Human renal carcinoma line transfected with interleukin-2 and/or interferon alpha gene(s): implications for live cancer vaccines.

BACKGROUND: Combination therapy with systemically administered interleukin-2 (IL-2) and interferon alpha (IFN-alpha) has resulted in long-term objective remissions in 30% of patients with metastatic renal cell carcinoma (RCC), but toxic effects are clinically significant. PURPOSE: We have thus investigated an alternative therapeutic approach--continuous intratumoral production of IL-2 and/or IFN-alpha by a cytokine-transfected human RCC tumor cell line. METHODS: Plasmid vectors were used to transfect the R11 RCC line with the genes for human IL-2 and/or IFN-alpha by the calcium phosphate precipitation method. Biologic characteristics of the cytokine-transfected tumor cells were determined by assays of thymidine incorporation and cytotoxicity, fluorescence-activated cell-sorter analysis, Northern blotting, and in vivo studies in C3Hf/Sed/Kam mice rendered T-cell deficient. RESULTS: The transfected cell lines produced the following amounts of cytokine per 10(6) cells per day: R11-IL-2 (220 U), R11-IFN-alpha (10,240 U), and R11-IL-2 + IFN-alpha (95 U + 1270 U, respectively). Gamma irradiation did not eliminate cytokine secretion. Morphology and growth rates were identical to those for the parental R11 cell line, except for IFN-alpha-producing clones, which showed significant growth inhibition. All cytokine-producing cells demonstrated increased susceptibility to cell killing by peripheral blood leukocytes (PBL). IFN-alpha producers exhibited enhanced HLA antigen expression and suppressed c-myc messenger RNA expression; when cocultured in vitro, they induced similar changes in parental R11 cells. IL-2 producers could stimulate growth and cytotoxicity of naive (i.e., freshly isolated, uncultured) and activated PBL. All cytokine-producing cells lost their tumorigenicity, as evidenced by failure to grow in the T-cell-depleted mice. When co-injected at a local site but not at a distant site, these cells prevented growth of parental R11 cells. Histologic examination of the injection sites revealed a substantial influx of macrophages. Intraperitoneal administration of IL-2 and/or IFN-alpha could not, however, prevent growth of the parental R11 tumors. CONCLUSION: Local production of high concentrations of IL-2 and IFN-alpha at the tumor site is more effective in preventing tumor growth than systemic administration. IMPLICATION: Continuous local delivery of cytokines via transfer of cytokine genes into tumor cells for use as live cancer vaccines is a novel strategy for manipulation of host-mediated antitumor immune response in patients with advanced RCC.

Erythropoietin increases hemoglobin in cancer patients during radiation therapy.

PURPOSE: The hemoglobin (Hgb) level of patients during radiation therapy is associated with both survival and local tumor control in several organ sites. This clinical trial tested whether administering recombinant human erythropoietin (r-HuEPO) to cancer patients would increase their Hgb level during the course of radiation therapy without adverse effects. METHODS AND MATERIALS: The 40 participating patients had a Hgb value < 13.5 g/dL and a malignant tumor located above the diaphragm without evidence of distant metastasis for which they were scheduled to undergo a 5-8 week course of daily radiation therapy. All 40 patients were given oral ferrous sulfate throughout their radiation therapy course. Half the patients also received 150-300 mg/kg of r-HuEPO subcutaneously three times per week starting 0-10 days prior to the first dose of radiation. RESULTS: The r-HuEPO and control groups did not differ significantly in patient age, gender, tumor type, initial hemoglobin, erythropoietin, or iron bioavailability. The Hgb level increased more than 6% during radiation therapy in all 20 of the r-HuEPO patients but in only 2/20 of the control patients (p < 0.001). The Hgb rose from a mean +/- standard deviation of 11.9 +/- 1.3 g/dL to > 14 g/dL during radiation therapy in 80% of the r-HuEPO group compared to in 5% of the control group (p < 0.001). The mean +/- s.d. change in Hgb concentration during radiation therapy was 27 +/- 15% (an average rise of 5% per week) in the r-HuEPO group and 0 +/- 6% in the control group (p < 0.001). r-HuEPO had no significant measurable effect on blood pressure, white blood cell, neutrophil or platelet count, or liver or renal function. The only reported adverse effect of r-HuEPO administration was an asymptomatic skin rash in one patient. CONCLUSION: r-HuEPO with ferrous sulfate significantly increased the Hgb level in cancer patients without interfering with their course of radiation therapy, whereas ferrous sulfate alone did not. r-HuEPO appears to be a safe and effective means of increasing red cell mass during radiation therapy.

