individualizing breast cancer treatment—the dawn of personalized medicine
TRANSCRIPT
Q1
242526272829303132333435363738394041424344454647484950515253545556575859606162636465666768697071727374757677787980818283
Available online at www.sciencedirect.com
journal homepage: www.elsevier.com/locate/yexcr
E X P E R I M E N T A L C E L L R E S E A R C H ] ( ] ] ] ] ) ] ] ] – ] ] ]
0014-4827/$ - see frohttp://dx.doi.org/10.1
nCorresponding autE-mail addresses
cellbiology.ncri@gma
Please cite this art(2013), http://dx.d
Review Article
Individualizing breast cancer treatment—The dawnof personalized medicine
Argha Nandya, Sudeshna Gangopadhyaya, Ashis Mukhopadhyayb,n
aDepartment of Cell Biology, Netaji Subhas Chandra Bose Cancer Research Institute, 16A, Park Lane; Kolkata 700 016, IndiabDepartment of Medical Oncology, Netaji Subhas Chandra Bose Cancer Research Institute, 16A, Park Lane, Kolkata 700 016, India
a r t i c l e i n f o r m a t i o n
Article Chronology:
Received 21 May 2013Received in revised form28 August 2013Accepted 3 September 2013
Keywords:
Breast cancer stratificationPrecancer nicheHeterogenic tumor µenvironmentCirculating tumor cellNetwork medicinePersonalized oncology clinical trial
nt matter & 2013 Elsevier016/j.yexcr.2013.09.002
hor. Fax: þ91 33 2226 4704: [email protected] (A. Mukhopadhyay)
icle as: A. Nandy, et al., Ioi.org/10.1016/j.yexcr.201
a b s t r a c t
Identification of breast cancer not being a single disease but backed by multiple heterogeneousoncogenic subpopulations is of growing interest in developing personalized therapies to provideoptimal outcomes. Through this review, we bring attention to evolution of tumor and microenviron-ment heterogeneity as a predominant challenge in stratifying therapies. Establishment of a ‘precancerniche’ serves as a prerequisite for genetically initiated cells to survive and promote neoplastic evolutiontowards clinically established cancer through development of tumor and its microenvironment.Additionally, continuous evolutionary interplay between tumor and recruited stromal cells along withmany other components in the tumor microenvironment adds up to further complexity in developing
targeted therapies. However, through continued excellence in developing high throughput technologiesincluding the advent of single-nucleus sequencing, which makes it possible to sequence individualtumor cells, leads to improved abilities in decoding the heterogenic perturbations through reconstruc-tion of tumor evolutionary lineages. Furthermore, simple liquid-biopsies in form of enumeration/characterization of circulating tumor cells and tumor microvesicles found in peripheral circulation, shedfrom distinct tumor lesions, show great promise as prospective biomarkers towards better prognosis intailoring individualized therapies to breast cancer patients. Lastly, by means of network medicinalapproaches, it is seemingly possible to develop a map of the cell's intricate wiring network, helping toidentify appropriate interconnected protein networks through which the disease spreads, offering amore patient-specific outcome. Although these therapeutic interventions through designing persona-lized oncology-based trials are promising, owing to continuous tumor evolution, targeting genome
instability survival pathways might become an economically viable alternative.& 2013 Elsevier Inc. All rights reserved.
Contents
Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2Science driving personalized breast cancer therapies – advancements and future perspectives . . . . . . . . . . . . . . . . . . . . . . . . . . 2
Identifying novel genetic subsets of breast cancer genome: disease stratification . . . . . . . . . . . . . . . . . . . . . . . . . 2Identifying challenges in individualizing therapies: heterogeneity between tumors, within tumors and in themicroenvironment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
Inc. All rights reserved.
.n (A. Nandy), [email protected] (S. Gangopadhyay),.
ndividualizing breast cancer treatment—The dawn of personalized medicine, Exp Cell Res3.09.002
84858687888990919293949596979899
100101102103104105106107108109110111112113114115116117118119120121122123124125126127128129130131132133134135136137138139140141142143
E X P E R I M E N T A L C E L L R E S E A R C H ] ( ] ] ] ] ) ] ] ] – ] ] ]2
Please cite(2013), http
Identifying approaches towards addressing heterogeneity: lines of control . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4Identifying approaches in directing therapies towards novel protein targets for individualizing therapies. . . . . . 5Seeking help from the emerging ‘Network Medicine’ – developing better individualized therapeutics . . . . . . . . 6Towards personalized oncology-based clinical trials: where are we heading?. . . . . . . . . . . . . . . . . . . . . . . . . . . . 7
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7Conflict of interest statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8Funding . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
Fig. 1 – IHC based breast cancer subtype classification. Twomajor subtypes are the luminal (ERþ or PRþ) and non-luminaltumors (ER- and PR-). These are further sub-classified intoluminal 1 (HER2-), luminal 2 (HER2þ), non-luminal HER2þand triple-negative phenotype (TNP). The luminal 1 and 2types were further sub-classified into luminal basal-, luminalbasalþ and the TNP into core basal group (CBP: CK5/6 orEGFRþ) and 5 negative phenotype (5NP: ER-, PR-, HER2-,cytokeratin (CK) 5/6-, and EGFR-) (Reprinted from PublicLibrary of Science: [Plos Medicine] Blows et al. [13]).
Introduction
Breast cancer, the most common form of cancer in women, anestimated 232,340 new cases and 39,620 deaths in 2013 in the USalone [1] with worldwide more than 1,300,000 cases and 450,000deaths each year [2] was first documented in ancient Egyptianwritings, the Edwin Smith Papyrus (copy of trauma surgery) where8 cases of breast malignancy were recorded. Treatments at that timewere restricted to cauterization of the breast at the tumor site untilthe late seventeenth century when French surgeon Jean Louis Petitand Scottish surgeon Benjamin Bell were able to remove the diseasedbreast tissue and lymph node along with the malignant tumor.This was initially followed by mastectomy and then, with subsequentimpressive therapies such as lumpectomy, chemo and radiotherapy,providing better treatment options; increasing average survival ratesof breast cancer patients. Interestingly, in spite of enormous failurerate and lethal side effects, chemotherapy still finds the widestapplication in terms of treatment amongst all established therapies.Prominent chemotherapeutic drugs including doxorubicin, cisplatin,gemcitabine, bevacizumab and trastuzumab are in practice today forbreast cancer management. However, worldwide usages of allestablished cancer therapies have reported substantial inter-patientdifferences in therapeutic response. Any particular therapy can proveto be effective in some patients but ineffective in others with someexperiencing adverse drug reactions (ADRs) resulting in patientmorbidity and mortality, while some remain unaffected [3]. Eventhe blockbuster drugs (generating more than $1 billion dollarannually) show efficacies in 40–60% of the patients whereas 50%(estimated) of cancer patients fail to get benefited from chemother-apy. Reasons include intrinsic or acquired multidrug resistance (MDR)[4,5], DNA polymorphisms [6] and most importantly the presence ofinter-tumor heterogenic subpopulations, responding to radio, chemoand targeted therapies differently amongst different individualswithin the same cancer type. This inter-individual difference inresponse to drug treatment thus strongly commends a paradigmshift from “One Drug Fits All” strategy towards “PersonalizedMedicine”. This can be achieved through identifying genetic variantsof the same disease and tailoring targeted therapies towards theentire spectrum of mutations that collectively represent individualtumor sub-populations [7,8], providing improved therapeuticoutcomes.
Science driving personalized breast cancer therapies –
advancements and future perspectives
Identifying novel genetic subsets of breast cancer genome:disease stratificationOf late, breast cancer is considered not to be a single disease buta conglomerate of multiple subsets of genetically definable or
this article as: A. Nandy, et al., Individualizing breast can://dx.doi.org/10.1016/j.yexcr.2013.09.002
molecularly distinct syndromes, exhibiting different natural his-tories requiring different treatments for every patient or a parti-cular patient group. Thus a continuous approach has been todecode the heterogeneity into better characterized smaller subsetsto be used in predicting therapeutic and prognostic outcomesmaking way for personalized therapies [9]. Presently, this hetero-geneity is categorized clinically into estrogen receptor positive(ERþ), human epidermal growth factor receptor 2 (HER2; alsocalled ERBB2 or neu) positive (HER2þ) and triple-negative (ER� ,progesterone receptor negative (PR-) and HER2-) [2,10] with sixindependent intrinsic molecular subtypes such as normal-like,HER2-enriched (HER2E), luminal (A and B), basal A/basal-like andbasal B/claudin-low being reported in the past few years [10–12].Although a standard approach towards classifying breast cancersrelies on gene expression patterns, immunohistochemistry (IHC)based classification is preferred many a time (Fig. 1) [13] due tohigh expense and technical difficulties involved with gene expres-sion methods. However, to bring greater clarity into the robustnessof sensitive and accurate sub-classifications, different approachessuch as pathway-assisted clustering of plasma samples of breastcancers [14] and newly developed three-gene subtype classificationmodel (SCM), SCMGENE have been explained [15]; elucidatinglesser variability and simplicity over other complex classifiers.
