Bcl-2 prevents topoisomerase II inhibitor GL331-induced apoptosis is mediated by down-regulation of poly(ADP-ribose)polymerase activity

Poly (ADP-ribose) polymerase (PARP), a nuclear enzyme responsible for DNA strand breaks, has been recently suggested to be crucial for apoptosis induced by a number chemotherapeutic drugs. In this study, we demonstrated that the PARP activity could be evidently elevated with a peak at 6 h when HL-60 cells were treated with a new anticancer drug GL331. Coincident with the peak of PARP activity, an apparent DNA fragmentation and apoptotic morphology were observed in cells treated with GL331. The subsequent apoptotic DNA fragmentation induced by GL331 could be completely blocked by transfecting cells with anti-sense PARP retroviral vector or by treating cells with PARP inhibitor, 3-aminobenzamide (3-AB). This blocking effect thus suggests that activation of PARP was critically involved in GL331-induced apoptosis. The fact that Bcl-2 has been found to antagonize cell death induced by a wide variety of agents, accounts for why we examined whether if Bcl-2 could antagonize GL331 effects. Interestingly, ectopic overexpression of Bcl-2 in either HL-60 or U937 cells caused in resistance towards GL331-elicited DNA fragmentation and cytotoxic effect. Additionally, Bcl-2 also attenuated the poly(ADP-ribosyl)ation of PARP itself as well as Histone H1 at the early period of drug treatment. However, Bcl-2 did not influence the extent of DNA strand breaks induced by GL331 in either control or Bcl-2-overexpressing cells. In addition, analysis of basal PARP activity in control and several Bcl-2 overexpressing clones revealed that Bcl-2 down-regulated PARP activity under the condition without DNA damages. Above findings suggest that poly(ADP-ribosyl)ation of nuclear targets is important for apoptosis induced by DNA-reactive anticancer drugs.

Megestrol acetate antagonizes cisplatin cytotoxicity

Megestrol acetate (MGA) is being widely used for the improvement of appetite and performance status in patients receiving chemotherapy, especially cisplatin-containing therapy. However, little is known about whether MGA has an effect on cisplatin cytotoxicity. We have investigated this using two transitional carcinoma cell lines, i.e. the cisplatin-sensitive parental line NTUB1 and the resistant daughter line NTUB1/P. Combined effects of MGA and cisplatin were assayed with a microculture chemosensitivity method. We explored the level changes of several cisplatin detoxification mechanisms, including metallothionein (MT), glutathione S-transferase-pi (GST-pi) and glutathione (GSH) levels in cells treated with or without MGA. After treatment with 10 microns MGA for 24 h, the cisplatin IC50s of NTUB1 and NTUB1/P increased 1.4- (p = 0.03) and 1.6- (p = 0.02) fold, respectively. By median effect analysis, the combinations of MGA and cisplatin in the two cells appeared to produce an antagonistic interaction. By Northern analysis, MT transcript levels in both cells were significantly upregulated after treatment with MGA, as compared to those without treatment. Exposure to MGA in either sensitive or resistant cells did not alter GST-pi levels as shown by immunoblotting analysis. Cellular GSH content was increased only in NTUB1/P (p = 0.0036) but remained unchanged in NTUB1 cells (p = 0.29) after MGA exposure. In conclusion, MGA may antagonize cisplatin cytotoxicity by upregulating cellular MT and GSH levels. Use of MGA in cisplatin-containing chemotherapy may impair tumor response by antagonizing cisplatin antitumor activity.

High-dose therapy with peripheral blood stem cell (PBSC) support using an innovative mobilization regimen in patients with high-risk primary or chemoresponsive metastatic breast cancers

High-dose therapy followed by peripheral blood stem cell (PBSC) support was performed in 29 patients with primary high-risk (Group I) or chemoresponsive metastatic (Group II) breast cancer patients. Group I patients had received PBSC mobilization within 4 weeks of modified radical mastectomy. Group II patients had to achieve minimal residual disease (MRD) by induction chemotherapy before being considered eligible for PBSC mobilization and high-dose therapy. An innovative FE120C regimen (5-FU 600 mg/m2, i.v., day 1; epirubicin 120 mg/m2, i.v., day 1; cyclophosphamide 600 mg/m2, i.v., day 1) plus G-CSF (300 microg/day, subcutaneous injection for 9 days, from day 4 post-FE120C) was used to mobilize PBSCs. After high-dose CTCb (cyclophosphamide 6,000 mg/m2, thiothepa 500 mg/m2, carboplatin 800 mg/m2, in 4 days), patients received PBSC infusion and daily C-CSF 300 microg subcutaneous injection. There were 19 and 16 patients enrolled into Group I and Group II, respectively. Ten of the Group II patients had achieved minimal residual disease (MRD) after induction chemotherapy. The median numbers of mobilized total CD34 + cells for Group I and Group II patients were 27.3 (9.2 to 114.1) x 10(6)/kg and 17.1 (5.9 to 69.1) x 10(6)/kg respectively. The median time to neutrophil recovery (ANC > or = 500/microL) was 8 and 9 days in Group I and II, respectively. The median time to platelet recovery (> or = 50,000/microL) was 10 and 15 days in Group I and II, respectively. No major treatment-related toxicities were noted. In Group I, 13 out of 19 patients (68.4%; 43-87%, 95% C.I.) remained recurrence-free with a median follow-up of 31 months (6 + to 55 + months). In Group II, 3 out of 10 patients (30%; 7-65%, 95% C.I.) remained progression-free at 33 +, 35 +, 39 + months from induction therapy. We suggest that the FE120C plus G-CSF is an effective and innovative regimen for PBSC mobilization in breast cancer patients, and high-dose CTCb therapy with PBSC support is a safe and well-tolerated treatment modality.

