Steve Dunn's CancerGuide has a page about prostate cancer resources:

CancerGuide: Prostate Cancer Resources
<http://cancerguide.org/prostate.html>

Selenium's and vitamin E's anticancer effects against prostate cancer
may at least partly be caused by the fact, that they at least to some
extent inhibit 5-lipoxygenase. Other natural 5-lipoxygenase inhibitors
include cocoa, GLA (from evening primrose oil for example), quercetin,
dried or semi-dried qinger powder and curcumin. More information in
the following articles:

Ghosh J.
Rapid induction of apoptosis in prostate cancer cells by selenium:
reversal by metabolites of arachidonate 5-lipoxygenase.
Biochem Biophys Res Commun. 2004 Mar 12;315(3):624-35.
PMID: 14975747 [PubMed - indexed for MEDLINE]
<http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=pubmed&cmd=Retrieve&dopt=Abstra
ctPlus&list_uids=14975747>

   "Recent clinical trials have documented that selenium
   significantly reduces the incidence of clinical prostate
   cancer. However, nothing is clearly known about the underlying
   molecular mechanisms by which selenium exerts its anti-cancer
   effect. This report provides evidence that selenium at micro-
   molar concentrations induces rapid apoptotic death in human
   prostate cancer cells, but not in normal prostate epithelial
   cells. Apoptosis involves activation of caspase 3 which plays a
   critical role in the cell death process. Interestingly, the
   apoptosis-inducing effect of selenium in prostate cancer cells
   is substantially alleviated by the 5-lipoxygenase metabolites,
   5(S)-HETE and its dehydrogenated derivative 5-oxoETE, but not
   by metabolites of 12-lipoxygenase (12(S)-HETE) or 15-
   lipoxygenase (15(S)-HETE). Apoptosis is also prevented by their
   precursor, arachidonic acid, an omega-6, polyunsaturated fatty
   acid, presumably by metabolic conversion through the 5-
   lipoxygenase pathway. These results indicate that selenium's
   anticancer effect may involve induction of apoptosis
   specifically in prostate cancer cells sparing normal prostate
   epithelial cells, and that 5-lipoxygenase may be a molecular
   target of selenium's anticancer action. The present report
   warrants that care should be taken about high intake of dietary
   fat containing arachidonic acid or its precursor fatty acids
   when selenium is used for the management of prostate cancer,
   and suggests that a combination of selenium and 5-lipoxygenase
   inhibitors may be a more effective regimen for prostate cancer
   control."

Hammarberg T, Kuprin S, Radmark O, Holmgren A.
EPR investigation of the active site of recombinant human
5-lipoxygenase: inhibition by selenide.
Biochemistry. 2001 May 29;40(21):6371-8.
PMID: 11371199 [PubMed - indexed for MEDLINE]
<http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=pubmed&cmd=Retrieve&dopt=Abstra
ctPlus&list_uids=11371199>

Werz O, Szellas D, Henseler M, Steinhilber D.
Nonredox 5-lipoxygenase inhibitors require glutathione peroxidase for
efficient inhibition of 5-lipoxygenase activity.
Mol Pharmacol. 1998 Aug;54(2):445-51.
PMID: 9687587 [PubMed - indexed for MEDLINE]
<http://molpharm.aspetjournals.org/cgi/content/full/54/2/445>

Werz O, Steinhilber D.
Selenium-dependent peroxidases suppress 5-lipoxygenase activity in
B-lymphocytes and immature myeloid cells. The presence of
peroxidase-insensitive 5-lipoxygenase activity in differentiated
myeloid cells.
Eur J Biochem. 1996 Nov 15;242(1):90-7.
PMID: 8954158 [PubMed - indexed for MEDLINE]
<http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=pubmed&cmd=Retrieve&dopt=Abstra
ctPlus&list_uids=8954158>

Zingg JM, Azzi A.
Non-antioxidant activities of vitamin E.
Curr Med Chem. 2004 May;11(9):1113-33. Review.
PMID: 15134510 [PubMed - indexed for MEDLINE]
<http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=pubmed&cmd=Retrieve&dopt=Abstra
ctPlus&list_uids=15134510>

   "... At the posttranslational level, alpha-tocopherol inhibits
   protein kinase C, 5-lipoxygenase and phospholipase A2 and
   activates protein phosphatase 2A and diacylglycerol kinase. ..."

