Margaret L. Evans

Quantifying regional hypoxia in human tumors with positron emission tomography of [18F]fluoromisonidazole: A pretherapy study of 37 patients

PURPOSE To assess pretreatment hypoxia in a variety of tumors using positron emission tomography (PET) after injection of the hypoxia-binding radiopharmaceutical [18F]fluoromisonidazole ([18F]FMISO). METHODS AND MATERIALS Tumor fractional hypoxic volume (FHV) was determined in 21 nonsmall cell lung cancer patients, 7 head and neck cancer patients, 4 prostate cancer patients, and 5 patients with other malignancies by quantitative PET imaging after injection of [18F]FMISO (0.1 mCi/kg). The FHV was defined as the proportion of pixels in the imaged tumor volume with a tissue:blood [18F] activity ratio > or = 1.4 at 120-160 min postinjection. A FHV > 0 was taken as evidence for tumor hypoxia. RESULTS Hypoxia was observed in 36 of 37 tumors studied with FMISO PET imaging; FHVs ranged from 0 to 94.7%. In nonsmall cell lung cancers (n = 21), the median FHV was 47.6% and the range, 1.3 to 94.7%. There was no correlation between tumor size and FHV. In the seven head and neck carcinomas, the median FHV was 8.8%, with a range from 0.2 to 18.9%. In the group of four prostate cancers, the median and range were 18.2% and 0 to 93.9%, while in a group of five tumors of different types the median FHV was 55.2% (range: 21.4 to 85.8%). CONCLUSIONS Hypoxia was present in 97% of the tumors studied and the extent of hypoxia varied markedly between tumors in the same site or of the same histology. Hypoxia also was distributed heterogeneously between regions within a single tumor. These results are consistent with O2 electrode measures with other types of human tumors. The intra- and intertumor variability indicate the importance of making oxygenation measures in individual tumors and the necessity to sample as much of the tumor volume as possible.

l Brief Communication QUANTIFYING REGIONAL HYPOXIA IN HUMAN TUMQRS WITH POSITRON EMISSION TOMOGRAPHY OF ('*F)FLUUROMISONIDAZOLE: A PRETHERAPY STUDY OF 37 PATIENTS

PURPOSE To assess pretreatment hypoxia in a variety of tumors using positron emission tomography (PET) after injection of the hypoxia-binding radiopharmaceutical [18F]fluoromisonidazole ([18F]FMISO). METHODS AND MATERIALS Tumor fractional hypoxic volume (FHV) was determined in 21 nonsmall cell lung cancer patients, 7 head and neck cancer patients, 4 prostate cancer patients, and 5 patients with other malignancies by quantitative PET imaging after injection of [18F]FMISO (0.1 mCi/kg). The FHV was defined as the proportion of pixels in the imaged tumor volume with a tissue:blood [18F] activity ratio > or = 1.4 at 120-160 min postinjection. A FHV > 0 was taken as evidence for tumor hypoxia. RESULTS Hypoxia was observed in 36 of 37 tumors studied with FMISO PET imaging; FHVs ranged from 0 to 94.7%. In nonsmall cell lung cancers (n = 21), the median FHV was 47.6% and the range, 1.3 to 94.7%. There was no correlation between tumor size and FHV. In the seven head and neck carcinomas, the median FHV was 8.8%, with a range from 0.2 to 18.9%. In the group of four prostate cancers, the median and range were 18.2% and 0 to 93.9%, while in a group of five tumors of different types the median FHV was 55.2% (range: 21.4 to 85.8%). CONCLUSIONS Hypoxia was present in 97% of the tumors studied and the extent of hypoxia varied markedly between tumors in the same site or of the same histology. Hypoxia also was distributed heterogeneously between regions within a single tumor. These results are consistent with O2 electrode measures with other types of human tumors. The intra- and intertumor variability indicate the importance of making oxygenation measures in individual tumors and the necessity to sample as much of the tumor volume as possible.

