Cytokine levels and systemic toxicity in patients undergoing isolated limb perfusion with high-dose tumor necrosis factor, interferon gamma, and melphalan.

PURPOSE: Isolated limb perfusion (ILP) with tumor necrosis factor (TNF), interferon gamma, and melphalan (M) has been reported to result in high response rates for extremity melanoma and sarcoma. We have evaluated the relationship of systemic TNF exposure to induction of several secondary mediators and incidence of systemic toxicity. PATIENTS AND METHODS: Nineteen patients with extremity melanoma (n = 16) or sarcoma (n = 3), underwent 90-minute ILP with TNF-alpha, interferon gamma (0.2 mg), and M (10 to 13 mg/L of limb volume) (TNF/IFN/M) (n = 12), or M alone (n = 7). Continuous intraoperative monitoring (CIM) for systemic leak from the perfusion circuit was performed using radioactive iodine-131 albumin. Cytokine levels in the perfusate and systemic circulation during and after ILP were measured by enzyme-linked immunosorbent assay. RESULTS: Systemic leaks > or = 1% from the perfusion circuit occurred in six patients who received TNF/IFN/M and in four who received M alone. Hypotension that required vasopressor support occurred in six of six patients with evidence of a leak (> or = 1%) and zero of six patients without a leak (< 1%). These six patients had significantly higher peak systemic TNF levels during and after perfusion than patients without a leak (2.8 and 8.2 ng/mL v 0.7 and 2.0 ng/mL, respectively; P < .05). All patients who received TNF/IFN/M had significantly greater increases in systemic interleukin-6 (IL-6) levels than in patients with M alone (12,395 +/- 10,374 pg/mL v 79.4 +/- 7.2 pg/mL, respectively; P < .001). Intracellular adhesion molecule (ICAM), IL-8, and TNF-R levels were also increased after ILP with TNF/IFN/M. CONCLUSION: ILP with TNF/IFN/M can be safely performed, as I131 albumin provides a sensitive measure of systemic leakage from the perfusion circuit. Patients with a measured leak of > or = 1% develop mild and transient postoperative hypotension with significantly higher systemic TNF levels and lower perfusate TNF levels than in patients without leaks.

Effects of continuous therapeutic ultrasound on growth and metastasis of subcutaneous murine tumors.

BACKGROUND AND PURPOSE: The use of therapeutic ultrasound (US) in the presence of malignant neoplasms has been contraindicated in physical therapy practice despite a lack of convincing scientific evidence. Some studies have shown that high levels of US, which increase tissue temperatures greater than 42 degrees C, can kill tumors. We sought to determine whether the application of continuous therapeutic US would alter the growth or metastasis of methylcholanthrene-induced solid tumors in mice. SUBJECTS: Seventy-one female C57BL/6 mice, age 6 to 8 weeks, received subcutaneous injections of 5 x 10(5) tumor cells. METHODS: When tumors grew to 0.5 cm in diameter, the mice were randomly assigned to either a control group (n = 34) or an experimental group (n = 37). The experimental group received 10 treatments over a 2-week period of 3-MHz continuous US at 1.0 W/cm2 for 5 minutes, using a 0.5-cm2 sound head directly over the tumor. The control group received the same handling except for the US treatment. Tumor dimensions were measured on days 1 (baseline), 7 (midtreatment), and 14 (preexcision and postexcision). Tumors were weighed after excision, and the mice were evaluated by necropsy and histopathology of regional lymph nodes. RESULTS: All tumors grew larger over time, but final tumor volume and weight were larger in the experimental group (789 mm3 and 0.932 g) than in the control group (395 mm3 and 0.506 g). No significant difference existed in the number of metastatic lymph nodes between groups. CONCLUSION AND DISCUSSION: Continuous therapeutic US increased the volume and weight of subcutaneous murine tumors in mice. We urge caution in the use of continuous therapeutic US in the areas of tumors or suspected tumors.

IL-1 receptor antagonist (IL-1ra) augments IL-2-induced pulmonary vascular leak.