Schewe T, Kuhn H, Sies H.
Flavonoids of cocoa inhibit recombinant human 5-lipoxygenase.
J Nutr. 2002 Jul;132(7):1825-9.
PMID: 12097654 [PubMed - indexed for MEDLINE]
<http://www.ajcn.org/cgi/content/full/81/1/304S>

Suekawa M, Yuasa K, Isono M, Sone H, Ikeya Y, Sakakibara I, Aburada M,
Hosoya E.
[Pharmacological studies on ginger. IV. Effect of (6)-shogaol on the
arachidonic cascade]
Nippon Yakurigaku Zasshi. 1986 Oct;88(4):263-9. Japanese.
PMID: 3098654 [PubMed - indexed for MEDLINE]
<http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=pubmed&cmd=Retrieve&dopt=Abstra
ctPlus&list_uids=3098654>

   "(6)-Shogaol, a pungent component of ginger, which is contained in
   semi-dried ginger but is rarely found in fresh ginger inhibited
   carrageenin-induced swelling of hind paw in rats and arachidonic
   acid (AA)-induced platelet aggregation in rabbits.

   [...]

   (6)-shogaol exhibited an inhibitory action on 5-lipoxygenase
   activity."

Grzanna R, Lindmark L, Frondoza CG.
Ginger--an herbal medicinal product with broad anti-inflammatory
actions.
J Med Food. 2005 Summer;8(2):125-32. Review.
PMID: 16117603 [PubMed - indexed for MEDLINE
<http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=pubmed&cmd=Retrieve&dopt=Abstra
ctPlus&list_uids=16117603>

   "... Ginger suppresses prostaglandin synthesis through
   inhibition of cyclooxygenase-1 and cyclooxygenase-2. An
   important extension of this early work was the observation that
   ginger also suppresses leukotriene biosynthesis by inhibiting
   5-lipoxygenase. This pharmacological property distinguishes
   ginger from nonsteroidal anti-inflammatory drugs. This
   discovery preceded the observation that dual inhibitors of
   cyclooxygenase and 5-lipoxygenase may have a better therapeutic
   profile and have fewer side effects than non-steroidal anti-
   inflammatory drugs. The characterization of the pharmacological
   properties of ginger entered a new phase with the discovery
   that a ginger extract (EV.EXT.77) derived from Zingiber
   officinale (family Zingiberaceae) and Alpina galanga (family
   Zingiberaceae) inhibits the induction of several genes involved
   in the inflammatory response. These include genes encoding
   cytokines, chemokines, and the inducible enzyme
   cyclooxygenase-2. This discovery provided the first evidence
   that ginger modulates biochemical pathways activated in chronic
   inflammation...."

Hong J, Bose M, Ju J, Ryu JH, Chen X, Sang S, Lee MJ, Yang CS.
Modulation of arachidonic acid metabolism by curcumin and related
beta-diketone derivatives: effects on cytosolic phospholipase A(2),
cyclooxygenases and 5-lipoxygenase.
Carcinogenesis. 2004 Sep;25(9):1671-9. Epub 2004 Apr 8.
PMID: 15073046 [PubMed - indexed for MEDLINE]
<http://carcin.oxfordjournals.org/cgi/content/full/25/9/1671>

   "... Curcumin (20 micro M) significantly inhibited LPS-induced
   COX-2 expression; this effect, rather than the catalytic
   inhibition of COX, may contribute to the decreased PGE(2)
   formation. Without LPS-stimulation, however, curcumin increased
   the COX-2 level in the macrophage cells. Studies with isolated
   ovine COX-1 and COX-2 enzymes showed that the curcuminoids had
   significantly higher inhibitory effects on the peroxidase
   activity of COX-1 than that of COX-2. Curcumin and THC potently
   inhibited the activity of human recombinant 5-LOX, showing
   estimated IC(50) values of 0.7 and 3 micro M, respectively. The
   results suggest that curcumin affects arachidonic acid
   metabolism by blocking the phosphorylation of cPLA(2),
   decreasing the expression of COX-2 and inhibiting the catalytic
   activities of 5-LOX. These activities may contribute to the
   anti-inflammatory and anticarcinogenic actions of curcumin and
   its analogs."

