oncology

Research www.AJOG.org

ONCOLOGY

A chemoresponse assay for prediction of platinum resistancein primary ovarian cancerThomas C. Krivak, MD; Shashikant Lele, MD; Scott Richard, MD; Angeles Alvarez Secord, MD; Charles A. Leath III, MD;Stacey L. Brower, PhD; Chunqiao Tian, PhD; Richard G. Moore, MD

OBJECTIVE: Recurrence following primary platinum-based chemo- nonresistant (sensitive þ IS) tumors (median PFS: 11.8 vs 16.6
therapy remains a challenge in the treatment of patients withadvanced-stage epithelial ovarian cancer. This study examineswhether a chemoresponse assay can identify patients who areplatinum-resistant prior to treatment.
STUDY DESIGN: Women (n ¼ 276) with International Federation ofGynecology and Obstetrics stage III-IV ovarian, fallopian, and perito-neal cancer were enrolled in an observational study, and theresponsiveness of their tumors was evaluated using a chemoresponseassay. All patients were treated with a platinum/taxane regimenfollowing cytoreductive surgery. Assay responses to carboplatin orpaclitaxel were classified as sensitive, intermediate sensitive (IS), orresistant. Association of assay response with progression-free survival(PFS) was analyzed using the Kaplan-Meier method and a Coxregression model.

RESULTS: Patients whose tumors were resistant to carboplatin were atincreased risk of disease progression compared to those with

From the Division of Gynecologic Oncology, Department of Obstetrics and GCenter (Drs Krivak and Richard), and Department of Product Development, PSurgical Subspecialties, Department of Gynecologic Oncology, Roswell ParkDepartment of Obstetrics and Gynecology, Duke Cancer Institute, Duke UniOncology, Department of Obstetrics and Gynecology, University of Alabama Sof Gynecologic Oncology, Department of Obstetrics and Gynecology, WomKrivak is now affiliated with the Division of Gynecologic Oncology, DepartmentPA. Dr Richard is now affiliated with the Division of Gynecologic Oncology, DPhiladelphia, PA.

Received Nov. 7, 2013; revised Jan. 23, 2014; accepted Feb. 11, 2014.

This study was supported by Precision Therapeutics Inc.

C.A.L. is a member of the speakers bureau for Precision Therapeutics Inc. S.authors report no conflict of interest.

Presented in poster format at the 42nd annual meeting of the Western Asso

Reprints: Becky Yacht, 2516 Jane St., Pittsburgh, PA 15203. byacht@ptilab

0002-9378/$36.00 � ª 2014 Mosby, Inc. All rights reserved. � http://dx.doi.org/10.1

68.e1 American Journal of Obstetrics & Gynecology JULY 2014

months, respectively, P < .001), and the association was confirmedafter adjusting for other clinical factors (hazard ratio, 1.71; 95%confidence interval, 1.12e2.62; P ¼ .013). Association of assayresponse to paclitaxel with PFS trended in multivariate analysis (hazardratio, 1.28; 95% confidence interval, 0.84e1.95; P ¼ .245). Fortumors resistant to carboplatin, 59% were sensitive or IS to at least 1other commonly used agent, demonstrating the ability of the assay toinform treatment decisions beyond the standard platinum/taxaneregimen.

CONCLUSION: Assay resistance to carboplatin is strongly associatedwith shortened PFS among advanced-stage epithelial ovarian cancerpatients treated with carboplatin þ paclitaxel therapy, supporting useof this assay to identify patients likely to experience early recurrence onstandard platinum-based therapy.

