Markov-modeling for the administration of platinum analogues and paclitaxel as first-line chemotherapy as well as topotecan and liposomal doxorubicin as second-line chemotherapy with epithelial ovarian carcinoma.

PURPOSE: So far there is no analysis available on the cost effectiveness of the paclitaxel/platinum-analogue combination versus carboplatin monotherapy with ovarian cancer. Up-to-now only a cost-utility analysis on ovarian carcinoma has been published (Ortega et al. in Gynecol Oncol 66(3):454-463, 1997), which in addition to the first-line chemotherapy included second-line chemotherapy with effectiveness and cost data in the analysis. Therefore, within the scope of our study the cost effectiveness of platinum analogues and paclitaxel as first-line chemotherapy as well as topotecan and liposomal doxorubicin as second-lie chemotherapy was to be determined with epithelial ovarian carcinoma. METHODS: For this purpose a decision-making Markov model was developed which represents the medical and economic consequences of the administration of paclitaxel and platinum derivatives in first-line chemotherapy and the administration of topotecan and liposomal doxorubicin in second-line chemotherapy in the treatment of epithelial ovarian carcinoma by means of data from the literature. Patients were treated either in the early (FIGO stage I-IIa) or advanced stage (FIGO stage IIb-IV). RESULTS: The therapeutic strategy caboplatin followed by topotecan costs 20,123.91 euros, the therapeutic strategy carboplatin followed by liposomal doxorubicin 22,336.57 euros, the therapeutic strategy carboplatin/pactlitaxel followed by liposomal topotecan 29,820.64 euros and the therapeutic strategy carboplatin/paclitaxel followed by liposomal doxorubicin 31,560.47 euros from the time of diagnosis until death or survival within 5 years. With lives saved, accordingly of 2.55, 2.70, 2.60 and 2.65 years' costs amounted to 7,891 euros, 8,270.35 euros, and 11,453.62 euros per year of life saved. CONCLUSIONS: Based on the threshold value of social willingness to pay 45,500 euros per year of life saved, the therapeutic strategy carboplatin followed by topotecan, the therapeutic strategy carboplatin followed by liposomal doxorubicin, the therapeutic strategy carboplatin/paclitaxel followed by topotcan and the therapeutic strategy carboplatin/paclitaxel followed by liposomal doxorubicin can be evaluated to be cost effective.

Antioxidant activity of hydroalcoholic leaf extract of ocimum sanctum in animal models of peptic ulcer.

In the present study, a hydroalcoholic extract of ocimum sanctum leaves has been investigated for its antioxidant activity in animal models of peptic ulcer with the aim of exploring a possible correlation between its antioxidant and antiulcer activities. Gastric ulcers were produced in rats by ethanol treatment and pyloric ligation whereas duodenal ulcers were produced in guinea pigs by histamine treatment. The animals were divided into six groups of six animals each in all these three models of peptic ulcer. Group I served as diseased control in which distilled water (10 ml/kg) orally was administered as placebo. Group II, III and IV received the test drug (ocimum sanctum leaf extract) in doses of 50 mg/kg, 100 mg/kg and 200 mg/ kg respectively orally once daily for 7 days. Group V was administered ranitidine (10 mg/kg orally) once daily for 7 days and served as standard for comparison. Group VI consisted of healthy control for baseline malondialdehyde (MDA) and superoxide dismutase (SOD) levels. The antioxidant activity was by evaluated estimating plasma MDA in ethanol treated rats and histamine treated guinea pigs and estimating SOD in pyloric ligated rats and histamine treated guinea pigs. In ethanol treated rats, ocimum sanctum leaf extract (100 mg/kg & 200 mg/kg) significantly decreased the levels of MDA to 2.45 +/- 0.29 nmole/ml and 2.40 +/- 0.14 nmole/ml respectively in comparison to 4.87 +/- 0.06 in the diseased control. Similarly, in the histamine treated guinea pig group, the same doses of the extract significantly lowered the levels of MDA to 2.45 +/- 0.12 nmole/ml and 2.37 +/- 0.16 nmole/ml respectively when compared to 4.66 +/- 0.11 in the diseased control. The extract (100 mg/kg & 200 mg/ kg) also increased the levels of SOD in pyloric ligated rats to 1.78 +/- 0.12 U/ml and 1.89 +/- 0.08 U/ml respectively when compared to 1.29 +/- 0.06 U/ml in the diseased control. In the histamine treated guinea pig group also, the same doses of the extract produced a rise in the SOD levels to 2.10 +/- 0.11 U/ml and 2.20 +/- 0.14 U/ml respectively when compared to 1.32 +/- 0.07 in the diseased control. Since lowered levels of MDA and increased levels of SOD signify antioxidant activity, the antiulcer activity of ocimum sanctum might be due to this mechanism.