Renal clearance, tissue distribution, and CA-125 responses in a phase I trial of suramin

Suramin was administered to 49 patients in a Phase I cancer trial with real-time pharmacokinetic monitoring and dose individualization to achieve targeted mean plasma concentrations of 210 and 155 mg/liter during the 7-day period between days 15 and 22. Pharmacokinetic sampling after doses on days 1, 3, 5, and 8 was used to modify weekly suramin doses, beginning on day 15, in an attempt to achieve specific averaged plasma concentrations of 210 and 155 mg/liter. A 200-mg test dose was not effective in prospectively determining individual pharmacokinetic parameters and dosage requirements. Patients with peak plasma suramin concentrations in excess of 350 mg/liter may be more likely to experience neurotoxicity (P = 0.06), but there was no statistically significant effect of peak suramin concentration or of cumulative dose. Biopsy and autopsy tissue samples demonstrated low penetration of suramin into brain tissue and muscle but good penetration into prostate and other visceral organs. Prospective use of surrogate substrates for CYP1A2, CYP3A3/4, and CYP2D6 showed no consistent effect of suramin on these enzymes. Although a correlation between creatinine clearance and suramin renal clearance was found (r2 = 0.38; P < 0.00005), there was no correlation between creatinine clearance and total suramin clearance (P = 0.21). No suramin dose modification for renal or hepatic dysfunction can be supported at this time. Three of four ovarian cancer patients demonstrated a drop in CA-125 serum concentrations during suramin treatment.

Relatively low expression of multidrug resistance-1 (MDR-1) and its possible clinical implication in gastric cancers

The mechanism of drug resistance of gastric cancer cells has rarely been investigated. We specifically examine the magnitude and the biologic significance of multidrug resistance-1 (MDR-1) expression in human gastric cancer. All patients had previously been treated in prospective clinical trials for advanced gastric cancer in our institution. Patients with adequate prechemotherapy gastric cancer tissues for immunohistochemical studies by a C219 monoclonal antibody were selected for the determination of the expression rate of MDR-1. The results were designated as negative or positive by the independent interpretation of two pathologists. A subgroup of patients who had been treated with doxorubicin- or etoposide-containing regimens were selected for further correlation with drug sensitivity. Between 1990 and 1996, a total of 60 patients, 38 men and 22 women with a median age of 55 years, were studied. Eight (13.3%; 95% confidence interval, 6%-25%) of them had MDR-1 expression. None of the pertinent clinicopathologic features, including the histopathologic types of the tumors and the extent of the diseases, correlated with the expression of MDR-1. Among the 30 patients who had received doxorubicin- or etoposide-containing combination chemotherapy, 3 (10%; 95% confidence interval, 3%-27%) were designated positive for MDR-1 expression. None of the 3 patients responded to chemotherapy, whereas 19 (70.4%) of the 27 patients who had not expressed MDR-1 did respond (p=0.041 by Fisher's exact test). We conclude that the expression of MDR-1 in gastric cancer is relatively low. Its expression, however, is clinically relevant and is useful in predicting the chemoresistance of patients with gastric cancer receiving doxorubicin- or etoposide-containing combination chemotherapy.

High expression of thymidylate synthase is associated with the drug resistance of gastric carcinoma to high dose 5-fluorouracil-based systemic chemotherapy

BACKGROUND

In the past 4 years, the weekly 24-hour infusion of high dose 5-fluorouracil (5-FU) and leucovorin in the treatment of patients with advanced gastric carcinoma has been prospectively studied at the authors' institution. This has enabled them to explore the possibility that the level of expression of thymidylate synthase (TS), the target enzyme of 5-FU, is related to the drug sensitivity of gastric carcinoma to 5-FU-based chemotherapy.

METHODS

To be eligible for this study, patients were required to have received high dose 5-FU and leucovorin chemotherapy (weekly 24-hour infusions of 5-FU, 2,600 mg/m2, and leucovorin, 300 mg/m2) and to have had adequate prechemotherapy gastric carcinoma tissues for immunohistochemical study. TS106 monoclonal antibody was used to detect the expression of TS. A visual scoring system, which ranged from 0 to 3+, was adopted by 2 independent pathologists to semiquantitate the intensity of TS expression.

RESULTS

Between 1993 and 1996, a total of 30 patients, 18 men and 12 women, with a median age of 61.5 years, were enrolled. Of these patients, 16 (53.3%) and 14 (46.7%) had high and low expression of TS, respectively. Two of the 16 patients (12.5%) with high expression of TS and 13 of the 14 patients (92.9%) with low expression of TS responded to chemotherapy (P < 0.001, chi-square test). The median overall survival was 10 months for patients with low TS expression and 4 months for patients with high TS expression (P < 0.01, log rank test).