Key words: chemoresponse assay, ovarian cancer, platinumresistance

Cite this article as: Krivak TC, Lele S, Richard S, et al. A chemoresponse assay for prediction of platinum resistance in primary ovarian cancer. Am J Obstet Gynecol2014;211:68.e1-8.

n 2013, it was estimated that there
I will be 22,240 new cases of ovariancancer and 14,030 deaths due to thisdisease in the United States; epithelialovarian cancer (EOC) represents theleading cause of death from gynecologicmalignancies.1 The poor prognosis
observed with EOC is largely attributedto late detection of the disease (ie, once ithas already advanced to late stages), aswell as intrinsic drug refractory and/oremerging drug resistance to initialchemotherapy. Evidence from random-ized clinical trials has established the

ynecology, Magee-Womrecision Therapeutics Inc (Cancer Institute, Buffalo

versity Medical Center, Duchool of Medicine at Birm

en and Infants Hospital, Bof Obstetrics andGynecoepartment of Obstetrics a

L.B. and C.T. are employe

ciation of Gynecologic On

s.com.

016/j.ajog.2014.02.009

platinum/taxane combination regimenas standard first-line chemotherapyfor patients with advanced-stage EOC,yielding response rates of 60-70%.2,3

However, most such patients experi-ence relapse within 1-2 years, and only30% live>5 years.4 It is clear that EOC is

ens Hospital of University of Pittsburgh MedicalDrs Brower and Tian), Pittsburgh, PA; Division of, NY (Dr Lele); Division of Gynecology Oncology,rham, NC (Dr Secord); Division of Gynecologicingham, Birmingham, AL (Dr Leath); and Divisionrown University, Providence, RI (Dr Moore). Drlogy,Western Pennsylvania Hospital, Pittsburgh,nd Gynecology, Hahnemann University Hospital,

es of Precision Therapeutics Inc. The remaining

cologists, Seattle, WA, June 26-29, 2013.

http://dx.doi.org/10.1016/j.ajog.2014.02.009

http://www.AJOG.org

http://www.AJOG.org

http://www.AJOG.org

http://www.AJOG.org

mailto:[email protected]

www.AJOG.org Oncology Research

a heterogeneous disease, and a platinum/taxane combination is not the optimalchemotherapy regimen for all patients.

Efforts have been taken to improvetoxicities, response rates, and survivalthrough the use of alternate chemother-apies, the use of different treatmentschedules, or the incorporation of bio-logic agents, with encouraging datarecently reported for the latter 2 ap-proaches.5-7 Over the last 2 decades,multiple clinical studies have attemptedto identify chemotherapy regimens su-perior to platinum/taxane in the first-linetreatment of advanced-stage EOC.3,8-10

Although progression-free survival(PFS) and overall survival (OS) observedin these alternate regimens are no better(and, inmany studies, are noworse) thanthose observed with the platinum/taxanestandard, the alternate regimens may beconsidered to be equivalent in clinicalpractice.

In EOC, clinically useful markers thatidentify platinum-resistant tumors,among the overall high number of che-mosensitive patients, remain a criticalneed. If identified early, platinum-resistant EOC patients could benefitfrom alternate and/or additional thera-peutic options in first-line therapy.Moreover, reliable early identification ofplatinum resistance may allow thedevelopment of clinical trials specificallytargeting this population with novelalternate therapies.

Chemoresponse assays have beeninvestigated as a method for individual-izing chemotherapy treatment decisionsand improving outcomes in cancer pa-tients. Recently, a prospective studydemonstrated that women with persis-tent or recurrent EOC who were treatedwith an assay-sensitive therapy experi-enced significantly improved PFS andOScompared to those treated with assay-resistant therapies.11 To further evaluatethe clinical relevance of this assay in theprimary setting, and in accordance withstandards for the reporting of diagnosticaccuracy criteria,12 an observationalstudy was conducted among womenwith stage III/IV EOC treated bystandard-of-care chemotherapy. Theprimary objective of this study is todetermine whether assay response to

carboplatin or/and paclitaxel is associ-ated with disease progression amongpatients with primary EOC followinginitial treatment with platinum/taxaneregimen. Furthermore, this study willevaluate whether this assay can be used toidentify patients who are resistant toplatinum-based treatment and at highrisk of early progression.