The molecular architecture of breast cancer genome is howeverrevisited again and again with the help of high-throughputtechnologies such as the ever-evolving next-generation sequen-cing (NGS) shown in Figs. 2 and 3 [16], to extract the quantum of
cer treatment—The dawn of personalized medicine, Exp Cell Res
144145146147148149150151152153154155156157158159160161162163164165166167168169170171172173174175176177178179180181182183184185186187188189190191192193194195196197198199200201202203
Fig. 2 – Progress in DNA sequencing till the past decade andinto the future. Over the past three decades, rate of sequencinghas terrifically progressed by more than a million folds and isstill continuing to progress through the advancement ofsecond-generation sequencing platforms (Reprinted bypermission from Macmillan Publishers Ltd: [Nature] Strattonet al. [16]).
Fig. 3 – Drop down in genome sequencing costs. Cost ofgenome sequencing went down across several folds with theuse of next generation sequencing since January 2008,bringing down both cost and time of sequencing an entiregenome from $3 billion and 13 years till 2003 to just below$10,000 by 2013. (Wetterstrand KA. DNA Sequencing Costs:Data from the NHGRI Genome Sequencing Program (GSP)Available from: www.genome.gov/sequencingcosts (accessed22.07.13).
E X P E R I M E N T A L C E L L R E S E A R C H ] ( ] ] ] ] ) ] ] ] – ] ] ] 3
information that's elusive as well as evolving with time; aidingthe progress of personalized medicine [17]. Supportive to theseprogresses are a number of recent landmark studies acrossthousands of breast tumors unraveling the complex and diversegenomic and transcriptomic landscape [18–23]; suggestingrefined tumor classification and potential new targets throughuncovering of diverse somatic mutational hideouts. These somaticmutations which are often categorized into driver mutations: aidin oncogenesis and passenger mutations: unable to drive tumorgrowth or survival on their own [16,24–26] are sometimes acquiredwhen cells are in normal state or else after a cell has undergoneneoplastic transformation. Importantly, considering the first case,
Please cite this article as: A. Nandy, et al., Individualizing breast can(2013), http://dx.doi.org/10.1016/j.yexcr.2013.09.002
later the onset of cancer, more are the normal cells exposed tosequential acquisition of the drivers leading to a higher number ofmutations, suggesting that a majority of mutations in breastcancers occur after the initial driver event [18]. Also a numberof studies have demonstrated the presence of multiple drivers for8p11–12 [27–32], 11q13, 17q12 [33] and 19q12 amplicons [34].It can thus be inferred that breast tumors in general harbor a highernumber of mutations. Further investigations related to whether anydifference in tumor characteristics between both categories ofmutational acquisition amongst different sub-types, make thetumors more deadly or diversified, will lead to enhanced therapeuticand personalized outcomes.
Identifying challenges in individualizing therapies: heterogeneitybetween tumors, within tumors and in the microenvironmentAlthough new cancer genes are identified every year throughcontinued high-throughput efforts stating involvement of almost2% of genes in the human genome contributing to cancerdevelopment on getting mutated [33], the number is consistentlyrising. A possible reason might be the involvement of manyinfrequently mutated genes that significantly contribute to onco-genesis through innumerable combinations of multiple somaticevents [18]. This ultimately leads to heterogeneity of the disease,which indeed is a challenge to personalized medicine andbiomarker development. Even if this heterogeneity is oftenconfusing as to whether considering it as an outcome of aberrantdifferentiation or a contest amongst tumor cells with diversephenotypes [35], comprehensive understanding of thus bothinter- and intra-tumor heterogeneities exhibiting diverse char-acteristics at different stages of tumor progression will lead tobetter personalized outcomes. Now what drives these heterogenicevents in breast cancers? Well in case of inter-tumor hetero-geneity, two hypothetical models such as ‘same cell of origin’backed by different genetic and epigenetic mutations leading todifferent subtypes and ‘different cells of origin’ leading todifferent subtypes have been proposed [35,36]. However, acombination of the two sometimes might also take the centre-stage making a specific subtype more diverse and complicatedone to treat. Moving towards intra-tumor heterogeneity, modelssuch as ‘cancer stem cell model’ and ‘clonal evolution model’ areoften considered. While the former suggests precursor cancerstem cells (CSCs) having the capacity to self-renew and differ-entiate, give rise to different subpopulations in a tumor, the‘clonal evolution model’ suggests continuous evolution of tumorsoccurring through expansion of monoclonal or polyclonal sub-populations with most favorable characteristics [35,37,38]. Alsothese are supported by some varied models related to thepolyclonal evolution [37]. Moreover, intra-tumor phenotypicheterogeneity has given rise to multiple theories stating thisevent to be driven or accompanied by silent mutations increasingthe phenotypic plasticity, differentiation hierarchies in tumorpopulations, high stem cell marker expression leading to ther-apeutic resistance and metastases and presence of intra-tumorheterogenic microenvironments exerting different selective pres-sures inside the same tumor aiding in tumor evolution with full oftwists and turns [39,40]; something that’s going to greatly upsetthe progress of personalized medicine.Interestingly, from the metastases point of view, two hypothetical
models, ‘stepwise progression model’ and ‘parallel progressionmodel’ give rise to plausible trajectory of tumor heterogeneity in
cer treatment—The dawn of personalized medicine, Exp Cell Res
204205206207208209210211212213214215216217218219220221222223224225226227228229230231232233234235236237238239240241242243244245246247248249250251252253254255256257258259260261262263
E X P E R I M E N T A L C E L L R E S E A R C H ] ( ] ] ] ] ) ] ] ] – ] ] ]4
the metastatic spread as in the later case, tumor disseminationoccurs independently in parallel with primary tumor progressionresulting in spread of the disease from early stages [39,41]. Alsostudies describe that in spite of common ancestry, metastatic lesionsmany a times display additional mutations which are absent in theprimary ones suggesting further independent genetic divergenceand evolution [42,43]. This complexity can be further clarified by‘branched evolution model’ [42,44–46] or a quite similar and morepractical ‘trunk-branch model’ where on comparing intra-tumorheterogeneity with a growing tree, the trunk carries the foundingubiquitous driver mutations present in every regional sub-clonewhile branches spread, evolve and diversify into clones in geogra-phically isolated metastatic sites with the probability of passengersbecoming drivers and vice versa on some occasions [47]. Thissuggests, more branched is the evolution, more chances of hetero-genic mutative events (may be in the form of low-frequencysubpopulation) outnumbering the ubiquitous mutations which ingeneral are the targets for therapies; thus giving rise to drugresistance and failure of most advanced treatment options, puttingforth a diverse intra-patient phenotypic heterogeneity of tumorscoming from the same seed, jeopardizing the aims of personalizingtherapies.Another important aspect that challenges personalized medi-
cine considerably is the evolutionary interplay between tumorand its heterogenic microenvironment. Diverse interactions of theinitiating mutations with recruited stromal cells forming tumor-associated macrophages (TAMs), cancer-associated fibroblasts(CAFs) or vascular and perivascular cells [48] along with manyother components such as immune cells, extracellular matrix,secreted factors including chemokines, cytokines and exosomes,bone-marrow-derived-cells and lymphatic growth factors [49]add up to complexity of the tumor microenvironment (TME). Thisin turn actively promotes tumor growth and progression throughdiverse processes of neovascularisation mediated genomic insta-bility, resistance to therapies and protecting tumors from hostimmunity to name a few [49,39]. Thus, hunt for proper under-standing of the TME landscape over the past decades hasintriguingly led to recent findings which propose certain rate-limiting steps as building blocks of TME, discovering establish-ment and evolution of a ‘precancer niche’ through processes ofconstruction, expansion and maturation [50]. Construction of thisprecancer niche is proposed to be the earliest stage of carcino-genic stimulation of non-cancerous cells followed by expansionand maturation which eventually promotes neoplastic evolutiontowards a clinically established cancer, thereby initiating newmicroenvironments (premetastatic niche) at distant sites (Fig. 4).This probably infers that dissemination and establishment ofnewer premetastatic niche thrives to be a much early eventwhich creates a favorable premalignant environment at distantsites even before the primary tumor starts spreading its branchesas discussed in the ‘parallel progression model’ above. Thisconcept also goes well with the findings that metastatic lesionsmany a times harbor additional mutations which are absent in theprimary tumor as the disseminated premetastatic niches will havethe liberty to establish different microenvironments encompassingdifferent sets of neoplastic events in addition to the ones thatarrive through the primary tumor branches later. Hence, all theseintricacies altogether give rise to evolutionary and heterogeneousinteractions between the tumors and their microenvironments,putting forth diverse and difficult-to-treat recipes.