Risk of malignancies in patients with insulin-treated diabetes mellitus: results of a population-based trial with 10-year follow-up (JEVIN).

BACKGROUND: Studies involving diabetes mellitus and malignancies show contradictory results: Many of them have found incidences of malignancies that are comparable or lower, other studies have found higher rates than those of non-diabetic subjects. Hence, the goal of the present trial was to study the possible association between diabetes mellitus and the incidence of malignancies and its outcome in a selection-free population over a longer period of time. - PATIENTS AND METHODS: All the patients (n=291) who participated in the JEVIN (Jena's St. Vincent)- trial (a prospective, 10 year follow-up, population-based intervention survey of all insulin-treated patients with type 1 and type 2 diabetes mellitus aged 16 to 60 years and living in the city of Jena [about 100,000 inhabitants], Thuringia, Germany) were assessed. The baseline examination took place in 1989/90, follow-up examinations were performed in 1994/95 and 1999/2000. - RESULTS: Up to 1999/2000, 2 patients with type 1 and 5 patients with insulin-treated type 2 diabetes mellitus developed a malignancy (incidence 0.0241). The most frequent malignancies were cancer of the colon and rectum (3 of 291 patients, incidence 0.0103). Comparing these data with the incidence of carcinoma of the colon and rectum reported by the Robert-Koch-Institute of Germany (incidence 0.0012) diabetic patients showed a 9.9-fold increased risk (p=0.042). There were no significant differences regarding incidence of total malignancies or carcinoma of the breast, the lung, renal cells or gonads. Correlation and multivariate analyses revealed no associations between the development of malignancies and patients' outcome and diabetes duration, the duration of insulin therapy, insulin dosage, the quality of diabetes control or the presence of diabetes-related long-term complications. - CONCLUSIONS: Conclusive to other data derived from selected cohorts, population-based the JEVIN-trial demonstrate an increased incidence of malignancies of the colon and rectum in insulin-treated patients with diabetes mellitus. However, for further confirmation of these interesting results more studies in larger populations over longer periods of time are necessary to explain the heterogeneous findings in patients with diabetes mellitus of an increased incidence for some cancer entities, but not for others. Knowledge of these mechanisms should have important implications for the direction of strategies to prevent the development of malignancies, or to enhance ability to make an earlier diagnosis and more effective therapies.

Phase I clinical and pharmacokinetic study of combination chemotherapy with topotecan and ifosfamide in patients with progressive or relapsed solid tumors.

PURPOSE: Topotecan and ifosfamide are effective in the treatment of various solid tumors. Up to the time of this study, the two drugs have been combined just once (Smith et al. 1998). Due to its hematotoxicity, topotecan has been used predominantly within monochemotherapy protocols. However, the combination of topotecan and alkylating agents is supra-additive in many preclinical models. This phase-I trial was primarily performed to evaluate the maximum tolerable dose (MTD) and dose-limiting toxicity (DLT) of a combination chemotherapy with topotecan and ifosfamide using a 5-day schedule. A secondary goal was to estimate the response rate in a group of heavily pretreated patients with advanced solid tumors. METHODS: The pharmacokinetics of topotecan were preliminarily determined in some of the patients. A total of 12 patients (three female/nine male), median age 49 years (range 19-69), 11 with prior chemotherapy, received a total of 24 courses of chemotherapy at three dose-levels of topotecan. Ifosfamide was administered by intravenous infusion over 3 h immediately followed by a 30-min infusion of topotecan. Mesna (3 x 300 mg x m(2) x day) was given routinely during chemotherapy as a uroprotector. G-CSF (filgrastim) was permitted only in cases of febrile neutropenia (FN). RESULTS: The major toxicity was non-hematologic; severe liver and renal toxicity were observed in three out of 11 patients. Two treatment-related deaths occurred. No clinical remissions occurred in 11 evaluable patients. The pharmacokinetics of topotecan were relatively similar in our patients and supported findings in recent literature. The MTD of this combination was defined at dose-level 2 (1.0 mg/m(2) of topotecan and 750 mg/m(2) of ifosfamide). CONCLUSIONS: Further trials should not exceed this dose. The pharmacological causes for the pronounced toxicity have to be clarified.