Dose, time, and fraction size issues for late effects in head and neck cancers.

This paper contains a statistical analysis of the dose-time factors influencing late complications in 784 patients with squamous cell carcinomas of the pharynx or larynx treated with external beam irradiation only at the University of Florida. The patients include 560 who received continuous course once-a-day therapy, 116 who received twice-a-day treatment, and 108 who received a once-a-day split course regimen. Both 2+ and 3+ complications were considered. Fifty-six patients developed either of these complications. The factors included in the analysis were site and size of the primary, total dose, fraction size, and treatment time. The linear-quadratic model was used to incorporate fraction size into the analysis. Proportional hazards analysis, which models the time to occurrence of the late complication, was used to quantify the joint influence of the above patient and fractionation variables on the incidence of late effects. The occurrence of the late effects was heterogeneous, with only a weak relationship to the patient and fractionation variables. The influence of the size of the primary was significant, with larger primaries associated with higher complication rates independent of fractionation variables. For oropharynx primary sites there was no significant effect of the fractionation variables. For larynx and hypopharynx, excluding T1-T2 true vocal cord, there was a significant effect of total dose and fraction size. The alpha/beta ratio was estimated to be 7.8 Gy (95% confidence interval, 3.0, infinity). There was no significant effect of overall treatment time. The estimated 2+ complication rate at 1 year from 68 Gy given in 2 Gy fractions in 50 days is 0.1% for T 1-2 vocal cord, 4.1% for T1-T2 supraglottic larynx, 3.8% for T3 supraglottic larynx and vocal cord, 14.9% for T4 supraglottic larynx, 6.7% for T1-T2 tonsil and soft palate, 7.6% for T3-T4 tonsil and soft palate, 7.0% for T1-T2 pyriform sinus and pharyngeal wall, and 13.0% for T3-T4 pyriform sinus and pharyngeal wall.

The influence of dose and time on wound complications following post-radiation neck dissection.

Data on 205 patients who were treated with a planned unilateral neck dissection following radiation therapy were analyzed with the purpose of understanding how treatment factors affect the incidence of wound complications. There were 27 occurrences of wound complication in the patient series. Logistic regression was used to analyze the data. We found that the surgical technique of flap reconstruction gave a significant increase in wound complications. There was a suggestion, although not statistically significant, that higher total doses increased the complication rate, that lower fraction sizes reduced the complication rate, and that longer overall radiotherapy treatment times were associated with higher complication rates. There was no association between the incidence of complications and the time interval between the end of radiotherapy and surgery.

The effect of hyperfractionation on spinal cord response to radiation.

The T10-L2 level of the spinal cord of C3Hf mice was irradiated using a conventionally fractionated regimen of 2.0 Gy once daily or a hyperfractionated regimen of 1.2 Gy twice daily separated by 8 hr. After a fractionated dose of 24-60 Gy given by either regimen, a top-up dose of 15 Gy was given. Hind limb strength was then measured weekly for 15 months. The time to onset of paralysis was inversely associated with the total dose. Overall, the spinal cord was not spared by hyperfractionation to the extent predicted by the modified Ellis power law or the linear-quadratic model. The threshold dose for the development of paralysis was higher in the hyperfractionated than in the conventionally fractionated group. However, the latent period for paralysis and the dose producing hind limb paralysis in 50% of the mice (ED50) were not significantly different between the two regimens. The continuation of the process of sublethal damage (SLD) repair in the spinal cord beyond 8 hr after irradiation may have influenced these results. The slow component of SLD repair should be considered in the design of hyperfractionated or accelerated radiation therapy schedules for clinical use.