Please cite this article as: A. Nandy, et al., Individualizing breast can(2013), http://dx.doi.org/10.1016/j.yexcr.2013.09.002
Identifying approaches towards addressing heterogeneity: linesof controlOverall, the challenge to personalized medicine lies in betterunderstanding the evolutionary dynamics of the tumor and itsmicroenvironment in terms of a patient's inter- and intra-tumorheterogeneity. Even the presence of a minor tumor sub-popula-tion might stimulate the growth of tumor and support differentheterogenic events such as tumor microenvironment mediatedinnate resistance mutations leading to cancer drug resistance andtreatment failure [51]. Hence, decoding the diverse heterogenicevents will lead to proper understanding of tumor evolutionwhich can be done by isolating tumor sub-populations throughlikes of regional macro-dissection, laser capture microdissection[52], ploidy and allelic imbalance profiling [44], Sector-Ploidy-Profiling (SPP) [37] and more advanced plasma circulating cell-free tumor DNA sequencing technology [53,54] or by sequencinggenomes of individual tumor cells. However, in case of sequencingindividual tumor cells, recently single-nucleus sequencing (SNS)has shown promise by quantifying genomic copy numbers insingle cells. On profiling 100 single cells from a heterogeneoustriple-negative breast tumor along with 100 single cells from ahomogeneous primary breast tumor and its paired liver metas-tasis, it was possible to reconstruct the tumor evolutionarylineage towards better understanding of its pattern of progress[52]; decoding mode of intra-patient heterogenic tumor progressand predicting therapy, which is impressive.
On the contrary, targeting the heterogenic tumor ecosystem, i.e.the TME, is increasingly gaining momentum as scientists aredesperately in search for therapies that will detain this tumorsupport network system. Several TME targeted strategies: inter-fering with proangiogenic factor(s) mediated neovascularisation,chemokine mediated TAM polarization towards T-helper cell type2-driven protumorigenic inflammation, protumorigenic inflam-matory pathways, tumor cell-stroma communications, and hypoxiamediated therapeutic resistance [48] are already being evaluated indifferent phase clinical trials. Besides, CAFs in general make up thebulk of tumor stroma and studies demonstrate CAFs as the mostfrequent component of breast tumor stroma [55]. Thus targetingCAFs in breast TMEs have grabbed much attention with a range ofCAF targeted therapies currently under investigation both in pre-clinical and clinical trials [55]. However, heterogeneity between CAFswithin the same cancer type due to different cellular originsinteracts differently with different stromal cells leading to diversetumor survival processes; posing a striking challenge in personaliz-ing therapies. Also studies reveal that many TME targeted agentsalter the homeostatic balance in normal tissues by targeting cellularpathways not involved in tumorigenesis [48]. Hence, developingreliable biomarkers signifying accurate TME characteristics is aprerequisite for TME targeted therapies to succeed.
Another futuristic approach towards personalizing therapiesmight be the detection of circulating tumor cells (CTCs) in peripheralblood as CTCs might be released from distinct tumor lesions [56]depicting intra-patient inter- and intra-tumor heterogeneities. More-over, this painless liquid-biopsy is thought to represent the meta-static characteristics better than primary tumors. Therefore, a CTCcount across different time-points during metastatic breast cancertreatment might serve as a strong prognostic factor [56,57] formonitoring mutations from both overt and occult metastatic sites.Hence, continuous efforts in better characterization of CTCs [58]in measuring HER2 (including phosphorylated HER2 [pHER2]),
cer treatment—The dawn of personalized medicine, Exp Cell Res
264265266267268269270271272273274275276277278279280281282283284285286287288289290291292293294295296297298299300301302303304305306307308309310311312313314315316317318319320321322323
Fig. 4 – Establishment and evolution of a precancer niche. Construction of precancer niche is the earliest step of carcinogenicstimulation of non-tumorigenic cells, supporting survival of an initiated clone. This is followed by niche expansion where itgenerates secreted factors (chemokines, cytokines, exosomes), followed by recruitment of bone-marrow-derived myeloid cells,fibroblasts and other inflammatory cells thus initiating new microenvironments (premetastatic niche) at distant sites. Maturationis the final step where the expanded niche matures into the tumor microenvironment. Step by step, all these events collectivelypromote neoplastic evolution towards a clinically established cancer. CAF: cancer associated fibroblast; IMC: immature myeloid cell(Reprinted by permission from Macmillan Publishers Ltd: [Nature Reviews Cancer] Barcellos-Hoff et al. [50]).
E X P E R I M E N T A L C E L L R E S E A R C H ] ( ] ] ] ] ) ] ] ] – ] ] ] 5
ER, EGFR, phosphorylated EGFR (pEGFR) [56], different breast cancerstem cell marker [59] and different multidrug resistance-relatedprotein expression levels [60] are being explored. Interestingly,keeping in mind the characterization techniques being tumorbiomarker-specific which can be utilized for specific cancers only,recent development of CTC-iChip, a multistage microfluidic device,declares CTC isolations involving strategies either dependent orindependent of surface antigen epithelial cell adhesion molecule(EpCAM) and thus applicable to almost all cancer types (epithelialand non epithelial) including the breast [61]. This will also outdo thedrawback of lower CTC detection levels (Z1 toZ5 CTCs/7.5 ml ofblood in breast cancer patients) [56] through prior technologiesbased on EpCAM-positivity of tumor cells. Supportive to these,important pivotal trials such as STIC CTC METABREAST and Endo-crine Therapy Index (ETI) for evaluating CTCs as prognostic markers,SWOG0500 and CirCe01 for evaluating CTCs as early surrogate ofchemotherapy efficiency and phase II Treat CTC, phase III DETECT III,and phase II CirCe XXX1 for evaluating CTCs as an indicator of tumorbiology are already making their mark towards CTC driven perso-nalized breast cancer therapies [57]. However, with respect to earlydisease detection which has been a major goal of cancer research, itis still mystifying whether CTCs can be found in detectable amountsat early stages of the disease with measurable amounts only foundat advanced stages. Conjunction to this, tumor microvesicles (TMVs)shed from tumor cell surfaces, in substantial amounts (million to
Please cite this article as: A. Nandy, et al., Individualizing breast can(2013), http://dx.doi.org/10.1016/j.yexcr.2013.09.002
several billion per ml of blood) only on acquisition of malignantphenotypes, seem to be highly promising; allowing early tumordetection [62,63]. Nonetheless, owing to the heterogenic and evolu-tionary nature of TMVs [63], a magnetic nanosensor technology hasrecently shown promise in measuring composition of TMVs besidesquantifying their numbers, in brain cancer patients [64]. Furtherinvestigation of this technology on other cancer types is currentlyunderway. Hopefully, these advancements will someday result inmere simple blood tests leading to better prognosis and persona-lized treatment decisions.
