Mammalian cancer cells are efficiently killed by β particles emitted from radionuclides. The ability of ¹³¹I to home to thyroid tissue has been exploited for decades as a therapeutic strategy against thyroid cancer and Graves’ disease, and this strategy is still used in over 50% of thyroid cancer patients in the USA today.¹ Similarly, ¹³¹I-Bexxar and ⁹⁰Y-Zevalin are used to treat non-Hodgkin’s lymphoma based on antibodies against the CD20 cell surface antigen.^2,3 Moreover, ⁹⁰Y-radiolabeled somatostatin is also used to treat neuroendocrine tumors.⁴ [³²P] ATP emits electrons with an energy level intermediate between that of ⁹⁰Y and ¹³¹I; these electrons have a path length of up to 5 mm in tissues. Thus, each electron can penetrate thousands of cells. The resulting cross-fire results in a “bystander effect” that greatly amplifies the killing power of each ³²P atom in or near a tumor. In addition, ³²P possesses a much longer half-life than ⁹⁰Y or ¹³¹I: this is an advantage, since radioactivity levels in tumors do not diminish as rapidly from natural decay.

Anticancer therapeutics are often assessed by their ability to inhibit xenografted tumors in nude mice.^5,6 Successful drugs tested in this fashion have included Rituxan and other antibodies directed against cell surface molecules.⁷ Examples of drugs designed against broader targets include the anti-VEGF antibody Avastin, which inhibits the establishment of the tumor microenvironment, and cisplatin, which targets rapidly growing cells.⁸ Small molecules present many advantages as anticancer therapeutics, such as better tumor penetration and low immunogenicity. Previously, we reported that a single, low-dose intravenous injection of [³²P] ATP inhibited the growth of xenografted tumors in nude mice.⁹ In order to make this assay more clinically relevant, we ensured that our xenografts had become established and well-vascularized prior to dose administration; in this setting, a narrow window of dose-effectiveness was observed. The current report expands on this previous study by describing the biodistribution of [³²P] ATP in nude mice bearing HeLa-derived xenografted tumors as well as by exploring the effectiveness multi-dose regimens of [³²P] ATP.

The current report confirms that intravenous injection of low doses of [³²P]ATP rapidly slows xenografted tumor growth in nude mice for several weeks, although the mechanism responsible for this reproducible finding is not yet understood. Our biodistribution results establish that although levels of [³²P]ATP in tumors are initially low relative to normal organs, tumor levels decrease more slowly than in any other tissues. Unlike most potential anticancer drugs, which are based on antibodies or alien chemical compounds, ATP is relatively unique in that it is a completely natural compound, although the vast majority of ATP in humans is intracellular. The initial low level of [³²P]ATP that was seen in the xenografted tumor was expected, because the tumors were essentially white in color, demonstrating inferior levels of vascularization when compared with the red blood-rich organs, such as the liver, kidney, heart and others. This inhibition of xenografted tumor may result from direct cytotoxic ability of the radioisotope, significant injury to B-cell lymphocytes and natural killer cells or, possibly, the well-documented phenomenon of low-dose radiation hypersensitivity.¹⁰ Hematoxylin and eosin staining of sections of the xenografted tumors showed no immune cells present in tumors from mice that received the highest doses of [³²P]ATP, while the small, disrupted tumors examined from mice receiving optimal [³²P]ATP doses did contain infiltrating immune cells (data not shown).

In the 1920s, Otto Warburg showed that tumor tissues metabolize glucose at a rate approximately ten times greater than do normal tissues under aerobic conditions.^11,12 This finding became known as the Warburg effect and is centered on the mechanisms used by cancer cells to generate energy for the cell in the form of ATP, which was first discovered in 1929.¹¹ The gamma phosphate of each ATP molecule is readily removed and restored over 1,000 times per day in the normal metabolic functions of cells, and the quantity of the ATP turnover is so great that it is roughly equal to a human’s total body weight.¹³

Inorganic elemental ³²P has been used to treat essential thrombocythemia and polycythemia vera, and up to 15 mCi of inorganic ³²P-sodium phosphate have been administered to each patient with chronic leukemia.¹⁴ Levels of organic ³²P used here correspond, on a mouse to human weight basis, favorably to the levels of inorganic ³²P that historically have been used previously to treat humans. In the United States, several ongoing clinical trials utilize inorganic ³²P; moreover, nanotechnology approaches have been attempted to target this radioisotope against various cancer types. However, to our knowledge, organic ³²P (as in [³²P]ATP) has never been tested as a therapeutic agent in any disease.

Extracellular cold ATP occurs spontaneously at concentrations above 100 μM in xenografted mouse tumors, but it is undetectable in healthy tissues.¹⁵ Therefore, [³²P]ATP may constitute a naturally targeted anticancer therapeutic agent, perhaps due to cancer-related inflammation and the energy-demanding extracellular tumor microenvironment.¹⁶ Additional potential advantages of [³²P] ATP as an anticancer agent include its ready availability, low cost, ease of handling and transport. and relatively long half-life. Furthermore, there is the consensus that the smaller the anticancer molecule, the better the tumor penetration.¹⁷ In this context, [³²P]ATP has a molecular weight of only 508 Daltons, making it much smaller than most anticancer agents. Finally, its apparent tumor-homing ability and the apparent requirement of the tumor microenvironment for exogenous ATP make [³²P]ATP an attractive potential anticancer treatment strategy.