Changing acidic niches

Mechanistic thesis

Tumour acidity, cell state and treatment response

Four intervention classes, different mechanisms, and a testable question: can changing acidic niches alter therapeutic response?

Tumour acidity is often discussed as a number to correct. A more useful starting point is to ask what maintains the acidic environment, which cells experience it, and whether changing it alters a treatment-relevant behaviour.

The thesis is that interventions affecting perfusion, extracellular buffering, acid–base regulation or cellular state may modify acidic tumour niches and, in selected settings, change treatment response. The evidence supports several parts of this proposition. It does not yet establish a single causal sequence connecting all four intervention classes, or a general benefit from combining them.

Extracellular pH, intracellular pH and oxygen tension are different measurements. A tumour can contain regions with different combinations of all three. Any account that substitutes improved oxygenation for measured alkalinisation, or a whole-tumour average for the conditions surrounding individual cells, risks missing the biology it is trying to explain. [2] [3] [12]

Several processes maintain an acidic niche

Metabolic acid generation, membrane transport, extracellular buffering and removal through the circulation jointly influence tumour pH. Matrix-associated solid stress can compress blood vessels and impair perfusion; this provides one plausible route by which stromal architecture affects the removal of metabolic products. Hypoxia can also change metabolism and acid–base regulation. These processes interact rather than forming a universal four-step chain. [1] [3] [5]

CAIX is one component of that system. Its extracellular catalytic domain accelerates the reversible conversion of carbon dioxide and water into bicarbonate and a proton. Bicarbonate uptake is performed by transporters, not by CAIX itself. Coupling enzyme activity to transport can support intracellular pH regulation while contributing to extracellular acidity. This is a more precise description than saying that one enzyme builds the entire gradient. [5]

Four intervention classes that may influence tumour acidity

These classes describe different points of intervention. They are neither an exhaustive list nor an established ranking of potency or durability. For the fourth class, particularly 4-PBA, a change in tumour acidity remains a hypothesis requiring direct measurement.

Class 1 · stromal and perfusion modification

Losartan: a possible influence on acid removal

In desmoplastic mouse tumours, losartan reduced collagen and hyaluronan, relieved solid stress and improved vascular perfusion. In an E0771 breast-tumour imaging study, extravascular oxygen tension was 61% higher than controls on day five, without a corresponding increase in intravascular oxygenation. [1] [2]

Interpretation: improved perfusion could affect acid clearance and hypoxia-dependent metabolism. The cited oxygenation findings do not demonstrate a rise in extracellular tumour pH. Losartan addresses a possible contributor to acidity, rather than an established single cause.

Class 2 · systemic buffering

Bicarbonate: modifying extracellular acid–base conditions

Bicarbonate has raised tumour-region extracellular pH in mouse experiments. Martin and colleagues modelled an approximately 0.04-unit increase in human tumour pH under specified assumptions. That figure is a prediction, not a measured human treatment response. [3]

Interpretation: buffering can alter the environment experienced by cells. Its biological impact depends on the achieved local pH change and the response of those cells; a predicted small mean shift cannot establish either clinical benefit or a particular signalling gain.

Class 3 · inhibition of acid–base regulatory activity

CAIX inhibition: interfering with an adaptive mechanism

Faes and colleagues found that CAIX knockdown and treatment with acetazolamide enhanced rapamycin activity in mouse tumour models. The genetic result supports CAIX involvement; acetazolamide inhibits multiple carbonic-anhydrase isoforms, so its effects cannot be assigned exclusively to CAIX. [5]

Interpretation: disrupting acid–base regulation may weaken hypoxic tumour compartments. The combination results do not establish that extracellular alkalinisation and restoration of mTORC1 activity are the sole mechanism.

Class 4 · chromatin and cellular-state modulation

HDAC inhibitors and 4-PBA: a link that needs to be resolved

Ledaki and colleagues linked CA9 expression to chromatin accessibility and showed that sodium butyrate and SAHA reduced CAIX and selected stem-cell markers. Separate studies show that phenylbutyrate can alter differentiation, stemness-associated expression and EMT. [6] [7] [8] [9]

Interpretation: cellular-state modulation is a plausible way to alter adaptation to an acidic niche. Direct repression of CA9/CAIX by native 4-PBA, and a resulting change in tumour pH, were not established by the studies identified in this review.