Identifying approaches in directing therapies towards novelprotein targets for individualizing therapiesWith the help of advanced protein analytical tools, researchers aretrying to drill down and understand the real villains involved inbreast tumorigenesis [65]. HER2 overexpression being a commonphenomenon, thus following successful phase II NEOSPHERE trial[66], recent phase III CLEOPATRA trial using newly developedpersonalized drug pertuzumab in combination with trastuzumaband docetaxel has demonstrated better patient-treatment out-comes in case of HER2þ metastatic breast cancer (mBC) bytargeting the potent HER2-HER3 heterodimer [67–71]. However,this new combinatory regimen accepted by FDA as a first-linetreatment for treatment-naive HER2þ mBC patients [72], if
cer treatment—The dawn of personalized medicine, Exp Cell Res
324325326327328329330331332333334335336337338339340341342343344345346347348349350351352353354355356357358359360361362363364365366367368369370371372373374375376377378379380381382383
E X P E R I M E N T A L C E L L R E S E A R C H ] ( ] ] ] ] ) ] ] ] – ] ] ]6
administered during pregnancy or if the patient becomes preg-nant during treatment, the fetus might have to face a potentialhazard as exposure to pertuzumab can result in embryo-fetaldeath and birth defects [73]. Another combinatorial regimen(trastuzumab, DC101 (an anti-murine VEGF receptor-2 (VEGFR2)antibody) and lapatinib) has been recently reported by Kodacket al. (2012) [74] for treating a more concentrated HER2þ
subpopulation, the HER2-amplified breast cancer brain metas-tases subgroup. Although this site-specific regimen by means ofcombined inhibition of VEGFR2 and HER2 demonstrated incre-ased anti-angiogenesis mediated prolonged median overall survi-val, further testing is compulsory. This will disclose whether thisanti-VEGF therapy contributes to increased tumor invasion andmetastases unlike bevacizumab and sorafenib/sunitinib (inhibi-tors for VEGF receptor) by enriching breast cancer stem cells(BCSCs), elucidating the potentiality of this combination therapyas a prospective future personalized regimen for this sub-population [75]. Choi et al.[76] on the other hand brought HER2progression to a standstill along with tumor senescence throughshutting down of cell cycle protein (cyclin D1) and inhibitingcyclin D-associated kinase (CDK) activity; revealing cell cyclecomponents as potential targets. Thus restricting tumor subsetsat the protein level through distinctive approaches is continu-ously experimented which led the FDA in approving a newpersonalized drug, Trastuzumab emtansine (T-DM1) from thephase III EMILIA study [77] in February 2013. This antibody-drug conjugate made of trastruzumab and chemotherapy DM1(derivative of maytansine) allows targeted delivery specifically toHER2þ mBC; minimizing exposure to normal cells. T-DM1, com-pared to lapatinib plus capecitabine treatment, showed improvedmedian progression-free survival (9.6 months versus 6.4 months)and median overall survival (30.9 months versus 25.1 months).However, although there has been a lot of advancement in
novel protein identification in malignant tumors, antibodies stillremain as a major tool for target validation at the protein level.This high reliance on antibody mediated target inhibition is due totheir efficient binding to large protein–protein interfaces. How-ever, antibodies are unfortunately backed with some majorlimitations of poor tissue-penetration power, difficulty in deliveryinto cells and involvement of huge time, money and labor ingenerating multi targeted monoclonal antibodies [78]. Related tothis, camelid derived nanobodies (Nbs), demonstrating high rateof similarity with human antibodies with higher efficacy inpenetrating tissues, recognizing uncommon, hidden epitopesinaccessible by conventional antibodies and easy to produce inlarge quantities in addition to re-formatting into multivalentproteins [79–82], are already being developed for superior out-comes. These outstanding features are thus being well exploitedin developing radioiodinated HER2-specific 5F7GGC Nb for ima-ging HER2 status prior trastuzumab-therapy [83], anti-ERBB1 Nbscoupled to PEG-liposomes on a multivalent platform [84] and arange of anti-HER2, overlapping (displacing trastuzumab from itsepitope on HER2) and non-overlapping (targeting epitopes onHER2 other than that targeted by trastuzumab) Nbs [85]; all ofwhich are stepping towards personalized care for the HER familysubtypes. Interestingly, Sukhanova et al. [86] reported conjuga-tion of Nbs with quantum dots as new generation ultra-smallcancer-diagnostic nanoprobes in targeting carcinoembryonic anti-gen found in colorectal, breast, ovarian, lung and pancreaticcancers.
Please cite this article as: A. Nandy, et al., Individualizing breast can(2013), http://dx.doi.org/10.1016/j.yexcr.2013.09.002
Thus all these novel findings point out to two things, innovativeapproach and involvement of intelligent technology in diggingdeep into individual cancer genomes for not finding which wellknown genes are mutated in a patient but to find which uniquelymutated genes does different patients carry for rapidly bring inindividualized treatment regimens to the clinic. Technology doesnot has to be intelligent only but also needs to be flawless, as thewidely used IHC assay for routine identification of HER2 over-expression, many a times result in generating false-positive orfalse-negative outcomes leading to improper diagnosis and treat-ments [71]. This erroneous diagnosis related treatments howevercan now be circumvented by reverse-phase protein microarray(RPMA) mediated functional protein pathway activation mapping,pinpointing existence of a HER2- subpopulation exhibiting pHER2levels similar to that of HER2þ tumors, the HER2�/pHER2þ
subgroup. This remarkable finding however could not be identi-fied by IHC or fluorescence in situ hybridization (FISH) analysis[87]; suggesting activation of HER2 receptor in absence of HER2overexpression. Additionally, RPMA determines activation/phos-phorylation status of other HER family receptors along withdownstream signaling endpoints, helping to map HER2 phosphor-ylation and signaling network. This essentially will lead to betterindividualized therapies as HER network could be mapped outdeciphering the mechanistic pathways in action; a network-basedapproach described in the next section.
Finally, as in the recent times, it has been understood that CSCsremain unaffected by chemo and radiotherapies reducing justbulk of the tumors thereby generating more CSCs and tumorrecurrence [75,88–90], thus a more concentrated focus should beto find out those genes or proteins that induce CSC developmentwithin a tumor subset. Conjunction to this, scientists havereported the discovery of an oncogenic transcription factor,Y-box binding protein-1 (YB-1), for the first time that inducesBCSC marker CD44 in TNBC tumors [91]. More such discoveries indifferent breast tumor subsets should be looked up to, as theymight serve as the most prominent and reliable novel therapeutictargets for eliminating BCSCs leading to not only highly effectivepersonalized patient outcomes but might also serve in removal ofthis deadly disease from its roots completely.
Seeking help from the emerging ‘Network Medicine’ – developingbetter individualized therapeuticsWith the evolution of our understanding in medical therapeutics,it has been noticed that a majority of drugs do not cure the exactdisease, but only aid in altering its respective signs and symptoms.Moreover, proteins function by interacting with other proteinsand biomolecules in the system, operating through networks.With �25,000 protein-coding genes, �1000 metabolites, unde-fined number of distinct proteins and RNAs along with 4100,000highly connected nodes/hubs constituting the human system; it isoften highlighted that a particular genetic abnormality not onlyresults in abnormal activity of its gene product but can alsospread through the links of the network, altering normal func-tioning of gene products having no prior defects. Thus bearingthis concept that a particular disease outcome involves severalpathobiological processes in a complex network and is not limitedto abnormal behavior in a single gene product, it has becomeincreasingly necessary to understand the human protein interac-tion network (interactome) for determining whether changes inthe interactome architecture can be utilized for predicting patient
cer treatment—The dawn of personalized medicine, Exp Cell Res
384385386387388389390391392393394395396397398399400401402403404405406407408409410411412413414415416417418419420421422423424425426427428429430431432433434435436437438439440441442443
E X P E R I M E N T A L C E L L R E S E A R C H ] ( ] ] ] ] ) ] ] ] – ] ] ] 7
outcomes. The human interactome includes protein interaction,metabolic, regulatory and RNA networks. Also of importance isthe understanding of ‘disease module’ which implies that directinteractors/components of a gene or molecule involved in adisease or process, are also involved in the same disease orprocess and are likely to be located in the vicinity of the network.This implicates that each disease will have linkage to a distinctneighborhood in the interactome. However, a gene, protein ormetabolite might be associated to several disease modules. Thisgives rise to possibility of overlapping of several disease modulesand thus identifying the one with exact pathophenptype of ourinterest will aid in tracking down all the disease related genes inthe network leading to development of better individualizedtherapeutics. To this end, besides the availability of linkagemapping and genome-wide associated (GWA) studies, a numberof highly advanced potential disease gene identification network-based tools in the form of ‘linkage methods’, ‘disease module-based methods’ and ‘diffusion-based methods’ have been recentlydeveloped. Thus with the help of continued GWA and whole-genome sequencing findings, these tools will serve in narrowingthe vast search space provided by the interactome, aiding inuplifting the process of individualizing therapies to the next level[92–94].
Utilizing somewhat similar concepts, a newly discovered tech-nology, DyNeMo (dynamic network modularity), that aids inanalyzing breast tumors in patients for determining a patient'sbest treatment options has been unveiled [95]. This technologyhas a software application consisting of an algorithm and aninteractome database that analyzes protein networks in malig-nant cells and has the ability of predicting a particular patient'sprobability of recovering from breast cancer with 480% accuracy.This allows oncologists to provide an individualized analysis totheir patients for selection of treatment options that will suit bestto their requirements. This tool also has a unique capability ofmaking inferences about the entire protein interaction network(human interactome size estimate of �130,000 interactions) atonce based on expression ratios of the interacting proteins,making it different from other biomarker assessment tools[96–98]. However, Tyagi et al. [99] reported that current datasetsfrom large-scale mapping of the human interactome by means ofhigh-throughput methods, still suffer from a high rate of false-positives and low coverage, thus insisting the use of a frameworkwhich will aid in consistent inference along with the bindinginterfaces. Hence, by means of network medicine, one can get abird's eye view of the human interactome for proper understand-ing of cellular interconnectivity that aids in disease progression.Thereafter one can zoom into a particular disease module-basedlocality of concern for identifying the entire range of genes andpathways associated with the disease which will help in develop-ing multi-targeted therapeutics on an individualized basis.