CONCLUSIONS

The data from this study suggest that the expression of TS, as determined by immunohistochemistry, is a relatively reliable indicator of whether 5-FU should be used in the treatment of patients with gastric carcinoma.

Systemic chemotherapy alone for patients with non-acquired immunodeficiency syndrome-related central nervous system lymphoma: a pilot study of the BOMES protocol

BACKGROUND

Anecdotal reports have suggested that systemic chemotherapy with agents that better cross the blood-brain barrier may result in long term disease remission in some patients with central nervous system (CNS) lymphoma. This treatment strategy has the advantage of sparing patients the late neurologic complications from brain irradiation.

METHODS

Eligible patients were required to 1) have tissue-proven and measurable non-acquired immunodeficiency syndrome (AIDS)- related primary or metastatic CNS lymphoma; 2) have normal hemogram, renal function, and hepatic function; 3) be age < or = 75 years; and 4) have provided informed consent. Patients with lymphoblastic lymphoma or patients who previously had been exposed to nitrosoureas, etoposide, or high dose methotrexate were not eligible. The systemic chemotherapy (BOMES regimen) included carmustine, 65 mg/m2/day, intravenously (i.v.) on Days 1-2; vincristine, 2 mg/day, i.v. on Days 1 and 8; methotrexate, 1.5 g/m2, i.v. on day 15 followed by leucovorin rescue; etoposide, 50 mg/m2/day, i.v. on Days 1-5; and methylprednisolone, 200 mg/day, i.v. on Days 1-7; repeated every 4 weeks (BOMES regimen). Four doses of intrathecal methotrexate were given to patients who had involvement in the cerebrospinal fluid.

RESULTS

Between March 1991 and March 1997 a total of 19 patients were enrolled on the study. There were 13 men and 6 women, with a median age of 57 years. Fourteen patients had primary CNS lymphoma and 5 patients had concurrent extra-CNS lymphoma. Nine patients previously had been treated by radiotherapy (four patients), chemotherapy (three patients), or both (two patients). There were 11 complete remissions (CR) (57.9%) and 5 partial remissions (26.3%), with a total remission rate of 84.2%. One patient had had progressive brain lymphoma during systemic chemotherapy with the conventional cyclophosphamide, doxorubicin, vincristine, and prednisolone regimen, but achieved CR soon after the regimen was changed to BOMES. The median time to progression of the responders was 6 months. At last follow-up, 4 patients were alive without lymphoma at 10, 47, 64, and 66 months, respectively. There were two treatment-related deaths due to sepsis. Another two patients died of fulminant hepatitis that most likely was chemotherapy-related reactivation of chronic B viral hepatitis.

CONCLUSIONS

The authors believe systemic chemotherapy alone may result in long term disease remission in some select patients with non-AIDS-related CNS lymphoma. Further investigation for better protocols is mandatory.

High expression of thymidylate synthase is associated with the drug resistance of gastric carcinoma to high dose 5-fluorouracil-based systemic chemotherapy

BACKGROUND

In the past 4 years, the weekly 24-hour infusion of high dose 5-fluorouracil (5-FU) and leucovorin in the treatment of patients with advanced gastric carcinoma has been prospectively studied at the authors' institution. This has enabled them to explore the possibility that the level of expression of thymidylate synthase (TS), the target enzyme of 5-FU, is related to the drug sensitivity of gastric carcinoma to 5-FU-based chemotherapy.

METHODS

To be eligible for this study, patients were required to have received high dose 5-FU and leucovorin chemotherapy (weekly 24-hour infusions of 5-FU, 2,600 mg/m2, and leucovorin, 300 mg/m2) and to have had adequate prechemotherapy gastric carcinoma tissues for immunohistochemical study. TS106 monoclonal antibody was used to detect the expression of TS. A visual scoring system, which ranged from 0 to 3+, was adopted by 2 independent pathologists to semiquantitate the intensity of TS expression.

RESULTS

Between 1993 and 1996, a total of 30 patients, 18 men and 12 women, with a median age of 61.5 years, were enrolled. Of these patients, 16 (53.3%) and 14 (46.7%) had high and low expression of TS, respectively. Two of the 16 patients (12.5%) with high expression of TS and 13 of the 14 patients (92.9%) with low expression of TS responded to chemotherapy (P < 0.001, chi-square test). The median overall survival was 10 months for patients with low TS expression and 4 months for patients with high TS expression (P < 0.01, log rank test).

CONCLUSIONS

The data from this study suggest that the expression of TS, as determined by immunohistochemistry, is a relatively reliable indicator of whether 5-FU should be used in the treatment of patients with gastric carcinoma.

