Transfer factor from guinea pigs sensitive to dinitrochlorobenzene: absence of superantigen properties.

Transfer factor from guinea pigs sensitive to dinitrochlorobenzene was not bound to an immunoadsorbent column that is specific for the dinitrophenyl determinant. The absence of the dinitrophenyl determinant on transfer factor suggested that the factor does not function as superantigen. The duration of the adoptive sensitivity, the small molecular weight, and the polypeptide or polynucleotide (or a combination) composition of the transfer factor are consistent with a derepressor function of the molecule.

Immune response to melanoma extracts in three melanoma-prone families.

Immune reactivity to melanoma extracts was measured by the leukocyte adherence inhibition (LAI) test in 40 members of 3 melanoma-prone families. The melanoma patients had a wide range of responsiveness to the extract, the highest responder being a 10-year survivor. As a group, family members (including spouses) without disease had significantly elevated LAI responses compared to those of unrelated controls (P less than 0.01). Within the families, members with close exposure to melanoma patients for 10 years or more had a significantly higher response to melanoma antigen than did members with 0-5 years of close exposure (P less than 0.05). Responses of spouses and members at high risk of developing melanoma (B-K mole syndrome) also correlated with length of exposure to patients, which suggests that the elevated LAI response was not genetically determined. The high frequency of positive responses to melanoma antigens in these families, particularly in spouses, suggests the presence of transmissible melanoma-associated material capable of immunizing persons in contact with melanoma patients.

Flow cytofluorometric detection of tumor-specific rosette-forming cells in patients with squamous cell carcinoma of the head and neck.

Tumor-specific rosette-forming cells reactive to solubilized tumor antigens conjugated to autologous erythrocytes were quantitated by flow cytofluorometry. Leukocytes from a high frequency of the patients (greater than 70%) with squamous cell carcinoma of the head and neck (SQCC) formed rosettes to the conjugated SQCC tumor antigens but not to other histologically distinct tumor antigens (melanoma and colon carcinoma). Healthy control subjects or tumor patients with other cancers were mostly unreactive to the SQCC tumor extract [1 to 21 (5%) and 1 of 14 (7%) for controls and tumor patients, respectively]. Rosette-forming activity was observed in SQCC patients with primary cancers [22 of 30 (73%)] or in remission [5 of 6 (83%)], whereas patients with tumor recurrence were uniformly unresponsive [0 to 9 (0%)]. Tumor-specific rosette formation was mediated predominantly by monocytes, as identified by histochemical techniques and physiological properties. Rosette formation in reactive patients was abrogated by short-term culture, but the abated response could be restored by incubation with autologous serum or sera from other rosette-forming cell-positive patients. However, responsiveness of nonreactive patients with SQCC recurrence could not be constituted by rosette-forming cell-positive sera. These observations suggested the presence of tumor-reactive monocytes in a high frequency of patients with primary cancer or in remission but not in patients with recurrent disease.

Assessment of the mechanism of the leukocyte adherence inhibition test.

This study was designed to elucidate the mechanism of the leukocyte adherence inhibition (LAI) test in man. To identify the reactive cell types, enriched leukocyte populations (dextran-separated leukocytes and Hypaque-Ficoll-isolated mononuclear cells and neutrophils, as well as rosette-isolated B- and T-lymphocytes) were tested for leukocyte adherence in the absence of serum to tumor-specific antigens. LAI reactivity was not restricted to any of the enriched populations, suggesting the involvement of multiple cell types. Attempts to demonstrate soluble lymphocyte factors in the LAI mechanism have been uniformly negative. In contrast, factors in serum of immune donors were able to arm naive cells to be specifically responsive. This suggests a role for serum factors in the mechanism of LAI reactivity and partially explains the participation of multiple cell types in the responses observed. In additional studies, we could not document a correlation between the magnitude of the dermal test (delayed cutaneous hypersensitivity) and the magnitude of the LAI response in patients with squamous cell carcinoma of the head and neck. In 34 of 54 of these patients, there was agreement between the two tests (both positive, 27 of 54; both negative, 7 of 54). In the remaining 20 patients, the dermal test was greater than 5 mm while the LAI test was negative (less than 30% inhibition).

Association of melanoma tumor antigen activity with beta2-microglobulin.