Towards personalized oncology-based clinical trials: where arewe heading?In 2008, signing of the Genetic Information Nondiscrimination Act(GINA) into law by the WHITE HOUSE, research towards persona-lizing medications took a big leap. This act ensured to protect theprivacy of the personal genetic information derived from anindividual's genetic tests so as to prevent the genetic discrimina-tions such as decisions related to employment and health insur-ance coverage. Indeed people could now take full advantage of the
Please cite this article as: A. Nandy, et al., Individualizing breast can(2013), http://dx.doi.org/10.1016/j.yexcr.2013.09.002
promising personalized care without the need of getting afraid ofany such discrimination [100,101]; bringing researchers hope forpersonalized oncology-based research on a large scale.However, in spite of aforesaid major advancements, personalized
oncology faces some strong challenges. One of the biggest chal-lenges is to conduct clinical trials of newly developed targetedtherapies which demands transforming the design and set up ofthe present population oncology-based clinical trials towards morepersonalized oncology-based trials [71]. A set of different pre-screening strategies in designing effective early stage trials forappropriate selection of patient sub-population in assessingtargeted drug candidates has been depicted by Rodon et al. [102].Additionally, availability of a range of proper diagnostics forauthentic drug-target identification will reduce both time and costof clinical trials in the era of personalized oncology. This will allowcorrect pre-screening of individuals in early phase clinical trialsbased on enrichment biomarkers that have the potential of turninginto predictive biomarkers in late stage trials. Hence, approachingtowards an era of personalized oncology, identification of properbiomarkers might serve in improving success rates of breast cancerclinical trials as trials targeting the HER2 biomarker revealed anescalation in drug development success rate from 15% to 23%,projecting a cost reduction by millions of dollars [103]. Somewhatsimilar ideologies are being executed in the ongoing phase II I-SPY2trial for advanced breast cancers where five or more investigationaldrugs from different companies are being tested simultaneouslyand are matched to standard, qualifying and exploratory biomar-kers. Following this, they are either dropped out or promoted tothe next level (phase III), thereby cycling in new drug candidates;reducing time and cost of carrying out separate trials with a visionof 300-patient instead of 3000-patient phase III trial [104–106].This move towards speedy personalized treatments will expectablybring down the huge costs involved and the number of unsuccessful,expensive phase III trials. However, concerns about failure of suchimpressive trials due to continuous tumor evolution loom overlargely, challenging correct patient selection in trials. In this regard,accurate CTC or TMV characterizations at frequent time-intervalsmight prove to be a trouble-free and highly beneficial real-timemonitoring of tumors' progress, leading to near-accurate patient anddrug selection in prospective stratified medicine trials.
Conclusion
To conclude, detailed understanding of the evolutionary driverevents in tumor initiation, progression and identifying those sub-populations having ability of swapping from one morphology tothe other [107] along with accurate CTC or TMV characterizations,might in not-so-distant future lead to improved understanding ofthe degree of intra-patient tumor heterogeneity, which is yetdifficult to figure out. Importantly, targeting the tumors or theirmicroenvironments separately will only lead to partial therapeuticsuccess and thus, while addressing an individual's heterogenictumor-TME evolutionary processes, combinatorial or multi-targeted therapeutics are a prerequisite for optimal outcomes.Additionally, developing high quality and reliable statistical orcomputational tools through network medicinal approaches willlead to integrated understanding towards developing a map ofthe cell's intricate wiring diagram. This will ultimately aid in
cer treatment—The dawn of personalized medicine, Exp Cell Res
444445446447448449450451452453454455456457458459460461462463464465466467468469470471472473474475476477478479480481482483484485486487488489490491492493494495496497498499500501502503
E X P E R I M E N T A L C E L L R E S E A R C H ] ( ] ] ] ] ) ] ] ] – ] ] ]8
locating the different disease modules affected along with thoseprone to get affected through tumor evolution and progress.Besides two of the newly developed diagnostics associated with
CTCs and TMVs discussed above, a range of personalized biomar-ker based diagnostics including Mammostrats [108], SPOT-Lights
HER2 CISHTM Kit [109], HercepTest™ [110], MammaPrints [111],BluePrintTM [112], Oncotype DXs [113], Breast Cancer IndexSM
[114] and TargetPrints [115] have been developed to stratifypatient sub-populations accurately. But, owing to the largenumber of brand new mutative events that we come across everyyear due to continuous tumor evolution and the exceptionalhigher average cost of about USD$1.8 billion [116] and an averagetime of 10–16 years in drug development, developing nextgeneration anticancer drugs against each and every mutation,suited to benefit only a few patient sub-populations for apersonalized outcome seems to be a nightmare. Thus, instead oftargeting each tumor clone individually, alternative approaches infinding mutations which might have co-dependencies on eachother (such as ERþ and ER-, HER2þ and HER2�) can be targetedwith combinatory drugs which will attack both essential mutantclones and their community collectively [107]. Another approachmight be in identifying events which drive tumor survival path-ways helping in tumor progress and heterogeneity. Inactivatinggenome-instability survival pathways such as cellular metabolicprocesses [117] thus might turn out to be economically viable inthat sense.
Conflict of interest statement
The authors declare no known conflicts of interest associated withthis publication.
Funding
This review did not receive any specific grant from any fundingagency in the public, commercial or not for-profit sector.
Acknowledgments
The authors take full responsibility for the content of thispublication and confirm that it reflects their viewpoint and medicalexpertise. They also wish to acknowledge all the staff members ofNetaji Subhas Chandra Bose Cancer Research Institute, Kolkata,India for their support.
r e f e r e n c e s
[1] R. Siegel, D. Naishadham, A. Jemal, Cancer Statistics, 2013, CACancer J. Clin. 63 (2013) 11–30.
[2] D.C. Koboldt, R.S. Fulton, M.D. McLellan, H. Schmidt, J. Kalicki-Veizer,J.F. McMichael, et al., Comprehensive molecular portraits of humanbreast tumours, Nature 490 (2012) 61–70.
[3] L. Mancinelli, M. Cronin, W. Sadée, Pharmacogenomics: thepromise of personalized medicine, AAPS PharmSci. 2 (2000) E4.
[4] S.X. Hu, T. Foster, A. Kieffaber, Pharmacogenomics and persona-lized medicine: mapping of future value creation, Biotechniques39 (2005) S1–S6.
Please cite this article as: A. Nandy, et al., Individualizing breast can(2013), http://dx.doi.org/10.1016/j.yexcr.2013.09.002
[5] C. O’Brien, G. Cavet, A. Pandita, X. Hu, L. Haydu, S. Mohan, et al.,Functional genomics identifies ABCC3 as a mediator of taxaneresistance in HER2-amplified breast cancer, Cancer Res. 68(2008) 5380–5389.
[6] Personalized Medicine: An Introduction, ⟨http://www.personalizedmedicinecoalition.org/sites/default/files/personalmed_backgrounder.pdf⟩ (accessed 18.07.13).
[7] R.S. Huang, M.J. Ratain, Pharmacogenetics and pharmacoge-nomics of anticancer agents, CA Cancer. J. Clin. 59 (2009) 42–55.
[8] E.A. Perez, Breast cancer management: opportunities and bar-riers to an individualized approach, Oncologist 16 (2011) 20–22.
[9] The Breast Cancer Research Foundation, ⟨http://bcrfcure.org/action_meetres_hortobagyi.html⟩ (accessed 18.07.13.
[10] J.I. Herschkowitz, K. Simin, V.J. Weigman, I. Mikaelian, J. Usary,Z. Hu, et al., Identification of conserved gene expression featuresbetween murine mammary carcinoma models and human breasttumors, Genome Biol. 8 (2007) R76.
[11] Z. Hu, C. Fan, D.S. Oh, J.S. Marron, X. He, B.F. Qaqish, et al., Themolecular portraits of breast tumors are conserved acrossmicroarray platforms, BMC Genomics 7 (2006) 96.
[12] R.M. Neve, K. Chin, J. Fridlyand, J. Yeh, F.L. Baehner, T. Fevr, et al., Acollection of breast cancer cell lines for the study of functionallydistinct cancer subtypes, Cancer Cell 10 (2006) 515–527.
[13] F.M. Blows, K.E. Driver, M.K. Schmidt, A. Broeks, F.E. van Leeuwen,J. Wesseling, et al., Subtyping of breast cancer by immunohis-tochemistry to investigate a relationship between subtype andshort and long term survival: a collaborative analysis of data for10,159 cases from 12 studies, PLoS Med. 7 (2010) e1000279.
