Preoperative radiotherapy in the management of retroperitoneal liposarcoma.

BACKGROUND: Histological subtype influences both prognosis and patterns of treatment failure in retroperitoneal sarcoma. Previous studies on the efficacy of neoadjuvant radiotherapy (NRT) have incorporated multiple histological types with heterogeneous tumour biology. The survival impact of NRT specifically for patients with retroperitoneal liposarcoma is poorly defined. METHODS: Patients who underwent resection with curative intent for retroperitoneal liposarcoma and who received NRT or surgery alone were identified in the US National Cancer Data Base (2004-2013). Cox regression was used to identify co-variables associated with overall survival. NRT and surgery-alone cohorts were matched 1 : 1 by propensity scores based on the survival hazard on Cox modelling. Overall survival was compared by Kaplan-Meier estimates. RESULTS: A total of 2082 patients with retroperitoneal liposarcoma were identified; 1908 underwent surgery alone and 174 received NRT before surgical resection. Median tumour size was 22·0 cm and 34·9 per cent of tumours were high grade. In the unmatched cohort, NRT was not associated with improved overall survival (χ2 = 3·49, P = 0·062). In the propensity score-matched cohort, NRT was associated with an improvement in survival (median overall survival 129·2 versus 84·3 months; P = 0·046; hazard ratio (HR) 1·54, 95 per cent c.i. 1·01 to 2·36). This effect appeared most pronounced for tumours with adjacent organ invasion (median overall survival not reached versus 63·8 months; P = 0·044; HR 1·79, 1·01 to 3·19). CONCLUSION: NRT improved survival in patients undergoing surgery for retroperitoneal liposarcoma, particularly those with high-risk pathological features.

Passive immunization against tumor necrosis factor partially abrogates interleukin 2 toxicity.

Passive immunization against TNF allowed non-tumor-bearing C3H/HEN mice and tumor-bearing C57BL/6 mice to tolerate significantly more doses of IL-2 before death (p less than 0.005 and p less than 0.001, respectively). The antitumor effect of IL-2 against both 3-d and 10-d pulmonary metastases was maintained in mice treated concurrently with neutralizing antibodies to TNF. In one experiment with 10-d pulmonary metastases, increased administration of IL-2 made possible by passive immunization against TNF significantly improved the antitumor response compared to equitoxic doses of IL-2 and control antibody. The results indicate that TNF is a mediator of IL-2 toxicity but contributes minimally to the antitumor effects of IL-2. Strategies to inhibit TNF may improve the therapeutic index of IL-2 as a neoplastic agent.

Tolerance to tumor necrosis factor in rats and the relationship to endotoxin tolerance and toxicity.

Treatment of rats with recombinant human TNF initially causes a marked decrease in food intake, a loss of body weight, and a negative nitrogen balance. These alterations normalize with continued twice daily intraperitoneal injections of the same dose. Rats tolerized to TNF in this manner are refractory to a lethal dose of TNF. Also, TNF-pretreated and -tolerized rats have prolonged survival and reversed histopathologic changes after injection of a lethal dose of endotoxin compared with control animals. The TNF-tolerant state is dependent on the dose of TNF used and the length of TNF pretreatment. TNF-induced tolerance is relatively short lived, being present 2-4 d after TNF pretreatment and dissipating by 2 wk. Rats made tolerant to endotoxin are also tolerant to a lethal dose of TNF. A bidirectional crossreacting tolerance exists between TNF and endotoxin. The mechanism of TNF tolerance is unclear, but it does not appear to be due to a humoral immune response or a perturbation of the uptake and clearance of injected TNF.

Passive immunization of mice against D factor blocks lethality and cytokine release during endotoxemia.

D factor, also known as leukemia inhibitory factor, is a pleiotropic cytokine whose role during acute injury and inflammation is not known. Intraperitoneal administration of Escherichia coli endotoxin induced D factor gene expression in mice, and passive immunization against D factor protected them from the lethal effects of endotoxin and blocked endotoxin-induced increases in serum levels of interleukin 1 and 6. Peak levels of tumor necrosis factor and interferon gamma were not affected. These results indicate that D factor is an essential early mediator of the inflammatory cytokine response and therefore may be important in the pathogenesis of the many inflammatory conditions, such as sepsis, arthritis, allograft rejection, and cancer immunotherapy.

