B7-H3 as an Adaptive Surface Phenotype

Quiet Biology Framework | Scientific Support Document

This paper proposes that B7-H3 is best understood as an adaptive surface phenotype — an emergent property of the tumour under androgen suppression, inflammatory signalling, and microenvironmental stress. It further argues that the Quiet Biology framework addresses the upstream conditions that produce this state, complementing rather than replacing direct B7-H3 targeting strategies.

Abstract

In prostate cancer, most therapeutic targets behave like lineage markers — their expression depends on androgen receptor (AR) signalling and cellular differentiation. As treatment pressure increases, these markers often become unreliable: downregulated, heterogeneously expressed, or lost entirely. B7-H3 (CD276) does not follow this pattern.

Across the full disease spectrum — from hormone-sensitive disease to castration-resistant prostate cancer (CRPC), neuroendocrine variants, and AR-indifferent phenotypes — B7-H3 remains consistently expressed. Data from the JHU-PANORAMA single-cell atlas, encompassing nearly one million cells from 213 patients, confirm low inter- and intra-tumoral variability and broad cross-compartment distribution.^[1] This persistence is not incidental. It reflects a biological state that emerges under therapeutic and environmental pressure.

01Framing the Question

The central question this paper addresses is not whether B7-H3 is a useful therapeutic target. The evidence from the PANORAMA study makes that case clearly. The question is what B7-H3 expression represents — what biological state it signals, and what produces it.

The standard framing treats B7-H3 as a surface antigen: present, accessible, targetable. This framing is clinically productive. B7-H3 (CD276) is an immune checkpoint member of the B7 and CD28 families, highly overexpressed across a wide range of solid cancers and associated with poor prognosis.^[2] Antibody-drug conjugates, bispecific antibodies, and CAR-T approaches all follow from this framing, and the PANORAMA study's bispecific pairing framework extends this logic usefully.

The alternative framing proposed here treats B7-H3 as a readout: a surface expression of a coordinated biological state that the tumour has entered under pressure. In this framing, the question is not only how to target B7-H3 but what conditions drive its expression — and whether those conditions can be modified upstream.

The components of this argument are individually documented in the primary literature. The synthesis presented here is the contribution.

B7-H3 does not mark what the cell is. It marks what the cell is experiencing.

02From Lineage Marker to State Marker

Traditional prostate cancer targets share a common profile. PSA, KLK2, and PSMA are AR-dependent, linked to luminal differentiation, and sensitive to treatment-induced lineage shifts. As therapeutic pressure mounts, these targets shift with the tumour — downregulated under castration, lost in neuroendocrine transition, unreliable in double-negative disease. They mark what the cell has differentiated into, and when differentiation changes, they change with it.

B7-H3 is different in a specific way. Its low inter- and intra-tumoral variability — confirmed across the full disease spectrum in the PANORAMA atlas^[1] — is not consistent with a lineage marker. Lineage markers vary as lineages diverge. B7-H3 persists. This persistence, combined with its expression in both tumour epithelial cells and stromal compartments including endothelial cells, tumour-associated macrophages, and myeloid-derived suppressor cells,^[3] is more consistent with a coordinated tissue-level response than a clonal genetic trait.

The distinction matters therapeutically. A lineage marker becomes unreliable as the disease evolves. A state marker remains informative as long as the state persists — and the state B7-H3 reflects is one that treatment actively promotes.

