Search

We investigated the activity and toxicity of a combination of vinorelbine 25 mg/m2 on days 1 and 15; epirubicin 25 mg/m2 on days 1, 8, 15; and 5-fluorouracil continuous infusion at 200 mg/m2 every day, administered as first-line chemotherapy in anthracycline-naive metastatic breast cancer patients. Fifty-three patients entered the study. Cycles were repeated every 28 days. Objective response was 60% by World Health Organisation (WHO) criteria and 63% by Response Evaluation Criteria in Solid Tumours (RECIST). The median time to progression was 12.7 months (17.6 months in responders) and the median survival duration was 32.9 months. The dose-limiting toxicity was leucopenia (grade 3/4 in 36% of patients). Grade 3/4 non-haematological toxicities included mucositis in 11% of patients, skin and cardiac toxicity in 4% and 2%, respectively. The combination of vinorelbine, epirubicin and 5-fluorouracil continuous infusion was found to be an active and manageable first-line regimen for metastatic breast cancer patients.

Background: To test safety and activity of 3-weekly doses of docetaxel and a weekly dose of trastuzumab in women with HER2-overexpressing advanced breast cancer., Patients and Methods: Forty-two women, median age 53 years (range 36-73 years), with HER2-overexpressing advanced breast cancer were enrolled in a study of docetaxel, 75 mg/m(2) q3w for 6 cycles, and trastuzumab, 4 mg/kg loading dose, 2 mg/kg weekly thereafter. Thirty-four patients (81%) had visceral metastatic involvement. Thirty-five patients had received prior chemotherapy as part of their treatment: adjuvant/neoadjuvant (26), metastatic (2) and both (7). Thirty-one patients had been previously exposed to an anthracycline and 11 to paclitaxel. Four patients had previously received high-dose chemotherapy followed by autologous stem cell transplant., Results: 226 cycles (median 6, range 1-6) were administered. The median delivered dose intensity for docetaxel was 24 mg/m(2)/week (range 16-25 mg/m(2)/week). The intent to treat overall response rate was 67% (95% confidence interval, 52-79%). Median progression-free survival, time to treatment failure, and duration of response were 9, 8 and 12 months, respectively. Symptomatic cardiotoxicity (grade 3) occurred in 1 patient. The most common grade 3/4 toxicity was neutropenia (76% of the patients), although febrile neutropenia did not occur., Conclusions: Three-weekly doses of docetaxel and a weekly dose of trastuzumab is an active and safe combination in patients with HER2-overexpressing advanced breast cancer., (Copyright 2004 S. Karger AG, Basel)

Background: To evaluate the feasibility and activity of gemcitabine and vinorelbine as a second/third-line approach in patients with advanced breast cancer., Methods: Entered into the study were 51 consecutive patients. All had been previously treated with anthracyclines. Of these 51 patients, 36 had experienced failure or relapse after one chemotherapy line for advanced disease, and 15 after two chemotherapy lines. The dominant sites of involvement were brain in 4 patients (7.8%), liver in 22 (43.2%), lung in 10 (19.6%), bone in 10 (19.6), and soft-tissue in 5 (9.8%). Treatment consisted of vinorelbine 25 mg/m(2) and gemcitabine 1000 mg/m(2) administered on days 1 and 8 every 21 days., Results: The scheme was well tolerated. Grade 3/4 neutropenia was observed in 11% of patients. Grade 3 nausea and vomiting occurred in 6%, and grade 2 neurotoxicity in 6%. No patients experienced grade 3/4 alopecia. The median relative dose intensity was 94.6% (49.7-100%) and 90.0% (23.1-100%) for vinorelbine and gemcitabine, respectively. Two patients (3.9%) were not evaluable for disease response, 4 (7.8%) attained a clinical complete response, 13 (25.5%) a partial response (for an overall response rate of 33.3%, 95% coefficient interval 20.0-46.0%), 23 (45.2%) showed stable disease, and 9 (17.6%) progressed. The median time to progression of responding patients was 10.8 months, and the median overall survival of the entire population was 17.8 months., Conclusions: Vinorelbine and gemcitabine is a manageable scheme with moderate activity in pretreated patients with advanced breast cancer.