Severe anemia is associated with poor tumor oxygenation in head and neck squamous cell carcinomas.

PURPOSE: To investigate the relationship between tumor oxygenation and the blood hemoglobin (Hb) concentration in patients with squamous cell carcinoma of the head and neck (SCCHN). METHODS AND MATERIALS: A total of 133 patients with SCCHN underwent pretreatment polarographic pO2 measurements of their tumors. In 66 patients measurements were also made in sternocleidomastoid muscles. The patients were divided into three groups according to their Hb concentration-severe anemia (Hb < 11.0 g/dl), mild anemia (female: Hb 11.0-11.9 g/dl; male: Hb 11.0-12.9 g/dl), and normal Hb concentration (female: Hb > or =12.0 g/dl; male: > or =13.0 g/dl). RESULTS: No significant difference in tumor oxygenation could be detected between mildly anemic patients and patients with a normal Hb level. However, the tumor oxygenation in the severely anemic group was significantly below that of each of the other two groups (p < 0.0001). There was no significant difference between the Hb groups in oxygenation of sternocleidomastoid muscles. In a multivariate analysis including Hb group, tumor volume, smoking habits, gender, T-stage, N-stage, and histologic grade a Hb level < 11 g/dl was found to be the strongest predictor for a poor tumor oxygenation. Smoking also had a marginal influence on median pO2. CONCLUSION: Our data suggest that a low Hb concentration and cigarette smoking contribute to inadequate oxygenation of SCCHN and thus for increased radioresistance. Consequently, Hb correction and abstinence from smoking may significantly improve tumor oxygenation.

The extent, time course, and fraction size dependence of mouse spinal cord recovery from radiation injury.

PURPOSE: This experiment was designed to assess: (a) the influence of fraction size and time interval between fractions on the tolerance of the spinal cord to high cumulative doses of radiation; and (b) the influence of the long-term recovery process on the tolerance of the spinal cord to reirradiation. METHODS AND MATERIALS: The T10-L2 level of the spinal cord of C3Hf mice was irradiated using a conventionally fractionated regimen of 2.0 Gy once daily, a prolonged fractionated regimen of 1.2 Gy once daily, a hyperfractionated regimen of 1.2 Gy twice daily, or a single dose of 12 Gy followed 0-190 days later by a second dose of 5-20 Gy. Mice in the multifractionated regimen groups were given a single 15 Gy top-up dose 24 h after reaching a cumulative fractionated dose of 24-70 Gy. Hind limb strength was measured weekly for 2 years after the completion of irradiation. RESULTS: Paralysis occurred in a bimodal time distribution, with peaks at 5-10 months and 15-23 months after the completion of irradiation. The cumulative radiation dose was directly associated with the incidence of paralysis in each radiation schedule (p < 0.0001) and inversely associated with the time to onset of paralysis in the 1.2 Gy b.i.d. (p = 0.0001) and 2.0 Gy q.d. schedules (p = 0.03). The median latency of paralysis in each group was inversely associated with the incidence of paralysis in that group (p < 0.001). Decreasing the fraction size from 2.0 to 1.2 Gy once daily markedly increased the radiation tolerance of the spinal cord (p < 0.0001), consistent with a very small alpha-beta value of -0.30 Gy (approximately 95% confidence interval -0.72, +0.18) in the linear-quadratic model. Decreasing the time interval from 24 h to alternating 8 and 16 h periods produced an offsetting diminuation in cord tolerance (p < 0.0001). The 1.2 Gy once daily schedule resulted in ED20 and ED50 values that were approximately double those of the 2.0 Gy once daily and the 1.2 Gy twice daily schedules and a relative risk of paralysis from a given dose that was 0.03 times the risk associated with the other two regimens (p < 0.0001). There was no significant difference between the 2.0 Gy once daily and the 1.2 Gy twice daily dose-paralysis curves (p = 0.86). The residual from a single 12 Gy radiation dose was 17% after 190 days, leaving the retreatment ED50 only 10% below the ED50 of previously unirradiated spinal cord. The relative risk of paralysis after 12 Gy plus a second radiation dose decreased from 1.00 with no time interval between doses to 0.51-0.73 with a 0.25, 1 or 3 day interval, 0.32 with a 7 day interval, 0.11 with a 30 day interval, and 0.06 with a 190 day interval. CONCLUSION: The increased radiation tolerance of the murine spinal cord produced by decreasing the fraction size from 2.0 to 1.2 Gy was offset by the diminished tolerance produced by decreasing the time interval between fractions from 24 to 8-16 h, resulting in no significant difference in the dose-paralysis curves of conventional and hyperfractionated radiation schedules. The rodent spinal cord eliminates the majority of the occult radiation injury produced by a radiation dose equal to half the ED50 during the months following irradiation. This permits retreatment of previously irradiated spinal cord to high doses without the induction of myelopathy.