[14] F. Zhang, J.Y. Chen, Breast cancer subtyping from plasma proteins,BMC Med. Genomics 6 (2013) S6.
[15] B. Haibe-Kains, C. Desmedt, S. Loi, A.C. Culhane, G. Bontempi,J. Quackenbush, C. Sotiriou, A three-gene model to robustlyidentify breast cancer molecular subtypes, J. Natl. Cancer Inst.104 (2012) 311–325.
[16] M.R. Stratton, P.J. Campbell, P.A. Futreal, The cancer genome,Nature 458 (2009) 719–724.
[17] M. Wadelius, A. Alfirevic, Pharmacogenomics and personalizedmedicine: the plunge into next-generation sequencing, GenomeMed. 3 (2011) 78.
[18] P.J. Stephens, P.S. Tarpey, H. Davies, P. Van Loo, C. Greenman,D.C. Wedge, et al., The landscape of cancer genes and mutationalprocesses in breast cancer, Nature 486 (2012) 400–404.
[19] M.J. Ellis, L. Ding, D. Shen, J. Luo, V.J. Suman, J.W. Wallis, et al.,Whole-genome analysis informs breast cancer response toaromatase inhibition, Nature 486 (2012) 353–360.
[20] S. Banerji, K. Cibulskis, C. Rangel-Escareno, K.K. Brown,S.L. Carter, A.M. Frederick, et al., Sequence analysis of mutationsand translocations across breast cancer subtypes, Nature 486(2012) 405–409.
[21] S.P. Shah, A. Roth, R. Goya, A. Oloumi, G. Ha, Y. Zhao, et al., Theclonal and mutational evolution spectrum of primary triple-negative breast cancers, Nature 486 (2012) 395–399.
[22] C. Curtis, S.P. Shah, S.F. Chin, G. Turashvili, O.M. Rueda,M.J. Dunning, et al., The genomic and transcriptomic architectureof 2000 breast tumours reveals novel subgroups, Nature 486(2012) 346–352.
[23] L. Ding, M.J. Ellis, S. Li, D.E. Larson, K. Chen, J.W. Wallis, et al.,Genome remodelling in a basal-like breast cancer metastasis andxenograft, Nature 464 (2010) 999–1005.
[24] P.A. Futreal, L. Coin, M. Marshall, T. Down, T. Hubbard,R. Wooster, et al., A census of human cancer genes, Nat. Rev.Cancer 4 (2004) 177–183.
[25] I. Bozic, T. Antal, H. Ohtsuki, H. Carter, D. Kim, S. Chen, et al.,Accumulation of driver and passenger mutations during tumorprogression, Proc. Natl. Acad. Sci. U. S. A 107 (2010) 18545–18550.
[26] P.J. Stephens, D.J. McBride, M.L. Lin, I. Varela, E.D. Pleasance,J.T. Simpson, et al., Complex landscapes of somatic rearrangementin human breast cancer genomes, Nature 462 (2009) 1005–1010.
cer treatment—The dawn of personalized medicine, Exp Cell Res
504505506507508509510511512513514515516517518519520521522523524525526527528529530531532533534535536537538539540541542543544545546547548549550551552553554555556557558559560561562563
E X P E R I M E N T A L C E L L R E S E A R C H ] ( ] ] ] ] ) ] ] ] – ] ] ] 9
[27] D.G. Holland, A. Burleigh, A. Git, M.A. Goldgraben, P.A.Perez-Mancera, S.F. Chin, et al., ZNF703 is a common Luminal Bbreast cancer oncogene that differentially regulates luminal andbasal progenitors in human mammary epithelium, EMBO Mol.Med. 3 (2011) 167–180.
[28] I. Bernard-Pierrot, N. Gruel, N. Stransky, A. Vincent-Salomon,F. Reyal, V. Raynal, et al., Characterization of the recurrent 8p11-12 amplicon identifies PPAPDC1B, a phosphatase protein, as anew therapeutic target in breast cancer, Cancer Res. 68 (2008)7165–7175.
[29] J.S. Reis-Filho, P.T. Simpson, N.C. Turner, M.B. Lambros, C. Jones,A. Mackay, et al., FGFR1 emerges as a potential therapeutic targetfor lobular breast carcinomas, Clin. Cancer Res. 12 (2006) 6652–6662.
[30] K.L. Streicher, Z.Q. Yang, S. Draghici, S.P. Ethier, Transformingfunction of the LSM1 oncogene in human breast cancers with the8p11–12 amplicon, Oncogene 26 (2007) 2104–2114.
[31] N. Turner, A. Pearson, R. Sharpe, M. Lambros, F. Geyer, M.A.Lopez-Garcia, et al., FGFR1 amplification drives endocrine ther-apy resistance and is a therapeutic target in breast cancer, CancerRes. 70 (2010) 2085–2094.
[32] F. Sircoulomb, N. Nicolas, A. Ferrari, P. Finetti, I. Bekhouche,E. Rousselet, et al., ZNF703 gene amplification at 8p12 specifiesluminal B breast cancer, EMBO Mol. Med. 3 (2011) 153–166.
[33] T. Santarius, J. Shipley, D. Brewer, M.R. Stratton, C.S. Cooper, A census of amplified and overexpressed human cancer genes,Nat. Rev. Cancer 10 (2010) 59–64.
[34] R. Natrajan, A. Mackay, P.M. Wilkerson, M.B. Lambros, D. Wetterskog,M. Arnedos, et al., Functional characterization of the 19q12 ampliconin grade III breast cancers, Breast Cancer Res. 14 (2012) R53.
[35] K. Polyak, Breast cancer: origins and evolution, J. Clin. Invest. 117(2007) 3155–3163.
[36] J.E. Visvader, Cells of origin in cancer, Nature 469 (2011) 314–322.[37] N.E. Navin, J. Hicks, Tracing the tumor lineage, Mol. Oncol. 4
(2010) 267–283.[38] S.Y. Park, H.E. Lee, H. Li, M. Shipitsin, R. Gelman, K. Polyak,
Heterogeneity for stem cell-related markers according to tumorsubtype and histologic stage in breast cancer, Clin. Cancer Res. 16(2010) 876–887.
[39] A. Marusyk, V. Almendro, K. Polyak, Intra-tumour heterogeneity:a looking glass for cancer?, Nat. Rev. Cancer 12 (2012) 323–334.
[40] A. Marusyk, K. Polyak, Cancer. Cancer cell phenotypes, in fiftyshades of grey, Science 339 (2013) 528–529.
[41] C.A. Klein, Parallel progression of primary tumours and metas-tases, Nat. Rev. Cancer 9 (2009) 302–312.
[42] S.P. Shah, R.D. Morin, J. Khattra, L. Prentice, T. Pugh, A. Burleigh,et al., Mutational evolution in a lobular breast tumour profiled atsingle nucleotide resolution, Nature 461 (2009) 809–813.
[43] S. Yachida, S. Jones, I. Bozic, T. Antal, R. Leary, B. Fu, et al., Distantmetastasis occurs late during the genetic evolution of pancreaticcancer, Nature 467 (2010) 1114–1117.
[44] M. Gerlinger, A.J. Rowan, S. Horswell, J. Larkin, D. Endesfelder,E. Gronroos, et al., Intratumor heterogeneity and branchedevolution revealed by multiregion sequencing, N. Engl. J. Med.366 (2012) 883–892.
[45] K. Anderson, C. Lutz, F.W. van Delft, C.M. Bateman, Y. Guo,S.M. Colman, et al., Genetic variegation of clonal architecture andpropagating cells in leukaemia, Nature 469 (2011) 356–361.
[46] X. Wu, P.A. Northcott, A. Dubuc, A.J. Dupuy, D.J. Shih, H. Witt, et al.,Clonal selection drives genetic divergence of metastatic medullo-blastoma, Nature 482 (2012) 529–533.
[47] T.A. Yap, M. Gerlinger, P.A. Futreal, L. Pusztai, C. Swanton,Intratumor heterogeneity: seeing the wood for the trees, Sci.Transl. Med. 4 (2012) 127ps10.
[48] H. Fang, Y.A. Declerck, Targeting the tumor microenvironment:from understanding pathways to effective clinical trials, CancerRes. 73 (2013) 4965–4977.
Please cite this article as: A. Nandy, et al., Individualizing breast can(2013), http://dx.doi.org/10.1016/j.yexcr.2013.09.002
[49] M.A. Swartz, N. Iida, E.W. Roberts, S. Sangaletti, M.H. Wong,F.E. Yull, et al., Tumor microenvironment complexity: emergingroles in cancer therapy, Cancer Res. 72 (2012) 2473–2480.
