Gynecologic Oncology 133 (2014) 624631

Contents lists available at ScienceDirect

Gynecologic Oncology

j ourna l homepage: www.e lsev ie r .com/ locate /ygynoReviewPlatinum resistant ovarian cancer: What is it, who to treat and how tomeasure benefit?Alison Davis a,, Anna V. Tinker b, Michael Friedlander c

a The Canberra Hospital and The Australian National University, Canberra, Australiab Vancouver Centre, British Columbia Cancer Agency, Vancouver, Canadac The Prince of Wales Clinical School, University of New South Wales, Sydney, Australia

H I G H L I G H T S

Platinum resistant ovarian cancer is comprised of a heterogeneous and complex spectrum of diseases. Detailing time to progression, method of determining progression, histotype and genomic information may all improve interpretation of clinical trial results. More active agents and better methods of assessing benefit are needed. Corresponding author at: Medical Oncology Unit, TheE-mail addresses: Alison.davis@act.gov.au, Alison_dav

http://dx.doi.org/10.1016/j.ygyno.2014.02.0380090-8258/Crown Copyright 2014 Published by Elseviea b s t r a c ta r t i c l e i n f oArticle history:Received 21 December 2013Accepted 25 February 2014Available online 4 March 2014

Keywords:Platinum resistantOvarian cancerHistotypeGenotypeTreatmentPlatinum resistant ovarian cancerwas historically defined as disease recurrencewithin 6 months of completionof first-line platinum-based chemotherapy, although this is now more broadly applied to also include patientsprogressingwithin 6 months aftermultiple lines of chemotherapy. However, this definition ignores the heteroge-neity and complexity of the spectrumof diseases that comprise platinum resistant ovarian cancer (PROC) and isinnatelyflawed as itwas initially derivedusingmethods of detection of recurrence thatwouldnowbe regardedasoutdated. The outcome of patients with PROC is generally poor, with low response rates to further chemotherapyand a median survival of less than 12 months, but this is unpredictable and can be quite variable from study tostudy. This reviewoutlines the complexity of PROC, examines how this impacts on the interpretation of the resultsof clinical trials, and explores how the definition may be improved. We also briefly describe the mechanisms ofplatinum resistance, the results of clinical trials to date as well as treatment options for patients with PROC andhighlight the need for better methods of assessing clinical benefit in this poor prognostic sub group of patients.

Crown Copyright 2014 Published by Elsevier Inc. All rights reserved.ContentsIntroduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 625Platinum resistant disease: what is it? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 625Relationship between specific histotype and response to chemotherapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 625The Role of BRCA Mutations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 626The BRCAness profile . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 626Mechanisms of resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 626Tumor heterogeneity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 627Platinum resistant disease: outcomes of therapy . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 627Who to treat, how to treat and for how long? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 628Chemotherapy-response assays . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 629How long to treat and how many lines of therapy? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 629What to measure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 629Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 629Conflict of interest statement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 630References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 630Canberra Hospital, Yamba Drive, Canberra ACT 2606, Australia.is90@hotmail.com (A. Davis).

r Inc. All rights reserved.

http://crossmark.crossref.org/dialog/?doi=10.1016/j.ygyno.2014.02.038&domain=pdfhttp://dx.doi.org/10.1016/j.ygyno.2014.02.038mailto:Alison.davis@act.gov.aumailto:Alison_davis90@hotmail.comhttp://dx.doi.org/10.1016/j.ygyno.2014.02.038http://www.sciencedirect.com/science/journal/00908258

625A. Davis et al. / Gynecologic Oncology 133 (2014) 624631Introduction

Globally, approximately 224,000 women are diagnosed with ovari-an, primary peritoneal or fallopian tube cancer (henceforth referred toas ovarian cancer) annually with 140,000 deaths [1]. At the time ofdiagnosis, most patients have advanced stage disease and despitedebulking surgery and platinum-based chemotherapy, the majoritywill experience disease recurrence requiring further treatment. Themost widely used predictor of the likelihood of response to subsequentplatinum-based chemotherapy has been the platinum free interval. Pa-tients with recurrent ovarian cancer are typically categorized as havingeither platinum resistant or platinum sensitive disease, based on aplatinum-free interval of less than or greater than 6 months, but somestudies use progression free interval (PFI) to define PROC. AlthoughPFI is often considered to be synonymous with platinum-free interval,it is important to note that these can be different. The PFI is generally de-fined as the interval from diagnosis to recurrence/progression, whichwould include the period on treatment. Clinical studies in ovariancancer often use platinum-free interval to define patient eligibility orsubgroup ie platinum sensitive or resistant, but PFI tomeasure outcome.However, it is arguably more useful to think of the probability of plati-num response as a continuum rather than fixating on arbitrary timepoints to define patient subsets and determine treatment approaches.

The aim of this review is to explore the complexity and heterogene-ity of patients with platinum-resistant ovarian cancer (PROC), the po-tential impact this has on interpretation of clinical trial results andoutline the Gynecologic Cancer InterGroups' (GCIG) recommendationsfor better categorizing patients with recurrent ovarian cancer in clinicaltrials. We also briefly review the mechanisms of platinum resistance,and treatment options as well as outcomes. Given that the prognosisformost patientswith PROC is poor, and treatment to date has hadmin-imal effect on overall survival, one of the main goals of therapy is palli-ation and symptom improvement. We discuss the international effortsthat are underway tomeasure the impact of chemotherapy on symptombenefit, which despite its importance, has mostly been neglected inmeasured outcomes of clinical trials to date.Platinum resistant disease: what is it?

Patientswho initially respond to platinum-based chemotherapy andwho subsequently relapse 6 months or more after their initial treat-ment have been classified as platinum sensitive andmostwill respondto further platinum-based chemotherapy with response rates rangingfrom 30% to 90% [25]. The median survival of patients with platinumsensitive recurrent ovarian cancer is 2 years, but can range from onlya fewmonths to more than a decade. Many patients will receive multi-ple lines of treatment over time, but with few exceptionswill ultimatelydevelop platinum resistant disease.

Patients who relapse within 6 months of completing first-linetherapy have been classified as being platinum resistant and typicallyhave low response rates to subsequent chemotherapy (b15%), with aprogression free survival of 34 months and a median survival undera year, see Table 1. The 3 studies originally used to define platinum re-sistance determined recurrence based on clinical symptoms, clinicallyTable 1Definitions of platinum response.

Definition

Platinum free interval last platinum dose until progressive disease is documentedPlatinum refractory progression during or within 4 weeks of platinum-based first linePlatinum resistant relapse b 6 months after 1st line platinum-based chemotherapy.

Platinum sensitive relapse 6 months after 1st line platinum-based chemotherapy.

PFS: progression free interval OS: overall survival.detectable disease and/or radiological evidence of disease recurrence[2,6,7]. These studies were carried out in an era prior to the widespreaduse of CA125 to detect recurrence and when radiological techniqueswere relatively crude compared to the high resolution CT, MRI andPET scans now available. It is very likely that the patients in the originalstudies were quite different to asymptomatic patients with a risingCA125 who may be found to have recurrent disease on a CT or PET CTwithin 6 months of completing first line chemotherapy. As both aCA125 rise and radiologic findings may precede the clinical recurrenceof ovarian cancer by several months [8], these patients would have his-torically been considered platinum-sensitive, but are now classified asplatinum-resistant.