MATERIALS AND METHODS

Study populationParticipants were prospectively enrolledin an observational study of women withgynecologic cancers. Tumor samplesfrom 54 institutions were submittedfor chemoresponse testing from 2006through 2010.Womenwith InternationalFederation of Gynecology and Obstetricsstage III-IV EOC, fallopian tube cancer,and peritoneal cancer treated withcarboplatin/paclitaxel-based chemothe-rapy following initial cytoreductive sur-gery were included in the study. Patientswith a time interval of >2 months be-tween surgery and initiation of chemo-therapy, chemotherapy duration >6months, and/or treatment consisting of>10 cycles of chemotherapy wereexcluded. The institutional review boardat each participating center approved thisstudy, and documented informed con-sent was obtained from all enrolledpatients.

Chemoresponse assayDetails regarding the chemoresponseassay employed in this study (ChemoFx;Precision Therapeutics Inc, Pittsburgh,PA) have been described elsewhere.13

Briefly, the inhibition of tumor growthwasmeasured at different concentrationsof each therapy. The survival fraction oftumor cells at each concentration wascalculated as compared to a control (nodrug). The summation of survival frac-tion values over 7 concentrations wascomputed as the drug response score,which represents the area under the dose-response curve (AUC). A smaller AUCscore indicates greater sensitivity to thetherapy. Chemoresponse is classified into1 of 3 categories according to the AUCscore: sensitive, intermediate sensitive(IS), or resistant. The classification cri-terion was defined based on the

JULY 2014 Ame

distribution of AUC scores among anexternal population of patients with pri-mary EOC. Specifically, the distributionsof AUC scores for carboplatin and pacli-taxel were established based on referentspecimens. Scores ranked at the 25th and75th percentiles were obtained. A tumorwith an AUC score <25th rank wasclassified as sensitive, between 25th-75thrank as IS, and >75th rank as resistant.

Statistical analysisThe primary endpoint of this study wasPFS, calculated from the start of chemo-therapy administration until the date offirst documented disease recurrence,death, or most recent follow-up. Com-monly utilized patient prognostic infor-mation was also collected, including: age,Eastern Cooperative Oncology Groupperformance status, histology, tumorgrade, stage, debulking status, and type ofchemotherapy administered. The physi-cian(s) at each institution reported allclinical information, which was qualitycontrolled according to a comprehensivemonitoring plan. Disease progression wasdetermined by clinical evidence, radio-logical examination, and/or cancerantigen 125. Optimal debulking wasdefined as residual tumor of �1 cm inmaximal dimension at the end of surgeryand was reported by enrolling physicians.PFS based on assay response was esti-mated using the Kaplan-Meier method,and the log rank test was used to comparethe differences among sensitive, IS, andresistant patients. Since the primaryobjective of the current study was toidentify platinum-resistant patients, sen-sitive and IS groups were combined forfurther analyses. The association of theassay and PFSwas also assessed using Coxregression model adjusted for clinicalcovariates (age, performance status [1-3 vs0], histology [high-grade serous vs non-high-grade serous], and stage/debulkingstatus [III-suboptimal/IVvs III-optimal]).The hazard ratio (HR) of disease pro-gression for patients treated with resistantvs nonresistant (sensitive þ IS) therapywas estimated. Subgroup analyses strati-fied by age group, performance status,histology/tumor grade, or stage/debulkingstatus were also conducted.

rican Journal of Obstetrics & Gynecology 68.e2

http://www.AJOG.org

FIGURE 1CONSORT diagram

Research Oncology www.AJOG.org

RESULTS

PatientsA total of 462 patients were enrolled inthis study, with 276 evaluable for inclu-sion in the analysis (Figure 1). Patientcharacteristics are displayed in Table 1.The median age of the study populationwas 61 years, and most patients had tu-mors that were classified as papillary se-rous (84%), poorly differentiated (83%),stage III (85%), and optimally debulked(72%) (Table 1). The majority (94%)completed 4-8 cycles of chemotherapy.The median follow-up period was 23months (range, 12e37months), and 193(70%) patients experienced disease pro-gression within this time frame. Themedian PFS was estimated to be 15.9months (95% confidence interval [CI],14.3e17.1 months).