[50] M.H. Barcellos-Hoff, D. Lyden, T.C. Wang, The evolution of thecancer niche during multistage carcinogenesis, Nat. Rev. Cancer13 (2013) 511–518.
[51] R. Fisher, L. Pusztai, C. Swanton, Cancer heterogeneity: implica-tions for targeted therapeutics, Br. J. Cancer 108 (2013) 479–485.
[52] H.G. Russnes, N. Navin, J. Hicks, A.L. Borresen-Dale, Insight intothe heterogeneity of breast cancer through next-generationsequencing, J. Clin. Invest. 121 (2011) 3810–3818.
[53] L.o.Y. Dennis, R.W. Chiu, Plasma nucleic acid analysis by mas-sively parallel sequencing: pathological insights and diagnosticimplications, J. Pathol. 225 (2011) 318–323.
[54] T. Forshew, M. Murtaza, C. Parkinson, D. Gale, D.W. Tsui, F. Kaper,et al., Noninvasive identification and monitoring of cancermutations by targeted deep sequencing of plasma DNA, Sci.Transl. Med. 4 (2012) 136ra68.
[55] Y. Mao, E.T. Keller, D.H. Garfield, K. Shen, J. Wang, Stromal cells intumor microenvironment and breast cancer, Cancer MetastasisRev. 32 (2013) 303–315.
[56] W. Onstenk, J.W. Gratama, J.A. Foekens, S. Sleijfer, Towards apersonalized breast cancer treatment approach guided bycirculating tumor cell (CTC) characteristics, Cancer Treat Rev. 39(2013) 691–700.
[57] F.C. Bidard, T. Fehm, M. Ignatiadis, J.B. Smerage, C. Alix-Panabières,W. Janni, et al., Clinical application of circulating tumor cells inbreast cancer: overview of the current interventional trials, CancerMetastasis Rev. 32 (2013) 179–188.
[58] M.J. Magbanua, E.V. Sosa, R. Roy, L.E. Eisenbud, J.H. Scott,A. Olshen, et al., Genomic profiling of isolated circulating tumorcells from metastatic breast cancer patients, Cancer Res. 73(2013) 30–40.
[59] N. Bednarz-Knoll, C. Alix-Panabières, K. Pantel, Clinical relevanceand biology of circulating tumor cells, Breast Cancer Res. 13(2011) 228.
[60] A. Gradilone, G. Naso, C. Raimondi, E. Cortesi, O. Gandini,B. Vincenzi, et al., Circulating tumor cells (CTCs) in metastaticbreast cancer (MBC): prognosis, drug resistance and phenotypiccharacterization, Ann. Oncol. 22 (2011) 86–92.
[61] E. Ozkumur, A.M. Shah, J.C. Ciciliano, B.L. Emmink, D.T. Miyamoto,E. Brachtel, et al., Inertial focusing for tumor antigen-dependentand -independent sorting of rare circulating tumor cells, Sci.Transl. Med. 5 (2013) 179ra47.
[62] Can we detect cancer earlier? ⟨http://www.health.harvard.edu/newsletters/Harvard_Health_Letter/2013/February/can-we-detect-cancer-earlier⟩ Harvard Health Letter, 2013 (accessed01.08.13).
[63] C. D’Souza-Schorey, J.W. Clancy, Tumor-derived microvesicles:shedding light on novel microenvironment modulators andprospective cancer biomarkers, Genes Dev. 26 (2012) 1287–1299.
[64] X. Shao, H. Chung, J. Balaj, L. Charest, A. Bigner, D.D. Carter BS, et al.,Protein typing of circulating microvesicles allows real-time mon-itoring of glioblastoma therapy, Nat. Med. 18 (2012) 1835–1840.
[65] Road to Personalized Cancer Treatments, ⟨http://www.mdanderson.org/publications/annual-report/issues/2004-2005/conquest-05-06ar.pdf⟩ (accessed 05.08.13).
[66] L. Gianni, T. Pienkowski, Y.H. Im, L. Roman, L.M. Tseng, M.C. Liu, et al.,Efficacy and safety of neoadjuvant pertuzumab and trastuzumab inwomen with locally advanced, inflammatory, or early HER2-positivebreast cancer (NeoSphere): a randomised multicentre, open-label,phase 2 trial, Lancet Oncol. 13 (2012) 25–32.
[67] J. Baselga, J. Cortés, S.B. Kim, S.A. Im, R. Hegg, Y.H. Im, et al.,Pertuzumab plus trastuzumab plus docetaxel for metastaticbreast cancer, N. Engl. J. Med. 366 (2012) 109–119.
[68] G.M. Keating, Pertuzumab: in the first-line treatment of HER2-positive metastatic breast cancer, Drugs 72 (2012) 353–360.
cer treatment—The dawn of personalized medicine, Exp Cell Res
564565566567568569570571572573574575576577578579580581582583584585586587588589590591592593594595596597598599600601602603604605606607608609610611612613614615616617618619620621622623
E X P E R I M E N T A L C E L L R E S E A R C H ] ( ] ] ] ] ) ] ] ] – ] ] ]10
[69] D. Mavroudis, Pertuzumab: a step forward in treating HER2-positive breast cancer, Transl. Cancer Res. 1 (2012) 117–118.
[70] A Study to Evaluate PertuzumabþTrastuzumabþDocetaxel vs.PlaceboþTrastuzumabþDocetaxel in Previously Untreated HER2-positive Metastatic Breast Cancer (CLEOPATRA), ⟨http://clinicaltrials.gov/show/NCT00567190⟩ (accessed 08.08.13).
[71] M.L. Maitland, R.L. Schilsky, Clinical trials in the era of persona-lized oncology, CA. Cancer. J. Clin. 61 (2011) 365–381.
[72] FDA approves Perjeta for type of late-stage breast cancer, ⟨http://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/ucm307549.htm⟩ (accessed 12.08.13).
[73] PERJETATM: Strengthen HER2n Suppression, ⟨http://www.perjeta.com/hcp/pdf/perjeta-core-data-brochure.pdf⟩ (accessed 17.08.13).
[74] D.P. Kodack, E. Chung, H. Yamashita, J. Incio, A.M. Duyverman,Y. Song, et al., Combined targeting of HER2 and VEGFR2 foreffective treatment of HER2-amplified breast cancer brainmetastases, Proc. Natl. Acad. Sci. U. S. A. 109 (2012) E3119–E3127.
[75] S.J. Conley, E. Gheordunescu, P. Kakarala, B. Newman, H. Korkaya,A.N. Heath, et al., Antiangiogenic agents increase breast cancerstem cells via the generation of tumor hypoxia, Proc. Natl. Acad.Sci. U. S. A. 109 (2012) 2784–2789.
[76] Y.J. Choi, X. Li, P. Hydbring, T. Sanda, J. Stefano, A.L. Christie, et al.,The requirement for cyclin D function in tumor maintenance,Cancer Cell 22 (2012) 438–451.
[77] S. Verma, D. Miles, L. Gianni, I.E. Krop, M. Welslau, J. Baselga, et al.,Trastuzumab emtansine for HER2-positive advanced breast cancer,N. Engl. J. Med. 367 (2012) 1783–1791.
[78] P.M. Watt, T.K. Heinrich, W.R. Thomas, Protein silencing withPhylomers: a new tool for target validation and generating leadbiologicals targeting protein interactions, Expert Opin. DrugDiscov. 1 (2006) 491–502.
[79] R.C. Roovers, T. Laeremans, L. Huang, S. De Taeye, A.J. Verkleij,H. Revets, et al., Efficient inhibition of EGFR signaling and oftumour growth by antagonistic anti-EFGR nanobodies, Cancer.Immunol. Immunother. 56 (2007) 303–317.
[80] K. Deffar, H. Shi, L. Li, X. Wang, X. Zhu, Nanobodies - the newconcept in antibody engineering, African J Biotechnology 8(2009) 2645–2652.
[81] F. Sheikholeslami, M.J. Rasaee, M.A. Shokrgozar, M.M. Dizaji,F. Rahbarizadeh, D. Ahmadvande, Isolation of a Novel NanobodyAgainst HER-2/neu Using Phage Displays Technology, Labmedicine41 (2010) 69–76.
[82] Understanding Nanobodies, ⟨http://www.ablynx.com/research-development/nanobody-technology/understanding-nanobodies/⟩ Accessed 19 August 2013.
[83] M. Pruszynski, E. Koumarianou, G. Vaidyanathan, H. Revets,N. Devoogdt, T. Lahoutte, et al., Targeting breast carcinoma withradioiodinated anti-HER2 Nanobody, Nucl. Med. Biol. 40 (2013)52–59.
