Interleukin 7 generates antitumor cytotoxic T lymphocytes against murine sarcomas with efficacy in cellular adoptive immunotherapy.

Interleukin 7 (IL-7) is a 25-kD cytokine that was initially described as a pre-B cell growth factor. This cytokine has also been shown to have T cell proliferative and differentiation effects. In this report, we demonstrate that antitumor cytotoxic T lymphocytes (CTL) generated by secondary in vitro sensitization of draining lymph node cells in IL-7 are effective in treating 3-day syngeneic methylcholanthrene (MCA) sarcoma pulmonary metastases in mice. In vivo titrations comparing IL-7 to IL-2 antitumor CTL show that they have equivalent potency in adoptive immunotherapy. IL-7 antitumor CTL generated against MCA sarcomas of weak immunogeneity are also tumor specific in their in vivo efficacy. This study represents the first successful use of a cytokine other than IL-2 for the generation of cells with in vivo efficacy in cellular adoptive transfer.

Oncostatin M is a proinflammatory mediator. In vivo effects correlate with endothelial cell expression of inflammatory cytokines and adhesion molecules.

Oncostatin M is a member of the IL-6 family of cytokines that is primarily known for its effects on cell growth. Endothelial cells have an abundance of receptors for oncostatin M, and may be its primary target. We determined if oncostatin M induces a key endothelial cell function, initiation of the inflammatory response. We found that subcutaneous injection of oncostatin M in mice caused an acute inflammatory reaction. Oncostatin M in vitro stimulated: (a) polymorphonuclear leukocyte (PMN) transmigration through confluent monolayers of primary human endothelial cells; (b) biphasic PMN adhesion through rapid P-selectin expression, and delayed adhesion mediated by E-selectin synthesis; (c) intercellular adhesion molecule-1 and vascular cell adhesion molecule-1 accumulation; and (d) the expression of PMN activators IL-6, epithelial neutrophil activating peptide-78, growth-related cytokine alpha and growth-related cytokine beta without concomitant IL-8 synthesis. The nature of the response to oncostatin M varied with concentration, suggesting high and low affinity oncostatin M receptors independently stimulated specific responses. Immunohistochemistry showed that macrophage-like cells infiltrating human aortic aneurysms expressed oncostatin M, so it is present during a chronic inflammatory reaction. Therefore, oncostatin M, but not other IL-6 family members, fulfills Koch's postulates as an inflammatory mediator. Since its effects on endothelial cells differ significantly from established mediators like TNFalpha, it may uniquely contribute to the inflammatory cycle.

Disseminated human malignant melanoma in congenitally immune-deficient (bg/nu/xid) mice.

Congenitally immune-deficient bg/nu/xid (BNX) mice are severely compromised in their ability to mount T-cell, B-cell, and lymphokine-activated killer (LAK) cell responses. Successful engraftment of BNX mice with human hematopoietic stem cells has been demonstrated recently. We have investigated the potential use of BNX mice for studies relating to the biology and immunotherapy of human malignant melanoma. The intravenous injection of fresh single-cell suspensions of human malignant melanomas into mice resulted in widely disseminated disease. Metastatic spread of human melanoma in BNX mice mimicked that observed in patients: eg, there were numerous tumor nodules identified in the subcutaneous tissues as well as in a variety of visceral organs, including spleen, kidneys, thyroid, adrenals, lungs, heart, and brain. BNX mouse lymph nodes were replaced consistently by human malignant melanoma cells. The presence of human tumor cells in these mice was confirmed by histologic analysis and microcytofluorometry analyses using human melanoma-specific monoclonal antibodies (MAbs). Moreover, human melanoma cells passaged in BNX mice remained lysable in vitro by specifically cytolytic, autologous human tumor-infiltrating lymphocytes (TILs). The capacity of fresh human malignant melanoma to disseminate widely in BNX mice may prove valuable not only for study of the biology of metastatic spread but also for studies of the immunotherapy of human melanoma using melanoma-specific MAbs and chemotherapeutic agents, as well as human TILs and LAK cells with or without retrovirus-mediated gene transfer modification.

Enhanced tumor growth of both primary and established human and murine tumor cells in athymic mice after coinjection with Matrigel.