Interleukin-2 (IL-2) therapy is dose limited by a severe vascular leak with resulting systemic and pulmonary toxicity. Although recognized as a mediator of septic shock and vascular leak, the relative role of IL-1 in IL-2 toxicity is unclear. We evaluated the effect of IL-1 receptor antagonist (IL-1ra) on IL-2 lethality, pulmonary vascular leak, and treatment of pulmonary metastases in a murine model. In vivo induction of mRNA for IL-1 alpha was evaluated in liver by Northern blots after 0, 5, 8, and 11 doses of IL-2 in C3H/HEN mice. The expression index for the IL-1 alpha gene increased from 0.16 to 0.74 after 5 doses of IL-2, and further increased to 1.04 after 11 doses of IL-2. C3H/HEN mice (n = 56) were randomized to receive phosphate-buffered saline (PBS), IL-1ra high dose (HD), or IL-1ra low dose (LD) by continuous subcutaneous infusion via Alzet mini-pumps. The biologic effectiveness of the dose and administration of IL-1ra was determined by the ability to block IL-1-induced IL-6 production in vivo. Mean serum IL-6 levels 3 hr after intraperitoneal IL-1 alpha (10 micrograms/kg) were: PBS, 3730 +/- 526 (mean +/- SEM pg/ml); IL-1ra (LD), 1156 +/- 398; and IL-1ra (HD), 594 +/- 30 (P < 0.01, IL-1ra HD or LD vs PBS). Pulmonary vascular leak was measured by iv I125 albumin after 8 doses of IL-2 (100,000 U ip q 8 hr).(ABSTRACT TRUNCATED AT 250 WORDS)

Prospective study of the use of intraarterial secretin injection and portal venous sampling to localize duodenal gastrinomas.

BACKGROUND: Duodenal gastrinomas producing Zollinger-Ellison syndrome (ZES) are rarely imaged on preoperative studies. Measurement of serum gastrin levels by transhepatic portal venous sampling (PVS) or by sampling from hepatic veins after intraarterial secretin injection have been advocated as useful tests to identify these tumors before operation. METHODS: As part of a prospective study, selective intraarterial secretin injection has been performed in 36 consecutive patients with ZES, PVS has been performed in 30 of these patients, and the results have been compared. RESULTS: Gastrinomas were found at laparotomy in 33 of 36 patients (92%). Duodenal tumors were found in 18 patients (50%). The remaining patients had liver, pancreatic, or nodal disease (n = 15). Thirty-two of 36 patients (89%) had positive results with intraarterial secretin injection study, whereas 18 of 30 (60%) had a positive PVS gradient (p = 0.02, Fisher's exact test). The most common positive gradient with intraarterial secretin injection was found with injections of the gastroduodenal artery, and the most common positive gradient with PVS was found in the inferior pancreaticoduodenal (IPDV) or superior pancreaticoduodenal vein (SPDV). Fourteen of 18 (78%) patients with duodenal gastrinomas had a positive GDA injection, whereas five of 18 (28%) without duodenal tumors had a positive GDA injection (p = 0.006). Five of 16 patients with duodenal gastrinomas had a positive gradient in the IPDV or SPDV, whereas four of 14 without duodenal tumors had a positive gradient in the IPDV or SPDV (not significant). CONCLUSIONS: Intraarterial secretin injection is more sensitive than PVS at localizing duodenal gastrinomas and should replace PVS in patients with ZES and occult tumors.

Effective regional therapy of experimental cancer with paralesional administration of tumour necrosis factor-alpha + interferon-gamma.