Gastric cancer associated with acute disseminated intravascular coagulation: successful initial treatment with weekly 24-hour infusion of high-dose 5-fluorouracil and leucovorin

Acute disseminated intravascular coagulation (DIC) is a severe complication of gastric adenocarcinoma, and most of the patients die within 1-3 weeks. We have treated five such patients with an empirical non-myelosuppressive HDFL regimen (weekly 24h infusion of high-dose 5-fluorouracil 2600 mg/m2 and leucovorin 300 mg/m2). Within 2 weeks of starting the treatment the clinical and laboratory evidence of acute DIC quickly resolved in all five patients. HDFL not only caused no further myelosuppression, but also resulted in normalization of the patient's haemogram within a few weeks. Other anti-cancer drugs could then be safely added. Three patients had a survival time of more than 6 months. We suggest that HDFL is an ideal initial treatment for gastric cancer complicated by acute DIC.

Detection of circulating cancer cells by nested reverse transcription-polymerase chain reaction of cytokeratin-19 (K19)--possible clinical significance in advanced gastric cancer

Intermediate filament cytokeratin-19 (K19) protein is expressed in normal and malignant gastrointestinal epithelial cells, but not in peripheral blood (PB). Small amount of circulating gastric cancer cells can be detected by a sensitive nested reverse transcription-polymerase chain reaction (RTPCR) with primers specific for K19 mRNA. Thirty-four PB samples obtained from patients with inoperable/metastatic gastric cancer were examined. The mononuclear cell (MNC) fraction was collected by Ficoll centrifugation, and followed by total RNA extraction by acid guanidinium thiocyanate-phenol-chloroform method. RNA from 8 gastric cancer cell lines and the mononuclear cells of 33 healthy adults were used as positive and negative controls, respectively. DNA fragment of 774 bp amplified by the internal primers was found to be a highly reliable marker for K19 mRNA expression. The sensitivity of detection was between 1 and 10 cells/10(6) normal MNCs. The K19 transcripts were detected in 20.6% (7/34; 8-37%, 95% C.I.) of PB samples. None of the other pertinent clinicopathological features, including the disease extent and the histopathologic types of the tumors, were related to the expression of K19 in PB. All 34 patients had been treated by systemic chemotherapy. Among the 17 non-responders to chemotherapy, the survival of the 4 patients with detectable K19 was significantly shorter than that of 13 patients without detectable K19 in their circulating blood (p = 0.014). However, the survival impact of K19 was less significant in the other 17 patients whose tumors had responded to systemic chemotherapy. Of the whole group of patients, the median survival of the 7 and 27 patients with and without detectable K19 in their circulating blood was 1 and 3.5 months, respectively (p = 0.368). We concluded that detecting circulating cancer cells by K19 nested RT-PCR is associated with poor prognosis of gastric cancer, particularly in those patients who are not responsive to systemic chemotherapy.

Cisplatin, etoposide, and weekly high-dose 5-fluorouracil and leucovorin infusion (PE-HDFL)--a very effective regimen with good patients' compliance for advanced gastric cancer

We have previously shown that weekly 24-hour infusion of high-dose 5-fluorouracil and leucovorin (HDFL), a regimen initially designed for the treatment of advanced colorectal cancer, is also effective in the treatment of gastric cancer. This HDFL regimen is unique in that it is virtually non-myelosuppressive, and thus provides a comerstone on which ideal protocols may be developed. In this prospective phase II study, we examined the efficacy and toxicity of PE (cisplatin, etoposide)-HDFL, a HDFL-based combination chemotherapy, in the treatment of advanced-gastric cancer. This regimen consisted of cisplatin 60 mg/m2, i.v., D1; etoposide 65 mg/m2, i.v., D1-3; and 5-fluorouracil 2600 mg/m2 plus leucovorin 300 mg/m2, 24-hour i.v. infusion by an ambulatory infusion pump, D2,9,16; repeated every 4 weeks. The major eligibility criteria of the patients included: a) a histologically confirmed, objectively measurable, recurrent or primary inoperable gastric adenocarcinoma; b) age > or = 75 years; c) a Karnofsky performance status > or = 50%; d) an absolute granulocyte count (AGC) > or = 2000/mm3, and a platelet count > or = 100,000/mm3; e) a serum bilirubin concentration < or = 2.0 mg/dl; f) a serum creatinine concentration < or = 1.5 mg/dl; and g) a signed informed consent. Between March 1992 and June 1996, a total of 42 patients were enrolled onto the study. There were 31 men and 11 women with a median age of 54 (24-75) years; these included 16 primary metastatic, 3 locally advanced and inoperable, and 23 postgastrectomy recurrent gastric cancer patients. ECOG (Eastern Cooperative Oncology Group) grade III/IV leukopenia and thrombocytopenia developed in 34.0% and 11.0% of a total of 229 courses given, respectively. There was no treatment-related death. Four patients developed a reversible neurotoxicity; and two of them refused further chemotherapy. Among the 40 patients evaluable for responses, 9 [22.5%; 12-38%, 95% confidence interval (C.I.)] patients achieved complete remission, and 20 [50.0%; 33-67%, 95% C.I.] patients achieved partial remission. The overall response rate was 72.5% [56-86%, 95% C.I.]. The overall median survival and median time to progression of the responders were 10 and 7 months, respectively. The overall median survival of the whole group was 9 months. We concluded that PE-HDFL is a highly effective treatment for advanced gastric cancer. The treatment-related toxicity was mild and the patients' compliance was satisfactory.

