Identification of Patients With Ovarian

original reports
Cancer Experiencing the Highest Benefit From
Bevacizumab in the First-Line Setting on the
Basis of Their Tumor-Intrinsic Chemosensitivity
(KELIM): The GOG-0218 Validation Study
Benoit You, MD, PhD1,2; Christopher Purdy, MA3; Larry J. Copeland, MD4; Elizabeth M. Swisher, MD5; Michael A. Bookman, MD6;
Gini Fleming, MD7; Robert Coleman, MD8; Leslie M. Randall, MD9; Krishnansu S. Tewari, MD10; Bradley J. Monk, MD11;
Robert S. Mannel, MD12; Joan L. Walker, MD12; Fabio Cappuccini, MD10; David Cohn, MD4; Mahvish Muzaffar, MD13; David Mutch, MD14;
Andrea Wahner-Hendrickson, MD15; Lainie Martin, MD16,17; Olivier Colomban, MSc1,2; and Robert A. Burger, MD16,17

PURPOSE In patients with high-grade ovarian cancer, predictors of bevacizumab efficacy in first-line setting are
abstract

needed. In the ICON-7 trial, a poor tumor intrinsic chemosensitivity (defined by unfavorable modeled cancer
antigen-125 [CA-125] ELIMination rate constant K [KELIM] score) was a predictive biomarker. Only the patients
with high-risk disease (suboptimally resected stage III, or stage IV) exhibiting unfavorable KELIM score , 1.0
had overall survival (OS) benefit from bevacizumab (median: 29.7 v 20.6 months; hazard ratio [HR], 0.78). An
external validation study in the GOG-0218 trial was performed.
METHODS In GOG-0218, 1,873 patients were treated with carboplatin-paclitaxel 6 concurrent-maintenance
bevacizumab/placebo. Patient KELIM values were calculated with CA-125 kinetics during the first 100 che-
motherapy days by the Lyon University team. The association between KELIM score (favorable $ 1.0, or
unfavorable , 1.0) and bevacizumab benefit for progression-free survival (PFS)/OS was independently
assessed by NGR-GOG using univariate/multivariate analyses.
RESULTS KELIM was assessable in 1,662 patients with $ 3 CA-125 available values. An unfavorable KELIM
score was associated with bevacizumab benefit compared with placebo (PFS: HR, 0.70; 95% CI, 0.59 to 0.82;
OS: HR, 0.87; 95% CI, 0.73 to 1.03), whereas a favorable KELIM was not (PFS: HR, 0.96; 95% CI, 0.79 to 1.17;
OS: HR, 1.11; 95% CI, 0.89 to 1.39). The highest benefit was observed in patients with a high-risk disease
exhibiting unfavorable KELIM, for PFS (median: 9.1 v 5.6 months; HR, 0.64; 95% CI, 0.53 to 0.78), and for OS
(median: 35.1 v 29.1 months; HR, 0.79; 95% CI, 0.65 to 0.97).
CONCLUSION This GOG-0218 trial investigation validates ICON-7 findings about the association between poor
ASSOCIATED tumor chemosensitivity and benefit from concurrent-maintenance bevacizumab, suggesting that bevacizumab
CONTENT may mainly be effective in patients with poorly chemosensitive disease. Bevacizumab may be prioritized in
Data Sharing patients with a high-risk and poorly chemosensitive disease to improve their PFS/OS (patient KELIM score
Statement calculator available on the Biomarker Kinetics website).
Data Supplement
J Clin Oncol 40:3965-3974. © 2022 by American Society of Clinical Oncology
Protocol
Creative Commons Attribution Non-Commercial No Derivatives 4.0 License
Author affiliations
and support
information (if INTRODUCTION standard first-line chemotherapy in patients with an
applicable) appear
In June 2018, the US Food and Drug Administration advanced stage III-IV ovarian cancer.2,3
at the end of this
article. approved bevacizumab for patients with stage III or IV Nevertheless, identifying the patients who should be
Accepted on August epithelial ovarian, fallopian tube, or primary peritoneal prescribed adjuvant bevacizumab treatment is still a
26, 2022 and cancer in combination with carboplatin and paclitaxel, matter of controversy.4,5 Indeed, the final survival
published at followed by single-agent bevacizumab, after initial analysis report of GOG-0218 did not find any benefit
ascopubs.org/journal/ surgical resection.1 This approval followed the out- in overall survival (OS) with bevacizumab in the whole
jco on October 17,
2022: DOI https://doi.
comes of two parallel phase III trials, GOG-0218 and population, but suggested a potential OS gain in
org/10.1200/JCO.22. ICON-7, that demonstrated a benefit in progression- patients with a stage IV disease (median OS: 42.8 v
01207 free survival (PFS) with the addition of bevacizumab to 32.6 months; hazard ratio [HR], 0.75; 95% CI, 0.59

Volume 40, Issue 34 3965

 You et al

CONTEXT
Key Objective
To confirm the predictive value of the tumor primary chemosensitivity (assessed by modeled cancer antigen-125 [CA-125]
ELIMination rate constant K [KELIM]) regarding the efficacy of bevacizumab previously reported in the ICON-7 trial in
patients with advanced ovarian carcinoma.
Knowledge Generated
This external validation analysis with GOG-0218 trial data confirmed that only the patients with poorly chemosensitive
disease (unfavorable KELIM score , 1.0) experienced overall survival benefit among those with a high-risk disease. The
effect of bevacizumab in patients with a low-risk disease characterized by high chemosensitivity (favorable KELIM
score $ 1.0) is uncertain.
Relevance (K.D. Miller)
The results from the ICON-7 and GOG-0218 trials show that primary tumor chemosensitivity predicts the overall survival
benefit of adding bevacizumab; these results informed the design of ongoing validation trials. If confirmed in those
ongoing studies, bevacizumab should be prioritized in patients with a high-risk disease characterized by poor che-
mosensitivity (unfavorable KELIM score) during the first three cycles of neoadjuvant or adjuvant chemotherapy (patient
KELIM score easily calculable on CA-125 Biomarker Kinetics in routine).*