Vang K, Ziboh VA.
15-lipoxygenase metabolites of gamma-linolenic acid/eicosapentaenoic
acid suppress growth and arachidonic acid metabolism in human
prostatic adenocarcinoma cells: possible implications of dietary fatty
acids.
Prostaglandins Leukot Essent Fatty Acids. 2005 May;72(5):363-72.
PMID: 15850718 [PubMed - indexed for MEDLINE]
<http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=pubmed&cmd=Retrieve&dopt=Abstra
ctPlus&list_uids=15850718>

   "Although gammalinolenic acid (GLA) and eicosapentaenoic acid
   (EPA) have independently been reported to suppress growth of
   cancer cells, their relative potencies are unknown. To
   determine the possible attenuating efficacies of dietary GLA or
   EPA on prostate carcinogenesis, we hereby report the in vitro
   effects of GLA, EPA and their 15-lipoxygenase (15-LOX)
   metabolites: 15(S)-HETrE and 15(S)-HEPE, respectively, on
   growth and arachidonic acid (AA) metabolism in human androgen-
   dependent (LNCaP) and androgen-independent (PC-3) prostatic
   cancer cells in culture. Specifically, both cells were
   preincubated respectively with the above PUFAs. Growth was
   determined by [3H]thymidine uptake and AA metabolism by HPLC
   analysis of the extracted metabolites. Our data revealed
   increased biosynthesis of prostaglandin E2 (PGE2) and 5-
   hydroxyeicosatetraenoic acid (5(S)-HETE) by both cells.
   Preincubation of the cells with 15(S)-HETrE or 15(S)-HEPE more
   markedly inhibited cellular growth and AA metabolism when
   compared to precursor PUFAs. Notably, 15(S)-HETrE exerted the
   greatest inhibitory effects. These findings therefore imply
   that dietary GLA rather than EPA should better attenuate
   prostate carcinogenesis via its in vivo generation of 15(S)-
   HETrE, thus warranting exploration."

Sies H, Schewe T, Heiss C, Kelm M.
Cocoa polyphenols and inflammatory mediators.
Am J Clin Nutr. 2005 Jan;81(1 Suppl):304S-312S. Review.
PMID: 15640495 [PubMed - indexed for MEDLINE]
<http://www.ajcn.org/cgi/content/full/81/1/304S>

Iversen L, Fogh K, Kragballe K.
Effect of dihomogammalinolenic acid and its 15-lipoxygenase metabolite
on eicosanoid metabolism by human mononuclear leukocytes in vitro:
selective inhibition of the 5-lipoxygenase pathway.
Arch Dermatol Res. 1992;284(4):222-6.
PMID: 1329675 [PubMed - indexed for MEDLINE]
<http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=pubmed&cmd=Retrieve&dopt=Abstra
ctPlus&list_uids=1329675>

   "The purpose of the present study was to determine the effect
   of the n-6 fatty acid, dihomogammalinolenic acid (DGLA, 20: 3,
   n-6) on arachidonic acid (AA) (C20: 4) metabolism by human
   peripheral mononuclear leukocytes (HPML). After incubation of
   HPML with A23187 (5 microM) and DGLA, the cyclooxygenase (CO)
   and lipoxygenase (LO) products were separated and quantified by
   reversed-phase high-performance liquid chromatography (RP-HPLC)
   combined with radioimmunoassay. DGLA led to no change in PGE2
   formation, but at similar concentrations there was a dose-
   dependent decrease in LTB4 formation (IC50 = 45.0 microM). The
   inhibition of LTB4 formation by DGLA was associated with a
   dose-dependent increase in its 15-LO metabolite 15-
   hydroxyeicosatraenoic acid (15-HETrE) and its CO metabolite
   prostaglandin E1 (PGE1). Incubation of HPLM with 15-HETrE
   (0-1.5 microM) alone did not result in a change in PGE2
   formation, whereas 15-HETrE was a much more potent inhibitor of
   LTB4 formation (IC50 = 0.5 microM) than DGLA. These results
   show that the addition of DGLA to HPML results in a selective
   inhibition of LTB4 formation, presumably via its metabolite
   (15-HETrE)."

Chilton-Lopez, Surette ME, Swan DD, Fonteh AN, Johnson MM, Chilton FH.
Metabolism of gammalinolenic acid in human neutrophils.
J Immunol. 1996 Apr 15;156(8):2941-7.
PMID: 8609415 [PubMed - indexed for MEDLINE]
<http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=pubmed&cmd=Retrieve&dopt=Abstra
ctPlus&list_uids=8609415>

   "... Exogenously provided DGLA was converted to one major
   metabolite during cell stimulation; this product migrated on
   reverse-phase HPLC with the 15-lipoxygenase product, 15-
   hydroxy-eicosa-trienoic acid (15-HETre). Both 15-HETre and DGLA
   (provided exogenously) inhibited the formation of leukotriene
   B4, (LTB4) and 20-hydroxy-leukotriene B4 (20-OH-LTB4). The IC50
   for 15-HETre inhibition of both LTR, and 20-OH-LTB4 in A23187-
   stimulated neutrophils was 5 microM. ..."