Inhibition of the membrane translocation and activation of protein kinase C, and potentiation of doxorubicin-induced apoptosis of hepatocellular carcinoma cells by tamoxifen

Hepatocellular carcinoma (HCC) is characterized by high drug resistance to currently available chemotherapeutic agents. In a prospective clinical study, we have demonstrated that high-dose tamoxifen significantly enhanced the therapeutic efficacy of doxorubicin in patients with far-advanced HCC. In a search for a possible mechanism, we found that tamoxifen at a clinically achievable concentration (2.5 microM) significantly enhanced doxorubicin-induced cytotoxicity and apoptosis of Hep-3B cells, a multidrug resistance (MDR)-1 expressing HCC cell line. This synergistic cytotoxic effect of tamoxifen, at this concentration, however, was not mediated by MDR inhibition. Instead, as evidenced by both western blot and immunofluorescence studies, tamoxifen inhibited the cytoplasmic-membrane translocation of protein kinase C (PKC)-alpha. 12-O-Tetradecanoylphorbol-13-acetate (TPA) restored the membrane translocation of PKC-alpha and abrogated the synergistic cytotoxicity of tamoxifen. We also showed that tamoxifen, at this concentration, did not directly affect the enzyme activity of PKC. Further, membrane translocation of other membrane-bound proteins, such as Ras protein, was similarly inhibited by tamoxifen, but could not be restored by the addition of TPA. Together, these data suggested that tamoxifen may act on the cytoplasmic membrane, and thereby inhibit PKC-alpha translocation to the membrane where it is activated. We hypothesize that high-dose tamoxifen may be an effective modulator of doxorubicin in the treatment of HCC, and suggest that biochemical modulation of PKC as a measure to improve systemic chemotherapy for HCC deserves further investigation.

Transforming growth factor beta1 attenuates ceramide-induced CPP32/Yama activation and apoptosis in human leukaemic HL-60 cells

Ceramide, a product of sphingomyelin turnover, is a novel lipid second messenger that mediates important cellular functions including proliferation, differentiation and apoptosis. This study demonstrates that the CPP32/Yama protease was activated during apoptosis induced by the membrane-permeable second messenger C2-ceramide in HL-60 cells. We also found that the addition of a specific tetrapeptide inhibitor of CPP32/Yama, Ac-DEVD-CHO, provided an effective protection against ceramide-induced cell death. These results suggested that CPP32/Yama has a central role in ceramide-mediated apoptosis. Furthermore a wide variety of cytokines were examined for their effect on ceramide-induced apoptosis. Only transforming growth factor beta1 (TGF-beta1) (1 ng/ml) exerted significant prevention of apoptosis induced by C2-ceramide, or by sphingomyelinase (increases intracellular ceramide). Consistently, TGF-beta1 abrogated the cleavage of poly(ADP-ribose) polymerase and the production of the CPP32/Yama active subunit, p17. However, TGF-beta1 treatment did not cause growth inhibition or alter the level of cyclin-dependent kinase inhibitor p27. It suggests that the preventive effect of TGF-beta1 is not mediated by growth arrest. Interestingly, we found that TGF-beta1 prevented the C2-ceramide-caused decrease of Bcl-2 protein. We thus propose that TGF-beta1 rescues ceramide-induced cell death, possibly by maintaining the constant level of Bcl-2, thereby abolishing CPP32/Yama protease activation.

Tax of the human T-lymphotropic virus type I transactivates promoter of the MDR-1 gene

The mdr-1 gene has been shown to confer resistance to chemotherapy of multiple drugs which share no obvious structural similarities. We and others have previously reported that some virus-associated malignant cells express high levels of MDR-1 (1,2), probably regulated by some viral proteins. In this study we have examined the role of Tax, the key protein of HTLV-1. An excellent correlation was found between the existence of HTLV-1 and the expression of MDR-1 among seven human T-cell lines. In the second part of the study, a 1. 76-kb DNA fragment representing the upstream regulatory elements of human mdr-1 gene was cloned into the CAT reporter plasmid. When the Tax expression plasmid was co-transfected with the MDR-1 reporter plasmid, a significant induction of CAT activity was observed. We conclude that Tax protein may up-regulate the expression of the mdr-1 gene.

A phase II study of weekly 24-hour infusion of high-dose 5-fluorouracil and leucovorin (HDFL) in the treatment of recurrent or metastatic colorectal cancers