The platinum-free interval may therefore be influenced by the fre-quency and types of investigations a patient receives during follow upand this has several important implications. Firstly, patients with earlyrecurrence of disease detected by PET/CT or high resolution CT maynot be treatedwith platinum-based therapy due to being inappropriate-ly classified as platinum-resistant. Secondly, this could impact on theinterpretation of results of clinical trials conducted in PROC patientsdue to the heterogeneous mix of patients included in these studies.There is a progressively higher likelihood of response to second-lineplatinum-based therapy, and also to non-platinum agents, as theplatinum-free interval increases, see Table 2 [2,9]. Therefore the activityof new drugs or drug combinations in PROC could appear falsely high ifstudies include greater numbers of asymptomatic patients diagnosedwith PROC based purely on a rising Ca125 or with low volume diseaseon imaging, as these patients may have inherently more chemotherapysensitive disease compared to patients who have symptomatic progres-sionwithin 6 months of completingfirst-line chemotherapy. To compli-cate matters further, the definition of platinum resistance has beenextended to include patients who relapsewithin 6 months after secondor multiple lines of chemotherapy and thus patients classified asplatinum resistant are a very heterogeneous group.

To avoid confusion and improve clarity in clinical trials, the GCIGhas proposed that all future trials specify how the recurrence wasestablished (rising CA125, radiologically or symptomatically) and de-scribe the populations in terms of the interval since last line of platinum,rather than labeling patients as platinum sensitive or resistant [10].Relationship between specific histotype and responseto chemotherapy

Themost common form of epithelial ovarian cancer is high grade se-rous cancer (HGSC). The relationship between time-to-recurrence andplatinum-sensitivity is best applied to patients with HGSC, and possiblyhigh grade endometrioid cancers (EC). Clear cell carcinoma (CCC),mucinous carcinoma (MC), low grade endometrioid carcinoma andlow grade serous carcinoma (LGSC) are less commonhistotypes of ovar-ian cancer which behave very differently to HGSC, with respect topresentation, clinical course, response to chemotherapy, genetic riskfactors (e.g not BRCA associated) andmolecular aberrations and drivers,see Table 3. In general, CCC, MC and LGSC tend to be resistant toplatinum-based chemotherapy. Patients with advanced or metastaticCCC or MC have poor clinical outcomes, with median survivals of 20Comments

subject to the types and frequency of investigations during follow upchemotherapy median OS 35 months

median PFI ~ 3 monthsmedian OS 912 monthsmedian PFI 912 monthsmedian OS 2436 months

Table 2Relationship between progression free interval and response to second-line therapy [2].

Secondary response rates based on duration of prior response

Duration of prior response(months)

No. of regimens(N = 211)

Response rate (%)

b12 51 331217 60 5518 100 75

Includes patients who received multiple courses of second-line, platinum basedtherapy.

626 A. Davis et al. / Gynecologic Oncology 133 (2014) 624631and 14 months, respectively [11]. Historically, all the subtypes ofovarian cancer were studied together. The inclusion of patients withthese inherently different subtypes of ovarian cancer in clinical studiesof PROC introduces bias and makes it difficult to interpret the results.At the very least, patients with different histotypes of ovarian cancershould be stratified and ideally, should be studied independently asunique diseases.

The Role of BRCA Mutations

Germ-line mutations in BRCA1 and BRCA2 are established riskfactors for the development of HGSC of the ovary and are present inapproximately 1520% of women with ovarian cancer and an evenhigher proportion of those with HGSC [12,13]. BRCA mutation carrierswith ovarian cancer have a more favorable clinical course (longer sur-vival rates) and higher responsiveness to platinum-based therapies[1417]. A recent paper byAlsop et al noted that BRCAmutation carrierswith early recurrence of ovarian cancer (relapse occurring b6 monthsfollowing primary therapy including CA125 and radiological imagingto diagnose recurrence) had a higher chance of response to second-linechemotherapy with platinum based chemotherapy compared toBRCA1/2 wildtype cases [12]. While intriguing, this study includedonly 17 patients with BRCA mutations and PROC, therefore the find-ings need to be confirmed in a larger cohort.

However, knowing the BRCA status could have an impact on clinicaldecision-making and based on Alsops' findings it seems reasonable toconsider retreating patients with BRCA mutations and early recurrencewith platinum-based therapy rather than assuming that they have plat-inum resistant disease. Clinical trial results may also be influenced bythe proportion of patients included who have BRCA mutations and todate this has not been readily available information. The GCIG consen-sus paper suggested that it would be important to report the proportionof patients that are germ-line BRCA mutation carriers in clinical trialsand document their response to the preceding platinum treatment(in addition to time since last platinum treatment) as this could helpinterpret the activity of new therapies.

The BRCAness profile

Many sporadic HGSC share phenotypic characteristics with tumorsassociated with germ-line BRCA mutations. This so-called BRCAnessphenotype may be due to a variety of mechanisms associated withTable 3Clinical features of the different ovarian cancer histotypes.

HGSC CCC

Portion of cases (%) 70 12Genetic Risk Factors BRCA1/2 HNPCCPrecursor Lesions/Cell of Origin STIC, p53 signatures EndometriosisCommon stage at presentation advanced earlyResponse to Platinum-based therapy Chemo-sensitive Chemo-resistant,Molecular aberrations p53, BRCA1, BRCA2, HR defects PI3K, ARID1A, MS

HGSC: high grade serous carcinoma, CCC: clear cell carcinoma, EC: endometrioid carcinomapolyposis colorectal cancer, STIC: serous tubal intraepithelial carcinoma, HR: homologous recodefective homologous recombination (HR) which include epigenetichypermethylation of the BRCA1 promoter [18,19], somatic mutation ofBRCA1/2 [20,21], amplification of EMSY [22] resulting in BRCA silencing,loss of function mutations or methylation of the FANCONI anemiacomplex [23] and hypermethylation of RAD51C, amongst others.

A correlation between disruption of HR and response to platinumand other DNR repair targeted therapies including poly(ADP-ribose)polymerase (PARP) inhibitors, has been shown in several in vitrostudies [2426]. More recently, a BRCAness gene expression profilebased on samples of BRCA1/2-mutated tumors was able to predict plat-inum and PARP inhibitor responsiveness [27]. This study also found thatBRCAnesswas of independent prognostic value in a small group of pa-tients with sporadic ovarian cancers,most of whomhadHGSC. If this, orother gene expression profiles, can be validated it would be a useful toolfor stratification and patient selection in clinical trials of treatment forrecurrent ovarian cancer. In addition to BRCAmutation status, reportingthe proportion of patients with a BRCAness phenotype in clinicalstudies of PROC may also be important, as this may also impact on thelikelihood of response to platinum-based chemotherapy. Combined, pa-tientswith BRCAmutations and BRCAness phenotype could representover 50% of patients with HGSC [28].

