Gemcitabine has been recently added to the 5-fluorouracil (5-FU) protocol in the treatment of both new and 5-FU refractory patients with pancreatic cancer. We compared the efficacy of gemcitabine versus 5-FU and leucovorin (LCV) in 82 patients with advanced pancreatic cancer. Forty-seven patients received 5-FU and LCV and 35 patients received gemcitabine. Objective responses were documented in 4% on the 5-FU and LCV arm, compared to 9% in the gemcitabine arm. No change was observed in 48% on the 5-FU and LCV arm compared to 20% in the gemcitabine arm. Clinical benefit was observed in 19% on the 5-FU and LCV arm compared to 48% in the gemcitabine arm (p<0. 001). Toxicity was mild and well tolerated in both arms. One-year survival was 32% on the 5-FU and LCV arm and 23% in the gemcitabine arm. These results indicate that gemcitabine had a significantly higher clinical benefit than 5-FU and should be the standard treatment in advanced pancreatic cancer.

Forty-seven patients with pancreatic cancer were treated with two different schedules of 5-fluorouracil (5FU) and leucovorin (LCV): standard and dose-intense schedules. The standard regimen was monthly and the dose intense was biweekly. Partial response was observed in one patient (4%), no change in 12 (48%) and progression of disease in another 12 patients. Clinical benefit, measured by symptomatic improvement, was observed in 19% of all the patients, in 12% of those treated with the standard regimen and in 27% of the intense group. Median survival was 8 months for all the patients. The 1-year survival rate was 32%. Toxicity was mild. There was no survival benefit for the dose intense regimen. These results indicate that clinical benefit can be obtained with 5FU and LCV regimens despite the lack of objective response and that a dose-intense schedule is of little benefit in treating pancreatic cancer.

Staging of gastric carcinoma depends on exact lymph node status. However, very small nodes are not easily found as they are obscured by the surrounding adipose tissue. The purpose of the present study was to demonstrate the usefulness of a Olymph node revealing solutionO (LNRS) in gastric cancer. The perigastric adipose tissue of ten OproblematicO cases of gastric carcinoma, in which <10 lymph nodes were found using the traditional method, was immersed in LNRS for 6-12 h. Subsequently, the lymph nodes stood out as white chalky nodules. They were excised and processed routinely. The traditional method yielded a total of 30 lymph nodes with a mean size of 6.69 +/- 3.43 mm. The LNRS revealed 89 additional nodes with a mean size of 3.03 +/- 3.43 mm, which was significantly smaller. The Node (N) stage was changed in four cases from Nx to N0, in one case from N1 to N2, and in one case from N0 to N2. LNRS seems to be the technique of choice for staging of patients with gastric adenocarcinoma in whom <10 lymph nodes were found with the traditional method and accurate staging was not possible.

Background. Tissue polypeptide specific antigen (TPS) measures a soluble fragment of cytokeratine 18 and may be regarded as a proliferative marker. Materials and Methods. TPS was measured in 173 consecutive patients with colorectal cancer. Median follow up time was 36 months. Of 137 evaluable patients 39 developed metastases (P.D.) and 98 remained with no evidence of disease (N.E.D.). Results. Initial TPS levels were elevated in 75% of P.D. patients compared to 32% of N.E.D patients (p<0.001), CEA levels were elevated in 26% of P.D. patients compared to 2% of N.E.D. In Dukes' B2 92.8% of PD. patients had elevated initial TPS compared to 35.5% of N.E.D patients (p <0.001) and CEA was elevated in 33.3% of the P.D. patients compared to 1.3% of N.E.D. patients (p<0.001). Survival and disease free survival were significantly shorter for patients with initial high TPS level. TPS was more sensitive than CEA in predicting relapse. Conclusions. These preliminary data suggest that TPS may be a prognostic factor for relapse and may help to allocate Dukes' B2 patients for adjuvant chemotherapy.

HLA class I antigens are composed of a major histocompatibility complex (MHC) encoded heavy chain that is associated non-covalently with alight chain beta-2 microglobulin (beta-2m). When the HLA complex is metabolized, beta-2m is shed into the serum. A large variety of human and experimental tumours have altered MHC class I expression. In a previous study we observed elevated mean beta-2m serum levels in breast cancer patients, as compared to controls. To study the relationship between tumour expression and serum levels, we examined 54 patients with breast cancer. Tumour beta-2m was determined by immunohistochemistry and serum levels by the ELISA technique. Of the 54 patients, 38 had low and 16 had high beta-2m expression on the tumour. There was a significant correlation between tumour beta-2m and serum beta-2m levels (P = 0.02), with patients whose tumours expressed high beta-2m having high serum beta-2m levels. There was an inverse correlation between tumour grade and tumour beta-2m expression which approached statistical significance (P = 0.06). These findings suggest that in a substantial number of patients the high serum levels derive from shedding of beta-2m from tumour cells. These levels may have implications for tumour growth and metastases due to influences on immunological responses.

