Tumour metabolism

Tumour metabolism & the microenvironment

Targeting the Tumour’s Blood Supply

How angiogenesis inhibitors and embolisation attack a tumour’s circulation—and why cutting off more blood is not always the better treatment.

The previous article, The Tumour’s Supply Problem, described how poor blood flow can leave a tumour short of oxygen, slow the removal of metabolites and obstruct treatment delivery. That creates an apparent contradiction: if inadequate circulation contributes to the problem, why would deliberately reducing blood supply help?

Because blood vessels support tumour growth as well as treatment delivery. Cancer cells need access to circulating resources, but drugs also need access to cancer cells. Changing the circulation can affect both sides of that relationship.

Medicine already exploits this dependence in two main ways: interfering with the signals that support tumour vessels, and selectively blocking arteries that feed a tumour. A third aim—improving the function of the vessels that remain—helps explain why some combinations work better than a simple starvation model would predict.

A tumour’s blood supply is a therapeutic target. The objective is better tumour control with acceptable harm to the patient; the lowest possible blood flow is not a universal measure of success.

How a tumour recruits a blood supply

Angiogenesis is the growth of new blood vessels from existing ones. It normally supports processes such as wound repair. Tumours can stimulate it by releasing signals that activate endothelial cells, the cells lining blood vessels. One major signal is vascular endothelial growth factor, or VEGF. [1]

Low oxygen can increase VEGF expression, creating a response to inadequate supply. Experimental work established this hypoxia–VEGF connection decades ago. It helps explain how a growing tumour can recruit vessels, but it does not mean that every new vessel carries blood efficiently. [2]

The signal and the circulation

Building vessels is not the same as delivering blood

1. A growth signal

Hypoxia and other tumour signals can increase the drive for vessel growth.

2. A vessel response

VEGF activates receptors on endothelial cells, supporting vessel growth and survival.

3. A functional outcome

The resulting network may still deliver blood unevenly. Vessel abundance and perfusion are different measurements.

A simplified mechanism. Antiangiogenic drugs interrupt signalling; their effects on actual blood flow must be measured separately. [1] [2] [6]

How angiogenesis inhibitors interfere with supply

These drugs act at different points in the signalling pathway. They can suppress new vessel formation and affect the support of existing tumour vessels. They do not physically seal every vessel, and they are not interchangeable.

Three ways to interrupt VEGF signalling

Approach	How it works	Example and boundary
Bind the signal	An antibody binds VEGF before it activates its receptors.	Bevacizumab targets VEGF. Its benefit depends on the cancer and treatment regimen. [3]
Block the receptor	An antibody binds VEGFR2 and prevents activating ligands from binding.	Ramucirumab acts at the receptor rather than binding circulating VEGF itself. [4]
Inhibit receptor signalling	A small molecule inhibits the receptor’s tyrosine-kinase activity.	Axitinib inhibits VEGFR1, VEGFR2 and VEGFR3. Other kinase inhibitors have different target profiles. [5]

There is direct evidence that this can reduce human tumour blood supply. In a small rectal-cancer study, a single bevacizumab infusion reduced tumour perfusion, vascular volume and microvascular density. Interstitial fluid pressure also fell, while the proportion of vessels covered by supporting cells called pericytes increased. Several features of the circulation changed together; vessel loss was only one of them. [6]

The clinical benefit requires separate evidence. In a landmark randomised trial of 813 patients with previously untreated metastatic colorectal cancer, adding bevacizumab to irinotecan, fluorouracil and leucovorin increased median overall survival from 15.6 to 20.3 months (hazard ratio for death 0.66). This historical trial demonstrated benefit from adding VEGF blockade to that chemotherapy regimen. It did not show that complete vascular shutdown was achieved or necessary. [7]

Why fewer vessels can sometimes mean better delivery

Abnormal vessels are not simply normal pipes multiplied. Their organisation, leakiness and surrounding pressure can impair delivery. Removing some poorly functioning vessels while improving the remaining network can, under suitable conditions, make transport more effective.

This is vascular normalisation. In experimental tumours, VEGFR2 blockade reduced interstitial pressure and improved drug penetration. Separate brain-tumour experiments identified a temporary period of improved oxygenation during which radiation treatment was more effective. These results made timing part of the mechanism: the condition of the vessels during combination therapy mattered. [8] [9]

Human observations support the possibility without making it universal. In a glioblastoma study using cediranib with chemoradiation, patients whose tumour perfusion increased also showed improved oxygenation and longer survival. That was an association within a treated group, not proof that increasing perfusion alone caused the survival difference. [10]

Different vascular effects

Growth inhibition, normalisation and shutdown

Limit vessel support

Suppress angiogenic signals and constrain expansion of the vascular network.

Improve vessel function

Under suitable conditions, improve delivery through the remaining network.