Differentiation factor/leukemia inhibitory factor protection against lethal endotoxemia in mice: synergistic effect with interleukin 1 and tumor necrosis factor.

Differentiation factor (D factor), also called leukemia inhibitory factor (LIF), is a glycoprotein that has been increasingly recognized to possess a wide range of physiological activities. We examined the possibility that the administration of D factor may confer beneficial effects and enhance host resistance against lethal endotoxemia. A single intravenous dose of recombinant human D factor completely protected C57/Bl6 mice from the lethal effect of Escherichia coli endotoxin (lipopolysaccharide [LPS]). The protective effects were dose dependent and observed when administered 2-24 h before LPS. Previous work has shown that interleukin 1 (IL-1) and tumor necrosis factor (TNF) also protect against a subsequent LPS challenge in a dose-dependent manner. When human D factor was combined with sub-protective doses of IL-1 beta or TNF-alpha, there was dramatic synergistic protection against a subsequent lethal LPS challenge.

Multiple cytokines and acute inflammation raise mouse leptin levels: potential role in inflammatory anorexia.

Several inflammatory cytokines, most notably tumor necrosis factor (TNF) and IL-1, induce anorexia and loss of lean body mass, common manifestations of acute and chronic inflammatory conditions. In C57BL/6 female mice, the administration of TNF, IL-1, and, to a lesser extent, leukemia inhibitory factor (LIF), produced a prompt and dose-dependent increase in serum leptin levels and leptin mRNA expression in fat. IL-10, IL-4, ciliary neurotrophic factor, and IL-2, cytokines not known to induce anorexia or decrease food intake, had no effect on leptin gene expression or serum leptin levels. After administration of Escherichia coli lipopolysaccharide (LPS), leptin gene expression and leptin levels were increased. These findings suggest that leptin levels may be one mechanism by which anorexia is induced during acute inflammatory conditions.

Treatment with recombinant human tumor necrosis factor-alpha protects rats against the lethality, hypotension, and hypothermia of gram-negative sepsis.

Tumor necrosis factor (TNF) is a peptide secreted by macrophages in response to endotoxin that can produce many of the changes seen in septic shock. After cecal ligation and puncture (CLP) rats gradually develop tachycardia, hypotension, tachypnea, and hypothermia. At 5 h post-CLP, rats have a peak in serum levels of endotoxin and 60% of rats have blood cultures that grow Gram-negative rods (Escherichia coli and Klebsiella pneumonia). At 20 h post-CLP all rats develop positive blood cultures. Serum levels of TNF are not reproducibly measurable in rats following CLP. Rats undergoing CLP have a 50-80% mortality with deaths usually occurring 24-72 h postinjury. Repetitive (twice daily x 6 d) i.p. injection of sublethal doses of recombinant human TNF-alpha (100 micrograms/kg) to rats undergoing CLP 1 d after the treatment period resulted in a significant reduction in mortality compared to control rats previously unexposed to rTNF (P less than 0.03). Animals treated with rTNF had no hypotension or hypothermia after CLP and regained normal food intake faster than control rats. 12 h after CLP the gene expression for manganous superoxide dismutase (MnSOD), an inducible mitochondrial metalloenzyme responsible for cellular resistance to injury from toxic reactive oxygen species, was higher in livers of rats treated with rTNF suggesting that the TNF treatment augmented expression of this protective enzyme. Unlike MnSOD, expression of the gene for copper-zinc SOD was not affected by CLP or rTNF treatment. The results suggest that prior treatment with recombinant TNF can ameliorate the lethality, hypotension, hypothermia, and anorexia of Gram-negative sepsis in rats and that the mechanism may be related to enhanced hepatic expression of the gene for MnSOD. Repeated administration of recombinant TNF may be a strategy to minimize mortality and morbidity of Gram-negative sepsis.

