Effect of Ozone/Oxygen-Pneumoperitoneum on Tumour Growth and Metastatic Spread of the Rabbit VX2 Head and Neck Cancer Model

About 6% of all newly diagnosed malignancies worldwide are cancers of the head and neck. They account for nearly 5% of all cancer-related deaths. Surgery, radiation therapy and chemotherapy as well as combinations of these three are currently regarded as the standard treatment. While high cure rates can be achieved for the early stage disease, cure and survival rates for locoregionally advanced and distant metastatic disease are still disappointing. Thus, the development of new treatment strategies is essential, particularly for the treatment of cancers with metastatic involvement of the lymph nodes. The VX2 auricle carcinoma of New Zealand White rabbits (NZWR) is an accepted animal model for human head and neck cancers, since both are similar in growth leading to early regional lymph node metastases and subsequent distant metastatic spread. Ozone, 2002 recognized as a biomolecule since it was found to be endogenously produced by human neutrophils, exhibited protective effects in a rat sepsis model when insufflated intraperitoneally as a gas mixture with oxygen. This finding suggests that ozone might have immunemodulatory properties. In vitro, ozone was found to inhibit selectively the growth of human cancer cells. However, until now there are no randomized controlled trials that show a clear therapeutic effect of ozone on cancer disease. The aim of this study was to test the effects of an intraperitoneally insufflated O3/O2 gas mixture (O3/O2-pneumoperitoneum, O3/O2-PP) on growth of the primary tumour and metastatic spread of the VX2 auricle cancer in NZWR. Fourteen days after inoculation of a VX2 tumour cell suspension into the subcutis of the right auricle rabbits were randomized to three experimental groups. The treatment was performed daily for five consecutive days, beginning on day 14 after tumour cell inoculation. Rabbits in Group A (n = 14, O3/O2 group) each day received an intraperitoneally insufflated volume of 80 ml/kg body weight of an O3/O2 gas mixture with an ozone concentration of 50 μg/ml gas mixture. Rabbits in Group B (n = 13, O2 group) received the same amount of pure oxygen and rabbits in Group C (n = 14, Sham group) were sham-treated. Arterial blood samples were taken on day 14 (prior to initiation of treatment) and on day 19 (after the last treatment) and standard haematological and clinical chemistry blood parameters were measured. O3/O2-PP significantly increased the 3-month survival probability (50.0%) and the complete remission rate (42.9%) compared with sham treatment (7.1% 3-month survival probability and complete remission rate). The 3-month survival probability (23.1%) and complete remission rate (15.4%) of the rabbits that received pure medical oxygen were neither significantly different from the sham group nor from the O3/O2 group. All animals with complete remission of the primary tumour had an enlarged sentinel lymph node (parotid lymph node) at the beginning of the treatment period that regressed simultaneously. None of these rabbits had detectable lung metastases at day 90 on computed tomography scans of the thorax. No severe side effects could be detected except for a mild transient body weight loss and a mild leukocytosis pointing to O3/O2-PP as a relatively safe method. Animals of the O3/O2 group with complete tumour remission were resistant to the tumour cells after reinoculation of the tumour cell suspension. However, when treating these animals with dexamethasone and cyclosporin A, rabbits became susceptible to reinoculation of the tumour again. This indicates a crucial role of the immune system in tumour eradication with subsequent resistance to reimplantation. However, prior to a potential clinical use of the O3/O2-PP method, further research is essential to clear up the exact mechanisms by which ozone exerts this antitumoural effects. Particularly, further clarification is needed regarding the dose response relationship, evaluation of yet undetected adverse effects and reproducibility of the results of the VX2 tumour model in other tumour model systems