Mechanisms of resistance

The variability in response to platinum chemotherapy of patientswith PROC is probably due tomultiple factors including the heterogene-ity of patients entered into clinical trials, how recurrence was deter-mined and is also likely to be related to the differing mechanism/s ofresistance. Despite an improved understanding of the complex inter-play between clonal selection and the microenvironment of resistanttumors we still have much to learn about the mechanisms of platinumresistance, which are complex and multiple [29].

The cytotoxic effect of cisplatin (and its analogs) involves active up-take into cells and subsequent binding to DNA, forming adducts whichthen leads to the production of intra-strand and inter-strand crosslinks,and in turn single or double strand DNA breaks. This DNA damage canresult in activation of the apoptosis cascade and cell death unless thecell can repair theDNA. Platinummay also causemitochondrial damage,which is associated with decreased ATPase activity and cell death [30].

Factors associated with resistance to platinum include those thatlimit the formation of cytotoxic platinum-DNA adducts and those thatprevent cell death occurring after platinum-adduct formation [31].The former may result from either reduced uptake of cisplatin intocells or increased efflux via alterations to transport proteins or byinactivation of intracellular cisplatin by conversion into cisplatin- thiolconjugates. The latter form of resistance may occur by increased DNArepair after adduct formation. There are five major DNA repair mecha-nisms including nucleotide excision repair (NER), mismatch repair(MMR), homologous recombination (HR), base excision repair (BER)and translesion synthesis. Alterations in various proteins associatedwith these repairmechanisms have been associatedwith platinum resis-tance, for example high levels of excision repair cross-complementationgroup 1 (ERCC1) protein [32], mutations or down-regulation of MLH1,MSH2 and MSH1 [33] and secondary mutations of BRCA1 or 2, whichEC MC LGSC

11 3 3HNPCC/BRCA none known none knownEndometriosis not known SBTearly early advanced.

radiosensitive Chemo-sensitive Chemo-resistant Chemo-resistantI PTEN, bcatenin, ARID1A, MSI KRAS, HER2 BRAF, KRAS, NRAS

, MC: mucinous carcinoma, LGSC: low grade serous carcinoma, HNPCC: hereditary non-mbination, SBT: serous borderline tumor, MSI: microsatellite instability.

Table 4Published GOG Phase 2 clinical trials with chemotherapy conducted in platinum-resistantovarian cancer.

Year ofpublication

Study Agent N RR (%)

2011 GOG-126R Nanoparticle, albumin-bound(nab) paclitaxel

47^ 23

GOG-186 F Docetaxel + trabectidin 53^/71# 302010 GOG 126 M Ixabepilone 49^ 14.3

GOG 146O Irofulven 61^^ 12.72009 GOG 126Q Pemetrexed 51^ 212008 GOG 186C Paclitaxel poliglumex 25^/49# 16

GOG 186D Karenitecin 18^/26 122006 GOG 126 L Cisplatin + gemcitabine 57^ 16

GOG 126 N Paclitaxel weekly 48^ 20.92005 GOG 146 L Capecitabine 27 8

GOG 126I 9-aminocampthothecan 58^ 142003 GOG 126 J Docetaxel 60^ 22.4

GOG 126 K Oxaliplatin 23^ 4.32001 GOG 126E Paclitaxel + valspodar (PSC833) 58^^ 8.62000 GOG 126G CI-958 25^ 41998 GOG 126C Altretamine

(hexamethylmelamine)36^ 10

response rate calculated for all evaluable patients.^ Patients with platinum free interval b 6 months.# b12 months.

612 months.^^ Platinum refractory.

627A. Davis et al. / Gynecologic Oncology 133 (2014) 624631can cause reversion to the BRCA genotype and reestablishment of BRCAfunction, hence increasing HR [34,35]. These various factors may eitherbe present at diagnosis or acquired over time.

Recently two models, the cancer stem cell (CSC) model and the en-vironment mediated drug resistance model (EMDR) have been pro-posed to explain the origin of chemotherapy resistant cells [36]. TheCSCmodel proposes that genetic and/or epigenetic alterations occurringin the multi-potent tissue-specific adult stem cells may result in malig-nant transformation into cancer stem cells. As these cells possess stemcell-like properties, including the ability for self renewal and celldivision, they multiply forming tumors and with further genetic or epi-genetic alterationsmay develop invasive properties allowing the tumorto metastasize to distant sites [37,38]. CSC appear to be intrinsically re-sistant to chemotherapy for a variety of different reasons and may rep-resent a major source of chemoresistant cells within tumors [3941]. Inthe EMDR model, resistance emerges as the cancer cells interact withtheir surrounding microenvironment and enter a quiescent state dueto the complex interplay between tumor and its microenvironment. Itis recognized that tumors that develop a prominent desmoplastic reac-tion have a poor prognosis and are associated with platinum resistance[42].

Patients with primary platinum refractory disease have intrinsicdrug resistance and either do not respond or progress very early ontreatment. Primary platinum refractory ovarian cancers are quite un-common and are usually seen with non-serous ovarian cancers suchas CCC or MC rather than the more common HGSC. It is likely that themechanisms of resistance between these various histotypes are verydifferent and beyond the scope of this review. Patients who have an ini-tial response to platinum chemotherapy are believed to have tumorswith a heterogeneous population of both intrinsically platinum resistantcells and also sensitive cells. The sensitive cells undergo apoptosisfollowing chemotherapy (tumor response) but the resistant subpopula-tion of cells persist and expand leading to early recurrence in plati-num resistant disease. Platinum sensitive patients may respondrecurrently to platinum, due to the regrowth of the sensitive population.Ultimately however, the sensitive cells may alter ormutate, renderingthem resistant, or the resistant cell population will outgrow thesensitive population.

Tumor heterogeneity

Beyond the recognition of different histotypes of ovarian cancer, thegenomic heterogeneitywithin tumors of the samehistotype has recent-ly been described. The Cancer Genome Atlas (TCGA) Research Networkrecently completed a thorough genomic analysis involving 489 HGSC[28]. They found p53 mutations were almost universal (96%) andBRCAmutations common (22%, both germ line and somaticmutations),but other mutations were uncommon (26%). However, there was astriking degree of genomic instability with frequent somatic copynumber alterations. Gene expression arrays demonstrated at leastfour distinct subtypes based on gene clustering, which they termed im-munoreactive, differentiated, proliferative andmesenchymal. They con-firmed the findings by Bowtell's groupwho performed gene expressionarrays on 285 serous and endometrioid ovarian cancers and identified 6separate molecular subtypes associated with different molecular, histo-logical characteristics as well as patient outcomes [43]. The subgroupsidentified by the two groups were shown to correlate [28]. Furtherwork is required to determine the impact of the different molecularsubtypes on likelihood of response to chemotherapy.

Heterogeneity also exists spatiallywithin the primary tumor and be-tween the primary and themetastases and has also been shown to existtemporally, when repeat biopsies are taken at different time points dur-ing the disease course [44,45]. This heterogeneity adds significantly tothe complexity of assessing or interpreting response to treatmentsand outcomes in patients. This is supported by anecdotal clinical obser-vations of patients with a differential response to treatment withprogression in one site and response in other sites. There are currentlyno established mechanisms for capturing how common this is, and noguidance or consensus on appropriate management of these patients.In future studies, particularly of targeted therapies, repeat biopsiesboth at the time of recurrence and following further treatment willbe essential to gain a better understanding of the mechanisms ofresistance.