Isolated brain metastases as the sole manifestation of a late relapse in breast cancer.

We report on a 62-year-old female patient suffering from breast cancer (invasive ductal, premenopausal, estrogen- and progesterone-receptor status unknown) first diagnosed in July 1991. After mastectomy and axillary lymphonodectomy (pT2 N2 (12/15) M0 G2), adjuvant chemotherapy consisting of six courses epirubicin and cyclophosphamide was performed. Since 1993, serum Ca 15-3 levels began to climb slowly. As of June 1996, Ca 15-3 was measured 50.1 U/ml (normal value <28.0 U/ml). In routine follow-ups, Ca 15-3 increased slowly, but no metastases of the breast cancer could be detected. In September 2000, the patient attended a routine follow-up examination with poor performance status, presenting with ataxia and a psychic syndrome. Computed tomography revealed multiple suprasellar brain metastases. We conclude that this is an unusual case of a very late clinical manifestation of sole brain metastases in a patient with breast cancer. A slow preceding increase of Ca 15-3-level over a period of 7 years was the only indicator of the upcoming brain metastases.

Regression of brain metastases from breast carcinoma after chemotherapy with bendamustine.

PURPOSE: Bendamustinehydrochloride (bendamustine) is an alkylator with anticipated antimetabolic activity. It has shown activity in malignant lymphoma and breast cancer. Up to now there are no reports about the activity of bendamustine in the treatment of brain metastases. We report on a 38-year-old woman with brain metastases from breast cancer. The first diagnosis of an invasive ductal, hormone receptor negative breast cancer was made in October 1997, stage pT1c pN2 (11/11)cM0 - G3. After lumpectomy and resection of axillary lymph nodes, the patient received adjuvant chemo- and radiotherapy. Twenty six months after the first diagnosis, bone metastases occurred and were treated with radiotherapy. One month later, multiple liver metastases developed which were treated with trastuzumab and paclitaxel. Four months later, progress of the liver metastases and malignant infiltration of bone marrow with thrombopenia occurred. METHODS: Chemotherapy with bendamustine at a dose of 150 mg/m(2) on day 1 and 2 was initiated. Two days later, the patient suffered from central facial palsy and subsequent computed tomography (CT) revealed three brain metastases in the frontal, parietal and occipital region. Because of the advanced liver metastases with clinical and laboratory signs of liver insufficiency and the reduced performance status of the patient, chemotherapy with bendamustine was continued and no local treatment of the brain metastases was performed. RESULTS: After two courses of bendamustine, ultrasound showed regression of the liver metastases. Liver enzymes decreased, platelets increased, and the patient's performance status improved. Additionally, two of the three brain metastases were no longer detectable by CT, the third had decreased compared to the time of diagnosis. CONCLUSION: To the best of our knowledge, this is the first report describing major activity of bendamustine in cerebral metastases. Thus, it may be considered as another therapeutic strategy against metastatic brain cancer. However, this finding warrants further investigation in clinical trials.

[Bendamustine, vincristine, prednisolone (BOP) in therapy of advanced low-grade non-Hodgkin lymphoma]]. / Bendamustin, Vincristin, Prednisolon (BOP) in der Therapie von fortgeschrittenen niedrig malignen Non-Hodgkin-Lymphomen.