Flow diagram describing exclusion criteria and their effect on total number of evaluable patients in

study.

C, carboplatin; CONSORT, Consolidated Standards of Reporting Trials; P, paclitaxel; PFS, progression-free survival.

Krivak. Chemoresponse assay for platinum resistance. Am J Obstet Gynecol 2014.

Chemoresponse assay and clinicaloutcomesAssay results for carboplatin were avail-able for 231 patients, with 44 (19.1%)patients identified as resistant to thistherapy in the chemoresponse assay.Assay data for paclitaxel were availablefor 226 patients, 49 (21.7%) of whomwere classified as resistant. Assay resis-tance by age, performance status, his-tology/grade, and stage/debulking statusis summarized in Table 2. There is noevidence that assay result for either car-boplatin or paclitaxel is correlated withpatient characteristics. Assay result forcarboplatin was significantly associatedwith clinical outcome (Figure 2). Themedian PFS was 16.6 and 11.8 monthsfor assay nonresistant (sensitive þ IS)and resistant tumors, respectively. Pa-tients displaying assay resistance to car-boplatin were at a higher risk of diseaseprogression as compared to those whowere nonresistant (HR, 1.87; 95% CI,1.29e2.70; P< .001). These results wereconsistent in multivariate analysis aftercontrolling for clinical covariates (HR,1.71; 95% CI, 1.12e2.62; P ¼ .013)(Table 3). Analysis of subgroups (agegroup, performance status, histology,stage/debulking status) was also con-ducted (Figure 3), and the associationbetween PFS and assay result forcarboplatin was suggested across all


http://www.AJOG.org

TABLE 1Patient characteristics(n [ 276)Characteristic No. patients %

Age, y

Median (range) 61 (30e85)

<50 40 (14.5)

50e59 78 (28.3)

60e69 97 (35.1)

�70 61 (22.1)

ECOG PS

0 144 (52.2)

1 63 (23.8)

2 11 (4.0)

3 2 (0.7)

Unknown 56 (20.3)

Tumor site

Ovarian 229 (83.0)

Peritoneal 32 (11.6)

Fallopian tube 15 (5.4)

Histology

Serous 232 (84.1)

Endometrioid 8 (2.9)

Clear cell 6 (2.2)

Mucinous 3 (1.1)

Others 27 (9.8)

Tumor grade

1 12 (4.4)

2 25 (9.1)

3 228 (82.6)

Unknown 11 (4.0)

Stage

III 234 (84.8)

IV 42 (15.2)

Debulking

Microscopic 64 (23.2)

Optimal 135 (48.9)

Suboptimal 59 (21.4)

Unknown 18 (6.5)

ECOG, Eastern Cooperative Oncology Group; PS, per-formance status.

Krivak. Chemoresponse assay for platinum resis-tance. Am J Obstet Gynecol 2014.

TABLE 2Assay resistance by patient characteristics

Characteristic

Carboplatin Paclitaxel

No.patients

No. resistant,n (%) P value

No.patients

No. resistant,n (%) P value

Age group, y .987 .213

<60 100 19 (19.0) 96 17 (17.1)

�60 131 25 (19.1) 130 32 (24.6)

ECOG PS .636 .526

0 126 24 (19.1) 120 25 (20.8)

�1 59 13 (22.0) 60 15 (25.0)

Tumor grade .342 .438

1-2 32 8 (25.0) 30 5 (16.7)

3 190 34 (17.9) 187 43 (23.0)

HGS .204 .270

Yes 165 28 (17.0) 161 38 (23.6)

No 66 16 (24.2) 65 11 (16.9)

Stage/debulking .145 .136

III-Optimal 143 24 (16.8) 139 27 (19.4)

III-Suboptimal/IV 76 19 (25.0) 74 21 (28.4)

All 231 44 (19.1) 226 49 (21.7)

ECOG, Eastern Cooperative Oncology Group; HGS, high-grade (grade-3) serous; PS, performance status.