Interrupt blood flow

Deprive a targeted region of supply, as intended with arterial embolisation.

These are different treatment effects, not three guaranteed stages. The same antiangiogenic drug can have different effects across tumours and over time. [6] [8] [11]

Normalisation therefore does not contradict targeting the blood supply. It changes the aim from simply removing vessels to changing how the circulation supports the tumour and its treatment.

Directly blocking the arteries that feed a tumour

Embolisation directly reduces blood supply. An interventional radiologist guides a catheter into selected arteries and delivers material that obstructs flow. The procedure depends on suitable anatomy and organ function.

Many hepatocellular carcinomas receive most of their blood from the hepatic artery; normal liver tissue receives most through the portal vein. This difference can allow selective treatment of tumour-feeding arteries, although healthy liver tissue can still be injured. [11]

• Transarterial embolisation (TAE) obstructs feeding arteries to produce ischaemia—insufficient blood flow—and tumour tissue death.
• Transarterial chemoembolisation (TACE) combines arterial obstruction with local chemotherapy, using both interrupted supply and drug exposure. [11]
• Radioembolisation delivers radioactive microspheres, commonly containing yttrium-90. Its defining therapeutic component is local radiation; it should not be described simply as another way of starving the tumour. [12]

Selection depends on preserving enough functioning liver and treating the correct territory. Local treatment does not automatically control distant disease. [11]

Why combinations need clinical testing

Vascular targeting is often paired with treatment directed at cancer cells or the immune response. The combination has to succeed as a whole; an appealing mechanism cannot establish the contribution of each component.

In the phase 3 IMbrave150 trial, atezolizumab plus bevacizumab improved overall survival compared with sorafenib in patients with unresectable hepatocellular carcinoma who had not received systemic treatment. The initial analysis reported a hazard ratio for death of 0.58 (95% CI 0.42–0.79). This supports that immunotherapy–anti-VEGF regimen in the studied setting. Because the trial compared different regimens, it does not isolate vascular normalisation or reduced blood flow as the reason for benefit. [13]

A recent lesson: delayed progression is not necessarily longer survival

The phase 3 EMERALD-1 trial tested TACE with durvalumab and bevacizumab in embolisation-eligible, unresectable hepatocellular carcinoma. The 2025 publication reported median progression-free survival of 15.0 versus 8.2 months for the three-part regimen versus TACE with placebo (HR 0.77, 95% CI 0.61–0.98). [14]

The final overall-survival analysis, reported in a 2026 conference abstract, did not show an overall-survival improvement. The progression result remains a result, but it must not be presented as proof that patients lived longer. [15]

Combining a procedure that interrupts blood flow with a drug that blocks vessel signalling is biologically reasonable. The outcome still depends on the disease, schedule, accompanying treatment and effects on the patient.

Theobromine: an experimental antiangiogenic approach

Theobromine offers a connection between adenosine signalling and the tumour’s drive to recruit blood vessels. Experiments with the compound itself have reported reduced tumour-induced angiogenesis and VEGF production. The evidence is preclinical, but it provides a reason to investigate this pathway beyond the established VEGF-targeting drugs.

In a 1993 study, injected theobromine reduced new-vessel formation induced by lung-carcinoma cells in mouse skin. This demonstrated an effect on a tumour-induced angiogenic response. Gil et al., 1993.

A 1998 study using human ovarian-cancer cells found reduced angiogenic activity and lower VEGF production, while production of basic fibroblast growth factor (bFGF) and interleukin-8 (IL-8) was unchanged. A follow-up study found inhibition of angiogenic activity from ovarian-cancer cells and lymphocytes isolated from patients’ ascitic fluid. Pharmacological comparisons implicated A2 adenosine-receptor signalling. These were experimental studies using human-derived cells, not clinical treatment trials. Barcz et al., 1998; Barcz et al., 2000.

The proposed connection: interfering with adenosine signalling may reduce a stimulus for VEGF production and vessel formation. The older studies support investigating that mechanism; they do not establish selective A2A or A2B blockade, or direct VEGFR2 inhibition, by theobromine.

That last distinction matters when reading newer research. Several potent VEGFR2 inhibitors described as “theobromine derivatives” are chemically modified compounds. Their measured potency cannot be assigned to unmodified theobromine, cocoa or chocolate. Example: a semisynthetic theobromine-derived VEGFR2 inhibitor, 2023.

How tumours escape vascular targeting

Not every tumour relies on growing new vessels. Some cancer cells expand along vessels already present in the organ—a process called vessel co-option. Research combining human colorectal liver-metastasis samples with experimental models linked this growth pattern to resistance to antiangiogenic therapy. Blocking vessel construction is less effective when the tumour can use an existing network. [16]

There is also the problem of the cells that survive reduced supply. In mouse models, antiangiogenic treatment increased hypoxia and was associated with more invasive growth and, in some settings, metastasis. These findings identify a possible escape response, not a reason to conclude that antiangiogenic treatment generally accelerates cancer in patients. Human benefit and harm must be judged from the relevant clinical evidence. [17]

The distinction matters for the previous article’s argument. Moderately reduced delivery may leave viable cells in a stressful environment; sufficiently severe, sustained interruption can kill tissue. Neither outcome can be inferred from a lower vessel count alone.