Platinum resistant disease: outcomes of therapy

There have been numerous phase 2 studies conducted in patientswith PROC. Table 4 summarizes the published Gynecologic OncologyGroup (GOG) phase 2 studies of chemotherapy in this patient popula-tion. These studies appear to show an incremental increase in the re-sponse rate (defined by the RECIST criteria) over time, which may beexplained in several ways. Firstly, this may represent real improve-ments in therapy, with more effective agents being studied over time.More likely, this observation may be an artifact of patient selectionand the changing population of PROC, for the reasons described in detailearlier. Furthermore, it is uncertain how many patients with germ-lineBRCA mutations or somatic impairment in HR (BRCAness) wereincluded, which as discussed, may also affect likelihood of response.

To date, Phase 2 results involving targeted therapies have been dis-appointing, with the exception of bevacizumab, with response ratesb10% in most studies (see Table 5). Bevacizumab, a recombinant hu-manized monoclonal antibody that targets vascular endothelial growthfactor (VEGF)-A has single agent activity in recurrent ovarian cancer ofbetween 16 and 21% [46,47] and also has shown encouraging activityin PROC when combined with oral cyclophosphamide or weeklypaclitaxel [4850].

Phase 3 studies involving combinations of non-platinum chemo-therapy agents or combinations of chemotherapy and targeted agentshave often included PROC patients as a subgroup only. Until recently,none had shown a significant improvement in PFI or overall survival(OS), with a consistentmedian PFI of 34 months and OS approximate-ly 12 months, see Table 6. Two recently presented studies, involvingchemotherapy plus and an anti-angiogenic agent, have shown an im-proved PFI in PROC. The AURELIA study (involving chemotherapycombined with bevacizumab) showed an almost doubling of thePFI (3.4 vs 6.7 months, HR 0.48, p b 0.001) with the addition of

Table 5Published GOG phase 2 clinical trials with targeted therapies conducted in platinum resistant ovarian cancer (generally defined as platinum free interval b12 months).

Year of publication Study Agent N RR (%) PFS 6# (%)

2012 GOG 126 T Belinostat/carboplatin 29 7.4 GOG 170G Lapatinib 28 0 8GOG170M Dasatinib 34 0 20.6GOG 170 N Urokinase-derived peptide (A6) 31 0 6.5

2011 GOG 170I Temsirolimus 40^/54 9.3 24.1GOG 170 F Sorafenib 50^/59 3.4 24GOG 170 J Enzastaurin 16^/27 7 11

2010 GOG 170 K Mifopristone 24 4.5 2008 GOG 170H Vorinostat 27 3.7 7.4

GOG 170E Imatinib 56 1.8 162007 GOG 170D Bevacizumab 26^/62 21 40.32005 GOG 170C Gefitinib 27 4 14.81998 GOG 126B Cisplatin + cyclosporine A 26^ 11.5

Based on evaluable patients and includes all those with platinum free interval b12 months.^ patients with platinum free interval b6 months.# PFS: progression free survival.

628 A. Davis et al. / Gynecologic Oncology 133 (2014) 624631bevacizumab, but no improvement in OS (13.3 vs 16.6 months, HR 0.85,p = 0.171 [51]. Of note, 40% of patients in the chemotherapy alonearm received bevacizumab on progression. The TRINOVA 1 study(paclitaxel chemotherapy combined with an angiopoietin 1/2 inhibitor,trebananib) also showed an improvement in PFI (5.4 vs 7.2 months, pb 0.001) [52]. Also, a randomized Phase 2 study, PRECEDENT, examin-ing the addition of EC145, a folate targeted vinca alkaloid to liposomaldoxorubicin has shown an improved PFI (2.7 vs 5 months, p = 0.031)with the addition of EC145 [70] and the subsequent Phase 3 study(PROCEED) is still accruing.

Who to treat, how to treat and for how long?

Using the widely accepted definition of platinum resistant diseaseie. recurrence diagnosed within 6 months of completion of first orTable 6Phase 3 clinical trials in platinum resistant ovarian cancer.

Drugs N RR (%) PFS(months)

PaclitaxelTopotecan

5960

6.713.3

PaclitaxelWeekly paclitaxel

5751

2619

TopotecanPLD

124130

6.512.3

3.1a

2.1a

PaclitaxelPaclitaxel + epidoxorubicin

9091

46.937.4

66

PLDGemcitabine

9699

8.36.1

3.13.6

TopotecanTreosulfan

5757

19.37

4.2a

2.2a

PLD or TopotecanCanfosfamide

229232

10.94.3

4.32.3

PLDPLD/Trabectedin

117115

3.74

PLDPatupilone 417412

7.915.5

3.73.7

OVATUREweekly carboplatin+/phenoxodiol

7270

10

4.6a

3.5a

AURELIAPLD/Topotecan/weekly Paclitaxel+/bevacizumab

182179

1231

3.46.7

TRINOVA 1weekly paclitaxel+/trebananib

919 total 3038

5.47.2

RR: response rate, PFS: progression free survival, OS: overall survival.PLD: pegylated liposomal doxorubicin.

a converted from weeks to months.subsequent-line platinum-based chemotherapy, this will includeasymptomatic patients with no detectable disease on imaging and a ris-ing CA125, those with small volume recurrence who are also usuallyasymptomatic and patients with rapidly progressive, larger volume,symptomatic disease. Within this heterogeneous group of patientsthere will be a subset that will respond to further platinum therapy,but predicting who will benefit is very difficult and based on bestguess rather than objective criteria. Until such timewhenwe can betterdetermine if a patientwill derive benefit from further platinum it seemsreasonable to consider platinum therapy either as a single agent or incombination with other agents on a case by case basis, particularlywhen patients fall into a grey zone, for example prior response to plat-inum based chemotherapy and progress within 46 months after lastplatinum treatment. Carboplatin is generally well tolerated but recur-rent lines of therapy are known to increase the risk of a hypersensitivityOS(months)

Comments and References

Subgroup data[60]

Subgroup data[61]

9.5a

8.2aSubgroup data[62]

Partial platinum resistance (73% b6mo)[63]

13.512.7

[64]

11.2a

7.3aPre-planned sub-stratum of platinum resistant [65]

13.58.5

A trial of 3rd line therapyexperimental arm inferior [66]

Subgroup analysis[67]

12.713.2

[68]

10.5a

8.8aClosed early[69]

13.316.6

PFS p = 0.001OS p = 0.17[51]

17.319

PFS p b 0.001OS NS[52]

629A. Davis et al. / Gynecologic Oncology 133 (2014) 624631reaction and combination therapy is generally more toxic than singleagent therapy. As there is no evidence for a survival benefit from earlyreintroduction of chemotherapy in asymptomatic patients with earlyrecurrence, as determined by Ca125 progression alone [53], treatmentshould be delayed until patients are symptomatic or based on imagingfindings expected to become symptomatic in the near future.