Schneider I, Bucar F.
Lipoxygenase inhibitors from natural plant sources. Part 1: Medicinal
plants with inhibitory activity on arachidonate 5-lipoxygenase and
5-lipoxygenase[sol ]cyclooxygenase.
Phytother Res. 2005 Feb;19(2):81-102. Review.
PMID: 15852496 [PubMed - indexed for MEDLINE]
<http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=pubmed&cmd=Retrieve&dopt=Abstra
ctPlus&list_uids=15852496>
<http://www3.interscience.wiley.com/cgi-bin/abstract/110477935/ABSTRACT?CRETRY=1&
SRETRY=0>
<http://www3.interscience.wiley.com/cgi-bin/fulltext/110477935/PDFSTART>

An excerpt:

   "As illustrated in Tables 1 and 2 over 180 different plant
   extracts and natural compounds from more than 50 families show
   a significant in vitro 5-lipoxygenase or 5- LOX/COX inhibitory
   activity. Most of them belong to the families of Asteraceae,
   Apiaceae, Lamiaceae and Fabaceae. In the following promising
   plant species and chemical classes of compounds are discussed.
   An inhibition of the 5-lipoxygenase enzyme can be attained in
   different ways: 5-lipoxygenase is very sensitive to
   antioxidants because of its non-heme iron atom in the active
   site of the enzyme. Many 5-lipoxygenase inhibitors simply act
   as nonselective antioxidants by reducing the active-site ferric
   iron and thus interrupting the catalytic cycle. Examples for
   such nonselective redox-inhibitors are numerous flavonoids and
   coumarins. Furthermore, compounds that are able to scavenge
   radical intermediates or to chelate the active site iron
   (compounds with ortho-dihydroxy functions) are often found
   among 5-lipoxygenase inhibitors. Arachidonic acid analogues
   have also been detected as good inhibitors of 5-lipoxygenase,
   due to their structural similarities they compete with
   arachidonic acid for the active site of the enzyme.
   
   The results obtained from the literature research obviously
   indicate that most of the 5-lipoxygenase and 5-LOX/COX
   inhibitors are found among the class of phenolic compounds and
   arachidonic acid analogues.

   In the class of phenolic compounds the most potent 5-
   lipoxygenase inhibitors are the flavonoids such as quercetin,
   isoquercitrin, apigenin, luteolin, siderito- flavon, gnaphalin,
   silibinin, centaureidin, baicalein or rhamnetin. A quite active
   coumarin was isolated from Peucedanum ostruthium root extract
   (IC50 = 0.25 µm) (Hiermann and Schantl, 1998), osthol,
   aesculetin, fraxetin and daphnetin are further coumarins with
   5-lipoxygenase inhibitory activity. Numerous articles also
   report on the 5-lipoxygenase inhibitory properties of other
   phenolic compounds such as magnolol (Magnolia obovata)
   (Hamasaki et al., 1999), belamcandol (Belamcanda chinensis)
   (Fukuyama et al., 1991), caffeic acid derivatives such as
   rosmarinic acid or bornylcaffeate/ferulate or the phenols
   isolated from Atractylodes lancea (Resch et al., 2001) and
   Plantago asiatica (Ravn et al., 1990).
   
   The presence of a catechol structure or a lipophilic
   substituent in polyhydroxylated compounds appears to be
   essential for 5-lipoxygenase inhibitory activity. Thus,
   flavonoids with an ortho-dihydroxy structure in ring B, e.g.
   wedelolacton (Eclipta alba) (Wagner and Fessler, 1986), or
   prenylated flavonoids such as artonin E, morusin,
   sophoraflavanone G, sanggenon B, papyriflavonol A, kenusanon A,
   karzinol B, kenusanon A, kuraridin or kurarinon are very good
   inhibitors of 5-lipoxygenase and 5-LOX/COX. Generally,
   lipophilic substituents enhance the 5-lipoxygenase inhibitory
   properties, whereas additional polar groups as in glycosides
   diminish them (Wagner, 1989).
   
   In the class of arachidonic acid analogues potent 5-
   lipoxygenase inhibitors were found among fatty acids (linoleic
   acid, oleic acid, fatty acids of Sabal serrulata (Paubert-
   Braquet et al., 1997)), thiosulfinates and sulfinyldisulfides
   (active principles in garlic (allicin) and onion (cepaenes))
   (Wagner et al., 1990; Sendl et al., 1992), among isobutylamides
   (isolated from Asiasarum sieboldii (Hashimoto et al., 1994)),
   Echinacea purpurea (Müller-Jakic et al., 1994; Wagner et al.,
   1989) and Spilanthes oleracea (Wagner et al., 1989) and
   polyacetylenes (falcarindiol, panaxynol, polyacetylenes from
   Bidens campylotheca (Redl et al., 1994) or Angelica sp. (Liu et
   al., 1998a, b)).
   