The majority of melanoma tumor antigen activity present in melanoma extracts derived from fresh tumor tissue binds to a Sepharose-anti-beta2-microglobulin adsorbent. Removal of HLA antigens from the extracts of melanoma tissue by using a KBr flotation technique did not reduce either the tumor antigen activity of the extracts or the binding of melanoma tumor antigen (MTA) activity to the Sepharose-anti-beta2-microglobulin adsorbent. The complete blocking of MTA activity by pretreating the anti-beta2-microglobulin adsorbent with beta2-microglobulin and the lack of detectable MTA binding to a Sepharose anti-normal human serum adsorbent demonstrated the specificity of the binding of MTA to the anti-beta2-microglobulin adsorbent.

Automated counting of T cell rosettes.

Human T cell rosettes were enumerated using an automated particle counter, the Bio/Physics Cytograf 6300A. An electronic oscilloscope representation of particle absorbance and scatter of a focused laser beam allows the separation and enumeration of both rosetted and non-rosetted lymphocytes. Repeated Cytograf sampling of a single rosette preparation gave highly reproducible results, and sampling from replicate tubes produced the same degree of variation as microscopic analysis. T cell rosettes prepared from 27 volunteers and compared by both methods of quantitation showed a high degree of correlation. This method can objectively measure at least 100 times as many cells for their rosette-forming capability as the tedious microscopic technique.

Subcultured human endothelial cells can function independently as fully competent antigen-presenting cells.

Recent evidence has suggested that dendritic cells, epidermal Langerhan's cells and endothelial cells (EC) as well as macrophages, fulfill the requirements of antigen-presenting cells. Despite a variety of controls, one weakness in the evidence that these latter cell types can independently serve as antigen-presenting cells is that the cell preparations may contain small numbers of contaminating macrophages or other cell types. The experiments described in this paper are directed towards providing firm evidence that human EC are independently capable of presenting antigen to T cells. EC were isolated from human umbilical veins and maintained continuously by serial subculture for periods of up to 8 months. The subcultured EC displayed classic EC morphology and uniform immunofluorescent staining for Factor VIII-related antigen. The subcultured EC (tested to the 18th subculture) presented both particulate and soluble antigens to macrophage-depleted T cells with an efficiency equivalent to freshly isolated cells. Monoclonal antibodies to HLA-DR and HLA-DS determinants inhibited antigen presentation by either autologous macrophages or EC. In addition, antigen presentation by the subcultured EC was not affected by the macrophage-specific monoclonal antibody Mac-120, which inhibited antigen presentation by autologous macrophages in the same experiments. These results are consistent with human EC being able to independently function as fully competent antigen-presenting cells.

Endothelial cell presentation of antigen to human T cells.

Activation of human T cells requires presentation of antigen by Ia (HLA-DR in man) bearing cells of the mononuclear phagocytic series (macrophages, MO, and more recently Langerhans cells, dendritic cells, and vascular endothelial cells. Since T cells must cross endothelial barriers to enter extravascular tissues during immune reactions, we investigated the role of endothelial cells in antigen presentation. Endothelial cells were cultured from human umbilical veins and identified by classic morphology and specific markers (factor VIII related antigen, and so on). Antigen-pulsed endothelial cells were used to present antigen to MO-depleted human T cells; activation was assessed by 3H-thymidine uptake. The HLA-DR compatible endothelial cells were as effective as MO in reconstituting MO-depleted T-cell responses. The endothelial cell reconstituted responses were antigen specific, HLA-DR restricted, and blocked by monoclonal antibodies to HLA-DR framework structures. Moreover, the T-cell responses were clonal with respect to HLA-DR. A monoclonal antibody completely eliminated MO reconstitution of the MO-depleted response without diminution of endothelial cell reconstitution of the same response. Fibroblasts and smooth muscle cells cultured from the same umbilical veins could not reconstitute the MO-depleted T-cell response. These data indicate that endothelial cells play an important and distinctive role in lymphocyte triggering.

The mechanism of antigen presentation by endothelial cells.