subgroups. The data also suggest thatpatients with assay resistance to pacli-taxel would experience shortened PFS,but the association did not reach thelevel of statistical significance (Table 3).Assay results for carboplatin and pacli-

taxel were highly correlated. For 220patients with assay data available for bothagents, 75.5% were nonresistant to bothagents and 15.9% were resistant toboth agents, while only 8.6% of patientswere resistant to only 1 agent (5.9% tocarboplatin and 2.7% to paclitaxel). Pa-tients resistant to both agents experiencedthe worst outcomes (HR, 1.66; 95% CI,1.10e2.52; P ¼ .017, as compared to pa-tients nonresistant to both agents).Multivariate analysis indicated the sametendency, although the association wasnot statistically significant (Table 3).

Pattern of assay responseTo estimate the proportion of patientswho may benefit from assay-informed

JULY 2014 Ame

therapy, the assay responses to 7 clini-cally relevant single agents (carboplatin,cisplatin, gemcitabine, pegylated lipo-somal doxorubicin, paclitaxel, doce-taxel, and topotecan) were analyzed for153 patients who had complete assaydata for all 7 of these agents (Figure 4).Only 7% of the patients displayed assayresistance to all 7 agents, while 5% weresensitive to all 7 agents. Thus, 93% ofthe patients were nonresistant (sensitiveor IS) to at least 1 agent. Specifically,35% were IS to at least 1 agent, and58% were sensitive to at least 1 agent.Of note, 18% of these tumors wereresistant to carboplatin but, of those,59% of them were nonresistant (sensi-tive or IS) to at least 1 other agent inthe chemoresponse assay.

COMMENT

The standard of care for first-line treat-ment of patients with advanced-stageEOC consists of aggressive cytoreductive


http://www.AJOG.org

FIGURE 2Kaplan-Meier survival curves

Progression-free survival (PFS) in advanced-stage epithelial ovarian cancer patients receiving first-

line carboplatin/paclitaxel-based chemotherapy and exhibiting resistant (R) vs nonresistant (sensitive

[S] þ intermediate sensitive [I]) chemoresponse assay results for carboplatin.

CI, confidence interval; HR, hazard ratio.


TABLE 3Univariate and multivariate analyses of factors affecting diseaseprogression-free survival

Assay result

Univariate analysis Multivariate analysis

HR (95% CI) P value HR (95% CI)a P value

Assay for carboplatin

Sensitiveb Referent Referent

Resistant 1.87 (1.29e2.70) < .001 1.71 (1.12e2.62) .013

Assay for paclitaxel

Sensitive Referent Referent

Resistant 1.43 (0.99e2.06) .055 1.28 (0.84e1.95) .245

Assay for carboplatinand paclitaxel

Sensitive to both Referent Referent

Resistant to one 1.42 (0.83e2.43) .206 1.38 (0.76e2.51) .297

Resistant to both 1.66 (1.10e2.52) .017 1.51 (0.94e2.42) .090

CI, confidence interval; HR, hazard ratio.

a HR for disease progression estimated from Cox model adjusted for age, Eastern Cooperative Oncology Group performancestatus, tumor grade/histology, and stage/debulking status; b Sensitive or intermediate sensitive assay result.




surgery followed by platinum/taxane-based chemotherapy14; however, in thistreatment approach, approximately 20-30% of patients will have platinum-resistant disease.15 If identified early,platinum-resistant EOC patients maybenefit from alternate and/or additionaltherapeutic options in first-line therapy.At the time of recurrence, clinicians willclassify patients as being platinum sensi-tive (EOC relapsing>6 months after theend of first-line chemotherapy) or plat-inum resistant (EOC relapsing within 6months after the end of first-linechemotherapy).16,17 This platinum sta-tus classification is the primary covariateused in determining future prognosis andsubsequent treatment strategies. How-ever, as with most clinical covariates, itsaccuracy is not absolute; additionalmeasures of platinum responsivenessmay be beneficial in further personalizingtreatment strategies. Using the currentstandard clinical approach, identificationof platinum-resistant disease is delayeduntil after the patient has already expe-rienced the costs and toxicities associatedwith first-line therapy. Earlier identifica-tion of effective first-line treatment mayimprove the disease course in EOC pa-tients, potentially allowing them todemonstrate response, avoid recurrencefor a longer time, and delay the onset ofdecline in overall health, thereby allowingmore therapies to be given that mayfurther extend OS.