Impact of hemoglobin level and use of recombinant erythropoietin on efficacy of preoperative chemoradiation therapy for squamous cell carcinoma of the oral cavity and oropharynx.

PURPOSE: We assessed the influence of hemoglobin level and r-HuEPO administration on response to chemoradiotherapy, locoregional tumor control, and overall survival in patients treated with neoadjuvant chemoradiotherapy and surgery for a squamous cell carcinoma of the oral cavity or oropharynx. METHODS AND MATERIALS: The 191 study patients were treated with mitomycin C (15 mg/m(2) day 1), 5-fluorouracil (750 mg/m(2)/day, days 1-5), and radiotherapy (50 Gy in 25 fractions weeks 1-5), followed by resection of the primary tumor bed and neck dissection at the General Hospital Vienna, Austria, between November 1989 and October 1998 for a T2-4, N0-3, M0 SCC of the oral cavity or oropharynx. Starting in May 1996, patients with a low hemoglobin (Hgb) before or during chemoradiotherapy received r-HuEPO 10,000 IU/kg s.c. 3-6 times/week until the week of surgery. RESULTS: On multivariate analysis, Hgb level and use of r-HuEPO were independent prognostic factors for response to chemoradiotherapy and locoregional tumor control (p < 0.01). Pathologic response to neoadjuvant therapy was also predictive of locoregional control (p < 0.001). Patients with a pretreatment Hgb > or = 14.5 g/dL had significantly higher complete response, locoregional control, and survival rates than the patients with a pretreatment Hgb < 14.5 g/dL who did not receive r-HuEPO (p < 0.05). The response, control, and survival rates in patients with a pretreatment Hgb < 14.5 g/dL given r-HuEPO were significantly higher than in low Hgb patients not given r-HuEPO (p < or = 0.001) and equivalent to patients with a pretreatment Hgb > 14.5 g/dL (p > or = 0.3). CONCLUSION: Low pretreatment Hgb is a negative prognostic factor for oral cavity and oropharyngeal SCCA patients, but was completely abrogated by r-HuEpo administration during neoadjuvant chemoradiotherapy. Randomized trials of radiation and/or chemotherapy with or without r-HuEPO for patients whose Hgb level is either low at the start of therapy or is anticipated to become low during therapy are indicated.

Time-dose factors in positive neck nodes treated with irradiation only.

This article describes an analysis of time-dose and clinical factors which affect the 2 year rate of control of cervical node metastases from squamous cell carcinoma of the head and neck following external beam radiotherapy in a series of 140 patients. We find that node diameter and normalized total dose are the most important factors, and that overall treatment time is not statistically significant.

