Dalice A. Sim

The effect of rate of heating or cooling prior to heating on tumor and normal tissue microcirculatory blood flow.

Single vessel responses to hyperthermia were studied in tumor and normal tissues using a transparent access window chamber. Rates of heating less than or equal to .68 degrees C/minute preserved relatively better vascular function in normal than tumor tissue. A rate of heating of 1.0 degrees C/minute lowered normal tissue statis temperatures so they were no different from tumor. Cooling to 30 degrees C prior to heating slowed normal arteriolar flows to less than 5% of 38 degrees C controls. Heating resulted in increased flow in those vessels, but maximum flows never exceeded 5% of flows achieved in similar vessels which were not cooled first. The implications of this work are that rate of heating and cooling prior to heating can alter normal tissue vascular response to heat in a way that could prove deleterious to maintaining efficient vascular function in that tissue relative to tumor.

PRELIMINARY RESULTS OF A PHASE III TRIAL OF SPONTANEOUS ANIMAL TUMORS TO HEAT AND/OR RADIATION: EARLY NORMAL TISSUE RESPONSE AND TUMOR VOLUME INFLUENCE ON INITIAL RESPONSE

A Phase III randomized trial was initiated to test the relative efficacies of heat alone, radiation alone and heat plus radiation using spontaneous malignancies in pet animals. Heat alone was inferior to the other two treatment arms as demonstrated by a significantly higher non-response rate and shorter response duration. The ratio of complete response rates (CR) for heat plus radiation to radiation alone or the thermal relative risk (TRR) was greater for tumors greater than 10 cm3 as compared to those less than 10 cm3 (TRR = 4.8 and 1.4, respectively). The overall TRR for complete responses was 2.3. The CR data for the combined therapy arm indicate at least an additive effect between heat and radiation for small tumors but most likely a synergistic effect in the larger tumor group. Based on the data currently available, no significant difference in response duration is observed between the two radiation arms, although a nonsignificant advantage to the combination therapy exists. Normal tissue effects were evaluated by incidence of full moist desquamation within the irradiated volume, late fibrosis and bone necrosis. Since the radiation skin dose depended upon the technique being used it was possible to estimate the dose to achieve moist desquamation in 50% of the animals (DD50) by a logistic regression model as being 3728 +/- 344 rad for radiation alone. Significant lowering of the DD50 was not observed for the addition of heat to radiation. Low patient numbers where intact skin was heated prevented an accurate analysis of the effect, however.

Institutional reports: Clinical evaluation of hyperthermia equipment: The University of Arizona Institutional Report for the NCI Hyperthermia Equipment Evaluation Contract

Two-hundred and fifteen independent sites in 203 patients were treated with hyperthermia at the University of Arizona from 10/81 through 3/86 under the auspices of this contract. In the head and neck, a site dominated by superficial tumors, air-coupled and water-coupled microwave applicators yielded the best results. Similarly in the thorax, also dominated by superficial tumors, water-coupled microwave applicators were best. In the abdomen and pelvis, sites dominated by deep tumors, only interstitial radiofrequency (RF) heating, an invasive technique useful only in selected cases, was capable of consistently producing therapeutic temperatures. Toxicity appeared to be site-related, and treatment discomfort was especially common in abdominal and pelvic sites. In conclusion, while superficial sites were readily heated using propagative electromagnetic devices, these devices were ineffective and poorly tolerated at deeper sites. Effective deep hyperthermia was best produced with interstitial techniques, and further development of these techniques using RF electrodes, implantable microwave antennas and thermoregulating ferromagnetic seeds, as well as scanned, focussed-ultrasound techniques, holds promise for effective heating of deep visceral sites.

