INBRX-106

INBRX-106 is Inhibrx Biosciences' hexavalent OX40 agonist antibody, and on May 11, 2026 it produced the first credible clinical validation this target class has ever seen. The HexAgon Phase 2 (NCT06295731) interim readout in first-line PD-L1 positive recurrent/metastatic head and neck squamous cell carcinoma showed a confirmed objective response rate of 44.0% for INBRX-106 plus pembrolizumab versus 21.4% for pembrolizumab monotherapy in a 53-patient evaluable population, an absolute delta of 22.6 percentage points, with three complete radiographic responses in the combination arm [10]. The bet the writeup has always described (bivalent OX40 agonists failed at Bristol Myers Squibb, GSK, MedImmune, and Pfizer because they could not cluster enough receptors, and a hexavalent format with six binding sites should fix that [1][7][11][12]) now has its first positive human signal. Inhibrx plans to start the Phase 3 portion of HexAgon in Q3 2026 and expects Phase 2 progression-free survival data in Q4 2026 [10]. The broader Phase 1/2 basket (NCT04198766) continues in NSCLC, melanoma, gastric, renal cell, and urothelial cancers [2]. The mechanism thesis is no longer speculative: the remaining questions are durability, safety at scale, overall survival, and whether Inhibrx Biosciences can execute a Phase 3 on a stretched balance sheet.

INBRX-106 is a novel biologic with no approvals anywhere and no publicly disclosed FDA breakthrough, fast-track, or orphan designations. The commercially decisive study is NCT06295731 (Inhibrx trade name: HexAgon), a Phase 2 in first-line PD-L1 positive recurrent/metastatic HNSCC, active but no longer recruiting, planned enrollment 410, primary endpoint objective response rate [3]. Positive interim data from 53 evaluable patients was reported May 11, 2026: 44.0% cORR in the INBRX-106 plus pembrolizumab arm versus 21.4% with pembrolizumab monotherapy, safety described as generally manageable, with pharmacodynamic evidence of up to 15-fold peripheral CD8+ and CD4+ T-cell proliferation in the combination arm [10]. Phase 3 initiation is guided for Q3 2026 and Phase 2 PFS for Q4 2026 [10]. The broader Phase 1/2 basket (NCT04198766) remains open at 340 planned patients across seven solid tumor types as monotherapy and in combination with pembrolizumab, tracking safety as its primary endpoint [2]. Publicly disclosed efficacy signals from that trial before the HexAgon readout were limited to preclinical and translational presentations at ASCO 2025 and Immunology 2025 [1], with no dose-expansion ORR figures released across seven years of enrollment, a gap that made the HexAgon readout the first real read on the platform. A smaller investigator-sponsored neoadjuvant TNBC study (NCT06353997, n=12) at Providence Health adds mechanistic color but is not commercially decisive [4]. Inhibrx Biosciences (Nasdaq: INBX) is the sponsor, spun out in 2024 after Sanofi acquired the original Inhibrx for the alpha-1 antitrypsin asset INBRX-101 and left the oncology programs as a separate publicly-traded entity [8]. Q1 2026 10-Q shows $161.7M cash and cash equivalents, roughly $30M per quarter operating burn ($25.2M R&D plus $5.7M G&A), and $175M total debt including a $75M term loan at a 17.5% effective interest rate [13].

OX40 is a docking site (receptor) on activated T cells. When another protein called OX40L (found on antigen-presenting cells) grabs onto OX40, it tells the T cell to keep going: divide more, live longer, remember the antigen. Think of it as pressing the accelerator on a T cell that has already been switched on. In tumors, most T cells that recognize cancer end up exhausted or fail to expand. An OX40 agonist is supposed to push those tumor-reactive T cells to proliferate, especially when paired with a PD-1 blocker like pembrolizumab that releases the brake at the same time. The catch: OX40 signaling requires many receptors clustering together on the cell surface. A standard antibody has two arms, which is not enough to force strong clustering. Every prior OX40 agonist that entered the clinic used a bivalent format and produced mediocre responses at best. BMS-986178 (Bristol Myers Squibb) Phase 1/2a showed no clear efficacy signal above the nivolumab and/or ipilimumab backbone [7]. MEDI0562 (MedImmune/AstraZeneca) Phase 1 produced one partial response in 19 evaluable patients [11]. GSK3174998 (ENGAGE-1) enrolled 138 patients across monotherapy and pembrolizumab combination arms and was formally discontinued when the data did not support continued development [12]. PF-04518600 (Pfizer) Phase 1 delivered a 5.8% monotherapy ORR and the program was terminated after Grade 3 or higher hepatotoxicity emerged during dose escalation [12]. INBRX-106 uses Inhibrx's single-domain antibody technology to build a hexavalent construct with six binding sites for OX40 [1]. In preclinical models the format drove stronger receptor clustering, T cell expansion, and tumor regression than bivalent controls [1]. The May 2026 HexAgon interim data are the first clinical confirmation that this receptor-clustering thesis translates: a doubling of confirmed ORR over the pembrolizumab benchmark, combined with a 15-fold expansion of peripheral CD8+ and CD4+ T cells in treated patients, is exactly the biology the preclinical work predicted [1][10].

