The objective of this study was to identify prognostic factors for survival in patients with advanced oesophageal cancer, who are treated with cisplatin-based combination chemotherapy. We analysed the baseline characteristics of 350 patients who were treated in six consecutive prospective trials with one of the following regimens: cisplatin/etoposide, cisplatin/etoposide/5-fluorouracil, cisplatin/paclitaxel (weekly) and cisplatin/paclitaxel (biweekly). Predictive factors in univariate analyses were further evaluated using multivariate analysis (Cox regression). The median survival of all patients was 9 months. The 1, 2 and 5-year survival rates were 33, 12 and 4%, respectively. The main prognostic factors were found to be WHO performance status (0 or 1 vs 2), lactate dehydrogenase (normal vs elevated), extent of disease (limited disease defined as locoregional irresectable disease or lymph node metastases confined to either the supraclavicular or celiac region vs extensively disseminated disease) in addition to the type of treatment (weekly or biweekly cisplatin/paclitaxel regimen vs 4-weekly cisplatin/etoposide with or without 5-fluorouracil). Although weight loss, liver metastases and alkaline phosphatase were significant prognostic factors in univariate analyses, these factors lost their significance in multivariate analyses. The median survival for patients without any risk factors was 12 months, compared to only 4 months in patients with WHO 2 plus elevated LDH and extensive disease. The performance status, extent of disease, LDH and the addition of paclitaxel to cisplatin are independent prognostic factors in patients with advanced oesophageal cancer, who are treated with cisplatin-based combination chemotherapy.

In a phase I study we demonstrated the feasibility of a bi-weekly combination of paclitaxel 180 mg m −2 with cisplatin 60 mg m −2 . In this study we further assessed toxicity and efficacy of this schedule in the treatment of advanced cancer of the oesophagus or the gastro-oesophageal junction. Patients received paclitaxel 180 mg m −2 administered over 3 h followed by a 3-h infusion of cisplatin 60 mg m −2 . Patients were retreated every 2 weeks unless granulocytes were <0.75 × 10 9 or platelets <75 × 10 9 . Patients were evaluated after three and six cycles and responding patients received a maximum of eight cycles. Fifty-one patients were enrolled into the study. The median age was 56 years (range 32–78). WHO performance status were: 0 (19 patients); 1 (29 patients); 2 (three patients). All patients received at least three cycles of chemotherapy and all were evaluable for toxicity and response. Haematological toxicity consisted of uncomplicated neutropenia grade 3 in 39% and grade 4 in 31% of patients. Five patients (10%) were hospitalised, three patients because of treatment related complications and two patients because of infections without neutropenia. Sensory neurotoxicity was the predominant non-haematological toxicity; grade 1 and 2 neurotoxicity was observed in 43 and 20% of patients, respectively. Response evaluation in 51 patients with measurable disease: complete response 4%, partial response 39%, stable disease 43% and progressive disease in 14% of the patients. The median duration of response was 8 months. The median survival for all patients was 9 (range 2–29+) months and the one-year survival rate was 43%. Four patients who received additional local treatment (two patients surgery and two patients radiotherapy) are still disease free after a follow-up of 20–29 months. This bi-weekly treatment of paclitaxel and cisplatin is well tolerated by patients with advanced oesophageal cancer. The toxicity profile of this regimen compares favourable to that of previously used cisplatin- and paclitaxel-based regimens. Trials are underway evaluating this bi-weekly regimen in a neo-adjuvant setting.