*Relevance section written by JCO Senior Deputy Editor Kathy D. Miller, MD.

to 0.95).6 In the ICON-7 trial, an OS benefit was found in absolute difference, 9.1 months; HR, 0.78; 95% CI, 0.58 to
the predefined high-risk population (including patients 1.04). An external validation of these data was needed.
with stage IV, and patients with stage III disease not The objective of the present project was to perform an
suitable for surgery or having suboptimal debulking sur- external validation study to assess the prognostic and
gery with postoperative residual lesions . 1 cm; median predictive value of KELIM regarding the benefit from
OS: 39.3 v 34.5 months; HR, 0.78; 95% CI, 0.63 to 0.97).7 bevacizumab in terms of PFS and OS in GOG-0218.
This discrepancy suggests that, beyond disease stage and
completeness of surgery, other covariates should be METHODS
identified to select the patients likely to benefit from
bevacizumab. Patients and Methods

Bevacizumab was shown to be active in patients with This study was a retrospective investigation of the GOG-0218
platinum-resistant recurrent ovarian cancer.8 Therefore, trial (ClinicalTrials.gov identifier: NCT00262847). It was an
the impact of the tumor chemosensitivity on the efficacy of international, multicenter, double-blind, placebo-controlled,
bevacizumab has been investigated. Several indicators of phase III trial, in which 1,873 women with incompletely
the tumor intrinsic chemosensitivity have been reported in resected (optimal or suboptimal debulking surgery) stage III
the literature.9 Among them, the modeled cancer antigen- to IV disease were randomly assigned 1:1:1—to arm 1:
125 (CA-125) ELIMination rate constant K (KELIM; on the intravenous carboplatin (area under the curve 6) and pac-
basis of the CA-125 longitudinal kinetics observed during litaxel (175 mg/m2) once every 3 weeks (1 cycle 5 3 weeks)
the first 100 treatment days and calculated using mathe- for 6 cycles; arm 2: same chemotherapy regimen plus
matical modeling) was shown to be a reproducible indicator concurrent bevacizumab (15 mg/kg once every 3 weeks on
of the tumor intrinsic chemosensitivity using data from cycles 2-6); or arm 3: chemotherapy plus concurrent and
more than 12,000 patients enrolled in 12 randomized trials, maintenance bevacizumab (15 mg/kg once every 3 weeks
the Netherlands Cancer Registry, and the Gynecologic on cycles 2-22). The initial reports, where the other inclusion
Cancer InterGroup (GCIG) meta-analysis database.9 and exclusion criteria were described, demonstrated a
benefit in PFS of about 4 months, but the second analysis
The relationship between KELIM and benefit from bev- did not find any benefit in disease-specific survival or OS.2,6
acizumab in the first-line setting was assessed in a post hoc
analysis of the ICON-7 trial. Patients with a highly chemo- Mathematical Modeling of Longitudinal CA-125 Kinetics
sensitive tumor, defined by a favorable KELIM score ($ 1.0), and Estimation of Patient Standardized KELIM by the
had no benefit from bevacizumab, whether they had a low- Lyon University Team (France)
risk or high-risk disease. However, a benefit in OS was found The trial design planned that CA-125 values would be
in patients with a high-risk disease exhibiting an unfavorable locally measured at every cycle. At least three available
KELIM score , 1.0 (median OS: 29.7 v 20.6 months; CA-125 values during the first 100 days of treatment are