Previous report suggested that weekly 24-hour infusion of high-dose 5-fluorouracil (5-FU) and leucovorin is a highly active and relatively low toxic regimen for the treatment of colorectal carcinoma (J Clin Oncol 9: 625-30, 1991). This phase II study was conducted to test this important observation by a slightly modified regimen in a larger group of patients. The weekly HDFL regimen consisted of 5-FU 2600 mg/m2/week and leucovorin 300 mg/m2/week (maximum 500 mg) in a 24-hour intravenous infusion. Between February 1992 and December 1995, a total of 42 patients with non-resectable, recurrent or metastatic colorectal adenocarcinoma were enrolled onto the study. Twenty-nine (69.0%) patients had prior exposure to lower-dose 5-FU. There were 22 men and 20 women with median age of 60 (20-75) years. They received a total of 855 and an average of 20.4 (4 to 65) courses of HDFL chemotherapy. Most patients were treated at outpatient clinics and the drugs were infused by an ambulatory pump system via a Port-A catheter. The median duration of follow-up was 22 months. ECOG Gr 2-3 stomatitis, diarrhea, nausea, and vomiting developed in 6 (14.3%), 6 (14.3%), 5 (11.9%), and 5 (11.9%) patients, respectively. Twenty (47.6%) patients had developed hand-foot syndrome. There was no hematological toxicities except 3 (7.1%) patients developed ECOG Gr 1-2 leucopenia. The overall response rate was 42.9% (28%-59%, 95% C.I.) with 2 complete responses and 16 partial responses. Eight (61.5%; 31%-86%, 95% C.I.) of 13 patients, who had no previous 5-FU exposure, responded (1 complete response, 7 partial responses). Ten (34.5%, 17%-54%, 95% C.I.) of 29 patients, who had had previous lower-dose 5-FU exposure, responded (1 complete response and 9 partial responses). The median duration of response was 5 months (1+ to 23+ months). The median overall survival of the whole group of 42 patients and the 18 responders was 10 and 22 months, respectively. Our data supported the original results of HDFL regimen in the treatment of colorectal cancers. HDFL regimen can be used either as first-line or second-line treatment for non-resectable, recurrent or metastatic colorectal cancers.

Stability of curcumin in buffer solutions and characterization of its degradation products

The degradation kinetics of curcumin under various pH conditions and the stability of curcumin in physiological matrices were investigated. When curcumin was incubated in 0.1 M phosphate buffer and serum-free medium, pH 7.2 at 37 degrees C, about 90% decomposed within 30 min. A series of pH conditions ranging from 3 to 10 were tested and the result showed that decomposition was pH-dependent and occurred faster at neutral-basic conditions. It is more stable in cell culture medium containing 10% fetal calf serum and in human blood; less than 20% of curcumin decomposed within 1 h, and after incubation for 8 h, about 50% of curcumin is still remained. Trans-6-(4'-hydroxy-3'-methoxyphenyl)-2,4-dioxo-5-hexenal was predicted as major degradation product and vanillin, ferulic acid, feruloyl methane were identified as minor degradation products. The amount of vanillin increased with incubation time.

Expression of P-glycoprotein and p53 in advanced hepatocellular carcinoma treated by single agent chemotherapy: clinical correlation

Hepatocellular carcinoma (HCC), a chemoresistant tumour, is the most common fatal cancer in Taiwan. Hepatocellular carcinoma frequently expresses a high level of P-glycoprotein (P-gp), which is a specific phenotype of a multidrug-resistance gene, and harbours mutations of the tumour suppressor gene p53. A modulatory relationship between p53 and P-gp has been reported. In this study, we analysed the expression of P-gp in relation to chemotherapeutic response and p5353 protein expression in advanced HCC. Prechemotherapeutic tumour samples were obtained from 25 patients with HCC which had been treated with either etoposide (VP-16) or doxorubicin. P-glycoprotein and p53 in HCC were visualized by immunohistochemical staining using the monoclonal antibodies JSB-1 and DO1, respectively. We investigated the correlation of P-gp expression with chemotherapeutic responses, clinicopathological features and p53 protein expression. In our study, seven cases achieved partial remission, and the remaining 18 cases had a poor response to chemotherapy. Expression of P-gp was observed in 13 tumours (52%). Positive P-gp protein expression was significantly associated with non-responders (8% or 1/13 vs 50% or 6/12, P = 0.03). Thus, P-gp expression inversely correlated with chemotherapeutic response. Expression of p53 protein was seen in 12 cases and did not correlate with chemosensitivity or P-gp expression. In summary, P-gp expression correlates with the chemosensitivity of HCC that has been treated with VP-16 or doxorubicin and p 53 mutations do not appear to be a major determinant of P-gp expression in advanced HCC.

Factors associated with the therapeutic efficacy of retinoic acids on malignant lymphomas

We recently reported the successful use of retinoic acids in the treatment of refractory lymphoma. The biologic determinants predicting response of lymphomas to retinoic acid remain unknown. This study was conducted to explore this question using in vitro models. Sensitivity of representative lymphoma cells to 13-cis-retinoic acid was determined. Sensitive and resistant cell lines were then compared for their baseline and/or retinoic-acid-regulated expression of total cellular retinoic acid binding protein, retinoic acid receptor (RAR)-alpha, RAR-beta, RAR-gamma mRNA, retinoid X receptor (RXR)-alpha, RXR-beta, RXR-gamma mRNA, transforming growth factor (TGF)-beta 1 and TGF-beta 1 receptors, and Fas (Apo-I) mRNA. The results showed that four of five T, two of three Hodgkin's, and none of six B cell lymphoma cell lines were sensitive (IC30 < 1.5 mmol/L) to 13-cis-retinoic acid. Further analyses revealed several of the above-mentioned parameters may be relevant to retinoic acid sensitivity. Baseline expression of TGF-beta 1 receptors was present in all of the five sensitive cell lines examined, but in only one of the four resistant cell lines. The correlation of Fas expression and retinoic acid sensitivity was good for B cell lines, but not apparent for T cell or Hodgkin's cell lines. On exposure to retinoic acid, an immediate and prolonged upregulation of RAR-alpha mRNA expression, lasting for more than 12 hours, occurred in all sensitive cell lines, but only minimal or transient induction was seen in resistant cells. Together, these data suggested that; 1) retinoic acid has a preferential effect on T cell and Hodgkin's lymphoma cell lines; 2) autoregulation of RAR-alpha by retinoic acids, and the presence of TGF-beta 1 receptors may be relevant to the response of lymphomas to treatment with retinoic acids.