Determining which patients with clearly platinum-resistant orplatinum-refractory ovarian cancer will benefit from treatment andhow to select patients for treatment is complex and is not clinicallywell defined. Clinicians typically rely on patient factors to direct suchdecisions including patient performance status and desire for additionaltherapy, presence/persistence of toxicity from prior lines of treatment,and symptoms needing palliation. On average, 1015% of patientswith PROC have an objective response to treatment and therefore atrial of therapy in appropriately selected, and informed, patients withan ECOG PS of 02 is reasonable and includes selecting patients whowould meet the eligibility criteria for inclusion into clinical trials. It isless certain what the benefit of treatment is in patients with a poor per-formance status and rapidly progressive disease and good palliative careis probably the most important intervention in this setting.

Chemotherapy is most often given as single agents, used sequential-ly in suitably selected patients. Commonly used chemotherapy drugs in-clude paclitaxel, liposomal doxorubicin, topotecan, gemcitabine andetoposide, amongst others. The choice of agent for an individual patientis dependent on the history of prior treatment, residual toxicities,patient preferences in terms of toxicity and the availability, cost andconvenience of treatments. Consideration should be given to addingbevacizumab to chemotherapy in select patients, although this requiresan open discussionwith the patient regarding the lack of proven surviv-al benefit, added toxicities and practical concerns such as potential cost.Whether other targeted agents will have a role in the future remains tobe determined.

Chemotherapy-response assays

Currently it is not possible to confidently predict the likelihood of anindividual patient's response to chemotherapy or targeted therapy.There have been a number of retrospective studies examining the corre-lation between chemo-response assays and patient outcomes in ovariancancer but results have been inconsistent [5456]. More recently, a pro-spective, multi-institutional study involving 262womenwith recurrentor persistent ovarian cancer, found that patients treated with an assaysensitive regimen had significantly improved PFS and OS, while no dif-ference in outcomes was seen between the intermediate and resistantgroups [57]. The majority of the patients (68%) had HGSC and 45%were platinum resistant. Patients were treated with one of 15 pro-spectively specified protocol treatments, determined by the treating on-cologist, who was blinded to the results of the assay during the initialtreatment phase, but was able to review assay results at diseaseprogression. The associationwith assay responsewas consistent in plat-inum sensitive and platinum resistant tumors. These results are inter-esting and if confirmed in prospective studies may provide a veryuseful tool for individualizing drug selection.

How long to treat and howmany lines of therapy?

The likelihood of clinical benefit or response to second-line platinumbased chemotherapy correlates with the time interval between primarychemotherapy and relapse.With subsequent lines of therapy it is recog-nized that the likelihood of response and the duration of response re-duces significantly. Hoskins et al retrospectively reviewed 120 womenwith recurrent ovarian cancer (both platinum sensitive and resistant)and determined that an interval of b6 months between two consecu-tive relapses was predictive of poor survival, and might be used todetermine whom not to treat or when to stop [58]. Griffiths et al subse-quently reviewed 274 women with PROC and confirmed that efficacydeclined rapidly with successive lines of therapy, in particular whenthe patient had primary platinum refractory disease. Patients were un-likely to benefit from further chemotherapy after failure of two linesof therapy. Given the high likelihood that further chemotherapy willbe ineffective in heavily pretreated patients with resistant/refractorydisease and the inevitable toxicities associated with treatment, it isimportant that clinicians discuss the option of supportive care alone asthis may well provide the best palliation.

What to measure

Standard endpoints for clinical trials of new agents or combinationsinclude RECIST and CA125 response as well as PFS and OS. Clinical trialsgenerally include patients with good performance status and youngerage, whichmay not be representative of the general patient population.As most studies have shown only a minority of patients with PROC willhave an objective response by RECIST or CA125 and that themedian PFSis only 34 months, how can we determine whether patients areobtaining a symptomatic or palliative benefit from treatment?

There have been very few studies in PROC that have included amea-sure of quality of life (QoL), and none that have shown improved QoL orsymptompalliation, despite this being themain goal of treatment. Thereis a general perception among oncologists that more patients appear tosubjectively benefit from treatment than are shown to meet criteria forresponse, but this has been poorly studied. In a small study, 5060% ofpatients appeared to benefit in terms of quality of life and emotionalwell being, while objective responses were seen in only 7 of 27 (26%)[59]. There is general consensus that we need an accurate measure ofthe palliative benefit of treatment to weigh against the toxicity of treat-ment, particularly in the setting of platinum resistant disease where re-sponse rates are low and of short duration. However, before we canassess changes in QoL and symptom benefit we need to better under-stand what the cancer related symptoms are, how frequently theyoccur, how important they are to the patient and whether we canreliably measure them. There are a number of validated QoL tools avail-able. Some are generic (SF-36, SF-12), others cancer specific (QLQ-C30,FACT-G, CARES-SF), disease specific (QOL-OVCA, QLQ-OV28, FACT-O forovarian) or symptom specific (FACT-Anemia and FACIT-Fatigue), butnone have been designed to specifically measure symptom benefit inthe recurrent disease setting. The GCIG SymptomBenefit study, is cur-rently recruiting, and aims to validate a measure of symptom benefitthat can be used as an endpoint in trials in PROC as well as prognosticmodels to predict which patients are most likely to benefit from furtherchemotherapy.

Conclusion

Platinum resistant ovarian cancer patients should not be regarded asa homogeneous and distinct population defined simply by the intervalfrom the last platinum chemotherapy or the time to recurrence. Ovariancancer is a complex malignancy and is comprised of multiple histologi-cal subtypes, with numerous genotypes and phenotypes within eachhistotype, which all potentially impact on biological behavior andresponse to chemotherapy. It is also a dynamic disease process with ge-netic and epigenetic alterations occurring and evolving both over timeand at different metastatic sites of disease.

The GCIG has proposed that, rather than making assumptions aboutlikelihood of response to treatment and labeling or categorizing recur-rent ovarian cancer as platinum sensitive or resistant, it is moreimportant to specify the time from last platinum therapy and use thisto categorize patients included in clinical trials as well as stratifyingfor the various histotypes. Information about BRCA, BRCAness pheno-type and any future relevant genomic information, should also be docu-mented where-ever possible, given the potential impact on response toplatinum-based chemotherapy.

630 A. Davis et al. / Gynecologic Oncology 133 (2014) 624631In the future it may be possible to determine whether a patient islikely to be sensitive or resistant to platinum or other agents, but untilthen we must rely on clinical parameters such as ECOG performancestatus, previous treatment history and residual toxicitieswhen selectingtreatment. Importantly, we also need to ensure that patients with PROCare deriving ameaningful benefit from palliative treatment, particularlygiven their poor prognosis.Conflict of interest statement

MF has received an honorarium for participation on a Roche advisory board. The otherauthors declare that they have no conflicts of interest to disclose.References

[1] GLOBOCAN 2008 v1.2 Cancer Incidence and Mortality Worldwide [Internet]; 2010.[2] Markman M, Rothman R, Hakes T, Reichman B, Hoskins W, Rubin S, et al. Second-

line platinum therapy in patients with ovarian cancer previously treated withcisplatin. J Clin Oncol 1991;9(3):38993.