Previously we found that the reconstituted basement membrane matrix Matrigel, when premixed with human small-cell lung carcinoma cells and injected subcutaneously into athymic mice, permitted tumor growth, whereas cells injected in the absence of Matrigel did not form tumors. In the present study, we examined additional cell types and determined some of the underlying mechanisms involved in the promotion of tumor formation by Matrigel. The tumor cell lines that we studied included transformed mouse Englebreth-Holm-Swarm tumor cells (T-EHS), human submandibular carcinoma A253 cells, mouse melanoma B16F10 cells, human epidermoid carcinoma KB cells, and human primary renal cell carcinoma cells. When coinjected subcutaneously with Matrigel, these cell lines formed rapidly proliferating tumors. Primary biopsy specimens of human colon carcinoma, when dispersed and coinjected with Matrigel, also formed tumors. Only A253, KB, and B16F10 cells formed small tumors in the absence of Martrigel, but a fivefold to tenfold increase in tumor size was observed in the presence of Matrigel. These data demonstrate a useful method for improving the growth of human tumors in athymic mice.

The use of congenitally immunodeficient mice to study human tumor metastases and immunotherapy.

Congenitally immunodeficient strains of mice have proven valuable in the development of relevant models to study human tumor biology, metastases, and immunotherapy. Local invasion and extensive multiorgan metastases in athymic mice have been obtained following orthotopic implantation or onplantation of histologically intact fragments of human tumors. In C.B-17 severe combined immunodeficient (SCID) mice or in triple immunodeficient, beige/nude/xid (BNX) mice, the development and spread of inoculated human leukemia/lymphoma and/or melanoma have mimicked, in some cases, those observed in patients. Reports of reconstitution of SCID and BNX mice with human myeloid or lymphoid cells have suggested that these models might be useful for the study of human immune responses to autologous tumors in vivo. The severe immunocompromised status of these mice have also led to evaluations of the therapeutic efficacy of adoptively transferred, tumor-reactive human T cells. In this report, we review the pertinent information currently available on the use of congenitally immunodeficient mice in studies of human cancer biology and treatment.

The persistence of human peripheral lymphocytes, tumor infiltrating lymphocytes, and colon adenocarcinomas in immunodeficient mice.

The reconstitution of severely immunodeficient mice with human peripheral blood mononuclear cells (PBMCs) may represent a unique model system to evaluate human antitumor responses. To evaluate this possibility, we studied human PBMC reconstitution, human tumor infiltrating lymphocyte (TIL) persistence, and human colon adenocarcinoma propagation in beige/nude/xid (BNX) and in severe combined immunodeficient (SCID) mice. To evaluate human PBMC reconstitution, 75 mice received 1 x 10(7)-1 x 10(9) human PBMCs i.p. or i.v. and were studied at intervals ranging from 1 to 8 weeks by fluorescence-activated cell sorting (FACS) analysis and by measurement of circulating human immunoglobulin levels. By FACS analyses, only one of 75 mice had evidence of human PBMC persistence at 2 weeks in the spleen. Moreover, liver and peritoneum showed evidence of human cells in only 13 of 56 and 16 of 55 mice, respectively. In these mice, human cells comprised 1-77% of total cells recovered. Human immunoglobulin levels in mouse serum ranged from 0 to 34,000 micrograms/ml and correlated only weakly with evidence of human PBMC reconstitution in peripheral organs, but were generally higher in SCID mice than in BNX mice. Human TIL persistence was evaluated in BNX and SCID mice that were given 3 x 10(7) TILs i.v. (in divided doses) or 1 x 10(8) i.p. TILs along with interleukin-2 administration. At 1, 2, 7, and 14 days following TIL delivery, evidence of human TIL persistence in liver, lung, peritoneum, and spleen was evaluated by FACS analysis. Fresh organ suspensions did not contain human TILs. In mice given cyclophosphamide followed by human TILs i.p., the TILs were demonstrated at 7 days in the SCID peritoneum (leu 4 = 4%) and at 2 days in the SCID spleen (leu 4 = 2%). In BNX mice, 12 of 14 fresh human colon adenocarcinomas were propagated successfully at subcutaneous sites with latency periods ranging from 1 to 13 weeks. Enzymatic disaggregation of tumors greater than 1 cm following one passage yielded 6.5-47 x 10(6) cells with viabilities ranging from 13 to 85%. We conclude that limitations and variability exist in the use of BNX and SCID mice for human PBMC reconstitution, TIL persistence, and propagation of human colon adenocarcinomas.

Ductal carcinoma in situ (DCIS): a revised concept.