   Besides these two main classes of 5-lipoxygenase inhibitors,
   the triterpencarboxylic acid acetyl-11-keto- ß-boswellic acid
   isolated from Boswellia serrata (e.g. Wildfeuer et al., 1998)
   acts as a direct, specific nonredox triterpencarboxylic acids
   with 5-lipoxygenase inhibitory properties (masticadienolic
   acid, morolic acid, oleanolic acid) were isolated from Pistacia
   terebinthus (Giner- Larza et al., 2001, 2002). Quite active
   diterpenes (e.g. isolinaridial) were isolated from Linaria
   saxatilis (Benrezzouk et al., 1999) and alkaloids with 5-
   lipoxygenase inhibitory activity were found in Isatis
   indigotica (isaindigotone) (Molina et al., 2001), Isatis
   tinctoria (tryptanthrin) (Danz et al., 2002) or Chelidonium
   majus (chelerythrine, sanguinarine) (Vavreckova et al., 1996).
   Resveratrol, canniprene (EI Sohly et al., 1990) or amurensin F
   (Huang et al., 2000) are examples for 5-lipoxygenase inhibitory
   stilbenes and different lichen acids like protolichesterinic
   acid (Ingolfsdottir et al., 1994), lobaric acid (Ingolfsdottir
   et al., 1996) and diffractaic acid (Sunil Kumar and Müller,
   1999) have also been detected as 5-lipoxygenase inhibitors.
   
   The IC50 values of plant extracts and natural compounds with 5-
   lipoxygenase and 5-LOX/COX inhibitory effects are mainly
   between 1 and 50 µm. However, the literature research also
   revealed several plant constituents showing IC50 values below 1
   µm.
   
   Examples for plant species containing 5-lipoxygenase inhibitory
   compounds with IC50 values below 1 µm are: Atractylodes lancea
   (Resch et al., 2001), Isatis tinctoria (Danz et al., 2002),
   Isatis indigotica (Molina et al., 2001), Linaria saxatilis
   (Benrezzouk et al., 1999), Haplophyllum hispanicum (Prieto et
   al., 2002), Plantago asiatica (Ravn et al., 1990), Chelidonium
   majus (Vavreckova et al., 1996), Artocarpus communis (Reddi et
   al., 1991), Salvia aethiopis (Benrezzouk et al., 2001),
   Belamcanda chinensis (Fukuyama et al., 1991), Glycyrrhiza
   inflata (Yoshiyuki, 1995), Dalbergia odorifera (Sunil Kumar and
   Müller, 1998; Miller et al., 1989) and Asiasarum sieboldii
   (Hashimoto et al., 1994)."
   
Schneider I, Bucar F.
Lipoxygenase inhibitors from natural plant sources. Part 2: medicinal
plants with inhibitory activity on arachidonate 12-lipoxygenase,
15-lipoxygenase and leukotriene receptor antagonists.
Phytother Res. 2005 Apr;19(4):263-72. Review.
PMID: 16041764 [PubMed - indexed for MEDLINE]
<http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?itool=abstractplus&db=pubmed&cmd=R
etrieve&dopt=abstractplus&list_uids=16041764>
<http://www3.interscience.wiley.com/cgi-bin/fulltext/110477935/PDFSTART>

Excerpts:

   "The 12-lipoxygenase pathway is of particular interest with
   regard to its involvement in the regulation of tumour
   progression, angiogenesis and metastasis. A number of studies
   have implicated the 12(S)- lipoxygenase metabolite, 12(S)-HETE,
   in the modulation of tumour cell proliferation and apoptosis
   and it was shown to be a key regulator of several parameters
   related to the metastatic potential of tumour cells. An
   increased expression of 12(S)-lipoxygenase has been found in
   various cancers such as prostate cancer, pancreatic cancer,
   breast cancer and lung cancer among others (Nie and Honn,
   2002). The expression level correlates with the clinical stage
   of the disease. It has been demonstrated that inhibitors of
   12(S)-lipoxygenase cause tumour cell apoptosis, reduce tumour
   cell motility and invasiveness, or decrease tumour angiogenesis
   and growth (Nie et al., 2001; Nie and Honn, 2002).
   
   [...]
   