The commercially relevant study is NCT06295731, branded HexAgon by Inhibrx, an active-not-recruiting Phase 2 in first-line recurrent or metastatic HNSCC selected for PD-L1 positivity (Combined Positive Score, or CPS, above defined thresholds; CPS is a ratio of PD-L1 staining tumor plus immune cells to viable tumor cells, multiplied by 100, with higher scores predicting stronger pembrolizumab benefit) [3]. Design: INBRX-106 plus pembrolizumab versus pembrolizumab monotherapy, planned enrollment 410, primary endpoint objective response rate. The comparator is the right one because pembrolizumab monotherapy is standard of care in this population, established by KEYNOTE-048 [6]. The randomized design with a within-trial pembrolizumab control arm is a smart choice: it neutralizes the historical variability in cross-trial pembrolizumab ORR estimates and delivers a clean apples-to-apples delta, which is exactly what the 44.0% versus 21.4% May 2026 interim reported [10]. The ORR primary endpoint gets Inhibrx to a signal quickly but does not resolve overall survival, which is what regulators and payers will require for full first-line approval. The Q4 2026 PFS readout is the next tell on durability. The Phase 3 portion planned for Q3 2026 will need to be OS-powered to convert this into an approvable package. NCT04198766 is the older Phase 1/2 dose-finding basket across seven tumor types, still recruiting to 340 patients as monotherapy and pembrolizumab combination, tracking safety as primary endpoint [2]. Seven years of enrollment without a public efficacy readout from this trial was, until May 2026, the strongest yellow flag on the program. NCT06353997 is a 12-patient neoadjuvant TNBC study run by Providence Health, useful for on-treatment biomarker work but not powered for efficacy claims [4]. Elpiscience's ES102, the other hexavalent OX40 agonist in development, is running an open-label single-arm Phase 2 in advanced NSCLC combined with toripalimab (NCT06623136, planned enrollment 40, active as of April 2025) and is now the only genuine head-to-head competitor in the hexavalent format [9].

Our model estimates a 4% 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 larger-than-typical enrollment for this phase and more secondary endpoints than usual; 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.

Confirmatory risk is now the biggest one. Phase 2 ORR readouts historically deflate meaningfully in Phase 3, especially when the Phase 2 sample size is small (n=25 in the treatment arm at the interim cutoff). The Q4 2026 Phase 2 PFS readout is the next meaningful test: a strong PFS delta would tighten the case, a modest one would raise durability concerns even with the positive ORR. Safety risk: OX40 agonism is on-target immune activation, so failure modes are cytokine release, hepatitis, colitis, and other immune-related adverse events that overlap with pembrolizumab's toxicity profile and can be additive. The interim safety readout was described as generally manageable, but the PF-04518600 program was terminated for Grade 3+ hepatotoxicity that emerged only during dose escalation [12], so extended Phase 3 follow-up is still where nasty liver signals typically show up. Execution and financing risk: Inhibrx Biosciences ended Q1 2026 with $161.7M cash and roughly $30M per quarter operating burn, giving approximately 5 to 6 quarters of runway at current burn, into H2 2027 [13]. That is thin against a Phase 3 HNSCC trial that will typically cost $150M+ over its life. The balance sheet also carries $175M in debt with a 17.5% effective interest rate and a $15.75M final fee obligation, which compounds runway pressure. A positive interim readout of this magnitude usually opens strategic options (secondary offering, partnership, or acquisition), but the financing overhang is real and any Phase 2 PFS disappointment would tighten the noose fast. Commercial risk: even with approval, first-line PD-L1 positive HNSCC is a modest addressable market. Approximately 66,000 new head and neck cancer cases annually in the US, with roughly 50-60% ultimately PD-L1 positive and eligible for pembrolizumab-based therapy, implies a first-line addressable population in the low tens of thousands. Payers will demand clear OS benefit before covering an added biologic on top of pembrolizumab. Regulatory risk: no FDA designations disclosed, and no public confirmation of a pre-Phase 3 Type B meeting on HexAgon design. The Phase 3 pivotal design and endpoint alignment with FDA is the next public signal to watch alongside the Q4 2026 PFS readout.

The story has changed. Before May 11, 2026 this program was the last credible attempt to prove that OX40 agonism could work as a checkpoint combination partner, and the base case was another funeral for the class. The HexAgon interim readout (44.0% vs 21.4% cORR, three complete responses, 15-fold peripheral T-cell expansion, manageable safety) is the first positive human signal any OX40 agonist has ever produced, and it validates both the mechanism and the hexavalent format thesis [10]. The next signals to watch: (1) Q4 2026 Phase 2 PFS readout from HexAgon: durability, not just response rate. (2) Q3 2026 Phase 3 initiation and its exact design, which will indicate whether FDA aligned on an accelerated approval pathway or requires a full OS-powered pivotal. (3) The Phase 1/2 basket (NCT04198766) NSCLC expansion cohort readout, guided for H2 2026, which will indicate whether the HNSCC signal generalizes. (4) Financing action: a secondary offering, licensing deal, or acquisition proposal is the natural sequel to a positive Phase 2, and its terms will price the market's read on the data. Given active-not-recruiting status and the interim readout already in hand, the ASCO 2026, ESMO 2026, and SITC 2026 cycles are the venues to watch for expanded datasets. Inhibrx has been announcing program milestones through 8-K filings [8], so the H2 2026 8-K cadence is the primary source of dated guidance.