Statement of Problem: Fiber-reinforced composites (FRC) are used in direct intra-oral applications as periodontal splints and chairside tooth replacement by bonding them to etched enamel with resin adhesives and composites. There is little information regarding the effect of FRC on the shear bond strength of composite to etched enamel., Purpose: The purpose of this study was to examine the effect of resin preimpregnated and non-preimpregnated fiber-reinforced composites on enamel to composite shear bond strength (SBS)., Material and Methods: Specimen groups (n = 12) consisted of a control (composite with no fiber reinforcement), Ribbond, Splint-It Unidirectional, Splint-It Woven, and Connect, which were bonded to 37% phosphoric acid etched Prime and Bond NT adhesive-treated bovine enamel surfaces on a bed of Tetric Flow composite. Specimens were thermocycled 1000 times between 5 degrees and 55 degrees C and loaded on a universal testing machine in shear at a linear increasing load until fracture (MPa). The fractured surfaces of the debonded specimens were evaluated to determine the nature of the fracture with a light binocular microscope (x10). Shear bond strength data were subjected to analysis of variance (ANOVA) and Student-Newman-Kuels tests (P <.05)., Results: Mean MPa +/- SD for the test groups were as follows: Control, 15.6 +/- 2.4; Splint-It Unidirectional, 15.3 +/- 2.4; Splint-It Woven, 16.5 +/- 1.8; Connect, 18.8 +/- 1.5; and Ribbond, 15.8 +/- 2.2. The Connect FRC group had significantly higher (P <0.05) enamel SBS than all other groups. Fracture analysis showed varying types of failures among the groups, with cohesive fractures within the fiber reinforcement of Splint-It Unidirectional and Connect, cohesive fractures within the bonding resin/flowable composite for Ribbond and the control, and adhesive fracture at the fiber reinforcement interface with Splint-It Woven., Conclusion: Within the limitations of this study, no differences in SBS were observed with the addition of 3 of the 4 FRCs compared to composite without FRC, with the exception of the Connect product which provided significantly higher SBS values.

We conducted a pilot phase II trial of trastuzumab administered concurrently with docetaxel in women with HER2-overexpressing advanced breast cancer. Twenty-five women with HER2-positive (3+ by immunohistochemistry = 16, 2+ = 9) metastatic breast cancer received docetaxel (75 mg/m every 3 weeks for 6 cycles) and trastuzumab (4 mg/kg loading dose, 2 mg/kg weekly thereafter). Twenty-three patients (92%) had visceral metastatic involvement. Twenty-three patients had received prior chemotherapy as part of adjuvant (18), metastatic (2), and both (3) treatment. The number of cycles administered was 121 (median 6, range 1-6). Symptomatic cardiotoxicity (GIII) occurred in one patient. The most common grade GIII/IV toxicity was neutropenia (80% of the cycles), although febrile neutropenia did not occur. No other GIII/IV toxicities were observed. Response rate was 70% (1 complete response and 15 partial responses) in 23 evaluable patients. The combination of docetaxel and trastuzumab is well tolerated and has clinically meaningful antitumor activity.

Purpose: To evaluate the feasibility and activity of vinorelbine in association with protracted infusional fluorouracil in patients with advanced breast cancer who were previously treated with anthracycline-containing regimens., Patients and Methods: Eighty-three consecutive patients were entered onto the study. Forty-three patients experienced treatment failure or relapse after anthracycline-based, first-line chemotherapy for advanced disease and 29 experienced treatment failure or relapse after first- and second-line approaches; 11 patients experienced progressive disease within 6 months of completion of adjuvant anthracycline therapy. Sites of involvement were as follows: liver involvement, 42 patients (50.6%); lung 24 (28.9%); bone, 49 (59.0%); and skin/lymph nodes, 21 (25.3%). Treatment consisted of vinorelbine 30 mg/m(2) administered on days 1 and 15 every 28 days and fluorouracil 200 mg/m(2)/d given continuously over a 24-hour period., Results: Toxicity was recorded for 441 cycles. The scheme was well tolerated: grade 1/2 nausea/vomiting occurred in 13 patients (15.6%), grade 1/2 diarrhea in nine (10.8%), and grade 2/3 stomatitis in six (7.2%). Three patients (3.6%) experienced grade 3/4 leukopenia and four (4.8%) experienced grade 2/3 anemia. Grade 2/3 neurologic toxicity was observed in three cases (3.6%), and grade 2/3 hand-foot syndrome was observed in three (3.6%). The median relative dose-intensity was 92% and 100% for vinorelbine and fluorouracil, respectively. Six patients (7.2%) attained a complete clinical response and 45 (54.2%) attained a partial response, for an overall response rate of 61.4% (95% confidence interval, 50.9% to 71.9%). Twenty-one patients (25.3%) obtained disease stabilization, and 11 (13.3%) experienced disease progression. Median time to progression in responding patients was 15 months; median overall survival of the entire population was 22 months., Conclusion: Vinorelbine associated with protracted infusional fluorouracil is an active and manageable scheme in advanced breast cancer patients previously treated with anthracyclines. The response obtained is durable.