   The findings of the literature research are summarized in
   Tables 1–5. The presented information has been compiled from
   literature of the years 1989–2003. About 35 plant species with
   12-lipoxygenase inhibitory, 15- lipoxygenase inhibitory and
   leukotriene receptor antagonistic activities are included in
   Tables 1, 3 and 5, respectively. Over 25 natural compounds with
   12- and 15-lipoxygenase inhibitory properties are listed in
   Tables 2 and 4. The active medicinal plants are of a wide range
   of families and origins and contain a variety of chemical
   classes of compounds.
   
   The tables present information about the family, the used part,
   the analysed extract, the active constituents, the used in
   vitro assay and the inhibitory activity of the medicinal plant.
   Plant extracts and natural compounds with an IC50 up to 150
   µg/mL or 150 µm are included in the tables
   
   [...]
   
   As illustrated in Tables 1 and 2, over 35 plant extracts and
   natural compounds were shown to inhibit 12- lipoxygenase
   activity in vitro. The results obtained from the literature
   research clearly demonstrate that the most effective 12-
   lipoxygenase inhibitors are found in the class of phenolic
   compounds, especially among flavonoids such as baicalein,
   quercetin, quercitrin, daphnodorin A/C, sideritoflavon,
   leucocyanidol, hypolaetin or isorhamnetin. Besides, prenylated
   flavonoids were identified as compounds with 12-lipoxygenase
   inhibitory activity. As these molecules have an isoprenyl
   moiety as part of their flavonoid backbone structure, they are
   usually more hydrophobic than conventional flavonoids
   suggesting easy penetration through the cell membrane and good
   12-lipoxygenase inhibitory properties. Prenylated flavonoids
   such as artonin E, morusin, sophoraflavonone G, papyriflavonol
   A, sanggenon D or kuwanon C were shown to act as 12-
   lipoxygenase inhibitors. Some articles also report on the 12-
   lipoxygenase inhibitory activity of other phenolic compounds
   such as rosmarinic acid, e.g. isolated from Perilla frutescens
   (Noriko and Keizo, 1997), 2-(3,4- dihydroxyphenyl)-ethanol
   (hydroxytyrosol) derived from olives (De la Puerta et al.,
   1999) or caffeic/ferulic acid derivatives isolated from the
   fruit peel of Citrus reticulata (Nogata et al., 2001). Another
   class of 12- lipoxygenase inhibitors are the arachidonic acid
   analogues. The majority of substances in this class contain
   acetylenic groups, which are responsible for the irreversible
   inactivation of the enzyme. In the literature the
   polyacetylenes falcarindiol and panaxynol are reported as 12-
   lipoxygenase inhibitors (Alanko et al., 1994). Other substance
   classes that possess 12- lipoxygenase inhibitory properties are
   alkaloids such as chelerythrine and sanguinarine isolated from
   Chelidonium majus (Vavreckova et al., 1996), coumarins such as
   esculetin (Sekiya et al., 1982) and tropolones such as
   hinokitiol (Suzuki et al., 2000).
   
   The IC50 values of the plant extracts and natural compounds
   with 12-lipoxygenase inhibitory activity are rather high
   compared with those of natural products with 5-lipoxygenase or
   cyclooxygenase inhibitory properties. Most of the 12-
   lipoxygenase inhibtors of plant origin show IC50 values between
   10 and 100 µm. The tropolone hinokitiol is an example of a
   compound with a clearly lower IC50 value (0.1 µm). Baicalein
   was also reported to show a comparable low IC50 value (Sekiya
   and Okuda, 1982)."

Both above links are on the page:

Nutrition and Lifestyle - Prostate Cancer Foundation
<http://www.prostatecancerfoundation.org/site/c.itIWK2OSG/b.47431/k.8A27/Nutritio
n_and_Lifestyle.htm>

  and don't forget broccoli and other cruciferous vegetables.

Pomegranate juice and vitamin D3 trials:

Pantuck AJ, Leppert JT, Zomorodian N, Aronson W, Hong J, Barnard RJ,
Seeram N, Liker H, Wang H, Elashoff R, Heber D, Aviram M, Ignarro L,
Belldegrun A.
Phase II study of pomegranate juice for men with rising
prostate-specific antigen following surgery or radiation for prostate
cancer.
Clin Cancer Res. 2006 Jul 1;12(13):4018-26.
PMID: 16818701 [PubMed - in process]
<http://clincancerres.aacrjournals.org/cgi/content/abstract/12/13/4018>

Woo TC, Choo R, Jamieson M, Chander S, Vieth R.
Pilot study: potential role of vitamin D (Cholecalciferol) in patients
with PSA relapse after definitive therapy.
Nutr Cancer. 2005;51(1):32-6.
PMID: 15749627 [PubMed - indexed for MEDLINE]
<http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?db=pubmed&cmd=Retrieve&dopt=Abstra
ctPlus&list_uids=1574962>
<http://www.leaonline.com/doi/abs/10.1207%2Fs15327914nc5101_5>

