Expert Reviews - Expert Review of Neurotherapeutics - = 8(4):515

Drug = cocktails=20 for effective treatment of glioblastoma=20 multiforme

Sections:

INTRODUCTION	Top of pageINTRODUCTION = <<References

Glioblastoma multiforme (GBM) is the = most=20 prevalent primary brain tumor and also one of the most = difficult human malignancies to manage. In fact, only=20 7 months was added to the mean survival rate of = patients=20 with GBM during the last seven decades [1,2]. = The=20 location of tumor, which impacts on poor penetration = of drugs=20 through the blood=E2=80=93brain barrier and also = within the tumor;=20 well-known resistance to chemotherapy and radiation = therapy=20 also in part because of the location; the infiltrative = nature=20 reflected by tumor cells migrating away from the = primary site=20 and embedding into normal brain parenchyma; genetic,=20 morphological and molecular heterogeneity; and = extremely well=20 developed although improperly working neovasculature = all=20 contribute to a dismal outcome of GBM patients. Recent = years=20 have brought more interest in studying GBM among = cancer=20 researchers generating new invaluable knowledge. This = new=20 knowledge at genetic, morphological, cellular and = molecular=20 levels provides a basis for rational drug development=20 approaches. The prospect of finding better treatments = for GBM=20 is now certainly more realistic.

Just about everything that relates to GBM = pathobiology and=20 its clinical course invites thinking about specific = targeting=20 of more than one tumor compartment/target and more = than one=20 mechanism controlling the pathobiology of GBM, hence = its=20 maintenance and progression. This truly prompts = researchers to=20 consider combinatorial therapy or a cocktail of drugs. = The=20 current standard of care itself is a combination of = different=20 modalities of treatment, such as surgery, chemotherapy = and=20 radiation, but they do not address fully, as they = cannot, the=20 formidable complexity of GBM. In this editorial, it is = argued=20 that combinatorial treatment can and should be applied = utilizing various classes of drugs that are directed = against=20 different tumoral compartments and different aspects = of=20 cellular/molecular composition and pathobiological = importance.=20 A combinatorial (i.e., a cocktail) drug approach would = be=20 relevant within a specific group of drugs as well as = among=20 them. This would imply that perhaps not only a = cocktail of=20 drugs (nowadays rationally designed) against specific = tumor=20 compartments/functions will be needed, but also such a = cocktail will need to include cocktails from each = specific=20 type of agents.

A novel approach to GBM treatment is the use of = molecularly=20 targeted recombinant cytotoxins. Recombinant = cytotoxins are=20 targeted to GBM tumor cells by the means of a ligand = or=20 antibody that binds an internalized plasma membrane = receptor.=20 They deliver modified bacterial toxins as active, = catalytic=20 portions of the cytotoxins [3]. Several = cytotoxins=20 have been in clinical trials during the last decade or = so even=20 though none of them were designed for the treatment of = brain=20 tumors [3]. The most = striking=20 preclinical and clinical results were obtained with a=20 cytotoxin composed of a wild-type IL-13 and PE38QQR = derivative=20 of Pseudomonas exotoxin A [4]. An = uncommon=20 efficacy seen in patients with recurrent GBM in = Phase I=20 and II trials with IL13-PE38QQR [5] prompted=20 development of Phase III trial that ended in late = 2006. The=20 drug demonstrated significantly better = progression-free=20 survival in patients receiving the cytotoxin versus = current=20 standard of care [6]. This = efficacy was=20 obtained even though intratumoral drug delivery = through=20 convection-enhanced delivery was not monitored, and = patients=20 were not selected for the overexpression of one of the = drug=E2=80=99s=20 targets, IL-13 receptor-=CE=B12 (IL-13R=CE=B12). Here = is the first=20 reason why we should think about a cocktail of = cytotoxins=20 rather than an individual one. IL-13R=CE=B12 is = overexpressed in=20 approximately 75% of patients with GBM [7=E2=80=9310]=20 although its gene is expressed at lower levels [11], not an = uncommon=20 scenario. Thus, at least a quarter, and perhaps even = more than=20 that, of patients with GBM are expected to be poor or=20 nonresponders to therapies targeting IL-13R=CE=B12. = This calls for=20 prescreening patients before applying therapy, but = then a=20 sizable subset of patients will not have a chance to = receive a=20 drug based on a promising therapeutic strategy. One = can=20 envision that addition of another cytotoxin would make = a=20 difference. The prerequisite would be that there is = another=20 target in GBM that is not completely overlapping with = the=20 expression of IL-13R=CE=B12 and its distribution. = EphA2 tyrosine=20 kinase receptor is the second receptor that is highly=20 overexpressed in patients with GBM and its expression = is=20 linked to a worse outcome [12,13]. A vast=20 majority of GBM patients overexpress either = IL-13R=CE=B12 or EphA2,=20 but importantly approximately 95% overexpress two = receptors [10]. Hence, = a very=20 small fraction of patients would be potential = nonresponders if=20 a cocktail of two cytotoxins, the ones targeting = IL-13R=CE=B1 2 and=20 EphA2, was administered. A prototype cytotoxin against = EphA2=20 receptor has already been generated and tested [14]. In = addition, a=20 combination of two cytotoxins or double-targeting in = GBM=20 treatment makes the cytotoxins somehow more effective=20 antitumor agents [15,16].

