Search

Mechanical ventilation has become an indispensable therapeutic modality for patients with respiratory failure. However, a serious potential complication of MV is the newly recognized ventilator-induced acute lung injury. There is strong evidence suggesting that matrix metalloproteinases (MMPs) play an important role in the development of acute lung injury. Another factor to be considered is extracellular matrix metalloproteinase inducer (EMMPRIN). EMMPRIN is responsible for inducing fibroblasts to produce/secrete MMPs. In this report we sought to determine: (1) the role played by MMPs and EMMPRIN in the development of ventilator-induced lung injury (VILI) in an in vivo rat model of high volume ventilation; and (2) whether the synthetic MMP inhibitor Prinomastat (AG3340) could prevent this type of lung injury. We have demonstrated that high volume ventilation caused acute lung injury. This was accompanied by an upregulation of gelatinase A, gelatinase B, MT1-MMP, and EMMPRIN mRNA demonstrated by in situ hybridization. Pretreatment with the MMP inhibitor Prinomastat attenuated the lung injury caused by high volume ventilation. Our results suggest that MMPs play an important role in the development of VILI in rat lungs and that the MMP-inhibitor Prinomastat is effective in attenuating this type of lung injury.

Effective therapy is needed to improve the survival of patients with advanced lung cancers. We studied the effects of a selective metalloprotease inhibitor, AG3340, on chemoresistant human non-small cell lung cancer tumors (line MV522) in vivo. Mice bearing s.c. tumors were given twice-daily oral doses of AG3340. As a single agent, AG3340 inhibited angiogenesis (up to 77%) and tumor growth (up to 65%) in a dose-dependent manner at well-tolerated daily doses up to 400 mg/kg/day and induced significant tumor necrosis. In contrast, tumors were relatively insensitive to carboplatin with approximately 25% growth inhibition observed at a maximum tolerated dose of approximately 30 mg/kg/week (given i.p., twice weekly). Carboplatin inhibited tumor growth markedly only at toxic doses, demonstrating a superior therapeutic index of AG3340 to carboplatin in this tumor model. A suboptimal dose of AG3340, when used in combination with an ineffective maximum tolerated dose of carboplatin, resulted in greater tumor growth inhibitions than those produced by either agent alone. Similarly, growth inhibition was enhanced when AG3340 was used in combination with paclitaxel. Cotreatment with carboplatin did not alter AG3340 plasma concentrations achieved acutely after oral dosing. These data demonstrate an antiangiogenic and antitumor effect of AG3340 when used as a single agent and enhanced growth inhibitions when AG3340 is used in combination with cytotoxic agents. These data suggest that treatment with this novel matrix metalloprotease inhibitor may be beneficial in advanced lung cancers and other chemoresistant malignancies.

Synthetic matrix metalloproteinase (MMP) inhibitors have activity against a variety of tumors in preclinical models but have not been studied in gliomas. We determined the effect of AG3340, a novel synthetic MMP inhibitor with Ki values against gelatinases in the low picomolar range, on the growth of a human malignant glioma cell line (U87) in SCID-NOD mice. Mice were injected s.c. with U87 cells. Tumors were allowed to grow to a size of approximately 0.5 x 0.5 cm (after about 3 weeks), and the mice were randomized to receive either: (a) 100 mg/kg AG3340 in vehicle; or (b) vehicle control (0.5% carboxymethyl cellulose, 0.1% pluronic F68), both given daily i.p. Tumor area was measured twice weekly, and animals were sacrificed when moribund, or earlier if premorbid histology was examined. In vivo inhibition of tumor growth was profound, with AG3340 decreasing tumor size by 78% compared with controls after 31 days (when controls were sacrificed; P < 0.01, Wilcoxon test). Control animals survived 31 days after the i.p. injections began, and AG3340 mice survived 71 days, representing a >2-fold increase in survival associated with tumor growth delay. Histological examination found that AG3340-treated tumors were smaller, had lower rates of proliferation, and significantly less invasion than control-treated tumors. Hepatic or pulmonary metastases were not seen in either group. In a separate experiment, the tumors were smaller and sampled after a shorter duration of treatment; the changes in proliferation were more marked and occurred earlier than differences in tumor invasion between the two groups. Furthermore, in vitro cell growth was not inhibited at AG3340 concentrations of <1 mM. AG3340 plasma concentrations in vivo, 1 h after administration, ranged from 67 to 365 nM. Thus, AG3340 produced a profound inhibition of glioma tumor growth and invasion. AG3340 markedly increased survival in this in vivo glioma model. Treatment with AG3340 may be potentially useful in patients with malignant gliomas.

