in advanced solid tumors and was discontinued [8]; PF-04518600 (Pfizer) returned 5.8% ORR with grade 3+ hepatotoxicity emerging early and was attributed to insufficient FcγR cross-linking [9]. Residual Treg activity and limited intratumoral OX40+ T cell density are recurring mechanistic explanations [9]. INBRX-106's hexavalent thesis directly addresses the clustering/cross-linking problem; whether it also addresses Treg modulation is unclear. Safety risk is real but to date appears manageable: the interim disclosure described 'predominantly low-grade' treatment-related adverse events (rash, diarrhea, fatigue, infusion reactions) and no treatment-related deaths in either arm [1]. Execution risk is acute. Inhibrx Biosciences has $161.7M cash at Q1 2026 against $175M long-term debt at a 17.5% effective rate; quarterly burn is approximately $30M, implying runway into 2027 absent further raises but with material debt-service overhang [4][5]. Equity dilution risk is elevated given the debt cost. Commercial risk: 1L HNSCC PD-L1-high is a crowded combination space dominated by pembrolizumab plus chemotherapy. To displace standard of care, INBRX-106 will need durable benefit, not just an interim ORR delta. The neoadjuvant TNBC trial is too small to matter commercially. Competitive risk: ES102 and emerging PD-L1/OX40 bispecifics could pressure the differentiation story if any show comparable signals.

Reweight upward but stay disciplined. The May 2026 interim is the first credible OX40-class signal in a decade and the differential is large (44% vs 21.4% with 3 CRs vs 0). That meaningfully changes the prior - geometry may actually be doing what the bivalent programs couldn't. But the numbers are small (25 vs 28), follow-up is undisclosed, and class history has eaten plausible-looking interim signals before. The single data point that would harden the thesis: duration of response holding up at 12+ month follow-up with the same delta over pembrolizumab. The other thing to watch: whether INBRX-106 produces clean single-agent activity in the Phase 1 backbone (NCT04198766) - no prior OX40 agonist has, and that would be the strongest possible validation of the hexavalent thesis. The Phase 1 data has not been publicly disclosed in detail despite enrollment completion, which is itself notable. On the company: INBX is a small-cap binary bet with material debt service. The parent's INBRX-101 was the asset Sanofi paid for [7]; what's left in INBX is a thinner pipeline where INBRX-106 carries most of the weight. Check back when (a) full Phase 2 HNSCC data publishes with duration of response, (b) ES102 reports any combination data from NCT06623136 [12], and (c) INBX reports Q2/Q3 2026 cash position. The 8-K cadence around financings is the cash-runway tell [5].