An ideal scenario would be to have a cocktail of = anti-GBM=20 cytotoxins targeted against receptors that are = overexpressed=20 in 100% of patients at any given time. The third = receptor,=20 besides IL-13R=CE=B12 and EphA2, highly overexpressed = in GBM but=20 not normal brain is currently being searched for as a = suitable=20 candidate for the development of another specific and = potent=20 cytotoxin. It is likely that we will identify such a = target,=20 since a protein, a transcription factor, the = overexpression of=20 which complements that of IL-13R=CE=B12 and EphA2, has = been found.=20 Fos-related antigen 1, together with IL-13R=CE=B12 and = EphA2=20 constitutes the first trimolecular signature of GBM = identified=20 as such [10]. A = combination of=20 drugs targeting this trio of factors will not miss a = single=20 patient with GBM, but obviously a cytotoxin cannot be=20 generated against Fra-1.

It is now possible to specifically target a = neovasculature=20 (vessels formed during tumor initiation and = progression) of=20 GBM with drugs. Clinical trials demonstrate that such = a=20 treatment tightens up neomicrovasculature and tumor = vessels=20 become even less permeable [17]. One = could=20 consider intratumoral delivery of recombinant = cytotoxins under=20 such conditions in expectation of longer residence = time within=20 the tumor and its vicinity with less leakage into the=20 periphery. This requires having cytotoxins of neutral = activity=20 towards normal brain, such as those targeted against = IL-13R=CE=B12=20 and EphA2. Clinical examination of a cocktail of=20 cytotoxins/antiangiogens is warranted.

We hypothesized that GBM is powered, like a sports = car, by=20 multiple engine cylinders and it might be necessary to = knock=20 down most of them in order to effectively stop or at = least=20 weaken it [18]. = Biological=20 forces moving this engine are the intracellular = signaling=20 pathways. Recent extensive research in this area of=20 investigation fully supports the notion that an = Achilles heel=20 of GBM has not yet been identified, but rather = multiple=20 complex mechanisms operating in GBM that maintain the = tumor=20 and cause it to progress [19]. Thus, = inhibiting=20 one activated system may slow the work temporarily but = will=20 not stop the engine. Preferably, two or three pathways = should=20 be attacked concomitantly or consecutively. This is = another=20 example of combinatorial therapy in which a cocktail = of=20 intracellular pathway inhibitors might need to be = employed in=20 the treatment of GBM.

And finally, I would like to make a = case for=20 having drugs for the treatment of GBM available = off-the-shelf.=20 Our current understanding of the disease and current = ability=20 to generate therapeutic compounds makes combinatorial = therapy=20 an approach of choice. The next decade or so of = research will=20 indicate which combinations/cocktails of drugs = belonging to=20 one group of compounds, such as small molecule = inhibitors or=20 recombinant cytotoxins, will work most effectively in=20 patients. During this process, combining a cocktail of = recombinant cytotoxins with a cocktail of small = molecule=20 inhibitors may make therapeutic sense in that the = cytotoxins=20 will eliminate as many tumor cells as they can, but = even a=20 residual disease will ascertain the recurrence, and = hence=20 needs to be taken care of. Small molecule inhibitors = could=20 then be introduced into the treatment regimen. Or, a=20 multivalent vaccine following the = cytotoxins/inhibitors will=20 be necessary to be combined with the preceding = treatment [20]. This = is again in=20 order to eliminate surgically inaccessible,=20 chemotherapy-resistant/inaccessible and=20 radiation-resistant/inaccessible tumors. The cost of=20 developing off-the-shelf cocktails might actually be = less than=20 making individualized target identification for = choosing an=20 individual patient-profiled cocktail, and certainly = more=20 applicable in parts of the world where individualized = medicine=20 will not have economic support for a long time to = come.