Oral administration of AG3340, a novel metalloprotease (MMP) inhibitor, suppresses the growth of human colon adenocarcinoma (COLO-320DM) tumors in vivo (Proc Am Assoc Cancer Res 39: 2059, 1998). In this report, we tested the hypothesis that the growth inhibition of these tumors is associated with maintaining minimum effective plasma concentrations of AG3340. Nude mice were given a total oral daily dose of 25 or 200 mg/kg; 6.25 mg/kg was given four times per day (QID) (25 mg/kg/day), and 100 mg/kg was given in two daily doses (BID) (200 mg/kg/day). Peak plasma concentrations (Cmax) of 83 +/- 43 (mean +/- SD) and 1998 +/- 642 ng/ml were detected 30 min after a single dose with 6.25 mg/kg and 100 mg/kg AG3340, respectively. AUC(0-24 h) values estimated from dosing with 25 and 200 mg/kg/day AG3340 were 672 and 10882 ng*h/ml, respectively. Importantly, both regimen inhibited tumor growth equivalently (74 to 82%). Efficacy was also compared at a total daily dose of 25 mg/kg by giving AG3340: QID (6.25 mg/kg per dose), BID (12.5 mg/kg per dose), and once daily (25 mg/kg per dose). The Cmax of these regimens was 83 +/- 43, 287 +/- 175 and 462 +/- 495 ng/ml, respectively. AG3340 did not inhibit tumor growth with the latter two regimens. The efficacy of 6.25 mg/kg QID (25 mg/kg/day) was superior to the efficacy of 25 mg/kg BID (50 mg/kg/day), substantiating the independence of efficacy from the total daily dose and Cmax. Expectedly, peak to trough fluctuations were significantly smaller with the QID regimen than with BID and QD dosing. After 24 h, the trough was greater than 1 ng/ml with QID dosing but was less than 1 ng/ml after QD and BID dosing. These results suggest that the antitumor efficacy of AG3340 was associated with maintaining minimum effective plasma concentrations of AG3340 and demonstrate that the antitumor efficacy of AG3340 was independent of the total daily dose, peak plasma concentration, and drug exposure in this tumor model.

We have identified a novel retinoid, ALRT1550, that potently and selectively activates retinoic acid receptors (RARs). ALRT1550 binds RARs with Kd values of approximately equal to 1-4 nM, and retinoid X receptors with low affinities (Kd approximately equal to 270-556 nM). We studied the effects of ALRT1550 on cellular proliferation in squamous carcinoma cells. ALRT1550 inhibited in vitro proliferation of UMSCC-22B cells in a concentration-dependent manner with an IC50 value of 0.22 +/- 0.1 (SE) nM. 9-cis-Retinoic acid (ALRT1057), a pan agonist retinoid that activates RARs and retinoid X receptors, inhibited proliferation with an IC50 value of 81 +/- 29 nM. In vivo, as tumor xenografts in nude mice, UMSCC-22B formed well-differentiated squamous carcinomas, and oral administration (daily, 5 days/week) of ALRT1550, begun 3 days after implanting tumor cells, inhibited tumor growth by up to 89% in a dose-dependent manner over the range of 3-75 micrograms/kg. ALRT1550 (30 micrograms/kg) also inhibited growth of established tumors by 72 +/- 3% when tumors were allowed to grow to approximately equal to 100 mm3 before dosing began. In comparison, 9-cis retinoic acid at 30 mg/kg inhibited growth of established tumors by 73 +/- 5%. Interestingly, retinoids did not appear to alter tumor morphologies in UMSCC-22B tumors. Notably, ALRT1550 produced a therapeutic index of approximately equal to 17 in this model, indicating a separation between doses that inhibited tumor growth and that induced symptoms of hypervitaminosis A. In summary, ALRT1550 potently inhibits cellular proliferation in vitro and in vivo in this squamous cell carcinoma tumor model. These data support additional study of ALRT1550 for its potential for improving anticancer therapy in human clinical trials.

