Effect of lithium gamma-linolenate on the growth of experimental human pancreatic carcinoma

Cytotoxicity of unsaturated fatty acids in fresh human tumor explants: concentration thresholds and implications for clinical efficacy

Background

Unsaturated fatty acids (UFAs) exhibit in vitro cytotoxicity against many malignant cell lines and yield decreased cancer incidence and reduced tumor growth in animal models. But clinical and animal studies to date have achieved response using only localized delivery methods such as intratumoral infusion. To explore possibilities for enhanced clinical efficacy, fresh surgical explants of tumors from 22 patients with five malignancies were exposed to γ-linolenic acid (GLA) and α-linolenic acid (ALA) and analyzed with an in vitro chemosensitivity testing system, the Fluorescent Cytoprint Assay (FCA). A total of 282 micro-organ cultures derived from these malignant tumors were exposed to GLA and ALA at different concentrations.

Results

GLA and ALA exhibited greater than 90% cytotoxicity at a sharp concentration threshold between 500 μM and 1 mM against all but two malignant micro-organ cultures tested in 5-10% serum. In tests using 30-40% serum, GLA and ALA killed tumor at concentrations of 2 mM and above.

Conclusions

The concentration threshold of 500 μM to 2 mM exhibited for antitumor activity by GLA and ALA is much higher than that observed in most previously reported cell culture studies but consistent with physiological concentrations found to kill tumor clinically and in animals. A mechanism of antitumor activity by unsaturated fatty acids through selective destabilization of the malignant plasma membrane is considered. An oral regimen is proposed for phase I clinical testing that could push the area under the curve for serum concentration of unbound unsaturated fatty acids over time to much higher levels than previously achieved for systemic administration and into the range that could yield antitumor response.

Impact of polyunsaturated fatty acids on hepato-pancreatic prostaglandin and leukotriene concentration in ductal pancreatic cancer -- is there a correlation to tumour growth and liver metastasis?

Type and composition of polyunsaturated fatty acids (PUFAs) are suspected to play an important role in carcinogenesis. Thus we investigated the effects of n-3, n-6 and n-9 PUFAs on tumour growth, liver metastasis and concentration of prostaglandins (PG) and leukotrienes (LT) in experimental ductal pancreatic adenocarcinoma. Ninety male hamsters were randomised into six groups (Gr.) (n=15). While Gr. 1-3 were healthy control groups, Gr. 4-6 weekly received subcutaneous injections of 10mg N-nitrosobis-2-oxypropylamine (BOP)/kg body weight for 12 weeks in order to induce ductal pancreatic adenocarcinoma. Between week 1 and 16 all animals were fed with a standard diet with a raw fat content of 2.9%. In week 17 Gr. 1-6 were allocated to three types of diets: Gr. 1: standard high fat (=SHF diet, rich in n-6 PUFAs)/Gr. 2: FISH-OIL (rich in n-3 PUFAs)/Gr. 3: SMOF (=mixture of n-3, n-6 and n-9 PUFAs)/Gr. 4: BOP+SHF/Gr. 5: BOP+SMOF/Gr. 6: BOP+FISH-OIL. After 32 weeks all animals were sacrificed and pancreas as well as liver were analysed histologically. Furthermore pancreatic and hepatic concentrations of prostaglandins (PGF1alpha, PGE(2)) and LT were measured. FISH-OIL decreased number of macroscopically visible pancreatic tumours (Gr. 4-6: 54.5% vs. 45.5% vs. 9.1%, P<0.05) as well as incidence of liver metastasis (Gr. 4-6: 90.9% vs. 72.7% vs. 36.4%, P<0.05). Furthermore concentration of PGF(1)(alpha), PGE(2) and LT were significantly increased in pancreatic carcinoma compared to tumour-free tissue. Moreover levels of PGF(1)(alpha) and PGE(2) were higher in liver metastasis than in extrametastatic hepatic tissue. However, in Gr. 6 (FISH-OIL) intrametastatic concentration of LT was significantly lower than in non-metastatic hepatic tissue as well as in Gr. 4 and Gr. 5. FISH-OIL decreased number of visible pancreatic tumours and incidence of histological proven liver metastasis. This effect might be caused by a decrease of intrametastatic concentration of LT compared to extrametastatic hepatic tissue.