Financial & = competing=20 interests disclosure=20

Dr Waldemar Debinski is a consulting scientific = advisor=20 and a shareholder in Targepeutics, Inc. Dr Debinski is = an=20 inventor on the patent applications submitted or = issued, which=20 are also on the subject of this paper (however, Penn = State=20 University and Wake Forest University own the = patents).=20 Dr Debinski=E2=80=99s work has never been = supported by any=20 industrial entity. The author has no other relevant=20 affiliations or financial involvement with any = organization or=20 entity with a financial interest in or financial = conflict with=20 the subject matter or materials discussed in the = manuscript=20 apart from those disclosed.

No writing assistance was utilized = in the=20 production of this manuscript.

References	Top of pageINTRODUCTIONReferences = <<

Papers of special note have been highlighted as: = =E2=80=A2 of=20 interest =E2=80=A2=E2=80=A2 of considerable = interest

1	Stupp=E2=80=8C<= /SPAN>=20 R, Mason WP, van den Bent MJ et al.=20 Radiotherapy plus concomitant and adjuvant = temozolomide=20 for glioblastoma. N. Engl. J. Med. 352, = 987=E2=80=93996=20 (2005).
2	Gallia=E2=80=8C<= /SPAN>=20 GL, Brem S, Brem H. Local treatment of malignant = brain=20 tumors using implantable chemotherapeutic = polymers.=20 J. Natl Compr. Canc. Netw. 3, = 721=E2=80=93728=20 (2005).
3	Debinski=E2=80=8C<= /SPAN>=20 W. Local treatment of brain tumors with targeted = chimera=20 cytotoxic proteins. Cancer Invest. 20, = 801=E2=80=93809=20 (2002).
4	Debinski=E2=80=8C<= /SPAN>=20 W, Obiri NI, Powers SK, Pastan I, Puri RK. Human = glioma=20 cells over-express receptor for IL13 and are = extremely=20 sensitive to a novel chimeric protein composed = of IL13=20 and Pseudomonas exotoxin. Clin. Cancer = Res. 1, 1253=E2=80=931258 (1995).
5	Kunwar=E2=80=8C<= /SPAN>=20 S, Prados MD, Chang SM et al. Direct=20 intracerebral delivery of cintredekin besudotox=20 (IL13-PE38QQR) in recurrent malignant glioma: a = report=20 by the Cintredekin Besudotox Intraparenchymal = Study=20 Group. J. Clin. Oncol. 25, = 837=E2=80=93844 (2007).
6	Kunwar=E2=80=8C<= /SPAN>=20 S, Westphal M, Medhorn M et al. Results = from=20 PRECISE: a randomized Phase 3 study in patients = with=20 first recurrent GBM comparing Cintredekin = Besudotox=20 administered via convection-enhanced delivery = with=20 Gliadel wafers. Neuro-oncology 9, 531 = (2007). =E2=80=A2=E2=80=A2 As in [5], this study=20 establishes clinical efficacy of a = recombinant=20 cytotoxin in patients with glioblastoma = multiforme=20 (GBM).
7	Saikali=E2=80=8C<= /SPAN>=20 S, Avril T, Collet B et al. Expression of = nine=20 tumour antigens in a series of human = glioblastoma=20 multiforme: interest of EGFRvIII, IL-13R=CE=B12, = gp100 and=20 TRP-2 for immunotherapy. J. Neurooncol. = 81,=20 139=E2=80=93148 (2007).
8	Zhang=E2=80=8C<= /SPAN>=20 JG, Eguchi J, Kruse C et al. Antigenic = profiling=20 of glioma cells to generate allogeneic vaccines = or=20 dendritic cell-based therapeutics. Clin. = Cancer=20 Res. 13, 566=E2=80=93575 (2007). =E2=80=A2 Demonstrates an interesting=20 profile of antigens found in GBM suitable for = targeting=20 using multivalent vaccines.