BACKGROUND AND OBJECTIVE: Low grade non-Hodgkin lymphomas (l-NHL) are rarely showing complete or sustained remissions to conventional chemotherapy. Thus, many therapeutic strategies try to improve the remission rates and outcome in relapsed and refractory l-NHL. Bendamustine (B) is a non-cross resistant alkylating agent shown to be highly effective in lymphoproliferative and other malignant diseases. In an open phase-II study we evaluated the efficacy and toxicity of B in combination with vincristine (O) and prednisolone (P) in heavily pretreated relapsed or refractory l-NHL. PATIENTS AND METHODS: 22 patients (median age 61.5 years, range 39-77 years) with relapsed or refractory low grade NHL: immunocytoma (IC) n = 11, centroblastic-centrocytic (CB-CC) n = 6, centrocytic (CC) n = 2, others n = 3, were treated with BOP as follows: patients up to 75 years: 60 mg/m2 B for 5 days; patients over 75 years: 50 mg/m2 B for 5 days. All patients received 2 mg vincristine (O) on day 1, 100 mg/m2 prednisolone (P) on day 1-5; repetition day 29. Prior to BOP patients were pretreated with 1-4 chemotherapy protocols. An average of 5 courses of BOP were administered (range 2-8). In most patients BOP was followed by a maintenance therapy (IFN-alpha n = 11, chlorambucil n = 4, etoposide n = 2). RESULTS: Objective remission was achieved in 19/22 (86%) patients, complete remission (CR) in 10/22 (45%), partial remission (PR) in 9/22 (41%) and no change (NC) in 3/22 (14%) patients. The mean duration of remission was 16.1 months. Predominant features of side effects of the BOP protocol were myelotoxicity of WHO grade III/IV in 8 of 109 cycles leukopenia (8%), thrombocytopenia 3 cyles (3%) and anaemia in 4 cycles (4%). We observed one WHO grade IV infectious episode. Other side effects were mild and rare. There was a decline of the CD4/8 in more than 50% of patients. However, these changes were not accompanied by a higher rate of infectious episodes. CONCLUSION: Salvage therapy of refractory and relapsed l-NHL with BOP results in a high objective remission rate. Together with a maintenance therapy most patients achieved a long-term disease-free survival. Myelotoxicity and the inversion of the CD4/CD8 ratio were frequently observed side effects.

Bendamustine monotherapy in advanced and refractory chronic lymphocytic leukemia.

Bendamustine, an alkylating agent without cross-resistance to cyclophosphamide is active in a variety of lymphoproliferative and other malignancies. In an open phase-II study we treated 23 patients with a median age of 62 years at study entry (43-86 years) with advanced, refractory or relapsed (Rai stage III n = 9, Rai stage IV n = 14) chronic lymphocytic leukemia (CLL) with bendamustine. At study entry, only 13 patients were chemotherapy-naive. The treatment schedule with bendamustine was as follows: for patients up to 70 years 60 mg/m2 for 5 days, for patients over 70 years 50 mg/m2 for 5 days, repetition at day 29. Remission criteria were used according to Cheson et al. (1996). All patients were evaluable for toxicity and 20 for response. An objective remission was achieved in 15/20 patients (75%), including six patients with complete remission (CR). Three of the complete responders had no chemotherapy prior to bendamustine. No change (NC) occurred in 5/20 patients (25%). Median overall survival after bendamustine treatment is 13.6 months (1-46 months) and 16.6 months (1-46 months) in patients responding to bendamustine. In total, 74 courses of bendamustine were applied. Therapy-related anemia and thrombocytopenia were rare. However, WHO grade III/IV leukocytopenia occurred in 38/74 cycles (51%), resulting in treatment-related mortality in 3/23 patients (13%). These patients were severely immunocompromised due to pretreatment or the underlying disease. As a corollary of the study, a general prophylactic antibiotic treatment (trimethoprim/ sulfamerazine) was instituted. A general feature was the decline of the CD4/CD8 ratio: mean before therapy: 1.36; after two courses: 0.98; after four courses: 0.6, as documented in all patients who received at least two courses of bendamustine (n = 12). All evaluable patients showed a decline in the CD4/8 ratio. However, this decline was not clearly related to an increased risk of infectious episodes. We observed mainly cutaneous allergic reactions (three WHO grade I; one WHO grade II) leading to a cessation of bendamustine treatment in 4/23 patients (18%). Bendamustine is highly effective in advanced or refractory CLL. In multiple pretreated or otherwise severely immunocompromised patients bendamustine might lead to additional immunosuppression with subsequent infectious complications.

[Cardiac manifestations of peripheral primitive neuroectodermal tumor (pPNET): a rare case]. / Kardiale Manifestation eines peripheren primitiven neuroektodermalen Tumors (pPNET)--eine Rarität.