3966 © 2022 by American Society of Clinical Oncology Volume 40, Issue 34

 KELIM and Bevacizumab in GOG-0218

required to ensure an accurate assessment of KELIM by Cox hazard-ratio regression analyses and chi-square tests
the model. The mathematical modeling of the early CA- were performed to assess the interactions between KELIM
125 kinetics with a nonlinear mixed-effect model was and treatment arms regarding PFS and OS benefit from
described previously.10-12 To normalize the distribution of bevacizumab in the whole population, and in patients with a
CA-125 concentrations, and to eliminate right-skewness high-risk disease.
in this distribution, CA-125 levels were log-transformed. All survival analyses were implemented with a landmark
Basic details about the semimechanistic kinetic- time point set at 100 days after the start of neoadjuvant
pharmacodynamic model adjustment, and qualification, chemotherapy or at the surgery date, whichever occurred
are presented in the Data Supplement (online only).13 first. As CA-125 was modeled from day 0 to 100, the pa-
In this study, individual KELIM values were estimated with tients who progressed during the first 100 days were ex-
the model implemented in the online calculator.14 Indi- cluded to avoid bias related to the links between early
vidual KELIM values were computed using empirical Bayes progression and CA-125 kinetics, or radiologic tumor re-
estimates. As assessed in previous studies,15,16 KELIM was sponses.17 The median follow-up was computed using
standardized by the prespecified optimized cutoff in an reverse Kaplan-Meier method.
adjuvant setting (cutoff, 0.07/days; prespecified in all
Statistics and Computing Process
studies, including the initial ICON-7 trial), as a way of
providing an easy reading of patient KELIM outcome, with All tests were implemented using a two-sided 0.05 alpha
the following equation: Standardized (std) KELIM 5 KELIM risk. NONMEM 7.5 (ICON Development Solutions, Ellicott
estimated by the model/cutoff. As a consequence, std City, MD) software program was used to fit the semi-
KELIM was a continuous covariate centered on 1.0. To help mechanistic model to CA-125 kinetic data.18 XPOSE4 pro-
the interpretation of KELIM for prognostic analyses, KELIM gram was used for graphical evaluation of model fits.19 The
was dichotomized into a KELIM score: std KELIM , 1.0 was Kaplan-Meier analyses (LIFETEST) and Cox regression an-
considered as unfavorable, whereas std KELIM $ 1.0 was alyses (PHREG) were performed using SAS 9.4 (Cary, NC).20
considered as favorable.
The data set with patient KELIM scores (favorable or un- RESULTS
favorable) was sent to the NRG-GOG team to ensure an Patient Characteristics
independent assessment of KELIM score prognostic and Out of 1,873 enrolled patients, 1,662 patients with $ 3 CA-
predictive value. 125 available values during the first 100 days of treatment
Assessment of the Prognostic and Predictive Values of were assessable for KELIM. Among them, 1,644 and 1,657
KELIM Regarding PFS and OS patients were assessable for KELIM prognostic/predictive
The main objective was to confirm a benefit in OS with value regarding PFS and OS, respectively (87.8% and
bevacizumab addition in patients with a high-risk disease 88.4%, respectively; Data Supplement). The characteris-
characterized by an unfavorable KELIM, as suggested in tics of the 1,657 patients assessable for OS are presented in
the ICON-7 trial. Table 1. Among them, 575 (34.7%) had stage III disease
operated with optimal debulking surgery (postoperative
The prognostic value of KELIM score for PFS and OS was residual lesions # 1 cm); 660 (39.8%) had stage III disease
assessed using univariate and multivariate tests (C-index, operated with suboptimal debulking surgery (postoperative
Kaplan-Meier method, log-rank, and Cox model). In ad- residual lesions . 1 cm); and 422 patients (25.5%) had
dition to KELIM score, the other prognostic factors assessed stage IV disease.
in univariate and multivariate analyses were disease stage
and quality of debulking surgery (stage III operated with Model Qualification
optimal surgery with postoperative residual lesions # 1 cm; Typical parameter estimates, along with the qualification
stage III operated with suboptimal surgery with postoper- analyses from the final semimechanistic models, are pre-
ative residual lesions . 1 cm, or stage IV); pathologic sented in the Data Supplement.
subtype (serous 1 endometrioid, v others); histologic grade
The median value of KELIM was 0.063 days–1 (95% CI,
(grade 1, 2, or 3); ascites (yes or no); and treatment arm
0.061 to 0.065). Std KELIM was not different across
(arm 1, arm 2, or arm 3). The final C-index and Cox survival
treatment arms, similar to what has been observed in
models were obtained using backward selections.
previous studies (Data Supplement).
Moreover, the predictive value of KELIM regarding the
benefit from arm 3 (bevacizumab-concurrent-maintenance Prognostic Value of the KELIM Score Regarding PFS
group) compared with arm 1 (placebo group) was assessed and OS
in the whole population, and then in the population of The median follow-up was 98.1 months for PFS (95% CI,
patients with a high-risk disease (stage IV, and stage III 94.9 to 102.9), and 100.0 months for OS (95% CI, 97.5 to
disease operated with suboptimal debulking surgery) using 103.1). The results of the univariate C-index and log-rank
Kaplan-Meier method and log-rank tests. tests for PFS and OS are presented in the Data Supplement.

Journal of Clinical Oncology 3967

3968 © 2022 by American Society of Clinical Oncology

TABLE 1. Characteristics of the Patients Assessable for Overall Survival (N 5 1,657 patients)
All (N 5 1,657), Placebo-Arm 1 Bevacizumab-Arm 2 Bevacizumab-Arm 3
Variable Level No. (%)a,b,c (n 5 548), No. (%) (n 5 560), No. (%) (n 5 549), No. (%)
Treatment arm Arm 1: control (CT 1 P → P) 548 (33.1)
Arm 2: bevacizumab—concurrent (CT 1 B → P) 560 (33.8)
Arm 3: bevacizumab—concurrent and maintenance (CT 1 B → B) 549 (33.1)
Disease stage and debulking status III (optimal, # 1 cm) 575 (34.7) 194 (35.4) 189 (33.7) 192 (34.9)
III (suboptimal, . 1 cm) 660 (39.8) 221 (40.3) 227 (40.5) 212 (38.6)
IV 422 (25.5) 133 (24.3) 144 (25.7) 145 (26.4)
Ascites No ascites 411 (24.8) 134 (24.5) 135 (24.1) 142 (25.9)
Ascites 1,203 (72.6) 403 (73.5) 406 (72.5) 394 (71.8)
Missing 43 (2.6) 11 (2.1) 19 (3.4) 13 (2.4)
Tumor histology Serous or endometrioid 1,470 (88.7) 499 (91.1) 482 (86.1) 489 (89.1)