Systemically administered tumour necrosis factor (TNF) has anti-tumour effects in animal tumour models but its clinical application is limited by severe toxicity. Interferon-gamma(IFN-gamma) has been shown to augment the anti-tumour effect of TNF. We evaluated the effect of paralesional (p.I.) injections of TNF plus IFN-gamma in a murine tumour model and compared the toxicity and anti-tumour effect with that seen with systemic administration. C57BL6 mice with 10-day subcutaneous MCA sarcomas were treated with daily p.I. injections of recombinant huTNF +/- IFN-gamma for 5 days. Optimal mean survival and 30-day cure rate was seen with doses of 5 micrograms TNF-alpha + 5000 U IFN-gamma (P < 0.05 vs. control or IFN-gamma alone). Tumour response after a single i.v. injection of 0-15 micrograms TNF + 5000 U IFN-gamma was then compared with five daily p.I. injections of the same dose of TNF-alpha and IFN-gamma. All animals with p.I. injections of > 5 micrograms TNF had initial complete necrosis of tumour with a variable degree of surrounding tissue necrosis, with rapid regrowth of tumour seen in some animals. Although treatment-related mortality was similar between i.v. and p.I. therapy, there was a higher percentage of animals cured with p.I. injections with overall cure rates in treated animals at 30 days of 17% vs. 72% (i.v. vs. p.I., P < 0.01) and 13% vs. 67% (P < 0.04) in a repeat study. 2+ clinical applications.

Tumour necrosis factor mediates the survival benefit of interleukin 2 in a murine pulmonary metastases model.

Antibody to tumour necrosis factor (TNF Ab) markedly decreases the toxicity of systemic interleukin-2 (IL-2) in mice but does not completely abrogate the anti-tumour response in terms of number of pulmonary metastases. Experiments were performed with a murine model of pulmonary metastases treated with high-dose IL-2 and concomitant TNF Ab or control antibody (CON Ab) to determine the effects of TNF Ab on survival. Mice were given either equal doses of IL-2 and TNF Ab or CON Ab or equitoxic doses of IL-2. In four consecutive experiments mice given TNF Ab tolerated 5 to 6 additional IL-2 doses (a 40-60% increase in total doses) in the equitoxic IL-2 dose group compared to the maximally tolerated dose with CON Ab. In all four experiments TNF Ab-treated mice had decreased survival compared to the CON Ab group given equal doses of IL-2 and in two of four experiments this difference was statistically significant (P2 < 0.01). Mice given 40-60% additional doses of IL-2 with TNF Ab had no improvement in survival compared with equitoxic doses of IL-2 with CON Ab in three of four experiments (P2 = 0.32, P2 = 0.67, P2 = 0.69). The TNF Ab preparation had no direct inhibition of IL-2 activity in an in vitro IL-2 proliferation bioassay. TNF Ab consistently blocks IL-2 toxicity and it also abrogates IL-2 therapeutic efficacy such that survival parallels treatment toxicity in this experimental model.

Location, incidence, and malignant potential of duodenal gastrinomas.

Duodenal gastrinomas are increasingly found at surgery, yet information about their location and characteristics is based on the results of either pooled series or retrospective reviews of small numbers of selected cases. To address these issues we have analyzed the location, incidence, and malignant potential of duodenal gastrinomas in 65 consecutive patients who underwent removal of all tumor as part of a 10-year prospective study to resect gastrinomas in patients with sporadic Zollinger-Ellison syndrome. The primary gastrinoma was located in the duodenum in 24 patients (37%). There were 19 men and five women aged 32 to 69 years (mean 49.4 years), with symptoms for 0.6 to 35 years (mean 7.9 years). Preoperative studies included serum gastrin levels of 114 to 35,798 pg/ml (mean 2060 pg/ml), basal acid output of 7 to 95 mEq/hr (mean 37.6 mEq/hr), and a positive secretin test result in 22 patients. Preoperative imaging studies identified tumor in the duodenal area in 11 patients (46%), but most positive imaging findings were metastatic gastrinoma in lymph nodes, and the primary duodenal tumor itself was identified in only two patients. Portal venous sampling had a localizing gastrin gradient in the inferior or superior pancreaticoduodenal vein in 17 of 23 patients (74%). Each of the 24 patients had a single, small duodenal wall tumor of 2.8 to 10.1 mm diameter (mean 6 mm). Each tumor stained positive for gastrin by immunohistochemistry. Seventeen tumors (71%) were located in the first portion of the duodenum, five (21%) in the second, and two (8%) in the third. Each tumor originated in the submucosa, and 13 (54%) were limited to the submucosa, whereas 11 (46%) were locally invasive, four (16%) extending into the muscularis mucosa and seven (29%) into the muscularis propria. Thirteen patients (54%) had spread to regional lymph nodes, whereas two (8%) had liver metastases. Lymph node metastases were seen with larger duodenal tumors (mean 7.1 vs 5.4 mm; p less than 0.01). The data suggest that a single duodenal wall gastrinoma is a common cause of Zollinger-Ellison syndrome (37%). These small (less than 1 cm) tumors are located in the submucosal layer of the proximal duodenum (92%) and are malignant more often than previously thought (54%).