The in situ concept was introduced in an effort to clarify the transition between benign epithelium and invasive cancer, and for that reason it focused on histologic changes. Lobular carcinoma in situ was first described in these terms and continues to be considered a purely microscopic lesion that never makes a mass in the breast. A very different situation exists in cases of ductal carcinoma in situ (DCIS), because both gross and microscopic disease exists together. As a result, it has been difficult to evaluate competing treatment options for the DCIS lesion. This study was undertaken to better characterize patients with DCIS lesions. Seventy consecutive patients with DCIS who underwent treatment at our institution were analyzed and two subgroups were identified. The method of presentation and the distribution of cancer in the breast as well as in the regional lymph nodes were examined. The study shows that differentiation between gross and purely microscopic DCIS is feasible and must be accomplished if treatment recommendations are to be made on a rational basis.

Primary antiphospholipid syndrome in a child with lower extremity arterial thrombosis.

Primary antiphospholipid syndrome is characterized by venous and arterial thrombosis, fetal loss, or thrombocytopenia in association with antiphospholipid antibodies and no associated disease process. The authors report a case of lower extremity thrombosis in a 12 year old who had primary antiphospholipid syndrome. To our knowledge, this is the first report of peripheral arterial thrombosis in primary antiphospholipid syndrome in the pediatric population.

Interleukin-7 mediates the generation and expansion of murine allosensitized and antitumor CTL.

Interleukin-7 (IL-7) is a 25-kDa cytokine that was initially described as a pre-B cell growth factor and more recently has been shown to cause T cell proliferation. We have investigated the in vitro effects of IL-7 on mature T cells to include the generation and further expansion of allospecific and antitumor CTL. B6 anti-DBA allospecific CTL were generated in the presence of IL-7, IL-2, the combination IL-7 plus IL-2, or no cytokine. IL-7 alone or when combined with IL-2 enhanced the generation of allospecific CTL. To evaluate the proliferative effects of IL-7, 4-day B6 anti-DBA cultures were cultured in IL-7, IL-2, or no cytokine. Cell proliferation and duration of growth of cells cultured in IL-7 were significantly greater than cells cultured in IL-2 or in the absence of cytokine. Allospecific cytolytic activity was maintained during proliferation in IL-7 to a maximum of 60 days. In contrast with the ability of IL-2 to generate LAK cells, murine splenocytes cultured at varying doses of IL-7 (1 to 10,000 ng/ml), resulted in minimal LAK cell activity. The effect of IL-7 in the generation of CTL with antitumor activity was also studied. Seven days after footpad injection of MCA 203 or 205 sarcoma, draining lymph nodes (DLN) were harvested and restimulated in vitro with MCA 203 or 205, respectively, and maintained in culture with either IL-7, IL-2, the combination of IL-7 plus IL-2, or no cytokine. After 10 days in culture, cells generated in IL-7 or IL-2 exhibited similar cytotoxicity against the syngeneic autologous MCA tumor. IL-7 generated cells, however, showed specificity when tested by 51Cr release which was not seen with IL-2-generated cells. Cells generated in IL-7 plus IL-2 were more cytolytic than cells cultured with either cytokine alone. To further define the mechanism of action of IL-7 in antitumor CTL cultures, a monoclonal antibody, S4B6.1, capable of blocking murine-specific IL-2 was employed. The partial inhibition by this mAb of the generation of antitumor CTL demonstrated that IL-7 acted, in part, by an IL-2-dependent mechanism. Finally, IL-7 cultures restimulated at Day 11 with autologous MCA 203 showed greater proliferation than IL-2 cultures and remained lytic at Day 21 of culture.(ABSTRACT TRUNCATED AT 400 WORDS)

Durability of cross-femoral grafts after aortic graft infection: the fate of autogenous conduits.