Excerpts:

   "New studies of the relationship between polyunsaturated fatty
   acid metabolismand carcinogenesis have led to novel molecular
   targets for cancer chemoprevention research. These targets
   include procarcinogenic lipoxygenases (LOXs), including 5-, 8-,
   and 12-LOX, and anticarcinogenic LOXs, including 15-LOX-1 and
   possibly 15-LOX-2. Recent studies indicate that 15-LOX-1 is
   down-regulated in colorectal cancer cells and that the ability
   of nonsteroidal anti-inflammatory drugs, a class of clinically
   active cancer chemopreventive agents, to induce apoptosis and
   growth inhibition in these cells was dependent on the induction
   of 15-LOX-1 and its metabolic product 13-S-
   hydroxyoctadecadienoic acid. Consistent with the colorectal
   studies, 15-LOX very recently has shown anticarcinogenic
   activity in esophageal and prostatic carcinogenesis. Inhibitors
   of other LOXs (e.g., 5-LOX) have preclinical anticarcinogenic
   activity and are being developed for clinical chemoprevention
   study. These and other LOX data led us to propose that the
   various LOX pathways exist in a dynamic balance that shifts
   during carcinogenesis toward 5-, 8-, and 12-LOX (and
   cyclooxygenase-2) and away from 15-LOX. A novel approach for
   cancer chemoprevention would involve LOX modulators, i.e.,
   agents that can induce the anticarcinogenic and/or inhibit the
   procarcinogenic LOXs, thereby shifting the balance of LOX
   activities from procarcinogenic to anticarcinogenic metabolism
   of polyunsaturated fatty acids.

   [...]
   
   Several LOXs form different metabolites within the arachidonic
   acid pathway that appear to enhance tumorigenesis. These LOXs
   and metabolites include 5-LOX and its products 5-S-HETE and
   LTB4; 8-LOX and 8-S-HETE; 12-S-LOX and 12-S-HETE; and 12-R-LOX
   and 12-R-HETE. Although 15-LOX-2 also metabolizes arachidonic
   acid to form 15-S-HETE, recent data suggest that this LOX and
   product may be anticarcinogenic, and they are discussed later.
   
   [...]
   
   Several studies have suggested a link between 5-S-HETE
   formation and carcinogenesis in various organs. Prostate, lung,
   and other cancer cell lines express 5-LOX and FLAP mRNA (28 ,
   29) . 5-LOX overexpression recently has been documented in
   human prostate cancer tissue (30) , and 5-S-HETE formation and
   inhibition respectively promote and inhibit the growth of
   prostate cancer cells (31) . 5-S-HETE but not other HETE
   products (LTB4, 12-, or 15-HETE) can also inhibit apoptosis
   induction by MK-886 (a specific FLAP inhibitor) in prostate
   cancer cell lines (32) . Similarly, 5-LOX metabolism of
   arachidonic acid promotes the growth of lung cancer cells, and
   5-LOX inhibitors suppress cell proliferation and induce
   apoptosis in a variety of these cell lines (29) . 5-LOX and
   FLAP inhibitors can reduce tumorigenesis induction by 4-
   methylnitrosamino-1-(3-pyridyl)-1-butanone, a tobacco
   carcinogen (33) . Ding et al. (34)  found that 5-LOX mRNA is
   expressed in pancreatic cancer but not in normal pancreatic
   cells. They confirmed the specific mechanistic role of 5-LOX in
   promoting pancreatic cancer cell proliferation by blocking 5-
   LOX expression with an antisense method, which inhibited
   proliferation, and then adding back 5-S-HETE, which reversed
   the beneficial effects of 5-LOX inhibition. 5-LOX inhibitors
   also can inhibit the growth of mouse colon adenocarcinoma cell
   lines in vitro and in vivo (19 , 20) .

   [...]
   
   Not all studies, however, support the view that 5-LOX
   contributes to carcinogenesis. The naturally occurring
   carcinogen 1-hydroxyanthraquinone induces tumorigenesis in rat
   colon mucosa without affecting 5-LOX expression (35) . The 5-
   LOX inhibitor acetyl-11-keto-ß-boswellic acid induces apoptosis
   in cells that lack 5-LOX expression, possibly through
   topoisomerase I inhibition (36) . The FLAP inhibitor MK-886
   induces apoptosis in human chronic lymphocytic cancer cells
   lacking FLAP expression. Down-regulating FLAP expression by an
   antisense method in cells expressing FLAP had the minor effect
   of increasing the percentage of apoptotic cells from 4.4 to
   10.7% (37) . Therefore, although the role of 5-LOX in promoting
   prostate, lung, and pancreatic carcinogenesis is strongly
   supported by experimental data, 5-LOX may not promote
   carcinogenesis in all models.