Partial laryngectomy for glottic cancer after high-dose radiotherapy.

A vertical partial laryngectomy (VPL) for salvage was performed on 25 patients with locally persistent or recurrent squamous cell carcinoma of the vocal cord(s) after high-dose radiotherapy at the UCLA Medical Center between 1969 and 1988. Patients were followed for a minimum of 2 years and a median of 4.4 years after VPL. Ninety-six percent of patients remained free of disease. Tumor was controlled in patients with impaired vocal cord mobility and involvement of the contralateral cord or false cord. The actuarial survival rate was 80% at 5 years. There were no serious wound healing problems. A permanent tracheostomy was required in one patient due to recurrent aspiration pneumonia. Swallowing and voice function were satisfactory in all other patients. These results indicate that the selection criteria for initial VPL can be applied to the salvage situation with similar success.

Efficacy and morbidity of partial laryngectomy and postoperative radiation therapy.

In recent years, we have seen increasing use of partial laryngectomies for larger lesions that were previously treated by total laryngectomy. The resultant closer margins have made postoperative radiation therapy an important adjuvant treatment to conservation laryngeal surgery. We review the University of California, Los Angeles, experience with combination partial laryngectomy and postoperative radiation therapy between 1973 and 1987 for treatment of carcinoma of the larynx. Twenty-four such patients who underwent partial laryngectomies and postoperative radiation therapy are examined. Techniques of treatment, complications, and the functional ability of the remaining larynx are discussed. The locoregional control rate at 5 years was 80%. Risk factors associated with an increased risk of recurrence were positive margins, vascular invasion, and extranodal spread. There were no major problems with postoperative wound healing or airway management during the radiation treatment. Vocal and swallowing function were well preserved in most cases. We conclude that combination partial laryngectomy and radiation therapy permits preservation of laryngeal function without serious complications, and therefore is an effective treatment for selected patients with carcinoma of the larynx.

Potential therapeutic misadministration due to inappropriate electron beam field shaping.

Lead or cerrobend blocking strips are used to shape electron treatment fields when an appropriate custom insert is not available. For the Varian 2100C accelerator, the structural supports of the electron applicators impede the free placement of these field-shaping strips on the open custom insert frame while placement at the top of the applicator is unimpeded. We have investigated the dosimetric ramifications of placing field shaping strips at the top level of the 15x15 applicator for 6, 9, and 16 MeV electrons. Our results demonstrate as much as a 30% dose decrease and 2 cm penumbral increase when this is done compared to field shaping at the insert level. The magnitude of this dosimetric error qualifies as a therapeutic misadministration in many states depending on how many treatments are delivered in this manner. Based on this finding, we recommend that routine use of lead strip blocking be discouraged in favor of custom inserts due to the potential for inappropriate placement on some linear accelerators.

Clinical trial experience using erythropoietin during radiation therapy.

Oncologists have several reasons for trying to maintain or increase hemoglobin levels in their patients during therapy. Relief of the symptoms of anemia, including fatigue and dyspnea, are traditional, well-accepted indications. A newer rationale is to enhance the efficacy of radiation therapy and/or chemotherapy in controlling tumors. A laboratory animal study found that administration of recombinant human erythropoietin (rHuEPO) increased intratumoral median oxygen levels and diminished the proportion of measurements in the very low (< 3 mm Hg) range. Hemoglobin level is a strong independent prognostic factor for tumor control by radiation therapy. The hemoglobin level at the end of radiation therapy is a stronger prognostic factor than is the hemoglobin level at the start of therapy. Numerous clinical trials have utilized rHuEPO during radiation with or without concurrent chemotherapy. All 4 trials which enrolled patients with low hemoglobin levels (< 12 to 13.5 g/dl) found that rHuEPO significantly increased hemoglobin within 2 weeks and that hemoglobin levels continued to rise until the end of rHuEPO treatment. rHuEPO was efficacious in limiting the decrease in hemoglobin and use of packed red blood cell transfusion in the one reported trial in which it was used in patients with initially normal hemoglobin levels during intensive concurrent radiation and chemotherapy. One trial found a statistically significant improvement in complete pathologic response rate after neoadjuvant chemoradiotherapy with the use of rHuEPO. rHuEPO has a potentially large to play in the care of the cancer patient.