What the 4-PBA evidence supports

The strongest defensible claim is that phenylbutyrate can change cancer-cell phenotype in specific experimental settings. Rubio and colleagues treated bladder-cancer/macrophage co-cultures with 1 mM PBA for 24 hours and observed lower SOX2, NANOG and MYC expression. Sorted hybrid cells also showed reductions in miR-21 and miR-302, while hybrid formation was not reduced. These findings support a change in a stemness-associated programme; they do not demonstrate elimination of tumour-initiating cells. [7]

In medulloblastoma cells, phenylbutyrate increased differentiation markers and histone acetylation, with substantial differences between cell lines. Sodium phenylbutyrate reduced EMT-associated features in oral-cancer models. More recently, 4-PBA reduced sphere formation and stemness-associated proteins in NHE7-overexpressing endometrial-cancer cells, where it was used to suppress endoplasmic-reticulum stress. None of these findings establishes CAIX repression as the mechanism. [8] [9] [10]

The direction is also context-dependent. In hepatocellular-carcinoma models, Chen and colleagues found that 4-PBA increased stem-like features and tumour-initiating capacity through a pathway involving PPAR-α and Wnt/β-catenin signalling. It would therefore be inaccurate to describe 4-PBA as a general suppressor of the stem-cell programme. [11]

The fourth class remains a research proposition: 4-PBA can modulate cellular state, but whether it suppresses CAIX, changes extracellular acidity and thereby improves a particular treatment response must be demonstrated in the same model.

mTORC1 provides a concrete example, with limits

Balgi and colleagues showed rapid, reversible mTORC1 suppression after acidification in MCF-7 cells and fibroblasts, with a requirement for TSC2 in the fibroblast experiments. Faes and colleagues subsequently found that acidic conditions reduced rapamycin's antiproliferative effect in several cancer-cell models. In mice, bicarbonate increased phospho-S6 staining and improved the effect of rapamycin. This provides a preclinical rationale for modifying the environment in which a pathway inhibitor acts. [12] [4]

The appropriate interpretation is reduced pathway activity and altered dependence on that pathway, rather than physical absence of the drug target. The bicarbonate–rapamycin rationale was already advanced by Faes; it is not a new therapeutic principle introduced by this thesis. [4]

Nor does acidity invariably switch mTOR signalling off. Peppicelli and colleagues reported active mTOR signalling and sensitivity to everolimus in melanoma cells exposed or adapted to acidic conditions. Cell identity, adaptation and experimental conditions therefore matter. [14]

In CD8 T cells, acidity also perturbs IL-2 responsiveness, mTORC1 and c-Myc. However, the 2024 study found that the suppression of mTORC1 was not dependent on the same TSC2 mechanism, and enforcing mTORC1 activity did not restore proliferation. Tumour-cell and immune-cell outcomes must be measured separately; one shared signalling marker does not imply one shared mechanism or a net immune benefit. [13]

Small pH changes: a hypothesis about location, not a measured amplification

Martin's modelling study predicted an approximately 0.07-unit mouse tumour-pH increase, consistent with earlier imaging measurements. Faes separately reported increases in phospho-S6 immunostaining of 57.8% in HT29 and 52.9% in MC-38 tumours after bicarbonate. The pH and signalling values came from different experimental systems and were not paired measurements in the same tumours. They cannot be combined into a quantitative amplification factor or scaled into an expected human response. [3] [4]

A more useful hypothesis remains: the importance of a pH change may depend on which cells experience it and where they lie on a pH–response curve. A modest average shift could conceal a consequential change in some regions, or little useful change anywhere. Spatially resolved measurements are needed to distinguish those possibilities. This is a proposed explanation to test, not an effect size already demonstrated by the cited studies.

Persistence must be measured after treatment withdrawal

The four classes do not establish a hierarchy in which deeper intervention automatically means longer-lasting benefit. Ledaki's six-month observations concerned the behaviour of sorted CAIX-positive and CAIX-negative populations. They did not demonstrate that a drug had produced a six-month reprogramming effect. Reduced CAIX in colonies after a ten-day recovery from SAHA is a more relevant persistence observation, but it cannot be transferred to 4-PBA. [6]

Similarly, rapid signalling recovery after a laboratory pH switch does not determine how long an oral buffer's effect lasts in a tumour. Durability requires its own measurements of drug exposure, pH, cellular phenotype and function after withdrawal. [12]

How to test the thesis

The contribution of this framework is to compare mechanisms using connected measurements. A useful experiment would establish the following sequence within the same model, rather than infer it by joining results from different studies:

1. Perturb the proposed mechanismConfirm effects on perfusion, buffering, enzyme activity or cellular state.
2. Measure the local consequenceMap extracellular pH and relevant intracellular responses over time.
3. Test the treatment interactionMeasure pathway inhibition, tumour behaviour and immune function.