Weekly 24-hour infusion of high-dose 5-fluorouracil and leucovorin in the treatment of advanced gastric cancers. An effective and low-toxic regimen for patients with poor general condition

Systemic chemotherapy for advanced gastric cancer is frequently associated with significant treatment-related toxicity, which is particularly serve in patients presenting with a poor general condition. A search for effective and low-toxic regimens for this group of patients is mandatory. A weekly 24-hour infusion of high-dose 5-fluorouracil (5-FU) and leucovorin (HDFL) has previously been demonstrated to be an effective treatment for advanced colorectal cancer with minimal toxicity. In the past 3 years, this regimen has been tested at our institutes in patients with advanced gastric cancer, the general condition of whom had made the use of intensive combination chemotherapy impossible. The regimen consisted of a weekly 24-hour infusion of 2,600 mg/m2 of 5-FU and 300 mg/m2 of leucovorin. From August 1992 to December 1995, 34 patients had been treated with this regimen for a total of 488 courses (average: 14.4 per patient). Hematological toxicity of this regimen was minimal, with grade 3 or 4 leukopenia developing in only 1 (2.9%) patient. Other nonhematological toxicities were also negligible except a reversible neurotoxicity which developed in 2 patients. Twenty-five patients were eligible for response analysis. One complete response, 11 partial responses, 5 stable diseases, and 8 progressive diseases were observed. The response rate was 48% (32-72%, 95% CI). The median overall survival (OS) of the whole group was 7 months (range: 1-18+). The median OS and time to progression of the responders were 8.5 months (range: 2-18) and 5 months (range: 2-10+), respectively. The palliative effect was satisfactory with the Karnofsky performance status of the responders improving from a median of 50% (range: 20-90%) to 70% (range: 50-100%). Our retrospective data suggested that HDFL is an effective and low-toxic palliative treatment even in patients with a very poor general condition. We advocated that this regimen should be further tested in ordinary patients with advanced gastric cancer.

Weekly CAF chemotherapy for advanced breast cancer patients

In a prospective phase II study, 102 women with advanced breast cancer were treated with low doses of cyclophosphamide, Adriamycin and 5-fluorouracil (CAF) at weekly intervals by intravenous injection. Seventy-five patients were evaluable for treatment response and the overall response rate was 52% (95% confidence interval, 41-63%). Of the evaluable patients, 15% had complete response and 37% had partial response. The median survival after therapy was 15.6 months, the median time to progression was 6.8 months and the median duration of response was 9.1 months. The main toxicities were mild vomiting and moderate myelosuppression. There was only 1 patient who experienced heart failure. Weekly CAF appears to have an efficacy with tolerable side effects comparable to standard CAF with an every-3-week schedule.

Prognostic significance of immunophenotypes in adult lymphoblastic lymphomas

Adult lymphoblastic lymphoma (LBL) can be of T-cell or B-cell lineage. However, the clinical significance of immunophenotypes is largely unknown. We conducted a retrospective study to compare T-cell LBL with its B-cell counterpart. Between 1983 and 1995, 50 adult patients were diagnosed as LBL at National Taiwan University Hospital. Twenty-seven patients (T-LBL:20 and B-LBL:7) had adequate clinical information and formed the basis of final analysis. Pertinent characteristics, including sex, age, and lymphoma stages of these two groups of patients were identical. Detailed clinical features were compared. Systemic involvements of lymphoma were similar except that T-cell LBL had significantly more mediastinal involvement (T:B = 70%:14.3%, p = 0.011). CNS involvement was high in both groups (T:B = 50%:28.6%, p = NS). B-cell LBL had a better overall survival than T-cell LBL, although the survival benefit became less significant after 30 months. The median survival of T- and B-cell LBL was 8 and 31 months, respectively. Both groups taken together, patients who had received prophylactic cranial irradiation had a better overall survival (p < 0.01). We suggest that: a) B-cell LBL has a relatively favorable prognosis than T-cell LBL, at least in the initial 2 to 3 years; b) except for mediastinal involvement, the clinical presentation of T- and B-cell LBL appears to be similar; c) treatment policy, such as the need of prophylactic cranial irradiation, of these two groups may also be similar.