A further distinction concerns when B7-H3 becomes relevant. It is not solely a treatment-induced phenomenon. Roth et al. demonstrated that B7-H3 is uniformly expressed across all prostate adenocarcinomas examined at radical prostatectomy, including low-grade foci, and is present in benign prostatic epithelium at a reduced but detectable baseline level.^[12] Inamura et al. confirmed B7-H3 expression in treatment-naïve diagnostic biopsies from localised prostate cancer, with high expression less frequent than in metastatic disease but present and prognostically significant.^[13] Most strikingly, PCRP-funded research comparing pre-treatment diagnostic biopsies with later CRPC specimens from the same patients found no significant change in B7-H3 protein expression from early to late stage in patients with aggressive disease — indicating that B7-H3 is established early in the biology of tumours destined to progress, not acquired later under treatment pressure.^[1]

This means the ADT/B7-H3 amplification argument and the baseline expression argument are not in tension. They describe two different but related phenomena: B7-H3 is present and informative from the point of diagnosis in aggressive-trajectory disease, and it is further amplified by the treatment that follows. The state marker is already forming when the man first presents for biopsy.

2aB7-H3 in Normal Tissue: A Brief Note on Biological Context

B7-H3 (CD276) is not exclusive to prostate tissue or to cancer. At the mRNA level it is ubiquitously expressed across human tissues including liver, heart, prostate, spleen, and thymus.^[2] At the protein level, however, a tight post-transcriptional regulatory mechanism — likely microRNA-mediated^[11] — suppresses translation in most normal tissues at steady state. The result is a wide therapeutic window: B7-H3 protein is largely absent or low in healthy tissue even where the transcript is present, but pathologically upregulated across a broad range of solid tumours.

In normal physiology, B7-H3 has established homeostatic functions independent of immune regulation. Picarda et al. demonstrated that B7-H3 is most highly expressed of all 54 tissues examined in adipose tissue, specifically in adipocyte progenitor cells, where it regulates glycolytic and mitochondrial activity. Mice knocked out for B7-H3 develop spontaneous obesity, metabolic dysfunction, and adipose tissue inflammation.^[14]

The metabolic relevance to the QB framework is worth noting. If B7-H3 plays a regulatory role in adipocyte energy metabolism at steady state, and if metabolic dysregulation — which QB argues sits upstream of the cascade described in this paper — disrupts normal B7-H3 regulation in adipose tissue, it is plausible that upstream metabolic correction influences B7-H3 biology beyond the tumour compartment. This is stated as a mechanistically coherent inference rather than an established finding.

03The AR Suppression Axis

The PANORAMA study confirmed that B7-H3 expression is negatively regulated by AR signalling.^[1] This single finding reframes the relationship between standard-of-care treatment and immune evasion.

Under normal AR signalling, glandular structure is maintained, inflammatory signalling is constrained, and the cellular environment is associated with more ordered metabolic signalling. When AR signalling is suppressed by androgen deprivation therapy (ADT), this ordered state is disrupted. The cell enters an adaptive survival configuration: stressed, plasticity-prone,^[4] and primed toward inflammatory pathway activation, most notably NF-κB.

In this context, B7-H3 expression becomes advantageous. Its upregulation under ADT is not a side effect of treatment. It is an adaptation to the biological state that treatment creates. The tumour is not simply surviving ADT. It is responding to the new environment in a coherent, predictable way.

The reported synergy between ADT and B7-H3-targeted therapies follows directly from this logic. ADT induces the adaptive state. B7-H3 targeting blocks the escape mechanism that state produces. This is sequential constraint — mechanistically distinct from simple combination therapy.

A related inference extends this argument upstream of treatment entirely. Age-related androgen decline — driven by progressive Leydig cell insufficiency, rising sex hormone-binding globulin, and increasing aromatisation — produces a sustained, gradual reduction in AR tone that accumulates over decades. If the mechanism is the same as pharmacological ADT, the direction of effect on B7-H3 expression should be the same: reduced AR signalling progressively disinhibits B7-H3 expression in prostatic tissue. The process would be slower and lower in amplitude than acute ADT, but operating continuously from middle age onward.

This has two implications. First, the immunosuppressive TME drift that B7-H3 upregulation represents may begin well before any diagnosis or treatment — as part of the same ageing biology that the QB Primer identifies as erosion of host containment systems. Second, men with more pronounced age-related androgen decline, compounded by metabolic dysregulation increasing NF-κB tone, may enter the disease course with a B7-H3 expression environment already shifted toward immune evasion permissiveness. ADT then accelerates a process already under way rather than initiating it. This is a mechanistically plausible inference from the established AR/B7-H3 regulatory relationship; direct evidence in the context of physiological androgen decline rather than pharmacological suppression has not been established and would require dedicated investigation.