Effects of verapamil and degradable starch microspheres during hepatic artery infusion of doxorubicin.

Chemotherapeutic tumoricidal activity may be enhanced by altering regional blood flow between host organ and tumor by transient embolization with degradable starch microspheres (DSMs) or by altering cellular drug transport with drugs such as verapamil. This study evaluated the effects of intraarterial verapamil and DSMs on hemodynamics and doxorubicin tissue levels. Rabbits (n = 34) with hepatic VX-2 carcinomas underwent hepatic artery infusion of doxurubicin (3 mg/kg) alone, with two doses of verapamil, and with verapamil plus DSMs. Blood pressure and heart rate were monitored, and heart, liver, and tumor tissue was obtained after 30 or 90 minutes. After verapamil (2 mg/kg) alone, mean heart rate decreased but infusion of verapamil plus DSMs resulted in only a 14% decrease in mean heart rate. Thirty minutes after drug infusions, mean hepatic tissue levels of doxorubicin were increased significantly by 2 mg/kg of verapamil compared with results with 1 mg/kg of verapamil and those in control groups (p less than 0.05). Mean tumor doxorubicin levels were not significantly different with verapamil alone. Verapamil plus DSMs or DSMs alone resulted in significantly lower mean hepatic and myocardial levels and increased mean tumor/liver ratios of doxorubicin after 90 minutes, compared with results in control or verapamil-alone groups. These results suggest that hepatic tumor drug levels are more affected by changes in regional blood flow with DSMs than by cellular drug transport changes caused by calcium channel blockers such as verapamil.

Regional hepatic arterial infusion of degradable starch microspheres increases fluorodeoxyuridine (FUdR) tumor uptake.

Hepatic metastases of colorectal carcinoma demonstrate a dose response to chemotherapy. In animal studies hepatic arterial infusion of chemotherapeutic agents with degradable starch microspheres (DSMs) produces a redistribution of blood flow between tumor and liver and an increase in tumor drug levels. In this prospective clinical study in patients with colorectal metastases, we evaluated the effect of DSMs on liver and tumor levels of fluoropyrimidines after intraoperative administration through the hepatic artery. Fourteen patients underwent infusion of radiolabeled fluorodeoxyuridine, 14C-FUdR (0.15 mg/kg, 0.5 microCi/kg), followed 2 to 5 minutes later by infusion of 3H-FUdR (0.15 mg/kg, 1.0 microCi/kg) without (n = 3) or with (n = 11) DMS. Seven of the later patients underwent major hepatic resection and tissue mapping of drug distribution, and four patients underwent biopsy procedures to remove specimens of liver and tumor 5 minutes after microsphere infusion. Administration of DSMs with FUdR increased tumor drug levels as measured by 3H-FUdR (5.9 +/- 4.4 vs 17.1 +/- 9.4 nmol/gm, p = 0.07) without altering hepatic drug levels (35.7 +/- 10.9 vs 30.2 +/- 20.9 nmol/gm, p = NS) and significantly increased the tumor/liver drug ratio of tritiated fluoropyrimidines (0.16 +/- 0.09 to 0.63 +/- 0.13, p = 0.03). Fluoropyrimidine levels in tumor and liver correlated with blood flow as measured by technetium-99m macroaggregated albumin retention. Thus, hepatic arterial administration of DSMs in human beings enhances tumor FUdR levels and may be useful in increasing tumor cytotoxicity and decreasing systemic toxicity during regional hepatic infusion.

Arginine, protein malnutrition, and cancer.