We performed this dose-finding study with a fixed dose of cisplatin and increasing doses of paclitaxel given every 2 weeks to determine the maximum tolerable dose of this schedule. Sixty-four patients with advanced oesophageal cancer were treated with a cisplatin dose of 60 mg m(-2) and increasing doses of paclitaxel from 100 mg m(-2) up to 200 mg m(-2) both administered over 3 h for a maximum of six cycles in patients with stable disease or eight cycles in responding patients. Patients were retreated when the granulocytes were > 0.75 x 10(9) l(-1) and the platelets > 75 x 10(9) l(-1). The dose of paclitaxel could be increased to 200 mg m(-2) without encountering dose limiting haematological toxicity. At the dose levels 190 mg m(-2) and 200 mg m(-2) of paclitaxel cumulative sensory neurotoxicity became the doselimiting toxicity. The dose intensity of paclitaxel calculated over six cycles rose from 50 mg m(-2) per week to 85 mg m(-2) per week. Only three episodes of granulocytopenic fever were encountered out of a total of 362 cycles of treatment. Of the 59 patients evaluable for response, 31 (52%) had a partial or complete response. In a biweekly schedule with a fixed dose of 60 mg m(-2) cisplatin it is possible to increase the dose of paclitaxel to 180 mg m(-2). At higher dose levels, neurotoxicity becomes the dose-limiting toxicity. The observed response rate warrants further investigation of this schedule.

We evaluated the use of two tumour markers Cyfra 21.1 and tissue polypeptide antigen (TPA) for disease monitoring. Assessment of response to WHO criteria was compared to response assessment according to changes in the tumour marker levels. The criteria defined for marker response were a 65% decrease for a partial response and a 40% increase for progressive disease. When response evaluations with a positive lead time were included, 72% of 115 evaluations for Cyfra 21.1 and 59% of 107 evaluations for TPA yielded the same result. Most discordant evaluations were caused by those evaluations whereby the patient achieved a partial response according to the WHO criteria and had normalisation of the marker. Less cases with a positive lead time, more negative lead times, and more patients with progressive disease without an increase of the marker were seen with TPA compared to Cyfra 21.1. In conclusion, Cyfra 21.1 follows the changes in the tumour load better than TPA. Rising levels of both markers nearly always indicate disease progression, and such knowledge easily obtained may prevent the continuation of ineffective treatment.

Summary. We report a patient who developed acute myeloid leukaemia (M2) shortly after successful treatment of a mediastinal germ cell tumour. The leukaemia was preceded by a documented myelodysplastic phase. Complex cytogenetic abnormalities were found in bone marrow and peripheral blood cells including +i(12p), typical of germ cell malignancy. Fluorescence in situ hybridization revealed the presence of +i(12p) in myeloblasts, erythroblasts and neutrophils but not in lymphocytes. This case provides further evidence for a common clonal origin of haematological malignancies and mediastinal germ cell tumours.

A total of 24 patients presenting with small-cell lung cancer either resistant to or relapsing within 3 months after first-line treatment were entered in a phase II study of 1-(4-amino-2-methyl-5-pyrimidinyl)-methyl-(2-chloroethyl)-3- nitrosourea hydrochloride (ACNU). ACNU was given i.v. at a dose of 75 mg/m2 every 6 weeks. We observed a partial response of 7 months' duration in one patient and one case of stable disease that lasted for 6 months; all other subjects exhibited progressive disease. Two patients developed brain metastases during treatment. The toxicity of ACNU consisted mainly of bone marrow suppression, especially thrombocytopenia. At this dose and on this schedule, ACNU shows minimal activity as second-line treatment in small-cell lung cancer.

In 28 patients with transitional carcinoma of the urinary tract, all treated with chemotherapy, serial measurements of serum tissue polypeptide antigen (TPA) were performed and correlated to clinical evaluations of response. At the start of chemotherapy elevated levels of TPA were found in 4 out of 14 patients with T2-4NO-2MO tumours and in 7 out of 14 patients with distant metastases. In most patients with elevated TPA levels who responded to chemotherapy, TPA levels rapidly returned to normal. False positive elevations of TPA were observed in 2 patients. It is concluded that serial measurement of TPA for monitoring disease activity has limited value because of the low sensitivity of TPA, especially for patients with early-stage cancer, and because of the occurrence of false positive results.

Between 1987 and 1990 21 patients with proven adenocarcinoma of the pancreas were treated with chemotherapy in four different phase II studies. For 14 patients, serial measurements of CA19-9 serum levels and clinical evaluations of response by computed tomography scan and/or ultrasound were available. Clinical stable disease and progressive disease were accompanied by stable or exponentially rising serum levels of CA19-9. One patient with clinical partial remission showed a 90% decline of CA19-9. However, a 75% decline of CA19-9 was also observed in a patient with rapidly progressive disease. These data seem to indicate that the CA19-9 serum level may be used as an easy and sensitive tool to evaluate progressive disease during chemotherapy.

