I. Michl’s research while affiliated with Medical University of Vienna and other places

In order to study adhesion-molecule expression and its consequences for cellular recognition, the presence of adhesion molecules ICAM-1, VCAM-1, VLA-4, LFA-1, alpha, LFA-1 beta, LFA-3, β1-integrin and β3-integrin was studied on specimens from breast tissue by immunohistochemistry and on cells from breast cell lines propagated in vitro. Breast-cancer tissue and the breast-cancer cell lines MCF-7, SK-BR-3 and ZR-75-1 showed expression of ICAM-1 and VLA-4 significantly lower than that of benign breast cells or normal breast epithelium. Of various cytokines tested, including recombinant human (rh) interleukin-6 (IL-6), rh tumor necrosis factor alpha (TNF-α), interleukin 2 (IL-2), granulocyte/macrophage-colony-stimulating-factor (GM-CSF), interferon-alpha (IFN-α) and interferon-gamma (IFN-γ), only TNF was able to re-induce expression of ICAM-1 on cells from MCF-7, SK-BR-3 and ZR-75-1. Further, the ability of either unstimulated or lymphokine-stimulated killer (LAK) cells to recognize and lyse native or TNF-stimulated breast-cancer cells was studied. Whereas neither unstimulated lymphocytes or LAK cells were able to lyse untreated breast-cancer cells deficient for ICAM-1 expression, pre-treatment of tumor cells with TNF led to increased tumor-cell lysis. Anti-ICAM-1 antibodies, and pre-treatment of tumor cells with anti-TNF-receptor antibodies, abrogated these findings, corroborating their specificity. We thus conclude that the defective expression of ICAM-1 in our model might constitute a mechanism by which breast-cancer cells escape immunologic recognition and lysis by appropriate effector cells. Int. J. Cancer 71: 1086-1090, 1997. © 1997 Wiley-Liss Inc.

Numerous clinical studies showed efficacy of biological response modifiers (BRM) in the treatment of advanced RCC. In this phase II study, we investigated efficacy, toxicity, and feasibility of lFN-γ in combinalion with 1L-2 in patients with advanced RCC. One cycle of therapy consisted of 100 mcg lFN-γ sc 3x/week for two weeks, 4.5 MIU 1L-2 sc over four consecutive days for the next two weeks followed by a two week rest period. 3x500 mg/day paracetamol are given orally for the first two weeks of therapy to prevent or mitigate flu-like symptoms. To date 40 patients (25 males/15 females) with a median age of 59 years (range: 44– 81 years) have been accrued to the study. A median of 3 therapy cycles (range 1–9) have been given to these 40 patients (median time of observation: 5 months, range: 1–18 months) and all patients are eligible tor feasibilitv and toxicitv evaluation, and 31 patients for response documentation, respectively. 32 of the patients were trained in self-application of the cytokines und 8 patients preferred the application of therapy through their family doctor. No WHO-grade III or IV toxicity has been documented so far. Side effects consisted of flu-like syndrome grade I/II in 21 patients despite prophylactic paracetamol, local crythemas after 1L-2 application in 18 patients and hypotension grade I mainly after IL-2 application in 4 patients. Myelotoxicity was mild in general and consisted of grade I/ll leucopenia and thrombocylopenia only. No red cell transfusions had to be given. Response data of 32 patients after a median time of observation of 10 months (range: 5–16 months) are as lollows: CR: n=2 (6%), PR: n=3 (9%), SD: n=14 (44%), PD: n=13 (41%). Our preliminary data show that a treatment with lFN-γ/IL-2 is not only associated with low toxicity, but also very practical lor out, patients, and offers therefore a good quality of life. The antitumor activity of this therapy in patients with advanced RCC is proven by the response rate (CR+PR+SD) of 59%.

Adhesion molecules are membrane proteins responsible for cell adhesion and cellular recognition. In order to further our knowledge about the role of adhesion molecules (AM) in the context of breast cancer, the following experiments were designed. By using cell lines ZR-75–1, MCF-7, SK-BR-3, and HBL-100, the expression of AMS, ICAM-1, V-CAM, VLA-4, LFA-1 alpha, LFA-l beta, LFA-3, beta-1-integrin and beta-3-integrin using monoclonal antibodies was analyzed. Cytokines Interleukin-2 and -6, Interferons alpha and gamma, Tumor Necrosis Factor-alpha (TNF) and Granulocyte/Macrophage-Colony Stimulating Factor (GM-CSF) were tested for the ability to induce adhesion molecules. ICAM-1 and VLA-4 were expressed in a significantly lower degree by malignant breast cancer cell lines, as compared to the benign cell line HBL-100. A discordant picture was found concerning the ability of various cytokines to further increase the expression of adhesion molecules: Thus, TNF alpha increased the expression of ICAM-1 in a dose dependent manner on the HBL-100, the SKBR-3 and the ZR-75–1 cell lines, albeit to a different degree. GM-CSF was able to induce ICAM-1 expression in the SKBR-3 cell line only, whereas interleukin-2 induced the expression of ICAM-1 on the MCF-7 cell line. Concerning the expression of ICAM-1 on tumor cells, induced by TNF, lymphokine-activated killer cells result in lysis of target-cells in the presence of monoclonal anti-ICAM-1 antibody. This did not occure in the tumor cells that express ICAM-1 in a much lower degree. However, the defect to express adhesion molecules may constitute a reason for the metastatic potential and the spread of malignant tissue in vivo and the defective recognition and subsequent destruction of tumor cells by the immune system.