You et al
Other histology 187 (11.3) 49 (8.9) 78 (13.9) 60 (10.9)
Histologic grade Well differentiated (grade 1) 70 (4.2) 31 (5.7) 24 (4.3) 15 (2.7)
Moderately differentiated (grade 2) 264 (15.9) 93 (16.9) 80 (14.3) 91 (16.6)
Poorly differentiated (grade 3) 1,201 (72.5) 395 (72.1) 406 (72.5) 400 (72.9)
Missing 122 (7.4) 29 (5.3) 50 (8.9) 43 (7.8)
KELIM score KELIM score unfavorable , 1.0 962 (58.1) 330 (60.2) 324 (57.9) 308 (56.1)
KELIM score favorable $ 1.0 695 (41.9) 218 (39.8) 236 (42.1) 241 (43.9)

Abbreviations: →, followed by; B, bevacizumab; CT, chemotherapy; KELIM, ELIMination rate constant K; P, placebo; PFS, progression-free survival.
a
Patients in at least one of the data sets (PFS_Landmark or OS_Landmark).
b
Landmark PFS data set: n 5 1,644.
c
Landmark OS data set: N 5 1,657.
Volume 40, Issue 34
 KELIM and Bevacizumab in GOG-0218

TABLE 2. Results of the Multivariate Analysis Regarding Overall Survival (N 5 1,657 patients)
Covariates Level No. HR 95% CI P Type III Tests C-Index (95% CI)
KELIM score KELIM score unfavorable , 1.0 868 Ref Ref Ref , .001 0.65 (0.61 to 0.70)
KELIM score favorable $ 1.0 629 0.55 0.48 to 0.62 , .001
Histologic grade Well differentiated (grade 1) 68 Ref Ref Ref 0.023
Moderately differentiated (grade 2) 261 1.53 1.10 to 2.12 .012
Poorly differentiated (grade 3) 1,168 1.54 1.13 to 2.09 .006
FIGO stage/debulking status III (optimal, # 1 cm) 526 Ref Ref Ref 0.002
III (suboptimal, . 1 cm) 591 1.23 1.08 to 1.41 .003
IV 380 1.28 1.10 to 1.49 .002
Ascites No ascites 376 Ref Ref Ref , .001
Ascites 1,121 1.30 1.14 to 1.49 , .001

NOTE. Subjects with missing values for any of the independent variables were not used in the estimation of parameters.
Abbreviations: C-Index, concordance index; FIGO, International Federation of Gynecology and Obstetrics; HR, hazard ratio; ref, reference; KELIM,
ELIMination rate constant K.

In the univariate analyses for both PFS and OS, KELIM KELIM score (favorable $ 1.0, v unfavorable , 1.0; PFS:
score (favorable $ 1.0, v unfavorable , 1.0) was a sig- HR, 0.51; 95% CI, 0.46 to 0.57; OS: HR, 0.55; 95% CI, 0.48
nificant factor (Data Supplement), as were the disease to 0.62), disease stage and quality of debulking surgery,
stage and quality of debulking surgery (stage III operated histologic grade, and ascites (Table 2; Data Supplement).
with optimal surgery, v stage III operated with suboptimal Benefit From Bevacizumab According to KELIM Score
surgery), ascites (yes v no), and histologic grade (grade 1, and Disease Risk Groups
2, or 3), in addition to treatment arm for PFS only (arm 1,
In the whole assessable population, there was a PFS and OS
arm 2, or arm 3).
improvement with bevacizumab-concurrent-maintenance
In the final multivariate models of both PFS and OS, the arm 3 compared with the placebo arm 1 for patients with
following covariates were significant and independent: an unfavorable KELIM score (median PFS: 9.8 v 6.1 months;

KEL unfavorable (placebo) KEL unfavorable (bevacizumab)

KEL favorable (placebo) KEL favorable (bevacizumab)

1.0
+ Censored

0.8 Group Median Log-Rank HR (95% CI)

KEL unfavorable (placebo) 6.1 < .001 Ref
PFS (probability)

KEL unfavorable (bevacizumab) 9.8 0.70 (0.59 to 0.82)

KEL favorable (placebo) 13.6 0.679 Ref
0.6 KEL favorable (bevacizumab) 17.6 0.96 (0.79 to 1.17)
Total n = 1,088

0.4

0.2

0 12 24 36 48 60 72 84 96 108 120 132 144

Time (months)
No. at risk:
KEL unfavorable (placebo) 327 18 12 1 0
KEL unfavorable (bevacizumab) 304 20 11 5 0
KEL favorable (placebo) 218 54 32 10 0
KEL favorable (bevacizumab) 239 48 30 7 0

FIG 1. Kaplan-Meier curves of the PFS of patients according to treatment arm (arm 3 with bevacizumab concurrent-
maintenance, v arm 1 with placebo) in patients with unfavorable or favorable KELIM (KEL) score, in the whole population. HR,
hazard ratio; KELIM, ELIMination rate constant K; mPFS, median PFS (months); PFS, progression-free survival; Ref, reference.