Proposed validation sequence. It is not a demonstrated common mechanism for all four classes.

• For losartan: pair perfusion and oxygen measurements with extracellular pH in the same tumour regions. Improved oxygen delivery alone does not answer the acidity question.
• For bicarbonate: pair measured pH changes with signalling and treatment response. Distinguish a change in baseline pathway activity from drug-induced pathway inhibition.
• For CAIX inhibition: compare selective pharmacological inhibition with genetic perturbation, measuring extracellular and intracellular pH. This would help separate CAIX-dependent effects from those of broader carbonic-anhydrase inhibition.
• For 4-PBA: measure CA9 transcripts, surface CAIX, pH and stem-like functions under controlled oxygen and pH conditions. If CAIX decreases, test whether restoring its expression reverses the relevant phenotype. Include viability controls and measurements after washout.

Combination studies should include each intervention alone and the combination, with predefined interaction analyses. Shared mechanisms or a larger combination effect do not by themselves establish synergy. Tumour outcomes and immune-cell function also need separate assessment.

The resulting thesis is conditional and experimentally actionable: changing the processes that maintain an acidic niche may change how tumour and immune cells respond to treatment. Perfusion modification, buffering and CAIX inhibition provide distinct routes to investigate. Chromatin and cellular-state modulation add a fourth candidate route, with 4-PBA's connection to CAIX and tumour acidity still to be established. This is a framework for preclinical testing, not a validated four-drug treatment regimen.

References

1. Chauhan VP et al. Angiotensin inhibition enhances drug delivery and potentiates chemotherapy by decompressing tumour blood vessels. Nature Communications (2013). doi:10.1038/ncomms3516.
2. Multiphoton Phosphorescence Quenching Microscopy Reveals Kinetics of Tumor Oxygenation during Antiangiogenesis and Angiotensin Signaling Inhibition. Clinical Cancer Research (2022). doi:10.1158/1078-0432.CCR-22-0486.
3. Martin NK et al. Predicting the safety and efficacy of buffer therapy to raise tumour pH_e: an integrative modelling study. British Journal of Cancer (2012). doi:10.1038/bjc.2012.58.
4. Faes S et al. Acidic tumor microenvironment abrogates the efficacy of mTORC1 inhibitors. Molecular Cancer (2016). doi:10.1186/s12943-016-0562-y.
5. Faes S et al. Targeting carbonic anhydrase IX improves the anti-cancer efficacy of mTOR inhibitors. Oncotarget (2016). doi:10.18632/oncotarget.9134.
6. Ledaki I et al. Carbonic anhydrase IX induction defines a heterogeneous cancer cell response to hypoxia and mediates stem cell-like properties and sensitivity to HDAC inhibition. Oncotarget (2015). doi:10.18632/oncotarget.4989.
7. Rubio C et al. Toward Tumor Fight and Tumor Microenvironment Remodeling: PBA Induces Cell Cycle Arrest and Reduces Tumor Hybrid Cells' Pluripotency in Bladder Cancer. Cancers (2022). doi:10.3390/cancers14020287.
8. Li XN et al. Phenylbutyrate and phenylacetate induce differentiation and inhibit proliferation of human medulloblastoma cells. Clinical Cancer Research (2004). doi:10.1158/1078-0432.CCR-0747-3.
9. Qian K et al. Sodium Phenylbutyrate Inhibits Tumor Growth and the Epithelial-Mesenchymal Transition of Oral Squamous Cell Carcinoma In Vitro and In Vivo. Cancer Biotherapy and Radiopharmaceuticals (2018). doi:10.1089/cbr.2017.2418.
10. Yang S et al. Hypoxic glycolysis-driven histone lactylation activates NHE7 to promote endometrial cancer progression via COX6C-mediated endoplasmic reticulum stress. Apoptosis (2026). doi:10.1007/s10495-026-02262-w.
11. Chen SZ et al. 4-phenylbutyric acid promotes hepatocellular carcinoma via initiating cancer stem cells through activation of PPAR-α. Clinical and Translational Medicine (2021). doi:10.1002/ctm2.379.
12. Balgi AD et al. Regulation of mTORC1 signaling by pH. PLoS ONE (2011). doi:10.1371/journal.pone.0021549.
13. Vuillefroy de Silly R et al. Acidity suppresses CD8+ T-cell function by perturbing IL-2, mTORC1, and c-Myc signaling. The EMBO Journal (2024). doi:10.1038/s44318-024-00235-w.
14. Peppicelli S et al. Everolimus selectively targets vemurafenib resistant BRAF^V600E melanoma cells adapted to low pH. (2017). PubMed:28826720.