HISTORY: A 44-year-old man presented to his general practitioner with increasing exertional dyspnoea. After a syncope he was admitted to hospital. INVESTIGATIONS: Echocardiography on admission demonstrated a large pericardial effusion (45 mm over the right and 57 mm over the left ventricle) which, in view of its haemodynamic relevance, had to be removed by pericardiocentesis. Cytological examination of the fluid showed various blood constituents but no atypical cells. Computed tomography (CT) of the thorax revealed a homogeneous hyperdense structure in the area of the left atrium. After several more pericardiocenteses echocardiography showed a pericardial tumour between the left atrium and ventricle. Gastroscopy, coloscopy, abdominal CT, an octreotide scan and positron-emission tomography did not indicate an extracardiac tumour. Coronary angiography and video-assisted thoracoscopy were performed. They confirmed an epicardial tumour at the level of the left atrium. TREATMENT AND COURSE: The entire round and solid tumour of about 3 cm in diameter, at the tip of the left atrial appendage was removed under extracorporeal circulation. No adjuvant treatment was given. Histological and immunohistological tests of the resected specimen (CD-99 and NSE-positive) provided the diagnosis of malignant peripheral primitive neuroectodermal tumour (pPNET). CONCLUSION: There are at present no adequate data on optimal treatment of pPNET in the heart. This case demonstrates that even with a large such tumour long remission is possible after complete removal without any adjuvant treatment.

Malignancies in patients with insulin-treated diabetes mellitus.

In patients with diabetes mellitus, contradictory results have been reported indicating both increased and reduced risks of malignancies. In the present trial all insulin-treated diabetic patients (n = 2720) attending our centre since 1995 were studied. Of these patients, 28 (type 1/type 2: n = 1/27, 23 women) developed malignancies during insulin therapy: 11 patients developed cancer of the breast, 4 patients cancer of the pancreas, 3 patients cancer of the kidneys and 10 patients developed other malignancies. The characteristics of these patients [mean +/- SD (range)] were as follows: age 68.8 +/- 8.6 (52.0 87.0) years, diabetes duration 13.1 +/- 8.1 (0.5-29.0) years, duration of insulin therapy at the time of the diagnosis of malignancy 4.3 +/- 5.7 (0.5 24.0) years, insulin dosage 0.67 +/- 0.43 (0.11-1.72) IU/kg body weight, mean HbAlc 9.6 +/- 1.9 (6.8-14.9)% (HPLC, Diamat, normal range 4.4%-5.9%). The prevalences of nephropathy, retinopathy (non-proliferative: n = 7) and peripheral neuropathy were 35.7%, 25.0% and 46.4% respectively. When the features of the 27 patients with type 2 diabetes were compared with the characteristics of the type 2 diabetic patients (n = 117, 63 women) studied in a population-based survey of insulin-treated diabetic patients, also performed in the area of Jena [JEVIN; Schiel R et al. (1997a)] there were no significant differences in the duration of insulin therapy (JEVIN: 4.7 +/- 4.3 years, P = 0.64), insulin dosage (JEVIN: 0.55 +/- 0.27 IU/kg body weight, P = 0.08), mean HbAlc (JEVIN: 9.0 +/- 2.1%, P = 0.16) and the prevalences of long-term complications of diabetes. The quality of diabetes control in insulin-treated patients suffering from malignancies is comparable to that of a selection-free population of diabetic patients. Furthermore, in comparison to non-diabetic subjects our diabetic patients showed no altered risk for malignancies as a function of insulin dosage, the duration of diabetes or insulin therapy, the quality of diabetes control or the prevalence of long-term complications of the disease.

Gastric cancer in very young adults: apropos four patients and a review of the literature.