[3] Rose PG, Fusco N, Fluellen L, Rodriguez M. Second-line therapy with paclitaxel andcarboplatin for recurrent disease following first-line therapy with paclitaxel andplatinum in ovarian or peritoneal carcinoma. J Clin Oncol 1998;16(4):14947.

[4] Balbi G, Di Prisco L, Musone R, Menditto A, Cassese E, Balbi C, et al. Second-line withpaclitaxel and carboplatin for recurrent disease following first paclitaxel and plati-num compounds in ovarian carcinoma. Eur J Gynaecol Oncol 2002;23(4):3479.

[5] Hoekstra AV, Hurteau JA, Kirschner CV, Rodriguez GC. The combination of monthlycarboplatin and weekly paclitaxel is highly active for the treatment of recurrentovarian cancer. Gynecol Oncol 2009;115(3):37781.

[6] Blackledge G, Lawton F, Redman C, Kelly K. Response of patients in phase II studies ofchemotherapy in ovarian cancer: implications for patient treatment and the designof phase II trials. Br J Cancer 1989;59(4):6503.

[7] Gore ME, Fryatt I, Wiltshaw E, Dawson T. Treatment of relapsed carcinoma ofthe ovary with cisplatin or carboplatin following initial treatment with thesecompounds. Gynecol Oncol 1990;36(2):20711.

[8] Goonewardene TI, Hall MR, Rustin GJ. Management of asymptomatic patients onfollow-up for ovarian cancer with rising CA-125 concentrations. Lancet Oncol2007;8(9):81321.

[9] Bookman MA, Malmstrom H, Bolis G, Gordon A, Lissoni A, Krebs JB, et al. Topotecanfor the treatment of advanced epithelial ovarian cancer: an open-label phase II studyin patients treated after prior chemotherapy that contained cisplatin or carboplatinand paclitaxel. J Clin Oncol 1998;16(10):334552.

[10] Friedlander M, Trimble E, Tinker A, Alberts D, Avall-Lundqvist E, Brady M, et al.Clinical trials in recurrent ovarian cancer. Int J Gynecol Cancer 2011;21(4):7715.

[11] Mackay HJ, BradyMF, Oza AM, Reuss A, Pujade-Lauraine E, Swart AM, et al. Prognos-tic relevance of uncommon ovarian histology in women with stage III/IV epithelialovarian cancer. Int J Gynecol Cancer 2010;20(6):94552.

[12] Alsop K, Fereday S, Meldrum C, deFazio A, Emmanuel C, George J, et al. BRCA muta-tion frequency and patterns of treatment response in BRCA mutation-positivewomen with ovarian cancer: a report from the Australian Ovarian Cancer StudyGroup. J Clin Oncol 2012;30(21):265463.

[13] Schrader KA, Hurlburt J, Kalloger SE, Hansford S, Young S, Huntsman DG, et al.Germline BRCA1 and BRCA2 mutations in ovarian cancer: utility of a histology-based referral strategy. Obstet Gynecol 2012;120(2 Pt 1):23540.

[14] Ben David Y, Chetrit A, Hirsh-Yechezkel G, Friedman E, Beck BD, Beller U, et al. Effectof BRCAmutations on the length of survival in epithelial ovarian tumors. J Clin Oncol2002;20(2):4636.

[15] Cass I, Baldwin RL, Varkey T, Moslehi R, Narod SA, Karlan BY. Improved survival inwomen with BRCA-associated ovarian carcinoma. Cancer 2003;97(9):218795.

[16] Tan DS, Rothermundt C, Thomas K, Bancroft E, Eeles R, Shanley S, et al. BRCAnesssyndrome in ovarian cancer: a casecontrol study describing the clinical featuresand outcome of patients with epithelial ovarian cancer associated with BRCA1 andBRCA2 mutations. J Clin Oncol 2008;26(34):55306.

[17] Vencken PM, KriegeM, Hoogwerf D, Beugelink S, van der BurgME, HooningMJ, et al.Chemosensitivity and outcome of BRCA1- and BRCA2-associated ovarian cancer pa-tients after first-line chemotherapy comparedwith sporadic ovarian cancer patients.Ann Oncol 2011;22(6):134652.

[18] Baldwin RL, Nemeth E, Tran H, Shvartsman H, Cass I, Narod S, et al. BRCA1 promoterregion hypermethylation in ovarian carcinoma: a population-based study. CancerRes 2000;60(19):532933.

[19] Esteller M, Silva JM, Dominguez G, Bonilla F, Matias-Guiu X, Lerma E, et al. Promoterhypermethylation and BRCA1 inactivation in sporadic breast and ovarian tumors.J Natl Cancer Inst 2000;92(7):5649.

[20] Foster KA, Harrington P, Kerr J, Russell P, DiCioccio RA, Scott IV, et al. Somatic andgermline mutations of the BRCA2 gene in sporadic ovarian cancer. Cancer Res1996;56(16):36225.

[21] Hilton JL, Geisler JP, Rathe JA, Hattermann-Zogg MA, DeYoung B, Buller RE. Inactiva-tion of BRCA1 and BRCA2 in ovarian cancer. J Natl Cancer Inst 2002;94(18):1396406.

[22] Hughes-Davies L, Huntsman D, Ruas M, Fuks F, Bye J, Chin SF, et al. EMSY links theBRCA2 pathway to sporadic breast and ovarian cancer. Cell 2003;115(5):52335.

[23] Taniguchi T, Tischkowitz M, Ameziane N, Hodgson SV, Mathew CG, Joenje H, et al.Disruption of the Fanconi anemia-BRCA pathway in cisplatin-sensitive ovariantumors. Nat Med 2003;9(5):56874.[24] McCabe N, Turner NC, Lord CJ, Kluzek K, Bialkowska A, Swift S, et al. Deficiency inthe repair of DNA damage by homologous recombination and sensitivity topoly(ADP-ribose) polymerase inhibition. Cancer Res 2006;66(16):810915.

[25] D'Andrea AD. The Fanconi Anemia/BRCA signaling pathway: disruption in cisplatin-sensitive ovarian cancers. Cell Cycle 2003;2(4):2902.

[26] Teodoridis JM, Hall J, Marsh S, Kannall HD, Smyth C, Curto J, et al. CpG island meth-ylation of DNA damage response genes in advanced ovarian cancer. Cancer Res2005;65(19):89617.

[27] Konstantinopoulos PA, Spentzos D, Karlan BY, Taniguchi T, Fountzilas E, Francoeur N,et al. Gene expression profile of BRCAness that correlates with responsiveness tochemotherapy and with outcome in patients with epithelial ovarian cancer. J ClinOncol 2010;28(22):355561.

[28] Cancer Genome Atlas Research N. Integrated genomic analyses of ovariancarcinoma. Nature 2011;474(7353):60915.

[29] Galluzzi L, Senovilla L, Vitale I, Michels J, Martins I, Kepp O, et al. Molecularmechanisms of cisplatin resistance. Oncogene 2012;31(15):186983.

[30] Siddik ZH. Cisplatin: mode of cytotoxic action and molecular basis of resistance.Oncogene 2003;22(47):726579.