   [...]
   
   LTB4 is a more terminal product of the 5-LOX metabolic pathway
   than is 5-S-HETE. LTB4 inhibits apoptosis (38)  and has been
   shown to be procarcinogenic in several studies. The tobacco
   carcinogen 4-methylnitrosamino-1-(3-pyridyl)-1-butanone
   increased plasma LTB4 in animal lung tumorigenesis (33) .
   Bortuzzo et al. (18)  studied the effects of LTB4, LTB4 methyl
   ester, LTB5, 12-R-HETE, 12-S-HETE, and 15-S-HETE on the colon
   cancer cell lines HT-29 and HCT-15. Only LTB4 and 12-R-HETE (a
   12-R-LOX and cytochrome P450 product of arachidonic acid)
   stimulated colonic cell proliferation. SC-41930, a competitive
   antagonist of LTB4, blocked the LTB4 effects. Findings from the
   same laboratory, however, indicated that the LTB4 levels in
   surgical samples of 21 colon cancer patients and paired normal
   tissue were not significantly different (39) . One possible
   explanation for this finding is that the normal surgical
   samples may have contained substantial amounts of connective
   tissue and inflammatory cells containing LTB4 that confounded
   the results.
   
   Other evidence also supports a pro-tumorigenic effect of LTB4.
   In vitro, colon cancer cells produce LTB4 (40) , and LTB4
   synthesis can be triggered by putative tumorigenic agents, such
   as bile salts (41) . In vivo, LTB4 levels are increased in
   intestinal tumors versus in normal appearing intestinal mucosa
   of mice, and the NSAID sulindac blocks this increase (42) .

   [...]
   
   15-LOX-1 and -2 are two isoenzymes of 15-LOX that appear to
   exert important anticarcinogenic effects through the metabolism
   of polyunsaturated fatty acids. The preferred substrate for 15-
   LOX-1 is linoleic acid and for 15-LOX-2 is arachidonic acid
   (60) .

   [...]
   
   As summarized above, a group of LOXs joins COX-2 in promoting
   tumorigenesis by metabolizing arachidonic acid. Other LOXs (15-
   LOX-1 and -2) suppress carcinogenesis by metabolizing linoleic
   and arachidonic acids, respectively. This dichotomy in overall
   LOX effects suggests that a dynamic balance exists among the
   various LOXs. We propose that the pro and anticarcinogenic LOXs
   exist in a dynamic balance that is shifted during tumorigenesis
   from the metabolic activity of 15-LOX promoting cell
   differentiation, growth inhibition, and apoptosis to the
   metabolic activities of other LOXs and COX-2 promoting
   tumorigenesis through arachidonic-acid metabolites, such as
   LTB4, 12-S-HETE, and prostaglandin E2 (Fig. 2)Citation . Until
   recently, LOX cancer chemoprevention research focused
   exclusively on the tumor-promoting effects of LOXs and on
   inhibiting LOX in general and 5- and 12-LOX in particular (24 ,
   33) . Now we know that LOXs, i.e., 15-LOX-1 and -2, also can
   suppress tumorigenesis. Therefore, a novel approach for cancer
   chemoprevention would involve LOX modulators, i.e., agents that
   can induce the anticarcinogenic and/or inhibit the
   procarcinogenic LOXs, thereby shifting the balance of LOX
   activities from procarcinogenic to anticarcinogenic metabolism
   of polyunsaturated fatty acids.

   [...]
   
   LOX metabolism of linoleic and arachidonic acids leads to the
   formation of a variety of metabolically active products with
   different roles in carcinogenesis. Our understanding of these
   roles is steadily increasing. This increased understanding is
   helping to form a theoretical basis for developing new cancer
   chemoprevention approaches targeted on LOX activity within the
   polyunsaturated fatty acid metabolic pathway. The differential
   roles for the various LOXs during tumorigenesis should be
   incorporated within the theoretical framework of novel cancer
   chemoprevention strategies (101)."

Don't get me wrong, I'm not suggesting that lipoxygenase or COX-2
inhibitors could become a single agent therapy for cancer. But in
future they might have role as an adjunctive treatment or in primary
or secondary prevention. And it may not always even require taking
drugs, because there are a large number of natural lipoxygenase and
COX-2 inhibitors, as mentioned.