Endothelial cells line the vessels and lymphatics forming a barrier between circulating T cells and the extravascular tissue site of antigen. We have suggested that circulating T cells recognize antigen on the surface of endothelial cells, resulting in the activation of the endothelium such that the endothelial cells then release the key mediators of a cell-mediated immune response. To test this hypothesis, we have evaluated the extent to which endothelial cells can signal antigen-specific T cell activation. We have shown that cultured endothelial cells are as effective as macrophages in lymphocyte activation and that this activation is HLA-DR restricted. In additional experiments we have established that endothelial cells synthesize both Ia and IL-1 early in the signaling process. To eliminate any possible contribution of other cell types participating in the T cell-endothelial cell interaction, we have shown that cloned endothelial cells present antigen to cloned T cells. Moreover, there appeared to be a preference of selected T-cell populations for different types of antigen presenting cells. These experiments document that endothelial cells are independently competent antigen presenting cells.

Free jejunal mucosal patch graft reconstruction of the oropharynx.

We used microvascular anastomoses to transpose free pedicle jejunal mucosal patch grafts in seven patients. The procedure has been particularly helpful in rebuilding large intra-oral mucosal defects created by extensive resections for advanced carcinomas. Some of the benefits of this technique have included a one-stage procedure, which requires two to three weeks for healing; abundant donor tissue with characteristics similar to oral mucosa; near-normal facial appearance; and preservation of maximum tongue function. An unexpected benefit has been relief of annoying xerostomia by the jejunal mucous secretion. The most severe complication, which resulted in one death, was the excessive oral jejunal mucous secretion in the early postoperative period. It led to significant aspiration pneumonitis. To prevent this problem, we recommend a routine tracheostomy combined with rigorous pulmonary care whenever a jejunal patch graft is used.

Application of clinically valid cardiac risk factors to aortic aneurysm surgery.

The operative outcome of 97 consecutive nonruptured infrarenal aortic aneurysms is analyzed regarding clinically identifiable cardiac risk factors. Clinically evident coronary artery disease was present in 45 patients (46%). Operative mortality was 4% (four cardiac deaths) with an additional 4% nonfatal postoperative myocardial infarction rate. All cardiac complications occurred in patients with clinically evident coronary artery disease, while no mortality occurred in 52 patients lacking a preoperative history of myocardial infarction, congestive heart failure, or angina. Preoperative risk factors having a significant negative influence on outcome include a history of prior myocardial infarction and compensated congestive heart failure. Few patients with aneurysms who have clinical evidence of coronary artery disease are indicated for coronary arteriography and bypass prior to aneurysm repair. Furthermore, indications for invasive cardiac screening of the patient with an aneurysm who lacks cardiac symptoms are limited.

Efficacy of oral and systemic antibiotic prophylaxis in colorectal operations.

A cooperative Veterans Administration study of the septic complication rate during large-bowel surgery was undertaken in two groups of patients. The first group received oral neomycin and erythromycin base plus parenteral placebo; the second, the oral antibiotics plus parenteral cephalothin sodium. During a five-year period, 1,128 patients were studied. The overall septic complication rate was 7.8% in patients receiving only oral antibiotics, and 5.7% in patients receiving both oral and parenteral antibiotics. This difference was not significant. The only significant finding was a greater incidence of fever of unknown origin in patients receiving only oral antibiotics. None of those patients were treated with additional antibiotics, and all fevers cleared spontaneously. There seems to be no discernible benefit from adding parenteral antibiotic prophylaxis when performing elective colon surgery if appropriate mechanical cleansing and oral neomycin and erythromycin therapy are employed.

Correction of cervical esophageal stricture using an axial island cheek flap.

Chronic strictures of the cervical esophagus after laryngectomy and radiation therapy pose a difficult problem in reconstructive surgery. Most conventional operations for cervical esophageal reconstruction are not well suited to the treatment of stricture because of lack of mucosal surface, or because of bulky tissues that awkwardly fit around the tracheal stoma. This report describes our experience with a new operation designed to correct this defect. We transposed an island full-thickness cheek flap, which included an inner lining of mucosa and outer covering of skin. The flap was based on the facial artery and vein, and used as a patch to the stricture area. Normal swaLlowing and excellent cosmetic appearance were achieved. There has been minor numbness of the corner of the upper lip. This flap has excellent reach, and may have multiple applications to reconstruction problems in the head and neck.

Perforation of the colon in the immunocompromised patient.