Whether gastric cancer in young adults differs from gastric cancer in older patients has been a controversial issue. It has long been suspected that young patients with gastric cancer have different biological features with a more aggressive course of disease and a poorer prognosis than older patients. This, however, has not been firmly substantiated. We report on the clinical course of four patients (three female and one male) with locally advanced (n = 1) or metastasized (n = 3) non-resectable gastric cancer diagnosed under the age of 29 years (23, 25, 27, 28 years). Prior to diagnosis, all three women had recently been pregnant (1-22 months). Diagnosis was endoscopically biopsy-proven and staging work-up was performed by primary explorative surgery (n = 1), laparoscopy and explorative surgery (n = 1) or CAT scan and ultrasound (n = 2). The delay between initial symptoms and diagnosis was 8-22 weeks (median, 10 weeks). The histology was signet-ring cell (n = 2) or undifferentiated (n = 2) gastric cancer. All patients had the diffuse type of gastric cancer according to Lauren. Patients were treated with the FLAP polychemotherapy regimen consisting of leucovorin, 5-fluorouracil, doxorubicin and cisplatinum, as previously reported. The best response after chemotherapy was partial in two patients. Two patients showed progressive disease. Secondary surgery was performed in three responding patients (one of them responded only locally). One patient achieved no evidence of disease after complete tumor resection (R0). In two patients surgery was palliative (R2/exploration). Three patients died 6, 4 and 8 months after diagnosis. One patient is still alive. In our series, very young adults with gastric cancer had adverse clinical and pathological features. In accordance with other reports, we observed a predominance of female patients and a possible association with recent pregnancies. Though the delay between the first symptoms and diagnosis in our patients was no different from that reported for older patients, special emphasis should be given to prompt referral and diagnostic investigations, ensuring the diagnosis of gastric cancer early in the course of disease.

Analysis of Wilms tumor gene (WT1) expression in acute leukemia patients with special reference to the differential diagnosis between eosinophilic leukemia and idiopathic hypereosinophilic syndromes.

Continuous Wilms' tumor gene (WT1) expression is a typical feature of leukemic blasts in AML, ALL, and blast crisis CML patients. It is easily detectable by a variety of RT-PCR protocols, which differ mainly in their sensitivity. The nuclear WT1 protein can be found in blasts of approximately 50-60% of acute leukemia patients at diagnosis. Conversely, WT1 is only transiently expressed in normal hemopoiesis. Early CD34+ hemopoietic progenitors express WT1, whereas no WT1 mRNA transcripts can be found in mature blood cells and differentiation-induced committed CD34- progenitors. As a powerful complementary diagnostic tool, testing for WT1 expression can be helpful to discriminate between eosinophilic leukemia (EoL) patients and patients with idiopathic hypereosinophilic syndromes. Conflicting data about the usefulness of testing for WT1 expression to monitor minimal residual disease (MRD) in treated leukemia patients will be discussed. Finally, research strategies to circumvent shortcomings in detecting leukemia-associated WT1 expression will be outlined.

The significance of somatostatin analogues in the antiproliferative treatment of carcinomas.

Somatostatin is a cyclic tetradecapeptide hormone. It was initially isolated from bovine hypothalami. Somatostatin inhibits endocrine and exocrine secretion, as well as tumor cell growth, by binding to specific cell-surface receptors. Its potent inhibitory activity is limited, however, by its rapid enzymatic degradation and the consequently short plasma half-life. Octreotide is a short somatostatin analogue with increased duration of action compared with somatostatin. Preclinical studies have focused on the anticancer effects of octreotide and the related somatostatin analogues. In vitro, at nanomolar concentrations, these analogues inhibit the growth of tumor cells that express high-affinity somatostatin receptors. Accordingly, such analogues potently inhibit the growth of somatostatin receptor-positive tumors in various rodent models. The range of cancers susceptible to octreotide and related somatostatin analogues includes mammary, pancreatic, gastric, colorectal, prostate, thyroid, and lung carcinomas. Moreover, an indirect antiproliferative effect of somatostatin analogues is achievable in somatostatin receptor-negative tumors whose growth is driven by factors (e.g., gastrin, insulin-like growth factor-1) that become down-regulated by somatostatin. The clinical effect of somatostatin analogues in terms of tumor response in cancer patients is a subject of controversy, however. Most responses have been seen in patients with pancreatic cancers.

The role of LH-RH analogues in the adjuvant and palliative treatment of breast cancer.