Journal of Clinical Oncology 3969

 You et al

HR, 0.70; 95% CI, 0.59 to 0.82; median OS: 36.3 v significant for OS in the high-risk disease group (P 5 .1),
31.8 months; HR, 0.87; 95% CI, 0.73 to 1.03), which was potentially as the consequence of the limited statistical power
not observed in patients with a favorable KELIM (median related to the smaller number of patients, and the more
PFS: 17.6 v 13.6 months; HR, 0.96; 95% CI, 0.79 to 1.17; modest effects of bevacizumab on OS compared with PFS.
median OS; 55.7 v 56.7 months; HR, 1.11; 95% CI, 0.89 to
1.39; Figs 1 and 2). DISCUSSION
The maximum benefit from bevacizumab-concurrent- Many studies meant to find reproducible predictors of bev-
maintenance arm 3 compared with the placebo arm 1 acizumab activity in patients with ovarian carcinoma treated in
was observed in patients with a high-risk disease (stage IV, or first-line setting have been reported in the literature.4,5 These
stage III operated with suboptimal surgery) exhibiting an attempts highlight the need for biomarkers to identify which
unfavorable KELIM score (median PFS: 9.1 v 5.6 months; patients will benefit from this drug.4,5
HR, 0.64; 95% CI, 0.53 to 0.78; median OS: 35.1 v 29.1 After the publication by Oza et al7 showing an OS gain in
months; HR, 0.79; 95% CI, 0.65 to 0.97). However, as patients with a high-risk cancer (suboptimally resected stage
observed on Kaplan-Meier curves (Figs 3 and 4), the patients III, and stage IV disease) in the ICON-7 trial, and those by
with a high-risk and highly chemosensitive disease (favor- Tewari et al6 suggesting an OS benefit in patients with a stage
able KELIM score) did not experience benefit from bev- IV disease in GOG-0218, the most widely adopted parameter
acizumab (median PFS: 16.3 v 11.7 months; HR, 0.88; 95% for bevacizumab prescription in routine has been high
CI, 0.68 to 1.13; median OS: 59.6 v 49.4 months; HR, 1.05; disease bulk. However, the inconsistency between the two
95% CI, 0.79 to 1.39). studies suggested additional biomarkers of bevacizumab
By contrast, bevacizumab was associated with a potential efficacy were needed for improving the selection of patients.
nonsignificant deleterious effect in terms of OS in patients The AURELIA trial demonstrated that bevacizumab was
with a low-risk exhibiting favorable KELIM score (median active in women with platinum-resistant recurrent ovarian
OS: 51.3 v 68.6 months; HR, 1.18; 95% CI, 0.83 to 1.69; cancer,8 and supported the concept of an association be-
Data Supplement). tween bevacizumab efficacy and tumor chemosensitivity.
The interaction tests between KELIM and treatment arms were The investigation of the predictive value of KELIM in the
significant for PFS (in the whole population, P 5 .01; high-risk ICON-7 trial suggested that only the patients with poorly
disease group, P 5 .03) and OS in the whole population chemosensitive disease, characterized by an unfavorable
(P 5 .08; P 5 .04 by KELIM strates). However, it was not KELIM score, experienced a survival benefit (9 months) from

KEL unfavorable (placebo) KEL unfavorable (bevacizumab)

KEL favorable (placebo) KEL favorable (bevacizumab)

1.0 + Censored
Group Median Log-Rank HR (95% CI)
KEL unfavorable (placebo) 31.8 0.102 Ref
0.8 KEL unfavorable (bevacizumab) 36.3 0.87 (0.73 to 1.03)
OS (probability)

KEL favorable (placebo) 56.7 0.332 Ref

KEL favorable (bevacizumab) 55.7 1.11 (0.89 to 1.39)
0.6 Total n = 1,097

0.4

0.2

0 12 24 36 48 60 72 84 96 108 120 132 144

Time (months)
No. at risk:
KEL unfavorable (placebo) 330 137 54 7 0
KEL unfavorable (bevacizumab) 308 150 55 8 0
KEL favorable (placebo) 218 156 88 20 0
KEL favorable (bevacizumab) 241 168 88 20 0

FIG 2. Kaplan-Meier curves of the OS of patients according to treatment arm (arm 3 with bevacizumab concurrent-
maintenance, v arm 1 with placebo) in patients with unfavorable or favorable KELIM (KEL) score, in the whole
population. HR, hazard ratio; KELIM, ELIMination rate constant K; mPFS, median PFS (months); PFS, progression-free
survival; Ref, reference.

3970 © 2022 by American Society of Clinical Oncology Volume 40, Issue 34

 KELIM and Bevacizumab in GOG-0218

KEL unfavorable (placebo) KEL unfavorable (bevacizumab)

KEL favorable (placebo) KEL favorable (bevacizumab)

1.0 + Censored
Group Median Log-Rank HR (95% CI)
KEL unfavorable (placebo) 5.6 < .001 Ref
0.8
KEL unfavorable (bevacizumab) 9.1 0.64 (0.53 to 0.78)

PFS (probability)
KEL favorable (placebo) 11.7 0.312 Ref

0.6 KEL favorable (bevacizumab) 16.3 0.88 (0.68 to 1.13)

Total n = 705

0.4

0.2

0 12 24 36 48 60 72 84 96 108 120 132 144

Time (months)
No. at risk:
KEL unfavorable (placebo) 230 6 4 1 0
KEL unfavorable (bevacizumab) 203 12 6 3 0
KEL favorable (placebo) 122 23 13 7 0
KEL favorable (bevacizumab) 150 26 17 4 0

FIG 3. Kaplan-Meier curves of the PFS of patients according to treatment arm (arm 3 with bevacizumab concurrent-
maintenance, v arm 1 with placebo) in patients with favorable or unfavorable KELIM (KEL) score, in the population of
patients with a high-risk disease (stage IV 1 stage III operated with suboptimal surgery). HR, hazard ratio; KELIM,
ELIMination rate constant K; mPFS, median PFS (months); PFS, progression-free survival; Ref, reference.