Systemic IFN-beta gene therapy results in long-term survival in mice with established colorectal liver metastases.

Most patients succumbing to colorectal cancer fail with liver-predominant metastases. To make a clinical impact in this disease, a systemic or whole-liver therapy may be required, whereas most cancer gene therapy approaches are limited in their ability to treat beyond local disease. As a preclinical model for cancer gene therapy, recombinant adenovirus containing the human IFN-beta (hIFN-beta) cDNA was delivered systemically in nude mouse xenograft models of human colorectal cancer liver metastases. The vector targeted hepatocytes that produced high levels of hIFN-beta in the liver, resulting in a profound apoptotic response in the tumors and significant tumor regression. hIFN-beta gene therapy not only resulted in improved survival and long-term cure in a micrometastatic model, but provided similar benefits in a clinically relevant gross disease model. A similar recombinant adenovirus containing the murine IFN-beta (mIFN-beta) cDNA also resulted in a therapeutic response and improved survival in syngeneic mouse models of colorectal cancer liver metastases. Depletion studies demonstrate a contribution of natural killer cells to this therapeutic response. The toxicity of an adenoviral vector expressing murine IFN-beta in a syngeneic model is also presented. These encouraging results warrant further investigation of the use of cancer gene therapy for targeting metastatic disease.

Impact of tolerance on antitumor efficacy of tumor necrosis factor in mice.

Repetitive sublethal doses of tumor necrosis factor (TNF) can induce tolerance or tachyphylaxis to the toxic effects of TNF. Because tumor-bearing (TB) mice are more sensitive to the toxic effects of TNF, this study investigates whether similar tolerance occurs in TB mice and whether it affects the antitumor response of TNF. Nontumor-bearing C3H/Hen mice were treated with twice daily i.p. sublethal escalating doses of human recombinant TNF (2, 2, 3, 3, 4, and 4 micrograms i.p. every 12 h for 6 days) and were challenged 2 days later with a lethal i.v. dose (40 micrograms) of TNF. TNF-pretreated mice had 100% survival as compared to 0% survival in control mice previously treated with saline (P less than 0.01). Tumor-bearing C57BL/6 mice bearing an MCA-106 or MCA-102 sarcoma were treated with an identical TNF-tolerizing regimen (2, 2, 3, 3, 4, and 4 micrograms i.p. every 12 h for 6 days) beginning 3 days following tumor inoculation and were similarly more resistant to a subsequent 100% lethal i.v. treatment dose of TNF than control TB mice. A significantly greater percentage of TNF-pretreated mice bearing the MCA-106 sarcoma survived treatment doses of 8, 12, and 16 micrograms of TNF i.v. than control TB mice. Similarly, a significantly greater percentage of TNF-pretreated mice bearing the MCA-102 sarcoma survived treatment doses of 6 and 9 micrograms of TNF i.v. than control TB mice. However, the ability to administer higher doses of TNF i.v. to TNF-pretreated TB mice did not improve therapeutic efficacy. In mice bearing the MCA-106 tumor the most efficacious treatment responses were seen in animals that were previously naive to TNF, and treatment toxicity (lethality) correlated directly with antitumor efficacy such that larger treatment doses of TNF in tolerant mice resulted in similar antitumor effects as smaller treatment doses in control TB mice. In mice bearing the MCA-102 tumor, equitoxic treatment doses of TNF produced similar antitumor effects in both control and tolerant TB mice. There were no differences in cure rate for TNF-tolerant or control TB mice bearing either tumor. The results suggest that TNF tolerance occurs in TB mice and reduces the toxicity as well as the therapeutic efficacy of TNF.

Cachectin/tumor necrosis factor: a possible mediator of cancer anorexia in the rat.