Ovarian ablation has a long-standing history in the treatment of metastatic breast cancer. At the end of the past century, several reports appeared on the beneficial effect of surgical removal of the ovaries in premenopausal women with advanced breast cancer. Subsequently, this treatment became the standard systemic hormone therapy for such patients. However, the disadvantage is that only a minority of patients respond, and the remainder suffer unnecessary treatment morbidity. In a random premenopausal patient population, oophorectomy produced an average 33% objective response rate. In the search for alternatives, analogues of luteinizing hormone-releasing hormone (LH-RH) have been thoroughly investigated in pre- and postmenopausal women. The results obtained have established a place for LH-RH agonists in the treatment of premenopausal women with metastatic breast cancer. In addition, these substances have replaced oophorectomy where indicated in the adjuvant hormonal treatment of premenopausal breast cancer.

Severe non-haematological toxicity after treatment with gemcitabine.

The authors report that 4 out of a series of 56 patients (7.1%) treated with gemcitabine developed an unexplained non-cardiogenic pulmonary distress syndrome most likely related to gemcitabine. One further patient developed ventricular arrhythmia immediately after gemcitabine exposure, leading to cardiac arrest. Between 1995 and 1998 56 patients with locally advanced or metastatic carcinoma were treated with gemcitabine. The patients suffered from breast cancer (n = 17), pancreatic cancer (n = 17), lung cancer (n = 12), cancer of unknown primary (n = 5), ovarian cancer (n = 2), oral cavity cancer (n = 2) and cancer of the bladder (n = 1). Their median age was 55 years, and there were 33 female and 23 male patients. Fifteen patients had been pretreated with radiation therapy: 2 had received radiation therapy involving the mediastinum as treatment for non-small-cell lung cancer and cancer of unknown origin respectively, 11 patients had had prior neoadjuvant or adjuvant radiation therapy of the chest wall for breast cancer and 2 patients had received radiation therapy for head/neck cancer. All patients received gemcitabine on days 1, 8 and 15 and this was repeated on day 29 at a dose of 1000 mg/m(2) as a 30-min infusion in 250 ml isotonic NaCl. In 4 patients gemcitabine treatment was combined with cisplatinum, in 7 patients with a somatostatin analogue and in 1 patient with epirubicin. All other patients received gemcitabine as a single agent. We assume that the pulmonary or cardiac toxicity of 5 patients was related to gemcitabine. In 3 patients re-exposure resulted in repeated toxicity. One patient did not receive gemcitabine again because of the life-threatening nature of the primary response. Two patients had received prior radiation to the mediastinum with 62 Gy and 50 Gy respectively, 3 years and 1 year before gemcitabine application. In our experience pulmonary toxicity after gemcitabine treatment is more common than initially anticipated. Gemcitabine should be used with caution in patients who have received prior radiation to the mediastinum.

A phase II study of weekly high-dose 5-fluorouracil and leucovorin plus biweekly alternating doxorubicin and cisplatin for advanced gastric carcinoma.

On the basis of recent clinical data suggesting that high-dose continuous 5-fluorouracil (5-FU) is able to overcome resistance to 5-FU bolus application in gastric carcinoma, a phase II study was performed to evaluate the activity and toxicity of weekly high-dose 5-FU and leucovorin plus biweekly alternating doxorubicin and cisplatin as the first-line treatment in patients with advanced gastric carcinoma. Between October 1995 and September 1997, 24 consecutive patients with locally advanced (n=4) or metastatic (n=20) gastric carcinomas were treated with a combination of 500 mg/m2 leucovorin as a 2-h infusion, followed by 2.0 g/m2 5-FU as a 24-h continuous infusion once weekly for 6 weeks, plus 20 mg/m2 doxorubicin as a bolus application and 50 mg/m2 cisplatin as a 1-h infusion, week 1, 3 and 5 (FLAP regimen). Response, toxicity and survival data were evaluated. A total of 20 patients were evaluable for response and 24 for toxicity. Objective responses were observed in 11 patients (55%) with no complete remission. Four patients (20%) showed stabilization and 5 patients (25%) experienced progressive disease. The median time to disease progression was 8 months and the overall duration of survival was 14 months. Myelosuppression was significant. In 2 patients, grade 4 WHO thrombocytopenia and leukopenia/anaemia respectively were registered, but there were no treatment-related deaths. We conclude that the weekly alternating FLAP regimen is effective in advanced gastric carcinoma with tolerable toxicity. However, significant myelotoxicity and frequent hospitalization suggest that FLAP should not be preferred to other regimens used in metastatic disease. Currently we intend to establish this regimen in the neoadjuvant setting in patients with primary unresectable localized gastric carcinomas.