Cisplatin (DDP) is commonly used to treat head and neck tumors. Therapy frequently fails due to development of DDP resistance or toxicities associated with DDP therapy. In this study, effects of ALRT1057 [9-cis retinoic acid (9-cis RA)] on DDP cytotoxicity were studied in a human oral squamous carcinoma xenograft model. Mice bearing xenografts were dosed p.o. daily 5 days/week with 30 mg/kg 9-cis RA and/or i.p. twice weekly with 0.3-0.9 mg/kg DDP. Maximum tolerated doses of 9-cis RA and DDP were approximately 60 and >/=2.9 mg/kg, respectively, under their dosing schedules and routes of administration. Control tumors grew rapidly with mean doubling times of 4 +/- 1 days and reached mean volumes of 1982 +/- 199 (SE) mm3 after 24 days. DDP at doses of 0.3, 0.45, and 0.9 mg/kg inhibited tumor growth by 28, 47, and 86%, respectively, 24 days after tumor cell implantation. Thirty mg/kg 9-cis RA inhibited tumor growth by 25%. In combination, 0.3 mg/kg DDP + 30 mg/kg 9-cis RA inhibited tumor growth by 68%; 0.45 mg/kg DDP + 30 mg/kg 9-cis RA inhibited growth by 78%. These decreases were greater than those that would have been produced by either agent summed separately. Of importance, at doses of 9-cis RA that enhanced DDP cytotoxicity, no change in dose tolerance was observed as compared to tolerances observed for either agent alone, indicating that 9-cis RA increased sensitivity to DDP without altering systemic toxicity. In addition, 9-cis RA profoundly altered squamous cell carcinoma phenotypes by suppressing squamous cell differentiation, resulting in tumors with increased numbers of basal cells. In contrast, DDP selectively depleted proliferating basal cells from carcinomas. In combination, morphological changes produced by 9-cis RA alone predominated, suggesting a possible basis for enhanced DDP sensitivity in tumors exposed to both agents. These data demonstrate that 9-cis RA enhances tumor sensitivity to DDP, and suggest that this combination should be tested in Phase I-II clinical trials for its potential for improving anticancer therapy of squamous cell cancers.

9-cis retinoic acid (9-cis RA) is a retinoid receptor pan-agonist that binds with high affinity to both retinoic acid receptors (RARs) and retinoid X receptors (RXRs). Using a variety of in vivo and in vitro cancer models, we present experimental data that 9-cis RA has activity as a potential chemotherapeutic agent. Treatment of the human promyelocytic leukemia cell line HL-60 with 9-cis RA decreases cell proliferation, increases cell differentiation, and increases apoptosis. Induction of apoptosis correlates with an increase in tissue transglutaminase (type II) activity. In vivo, 9-cis RA induces complete tumor regression of an early passage human lip squamous cell carcinoma xenograft. Finally, 9-cis RA inhibits the anchorage-independent growth of the human breast cancer cell lines MCF-7 and LY2 (an antiestrogen-resistant MCF-7 variant). Transient co-transfection assays indicate that 9-cis RA inhibits estrogen receptor transcription of an ERE-tk-LUC reporter through RAR or RXR receptors. These data suggest that retinoid receptors can antagonize estrogen-dependent transcription and provides one possible mechanism for the inhibition of cell growth by 9-cis RA in breast cancer cell lines. In summary, these findings present evidence that 9-cis RA has a wide range of activities in human cancer models.