PURPOSE: Revascularization for the treatment of aortic graft infection is usually accomplished by remote prosthetic axillofemoral bypass combined with cross-femoral bypass. When infection at the femoral level precludes placement of a prosthetic cross-femoral graft, we have used a variety of autogenous tissue conduits to restore circulation to the contralateral leg. To determine which of these conduits offers the most durable reconstruction, we have reviewed 78 patients treated for aortic graft infection. METHODS: Between 1980 and 1991 we used either autogenous saphenous vein (ASV, n = 34), endarterectomized superficial femoral artery (SFA, n = 14), or direct ilioiliac anastomosis (iliac, n = 10) to provide cross-femoral flow. We compared the performance of these tissue conduits with a concurrent patient group with aortic graft infection in whom a prosthetic cross-femoral graft was used (prosthetic, n = 20). RESULTS: Follow-up was available for 98.7% of patients, average 3.8 +/- 2.9 years, and was not different between the four groups. Bleeding complications occurred exclusively in the ASV group (n = 3, 8.8%) and were all in the perioperative period. In addition one ASV and one iliac conduit developed multiple false aneurysms. Hemodynamic conduit failure (thrombosis or stenosis) occurred in nine (26.5%) ASV conduits, six (42.8%) SFA conduits, and one iliac conduit, but not in the prosthetic group. When all of these adverse events were combined for each conduit group, both ASV and SFA conduits had a higher rate of conduit failure when compared with the prosthetic conduits (p < 0.05, log-rank test). Limb loss resulting from cross-femoral conduit failure occurred in six (17.6%) patients in the ASV group, four (28.6%) patients in the SFA group, and one patient each in the iliac and prosthetic groups. These differences were not significant. CONCLUSIONS: We conclude that ASV and SFA conduits do not provide stable long-term cross-femoral revascularization and should be regarded as bridge grains until femoral infection is eradicated. When femoral infection mandates their use, frequent postoperative conduit surveillance is required. If ASV or SFA caliber is marginal, consideration should be given to the use of a larger autogenous conduit, such as superficial femoral vein.

Secondary aortoenteric fistula: contemporary outcome with use of extraanatomic bypass and infected graft excision.

PURPOSE: The standard treatment for secondary aortoenteric fistula (SAEF) has been infected graft removal (IGR) and extraanatomic bypass (EAB), an approach criticized for its high rate of death, amputation, and disruption of aortic closure. Recently, graft excision and in situ graft replacement has been proposed as a safer treatment alternative. Because the current outcome that can be achieved by use of the standard treatment of SAEF has really not been established, we reviewed the records of 33 patients treated for SAEF at our institution during a contemporary time interval (1980 to 1992). METHODS: Thirteen patients (39.4%) were admitted with evidence of gastrointestinal bleeding and infection, whereas nine (27.3%) only had bleeding, 10 (30.3%) only had signs of infection, and one SAEF was entirely occult (graft thrombosis). Four patients required emergency operation. The fistula type was anastomotic in 13 (39.4%) patients, paraprosthetic in 15 (45.5%), and not specified in 4 cases. Thirty-two patients underwent EAB followed immediately by IGR (n = 16, 48.5%) or followed by IGR after a short interval, averaging 3.9 days (n = 16, 48.5%). The final patient underwent IGR, followed by EAB. RESULTS: Follow-up on 31 patients (93.9%) averaged 4.4 +/- 3.7 years. There were nine deaths (27.3%) resulting from the SAEF, six perioperative and three late. Three patients (9.1%) had disrupted aortic closure. There were four amputations in three patients (9.1%), two perioperative and two late. Late EAB infection occurred in five patients (15.2%), leading to one death and one amputation. EAB failure occurred in six patients, two during operation and four late, leading to one amputation. The cumulative cure rate for this SAEF group was 70% at 3 years and thereafter. Compared with our earlier SAEF experience, this is a decline of 21% in the mortality rate, 19% in aortic disruption, and 27% in limb loss. CONCLUSIONS: We conclude that outcome reports based on SAEF series extending over long time intervals do not accurately represent the results that are currently achieved with standard SAEF treatment with use of EAB plus IGR. This improved outcome is attributed to wide debridement of infected tissue beds, reduced intervals of lower body ischemia, and advances in perioperative management. To determine whether any new treatment approach actually offers improved outcome in the management of SAEF, comparison with EAB plus IGR should be limited to patients treated within the last decade at most.

Effects of recombinant human leukocyte interferon treatment of endogenous interferon production in patients with chronic type-B hepatitis.

Interferon (IFN) added to cell culture systems alters the capacity of the cells to produce IFN when appropriately stimulated. To evaluate the effects of in vivo administration of IFN on the production of IFN by peripheral blood mononuclear cells (PBMCs), we studied patients with chronic type-B hepatitis who received doses of recombinant human leukocyte (alpha) IFN (IFN-alpha) ranging from 5 X 10(6) units daily to 60 X 10(6) thrice weekly. The production of endogenous IFN stimulated by specific inducers (Sendai virus for IFN-alpha; phytohemagglutinin for IFN-gamma) was studied in cell cultures containing PBMCs obtained from patients before or during courses of IFN treatment. In untreated controls, no change in the mean capacity of PBMCs to produce IFN-alpha was noted after 2 weeks. Priming of endogenous IFN-alpha production, as reflected by earlier production of IFN by PBMCs in culture, occurred in all treated patients irrespective of the dose of IFN-alpha received. Whereas mean 24-hour (total) endogenous IFN-alpha fell in all treatment groups, the response was highly variable in individual patients and half showed no change in total production. Individual variations in endogenous IFN-alpha production were unrelated to serum IFN levels achieved during treatment, changes in serum aminotransferase levels, reduction of hepatitis B virus replication during therapy, or the proportions of T and B lymphocytes in culture. In contrast to the changes in IFN-alpha production, IFN-gamma production by PBMCs was not affected by IFN-alpha treatment.(ABSTRACT TRUNCATED AT 250 WORDS)