Evaluation of oxygenation status during fractionated radiotherapy in human nonsmall cell lung cancers using [F-18]fluoromisonidazole positron emission tomography

PURPOSE Recent clinical investigations have shown a strong correlation between pretreatment tumor hypoxia and poor response to radiotherapy. These observations raise questions about standard assumptions of tumor reoxygenation during radiotherapy, which has been poorly studied in human cancers. Positron emission tomography (PET) imaging of [F-18]fluoromisonidazole (FMISO) uptake allows noninvasive assessment of tumor hypoxia, and is amenable for repeated studies during fractionated radiotherapy to systematically evaluate changes in tumor oxygenation. METHODS AND MATERIALS Seven patients with locally advanced nonsmall cell lung cancers underwent sequential [F-18]FMISO PET imaging while receiving primary radiotherapy. Computed tomograms were used to calculate tumor volumes, define tumor extent for PET image analysis, and assist in PET image registration between serial studies. Fractional hypoxic volume (FHV) was calculated for each study as the percentage of pixels within the analyzed imaged tumor volume with a tumor:blood [F-18]FMISO ratio > or = 1.4 by 120 min after injection. Serial FHVs were compared for each patient. RESULTS Pretreatment FHVs ranged from 20-84% (median 58%). Subsequent FHVs varied from 8-79% (median 29%) at midtreatment, and ranged from 3-65% (median 22%) by the end of radiotherapy. One patient had essentially no detectable residual tumor hypoxia by the end of radiation, while two others showed no apparent decrease in serial FHVs. There was no correlation between tumor size and pretreatment FHV. CONCLUSIONS Although there is a general tendency toward improved oxygenation in human tumors during fractionated radiotherapy, these changes are unpredictable and may be insufficient in extent and timing to overcome the negative effects of existing pretreatment hypoxia. Selection of patients for clinical trials addressing radioresistant hypoxic cancers can be appropriately achieved through single pretreatment evaluations of tumor hypoxia.

USE OF STEROIDS TO SUPPRESS VASCULAR RESPONSE TO RADIATION

A quantitative measure of the vascular permeability surface area product (PS) for albumin has been made using a double isotope technique. PS was significantly elevated in irradiated rat lung, heart, skin, and muscle, between 19 and 26 days following 18 or 25 Gray thorax irradiation. Administration of dexamethasone from 2 days before irradiation through the day of measurement suppressed the expected increase in PS in lung, heart, and muscle, but not in skin. Shorter periods of steroid administration were not as effective in suppressing this response to radiation exposure. Increased vascular permeability following radiation may be an essential element in the development of radiation fibrosis. We hypothesize that the ability to suppress this response could result in a long term reduction in the incidence of fibrosis.

Changes in Vascular Permeability Following Thorax Irradiation in the Rat

A double isotope technique was used to measure changes in the vascular permeability surface area product (PS) for albumin after irradiation. PS was measured in several tissues of the rat during the first 38 days following 11, 13.5, 18, or 25 Gy whole thorax irradiation. After 18 and 25 Gy most irradiated and nonirradiated (shielded) tissues showed elevated permeability at 1 day after radiation, which declined to control levels by Day 4. All irradiated tissues showed a second wave of increased permeability between 14 and 38 days after radiation that varied in onset and extent depending upon tissue and dose. Lung and heart showed a direct response to dose between 11 and 18 Gy during this period. Peak lung values averaged three times control values at 19 days after 18 Gy. Peak heart values averaged twice control values at the same time and dose. The double isotope technique has proven to be a reliable means of quantitatively determining vascular permeability response to radiation over time.

Pharmacological alteration of the lung vascular response to radiation

The role of endothelial cell damage in the development of radiation injury in the lung was investigated in rats. Vascular permeability-surface area product (PS) was measured as an indicator of the degree of endothelial cell damage in lungs of rats exposed to single dose hemithorax irradiation. Hemithorax irradiation was chosen to simulate clinical radiotherapy, in which only a portion of the lung is irradiated. In addition, it provided a control lung to compare to the irradiated lung. Radiation is postulated to lead to activation of several different biochemical pathways that result in lung injury and fibrosis. Many of these pathways can be specifically blocked with drugs. Thirteen different drugs were studied. Dexamethasone, indomethacin, cromolyn, cyproheptadine, Vitamin D3, theophylline, and diethylcarbamazine were all effective at reducing lung PS on the irradiated side. Dexamethasone, Vitamin D3, and indomethacin also significantly reduced lung PS in the unirradiated lungs and in sham-irradiated rats. Captopril, cobra venom factor, penicillamine, trapidil, epsilon-amino caproic acid, and dapsone had no significant effect on lung PS after hemithorax irradiation. We conclude that the major pathways involved in early post-radiation lung injury involve prostaglandin, leukotriene, and histamine release from macrophages and mast cells. Complement activation, proteolytic enzymes, and neutrophil migration do not seem to be important mediators of early post-radiation lung injury.