KEL unfavorable (placebo) KEL unfavorable (bevacizumab)

KEL favorable (placebo) KEL favorable (bevacizumab)

1.0 + Censored
Group Median Log-Rank HR (95% CI)
KEL unfavorable (placebo) 29.1 0.023 Ref
0.8 KEL unfavorable (bevacizumab) 35.1 0.79 (0.65 to 0.97)
KEL favorable (placebo) 49.4 0.748 Ref
OS (probability)

KEL favorable (bevacizumab) 59.6 1.05 (0.79 to 1.39)

0.6 Total n = 711

0.4

0.2

0 12 24 36 48 60 72 84 96 108 120 132 144

Time (months)
No. at risk:
KEL unfavorable (placebo) 232 87 31 4 0
KEL unfavorable (bevacizumab) 206 98 36 5 0
KEL favorable (placebo) 122 85 45 13 0
KEL favorable (bevacizumab) 151 106 58 16 0

FIG 4. Kaplan-Meier curves of the OS of patients according to treatment arm (arm 3 with bevacizumab concurrent-
maintenance, v arm 1 with placebo) in patients with favorable or unfavorable KELIM (KEL) score, in the population of
patients with a high-risk disease (stage IV 1 stage III operated with suboptimal surgery). HR, hazard ratio; KELIM,
ELIMination rate constant K; mPFS, median PFS (months); OS, overall survival; PFS, progression-free survival; Ref,
reference.

Journal of Clinical Oncology 3971

 You et al

bevacizumab among those with a high-risk disease.21 This interactions (between treatment and KELIM score) between
finding needs to be verified. the initial model and the stratified models were significant, the
results about the initial model were considered reliable. The
The current external validation study on the GOG-2018 trial
nonsignificativity of the interaction tests between KELIM score
data was performed in collaboration with the US NRG-GOG
and treatment arms regarding OS in the patients with high-risk
group. The statistical analyses, independently performed by
disease might not be related to a lack of discriminative power
the NGR-GOG statistician team, confirm the initial ICON-7
of KELIM score, but rather to the reduced number of patients,
trial findings. The tumor-intrinsic chemosensitivity assessed
and the lower impact of bevacizumab on OS compared with
by the KELIM score exhibits predictive value for benefit from
PFS. Moreover, GOG-0218 was conducted before the
the addition of bevacizumab in the first-line setting. An
emergence of poly(ADP-ribose) polymerase (PARP) inhibi-
unfavorable KELIM score , 1.0 is associated with benefit
tors, which have changed the prognosis and the standard
from bevacizumab, whereas a favorable KELIM score $ 1.0
management of patients with ovarian cancer. As a conse-
is related to a lack of benefit from bevacizumab.
quence, these results may not be applicable to the current
Consistent with the ICON-7 trial results, the maximum benefit patients who are frequently treated with PARP inhibitors,
from bevacizumab was observed in patients with a high-risk especially when bevacizumab is given concurrently with
disease associated with unfavorable KELIM score. The olaparib as done in the PAOLA-1 trial.4,25,26 The prognostic
superimposability of the OS curves according to treatment arm value of BRCA mutational status was not assessed in this
and KELIM scores in ICON-7 and GOG-0218 reconciles the study, but this covariate was already investigated by Tewari
data from these two large front-line phase III trials of bev- et al, and was not found to be associated with the benefit from
acizumab, and strongly suggests that KELIM optimizes the bevacizumab.6,27 Integrating this parameter in this study
identification of patients benefiting from this drug (Data would have led to very small subgroups with limited statistical
Supplement). Bevacizumab should be prioritized in patients power. However, a post hoc analysis of the VELIA trial, where
with a high-risk and poorly chemosensitive disease. By patients were treated with carboplatin-paclitaxel 6 concurrent
contrast, it may be avoided in patients with a low-risk disease and maintenance veliparib, suggested that KELIM could help
and favorable KELIM score, as the effect on OS is unclear select the patients with highly chemosensitive disease (fa-
(Data Supplement). vorable KELIM score), who would have maximum benefit from
These data are also consistent with a recent investigation of this PARP inhibitor.28
the predictive value of KELIM score in the ICON-8 trial, The objective of the present validation study on the GOG-
suggesting that the patients belonging to a poor prognostic 0218 trial was to externally confirm the predictive value of
group (characterized by an incomplete debulking surgery the KELIM score regarding the benefit from bevacizumab in
and an unfavorable KELIM score) were those who had the the first-line setting, as initially reported in the ICON-7 trial.
maximum benefit from the weekly dose-dense chemo- Bevacizumab should be prioritized in patients with a high-
therapy compared with the standard three-weekly regi- risk and poorly chemosensitive disease to improve their
men.22 Several preclinical studies have suggested that PFS and OS. Since patient KELIM score can easily be
metronomic taxane administration is associated with im- calculated on the basis of a minimum of three CA-125
proved drug delivery and antiangiogenic effects.23,24 values observed during the first three cycles of chemo-
The present analysis has several limitations. First, this is a post therapy using the online calculator (Biomarker Kinetics29
hoc analysis of the GOG-0218 trial, and the predictive value of for patients treated with neoadjuvant chemotherapy, and
KELIM was not prospectively assessed in this trial. The pre- Biomarker Kinetics14 for patients treated with adjuvant
specified KELIM score identified 58% of patients with an chemotherapy), this parameter could be used routinely for
unfavorable KELIM score and 42% of patients with a favorable adjusting the disease management.
KELIM score, which is slightly different from the distribution of A prospective validation of KELIM score utility is warranted.
the ICON-7 trial (42% and 58%, respectively). In both trials, The predictive value of KELIM score is being prospectively
the percentages of patients with unfavorable KELIM score assessed in the ongoing large international phase III trial
were higher among those with high-risk disease (63% of NIRVANA-1, which compares the efficacy of niraparib with/
patients in the GOG-0218 trial, and 53% of patients in the without bevacizumab in patients who have undergone
ICON-7 trial). As per inclusion criteria, a higher proportion of complete primary debulking surgery (ClinicalTrials.gov
patients with a high-risk disease were enrolled in the GOG- identifier: NCT05183984). Moreover, on the basis of the
0218 trial compared with the ICON-7 trial, which may explain consistent outcomes from ICON-7, ICON-8, and GOG-0218
this difference. The proportionality of hazard assumption was trials, a prospective trial will be conducted soon. It will
confirmed for Figure 4 (OS in patients with a high-risk dis- assess the efficacy of a salvage weekly chemotherapy
ease), but failed for Figures 1-3. To address this issue, the combined with bevacizumab and an innovative chemo-
interaction term was computed using the parameter estimates resistance reverser in patients with unfavorable KELIM
and the standard errors from the models stratified by KELIM. score during neoadjuvant chemotherapy and incomplete
Since the hazard-ratio results and the P values for the interim debulking surgery.