Human endothelial cells synthesize ENA-78: relationship to IL-8 and to signaling of PMN adhesion.

The interaction of endothelial cells and polymorphonuclear leukocytes (PMNs, neutrophils) is a critical determinant of the acute inflammatory response, and mirrors cell-cell interactions in other biologic systems. Adhesion molecules that tether the two cells together, and signaling factors that bind to receptors on the leukocytes and mediate their spatially-localized activation, govern PMN responses as they adhere to and traverse stimulated endothelial cells. Here we show that cultured human endothelial cells express two members of the C-X-C family of chemokines, epithelial neutrophil activating peptide-78 (ENA-78) and interleukin (IL)-8, when stimulated by IL-1 or certain other agonists. ENA-78, previously thought to be exclusively a product of epithelium, is expressed by in situ endothelium in inflamed human lung and other tissues as well as by cultured endothelial cells. The regulation of ENA-78 and IL-8 share certain features in common and they have overlapping biologic activities, including the ability to induce PMN adhesiveness. This was demonstrated in experiments in which we found that ENA-78 induces inside-out signaling of beta2 integrins on the PMN surface, as does IL-8. Antibody blocking experiments demonstrated that ENA-78 contributes to the proadhesive activity for neutrophils that is released by endothelial cells stimulated with IL-1 for a prolonged period and that it acts in concert with IL-8, which provides the major component of the signal for adhesion under this condition. We also show, however, that the temporal expression and secretion of ENA-78 and other characteristics of its handling by stimulated endothelial cells vary from the expression of IL-8, indicating that differential regulation of the two signaling chemokines occurs in this cell type.

Distinction between hyperaldosteronism due to bilateral hyperplasia and unilateral aldosteronoma: reliability of CT.

Hyperaldosteronism due to a unilateral adenoma must be distinguished from hyperaldosteronism due to bilateral hyperplasia to enable the proper choice between surgical treatment (for adenoma) or medical treatment (for hyperplasia). To compare the efficacy of computed tomography (CT) and adrenal venous sampling, both examinations were performed in 24 patients with primary aldosteronism. All patients with a diagnosis of adenoma based on findings at venous sampling underwent adrenalectomy. The CT-based diagnosis was unilateral aldosteronoma in 17 patients and hyperplasia in seven patients. On the basis of venous sampling, unilateral adenoma was diagnosed in 22 patients; this diagnosis was confirmed by means of unilateral adrenalectomy in 21 patients. The most common error was diagnosis of hyperplasia based on the presence of bilateral nodules on CT scans: In six of seven patients with such a diagnosis, venous sampling and subsequent surgery revealed a unilateral adenoma. In hyperaldosteronism with multiple bilateral nodules, CT cannot reliably permit distinction between hyperplasia and adenoma.

Pulmonary metastases: MR imaging with surgical correlation--a prospective study.

The sensitivity of magnetic resonance (MR) imaging for detection of pulmonary metastases in 11 patients scheduled for thoracotomy and curative resection of metastases was evaluated with a prospective, controlled study. MR imaging performed at 0.5 T was compared with chest radiography, computed tomography (CT), and thoracotomy in 12 cases. (One patient had two separate occurrences of pulmonary metastases.) All images were interpreted in blinded fashion. When all MR sequences were interpreted together, MR imaging enabled correct identification of all patients with pulmonary nodules (100%). CT enabled detection of at least one nodule in all 12 cases (100%) by design; the sensitivity of chest radiography was only 64%. For individual nodules, MR imaging was at least as sensitive as CT (P2 less than .25 [two-sided value]) and significantly more sensitive than chest radiography (P2 less than .01). Among all MR sequences, short inversion time inversion-recovery sequences had the highest sensitivity for detection of individual nodules (82%).