Vascular Response to Radiation Injury in the Rat Lung

Changes in relative left-to-right lung blood flow ratios were followed as an index of vascular radiation injury in left-hemithorax-irradiated Sprague-Dawley rats. Single doses of 11 to 21 Gy gamma radiation resulted in a dose-dependent decrease in relative blood flow to the irradiated lung from 3 to 5 weeks after exposure during the development of pneumonitis. Blood flow returned to near normal by 5 weeks after lower doses (11-13.5 Gy). After a single dose of 15 Gy the left-to-right blood flow ratio recovered to 75% of normal at 12 weeks and leveled off. Following 18 Gy irradiation a second period of reduced flow began 16 weeks after exposure. After 21 Gy irradiation flow to the irradiated side remained low for 1 year after exposure. Rats that received a single dose of 18 Gy to the left hemithorax were also treated with one or two of the following drugs: captopril, cyproheptadine, dexamethasone, diethylcarbamazine, penicillamine, or theophylline. Dexamethasone was most effective at preventing the decrease in blood flow to the irradiated lung when treatment was continued through the pneumonitis period and dose was not tapered until 8 weeks after radiation exposure. All other drugs and drug combinations were, for the most part, virtually ineffective after the pneumonitis period. There was a relatively poor correlation with earlier vascular permeability surface area product studies. This suggests that endothelial damage, as well as damage to other cell types, contributes to the development of post-irradiation fibrosis in the lung.

A simple, dual tracer method for the measurement of transvascular flux of albumin into the lung

A simple, first-order model of albumin kinetics in the rat lung is presented and validated with a more sophisticated model. The simple model assumes that intravascular concentration of tracer albumin is constant over 30 min after injection and transvascular flux of tracer albumin is unidirectional and proportional to the permeability-surface area product (PS). 125I-albumin is injected initially and 131I-albumin at 20 min. At 30 min the rat is sacrificed and plasma and tissue samples are obtained for gamma counting. Simultaneous equations are set up for the two tracers and solved for PS and plasma volume. The accuracy of this approach is examined with data generated from a more complete model. This model uses the concepts of hydraulic conductivity, solvent drag, reflection coefficients, hydrostatic and osmotic pressures, exclusion volumes, and lymph flow, as well as PS. Based on known PS and clearance rates from the complex model, the simple model estimates tracer albumin leakage rate with less than 5% error over the range of PS encountered in rat studies.

Vascular Response to Fractionated Irradiation in the Rat Lung

The effects of fractionated hemithorax irradiation on normal lung tissue were examined by measuring changes in the vascular permeability surface area product (PS) and relative lung blood flow in Sprague-Dawley rats. The rats received five daily fractions per week of either 3.0 or 4.0 Gy for 4 weeks to the left lung. Between 3 and 5 weeks after the start of irradiation, the average PS was approximately 50% above normal for the group of rats that received 3.0 Gy/day and 200-300% above normal in the group of rats that received 4.0 Gy/day. Treatment with cyproheptadine, indomethacin, or theophylline had no effect, but treatment with dexamethasone significantly reduced PS to near normal levels. Left-to-right blood flow ratios in the group of rats that received 3.0 Gy/day decreased to 66% of normal levels by 4 weeks. In the group of rats that received 4.0 Gy/day, blood flow decreased to 46% of normal levels by 4 weeks. Treatment with dexamethasone maintained normal blood flow until the drug dose was reduced. These results agree with earlier studies using single-dose irradiation and indicate that the methods used to measure PS and blood flow are sensitive at low doses.