Does enteral nutrition of dietary polyunsaturated fatty acids promote oxidative stress and tumour growth in ductal pancreatic cancer? Experimental trial in Syrian Hamster

BACKGROUND

Type and composition of dietary fat intake is supposed to play an important role in carcinogenesis. Thus we investigated the effects of n-3, n-6 and n-9 polyunsaturated fatty acids (PUFA) on oxidative stress (lipidperoxidation) and tumour growth in ductal pancreatic cancer.

METHODS

Ninety male hamsters were randomized into 6 groups (gr.) (n=15) and allocated to 3 main dietary categories: gr. 1 and 2 received a standard high fat diet (SHF, rich in n-6 PUFA), while gr. 3 and 4 were fed with a diet containing a mixture of n-3, n-6 and n-9 PUFA (SMOF) and gr. 5 and 6 had free access to a diet rich in n-3 PUFA (FISH-OIL). Gr. 1, 3 and 5 received weekly subcutaneous (s.c.) injections of 10 mg N-nitrosobis-2-oxypropylamine (BOP)/kg body weight in order to induce ductal pancreatic adenocarcinoma. Healthy control gr. 2, 4 and 6 were treated with 0.5 ml 0.9% sodium chloride s.c. After 32 weeks all animals were sacrificed. Removed pancreata were weighed and analysed histologically and biochemically. Activities of glutathionperoxidase (GSH-Px), superoxiddismutase (SOD) and levels of lipidperoxidation were measured in samples of pancreatic carcinoma as well as in tumour-free pancreatic tissue.

RESULTS

While different diets did not significantly alter the overall incidence of histologically proven pancreatic adenocarcinoma, the number of macroscopically visible tumours was decreased in the FISH-OIL-gr.

CONCLUSION

Different diets did not significantly influence the incidence of histologically proven pancreatic adenocarcinoma. However, administration of a diet rich in n-3 PUFA (FISH-OIL) resulted in a decrease of macroscopically visible tumours, thus indicating its beneficial effects in respect to attenuation of tumour growth.

From bench to the clinic: gamma-linolenic acid therapy of human gliomas

Malignant gliomas are among the most devastating of cancers and are a major cause of mortality in a young population with a median survival time of 9 months following cytoreductive surgery, radiotherapy and chemotherapy. Recent studies showed that polyunsaturated fatty acids especially gamma-linolenic acid (GLA) have selective tumoricidal action especially against malignant glioma cells both in vitro and in vivo. Limited open label clinical studies showed that intratumoral injection/infusion of GLA is safe and effective against malignant gliomas. In view of this, large-scale, double blind studies are needed to establish the usefulness of GLA in the treatment of malignant brain tumors.

Red oil A5 inhibits proliferation and induces apoptosis in pancreatic cancer cells

AIM: To study the effect of red oil A5 on pancreatic cancer cells and its possible mechanisms.

METHODS: Effect of different concentrations of red oil A5 on proliferation of three pancreatic cancer cell lines, AsPC-1, MiaPaCa-2 and S2013, was measured by ³H-methyl thymidine incorporation. Time-dependent effects of 1:32 000 red oil A5 on proliferation of three pancreatic cancer cell lines, were also measured by ³H-methyl thymidine incorporation, and Time-course effects of 1:32 000 red oil A5 on cell number. The cells were counted by Z1-Coulter Counter. Flow-cytometric analysis of cellular DNA content in the control and red oil A5 treated AsPC-1, MiaPaCa-2 and S2013 cells, were stained with propidium iodide. TUNEL assay of red oil A5-induced pancreatic cancer cell apoptosis was performed. Western blotting of the cytochrome c protein in AsPC-1, MiaPaCa-2 and S2013 cells treated 24 hours with 1:32 000 red oil A5 was performed. Proteins in cytosolic fraction and in mitochondria fraction were extracted. Proteins extracted from each sample were electrophoresed on SDS-PAGE gels and then were transferred to nitrocellulose membranes. Cytochrome c was identified using a monoclonal cytochrome c antibody. Western blotting of the caspase-3 protein in AsPC-1, MiaPaCa-2 and S2013 cells treated with 1:32 000 red oil A5 for 24 hours was carried out. Proteins in whole cellular lysates were electrophoresed on SDS-PAGE gels and then transferred to nitrocellulose membranes. Caspase-3 was identified using a specific antibody. Western blotting of poly-ADP ribose polymerase (PARP) protein in AsPC-1, MiaPaCa-2 and S2013 cells treated with 1:32 000 red oil A5 for 24 hours was performed. Proteins in whole cellular lysates were separated by electrophoresis on SDS-PAGE gels and then transferred to nitrocellulose membranes. PARP was identified by using a monoclonal antibody.