To determine the incidence and clinical relevance of neuronal autoantibodies in children with demyelinating syndromes.We conducted a prospective study including 31 consecutive children with demyelinating syndromes. Four patients with N-Methyl-D-aspartate receptor (NMDAR) encephalitis, 32 patients with Guillain-Barre syndrome, 13 children with benign childhood epilepsy, and 28 healthy children were used as controls. Prior to initiating immunomodulatory therapy, serum samples were tested for antibodies against NMDAR, α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor (AMPAR) 1, AMPAR2, leucine-rich glioma-activated protein 1, contactin-associated protein 2, gamma-aminobutyric acid B receptors, paraneoplastic ma antigen 2 (PNMA2/Ta), Yo, Ri, Hu, CV2, amphiphysin, and aquaporin-4 by indirect immunofluorescence assays.Three anti-neuronal antibodies were detected; NMDAR antibody in one with multiple sclerosis, PNMA2/Ta antibody in one with multiple sclerosis, and Yo antibody in one with clinically isolated syndrome. The positivity rate of neuronal autoantibodies in demyelinating syndrome was 10%. All seropositive patients were found to be negative for tumor screening. None of these patients exhibited symptoms of encephalitis.Children with demyelinating syndromes without symptoms of encephalitis can be positive for anti-neuronal antibodies.

Serum samples were collected from 115 small-cell lung cancer patients before each course of chemotherapy and during follow-up. Levels of neuronspecific enolase (NSE) were measured and compared to the clinical assessments of the course of the disease, which were done by the responsible physician without knowledge of NSE-values. The predictive accuracy of an increase or decrease of NSE for a major response (CR+PR), SD or PD was 98%. Importantly no falsepositive rises of NSE were observed. On the basis of this large number of data it seems justified to conclude that in common clinical practice the treatment of small-cell lung cancer patients can be monitored by serial measurements of NSE alone.

From 1980 till 1984 16 patients were entered into a non-randomized pilot study, to investigate the feasibility of five courses of adjuvant 5-fluorouracil, Adriamycin and mitomycin C 9FAM) after a curative resection of pancreatic or periampullary cancer. The survival of this group of patients was compared with that of 36 patients who underwent a curative resection alone between 1977 and 1984. Four patients received less than 20%, 4 patients 50%–60% and 7 patients≧80% of the calcuated dose of adjuvant chemotherapy. The chemotherapy was badly tolerated. Only 1 patient resumed some of his normal activity during chemotherapy. The 3-year actuarial survival after curative resection with and without FAM was similar, i.e. 24% and 28% respectively. These data suggest that adjuvant FAM after a Whipple's operation or total pancreatectomy is not feasible because of additive postoperative and chemotherapy-induced morbidity.

Upfront chemotherapy for locally invasive bladder cancer is a reasonable approach in view of the disappointing results with radical surgery (±radiation therapy). The aim is to cure micrometastasis not detectable but present at the time of diagnosis (neoadjuvant chemotherapy). Other goals of this approach are local tumor control and bladder preservation (additional therapy) or cure of the primary tumor and micrometastasis (definitive chemotherapy). This paper offers a critical review of our present knowledge in order to define the questions still open regarding the exact role of this approach in invasive bladder cancer.

Neuron-specific enolase (NSE) was measured in serum samples of 35 patients with small cell lung cancer and 10 control patients. The samples were collected during 10 days after the first course of chemotherapy, in order to investigate whether changes of NSE had a predictive value for tumour response. Three patterns of change of NSE were observed. Pattern 1 showed an increase of serum NSE with a maximum value more than 1.5 times the pretreatment level (n=17); pattern 2 involved no increase at all or less than 1.5 times the pretreatment level (n=14); pattern 3 showed a continuous decrease (n=5). No relationship between the three patterns of change and the tumour response was observed. Only an NSE level <10 ng/ml at the time of start of the second course predicted a major response.