Unfortunately, molecular character-ization of EOC has not yet been able tosubstitute for the clinically observedplatinum status classification. The cur-rent study evaluates the potential utilityof a chemoresponse assay in identifyingplatinum resistance in advanced-stageEOC patients undergoing standardfirst-line treatment. Determining plat-inum status earlier in the treatment ofadvanced-stage EOC may prevent thishigh-risk group of patients from beingexposed to multiple cycles of ineffectivetherapy and allow for more effectivealternate therapeutic options earlier inthe disease, with the ultimate goal ofimproving patient outcomes.

The results reported herein suggestthat the chemoresponse assay is an in-dependent factor in identifying platinum

http://www.AJOG.org

FIGURE 3Subgroup analysis of association between assay results andprogression-free survival

Hazard ratios (HR) (resistant vs nonresistant) estimated for each subgroup of patients stratified by

major clinical covariates.

CI, confidence interval; ECOG, Eastern Cooperative Oncology Group; HGS, high-grade serous; PS, performance status.



resistance in advanced-stage EOC pa-tients treated with standard first-linetherapy (carboplatin/paclitaxel). Pa-tients whose tumors were assay-resistantto carboplatin had an increased risk ofearly disease progression, as compared tothose whose assay results were nonre-sistant for carboplatin, recurring onaverage 5 months sooner. Furthermore,based on the Kaplan-Meier plot of thecurrent study (Figure 2), within 6months of the start of chemotherapy,25% of assay-resistant patients hadalready recurred, while <10% of assay-sensitive (nonresistant) had recurred.Likewise, at 18 months after the start ofchemotherapy, approximately 50% ofassay-sensitive patients had been freeof disease progression, while 80% ofassay-resistant patients had recurred.

Multivariate analysis of assay resultsfor paclitaxel demonstrated a positivetrend, and, further, patients who were

resistant to both agents demonstratedthe worst outcomes, which was signifi-cantly different from patients nonresis-tant to both agents. These results areconsistent with the notion that theplatinum portion of the standardregimen for advanced-stage EOC playsthe larger role in the clinical perfor-mance of that regimen.18,19 As such, it isexpected that assay results for paclitaxelare not as highly correlated with PFS asare those for carboplatin and carbo-platin þ paclitaxel. OS will be includedin future analyses.The ability of this assay to identify

patients likely to be platinum resistantcreates the opportunity to consideralternate treatments regimens for thesepatients earlier in the course of treatment.Alternate treatments may be consideredeither initially following surgery or uponfirst clinical indication of suboptimalperformance during standard first-line

JULY 2014 Ame

treatment. Earlier intervention mayallow for a reduction in toxicitiesincurred by the patient from ineffectivetherapy, as well as a reduction in theoverall costs of treatment.20 Mostimportantly, assay-informed treatmentdecisions may lead to earlier treatmentwith a more effective therapy, therebydelaying recurrence and potentiallylengthening the overall expected survivalduration for these high-risk patients.Identification of advanced-stage EOCpatients as platinum resistant prior totreatment could inform first-line treat-ment decisions in a variety of ways,including substitution of alternate activeagents, alteration of the planned first-linetherapy to a dose-dense approach, or theaddition of novel therapies that mayovercome the resistance observed.5-7,21-23

Results from various completed andongoing studies investigating alternatetreatment strategies to carboplatin þpaclitaxel should be referenced whenconsidering treatment different thancarboplatin þ paclitaxel.3,5-7,8-10,21-23