9	Liu=E2=80=8C<= /SPAN>=20 TF, Tatter SB, Willingham MC, Yang MY, Hu = J,=20 Frankel A. Growth factor expression varies among = high=20 grade gliomas and normal brain: epidermal growth = factor=20 receptor has excellent properties for = interstitial=20 fusion protein therapy. Mol. Cancer Ther. = 2(8),=20 783=E2=80=93787 (2003).
10	Wykosky=E2=80=8C<= /SPAN>=20 J, Gibo DM, Stanton C, Debinski W. IL-13 = Receptor =CE=B1-2,=20 EphA2, and Fra-1 as molecular denominators of = high-grade=20 astrocytomas and specific targets for = combinatorial=20 therapy. Clin. Cancer Res. 14, = 199=E2=80=93208 (2008). =E2=80=A2=E2=80=A2 Establishes first ever=20 documented trimolecular signature of = GBM.
11	Nutt=E2=80=8C<= /SPAN>=20 CL, Mani DR, Betensky RA et al. Gene=20 expression-based classification of malignant = gliomas=20 correlated better with survival than = histological=20 classification. Cancer Res. 63, = 1602=E2=80=931607 (2003).
12	Wykosky=E2=80=8C<= /SPAN>=20 J, Gibo DM, Stanton C, Debinski W. EphA2 as a = novel=20 molecular marker and target in glioblastoma = multiforme.=20 Mol. Cancer Res. 3, 541=E2=80=93551 = (2005).
13	Liu=E2=80=8C<= /SPAN>=20 F, Park PJ, Lai W et al. A genome-wide = screen=20 reveals functional gene clusters in the cancer = genome=20 and identifies EphA2 as a mitogen in = glioblastoma.=20 Cancer Res. 66, 10815=E2=80=9310823 = (2006).
14	Wykosky=E2=80=8C<= /SPAN>=20 J, Gibo DM, Debinski W. A novel, potent, and = specific=20 ephrinA1-based cytotoxin against EphA2=20 receptor-expressing tumor cells. Mol. Cancer=20 Ther. 6, 3208=E2=80=933218 (2007 )
15	Liu=E2=80=8C<= /SPAN>=20 TF, Willingham MC, Tatter SB et al. = Combination=20 fusion protein therapy of refractory brain = tumors:=20 demonstration of efficacy in cell culture. J. = Neurooncol. 65(1), 77=E2=80=9385 (2003).
16	Stish=E2=80=8C<= /SPAN>=20 BJ, Chen H, Shu Y, Panoskaltsis-Mortari A, = Vallera DA.=20 Anti-glioblastoma effect of a recombinant = bi-specific=20 cytotoxin co-targeting human IL-13 and EGF = receptors in=20 a mouse xenograft model. = J. Neurooncol.=20 87(1), 51=E2=80=9361 (2008). =E2=80=A2 Demonstrates better=20 efficacy of double-targeting of GBM with = recombinant=20 cytotoxins.
17	Batchelor=E2=80=8C<= /SPAN>=20 TT, Sorensen AG, di Tomaso E et al. = AZD2171,=20 a pan-VEGF receptor tyrosine kinase inhibitor,=20 normalizes tumor vasculature and alleviates = edema in=20 glioblastoma patients. Cancer Cell 11(1), = 83=E2=80=9395=20 (2007).
18	Debinski=E2=80=8C<= /SPAN>=20 W, Slagle-Webb B, Achen MG et al. = VEGF-D is=20 an X-linked/AP-1 regulated putative = onco-angiogen in=20 human glioblastoma multiforme. Mol. Med. = 7,=20 598=E2=80=93608 (2001).
19	Mischel=E2=80=8C<= /SPAN>=20 PS, Cloughesy T. Using molecular information to = guide=20 brain tumor therapy. Nat. Clin. Pract. = Neurol.=20 2(5), 232=E2=80=93233 (2006).
20	Debinski=E2=80=8C<= /SPAN>=20 W. Personalized, multivalent, and more = affordable: the=20 globalization of vaccines. Clin. Cancer = Res. 11,=20 5663=E2=80=935664 (2005).

Affiliations

Waldemar Debinski

Director of Brain Tumor = Center of=20 Excellence, Comprehensive Cancer Center of Wake Forest = University, Departments of Neurosurgery, Radiation = Oncology,=20 & Cancer Biology, Wake Forest University = School of=20 Medicine, Medical Center Boulevard, Winston-Salem,=20 NC 27157, USA. debinski@wfubmc.edu=20