RESULTS: Red oil A5 caused dose- and time-dependent inhibition of pancreatic cancer cell proliferation. Propidium iodide DNA staining showed an increase of the sub-G0/G1 cell population. The DNA fragmentation induced by red oil A5 in these three cell lines was confirmed by the TUNEL assay. Furthermore, Western blotting analysis indicated that cytochrome c was released from mitochondria to cytosol during apoptosis, and caspase-3 was activated following red oil A5 treatment which was measured by procaspase-3 cleavage and PARP cleavage.

CONCLUSION: These findings show that red oil A5 has potent anti-proliferative effects on human pancreatic cancer cells with induction of apoptosis in vitro.

Synergistic activity of gamma-linolenic acid and cytotoxic drugs against pancreatic adenocarcinoma cell lines

BACKGROUND

Gamma-linolenic acid (GLA) is growth inhibitory both in vitro and in vivo, at doses non-toxic to non-cancer cells. Chemotherapeutic agents have limited activity in pancreatic cancer. Interactions between GLA and cytotoxic drugs have not previously been investigated; any synergy might improve the therapeutic effect of these agents.

AIM

To investigate possible interactions between GLA and 5-fluorouracil (5-FU) or gemcitabine against pancreatic cancer cell lines in vitro.

METHODS

Two pancreatic cancer cell lines were exposed to GLA alone and in combination with 5-FU or gemcitabine. Residual viable biomass was measured using the MTT assay and the results analysed by the median effect method of Chou and Talalay [Adv Enzyme Regul 1984;22:27-55].

RESULTS

GLA concentrations of 3.9- 125 microg/ml had a synergistic or additive growth inhibitory effect on all tested concentrations of gemcitabine. Synergism was demonstrated between GLA and 5-FU only at concentrations of 62.5-125 microg/ml of 5-FU.

CONCLUSION

GLA has a synergistic effect with gemcitabine at concentrations that correspond to in vivo therapeutic doses. GLA with 5-FU is synergistic only at a tight range of high concentrations of 5-FU. GLA lacks toxic side effects and may be useful in combination with gemcitabine.

Intratumoral cancer chemotherapy and immunotherapy: opportunities for nonsystemic preoperative drug delivery

The recent literature documents the growing interest in local intratumoral chemotherapy as well as systemic preoperative chemotherapy with evidence for improved outcomes using these therapeutic modalities. Nevertheless, with few exceptions, the conventional wisdom and standard of care for clinical and surgical oncology remains surgery followed by radiation and/or systemic chemotherapy, as deemed appropriate based on clinical findings. This, in spite of the fact that the toxicity of conventional systemic chemotherapy and immunotherapy affords limited effectiveness and frequently compromises the quality of life for patients. Indeed, with systemic chemotherapy, the oncologist (and the patient) often walks a fine line between attempting tumour remission with prolonged survival and damaging the patient's vital functions to the point of death. In this context, it has probably been obvious for more than 100 years, due in part to the pioneering work of Ehrlich (1878), that targeted or localized drug delivery should be a major goal of chemotherapy. However, there is still only limited clinical use of nonsystemic intratumoral chemotherapy for even those high mortality cancers which are characterized by well defined primary lesions i.e. breast, colorectal, lung, prostate, and skin. There has been a proliferation of intratumoral chemotherapy and immunotherapy research during the past two to three years. It is therefore the objective of this review to focus much more attention upon intratumoral therapeutic concepts which could limit adverse systemic events and which might combine clinically feasible methods for localized preoperative chemotherapy and/or immunotherapy with surgery. Since our review of intratumoral chemoimmunotherapy almost 20 years ago (McLaughlin & Goldberg 1983), there have been few comprehensive reviews of this field; only one of broad scope (Brincker 1993), three devoted specifically to gliomas (Tomita 1991; Walter et al. 1995; Haroun & Brem 2000), one on hepatomas (Venook 2000), one concerning veterinary applications (Theon 1998), and one older review of dermatological applications (Goette 1981). However, none have shed light on practical opportunities for combining intratumoral therapy with subsequent surgical resection. Given the state-of-the-art in clinical and surgical oncology, and the advances that have been made in intratumoral drug delivery, minimally invasive tumour access i.e. fine needle biopsy, new drugs and drug delivery systems, and preoperative chemotherapy, it is timely to present a review of studies which may suggest future opportunities for safer, more effective, and clinically practical non-systemic therapy.