Cachectin/tumor necrosis factor (TNF) is a macrophage product which may have a role in cancer cachexia. Recombinant human cachectin/TNF (Cetus Corporation) was administered i.p. twice daily to male F344 rats at varying, nonlethal dosages for either 5 or 10 days, and daily rat food intake and body weight were measured. There was a dose-dependent cachectin/TNF-induced decline in food intake and body weight gain over the treatment period. However, after 1 day rats became tolerant to these effects and increased food intake and gained body weight despite receiving cachectin/TNF. Rats were subsequently inoculated with a transplantable methylcholanthrene-induced sarcoma, and survival was measured. Rats previously treated with high-dose (either 100 or 200 micrograms/kg/day) cachectin/TNF survived significantly longer following tumor inoculation than did control rats given saline or rats given 10 micrograms/kg/day of cachectin/TNF. Analysis of tumor growth curves and tumor weight indicated that high-dose cachectin pretreatment did not retard tumor growth. Analysis of food intake and tumor burden following tumor inoculation indicated that high-dose cachectin pretreatment decreased the reduction in food intake associated with progressive tumor growth and allowed rats to withstand a greater tumor burden at death. Rats immunized with low-dose human cachectin/TNF developed high IgG titers against human TNF, but failed to demonstrate the same protection against a methylcholanthrene-induced tumor challenge as rats made tolerant with repetitive twice daily high-dose cachectin/TNF. The observation of reduced cancer-associated anorexia and increased survival of tumor-bearing rats associated with previous tolerance to exogenous cachectin/TNF strengthens the contention that endogenously produced cachectin may be a factor in the pathogenesis of cancer anorexia in the tumor-bearing rat. The mechanism of this tolerance is unclear but does not appear to be a humoral immune response.

The roles of gamma-interferon and tumor necrosis factor alpha in an experimental rat model of cancer cachexia.

Administration of repetitive sublethal doses of either recombinant human tumor necrosis factor (TNF) or recombinant murine gamma-interferon (IFN) to non-tumor-bearing (NTB) rats caused a significant decline in food intake and body weight. After 3 days rats became resistant to the anorectic and weight loss effects of TNF but maintained persistent diminished food intake and diminished body weight gain while receiving recombinant murine IFN. Passive immunization against recombinant rat gamma-interferon allowed cachectic tumor-bearing (TB) rats to eat more food, have a lesser decline in body weight, live longer, and tolerate larger tumors than similar TB rats given nonspecific control antibody. TB rats treated with an antisera to recombinant murine TNF, which was 100% protective when given to NTB rats 6 h before a lethal endotoxin challenge, did not differ significantly from TB rats treated with control antibody with respect to food intake, body weight, survival, or tumor size. Serum levels of TNF or IFN were not detectable in cachectic tumor-bearing rats. The data indicate that the administration of exogenous IFN can simulate cachexia in NTB rats and that passive immunization against it can partially reverse the cachectic changes associated with cancer and prolong survival. These findings suggest that gamma-interferon may be an important mediator of cachexia in this rat tumor model.

Photofrin-mediated photodynamic therapy induces vascular occlusion and apoptosis in a human sarcoma xenograft model.

Photodynamic therapy (PDT) involves light activation of a photosensitizer, resulting in oxygen-dependent, free radical-mediated cell death. Little is known about the efficacy of PDT in treating human sarcomas, despite an ongoing clinical trial treating i.p. sarcomatosis. The present study evaluates PDT treatment of a human sarcoma xenograft in nude mice and explores the mechanism of PDT-mediated antitumor effect. Athymic nude mice, 6-8 weeks of age, were s.c. injected with 5 x 10(6) cells of the A673 human sarcoma cell line. Tumors were allowed to grow to a diameter of about 10 mm. Photofrin (PF), 10 mg/kg, was injected by tail vein, and 24 h later, 630 nm light was delivered to the tumor with fluences of 50, 100, 150, or 300 J/cm2 at a fluence rate of 250 mW/cm2. To assess the efficacy of PDT in the treatment of sarcomas, photosensitizer uptake/retention studies and dose-response studies were performed. Studies carried out to determine the mechanism of tumor response included tumor temperature measurements before, during, and after treatment; tumor vascular perfusion studies with laser Doppler; electron microscopic analysis of tumor sections for vascular occlusion; and analysis of tumor cryosections for endothelial cell damage, apoptosis, and necrosis. At all time points of analysis, photosensitizer levels were greater in tumor than in muscle. Dose-response studies showed that at 100 J/cm2, five of six mice had a complete response to treatment, one of six had a partial response, and no deaths occurred. Temperature measurements indicated that thermal injury did not contribute to tumor response. Vascular perfusion studies demonstrated a significant reduction in blood flow as early as 6 h after PDT. Electron micrographs revealed erythrostasis in tumor microvessels starting as early as 2 h after treatment and complete occlusion of blood vessels by 12 h. Starting as early as 4 h after PDT, apoptosis first appeared in endothelial cells lining the occluded blood vessels and became more widespread at later time points. PDT is an effective treatment for this human sarcoma xenograft in nude mice. The mechanism of tumor destruction in this model appears to be vascular damage with initial apoptosis in tumor endothelial cells and delayed tumor cell apoptosis. This therapy may be valuable in the treatment of patients with sarcomatosis.