30 patients with advanced non-small cell lung cancer were treated with cisplatin 80 mg/m2, day 1, and teniposide 100 or 120 mg/m2, days 1, 3 and 5, every 3 weeks. Myelotoxicity, nausea and vomiting and alopecia were the main side-effects. 8 patients of 26 evaluable had partial responses (31%): 6 had received 120 mg/m2 teniposide and 2 had received 100 mg/m2 teniposide. Overall median survival time was 251 days. Myelotoxicity was significantly lower in patients who received 100 mg/m2 teniposide. Although the number of patients is small and they were not randomly assigned to the two different teniposide doses, it appears that higher dose of teniposide determined a greater degree of myelotoxicity, and also a higher response rate.

Fifty evaluable patients with advanced lung cancer (28 small cell and 22 non-small cell carcinomas), mainly pretreated by chemotherapy, received 4'-epi-doxorubicin 90 mg/m2 every 3 weeks. Two partial responses were obtained in small cell lung cancer patients, which lasted 153 and 168 days. Leukopenia, emesis and alopecia were the most frequent side effects. Two patients who previously received anthracyclines died suddenly of cardiac failure, another patient had severe congestive heart failure, and four others had minor cardiac dysfunctions. 4'-epi-doxorubicin has a modest activity in advanced lung cancer, mainly pretreated by chemotherapy and is not devoid of significant cardiotoxicity in this patient population.

Nineteen evaluable patients with advanced breast cancer were treated with a combination of doxorubicin and etoposide. Patients had measurable disease, received only mild pretreatment and most had good general conditions at start of therapy. Strict criteria for dose adjustments according to nadir counts were applied. A 42% response rate was obtained. Toxicity was mild and treatment well-tolerated. Doxorubicin-etoposide is an active regimen for patients with breast cancer and warrants further testing in a larger patient population with less stringent criteria for evaluation and treatment monitoring.

Two hundred and eighty-one patients received 927 doses of cis-platinum, generally on an outpatient basis, at 55 mg/m2 every 3-4 weeks. Mannitol and 2.2501 of hydration with saline and 5% dextrose plus NaCl and KCl were given in 3-4 hr. No case of acute renal failure ensued and when azotemia occurred (3.5% of patients) it was easily reversible and controlled. An abnormal level of one or more electrolytes was detected in 194 patients (69%) during chemotherapy. K+, Na+, Ca2+ and Mg2+ values usually decreased in serum after DDP administration, but their depletion seldom caused symptoms. Hypomagnesemia developed in 20% of patients, but was symptomatic in only 1%. cis-Platinum, at the doses utilized, is safely given to outpatients, with the hydration program employed. Serum electrolyte decrease during chemotherapy must be expected, and rapidly corrected when symptoms develop.

The antibody titres in 250 subjects, aged 5 to 22 years, who were vaccinated with a mannitol-lyophilized antimeningococcal A + C vaccine, stable only when stored at -20 degrees C, were followed for two years. As measured by indirect haemagglutination (IHA) and indirect immunofluorescence (IF) techniques, titres for both A and C Neisseria meningitidis antibodies remained high. Two years after vaccination titres of antibodies against type A showed fourfold increase over the initial titres in from 46% to 100% of groups of subjects and against type C in from 42% to 80%. For 130 subjects vaccinated with a new lactose-lyophilized antimeningcoccal A + C vaccine (presumed stable at 4 degrees C) antibody titres were measured up to 16 months after inoculation with this vaccine stored at -20 degrees C and also after storage for several periods at 4 degrees C. Antibody titres in all these subjects had fallen to their initial titres by 16 months. The importance of evaluating the results on subjects showing low initial titres (less than or equal to 1/8 as measured by IHA) is discussed, as inclusion of high initial titres influences the extent of the response.