The serum levels of soluble interleukin-2 (sIL-2), sIL-2 receptors (sIL-2R), beta 2-microglobulin (beta 2M) and soluble intercellular adhesion molecule-1 (sICAM-1) were measured by the ELISA technique in 129 breast cancer patients and 40 controls. The median serum levels of sIL2-R, beta 2M and sICAM-1 were significantly higher and sIL-2 significantly lower than controls. sIL-2R, sICAM-1 and beta 2M levels were significantly higher in patients with metastatic disease compared to patients on long-term follow-up with no active disease. Initial study measurements of these markers could not identify patients at high risk for relapse. These findings suggest that the sIL-2R level is indicative of metastatic disease and together with other parameters of immune activation may be of help in monitoring disease activity in breast cancer patients.

Serum TPS and CEA levels were measured in 173 cancer patients with various disease stages. The patients were divided into 4 groups: preoperative group, 16 patients in whom TPS and CEA were measured pre- and postoperatively; group I, 66 newly diagnosed patients up to 40 days post surgery; group II, 86 patients whose blood was taken 40-365 days post surgery, and group III, 47 patients on long-term follow-up in whom blood was taken more than one year after surgery. The median preoperative TPS levels were significantly higher (p<0.01) than the postoperative levels. Patients with Dukes' D had significantly higher TPS (p<0.01) and CEA levels as compared to patients with Dukes' B or C. For both Dukes' B and C stages there was a significant increase in median TPS level between group I and II and III. There was no similar increase in median CEA level. Serial measurements in individual patients show a significant increase in TPS levels between groups I, II and III. In Dukes' D patients, individual TPS levels decreased in response to therapy. There was a significant correlation between TPS and CEA levels. These results suggest that TPS measurement is additive to CEA in follow-up of asymptomatic patients, and to monitor response to therapy in patients with metastatic disease.

Serum levels of P185-HER-2 were measured in 137 breast cancer patients and in 40 controls. The patients were divided into 4 groups: group A - 40 newly diagnosed patients; group B - 57 patients on long-term follow-up without active disease; group C - 26 patients with metastatic disease and group D - 5 patients with locally advanced inoperable tumors. The median level in controls was 4.8 U/l. The median P185 serum levels in groups C and D were significantly higher compared to groups A and B. In group C 60% and in group D 100% of patients had baseline elevated levels of serum P185 (>5 U/l) compared to 28% in groups A and B. Of the 14 patients in group A with elevated baseline levels of serum P185, 6 (43%) developed metastasis during the 24-month follow-up period. On serial measurements during follow-up in 23 patients of group A, 3 relapsed and the P185 level increased. In group C, serial measurements in patients with elevated baseline levels of P185 correlated with clinical response to therapy. These data suggest that serum levels of P185 are elevated in patients with metastatic disease. High initial P185 serum levels in new patients may have prognostic significance. Serial measurements of P185 in asymptomatic patients may help in monitoring disease state. In metastatic patients, serial P185 determination may be of benefit in assessing response to therapy.

The immunosuppressive effects of irradiation are well known; however, under certain circumstances irradiation also augments the local immune response by as yet undefined mechanisms. Because of the importance of HLA class I antigen in immune regulation and the fact that killing of tumor cells by cytotoxic T cells is HLA antigen-restricted, the authors studied HLA class I antigen expression in eight glioblastomas multiforme, four meningiomas, and four medulloblastomas. Twenty fragments of each tumor specimen were placed in short-term cultures immediately after resection. For each tumor, control Sample 1 was not irradiated. Sample 2 was irradiated on Day 1, and two groups of the remaining pieces of each tumor (specimens 3 to 10) were irradiated on two consecutive days. Escalating radiation doses were given, starting at 200 cGy/day for Sample 2 up to 1000 cGy/day for Sample 10. The total dose range was 200 to 2000 cGy. Corresponding nonirradiated tumor fragments served as controls. Four hours after irradiation, each sample was processed and stained for HLA class I antigen using the immunoperoxidase technique. The tumor cells were intensely stained in nonirradiated glioblastomas and meningiomas, whereas no staining was observed in medulloblastomas. In four of the eight glioblastomas and in all four meningiomas, irradiation augmented HLA class I antigen expression compared to controls. This effect was dose-dependent and was maximum in the 1200 cGy-treated specimens. No change was observed in the other four glioblastomas or in the medulloblastomas. The data suggest that irradiation does not decrease and may even induce HLA class I antigen expression in some brain tumors. This may be one of the mechanisms by which immunotherapy operates after irradiation. Further studies are required to elucidate optimum radiation doses and fractionation as well as optimum timing of immunotherapy.