Retinoids are promising agents for therapy of squamous cancers. In vitro, retinoids decrease expression of differentiation markers in head and neck squamous carcinoma cells. Little information is available on effects of retinoids on head and neck squamous carcinoma cell xenograft growth in vivo. To address this issue, head and neck squamous carcinoma cells (line 1483) were established as xenografts in nude mice. Control tumors grew rapidly with doubling times of 4-6 days to mean volumes of 1696 mm3 after 24 days. Histological analyses indicated the formation of well-differentiated squamous carcinoma cells exhibiting pronounced stratification (basal and suprabasal cells) and keratinization (keratin pearls) with abundant stroma. Cytokeratin 19 expression was restricted to the basal cell layers, and cytokeratin 4 expression was abundant in suprabasal cells. Mice were treated daily with 30 mg/kg 9-cis retinoic acid, 20 mg/kg all-trans-retinoic acid, or 60 mg/kg 13-cis retinoic acid by p.o. gavage on a schedule of 5 days/week over 4 weeks. Low micromolar (1.48-3.67 microM) and nanomolar (200-490 nM) concentrations of 9-cis retinoic acid and all-trans-retinoic acid were measured in plasmas and xenografts, respectively, 30 min after dosing. Retinoid treatment produced a marked suppression of the squamous cell differentiation of tumor cells manifest by decreased keratinization, loss of stratification, and accumulation of basal cells. This was accompanied by large decreases in the number of CK4-positive cells and concomitant increases of CK19-positive cells. REtinoic acid receptor-beta expression was also increased by 2.9-9.7-fold after chronic retinoid treatment. 9-cis retinoic acid and all-trans-retinoic acid decreased tumor volumes by 23 +/- 5 (SE) and 19 +/- 3%, respectively (P < or = 0.05); 13-cis retinoic acid was inactive. These retinoids did not decrease the rate of exponential tumor growth but increased the latent period until exponential growth began. These studies demonstrate that retinoids do not universally decrease tumor growth but profoundly suppress squamous cell differentiation in vivo in this xenograft model.

Taxol, a unique tubulin active agent, was found to demonstrate a marked schedule-dependent synergistic interaction with cisplatin (DDP) in the killing of human ovarian carcinoma 2008 cells in vitro as determined by median effect analysis. The interaction was highly synergistic when 19 h taxol exposure was followed by 1 h concurrent exposure to taxol and DDP. The combination indices (CIs) on this schedule were 0.11 +/- 0.1, 0.25 +/- 0.15 and 0.39 +/- 0.14 at 20%, 50% and 80% cell kill respectively. However, the interaction was antagonistic when 1 h exposure to DDP was followed by 20 h exposure to taxol, or when cells were exposed to DDP and taxol for 1 h concurrently. When taxol preceded DDP, synergy was also observed with the 11-fold DDP-resistant 2008/C13*5.25 subline, which yielded CI values of 0.21 +/- 0.02, 0.30 +/- 0.11 and 0.31 +/- 0.17 at 20%, 50% and 80% cell kill respectively. At an IC50 concentration, taxol had no effect on [3H]cis-dichloro(ethylenediamine) platinum uptake, on the permeability of the plasma membrane or on glutathione or metallothionein levels in 2008 or 2008/C13*5.25 cells. Mitotic arrest in these cells was observed only at taxol concentrations well above those required for synergy with DDP, suggesting that the mechanism underlying the synergistic interaction was not a taxol-induced alteration in cell cycle kinetics. Of additional interest was the fact that the 2008/C13*5.25 cells were hypersensitive to taxol, and that this was partially explained by an alteration in the biochemical pharmacology of taxol. Although cellular taxol accumulation reached steady state within 2 h in both cell lines, taxol efflux was slower and the taxol was more extensively bound in 2008/C13*5.25 cells than in 2008 cells. In addition, the 2008/C13*5.25 cells had only 55% of the parental levels of beta-tubulin content. However, in another pair of DDP-sensitive and -resistant ovarian cell lines no taxol hypersensitivity and no change in beta-tubulin content was observed, indicating that the DDP-resistant and taxol-hypersensitive phenotypes do not segregate together. We conclude that taxol interacts synergistically with DDP in a manner that is highly schedule dependent, and that the hypersensitivity of 2008/C13*5.25 cells no taxol is unrelated to the mechanism of synergy. These in vitro observations suggest that drug schedule will be an important determinant of the activity and toxicity of the DDP and taxol drug combination in clinical studies.