To investigate the relationship between the biochemical and behavioral deficits in phenylketonuria (PKU), we treated rats from Postnatal Days 2 through 21 with p-chlorophenylalanine plus L-phenylalanine to simulate PKU or with one of the following aromatic acid metabolites of phenylalanine phenylacetate, phenylpyruvate, phenyllactate, and mandelate. Behavioral tests were begun at about 9 weeks of age. Differences between experimental and control animals were found only in the water maze and in the open field. In the former, PKU rats required significantly more trials to reach criterion than controls. None of the single metabolite-treated groups exhibited a similar learning deficit. In the open field, PKU and mandelate-treated rats were hypoactive compared with controls, whereas phenylacetate- and phenylpyruvate-treated rats wee hyperactive. These results demonstrate a lack of congruence between morphological and behavioral effects of treatment, suggesting that performance deficits in PKU rats may be due to interactive effects of 2 or more metabolites.

No clear evidence of survival benefit has been definitely shown by chemotherapy in advanced non-small-cell lung cancer. We evaluated in a randomized trial the activity of the new drug lonidamine (up to 1050 mg/day) versus MVP (mitomycin C, 10 mg/m2, vinblastine, 5 mg/m2, cisplatin, 100 mg/m2). The preliminary findings on 25 patients showed that lonidamine can be easily administered at these dose ranges, and main toxicity was represented by myalgia and testicular pain. Tolerance to combination chemotherapy (MVP) was superimposable to our prior experience. Responses were recorded in both arms, and no survival difference was apparent. The study is in progress.

Sixty-one patients undergoing treatment with cis-platin-containing regimens were given prophylactically either metoclopramide or methylprednisolone, in order to reduce the gastrointestinal side effects. Vomiting occurred in 79% of the cycles (128/162), and had a distressing intensity in 39.5% of cycles (64/162). No significant differences were observed between metoclopramide and methylprednisolone with respect to number and duration of vomiting episodes and duration of nausea and anorexia. Two of 6 patients benefited from substitution of metoclopramide for methylprednisolone; only 1/11 benefited from the substitution of methylprednisolone for metoclopramide. Metoclopramide and methylprednisolone, at the dosage and schedule used, were well tolerated and moderately active in preventing nausea and vomiting induced by cis-platin; their use in combination could further improve these results.

A better therapeutic index has been obtained in breast cancer patients when vinblastine is given by a 5-day continuous infusion program than by i.v. bolus; the continuous infusion pharmacokinetics has been reproduced by an iv divided bolus at 0 and 48 h, which may be more easily applied to outpatients. We performed a broad phase II study in 97 advanced cancer patients in which vinblastine was administered by i.v. divided bolus at 0 and 48 h at the starting dose of 3.5-4 mg/m2, every 3 weeks. Our aim was to confirm the results achieved by continuous infusion and to investigate the toxicity pattern of this novel administration schedule. Neurotoxicity and myelosuppression were the main side effects: constipation and peripheral neurotoxicity respectively developed in 28% and 38% of patients and were severe in 5% and 1%. Leukopenia and thrombocytopenia respectively occurred in 70% and 40% of patients and were severe in 11% and 4%. Four partial responses, 38 no changes and 42 progression were obtained out of 84 evaluable patients. Responses were seen in tumors of breast, lung, and head and neck. Our results do not support the use of vinblastine in divided doses in advanced cancer patients.

Invasive thymomas are rare neoplasms arising from the epithelial cells of the thymus. In this paper, we report five cases of invasive thymoma treated with cisplatin either alone or combined with etoposide. Two partial responses (lasting 1 and 27+ months from the start of chemotherapy), two minor responses (lasting 1.5 and 13+ months from the start of chemotherapy), and one mixed response were observed. Further studies with cisplatin-containing regimens are warranted in the treatment of this rare tumor.

Forty-two evaluable patients with advanced non-small cell lung cancer were treated with teniposide at doses ranging from 120 to 180 mg/m2 on Days 1, 3, and 5 every 3 weeks. Thirty-four patients had received no prior chemotherapy. Seven partial responses (16.6%) were obtained (21% in chemotherapy-unexposed patients). Marrow toxicity was the main side effect: life-threatening thrombocytopenia occurred in 9% of patients, and 54.5% experienced severe leukopenia. Teniposide, at the doses and schedule employed, has moderate activity in non-small cell lung cancer.