Metastatic soft tissue sarcoma constitutes a major therapeutic problem potentiated by the development of drug resistance. The present pilot study was therefore undertaken in order to study the possibility of modulation of drug resistance by tamoxifen in patients with metastatic soft tissue sarcoma after the administration of combined cytostatic therapy consisting of ifosphamide (1.5 g/m2), adriblastin (50 mg/m2) and DTIC (200/m2), all on days 1-4; (IFADIC). 12 patients with soft tissue sarcoma who experienced progression under or following treatment with IFADIC underwent second-line chemotherapy with Epirubicine 60 mg/m2 (days 1 and 22). CCNU 80 mg/m2 (day 1) (ECC) and high-dose Tamoxifen (480 mg/day. day -1 to +1) in an endeavour to modulate drug resistance. Under this very well tolerated, low toxic regimen. I patient with lung metastases experienced complete remission (duration: 5 months +) and 1 patient partial remission (duration: 6 months). 1 final patient experienced stable disease for 12 months thus resulting in an overall response rate of 3 out of 12 patients. We conclude that this low-toxicity regimen might be effective in heavily pretreated patients with soft tissue sarcoma thus making the use of high-dose tamoxifen a possible canditate for modulation of drug resistance in metastatic soft tissue sarcoma.

Purpose The rationale of the present study was to validate the hypothesis of a favorable influence of a perioperative start of adjuvant chemotherapy in patients with operable breast cancer of various prognostic groups, as suggested by results of experimental data. Patients and methods Patients with operable breast cancer were entered into a prospective tria1 and randomized to receive epidoxorubicin (20 mg/sqm) and cyclophosphamide (200 mg/sqm; EC) either on days 1, 8 and 15 (= perioperatively) or on days 22, 29 and 36 (= postoperatively), day one being the day of surgery. Patients with lymph node involvement (N+, pre- and postmenopausal) and premenopausal patients without lymph node involvement (N-) and estrogen receptor (ER) negativity received 3 additional cycles of adjuvant chemotherapy with cyclophosphamide, methotrexate, fluouracil (CMF). All ER⁺ patients received 20 mg tamoxifen per day for two years. Results No increased toxicity or problems with wound healing were found in patients from the perioperative group. With a present median follow-up of 70 months, 76 out of 221 (34%) eligible patients have relapsed. The estimated percentage of disease-free survival (DFS) at 5 years was 66 (±3.2) in all patients, 59 (±5.4) in patients from the perioperative and 72 (±4.2) in those from the postoperative groups (p = 0.09). Conclusion An advantage in DFS resulting from an earlier start of adjuvant chemotherapy than on postoperative day 22 can be excluded at this point of the study.

Lymphedema of the arm is one of the most disabling and serious complications of breast cancer. Apart from tumor infiltration or fibrosis of lymphatic pathways, little is known about factors favoring the development of lymphedema. In the present study, we investigated the impact of rheologic parameters, e.g. red cell aggregation (EA) and plasma viscosity (PV), and of capillary morphology and capillary flow in patients with breast cancer with (n = 18) and without (n = 18) lymphedema. Patients with lymphedema showed a significant increase of red cell aggregation (p < 0.001) that indicates a systemic component of lymphedema and might offer a possibility of prevention and therapy of this condition. A hitherto unclassified protein factor favoring red cell aggregation and lymphedema might be postulated.

Patients with testicular cancer can be cured by cisplatin-based chemotherapy in many cases. Thus, concern about possible late toxicities of treatment is warranted. In this investigation, the absolute number of natural killer (NK) cells according to their CD56+ phenotype, NK cell activity, and antibody dependent cellular cytotoxicity (ADCC) were investigated in 29 patients with seminomas or nonseminomatous germ cell tumors (NSGCT) a median of 31 months (range, 5-73 months) after termination of chemotherapeutic treatment. No difference in the absolute number of NK cells, NK cell activity, and ADCC was found between patients with testicular cancer after either standard polychemotherapy consisting of bleomycin, etoposide, and cisplatin (BEP) or monotherapy with carboplatin and healthy control subjects. When patients were analyzed further using multivariate analysis, a significant (P < 0.05) detrimental influence of BEP polychemotherapy versus carboplatin monotherapy on NK cell activity was found. Moreover, NK cell activity also depended significantly (P < 0.05) on the time elapsed after chemotherapy was administered, but normalized with time. Because the absolute number of NK cells was not affected, is was assumed that polychemotherapy induced a functional defect. In a subanalysis of patients with NSGCTs, metastases at diagnosis resulted in a significant (p < 0.05) and persistent stimulation of NK cell activity but not of ADCC. Cytostatic chemotherapy in patients with testicular cancer did not lead to compromised lytic effector cell function as assessed by NK cell activity and ADCC. However, a short, time-dependent reduction was found that also depended on the intensity of chemotherapeutic treatment. This finding related to the observation of a long-lasting stimulus of NK cell activity by initial metastases in patients with NSGCTs.