[31] Tapia G, Diaz-Padilla I, editors. Molecular Mechanisms of Platinum resistance inOvarian cancer; 2013.

[32] Dabholkar M, Vionnet J, Bostick-Bruton F, Yu JJ, Reed E. Messenger RNA levels ofXPAC and ERCC1 in ovarian cancer tissue correlate with response to platinum-based chemotherapy. J Clin Invest 1994;94(2):7038.

[33] Helleman J, van Staveren IL, DinjensWN, van Kuijk PF, Ritstier K, Ewing PC, et al. Mis-match repair and treatment resistance in ovarian cancer. BMC Cancer 2006;6:201.

[34] Swisher EM, Sakai W, Karlan BY, Wurz K, Urban N, Taniguchi T. Secondary BRCA1mutations in BRCA1-mutated ovarian carcinomas with platinum resistance. CancerRes 2008;68(8):25816.

[35] Sakai W, Swisher EM, Karlan BY, Agarwal MK, Higgins J, Friedman C, et al. Secondarymutations as a mechanism of cisplatin resistance in BRCA2-mutated cancers. Nature2008;451(7182):111620.

[36] Zahreddine H, Borden KL. Mechanisms and insights into drug resistance in cancer.Front Pharmacol 2013;4:28.

[37] Dean M, Fojo T, Bates S. Tumour stem cells and drug resistance. Nat Rev Cancer2005;5(4):27584.

[38] Massard C, Deutsch E, Soria JC. Tumour stem cell-targeted treatment: elimination ordifferentiation. Ann Oncol 2006;17(11):16204.

[39] Mimeault M, Hauke R,Mehta PP, Batra SK. Recent advances in cancer stem/progenitorcell research: therapeutic implications for overcoming resistance to the most aggres-sive cancers. J Cell Mol Med 2007;11(5):9811011.

[40] Ma S, Lee TK, Zheng BJ, Chan KW, Guan XY. CD133+ HCC cancer stem cells conferchemoresistance by preferential expression of the Akt/PKB survival pathway.Oncogene 2008;27(12):174958.

[41] Hermann PC, Huber SL, Herrler T, Aicher A, Ellwart JW, Guba M, et al. Distinctpopulations of cancer stem cells determine tumor growth and metastatic activityin human pancreatic cancer. Cell Stem Cell 2007;1(3):31323.

[42] Helleman J, JansenMP, Burger C, van der BurgME, Berns EM. Integrated genomics ofchemotherapy resistant ovarian cancer: a role for extracellular matrix, TGFbeta andregulating microRNAs. Int J Biochem Cell Biol 2010;42(1):2530.

[43] Tothill RW, Tinker AV, George J, Brown R, Fox SB, Lade S, et al. Novel molecularsubtypes of serous and endometrioid ovarian cancer linked to clinical outcome.Clin Cancer Res 2008;14(16):5198208.

[44] Gerlinger M, Rowan AJ, Horswell S, Larkin J, Endesfelder D, Gronroos E, et al.Intratumor heterogeneity and branched evolution revealed bymultiregion sequenc-ing. N Engl J Med 2012;366(10):88392.

[45] Bashashati A, Ha G, Tone A, Ding J, Prentice LM, Roth A, et al. Distinct evolutionarytrajectories of primary high-grade serous ovarian cancers revealed through spatialmutational profiling. J Pathol 2013;231(1):2134.

[46] Cannistra SA, Matulonis UA, Penson RT, Hambleton J, Dupont J, Mackey H, et al.Phase II study of bevacizumab in patients with platinum-resistant ovarian canceror peritoneal serous cancer. J Clin Oncol 2007;25(33):51806.

[47] Burger RA, Sill MW, Monk BJ, Greer BE, Sorosky JI. Phase II trial of bevacizumab inpersistent or recurrent epithelial ovarian cancer or primary peritoneal cancer: aGynecologic Oncology Group Study. J Clin Oncol 2007;25(33):516571.

[48] Chura JC, Van Iseghem K, Downs Jr LS, Carson LF, Judson PL. Bevacizumab plus cyclo-phosphamide in heavily pretreated patients with recurrent ovarian cancer. GynecolOncol 2007;107(2):32630.

[49] Sanchez-Munoz A, Mendiola C, Perez-Ruiz E, Rodriguez-Sanchez CA, Jurado JM,Alonso-Carrion L, et al. Bevacizumab plus low-dose metronomic oral cyclophospha-mide in heavily pretreated patients with recurrent ovarian cancer. Oncology2010;79(12):98104.

[50] Markman M. Addition of bevacizumab to weekly paclitaxel significantly improvesprogression-free survival in heavily pretreated recurrent epithelial ovarian cancer.Gynecol Oncol 2012;124(1):171 [author reply -2].

[51] Witteveen P, Lortholary A, Fehm T, Poveda A, Reuss A, Havsteen H, et al. FInal overallsurvival (OS) results from AURELIA, an open-label randomised phase III trial ofchemotherapy (CT) with or without bevacizumab (BEV) for platinum-resistantrecurrent ovarian cancer (OS). European Cancer Congress LBA 5. Eur J Cancer2013;49(Suppl. 3):S3.

[52] Monk BJ, Poveda A, Vergote I, Raspagliesi F, Fujiwara K, Bae DS, et al. A phase III, ran-domized, double-blind trial of weekly paclitaxel plus the angiopoeitin 1 and 2 inhib-itor, trabananib, or placebo in women with recurrent ovarian cancer: TRINOVA-1.European Cancer Congress LBA 41. Eur J Cancer 2013;49(Suppl. 2):S18.

[53] Rustin GJ, van der Burg ME, Griffin CL, Guthrie D, Lamont A, Jayson GC, et al. Earlyversus delayed treatment of relapsed ovarian cancer (MRC OV05/EORTC 55955): arandomised trial. Lancet 2010;376(9747):115563.

http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0335http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0005http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0005http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0005http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0010http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0010http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0010http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0015http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0015http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0015http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0020http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0020http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0020http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0025http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0025http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0025http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0030http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0030http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0030http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0035http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0035http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0035http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0040http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0040http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0040http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0040http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0045http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0045http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0050http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0050http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0050http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0340http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0340http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0340http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0340http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0060http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0060http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0060http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0065http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0065http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0065http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0070http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0070http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0075http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0075http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0075http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0075http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0080http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0080http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0080http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0080http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0085http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0085http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0085http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0090http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0090http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0090http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0095http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0095http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0095http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0100http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0100http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0105http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0105http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0110http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0110http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0110http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0115http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0115http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0115http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0120http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0120http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0125http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0125http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0125http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0130http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0130http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0130http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0130http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0135http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0135http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0140http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0140http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0145http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0145http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0345http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0345http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0150http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0150http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0150http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0155http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0155http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0160http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0160http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0160http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0165http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0165http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0165http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0170http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0170http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0175http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0175http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0180http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0180http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0185http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0185http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0185http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0190http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0190http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0190http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0190http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0195http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0195http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0195http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0200http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0200http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0200http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0205http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0205http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0205http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0210http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0210http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0210http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0215http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0215http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0215http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0220http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0220http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0220http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0225http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0225http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0225http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0230http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0230http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0230http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0235http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0235http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0235http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0235http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0350http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0350http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0350http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0240http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0240http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0240http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0240http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0240http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0245http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0245http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0245http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0245http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0250http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0250http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0250

631A. Davis et al. / Gynecologic Oncology 133 (2014) 624631[54] Holloway RW, Mehta RS, Finkler NJ, Li KT, McLaren CE, Parker RJ, et al. Associationbetween in vitro platinum resistance in the EDR assay and clinical outcomes forovarian cancer patients. Gynecol Oncol 2002;87(1):816.