Standard-of-care treatment actively upregulates the very checkpoint it is now being combined with therapies to target. This is not a design flaw. It is the logic of adaptive biology — and it points toward upstream intervention.

04NF-κB and the Inflammatory Field

The transition from AR-on to AR-off is not purely intracellular. It propagates through the tumour microenvironment (TME), where NF-κB signalling acts as a central coordinator of the adaptive response. The downstream effects are multiple and self-reinforcing:

Increased IL-6 and pro-inflammatory cytokine output; increased IL-10 (anti-inflammatory; immunosuppressive in this context); expansion of myeloid-derived suppressor cells (MDSCs);^[5] polarisation of tumour-associated macrophages (TAMs) toward an M2 phenotype; and reinforcement of immune checkpoint expression, including B7-H3.

The result is a reinforcing loop. Therapeutic pressure activates NF-κB. NF-κB drives cytokine signalling. Cytokine signalling recruits and polarises myeloid cells. Polarised myeloid cells sustain the immunosuppressive environment in which B7-H3 expression is both produced and reinforced. B7-H3, in this loop, is not initiating immunosuppression. It is stabilising and expressing it at the interface of the tumour and immune system.

The biological logic here closely parallels the intratumoral microbiome findings documented elsewhere in the QB series. C. acnes drives IL-6, CXCL8, and PD-L1 expression through overlapping inflammatory routes; NF-κB and STAT3 are shared nodes. B7-H3 upregulation under ADT is a distinct but mechanistically convergent immunosuppressive axis — emerging from the same inflammatory field through different initiating signals.

05B7-H3 as a Microenvironmental Integrator

B7-H3's cross-compartment distribution — tumour epithelial cells, endothelial cells, TAMs, and MDSCs, all confirmed in the PANORAMA atlas^[1] — positions it not as a cell-autonomous checkpoint but as a surface integrator of a coordinated immunosuppressive state.^[3] Its functional consequences span the full immune-evasive programme: inhibition of CD8⁺ T-cell activation; promotion of regulatory T-cell (Treg) environments; support of angiogenesis and invasion — B7-H3 promotes VEGFA expression via NF-κB signalling,^[6] and B7-H3 expression on tumour vasculature distinguishes pathological from physiological angiogenesis;^[7] and reinforcement of metabolic adaptation.

This distribution is what makes B7-H3 a reliable target across disease stages. It is not clonally variable because it does not originate in a clonal process. It is produced by the microenvironmental state — and the microenvironmental state is produced by the same pressures that drive disease progression. As the disease becomes more advanced, the pressures that sustain the state intensify. B7-H3 expression follows.

B7-H3 is not simply a checkpoint. It is a surface expression of a coordinated microenvironmental state — and that state is one that current treatment actively promotes.

06Therapeutic Pressure and Compensatory Expression

The ADT and B7-H3 relationship illustrates a broader principle in prostate cancer biology: effective treatment often activates the very adaptive mechanisms that subsequent treatment must address. This is not a failure of the original treatment. It is the nature of adaptive biology under pressure.

What the PANORAMA data make clear is that this adaptation is predictable.^[1] B7-H3 upregulation under ADT is not a random resistance event. It is a mechanistically coherent response to a defined biological state change. Predictable adaptations are, in principle, interceptable — either by targeting the adaptation directly, as current B7-H3 programmes do, or by modifying the upstream conditions that make the adaptation necessary.

Both approaches are valid. They are not alternatives. A tumour in which B7-H3 expression has already established an immunosuppressive TME requires direct targeting. A treatment strategy that seeks to prevent or delay that establishment requires upstream intervention. The question of which is more relevant depends on where in the disease course the patient is — and what the biological state of the TME currently is.