Prolonged survival of tumor-bearing rats with repetitive low-dose recombinant tumor necrosis factor.

Tumor necrosis factor may be a mediator of the syndrome of cancer cachexia. Tachyphylaxis or tolerance to the cachectic effects of recombinant tumor necrosis factor (rTNF) has been previously described. In this study, we investigate whether repetitive exposure to rTNF can induce similar tolerance in tumor-bearing (TB) rats and ameliorate cachexia induced by the tumor. In experiment 1, non-tumor-bearing (NTB) and TB rats were randomized to either escalating low doses of rTNF or saline i.p. twice daily for 9 consecutive days. NTB rats treated with rTNF demonstrated a significant decline in food intake and weight change (P less than 0.00001) but soon developed tolerance to the cachectic effects of rTNF; they consumed significantly more food than on the first day of treatment and had weight change similar to NTB rats treated with saline. TB rats treated with rTNF showed a similar significant decline in food intake and weight change (P less than 0.0001) and also demonstrated similar tolerance to the cachectic effects of rTNF with continued treatment. Following treatment, TB rats that had been treated with rTNF ate significantly more and lost less weight than TB rats that had been treated with saline (P less than 0.00001). rTNF treatment of TB rats also demonstrated antineoplastic activity, as estimated tumor weight of tumors from rats treated with rTNF were significantly less than controls (P = 0.003). The anticachexia and antineoplastic effects of rTNF resulted in prolonged survival of TB rats treated with rTNF compared to control TB rats (P = 0.015). Experiment 2 utilized two different rTNF treatment regimens in TB rats: one group received 12 days of escalating doses of rTNF, and another group received 15 days of rTNF treatment. TB rats treated with rTNF again had a significantly greater food intake (P less than 0.00001) and delayed weight loss (P = 0.0001) posttreatment that was further augmented by additional doses of rTNF. Antineoplastic activity of rTNF was less clear, and overall tumor growth curves were not affected by rTNF treatment. Survival of TB rats treated with rTNF was again significantly increased in a dose-dependent manner (P = 0.006). Repeated administration of low doses of rTNF to TB rats induces mild reduction in tumor growth, tolerance to the cachectic effects of rTNF that results in tolerance to the cachectic effects of tumor, and prolongation of survival.

Cachectic effects of recombinant human tumor necrosis factor in rats.