In addition to identifying platinum-resistant patients, the current study alsodemonstrates the ability of the chemo-response assay to generally identifyagents to which a patient’s tumor issensitive. The results presented hereinshow that >90% of patient tumors weresensitive or IS to at least 1 of the 7 mostcommon agents utilized clinically totreat EOC. More importantly, for thosetumors resistant to carboplatin, >50%of them were identified to be sensitive orIS to at least 1 other agent. These resultsexemplify the ability of the assay toinform treatment decisions beyond thecarboplatin/paclitaxel standard of care.These findings are also consistent withthose from a recent prospective study ofpatients with recurrent EOC whodemonstrate an improvement in bothPFS and OS when treated with an assay-sensitive therapy compared to thosetreated with a nonsensitive agent,11

highlighting the clinical value of thisassay for individualized treatment ofEOC.

In summary, the chemoresponse assayevaluated herein is independently asso-ciated with PFS and may be used to


http://www.AJOG.org

FIGURE 4Distribution of assay results across 7 single-agent treatments

Patients (n¼ 153) were categorized as resistant to all 7 treatments (R7), intermediate or resistant to

all 7 treatments (I/R), or sensitive to anywhere from 1 (S1) to 7 (S7) treatments. Single-agent

treatments were: carboplatin, cisplatin, pegylated liposomal doxorubicin, gemcitabine, paclitaxel,

docetaxel, topotecan.



predict platinum resistance in patientswith advanced-stage EOC prior totreatment. Patients predicted for pooreroutcome (ie, platinum resistance) by theassay (and in conjunction with otherclinical factors) may be considered forinvestigation of alternate treatmentoptions. -

REFERENCES

1. Siegel R, Naishadham D, Jemal A. Cancerstatistics, 2013. CA Cancer J Clin 2013;63:11-30.2. Ozols RF, BundyBN,Greer BE, et al. Phase IIItrial of carboplatin and paclitaxel compared withcisplatin and paclitaxel in patients with optimallyresected stage III ovarian cancer: a gynecologiconcology group study. J Clin Oncol 2003;21:194-200.3. du Bois A, Luck HJ, Meier W, et al.A randomized clinical trial of cisplatin/paclitaxelversuscarboplatin/paclitaxel as first-line treatment

68.e7 American Journal of Obstetrics & Gynecolo

of ovarian cancer. J Natl Cancer Inst 2003;95:1320-9.4. Coleman RL, Monk BJ, Sood AK, Herzog TJ.Latest research and treatment of advanced-stage epithelial ovarian cancer. Nat Rev ClinOncol 2013;10:211-24.5. Katsumata N, Yasuda M, Isonishi S, et al.Long-term results of dose-dense paclitaxel andcarboplatin versus conventional paclitaxel andcarboplatin for treatment of advanced epithelialovarian, fallopian tube or primary peritonealcancer (JGOG 3016): a randomized, con-trolled, open-label trial. Lancet Oncol 2013;14:1020-6.6. Perren TJ, Swart AM, Pfisterer J, et al.A phase 3 trial of bevacizumab in ovarian cancer.N Engl J Med 2011;365:2484-96.7. Burger RA, Brady MF, Bookman MA, et al.Incorporation of bevacizumab in the primarytreatment of ovarian cancer. N Engl J Med2011;365:2473-83.8. Bookman MA, Brady MF, McGuire WP, et al.Evaluation of new platinum-based treatmentregimens in advanced-stage ovarian cancer: aphase III trial of the gynecologic cancer inter-group. J Clin Oncol 2009;27:1419-25.