Ovarian adenocarcinoma usually disseminates and recurs the abdominal cavity. The detection of central nervous system metastases is rare. In this report we describe a case of intracranial recurrence that developed during a complete local surgical and pathological response and was treated radically with surgery followed by radiation.

The relationship between seizure propensity and the vascular anatomy of the anterior cerebral arteries, or stroke induced by unilateral ligation of the common carotid artery, was investigated utilizing well differentiated colonies of seizure-prone (SP) and non-seizure-prone (NSP) gerbils. Twenty SP and 20 NSP gerbils were perfused with a latex dye and the vascular patterns of the anterior cerebral arteries (ACA) were analyzed. In addition, 30 SP and 30 NSP gerbils were subjected to unilateral ligation of the common carotid artery and the number developing stroke (determined by clinical observation as well as by microscopic examination of the brain) was recorded. We found that in all animals the ACAs formed a fused-central vessel which vascularized the olfactory bulbs, as well as two lateral vessels (rostral arteries) that vascularized a previously undescribed nasal plexus. Neither the vascular pattern of the ACA, nor the frequency of occurrence of stroke following unilateral ligation was related to seizure propensity.

Thirteen patients with small cell lung cancer responsive to chemotherapy were retreated with the same regimen at relapse, after a median off-therapy time of 30 weeks. Fifty per cent responded to reinduction; two out of six patients who had complete response to first chemotherapy obtained complete response again at reinduction. Median survival time from start of any therapy was 94 weeks. When induction chemotherapy has been effective and of short duration, the same chemotherapy can be attempted again with success at relapse and it may affect survival of relapsing small cell lung cancer patients.

Myelotoxicity is one of the most important side effects of antineoplastic drugs. Even in using the same dosages, the gravity of this toxicity varies greatly among different patients. With the aim of evaluating if an in vitro test may predict such an effect we have measured in bone marrow samples taken from 15 patients undergoing chemotherapy for small-cell lung cancer the in vitro uptake of 3H-thymidine in the presence or absence of the cytostatic drugs used for the clinical treatment of these patients. We did not find a clear correlation between the in vitro results and the myelotoxicity observed during the clinical course.

Infections are the most common cause of death in tumor patients. The risk of infection is progressively increased in relation to the severity of neutropenia. It is therefore essential to start empirical antibiotic therapy in these patients at the first sign of infection. Forty-three neutropenic tumor patients were entered into the above study when it was assumed that they had bacterial infections (temperature above 38.5 degrees C and/or signs and symptoms of infection). Patients with greater than 1000 neutrophils/mm3 were given piperacillin alone while those with more severe neutropenia (less than 1000/mm3) were given a combination of piperacillin plus amikacin. Of the 43 patients who had entered the study, 41 could be evaluated whereas the remaining two were considered dropouts either because of non-compliance with the study protocol or because the infection was found to be non-bacterial. In both groups of patients (greater than 1000 and less than 1000 neutrophils/mm3) infection resolved completely in a large percentage of cases (92% and 82%, respectively). The efficacy of piperacillin was therefore reconfirmed for the management of infection in oncologic patients with neutropenia, and proved to be an effective therapeutic resource in patients with both slight and severe neutropenia.

Fourteen evaluable advanced breast cancer patients, extensively pretreated by chemotherapy, received a combination of cisplatin (DDP) and etoposide (VP 16). DDP was given at 60 or 100 mg/m2 on day 1, and VP 16 at 120 mg/m2 on days 1, 2 and 3; cycles were repeated every 4 weeks. Major responses were never obtained; a minor response in 1 patient, no change in 7 patients, and progression in 6 patients were observed. Main side effects were nausea and vomiting (62% severe), and leukopenia (31% leukocytes less than 2,000/mm3). Two patients refused further treatment due to intense nausea and vomiting. DDP-VP 16 combination chemotherapy is ineffective and poorly tolerated in heavily pretreated breast cancer patients.