Local control and distant metastases in primary canine malignant melanomas treated with hyperthermia and/or radiotherapy

Forty-three dogs with primary malignant melanoma were randomized to receive radiotherapy alone (XRT) or hyperthermia plus radiotherapy (delta + XRT). Tumour responses were analysed in terms of complete response rates, rate of one year disease free survival and the incidence and time to develop distant metastasis. The frequency of complete responses (CR) was greater with adjuvant heat (76 per cent vs 21 per cent for XRT; P = 0.001). A trend towards an improvement in one year disease free survival was observed with delta + XRT (23.8 per cent) as compared with XRT (7.7 per cent), but the difference was not statistically significant. The frequency of distant metastases was not different between the two treatments. Descriptors of intratumoural temperatures achieved during therapy indicated that higher CR rates could be achieved with higher minima. When minima were less than and greater than 20 Equivalent minutes at 43 degrees C (Eq43) the CR rates were 64 and 90 per cent, respectively. One year disease free survival rates and frequencies of distant metastases seemed to be correlated with the intratumoural temperatures as well. This was reflected in analyses examining temperature minima and maxima. Examination of patterns of failure suggested that the most plausible explanation for the correlation between intratumoural temperature and metastases was the high local failure rate (70 per in the heated group). The results of this study emphasize the need for further investigation of the influence of local hyperthermia as a part of curative therapy on the frequency of distant metastases.

Radiotherapy prior to cortical allograft limb sparing in dogs with osteosarcoma: A dose response assay☆

Twenty-one dogs with spontaneously occurring appendicular osteosarcoma were given preoperative radiation therapy prior to a limb sparing procedure using a cortical allograft. Radiation doses were randomly assigned, ranged from 36-52 Gy in 4 Gy intervals, and were given in 10 equally-sized fractions on a M, W, F schedule. Seventeen of the 21 dogs underwent the limb sparing procedure approximately 3 weeks after completion of radiation therapy. Local tumor recurrence was documented in 4 of 17 dogs at mean and median times of 5.5 and 5.8 months, respectively, after initiation of radiation therapy. Three of 4 recurrences were in anatomic regions with sparse adjacent soft tissue which precluded wide excision. Complications were significant. Fixation device failure occurred in 9 of 17 dogs and was associated with host bone necrosis, muscle thinning and fibrosis of vessels and nerves in irradiated normal tissue. Incidence of host bone necrosis was directly related to radiation dose (Kendalls statistic, p = 0.005). Metastasis occurred in all 21 dogs. Mean and median times to metastasis in these dogs were 5.1 and 4.0 months, respectively, after initiation of radiation therapy. Local tumor control rates and survival times were higher in dogs developing allograft infection suggesting that infection acted as an immunostimulant. All local failures occurred in dogs that did not develop allograft infection and median survival times for uninfected versus infected dogs were 5 and 11 months, respectively (logrank test, p = 0.029). Increased tumor radiopacity following radiation therapy was significantly related to survival. Median survival in dogs whose tumors were characterized by decreased, unchanged or increased opacity after radiation therapy were 3.5 and 14 months, respectively (logrank test, p = 0.014). Based on the results of our study, radiation therapy can not be recommended as part of limb sparing treatments for patients with osteosarcoma at doses and dose per fraction values similar to those used herein.

The effect of hydralazine dose on blood perfusion changes during hyperthermia

Experiments were performed to determine the dose-related effects of the intravenous administration of a vasodilator (hydralazine) on normal muscle blood perfusion during localized hyperthermia. Fourteen anaesthetized outbred canines were investigated, seven receiving the recommended dose level of 0.5 mg/kg and seven receiving one-quarter of that level. The changes in blood perfusion were estimated using two methods: calculation of an effective blood perfusion magnitude and the use of state and parameter estimation techniques. Both methods showed that the changes in blood perfusion induced by the hydralazine were significant, and that the differences between the results for the two drug doses were not significantly different. This suggests that low doses may be useful in humans, giving the same resultant blood perfusion increase but with a decreased patient risk relative to standard therapeutic doses of hydralazine. While the trends in the blood perfusion changes were the same for both calculation methods the effective perfusion method frequently yielded blood perfusion magnitudes significantly different from those obtained using the state and parameter estimation technique. The differences are postulated to be due to the fact that the effective perfusion values include conduction effects, thus overpredicting the amount of perfusion present. Thus, while the effective blood perfusion can be used as a qualitative indication of blood perfusion changes under certain conditions, we do not recommend its use as a quantitative measure of perfusion.