Intraperitoneal (i.p.) administration of chemoterapeutic agents results in greater total drug exposures in the peritoneal cavity than in plasma. A study on the drug exposure for i.p. lymphatics of pigs, receiving 5-fluorouracil (5-FU), etoposide (VP-16) and carboplatin (CBDCA) by the i.p. route was conducted. Drug concentrations in peritoneal fluid, plasma, and thoracic duct lymph were monitored over the ensuing 3 h. 5-FU appeared rapidly in thoracic duct, lymph and plasma. The lymph concentration declined after 20 min while the plasma concentration remained stable. CBDCA reached a stable concentration in lymph and plasma after 60 min. VP-16 peaked in the lymph after 20 min, whereas the plasma concentration continued to rise for 150 min; the peritoneal half-life for VP-16 was too long for clearance to be defined. Total drug exposure (AUC) was for 5-FU 5.7-fold greater for lymph than for plasma and for CBDCA equal in both compartments. VP-16 had a 2.1-fold higher AUC for lymph than for plasma. The results indicate that the i.p. route of administration results in a greater exposure of the lower thoracic duct lymph than the plasma to 5-FU, produces only a marginally increased exposure to VP-16, and results in no difference for CBDCA. The efficacy of 5-FU is a function of total drug exposure. The results reported provide a strong rationale for evaluating the adjuvant use of i.p. 5-FU in colorectal and gastric carcinoma.

Gene amplification is one mechanism whereby tumor cells can become resistant to antineoplastic agents. Unstably amplified genes occur either on submicroscopic circular pieces of extrachromosomal DNA called episomes or on small acentric chromosomes called double minutes. Double minutes are frequently associated with cells containing amplified drug-resistance genes. Human epidermoid carcinoma KBV1 cells contain unstably amplified mdr1 genes and overexpress P glycoprotein, resulting in decreased intracellular drug accumulation. In this cell line, a nonlethal concentration of hydroxyurea accelerated the rate of loss of vinblastine resistance once vinblastine had been removed from the culture medium. After removal of vinblastine, KBV1 cells exposed to hydroxyurea for the time required to complete 12 cell doublings accumulated more vinblastine than control cells grown in the absence of hydroxyurea. In contrast, hydroxyurea had no effect on vinblastine sensitivity and accumulation in parental drug-sensitive KB-3-1 cells. Hydroxyurea also had no effect on sensitivity to cisplatin in cisplatin-resistant human ovarian carcinoma 2008/C13* cells, indicating that hydroxyurea's effect on drug sensitivity was specific for a drug-resistance phenotype associated with unstably amplified drug-resistance genes. These results indicate that, after removing selective pressure, hydroxyurea accelerates loss of resistance to vinblastine and increases accumulation of vinblastine in KBV1 cells, presumably by accelerating loss of amplified mdr1 genes and thus P glycoprotein.

We examined the ability of dipyridamole (DPM) to act synergistically with vinblastine (VBL) in HT1080 fibrosarcoma cells and a drug-resistant variant, HT1080/DR4, which lacks mdr1 expression, in order to determine whether DPM requires P-glycoprotein to modulate drug sensitivity. Median effect analysis of clonogenic assay was used to produce the combination index (CI50, values less than 1 indicate synergy). DPM was mildly synergistic with VBL producing a CI50 of 0.83 +/- 0.13 for HT1080 cells and 0.80 +/- 0.16 for HT1080/DR4 cells. HT1080 and HT1080/DR4 cells accumulated 6.7 +/- 0.7 and 5.6 +/- 0.9 pmol 3H-VBL mg cells-1 at steady state (Css) and 20 microM DPM elevated the Css by 1.8 and 2.9-fold, respectively. In comparison, the CI50 was 1.1 +/- 0.2 in parental KB-3-1 cells and 0.1 +/- 0.1 in mdr1-expressing KB-GRC1 cells. The KB-3-1 and KB-GRC1 cells had a Css of 3.8 +/- 0.8 and 0.7 +/- 0.2 pmol 3H-VBL mg cells-1, respectively, and DPM elevated the Css by 9.2-fold in KB-GRC1 cells. These studies demonstrate that DPM can produce synergy independently of mdr1 expression but that much greater levels of synergy are achievable in mdr1-expressing tumour cells.