Treatment of rats with either intermittent bolus i.v. injections or continuous i.v. infusions of the same sublethal daily dose of tumor necrosis factor (TNF) results in decreased food intake and decreased nitrogen balance compared to saline-treated control rats. After 4 days of treatment, rats treated with intermittent bolus doses of TNF develop tolerance to the nutritional effects and consume normal amounts of food and have nitrogen balance similar to those of saline treated rats. Rats receiving the continuous infusion of TNF do not. Rats treated with both routes of TNF lose more weight than pair fed rats who eat the same mean amount as the continuous TNF treated group. In addition, 56% of rats receiving continuous infusion TNF die during the 8-day experimental period while rats receiving either intermittent bolus TNF or similar food intake (pair fed) do not. Body composition studies of rats that completed the 8 days of treatment indicate that rats receiving either continuous infusion or intermittent bolus TNF have increased percentages of body water and reduced percentages of body solid compared to saline treated control rats. Rats pair fed to the food intake of continuous TNF treated rats also had increased percentages of body water and reduced percentages of body solid, but changes were significantly less than those observed in continuous TNF infused rats. Continuous TNF infusion reduced total body nitrogen and potassium while pair feeding did not reduce potassium and reduced nitrogen to a lesser degree. Pair feeding and continuous TNF infusion reduced total body fat to a similar extent. Twice a day administration of TNF resulted in lesser changes in carcass water, solid, nitrogen, lipid, and potassium than continuous infusion of the same dose of TNF. The results indicate that continuous infusion of TNF can produce anorexia, weight loss, edema, loss of body protein, lipid and cell mass, and lethality which is markedly ameliorated with bolus doses of TNF. The findings are consistent with the hypothesis that slow continuous secretion of sublethal amounts of TNF may mediate cancer cachexia.

Lymph node evaluation and survival after curative-intent resection of duodenal adenocarcinoma: A matched cohort study.

BACKGROUND: Lymph node (LN) metastasis in patients with duodenal adenocarcinoma is associated with poor prognosis; however, the optimal extent of LN assessment and the interaction between LN assessment and adjuvant systemic therapy is poorly understood. METHODS: Resected non-metastatic duodenal adenocarcinoma patients (n = 1743) were identified in the National Cancer Database (1998-2011). Logistic regression analysis identified covariates associated with LN metastasis. The influence of increasing LN cut-off points on overall survival (OS) was analysed using the log-rank test and Cox proportional hazards modelling. Adjuvant chemotherapy (AC) and surgery alone cohorts were matched (1:1) by propensity scores based on the likelihood of nodal metastasis or survival hazard on Cox modelling. OS in the matched cohort was compared by Kaplan-Meier estimates. RESULTS: LN metastases were present in 865 (49.6%) patients. Increasing LN assessment was associated with an increased likelihood of nodal involvement (P = 0.008). In node-negative patients, increasing LN assessment was associated with a decreased risk of death, with the largest actuarial survival differences observed for ≥15 LN (hazard ratio [HR] 0.63, 95% confidence interval [CI] 0.48-0.82, P = 0.001). In the propensity score-matched cohort of node-negative patients, AC was associated with non-significant improvements in 5-year actuarial (66.1% versus 58.7%, HR 0.79, 95% CI 0.53-1.18, P = 0.249), and did not vary by adequacy of LN counts (<15 LNs: HR 0.79, 95% CI 0.51-1.24, P = 0.305; ≥15 LNs: HR 0.90, 95% CI 0.35-2.30, P = 0.900). CONCLUSIONS: The extent of LN identification has prognostic significance in resected node-negative duodenal adenocarcinoma, but cannot be implicated in the selection of node-negative patients for AC.

Treatment of patients with melanoma of the extremity using hyperthermic isolated limb perfusion with melphalan, tumor necrosis factor, and interferon gamma: results of a tumor necrosis factor dose-escalation study.

PURPOSE: To evaluate response rates and systemic and regional toxicity of hyperthermic isolated limb perfusion (ILP) for treatment of in-transit metastases of extremity melanoma using escalating-dose tumor necrosis factor (TNF) in conjunction with melphalan and interferon gamma (IFN). PATIENTS AND METHODS: All patients received IFN 0.2 mg2 for 2 days followed by a 90-minute ILP with TNF and IFN (0.2 mg) given at time 0 and melphalan (10 mg/L limb volume) given at 30 minutes. Twenty-six patients were treated with 4 mg of TNF and 12 patients received 6 mg of TNF. All patients had assessable disease in the perfusion field and all but two patients were assessable for response at 1 month after treatment. RESULTS: Mean peak perfusate TNF levels in the 4-mg group were 4.8 micrograms/mL, compared with 7.4 micrograms/mL for the 6-mg group (P = .03). The complete response rate in the 4-mg TNF group was 76%, with an overall objective response rate of 92%, compared with 36% and 100% for the 6-mg group. Subgroup analyses showed that the lower complete response rate in the 6-mg TNF group was not explained by differences in disease burden or prior regional therapy. Systemic drug toxicity was short-lived, easily managed, and related to perfusate leak more than to TNF perfusate dose. Regional toxicity, particularly painful myopathy and neuropathy, was greater with the 6-mg dose level and was considered dose-limiting. CONCLUSION: ILP with 4 mg TNF, IFN, and melphalan can lead to complete local responses in the majority of patients with extremity melanoma. Escalating the TNF dose to 6 mg did not increase the complete response rate and increased regional toxicity.