[55] Gallion H, Christopherson WA, Coleman RL, DeMars L, Herzog T, Hosford S, et al.Progression-free interval in ovarian cancer and predictive value of an ex vivochemoresponse assay. Int J Gynecol Cancer 2006;16(1):194201.

[56] Herzog TJ, Krivak TC, Fader AN, Coleman RL. Chemosensitivity testing withChemoFx and overall survival in primary ovarian cancer. Am J Obstet Gynecol2010;203(1) [68 e1-6].

[57] Rutherford T, Orr Jr J, Grendys Jr E, Edwards R, Krivak TC, Holloway R, et al. A pro-spective study evaluating the clinical relevance of a chemoresponse assay for treat-ment of patients with persistent or recurrent ovarian cancer. Gynecol Oncol2013;131(1):276.

[58] Hoskins PJ, Le N. Identifying patients unlikely to benefit from further chemotherapy:a descriptive study of outcome at each relapse in ovarian cancer. Gynecol Oncol2005;97(3):8629.

[59] Doyle C, Crump M, Pintilie M, Oza AM. Does palliative chemotherapy palliate?Evaluation of expectations, outcomes, and costs in women receiving chemotherapyfor advanced ovarian cancer. J Clin Oncol 2001;19(5):126674.

[60] ten Bokkel Huinink W, Gore M, Carmichael J, Gordon A, Malfetano J, Hudson I, et al.Topotecan versus paclitaxel for the treatment of recurrent epithelial ovarian cancer.J Clin Oncol 1997;15(6):218393.

[61] Rosenberg P, Andersson H, Boman K, Ridderheim M, Sorbe B, Puistola U, et al.Randomized trial of single agent paclitaxel given weekly versus every three weeksand with peroral versus intravenous steroid premedication to patients with ovariancancer previously treated with platinum. Acta Oncol 2002;41(5):41824.

[62] Gordon AN, Fleagle JT, Guthrie D, Parkin DE, GoreME, Lacave AJ. Recurrent epithelialovarian carcinoma: a randomized phase III study of pegylated liposomal doxorubicinversus topotecan. J Clin Oncol 2001;19(14):331222.

[63] Buda A, Floriani I, Rossi R, Colombo N, Torri V, Conte PF, et al. Randomised controlledtrial comparing single agent paclitaxel vs epidoxorubicin plus paclitaxel in patientswith advanced ovarian cancer in early progression after platinum-based chemother-apy: an Italian Collaborative Study from the Mario Negri Institute, Milan, G.O.N.O.(Gruppo Oncologico Nord Ovest) group and I.O.R. (Istituto Oncologico Romagnolo)group. Br J Cancer 2004;90(11):21127.

[64] Mutch DG, Orlando M, Goss T, Teneriello MG, Gordon AN, McMeekin SD, et al.Randomized phase III trial of gemcitabine compared with pegylated liposomaldoxorubicin in patients with platinum-resistant ovarian cancer. J Clin Oncol2007;25(19):28118.

[65] Meier W, du Bois A, Reuss A, Kuhn W, Olbricht S, Gropp M, et al. Topotecan versustreosulfan, an alkylating agent, in patients with epithelial ovarian cancer and relapsewithin 12 months following 1st-line platinum/paclitaxel chemotherapy. A prospective-ly randomized phase III trial by the Arbeitsgemeinschaft Gynaekologische OnkologieOvarian Cancer Study Group (AGO-OVAR). Gynecol Oncol 2009;114(2):199205.

[66] Vergote I, Finkler N, del Campo J, Lohr A, Hunter J, Matei D, et al. Phase 3 randomisedstudy of canfosfamide (Telcyta, TLK286) versus pegylated liposomal doxorubicin ortopotecan as third-line therapy in patients with platinum-refractory or -resistantovarian cancer. Eur J Cancer 2009;45(13):232432.

[67] Monk BJ, Herzog TJ, Kaye SB, Krasner CN, Vermorken JB, Muggia FM, et al.Trabectedin plus pegylated liposomal Doxorubicin in recurrent ovarian cancer.J Clin Oncol 2010;28(19):310714.

[68] Colombo N, Kutarska E, Dimopoulos M, Bae DS, Rzepka-Gorska I, Bidzinski M, et al.Randomized, open-label, phase III study comparing patupilone (EPO906) withpegylated liposomal doxorubicin in platinum-refractory or -resistant patients withrecurrent epithelial ovarian, primary fallopian tube, or primary peritoneal cancer.J Clin Oncol 2012;30(31):38417.

[69] Fotopoulou C, Vergote I, Mainwaring P, Bidzinski M, Vermorken JB, Ghamande SA,et al. Weekly AUC2 carboplatin in acquired platinum-resistant ovarian cancer withor without oral phenoxodiol, a sensitizer of platinum cytotoxicity: the phase IIIOVATURE multicenter randomized study. (published online ahead of print Dec 10,2013) Ann Oncol; 2013. http://dx.doi.org/10.1093/annonc/mdt515.

[70] Naumann RW, Coleman R, Burger C, Sausville EA, Kutarska E, Ghamande SA, et al.PRECEDENT: A randomised phase II trial comparing Vintafolide (EC145) andpegylated liposomal doxorubicin (PLD) in combination versus PLD alone in patientswith platinum-resistant ovarian cancer. J Clin Oncol 2013;31(35):44008.

http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0255http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0255http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0255http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0260http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0260http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0260http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0355http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0355http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0355http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0360http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0360http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0360http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0360http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0270http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0270http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0270http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0275http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0275http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0275http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0280http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0280http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0280http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0285http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0285http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0285http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0285http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0290http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0290http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0290http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0295http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0295http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0295http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0295http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0295http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0295http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0300http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0300http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0300http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0300http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0305http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0305http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0305http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0305http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0305http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0310http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0310http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0310http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0310http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0315http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0315http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0315http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0320http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0320http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0320http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0320http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0320http://dx.doi.org/doi:10.1093/annonc/mdt515http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0330http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0330http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0330http://refhub.elsevier.com/S0090-8258(14)00202-9/rf0330

Platinum resistant ovarian cancer: What is it, who to treat and how to measure benefit?IntroductionPlatinum resistant disease: what is it?Relationship between specific histotype and response to chemotherapyThe Role of BRCA MutationsThe BRCAness profileMechanisms of resistanceTumor heterogeneityPlatinum resistant disease: outcomes of therapyWho to treat, how to treat and for how long?Chemotherapy-response assaysHow long to treat and how many lines of therapy?What to measureConclusionConflict of interest statementReferences