The concept of overcoming multidrug resistance using modulators is based on the hypothesis that there will be a synergistic interaction between the modulator and the cytotoxic agent. We examined the ability of dipyridamole (DPM) to synergistically enhance drug sensitivity in drug-sensitive KB-3-1 cells and their drug-resistant variants, KB-GRC1 and KBV1 cells, using median effect analysis to produce a quantitative measure of the extent of synergy. The drug-resistant variants were resistant to vinblastine (VBL), colchicine (COL), and etoposide (VP-16) in the order VBL greater than COL greater than VP-16 on the basis of 50% inhibitory concentration values obtained by clonogenic assay with continuous drug exposure. The extent of staining with the monoclonal antibody HYB-241, directed at a Mr 180,000 form of the mdrI gene product, correlated with drug resistance for all three drugs (r greater than or equal to 0.92). DPM and verapamil elevated the steady state content (Css) of VBL, but there was no correlation between elevation of Css and the extent of synergy observed. DPM enhanced the cytotoxicity of VBL and COL in a synergistic manner in KB-GRC1 cells, and in KBV1 cells DPM interacted synergistically with VBL. VPL was synergistic with VBL only in KB-GRC1 cells. No synergy was observed in the parental KB-3-1 line. These data indicate that, although both DPM and verapamil can increase Css in cells not expressing P-glycoprotein, such an increase was not associated with synergy. In cells expressing mdrl, synergy was observed, and it was greatest for the cytotoxic agent for which expression of mdrl produced the greatest fold-resistance and enhancement of Css. However, neither the level of resistance, the level of expression of mdrl, nor the ability of the modulator to alter Css accurately predicted whether the interaction would be truly synergistic. We conclude that additional factors determine the nature of the drug interaction.

Resistance of hypoxic cells to radiation and chemotherapy remains a major limitation to effective therapy of solid tumors. Misonidazole, a 2-nitroimidazole analogue, has been studied extensively as a radiosensitizer of hypoxic cells and has been shown to undergo bioreductive metabolism to exert preferential cytotoxicity against hypoxic cells. We have investigated the effects of misonidazole on the biosynthesis of prostaglandins (PGs) in a murine mammary adenocarcinoma cell line (No. 4526) under aerobic and hypoxic conditions in attempts to exploit modulation of PG levels under hypoxia as a means of improving therapeutic approaches for the treatment of solid tumors. We report a time-dependent inhibition of PG biosynthesis by the suspended cells under hypoxia induced by flushing sealed vials with N2 (1.5 liters/min). After 30 min of hypoxia, PG formation was inhibited by 50%. Indomethacin was able to further inhibit the PG formation in a concentration-dependent manner under hypoxia. Misonidazole, however, selectively increased the PGE2 biosynthesis under hypoxia by 49% at 100 microM. This increase was concentration dependent over the range of 25 to 100 microM and was blocked by indomethacin (0.1 microM). Imidazole, the heterocyclic moiety in misonidazole without the nitro function, had no effect on PG biosynthesis at these concentrations. These data suggest that arachidonic acid metabolism is sensitive to the differential oxygen levels which exist within solid tumors and that PG levels may be modulated by electron-affinic agents in hypoxic tumor cells.

We have reported that allicin, a constituent of garlic oil, has no effect on the activities of platelet cyclooxygenase or thromboxane synthase, or vascular PGI2 synthase. The effect of allicin on glutathione (GSH) dependent PGH2 to PGE2 isomerase is unknown. We therefore studied the effect of allicin on PGE2 biosynthesis in a murine mammary adenocarcinoma cell line (No 4526). Intact or sonicated cells were incubated with either 14C-arachidonic acid (AA) or 14C-PGH2, respectively. Following metabolism, products were extracted, separated by TLC and analyzed by radiochromatographic scan. PGE2 was predominantly formed with minimal amounts of PGF2 alpha and PGD2. Formation of 6-keto-PGF1 alpha or TXB2 was not detected indicating the absence of TXA2 and PGI2 synthase activity. Indomethacin and ibuprofen inhibited the PGE2 formation (p less than 0.05). The enzymatic PGE2 formation in sonicates was blocked by depletion of the cellular non-protein thiols by buthionine sulfoximine and was shown to be dependent on GSH. Allicin, over the range of 10-1000 microM, inhibited the formation of PGE2 in cells exposed to 2.0 microM 14C-AA for 20 min. and in sonicated cells incubated with 20.0 microM 14C-PGH2 for 2 min (p less than 0.05). Allicin did not alter cyclooxygenase-mediated oxygen utilization in ram seminal vessicle microsomes, suggesting that allicin selectively inhibits the GSH-dependent PGH2 to PGE2 isomerase in this adenocarcinoma cell line.