ITPP

ITPP is myo-inositol with three pyrophosphate bridges, made by Jean-Marie Lehn's group in Strasbourg. Its purpose is narrow: to lower the oxygen affinity of haemoglobin so that red blood cells unload more oxygen where tissue is hypoxic. Inositol hexaphosphate does the same thing but cannot cross the red cell membrane; ITPP can, which is what made it usable as a drug [1].

Most of the work on it is cancer research, on the idea that re-oxygenating a tumour makes it less aggressive and more vulnerable to treatment. In rodents it reversed tumour hypoxia, lowered HIF-1α and VEGF, normalised tumour vessels and roughly doubled survival when it preceded FOLFOX chemotherapy in colorectal liver metastasis [5]. It restricted pancreatic tumour growth and metastasis and improved the response to gemcitabine [6], and raised tumour oxygen tension in six different rodent models [7]. A 2022 review collects this work and argues the same mechanism should apply to ischaemic cardiovascular disease [8]. It also inhibited angiogenesis in chick embryo membranes and slowed implanted glioma [9].

The results are not uniform. In the six-model oxygenation study, ITPP on its own did not slow tumour growth at all, and combining it with radiotherapy helped in only one of the two models tested [7]. In a mouse glioblastoma model, ITPP was ineffective alone and appeared to reduce the benefit of radiotherapy [10]. Other groups found the opposite, with ITPP given before irradiation increasing radiation-induced DNA damage and slowing growth [11].

The reason anyone outside oncology knows about it is a 2009 study in mice. A single injection of 0.5–3 g/kg raised the oxygen tension at which haemoglobin is half saturated by up to 31%, and raised maximal exercise capacity by 57% in normal mice and 63% in mice with severe heart failure. Given in drinking water it still raised exercise capacity by 34% [3]. Later rodent work fits the same picture: eight injections over four weeks after a myocardial infarction halted left ventricular dilation and preserved ejection fraction in rats [12], a single dose restored right ventricular oxygen tension in rats with pulmonary hypertension [13], and five weeks of treatment relieved right ventricular hypoxia and reduced mortality in the same model, though it did nothing to the pulmonary vascular disease itself [14].

Haemoglobin's affinity for oxygen is tuned inside the red blood cell by 2,3-diphosphoglycerate, which binds in a pocket between the two beta chains and stabilises the low-affinity state. ITPP is a synthetic anion that acts at the same site, and it enters the red cell through the band 3 anion transporter: block band 3 with DIDS or NAP-taurine and both the uptake and the shift in the oxygen dissociation curve disappear [1]. Because band 3 is essentially confined to red cells, the compound is delivered where it is needed and largely nowhere else [1]. Phosphorus NMR of 2,3-DPG in intact erythrocytes shows ITPP also lowers intracellular pH, which contributes to oxygen release [15].

The consequence is a right shift of the oxygen–haemoglobin dissociation curve: the same arterial saturation delivers more oxygen to tissue with a low local pO2. In mice the p50 rose by up to 31% [3], and in rodent tumours oxygen tension rose within about two hours of a single injection [7]. How long that lasts depends on the schedule: oxygen-sensing implants in mouse melanoma and mammary tumours showed either a transient or a sustained rise of around 10 mmHg, depending on how many doses were given and how far apart [16].

Over days, a second and slower effect appears. Relieving hypoxia lowers HIF-1α and the VEGF and Lox signalling it drives, which remodels the chaotic tumour vasculature into something closer to normal and improves drug delivery [6]. This vascular normalisation outlasts the treatment by weeks [5], which is the basis for the clinical strategy of giving ITPP first and chemotherapy afterwards [2]. One caution about the simple story: in the six-model oxygenation study, ITPP raised tumour pO2 without changing tumour perfusion, and the authors attribute the early effect to reduced oxygen consumption rather than increased supply [7].

The only human safety data is the phase Ib trial. Twenty-eight patients with advanced abdominal cancers received nine 8-hour infusions over three weeks across eight dose levels. The maximum tolerated dose was 12,390 mg/m² per infusion; dose escalation stopped at 14,500 mg/m² after two patients had significant hypercalcaemia, one of them grade IV. Thirty-two adverse events were judged at least possibly ITPP-related, and they were almost all electrolyte disturbances: hypercalcaemia in 68% of patients, hypomagnesaemia in 18%, hypophosphataemia in 14%. All were asymptomatic and needed little or no treatment. Three older men with existing hypertension had a small rise in blood pressure that resolved after the infusion [2].

The hypercalcaemia is worth understanding, because it is an artefact of how the drug must be given. ITPP is a strong calcium chelator, so calcium chloride was infused with it to prevent hypocalcaemia; the excess calcium is what showed up in the blood [2]. Anyone taking an unregulated product without that adjustment has the opposite problem to worry about.

Nothing is known about repeated or long-term exposure in humans, about what it does in healthy people, or about the consequences of chronically shifting the oxygen dissociation curve. Animal studies report no evident toxicity [5][6], but none of them were designed as toxicology studies.

No human interaction studies exist. Its calcium chelation is the clearest theoretical concern: in the clinical trial it was strong enough that calcium had to be co-infused, and magnesium and phosphate also fell in some patients [2].

In rodents, the point of ITPP is that it changes how other treatments work. It improved the effect of gemcitabine in pancreatic cancer [6] and of FOLFOX in colorectal liver metastasis [5], and enhanced radiotherapy in some models [11] while reducing its benefit in a glioblastoma model [10]. That last result is the reason not to assume the direction of any interaction.