Impact of radiation therapy and/or chemotherapy on the risk for a second malignancy after breast cancer.

The risk of any second malignancy was determined for all patients treated for a primary cancer of the breast without evidence of distant metastasis at Duke University Medical Center between 1970 and 1981. The incidence, 10-year actuarial risk (AR), and relative risk (RR) of a second malignancy developing were calculated for the 407 patients who were treated with surgery alone, 226 who were treated with surgery followed by adjuvant chemotherapy (CT), 140 who were treated with surgery plus adjuvant radiation therapy (RT), and 308 who received all three modalities (CRT). The AR of a subsequent cancer (8.4% for CRT, 8.7% for CT, 8.7% for RT, and 11.7% for surgery only patients) did not differ significantly between treatment groups. The overall second cancer RR was 1.0% after CRT (95% confidence interval [CI], 0.4 to 2.0), 1.3% after RT (95% CI, 0.6 to 2.5), 1.6% after CT (95% CI, 0.9 to 2.6), and 1.7% after surgery alone (95% CI, 1.2 to 2.4). Contralateral breast cancers (RR of 4.2%; 95% CI, 2.7 to 6.3) account for the statistically significant excess of second malignancies among the surgery alone patients. The AR for contralateral breast cancer in the surgery group was higher than in either group receiving CT (P less than 0.01), but was not significantly different from the RT group. The RR for solid tumors other than breast cancer was not significantly different from unity in any of the treatment groups. The RR for acute leukemia was 16.7% in the CRT group (95% CI, 0.2 to 92.7), 11.1% in the CT group (95% CI, 0.1 to 61.8), 10.0% in the surgery alone group (95% CI, 1.1 to 36.1), and 0.0% in the RT group (95% CI, 0.0 to 61.1). This study indicated that inclusion of RT and/or CT in the initial treatment of breast cancer did not impact negatively on patients' overall risk for a subsequent malignancy during the first decade after therapy, and that adjuvant CT with or without RT may decrease their risk of a contralateral breast cancer.

Impact on second malignancy risk of the combined use of radiation and chemotherapy for lymphomas.

The risk of a second malignancy was determined for 999 patients given primary treatment using chemotherapy only, radiation therapy only, or both for Hodgkin's Disease or a non-Hodgkin's lymphoma at Duke University Medical Center between 1970 and 1981. The incidence, 10-year actuarial risk, and relative risk of developing an acute leukemia, solid tumor, or second lymphoma were determined by treatment modality and initial lymphoma type. Among the 313 Hodgkin's disease patients, the acute leukemia actuarial risk was 2.0% after chemotherapy, 1.4% after radiation therapy, and 0.9% after combined treatment with chemotherapy and radiation therapy. Their relative risk for acute leukemia was 51.3 overall (95% confidence interval [CI] 13.8 to 131.8) and was elevated in each treatment group. Among the 686 non-Hodgkin's lymphoma patients, the acute leukemia actuarial risk was zero after radiation therapy, 4.6% after chemotherapy, and 4.5% after the combined treatment, again not significantly different between treatment groups. The leukemia relative risk was 10.6 (95% CI 3.4 to 24.8) in the chemotherapy and 11.9 (95% CI 3.2 to 30.6) in the combined treatment group. Among both the Hodgkin's disease and non-Hodgkin's lymphoma populations, the combined treatment group had a lower actuarial risk for solid tumors than either the chemotherapy or radiation therapy group (P less than 0.02). Solid tumor actuarial risk did not differ significantly between the chemotherapy and radiation therapy groups. Hodgkin's disease patients had a solid tumor relative risk that was elevated significantly after radiation therapy (6.5; 95% CI 2.4 to 14.0) and to a lesser extent after chemotherapy (2.6; 95% CI 0.8 to 6.1) or combined treatment (1.7; 95% CI 0.2 to 6.0). Solid tumor relative risk among non-Hodgkin's lymphoma patients was 0.3 for the combined treatment, 0.8 for the chemotherapy, and 1.0 for the radiation therapy group. None of the Hodgkin's disease patients developed a non-Hodgkin's lymphoma. This study found no significant difference in leukemia risk among lymphoma patients treated with chemotherapy and the combined treatment. It also found that the overall risk of a second malignancy is no higher after treatment with the combined therapy than with chemotherapy or radiation therapy alone.