Perforation of the colon in the immunocompromised patient is a catastrophic and usually fatal event. The immunocompromised patient, like all patients, may suffer from the more common causes of colonic perforation, including diverticulitis, chronic inflammatory bowel disease, presence of a foreign body, and trauma. There also appears to be in these patients the unusual occurrence of spontaneous perforation, particularly in patients with renal allografts or on dialysis. In a retrospective multi-hospital review, 10 cases of apparent spontaneous perforation were found. The pathogenesis is unclear, but predisposing factors include immunosuppressive medications, uremia, discrete colon ulcerations, and fecal impaction. The reported mortality rate approaches 100 percent due to delayed recognition and impaired host defense mechanisms. In our patients, mortality was 40 percent. We attribute this improved survival to prompt surgical intervention and aggressive postoperative management, including daily dialysis, parenteral hyperalimentation, broad-spectrum antibiotics, and a high index of suspicion for ongoing sepsis with early repeat exploration.

Prognostic factors for survival after pancreaticoduodenectomy for malignant disease.

Forty patients underwent pancreaticoduodenectomy for malignant disease in a 25-year review of 6 Portland-area hospitals. After resection, 10 patients survived for greater than 5 years. Characteristics of survivors included a short prodrome of symptoms, minimal weight loss, tumor in the periampullary region, and female sex. Factors that did not influence survival included degree of jaundice, primary tumor size, and initial abnormal liver function studies.

Influence of serum blocking factors on cancer patients undergoing immunotherapy.

Blocking factors are small polypeptide molecules that may appear in the serum of patients with cancer. These factors block the transformation of lymphocytes in culture to nonspecific mitogens such as phytohemagglutinin or concanavalin A and, therefore, may reflect changes in the immunocompetence of the patient. Blocking factors were monitored during the clinical course of thirty-five patients with cancer. These factors did not develop in patients with response to therapy whereas they did develop in patients without response. A third group of patients without response to therapy after a previous remission showed an absence of lymphocyte responsiveness in culture that was not due to blocking factors, suggesting that immune clone consumption had occurred. Dermal responsiveness to tumor antigen correlated with a favorable clinical course and was usually absent when serum blocking factors were present.

Experience with colovesical fistula.

Experience with 37 patients with adult colovesical fistula over the past 19 years is reviewed. Specific guidelines for treatment of adult colovesical fistula are influenced by the location and cause of the fistula, the patient's general condition, the presence of a pelvic abscess and the presence of colonic obstruction. When criteria are met, a one-stage procedure is safe. The two-stage approach should enjoy wider application, with the three-stage approach reserved for patients who are unprepared or who have a large pelvic abscess. In patients with colovesical fistula due to cancer, the extent of tumor should be carefully evaluated and resection carried out whenever possible. Colovesical fistulas due to trauma, inflammatory bowel disease and iatrogenic causes are often unusual in location; thus treatment must be individualized.

Experience with total pancreatectomy.

Total pancreatectomy for benign disease should be considered only in highly selected patients and then only after lesser surgical procedures have failed. At present, truncal vagotomy and adequate gastrectomy should be part of the operation to prevent marginal ulceration. A multitude of undesirable problems, many requiring reoperation, may arise postoperatively and can compromise an otherwise excellent outcome with regard to pain control.

Hand infections in hospitalized patients.

Two hundred hospital admissions for hand infections are reviewed with regard to the causes of infection, modes of effective therapy and causes of residual impairment. Patients with human bite injuries and those who delayed seeking treatment most frequently had severe injuries. Appropriate antibiotics and prompt surgical therapy of hand abscesses are essential.

Problems in diagnosis and management of desmoid tumors.

Thirty-two primary desmoid tumors occurred in 29 patients. The median patient age was 32 years, and 55% of the patients were females. An antecedent history of trauma at the tumor site was elicited from 28% of patients. Thirty-one of 32 primary tumors were completely excised at initial presentation. Five tumors were treated with adjuvant radiotherapy. The overall recurrence rate for primary and recurrent lesions was 60%. The recurrence rate in children (88%) was more than twice that found in adults (38%). A single recurrence did not significantly increase the likelihood of a subsequent recurrence. Greater than 90% of all recurrences took place within 3 years of treatment. The rate of recurrence was not clearly influenced by the status of histologic margins, although this was examined in less than half the tumors. Desmoid tumors are aggressive neoplasms that exhibit a strong propensity for local recurrence. They should be treated as low-grade malignancies with documentation of histologic margins and close clinical follow-up within the framework of a tumor registry.