07Upstream Modulation: Quieting the Signal

If B7-H3 reflects an adaptive state, then understanding the cascade that produces it suggests where upstream intervention is possible. The argument is not that upstream modulation replaces B7-H3 targeting. It is that addressing the conditions that drive B7-H3 expression may reduce the rate at which the immunosuppressive state establishes — and may alter the TME in ways that complement direct targeting.

Three upstream mechanisms are relevant within the Quiet Biology framework:

Mitochondrial and NF-κB constraint — doxycycline. Doxycycline suppresses NF-κB activation via inhibition of the p38 MAPK and NF-κB signalling pathways^[8] — the mechanism documented in full in the Doxycycline paper, where IκB phosphorylation blockade prevents nuclear translocation. This directly dampens the inflammatory priming that drives myeloid polarisation and checkpoint upregulation. The expected effect is a reduction in the induction pressure for B7-H3 expression under conditions of AR suppression. Whether doxycycline modulates B7-H3 expression specifically has not been established in the literature and would require experimental verification.

Growth and signalling constraint — mTOR modulation. mTOR inhibition limits translational capacity and attenuates cytokine amplification within the TME, reducing the reinforcing loop between myeloid cells and checkpoint expression. Cyclic rapamycin, as described in the Chronic Activation vs Oscillation paper, does not permanently suppress mTOR but restores the oscillatory dynamic that chronic metabolic activation has flattened. The expected effect is attenuation of TME reinforcement rather than elimination of the B7-H3 signal.

p53 stabilisation. p53 and NF-κB are mutual negative regulators: constitutive NF-κB activation reduces p53 tumour suppressor activity, while p53 suppresses inflammatory cytokine programmes.^[9] p53 stabilisation under conditions of reduced AKT tone — via the MDM2 axis described in the MDM2 Convergence paper — may therefore reduce the probability that adaptive immune evasion programmes activate under the stress of androgen deprivation. Additionally, p53-regulated microRNAs including miR-34a^[10] may exert direct control over B7-H3 expression levels.^[11] These mechanisms remain at the level of plausible inference in the B7-H3 context; direct experimental evidence is not yet available.

In each case, the intervention is aimed not at B7-H3 itself but at the signalling environment in which its expression is induced. The expected effects are reductions in induction probability rather than elimination of expression. This is the upstream layer; direct B7-H3 targeting addresses the interface layer. Both are necessary in advanced disease.

08Reframing B7-H3 in Therapeutic Strategy

The current clinical approach treats B7-H3 as a targetable antigen. This is productive and the evidence supports it.^[3] The PANORAMA study's bispecific pairing framework — combining B7-H3 with TROP-2, NECTIN1, KLK2, or NECTIN4 — represents a multi-constraint logic that is structurally consistent with how the Quiet Biology framework approaches the metabolic layer: no single point of pressure, but overlapping constraints that close off escape routes.

The complementary argument presented here is about the layer above: targeting B7-H3 treats the manifestation. Modulating the upstream state alters the conditions of its emergence.

Targeting B7-H3 treats the manifestation. Modulating the upstream state alters the conditions of its emergence.

These are not competing strategies. They operate at different levels of the same cascade. The QB framework's contribution is to identify where in that cascade upstream intervention is mechanistically plausible — and to argue that preventing the stabilisation of the immunosuppressive state is a legitimate goal, distinct from and complementary to targeting its surface expression.

09Conceptual Model

The following cascade integrates the mechanistic argument. The upper portion describes the established biology. The lower portion describes the Quiet Biology intervention layer and its points of entry.

ADT / therapeutic stress → ↓ AR signalling → ↑ NF-κB activation → inflammatory cytokine output → ↑ MDSCs + TAM M2 polarisation → ↑ B7-H3 expression (tumour + stromal + endothelial compartments) → T-cell suppression → immune evasion → tumour persistence.