gy JULY 2014

9. duBoisA,WeberB,RochonJ, etal. Additionofepirubicin as a third drug to carboplatin-paclitaxelin first-line treatment of advanced ovarian cancer:a prospectively randomized gynecologic cancerintergroup trial by the ArbeitsgemeinschaftGynaekologische Onkologie Ovarian CancerStudy Group and the Groupe d’InvestigateursNationaux pour l’Etude des Cancers Ovariens.J Clin Oncol 2006;24:1127-35.10. Pfisterer J, Weber B, Reuss A, et al. Ran-domized phase III trial of topotecan followingcarboplatin and paclitaxel in first-line treatmentof advanced ovarian cancer: a gynecologiccancer intergroup trial of the AGO-OVAR andGINECO. J Natl Cancer Inst 2006;98:1036-45.11. Rutherford T, Orr J Jr, Grendys E Jr, et al.A prospective study evaluating the clinical rele-vance of a chemoresponse assay for treatmentof patients with persistent or recurrent ovariancancer. Gynecol Oncol 2013;131:362-7.12. Bossuyt PM, Reitsma JB, Bruns DE, et al.The STARD statement for reporting studies ofdiagnostic accuracy: explanation and elabora-tion. Clin Chem 2003;49:7-18.13. Brower SL, Fensterer JE, Bush JE. TheChemoFx assay: an ex vivo chemosensitivityand resistance assay for predicting patientresponse to cancer chemotherapy. In: Mor G,Alvero AB, eds. Methods in molecular biology.Vol. 414: apoptosis and cancer. Totowa, NJ:Humana Press Inc; 2008:57-78.14. National Comprehensive Cancer Network.NCCN clinical practice guidelines in oncology.Ovarian cancer: including fallopian tube cancerand primary peritoneal cancer. Version 1.2014.Available at: http://www.nccn.org/professionals/physician_gls/pdf/ovarian.pdf. Accessed March7, 2014.15. Markman M, Bookman MA. Second-linetreatment of ovarian cancer. Oncologist 2000;5:26-35.16. Gore ME, Fryatt I, Wiltshaw E, Dawson T.Treatment of relapsed carcinoma of the ovarywith cisplatin or carboplatin following initialtreatment with these compounds. GynecolOncol 1990;36:207-11.17. Markman M, Hoskins W. Responses tosalvage chemotherapy in ovarian cancer: a crit-ical need for precise definitions of the treatedpopulation. J Clin Oncol 1992;10:513-4.18. Muggia FM, Braly PS, Brady MF, et al.Phase III randomized study of cisplatin versuspaclitaxel versus cisplatin and paclitaxel in pa-tients with suboptimal stage III or IV ovariancancer: a gynecologic oncology group study.J Clin Oncol 2000;18:106-15.19. International Collaborative OvarianNeoplasm Group. Paclitaxel plus carboplatinversus standard chemotherapy with eithersingle-agent carboplatin or cyclophosphamide,doxorubicin, and cisplatin in womenwith ovariancancer: the ICON3 randomized trial. Lancet2002;360:505-15.20. Havrilesky LJ, Krivak TC, Mucensku JW,Myers ER. Impact of a chemoresponse assay on

http://refhub.elsevier.com/S0002-9378(14)00126-4/sref1





































































http://www.nccn.org/professionals/physician_gls/pdf/ovarian.pdf

http://www.nccn.org/professionals/physician_gls/pdf/ovarian.pdf




























http://www.AJOG.org


treatment costs for recurrent ovarian cancer. AmJ Obstet Gynecol 2010;203:160.e1-7.21. Monk BJ, Herzog TJ, Kaye SB, et al. Tra-bectedin plus pegylated liposomal doxorubicinin recurrent ovarian cancer. J Clin Oncol2010;28:3107-14.22. Sehouli J, Stengel D, Oskay-Oezcelik G,et al. Nonplatinum topotecan combinations

versus topotecan alone for recurrent ovariancancer: results of a phase III study of the North-Eastern German Society of GynecologicalOncology Ovarian Cancer Study Group. J ClinOncol 2008;26:3176-82.23. Buda A, Floriani I, Rossi R, et al. Ran-domized controlled trial comparing singleagent paclitaxel vs epidoxorubicin plus

JULY 2014 Ame

paclitaxel in patients with advanced ovariancancer in early progression after platinum-based chemotherapy: an Italian collaborativestudy from the Mario Negri Institute, Milan, G.O.N.O. (Gruppo Oncologico Nord Ovest)group and I.O.R. (Istituto OncologicoRomangnolo) group. Br J Cancer 2004;90:2112-7.


























http://www.AJOG.org

Documents

oncology