Correlation between granulocyte/macrophage-colony-forming units and CD34+ cells in apheresis products from patients treated with different chemotherapy regimens and granulocyte-colony-stimulating factor to mobilize peripheral blood progenitor cells.

We examined the efficiency of disease-specific "standard" chemotherapies epirubicin, cyclophosphamide (EC); cyclophosphamide, vincristine, doxorubicin, etoposide, prednisolone (CHOEP); epirubicin, ifosfamide (EPI/IFOS) for peripheral blood progenitor cell (PBPC) mobilization in comparison to well-characterized mobilization protocols, i.e. etoposide, ifosfamide, cisplatin, epirubicin (VIPE) and dexamethasone, carmustine, etoposide, cytarabine, melphalan (DexaBEAM). Twenty-seven patients with various malignancies underwent 75 apheresis procedures for PBPC collection. Median cell yields from all 75 aphereses were 1.18 x 10(5) mononuclear cells/kg [range (0.28-3.7) x 10)8)], 1.4 x 10(5) granulocyte/macrophage-colony-forming units (CFU-GM)/kg [range (0.2-11) x 10(5)] and 3.3 x 10(6) CD34+cells/kg [range (0.35-17.7) x 10(6). CD34+/ CD90+ cells could be mobilized by all mobilization regimens used. The difference observed in the mobilization of CD34+ cells was only of low significance when the mobilization regimens were compared, whereas the mobilizations of MNC and CFU-GM were significantly different between the groups. Breast cancer patients treated with the VIPE regimen (including pretreated women) had a significantly higher CFU-GM rate than patients treated with EC (P=0.0005). Mobilized CD34+ PBPC were correlated with CFU-GM in all apheresis products. The linear correlation coefficients differed for the various mobilization groups: DexaBEAM (r=0.9, P < 0.0001), VIPE (r=0.68, P=0.0024), CHOEP (r=0.52, P=0.022), EPI/ IFOS (r=0.34, P=0.11) and EC (r=0.23, P=0.2). We conclude that clonogenic assays can provide additional information about the autotransplant quality, particularly when alternative or new mobilization regimens are being investigated.

Pharmacoeconomic evaluation of high-dose chemotherapy and peripheral blood stem cell support in high-risk or poor-prognosis malignancies.

Discussion of the total costs and cost-effectiveness ratios of patients receiving high-dose chemotherapy (HDC) and peripheral blood stem cell support (PBSCS) is controversial. In Germany, no reliable data are available, whereas in other countries this issue has been extensively studied. We performed a pharmacoeconomic evaluation on all patients (n = 37) treated with HDC and PBSCS at our institution between July 1994 and June 1997. Patients suffered from high-risk or poor-prognosis breast cancer (n = 24), Hodgkin's disease (n = 3), high-grade non-Hodgkin's lymphoma (n = 4), multiple myeloma (n = 2), small-cell cervical cancer (n = 1), malignant hystiocytosis (n = 1) and testicular cancer (n = 2). For pharmacoeconomic evaluation, the period from initiation of induction chemotherapy (IC) until reconstitution after the last course of HDC and PBSCS was considered. A total of 18 patients received IC/HDC/PBSCS for locally advanced or systemic disease, and 19 patients received adjuvant or consolidation IC/HDC/PBSCS. Treatment protocols were heterogeneous. Patients were treated with two to five courses (median two) respectively of IC and sequential mono-HDC (n = 26), tandem-HDC (n = 10) or triple-HDC (n = 1). All patients received granulocyte/macrophage-colony-stimulating factor (G-CSF) for stem cell mobilisation and for amelioration of neutropenia after HDC. The relative costs (based on supplier prices) for the total amount of drugs prescribed during the in-patient period was 29.8% for G-CSF, 35.8% for blood products 18.5% for chemotherapy, 2.4% for antiemetics, 5.9% for antimicrobial drugs and 7.6% for other drugs. Contrary to expectations, antimicrobial drugs had only a minor pharmacoeconomic impact during IC/HDC/PBSCS in patients with high-risk or poor-prognosis malignancies, indicating that prolonged septic complications were uncommon in our institution. We conclude that pharmacoeconomic evaluations in IC/ HDC/PBSCS might be integrated into the effort to ensure quality control and monitoring.