QB upstream intervention acts at three points in this cascade: doxycycline^[8] → NF-κB constraint (p38 MAPK and IκB pathways) → reduced inflammatory priming; rapamycin → mTOR suppression → attenuated cytokine amplification and TME reinforcement; p53 stabilisation (via reduced AKT / MDM2 tone)^[9] → NF-κB antagonism + miR-34a^[10] + direct B7-H3 microRNA regulation^[11] → reduced checkpoint induction pressure.

The aim is not to eliminate B7-H3 expression but to reduce the probability that the full immunosuppressive cascade stabilises — particularly in the earlier phases of treatment, before the state has become self-sustaining.

10Implications and Open Questions

The framing developed in this paper generates a testable hypothesis: if inflammatory signalling, metabolic stress, and p53 dysregulation are constrained, does B7-H3 fail to upregulate under androgen deprivation?

This question is largely unexplored experimentally. It represents a junction between immunotherapy, metabolic therapy, and tumour ecology that no single research programme is currently positioned to address. It is the kind of question that the QB framework is designed to ask.

A secondary implication concerns B7-H3 as a dynamic biomarker. If its expression tracks with adaptive state rather than fixed tumour biology, serial B7-H3 measurement may signal when the TME has shifted toward immune evasion — not simply confirming tumour presence but indicating that the adaptive state has established. This would distinguish patients in whom direct targeting is the immediate priority from those in whom upstream intervention may still be effective.

A third implication concerns the doxycycline question specifically. The NF-κB mechanism is established. Whether doxycycline modulates B7-H3 expression is a tractable experimental question — potentially addressable in cell line or PDX models — and one that would have direct relevance to the QB protocol rationale.

Conclusion

B7-H3 is not merely a convenient therapeutic target. It is a signal of a specific biological state: AR-suppressed, inflammation-driven, microenvironmentally reinforced, and immune-evasive. Its consistent expression across the prostate cancer spectrum reflects the consistency of the adaptive biology that produces it, not a static molecular feature of the tumour.

Understanding B7-H3 as an adaptive surface phenotype does not diminish the value of targeting it directly. It extends the therapeutic logic upstream — to the conditions that make targeting necessary. In the Quiet Biology framework, the goal is not to eradicate cancer directly but to prevent the stabilisation of the states that allow it to persist. B7-H3 expression is one surface manifestation of such a state. The state itself is addressable.

The components of this argument are individually present in the literature. Researchers working in immunology, androgen receptor biology, and NF-κB signalling each hold a part of it. What is missing is not the evidence but the vertical assembly — the recognition that these components belong to the same cascade, that the cascade has a top, and that the top is accessible.

The tumour has not invented new biology. It has assembled existing biology into a stable state. Stability, unlike mutation, can potentially be interrupted.

Series Cross-References

Doxycycline (Paper 15) — NF-κB inhibition mechanism: p38 MAPK and IκB phosphorylation blockade, nuclear translocation, p53 stabilisation via MDM2 relief.

Tumour Ecology and Evolutionary Stability — immunosuppressive TME architecture; C. acnes / IL-6 / PD-L1 / STAT3 convergence.

MDM2 as Convergence Point (Paper 4) — AKT → MDM2 → p53 suppression axis; metabolic upstream of NF-κB tone.

Sirtuins, NAD⁺, and the QB Framework (Paper 7) — SIRT1 deacetylation of NF-κB p65; acetylation control layer operating alongside phosphorylation.

Chronic Activation vs Oscillation (Paper 2) — mTOR oscillation, rapamycin cycling rationale, signal rhythm restoration.