sICAM-1 and beta-2 microglobulin (beta-2M) serum levels were measured in 143 breast cancer patients and 43 controls. The patients were divided into three groups. A: new patients; B: patients on long term follow-up and C: metastatic patients. In all patients the mean +/-1SD sICAM-1 and beta-2M serum levels were significantly higher than normal controls (p <0.001 and p <0.0001, respectively). Analysis of the three groups showed that for both sICAM-1 and beta-2M, Groups A and C had similar serum levels, which differed significantly from Group B. sICAM-1 of Group B was similar to controls, while beta-2M of Group B was significantly higher than controls. During the study 20 patients relapsed. Initial high levels of sICAM-1 were observed in 20% and beta-2M in 50% of patients. These data suggest that sICAM-1 and beta-2M level indicate host cell-mediated immunity against tumor.

The expression of HLA class I antigens was studied using the immunoperoxidase technique on 30 patients with prostate cancer and 29 patients with benign prostatic hypertrophy (BPH). Forty-three percent of the tumors stained positive, in contrast to 21% of the BPH. An inverse relationship was noted between class I expression and degree of tumor differentiation: 88% of the well-differentiated tumors (8/9) expressed class I antigen, compared with 33% (3/9) of the moderately differentiated and 16% (2/12) of the poorly differentiated tumors. No significant survival difference was found between those with class I-positive and -negative tumors. However, when both class I expression and degree of differentiation were considered, those with positive tumors at each level of differentiation had better survival than those with negative tumors. These data suggest that HLA class I expression may serve as a finite prognostic factor and may have relevance in future immunotherapy.

Background - TPA is marker of proliferation, in contrast to CA-15.3 and beta-2m which are markers of differentiation and tumor bulk. We have compared TPS to CA-15.3 in breast cancer patients.Methods - Serum levels of TPS, CA-15.3 and beta-2 microglobulin (beta-2m) were measured in 120 breast cancer patients and in 59 controls, using ELISA assays.Results - There was a significant difference between patients and controls with respect to the 3 markers. When new patients were compared with patients on long-term follow-up and patients with metastatic disease, a significantly higher TPS level was found in the last group than in new patients and in those on follow-up. Similar pattern was also observed for CA-15.3. For beta-2m the new patients had similar levels to the metastatic patients and both differed significantly from those on long-term follow-up. In metastatic patients TPS was elevated in 85% compared with 64% for CA-15.3, and 67% for beta-2m. Twenty patients from the new and follow-up groups developed metastasis. Although elevated levels of the markers were seen in some patients at diagnosis of the disease, these levels could not predict relapse.Conclusions - TPS measurement may be useful in monitoring the course of the disease and identifying patients who develop metastases in whom CA-15.3 is not elevated. This will enable the response to therapy to be followed.

We measured soluble CD8 ((S)CD8) and microglobulin (beta-2M) in 128 breast cancer patients and in 200 controls by the ELISA method. Patient groups were: Group A-new patients; Group B-patients on follow-up; Group C-patients with metastases. The mean (S)CD8 and beta-2M were significantly higher in patients than in controls ((S)CD8 p<0.01, beta-2M p<0.0001). Both for (S)CD8 and beta-2M, groups A and C had high levels which differed significantly from Group B ((S)CD8 p<0.04; beta-2M p<0.0001). A significant correlation between (S)CD8 and B-2M was observed (r=0.379: p=0.0001). Twenty patients relapsed. In 14/20 (70%) an initial high (S)CD8 and in 10/20 (50%), high beta-2M was observed. High initial CD8 and beta-2M were able to identify 80% of relapsed patients. High (S)CD8 and beta-2M levels are indicative of tumor bulk and are able to identify patients at.