Inhibitory effect of locally produced and exogenous interleukin-6 on tumor growth in vivo.

In order to define the potential antitumor activity of the multifunctional cytokine interleukin-6 (IL-6), retrovirus-mediated gene transfer was used to introduce and express a cDNA encoding human IL-6 in the murine fibrosarcoma cell line Fsa-R. Although these genetically modified tumor cells appeared morphologically and phenotypically identical to control Fsa-R cells and had a similar plating efficiency in vitro, they were found to exhibit greatly reduced tumorigenicity in vivo following intravenous injection into syngeneic recipients. Exogenous IL-6 was shown to produce a similar inhibition of tumor growth in the lung if administered intraperitoneally. In contrast, tumor growth in subcutaneous sites was inhibited only if the tumor cells were engineered to express IL-6 locally, or if IL-6 was administered intratumorally. Intraperitoneal injection of IL-6 had no inhibitory effect. Tumors that did grow from IL-6-producing tumor cell inocula in subcutaneous sites were found to contain large numbers of macrophages. These results demonstrate that the antitumor activity of systemically administered IL-6 varies depending on the site of tumor growth and suggest an important role for IL-6 in the recruitment, proliferation and/or survival of tumor-associated macrophages.

Effect of nicotinamide and pentoxifylline on normal tissue and FSA tumor oxygenation.

Nicotinamide (NA) and pentoxifylline (PTX) sensitize experimental murine tumors to radiation without sensitizing normal tissues. They are presumed to exert this effect by reducing hypoxia in tumors. The present study evaluated the individual and combined effects of NA and PTX on oxygen levels in subcutaneous normal tissue and subcutaneous FSa fibrosarcoma tumors in the hind foot dorsum of C3H mice. Oxygen measurements were made using a polarographic needle electrode inserted into the tissue immediately before and/or 15-60 min after intraperitoneal administration of 500 mg/kg of NA, 50 mg/kg of PTX, or saline. The median tumor pO2 increased from a mean +/- S.E.M. of 4.1 +/- 1.1 mm Hg in saline-treated control mice to 6.8 +/- 1.9 mm Hg 15 min after NA, 7.6 +/- 1.4 mm Hg 60 min after PTX, and 6.7 +/- 1.1 mm Hg after NA and PTX in combination. PTX raised the median tumor pO2 level from 21% to 39% of the median subcutaneous normal tissue pO2 (p < 0.01). PTX also significantly reduced the proportion of tumor pO2 values < or = 2 mm Hg from 41 +/- 10% to 8 +/- 7% (p = 0.02). Although NA did increase the proportion of tumor that was well oxygenated, it did not significantly reduce the proportion of tumor pO2 values < or = 2 mm Hg (p = 0.34). The combination of NA and PTX did not add to the tumor oxygenation enhancement achieved by PTX alone. NA increased the median subcutaneous normal tissue pO2 by an average of 5.1 +/- 2.2 mm Hg from a baseline of 17.1 +/- 2.2 mm Hg (p = 0.04). PTX had no effect on the median normal tissue pO2 (p = 0.93). PTX showed greater therapeutic potential in this model system than did NA.