References

1. 01JHU-PANORAMA Consortium. B7-H3 as a pan-stage therapeutic target in prostate cancer: single-cell transcriptomic atlas and proteomic validation. Clinical Cancer Research. 2026. Single-cell atlas of 213 patients confirming B7-H3 expression across hormone-sensitive, CRPC, neuroendocrine, and double-negative subtypes, with low inter- and intra-tumoral variability, multi-compartment distribution, and demonstrated ADT/B7-H3 therapeutic synergy.
2. 02Picarda E, Ohaegbulam KC, Zang X. Molecular pathways: Targeting B7-H3 (CD276) for human cancer immunotherapy. Clinical Cancer Research. 2016;22(14):3425–3431. doi:10.1158/1078-0432.CCR-15-2428.
3. 03Kontos F, Michelakos T, Kurokawa T, et al. B7-H3: An attractive target for antibody-based immunotherapy. Clinical Cancer Research. 2021;27(5):1227–1235. doi:10.1158/1078-0432.CCR-20-2584.
4. 04Nyquist MD, Corella A, Coleman I, et al. Combined TP53 and RB1 loss promotes prostate cancer resistance to a spectrum of therapeutics and confers vulnerability to replication stress. Cell Reports. 2020;31(8):107669. doi:10.1016/j.celrep.2020.107669.
5. 05Gabrilovich DI. Myeloid-derived suppressor cells. Cancer Immunology Research. 2017;5(1):3–8. doi:10.1158/2326-6066.CIR-16-0297.
6. 06Wang R, Ma Y, Zhan S, et al. B7-H3 promotes colorectal cancer angiogenesis through activating the NF-κB pathway to induce VEGFA expression. Cell Death & Disease. 2020;11:55. doi:10.1038/s41419-020-2252-3.
7. 07Seaman S, Stevens J, Yang MY, et al. Genes that distinguish physiological and pathological angiogenesis. Cancer Cell. 2007;11(6):539–554. doi:10.1016/j.ccr.2007.04.017.
8. 08Santa-Cecília FV, Socias B, Ouidja MO, et al. Doxycycline suppresses microglial activation by inhibiting the p38 MAPK and NF-κB signalling pathways. Neurotoxicity Research. 2016;29(4):447–459. doi:10.1007/s12640-015-9592-2.
9. 09Gudkov AV, Gurova KV, Komarova EA. Inflammation and p53: a tale of two stresses. Genes & Cancer. 2011;2(4):503–516. doi:10.1177/1947601911409747.
10. 10Raver-Shapira N, Marciano E, Meiri E, et al. Transcriptional activation of miR-34a contributes to p53-mediated apoptosis. Molecular Cell. 2007;26(5):731–743. doi:10.1016/j.molcel.2007.05.010.
11. 11Xu H, Cheung IY, Guo HF, Cheung NK. MicroRNA miR-29 modulates expression of immunoinhibitory molecule B7-H3: potential implications for immune-based therapy of human solid tumours. Cancer Research. 2009;69(15):6275–6281. doi:10.1158/0008-5472.CAN-09-0542.
12. 12Roth TJ, Sheinin Y, Lohse CM, et al. B7-H3 ligand expression by prostate cancer: a novel marker of prognosis and potential target for therapy. Cancer Research. 2007;67(16):7893–7900. doi:10.1158/0008-5472.CAN-07-1068.
13. 13Inamura K, Amori G, Yuasa T, et al. Tumor B7-H3 expression in diagnostic biopsy specimens and survival in patients with metastatic prostate cancer. Prostate Cancer and Prostatic Diseases. 2021;24(3):767–774. doi:10.1038/s41391-021-00331-6.
14. 14Picarda E, Galbo PM Jr, Zong H, et al. The immune checkpoint B7-H3 (CD276) regulates adipocyte progenitor metabolism and obesity development. Science Advances. 2022;8(17):eabm7012. doi:10.1126/sciadv.abm7012.
15. 15Zang X, Thompson RH, Al-Ahmadie HA, et al. B7-H3 and B7x are highly expressed in human prostate cancer and associated with disease spread and poor outcome. Proceedings of the National Academy of Sciences. 2007;104(49):19458–19463. doi:10.1073/pnas.0709802104.

Finley Proudfoot | Quiet Biology Framework | April 2026