Antidepressants, eicosanoids and the prevention and treatment of cancer. A review

Among the mechanisms of carcinogenesis are oncogene synthesis and expression, upregulation of cyclooxygenase, accelerated cell replication, failed apoptosis, viral activation, disruption of signaling pathways, autoimmunity, immunosuppression, angiogenesis and metastasis. All fall within the orbit of eicosanoids and the enzymes that synthesize them. Antidepressants may be of benefit in the prevention and treatment of cancer, as they inhibit the synthesis, antagonize the actions and accelerate the degradation of such eicosanoids as prostaglandins and thromboxanes.

Randomized, dose-finding phase III study of lithium gamolenate in patients with advanced pancreatic adenocarcinoma

BACKGROUND

Chemotherapy for pancreatic cancer offers small survival benefits and considerable side-effects. Unsaturated fatty acids have an antitumour effect in experimental studies; in phase II studies few side-effects were seen.

METHODS

In this group-sequential, open-label, randomized study, 278 patients with a diagnosis of inoperable pancreatic cancer were treated with either oral (700 mg daily for 15 days), low-dose (0.28 g/kg) or high-dose (0.84 g/kg) intravenous lithium gamolenate (LiGLA). The primary endpoint was survival time from randomization using Kaplan-Meier estimates.

RESULTS

Median survival after oral and low-dose intravenous treatment was 129 and 121 days respectively. Median survival after high-dose intravenous treatment was 94 days. A good Karnofsky score and the absence of metastases were associated with increased survival. Haemolysis, a marker of rapid infusion, was associated with a median survival time of 249 days in the low-dose intravenous group.

CONCLUSION

Oral or low-dose intravenous LiGLA led to survival times similar to those of other treatments for pancreatic cancer although one subgroup (low-dose intravenous LiGLA with haemolysis) had longer survival. High-dose intravenous treatment appeared to have an adverse effect. Systemic treatment with LiGLA cannot be recommended for the treatment of pancreatic cancer.

Highly unsaturated (n-3) fatty acids, but not alpha-linolenic, conjugated linoleic or gamma-linolenic acids, reduce tumorigenesis in Apc(Min/+) mice

We showed previously that dietary eicosapentaenoic acid [EPA, 20:5(n-3)] is antitumorigenic in the APC:(Min/+) mouse, a genetic model of intestinal tumorigenesis. Only a few studies have evaluated the effects of dietary fatty acids, including EPA and docosahexaenoic acid [DHA, 22:6(n-3)], in this animal model and none have evaluated the previously touted antitumorigenicity of alpha-linolenic acid [ALA, 18:3(n-3)], conjugated linoleic acid [CLA, 77% 18:2(n-7)], or gamma-linolenic acid [GLA, 18:3(n-6)]. Stearidonic acid [SDA, 18:4(n-3)], the Delta6-desaturase product of ALA, which is readily metabolized to EPA, has not been evaluated previously for antitumorigenic efficacy. This study was undertaken to evaluate the antitumorigenicity of these dietary fatty acids (ALA, SDA, EPA, DHA, CLA and GLA) compared with oleic acid [OA, 18:1(n-9)] at a level of 3 g/100 g in the diets of APC:(Min/+) mice and to determine whether any alterations in tumorigenesis correspond to alterations in prostaglandin biosynthesis. Tumor multiplicity was significantly lower by approximately 50% in mice fed SDA or EPA compared with controls, whereas less pronounced effects were observed in mice fed DHA (P: = 0.15). ALA, CLA and GLA were ineffective at the dose tested. Although lower tumor numbers coincided with significantly lower prostaglandin levels in SDA- and EPA-fed mice, ALA and DHA supplementation resulted in equally low prostaglandin levels, despite proving less efficacious with regard to tumor number. Prostaglandin levels did not differ significantly in the CLA and GLA groups compared with controls. These results suggest that SDA and EPA attenuate tumorigenesis in this model and that this effect may be related in part to alterations in prostaglandin biosynthesis.