The amino acid arginine has anabolic and immunostimulatory properties. This study evaluated the potency of arginine in limiting the severe nutritional and immunological insults of protein calorie malnutrition and increasing tumor load. In protein-depleted A/J mice (n = 340) bearing either an immunogenic (C1300) or poorly immunogenic (TBJ) neuroblastoma, arginine supplementation [1%] significantly augmented T lymphocyte responses (mitogenesis, interleukin-2 production) compared with both a glycine-supplemented and nonsupplemented group. Arginine supplementation significantly retarded the growth of C1300 and prolonged median host survival. These results correlated with augmented autologous mixed lymphocyte tumor cell responses and enhanced specific cytotoxicity. This anti-tumor effect was not demonstrated in mice bearing TBJ where both arginine and glycine stimulated tumor growth compared with nonsupplemented mice. There was no significant difference between arginine and glycine in preservation of carcass weight. These studies suggest that the immunostimulatory effects of arginine are not due to supplemental nitrogen and that an associated antitumor effect is dependent on tumor antigenicity.

The use of quantitative perfusion fluorometry to measure relative tumor and liver blood flow after transient microembolization.

Optimal chemotherapy delivery to the tumor depends on regional drug concentration, tumor perfusion, tissue drug uptake, and metabolism. Modulation of tumor blood flow has been used to improve tumor response to treatment. Transient microembolization is one method to alter regional blood flow, but its effects on relative changes in tumor and liver blood flow have not been previously measured. This study used quantitative perfusion fluorometry (QPF) to evaluate blood flow distribution in liver and tumor before and after hepatic arterial infusion of degradable starch microspheres (DSMs) in 10 New Zealand white rabbits. QPF was compared with radioactive xenon-133 washout, an established method for measuring blood flow. Xenon-133 was injected intraparenchymally and the clearance rate was measured allowing calculation of relative blood flow. QPF was then used to measure liver and tumor blood flow in a hepatic VX-2 tumor model after hepatic artery injection of DSMs. Initial tumor blood flow was 55% of liver flow. DSMs produced a significant and transient decrease in hepatic blood flow that was decreased to 40% of baseline after 25 min. Changes in relative hepatic blood flow after DSMs as measured by QPF correlated strongly with results obtained by xenon-133 washout (R = 0.97, P less than 0.01). Fluorometry's simplicity and reliability may be clinically useful to evaluate tumor blood flow characteristics.

Modulation of doxorubicin uptake by epinephrine in the VX-2 carcinoma.

The purpose of regional chemotherapy is to deliver maximal drug concentrations to localized unresectable tumors, to minimize systemic toxicity, and to improve tumor response to treatment. Manipulation of regional blood flow by vasoactive agents could further increase tumor drug levels. This study evaluates the effect of hepatic artery infusion of epinephrine on tumor and liver doxorubicin uptake. New Zealand White rabbits (n = 17) were implanted with hepatic VX-2 carcinomas which were allowed to grow to 2-3 cm in diameter. Doxorubicin (3 mg/kg) +/- [14C]doxorubicin (2.5 microCi) was given alone (controls) or following low- (5 micrograms) or high- (15-50 micrograms) dose epinephrine infusion into the hepatic artery. Heart, liver, and tumor were obtained 30 min after infusion for determination of doxorubicin concentration. The doxorubicin liver concentrations were similar in all three groups but tumor levels decreased from 26.6 +/- 16.3 to 6.7 +/- 2.8 and 6.6 +/- 4.3 nmole/g in the epinephrine groups. Myocardial levels were reduced by the high-dose epinephrine (6.8 +/- 2.2 nmole/g) when compared to controls (11.3 +/- 2.1 nmole/g) or low-dose epinephrine (11.5 +/- 1.8 nmole/g). The liver concentration of doxorubicin was not significantly affected by intraarterial epinephrine. The myocardial doxorubicin concentration was affected only by epinephrine doses sufficient to cause severe systemic vasoconstriction. The decrease in tumor doxorubicin concentration to low-dose (5 micrograms) and high-dose (15-50 micrograms) epinephrine suggests that intraarterial epinephrine limits doxorubicin uptake by tumor.