3972 © 2022 by American Society of Clinical Oncology Volume 40, Issue 34

 KELIM and Bevacizumab in GOG-0218

AFFILIATIONS AUTHOR CONTRIBUTIONS

1
Université Lyon, Université Claude Bernard Lyon 1, Faculté de Conception and design: Benoit You, Larry J. Copeland, Elizabeth M.
Médecine Lyon-Sud, EMR UCBL/HCL 3738, Lyon, France GINECO, Swisher, Robert Coleman, Bradley J. Monk, Joan L. Walker, Fabio
Paris, France Cappuccini, Olivier Colomban
2
Medical Oncology, Institut de Cancérologie des Hospices Civils de Lyon Administrative support: David Mutch
(IC-HCL), CITOHL, GINECO-GINEGEPS, Lyon, France Provision of study materials or patients: Larry J. Copeland, Michael A.
3
Clinical Trial Development Division, Biostatistics and Bioinformatics Bookman, Leslie M. Randall, Bradley J. Monk, Robert S. Mannel, Joan L.
Department, Roswell Park Comprehensive Cancer Center, Buffalo, NY Walker, David Cohn, David Mutch
4
The Ohio State University, James Cancer Hospital, Columbus, OH Collection and assembly of data: Benoit You, Larry J. Copeland, Robert
5
Division of Gynecologic Oncology, Department of Ob/Gyn, University of Coleman, Leslie M. Randall, Krishnansu S. Tewari, Bradley J. Monk,
Washington, Seattle, WA Robert S. Mannel, Joan L. Walker, Mahvish Muzaffar, David Mutch,
6
Director, Gynecologic Oncology Therapeutics, Kaiser Permanente Andrea Wahner-Hendrickson, Lainie Martin
Northern California, San Francisco, CA Data analysis and interpretation: Benoit You, Christopher Purdy, Elizabeth
7
Hematology and Oncology, The University of Chicago Medicine, M. Swisher, Michael A. Bookman, Gini Fleming, Robert Coleman, Leslie
Chicago, IL M. Randall, Krishnansu S. Tewari, Bradley J. Monk, David Cohn, David
8
Chief Scientific Officer, US Oncology Research, The Woodlands, TX Mutch, Lainie Martin, Olivier Colomban, Robert A. Burger
9
Division of Gynecologic Oncology, Virginia Commonwealth University, Manuscript writing: All authors
School of Medicine, Richmond, VA Final approval of manuscript: All authors
10
UC Irvine Medical Center, Orange, CA Accountable for all aspects of the work: All authors
11
HonorHealth Research Institute, University of Arizona, Creighton
University, Phoenix, AZ
12
ACKNOWLEDGMENT
The University of Oklahoma, Oklahoma City, OK
13 The authors thank all the patients and their families, the investigators,
Internal Medicine, East Carolina University, Greenville, NC
14 study nurses, pharmacists, pathologists, and all study teams.
Washington University School of Medicine, Siteman Cancer Center, St
The following NRG Oncology/Gynecologic Oncology Group member
Louis, MO
15 institutions participated in the primary treatment studies: University of
Division of Medical Oncology, Mayo Clinic, Rochester, MN
16 Oklahoma Health Sciences Center, Washington University School of
Abramson Cancer Center at the University of Pennsylvania,
Medicine, Abington Memorial Hospital—Asplundh Cancer Pavilion,
Philadelphia, PA
17 University of Alabama at Birmingham, Women’s Cancer Center of
Mersana Therapeutics, Cambridge, MA
Nevada, Memorial Sloan-Kettering Cancer Center, University of
Kentucky, Sudarshan K Sharma MD Limited—Gynecologic Oncology,
CORRESPONDING AUTHOR Yale University, Metro-Minnesota CCOP, Roswell Park Comprehensive
Benoit You, MD, PhD, Centre Hospitalier Lyon-Sud, Centre Cancer Center, Women and Infants Hospital, Mount Sinai School of
d’Investigation de Thérapeutiques en Oncologie et Hématologie de Lyon Medicine, Northwestern University, Morristown Medical Center,
(CITOHL)/Lyon Investigational Center for Treatments in Oncology and Abramson Cancer Center of the University of Pennsylvania, Duke
Hematology, Service d’oncologie médicale, Chemin du Grand Revoyet, University Medical Center, University of Hawaii, Center of Hope at
69495 Pierre-Benite, France; Twitter: @benoityoulyon; e-mail: benoit. Renown Medical Center, Washington University School of Medicine,
[email protected]. Fox Chase Cancer Center, The Hospital of Central Connecticut, Ohio
State University Comprehensive Cancer Center, University of California
Medical Center At Irvine-Orange Campus, Mayo Clinic, Walter Reed
PRIOR PRESENTATION
National Military Medical Center, Saitama Medical University
Presented in part at the ASCO Annual Meeting, Chicago, IL, June 3-7,
International Medical Center, Gynecologic Oncology Network/Brody
2022.
School of Medicine, Rush University Medical Center, University of Iowa
Hospitals and Clinics, Fred Hutchinson Cancer Research Center,
SUPPORT University of Chicago, Cleveland Clinic Foundation, University of
Supported by Université Claude Bernard Lyon 1 (France), the employer of Mississippi Medical Center, University of North Carolina at Chapel Hill,
Olivier Colomban. Additionally, this project was supported by the Cooper Hospital University Medical Center, Indiana University
following grants received from the National Cancer Institute: Hospital/Melvin and Bren Simon Cancer Center, Seoul National
U10CA180822 (NRG Oncology SDMC) and U10CA180868 (NRG University Hospital, University of Colorado Cancer Center - Anschutz
Oncology Operations). Cancer Pavilion, University of California at Los Angeles Health System,
Wake Forest University Health Sciences, University of New Mexico,
Stony Brook University Medical Center, University of Virginia, Case
CLINICAL TRIAL INFORMATION
Western Reserve University, Fletcher Allen Health Care, Georgia Center
NCT00262847
for Oncology Research and Education (CORE), Cancer Research for the
Ozarks NCORP, Wayne State University/Karmanos Cancer Institute,
AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF University of Minnesota Medical Center-Fairview, Northern Indiana
INTEREST Cancer Research Consortium, Tufts-New England Medical Center,
Disclosures provided by the authors are available with this article at DOI University of Pittsburgh Cancer Institute (UPCI), State University of
https://doi.org/10.1200/JCO.22.01207. New York Downstate Medical Center, MD Anderson Cancer Center,
Moffitt Cancer Center and Research Institute, University of Wisconsin
Hospital and Clinics, University of Texas—Galveston, Gynecologic
DATA SHARING STATEMENT Oncology of West Michigan PLLC, Carle Cancer Center, Cancer
A data sharing statement provided by the authors is available with this Research Consortium of West Michigan NCORP, Central Illinois CCOP,
article at DOI https://doi.org/10.1200/JCO.22.01207. Virginia Commonwealth University, Saint Vincent Hospital, Penn State
Milton S Hershey Medical Center, New York University Medical Center,
Michigan Cancer Research Consortium Community Clinical Oncology
Program, Northern New Jersey CCOP, University of Cincinnati,