Effect of 1-(gamma)linolenyl-3-eicosapentaenoyl propane diol on the growth of human pancreatic carcinoma in vitro and in vivo

Essential fatty acids such as (gamma)linolenic (GLA) and eicosapentaenoic (EPA) acids have been proposed as anticancer drugs. The aim of this study was to test the effect of a lipid emulsion containing both GLA and EPA in a novel chemical formulation of 1-(gamma)linolenyl-3-eicosapentaenoyl propane diol on the growth of human pancreatic carcinoma in vitro and in nude mice. This compound had a dose-dependent growth-inhibitory effect on human pancreatic cancer cell lines MIA PaCa-2 and Panc-1 in vitro. The concentration necessary for 50% growth inhibition was 25 micromol/l for MIA PaCa-2 and 68 micromol/l for Panc-1 (95% CI 20-29 and 59-77 micromol/l respectively). Nude mice bearing subcutaneous pancreatic tumours produced with the MIA PaCa-2 cell line were treated with the maximum tolerated dose (6.75 mg GLA and 7.3 mg EPA per g of body weight) administered over 10 days by daily intravenous (i.v.) bolus injections. No antitumour effect or major alteration in tumour lipid fatty acid composition was seen in comparison with control animals. Concurrent treatment with parenteral iron (iron saccharate, 5 microg/gram body weight daily) did not make a significant difference. Further improvements in fatty acid delivery mechanisms are necessary before they can become useful anticancer agents.

Lipoxygenase inhibition induced apoptosis, morphological changes, and carbonic anhydrase expression in human pancreatic cancer cells

Epidemiologic and animal studies have linked pancreatic cancer growth with fat intake, especially unsaturated fats. Arachidonic acid release from membrane phospholipids is essential for tumor cell proliferation. Lipoxygenases (LOX) constitute one pathway for arachidonate metabolism. We previously reported that 5-LOX and 12-LOX are upregulated in human pancreatic cancer cells and that blockade of these enzymes abolishes pancreatic cancer cell growth. The present study was aimed at evaluating the effect of LOX inhibition on differentiation and apoptosis in pancreatic cancer cells in parallel with growth inhibition. Four human pancreatic cancer cell lines, PANC-1, MiaPaca2, Capan2, and HPAF, were used. Apoptosis was evaluated by three separate methods, including DNA propidium iodide staining, DNA fragmentation, and the TUNEL assay. Morphological changes and carbonic anhydrase activity were used to determine the effect of LOX inhibitors on differentiation. The general LOX inhibitor NDGA, the 5-LOX inhibitor Rev5901, and the 12-LOX inhibitor baicalein all induced apoptosis in all four pancreatic cancer cell lines, as confirmed by all three methods, suggesting that both the 5-LOX and 12-LOX pathways are important for survival of these cells. Furthermore, NDGA, Rev5901, or baicalein resulted in marked cellular morphological changes in parallel with increased intracellular carbonic anhydrase activity, indicating that LOX blockade induced a more differentiated phenotype in human pancreatic cancer cells. Together with our previous findings, this study suggests that perturbations of LOX activity affect pancreatic cancer cell proliferation and survival. Blockade of LOX enzymes may be valuable for the treatment of human pancreatic cancer.