[84] S. Oliveira, R.M. Schiffelers, J. van der Veeken, R. van der Meel,R. Vongpromek, P.M. van Bergen En Henegouwen, et al., Down-regulation of EGFR by a novel multivalent nanobody-liposomeplatform, J. Control. Release. 145 (2010) 165–175.
[85] F.R. Jamnani, F. Rahbarizadeh, M.A. Shokrgozar, D. Ahmadvand,F. Mahboudi, Z. Sharifzadeh, Targeting high affinity and epitope-distinct oligoclonal nanobodies to HER2 over-expressing tumorcells, Exp. Cell. Res. 318 (2012) 1112–1124.
[86] A. Sukhanova, K. Even-Desrumeaux, A. Kisserli, T. Tabary, B. Reveil,J.M. Millot, et al., Oriented conjugates of single-domain antibodiesand quantum dots: toward a new generation of ultrasmall diag-nostic nanoprobes, Nanomedicine. 8 (2012) 516–525.
[87] J.D. Wulfkuhle, D. Berg, C. Wolff, R. Langer, K. Tran, J. Illi, et al.,Molecular Analysis of HER2 Signaling in Human Breast Cancer byFunctional Protein Pathway Activation Mapping, Clin. Cancer.Res. 18 (2012) 6426–6435.
[88] L. Vermeulen, M.R. Sprick, K. Kemper, G. Stassi, J.P. Medema,Cancer stem cells–old concepts, new insights, Cell. Death. Differ.15 (2008) 947–958.
Please cite this article as: A. Nandy, et al., Individualizing breast can(2013), http://dx.doi.org/10.1016/j.yexcr.2013.09.002
[89] F. Pajonk, E. Vlashi, W.H. McBride, Radiation resistance of cancerstem cells: the 4 R’s of radiobiology revisited, Stem Cells 28(2010) 639–648.
[90] E. Vlashi, W.H. McBride, F. Pajonk, Radiation responses of cancerstem cells, J. Cell. Biochem. 108 (2009) 339–342.
[91] A.L. Stratford, K. Reipas, K. Hu, A. Fotovati, R. Brough, J. Frankum,et al., Targeting p90 ribosomal S6 kinase eliminates tumor-initiating cells by inactivating Y-box binding protein-1 in triple-negative breast cancers, Stem Cells 30 (2012) 1338–1348.
[92] A.L. Barabási, N. Gulbahce, J. Loscalzo, Network medicine: anetwork-based approach to human disease, Nat. Rev. Genet. 12(2011) 56–68.
[93] M. Ghoussaini, O. Fletcher, K. Michailidou, C. Turnbull,M.K. Schmidt, E. Dicks, et al., Genome-wide association analysisidentifies three new breast cancer susceptibility loci, Nat. Genet.44 (2012) 312–318.
[94] J.N. Hirschhorn, Genome wide association studies–illuminatingbiologic pathways, N. Engl. J. Med. 360 (2009) 1699–1701.
[95] New technology discovery at Mount Sinai Hospital holds promisefor improved breast cancer treatment, ⟨http://www.mountsinai.on.ca/about_us/news/2009/promise-for-improved-breast-cancer-treatment⟩ (accessed 18.08.13).
[96] Personalized healthcare through network biology, ⟨http://www.dynemobiosystems.com⟩ (accessed 18.08.13).
[97] K. Venkatesan, J.F. Rual, A. Vazquez, U. Stelzl, I. Lemmens,T. Hirozane-Kishikawa, et al., An empirical framework for binaryinteractome mapping, Nat. Methods 6 (2009) 83–90.
[98] I.W. Taylor, R. Linding, D. Warde-Farley, Y. Liu, C. Pesquita, D. Faria,et al., Dynamic modularity in protein interaction networks pre-dicts breast cancer outcome, Nat. Biotechnol. 27 (2009) 199–204.
[99] M. Tyagi, K. Hashimoto, B.A. Shoemaker, S. Wuchty, A.R. Pan-chenko, Large-scale mapping of human protein interactomeusing structural complexes, EMBO Rep. 13 (2012) 266–271.
[100] E.A. Feldman, The Genetic Information Nondiscrimination Act(GINA): public policy and medical practice in the age ofpersonalized medicine, J. Gen. Int. Med. 27 (2012) 743–746.
[101] Genetic Information Nondiscrimination Act (GINA) of 2008,⟨http://www.genome.gov/24519851⟩ (accessed 20.08.13).
[102] J. Rodón, C. Saura, R. Dienstmann, A. Vivancos, y. Ramón,S. Cajal, J. Baselga, et al., Molecular prescreening to selectpatient population in early clinical trials, Nat. Rev. Clin. Oncol. 9(2012) 359–366.
[103] Biological markers increase clinical trial success rate of newbreast cancer drugs, ⟨http://www.utm.utoronto.ca/main-news-research-news-general/biological-markers-increase-clinical-trial-success-rate-new-breast⟩ (accessed 20.08.13).
[104] I-SPY 2 TRIAL: Neoadjuvant and Personalized Adaptive NovelAgents to Treat Breast Cancer, ⟨http://clinicaltrials.gov/show/NCT01042379⟩ (accessed 20.08.13).
[105] A.D. Barker, C.C. Sigman, G.J. Kelloff, N.M. Hylton, D.A. Berry,L.J. Esserman, I-SPY 2: an adaptive breast cancer trial design inthe setting of neoadjuvant chemotherapy, Clin. Pharmacol. Ther.86 (2009) 97–100.
[106] Winslow R., A. New Rx for Medicine, ⟨http://online.wsj.com/article/SB10001424052748703882404575520190576846812.html⟩ Wall Street Journal, 2010 (accessed 21.08.13).
[107] S. Bhatia, J.V. Frangioni, R.M. Hoffman, A.J. Iafrate, K. Polyak, Thechallenges posed by cancer heterogeneity, Nat. Biotechnol. 30(2012) 604–610.
[108] J.M. Bartlett, K.J. Bloom, T. Piper, T.J. Lawton, C.J. van de Velde,D.T. Ross, et al., Mammostrat as an immunohistochemicalmultigene assay for prediction of early relapse risk in thetamoxifen versus exemestane adjuvant multicenter trialpathology study, J. Clin. Oncol. 30 (2012) 4477–4784.
[109] A. Cayre, F. Mishellany, N. Lagarde, F. Penault-Llorca, Compar-ison of different commercial kits for HER2 testing in breastcancer: looking for the accurate cutoff for amplification, BreastCancer Res. 9 (2007) R64.
cer treatment—The dawn of personalized medicine, Exp Cell Res
624625626627628629630631632633634635636
E X P E R I M E N T A L C E L L R E S E A R C H ] ( ] ] ] ] ) ] ] ] – ] ] ] 11
[110] HercepTest™, ⟨http://www.dako.com/dist/ar39/p101510/prod_products.htm?setCountry=true&purl=ar39/p101510/prod_products.htm?undefined&submit=Accept%20country⟩ (accessed21.08.13).
[111] D. Holmes, Genetic test gets FDA approval, Nat. Rev. Cancer 7(2007) 162.
[112] O. Krijgsman, P. Roepman, W. Zwart, J.S. Carroll, S. Tian,F.A. de Snoo, et al., A diagnostic gene profile for molecularsubtyping of breast cancer associated with treatment response,Br. Cancer Res. Treat. 133 (2012) 37–47.
[113] C.M. Kelly, S. Krishnamurthy, G. Bianchini, J.K. Litton,A.M. Gonzalez-Angulo, G.N. Hortobagyi, et al., Utility of onco-type DX risk estimates in clinically intermediate risk hormone
Please cite this article as: A. Nandy, et al., Individualizing breast can(2013), http://dx.doi.org/10.1016/j.yexcr.2013.09.002
receptor-positive, HER2-normal, grade II, lymph node-negativebreast cancers, Cancer 116 (2010) 5161–5167.
[114] Breast Cancer IndexSM, ⟨http://www.biotheranostics.com/healthcare-professionals/hcp/bci/⟩ (accessed 21.08.13).
[115] P. Roepman, H.M. Horlings, O. Krijgsman, M. Kok, J.M. Bueno-de-Mesquita, R. Bender, Microarray-based determination of estro-gen receptor, progesterone receptor, and HER2 receptor statusin breast cancer, Clin. Cancer Res. 15 (2009) 7003–7011.
[116] S. Morgan, P. Grootendorst, J. Lexchin, C. Cunningham,D. Greyson, The cost of drug development: a systematic review,Health Policy 100 (2011) 4–17.
[117] Y. Zhao, E.B. Butler, M. Tan, Targeting cellular metabolism toimprove cancer therapeutics, Cell Death Dis. 4 (2013) e532.
cer treatment—The dawn of personalized medicine, Exp Cell Res