The circulation belongs to the patient too

VEGF signalling also matters outside the tumour. Bevacizumab’s prescribing information identifies risks including hypertension, proteinuria, bleeding, arterial thromboembolic events, impaired wound healing and gastrointestinal perforation. These are part of why treatment selection, monitoring and coordination with surgery matter. [3]

The same principle applies to embolisation: blocking a tumour’s supply is useful only if the surrounding organ can tolerate the intervention. More vascular damage is not automatically a better anticancer effect.

What successful vascular targeting should demonstrate

The practical assessment has several parts:

• Which vascular dependence is being targeted? New vessel growth, signalling that supports existing vessels, or a selected arterial supply?
• What changed in the tissue? Blood flow, oxygenation, drug delivery, viable tumour or tissue death?
• Did the treatment improve meaningful outcomes? Tumour control, symptoms, progression-free survival and overall survival need to be distinguished.
• What did it cost the patient? Organ function and treatment toxicity belong in the assessment.

A tumour’s dependence on circulation is a real vulnerability. Angiogenesis inhibitors exploit its need for vascular support; embolisation can interrupt a selected blood supply directly. The most effective strategy depends on how the tumour is supplied, what treatment must still reach it, and how much normal tissue can safely be spared.

Evidence note: targeted literature review checked on 16 October 2026. Laboratory mechanisms, human physiological observations, randomised treatment trials and conference findings are distinguished in the text. Examples explain treatment principles rather than prescribe a regimen for an individual patient.

References

1. National Cancer Institute. Angiogenesis inhibitors.Patient information: basic mechanisms.
2. Shweiki D, et al. (1992). Vascular endothelial growth factor induced by hypoxia may mediate hypoxia-initiated angiogenesis.Nature. Experimental hypoxia–VEGF signalling.
3. Genentech. Avastin (bevacizumab): prescribing information.Drug mechanism and safety information.
4. Eli Lilly / DailyMed. Cyramza (ramucirumab): prescribing information.VEGFR2 antagonist; label updated June 2026.
5. Pfizer / DailyMed. Inlyta (axitinib): prescribing information.VEGF-receptor kinase inhibition.
6. Willett CG, et al. (2004). Direct evidence that the VEGF-specific antibody bevacizumab has antivascular effects in human rectal cancer.Nature Medicine. Small human physiological study.
7. Hurwitz H, et al. (2004). Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer.New England Journal of Medicine. Randomised phase 3 trial.
8. Tong RT, et al. (2004). Vascular normalization by vascular endothelial growth factor receptor 2 blockade induces a pressure gradient across the vasculature and improves drug penetration in tumors.Cancer Research. Preclinical vascular and transport experiments.
9. Winkler F, et al. (2004). Kinetics of vascular normalization by VEGFR2 blockade governs brain tumor response to radiation: role of oxygenation, angiopoietin-1, and matrix metalloproteinases.Cancer Cell. Preclinical timing and radiation-response experiments.
10. Batchelor TT, et al. (2013). Improved tumor oxygenation and survival in glioblastoma patients who show increased blood perfusion after cediranib and chemoradiation.PNAS. Human imaging and survival associations.
11. National Cancer Institute. Primary liver cancer treatment (PDQ), health professional version.Clinical treatment summary: TAE, TACE and patient selection.
12. National Cancer Institute. Radioembolization.Definition of internal radiation delivered through the arterial supply.
13. Finn RS, et al. (2020). Atezolizumab plus bevacizumab in unresectable hepatocellular carcinoma.New England Journal of Medicine. IMbrave150 randomised phase 3 trial.
14. Sangro B, et al. (2025). Durvalumab with or without bevacizumab with transarterial chemoembolisation in hepatocellular carcinoma (EMERALD-1).The Lancet. Randomised phase 3 trial; primary progression-free-survival report.
15. EMERALD-1 investigators (2026). 183O: Overall survival in EMERALD-1.Annals of Oncology 37, supplement 1, S72. Conference abstract reporting final overall-survival analysis.
16. Frentzas S, et al. (2016). Vessel co-option mediates resistance to anti-angiogenic therapy in liver metastases.Nature Medicine. Human tumour samples and experimental models.
17. Pàez-Ribes M, et al. (2009). Antiangiogenic therapy elicits malignant progression of tumors to increased local invasion and distant metastasis.Cancer Cell. Mouse models; not a general clinical outcome claim.