Isolated hepatic perfusion with tumor necrosis factor and melphalan for unresectable cancers confined to the liver.

PURPOSE: To evaluate the efficacy and systemic and regional toxicities of hyperthermic isolated hepatic perfusion (IHP) using tumor necrosis factor (TNF) and melphalan for the treatment of unresectable primary or metastatic cancers confined to the liver. PATIENTS AND METHODS: Thirty-four patients (18 men and 16 women; mean age, 49 years) underwent a 60-minute hyperthermic (39.5 degrees to 40.0 degrees C) IHP performed by laparotomy that used TNF 1.0 mg and melphalan 1.5 mg/kg. Perfusion inflow was through the gastroduodenal artery and outflow was from a cannula positioned in an isolated segment of retrohepatic inferior vena cava (IVC). Infrahepatic IVC and portal venous blood flow were shunted to the axillary vein using an external venoveno bypass circuit. Complete vascular isolation of the liver was confirmed by an I-131-labelled human serum albumin monitoring technique. RESULTS: There was no operative mortality. Seventy-five percent of patients had reversible grade III or IV (National Cancer Institute Common Toxicity Criteria) hepatic toxicity with one treatment-related mortality (3%) because of hepatic venoocclusive disease. In 33 assessable patients, the overall response rate was 75% (complete response, one patient [3%]; partial response, 26 patients [72%]). With a median potential follow-up of 15 months, the mean duration of response was 9 months (range, 2 to 30 months). CONCLUSION: IHP with TNF and melphalan results in significant regression of bulky hepatic cancers confined to the liver in the majority of patients. Based on these initial results, further refinement of this treatment technique is warranted; perhaps by the combination of IHP with other regional treatment strategies to provide long-term control of unresectable cancers confined to liver.

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.

A phase I-II study of isolated hepatic perfusion using melphalan with or without tumor necrosis factor for patients with ocular melanoma metastatic to liver.

There are no satisfactory treatment options for patients with ocular melanoma metastatic to liver, and after liver metastases are identified, median survival is only between 2 and 7 months. Because liver metastases are the sole or life-limiting component of disease in the vast majority of patients who recur, we reasoned that complete vascular isolation and perfusion of the liver might result in clinically meaningful regression of disease. Between September 1994 and July 1999, 22 patients (13 women and 9 men; mean age, 49 years) with ocular melanoma metastatic to liver were treated with a 60-min hyperthermic isolated hepatic perfusion (IHP) using melphalan alone (1.5-2.5 mg/kg, n = 11) or with tumor necrosis factor (TNF, 1.0 mg, n = 11). Via a laparotomy, IHP inflow was via the hepatic artery alone (n = 17) or hepatic artery and portal vein (n = 5) and outflow from an isolated segment of inferior vena cava. Most patients had advanced tumor burden with a mean percentage of hepatic replacement of 25% (range, 10-75%) and a median number of metastatic nodules of 25 (range, 5 to >50). Complete vascular isolation was confirmed in all patients using a continuous intraoperative leak monitoring technique with 131I radiolabeled albumin. There was one treatment mortality (5%). The overall response rate in 21 patients was 62% including 2 radiographic complete responses (9.5%) and 11 partial responses (52%). The overall median duration of response was 9 months (range, 5-50) and was significantly longer in those treated with TNF than without (14 versus 6 months, respectively; P = 0.04). Overall median survival in 22 patients was 11 months. These data indicate that a single 60-min IHP can result in significant regression of advanced hepatic metastases from ocular melanoma. TNF appears to significantly prolong the duration of response.