Angioimmunoblastic lymphadenopathy with dysproteinemia--lack of a prognostic value of clear cell morphology

It has been suggested that angioimmunoblastic lymphadenopathy with dysproteinemia (AILD) is closely related to peripheral T cell lymphoma (PTCL). However, the clinical course of AILD-like PTCL is notoriously unpredictable. A minor portion of patients enjoyed prolonged remission with steroid-only treatments (indolent AILD) while most others died rapidly despite the use of intensive chemotherapy (aggressive AILD). Recently, it has been suggested that histological features such as the presence or absence of clear cells and convoluted cells are of high prognostic value. The validity of this observation was addressed in this study. Eighteen patients who presented between 1977 and 1994 at the National Taiwan University Hospital were retrospectively studied. There were 11 men and 7 women, with a median age of 47 years. Twelve patients had received various regimens of systemic chemotherapy, and the other 3 patients had been treated with steroids alone. Eight patients had indolent AILD and 6 aggressive AILD. The follow-up period in 4 patients was too short to be analyzed. The histopathology of these cases was divided, according to the criteria of Aozasa et al., into group I (neither cells), 4 patients; group II (only convoluted cells), 1 patient, and group III (clear cells with or without convoluted cells), 13 patients. Contrary to others, our data revealed that group III patients were doing better than group I patients. Univariate analysis of other pertinent clinical features, including sex, age, lymphadenopathy, B symptoms, hepatosplenomegaly, hypergammaglobulinemia, elevated serum lactate dehydrogenase, and treatment regimens, revealed none of them to be prognostically relevant. However, patients who had achieved complete remission by steroids or other systemic chemotherapy had a significantly better prognosis than those who had not. Together, these preliminary data suggested that (1) the presence or absence of clear cells and convoluted cells failed to predict the clinical behavior, and (2) induction of complete remission by steroids or other chemotherapeutic agents is an important prognostic index.

MVP (mitomycin-C, vinblastine, cisplatin) salvage chemotherapy for relapsed and refractory metastatic breast cancer

Salvage chemotherapy for relapsed and refractory metastatic breast cancer is a challenging issue for oncologists. At our institution, the combination of mitomycin-C, vinblastine and cisplatin (MVP) is used for treatment. The records of 19 consecutive patients with refractory metastatic breast cancer treated with MVP between April 1992 and October 1995 were reviewed. The regimen consisted of mitomycin-C 6 mg/m2, vinblastine 6 mg/m2 and cisplatin 60 mg/m2, repeated every 3 to 4 weeks. The median age of patients was 49 years (range, 35-71 yr). All patients had clinically measurable or evaluable disease and a Karnofsky's performance status greater than or equal to 50%. The median number of prior chemotherapy regimens was two (range, 1-4). Eighteen (94.7%) patients had previously received an anthracycline/anthracenedione-containing regimen, and seven (36.9%) had progression of disease during these therapies. Sixteen patients had two or more sites of metastasis and 17 patients had visceral disease. The median duration of follow-up was 26 months. A mean of 3.5 courses (range, 1-7) was administered. One patient was lost to follow-up after one course of treatment. Of the remaining 18 patients, two complete and five partial responses were observed, for a total response rate of 37% (range, 17-61%, 95% CI). One of the partial responders had disease progression during anthracycline treatment. Treatment-related toxicities were relatively well tolerated. There were no treatment-related deaths. The median duration of response was 3 months (range, 2-11 mo). The median overall survival was 7 months (range, 1-32 mo). Our data suggest that the MVP regimen is an effective palliative treatment for patients with refractory metastatic breast cancer.

High-dose tamoxifen reverses drug resistance to cisplatin and etoposide in a patient with advanced large cell carcinoma of lung

A 64-year-old man was diagnosed as having large cell carcinoma of lung and suffered from recurrent tumors after repeated surgical resection and local irradiation. Systemic chemotherapy with chronic oral etoposide, and then 5 courses of intravenous cisplatin and etoposide (PE) resulted in disease progression with multiple new lung nodules and a severe superior vena cava syndrome. Empirical treatment with a PET regimen, which incorporated high-dose tamoxifen (TAM) 150 mg/m2/day into original PE regimen, was tried. Nearly total tumor necrosis was achieved after 5 courses of PET with remission duration of 4.5 months. The excellent response by simple addition of TAM into a previously inactive PE regimen advocates further investigation.

Thymic carcinoma with autoimmune syndrome: successful treatment with weekly infusional high-dose 5-fluorouracil and leucovorin

Thymic carcinoma is known for its poor clinical outcome and unsatisfactory response to conventional chemotherapy. A 53-year-old woman was diagnosed as having metastatic thymic carcinoma in 1989. She received systemic chemotherapy containing cisplatin, doxorubicin, and cyclophosphamide, and involved-field radiotherapy. A durable complete remission was achieved and lasted for 4 years. When the disease recurred in 1995, she was found to have an autoimmune syndrome in addition to pleural effusion, a posterior mediastinal mass and a left adrenal mass. The autoimmune manifestations were seen as scleroderma, high titers of rheumatoid factor and anti-nuclear antibody. We adopted a novel HDFL regimen, which is composed of weekly 24-hour infusion of high-dose 5-fluorouracil and leucovorin, for this patient. Complete remission was achieved again, and autoimmune syndrome was well controlled.