A framework published this month in Nature Biotechnology does not merely add a technique to the oncology toolbox. It dismantles a foundational assumption that has quietly shaped antibody-based trial design for two decades: that delivering a drug to a tumor is roughly uniform, and that target engagement follows from dosing.
The framework is called Single-cell Spatial Pharmacobiology (SSP). At its core, SSP maps drug delivery and target engagement at single-cell resolution within the tumor microenvironment. What it found, applied to solid tumor patients on antibody-based therapy, was pronounced spatial heterogeneity — specific cells receiving therapeutic concentrations while neighboring cells in the same biopsy received none. The heterogeneity followed a pattern. Periostin-rich extracellular matrix assemblies and fibroblast-activation-protein-positive cancer-associated fibroblast (CAF) neighborhoods acted as conserved stromal barriers, physically limiting antibody penetration into tumor beds. Sponsors designing Phase 1 trials have been measuring systemic exposure and calling it target engagement. SSP shows those two things can be completely decoupled inside the same tumor.
The Gap Between Dose and Delivery
Consider what a standard Phase 1 oncology trial actually measures. Plasma pharmacokinetics. Tissue biopsies read for biomarker expression. Receptor occupancy estimated from peripheral blood surrogates. None of these instruments resolve what SSP resolves: whether the antibody physically reached the cell carrying the target, and whether it engaged that target at the single-cell level in a spatially defined context.
The practical consequence is significant. A trial can report robust systemic exposure, clean PK curves, and even detect target expression in archival tumor tissue, while the drug is being physically blocked from its intended cellular targets by stromal architecture. The CAF neighborhoods identified in the SSP study are conserved across tumor types, which means this barrier problem transcends any single indication. Sponsors running antibody trials in pancreatic, breast, lung, or colorectal settings are all operating inside the same structural problem, with instruments that cannot see it.
Contrast this with immunoPET, the closest existing clinical-stage comparator. A 2023 review in Theranostics summarized immunoPET’s primary objectives as visualizing whole-body biodistribution of labeled antibodies — useful for understanding where a drug goes at the organ level, but operating at resolutions that cannot distinguish cell-to-cell variation within a tumor nodule. SSP does not replace immunoPET; it descends several orders of magnitude below it.
The molecular imaging adoption data underscores how long the field has operated without this resolution. A systematic review in PMC found that PET was used 4.7 times more frequently than SPECT in clinical drug development studies, with adoption accelerating markedly since 2010. The infrastructure investment is substantial. Yet all of that imaging capacity still cannot tell a trial team whether the antibody reached cell 47 in the tumor core or was stopped by a periostin matrix three cell-lengths from the vasculature.
Who Gets Exposed First
The most immediate operational impact falls on Phase 1 solid tumor programs using antibody-drug conjugates (ADCs), bispecifics, and checkpoint inhibitors in indications with known stromal density. The AMG 404 PD-1 antibody trial, a first-in-human, open-label, multicenter Phase 1 study in patients with advanced solid tumors, illustrates the exposure gap precisely: a clean systemic profile can coexist with heterogeneous intratumoral delivery that traditional endpoints never surface. Response heterogeneity across patients in antibody trials has always existed; SSP provides a mechanistic explanation at cellular resolution rather than a statistical description at cohort resolution.
Sponsors pursuing accelerated approval pathways face a compounding problem. The FDA’s March 2023 draft guidance, “Clinical Trial Considerations to Support Accelerated Approval of Oncology Therapeutics,” explicitly pushes away from single-arm designs toward randomized controlled trials, with renewed scrutiny on the quality of pharmacodynamic evidence underlying efficacy claims. If SSP data surfaces in the Phase 1 readout of a program headed toward accelerated approval, the FDA will reasonably ask whether the response-rate signal reflects genuine target engagement or whether a subset of non-responding tumors were simply undelivered. That is a data integrity question, and it now has a tool that can answer it.
The spatial proteomics market context matters here too. Grand View Research estimates the global spatial proteomics market at $87.5 million in 2024, growing to $199.8 million by 2030 at a 15.1% CAGR. That growth rate reflects instrument adoption, not regulatory integration. The commercial infrastructure for SSP-class analysis is scaling, but the regulatory frameworks for incorporating single-cell spatial pharmacobiology data into IND packages, CMC submissions, or pharmacodynamic endpoints have not been written yet.
What the Ops Team Should Do Now
If you are running a Phase 1 solid tumor antibody program with biopsies already in protocol, the window to incorporate SSP-compatible tissue handling is at amendment, not at analysis. The stromal barrier findings are conserved, which means the probability that your indication is affected is not low. The immediate practical step is a tissue procurement review: are your core biopsy specifications preserving spatial architecture at a resolution compatible with single-cell mapping? Standard FFPE protocols may not be sufficient. Engage your translational science lead before the next cohort opens, not after dose-escalation is complete and the tissue is already processed.
The harder strategic question is what SSP data means for go/no-go decisions at end of Phase 1. A program showing partial responses alongside SSP evidence of heterogeneous delivery has a more complex story than a response rate alone conveys. It may also have a more actionable one: if the barrier is stromal rather than target-negative, the path forward looks different than if the non-responding cells simply lack the antigen. Sponsors who capture this data early will be negotiating from a position of mechanistic clarity. Those who do not will be defending their Phase 2 design with aggregate endpoint data that the field is rapidly learning to distrust.
The signal to watch next is whether FDA’s Oncology Center of Excellence begins requesting SSP-class pharmacodynamic data in Type B meetings for antibody programs in stromal-dense indications. Pancreatic and triple-negative breast cancer programs are the highest-probability targets. The first advisory committee where this data appears in a sponsor briefing document will set the evidentiary standard. That meeting may be closer than the field’s current trial timelines suggest.
References
- Nature Biotechnology — “Single-cell spatial pharmacobiology for imaging antibody-based therapies in solid tumors”
- PubMed — Single-cell Spatial Pharmacobiology (SSP) framework and stromal barrier findings
- PMC — AMG 404 Phase 1 first-in-human study in advanced solid tumors
- FDA — “Clinical Trial Considerations to Support Accelerated Approval of Oncology Therapeutics,” draft guidance, March 24, 2023
- Grand View Research — Spatial Proteomics Market Report, July 15, 2025
- Theranostics, 2023 — “Advances and challenges in immunoPET methodology”
- PMC — Systematic review on molecular imaging adoption in clinical drug development
Moe Alsumidaie, MBA, MSF, is founder and Chief Editor of Vanguard Publications, which publishes Clinical Trial Vanguard, Pharma Vanguard and BullScope, and Head of Research at CliniBiz. He has two decades in clinical trial operations and data science, with earlier roles at Genentech, Abbott Vascular and Stanford University Medical Center, and is a guest lecturer in clinical trial sciences at Rutgers University.