Thirty patients with bulky advanced ovarian cancer surgically not resectable, received combination chemotherapy (median of 4.1 cycles; range, 3-7) including cisplatin or carboplatin, followed by a second surgical effort. Clinical CR + PR was observed in 24/30 (80%) patients after chemotherapy. Our study deals only with these 24 patients, and the 6 patients who did not respond to chemotherapy are not part of this report. At debulking, 7/24 (29.1%) patients had a complete macroscopic resection; 9/24 (37.5%) patients had a partial resection (residual tumor less than 2 cm). These data suggest that debulking is feasible and successful after chemotherapy containing cisplatin or its derivative. Overall median survival from diagnosis was 18.9 months; the 3-year survival rate was 28%. Median progression-free survival from diagnosis was 13.5 months. The results observed in our study indicate that the use of induction chemotherapy can play an important role in increasing the chances of optimal debulking in patients presenting with unresectable ovarian cancer.

Hepatocellular carcinoma has a poor prognosis and is poorly responsive to systemic chemotherapy. Here we describe a case with unresectable hepatocellular carcinoma who achieved a complete response and has survived for more than 2 years, following treatment with a combination of adriamycin and etoposide (VP 16-213).

A synergistic effect between alpha-Interferon and Bleomycin has been recently shown in in vitro and in vivo experimental systems. Although active, Bleomycin and other antineoplastic drugs give low response rates in non-small cell lung cancer, but, like the other antineoplastic agents, responses are short-lived. We treated 13 patients with advanced non-small cell lung cancer with the combination Bleomycin 15 mg/m2 i.v. at hour 0 and alpha-Interferon 9 X 10(6) U i.m. given at hours 6, 30 and 54. Major side effects were pyrexia, astenia and anorexia; only one case of moderate leukopenia was observed. No major responses were obtained and stable disease lasted a median of 5 months. Further study of this combination is not warranted in patients with pretreated non-small cell lung cancer.

4'-Deoxydoxorubicin was administered to 27 evaluable patients with refractory small cell lung cancer. The majority of patients had good initial performance status. One third of patients had never received doxorubicin before, and six had received a single drug alone (VM26). Myelotoxicity was the main side-effect, and leukopenia was more pronounced than thrombocytopenia. No significant non-hematological toxicity occurred apart from skin necrosis due to drug extravasation in one case. Two patients had partial response (7.4%; 95% confidence limits 0-17.2%). The low response rate obtained in this good prognosis patient population does not support further testing of this drug in small cell lung cancer.

Esorubicin was administered to 12 patients with soft tissue sarcoma, 1 osteosarcoma and 1 desmoid tumor. Seven patients had never received chemotherapy before. Myelotoxicity was the main side effect, leukopenia being more pronounced than thrombocytopenia. No significant nonhematological toxicity occurred. Six patients had no change of median duration of 98 days and 7 patients progressed. The patient suffering from multiple-site lesions of a desmoid tumor obtained a long-lasting partial response. This study does not support further testing of esorubicin in sarcoma patients.

Despite good results of chemotherapy in small cell lung cancer (SCLC), occurrence of brain metastases is frequent and unaffected by commonly employed antineoplastic drugs, mainly because they do not cross the blood-brain barrier. We treated eight patients with SCLC and cerebral metastases with VM26 at 120 mg/m2 given on days 1, 3 and 5 and repeated every 3 weeks. Two patients achieved complete response and one had partial response. Mean response duration was 8.2 months and survival was more than 9 months in responding patients. Toxicity was manageable. VM26 is an active drug in SCLC with brain metastases.

Fifty patients with small-cell lung cancer (SCLC) were treated with teniposide (VM26) at 120 to 140 mg/m2 on days 1, 3, and 5, every 3 weeks. Twelve elderly patients were administered VM26 as first-line chemotherapy. Toxicity was manageable, myelosuppression being the major side effect. The response rate for 44 evaluable patients was 34% (36% for untreated patients); the median durations of response and survival were 230 and 208 days, respectively. Effectiveness of prior chemotherapy and time from last administration was found to influence patient response to VM26: 42% of responders to prior chemotherapy responded to VM26, while 0% of the nonresponders to prior chemotherapy responded to the new agent. Moreover, among patients pretreated with chemotherapy, 12% of those recently treated (earlier chemotherapy ending less than or equal to 2.6 months before administration of VM26) responded to VM26, while 53% of patients treated greater than 2.6 months earlier responded to VM26. Survival was influenced by common prognostic factors (performance status, weight loss, prior chemotherapy exposure). Selection of pretreated patients by type of exposure to prior chemotherapy may help in the testing of new drugs in this disease.