Journal of Clinical Oncology 3973

 You et al

Memorial Sloan Kettering Cancer Center, University of Massachusetts Oncology Research Program, Missouri Valley Cancer Consortium CCOP,
Memorial Health Care, Aurora Women’s Pavilion of Aurora West Allis Delaware/Christiana Care CCOP, William Beaumont Hospital, Saint
Medical Center, Kansas City CCOP, Wisconsin NCI Community Louis-Cape Girardeau CCOP, and Wichita CCOP.

REFERENCES
1. US Food and Drug Administration: FDA approves bevacizumab in combination with chemotherapy for ovarian cancer. 2018. https://www.fda.gov/drugs/
resources-information-approved-drugs/fda-approves-bevacizumab-combination-chemotherapy-ovarian-cancer
2. Burger RA, Brady MF, Bookman MA, et al: Incorporation of bevacizumab in the primary treatment of ovarian cancer. N Engl J Med 365:2473-2483, 2011
3. Perren TJ, Swart AM, Pfisterer J, et al: A phase 3 trial of bevacizumab in ovarian cancer. N Engl J Med 365:2484-2496, 2011
4. Colombo N, Sessa C, Bois AD, et al: ESMO-ESGO consensus conference recommendations on ovarian cancer: Pathology and molecular biology, early and
advanced stages, borderline tumours and recurrent disease. Int J Gynecol Cancer 30:672-705, 2019
5. Collinson F, Hutchinson M, Craven RA, et al: Predicting response to bevacizumab in ovarian cancer: A panel of potential biomarkers informing treatment
selection. Clin Cancer Res 19:5227-5239, 2013
6. Tewari KS, Burger RA, Enserro D, et al: Final overall survival of a randomized trial of bevacizumab for primary treatment of ovarian cancer. J Clin Oncol 37:
2317-2328, 2019
7. Oza AM, Cook AD, Pfisterer J, et al: Standard chemotherapy with or without bevacizumab for women with newly diagnosed ovarian cancer (ICON7): Overall
survival results of a phase 3 randomised trial. Lancet Oncol 16:928-936, 2015
8. Pujade-Lauraine E, Hilpert F, Weber B, et al: Bevacizumab combined with chemotherapy for platinum-resistant recurrent ovarian cancer: The AURELIA open-
label randomized phase III trial. J Clin Oncol 32:1302-1308, 2014
9. You B, Freyer G, Gonzalez-Martin A, et al: The role of the tumor primary chemosensitivity relative to the success of the medical-surgical management in patients
with advanced ovarian carcinomas. Cancer Treat Rev 100:102294, 2021
10. Colomban O, Tod M, Leary A, et al: Early modeled longitudinal CA-125 kinetics and survival of ovarian cancer patients: A GINECO AGO MRC CTU study. Clin
Cancer Res 25:5342-5350, 2019
11. You B, Colomban O, Heywood M, et al: The strong prognostic value of KELIM, a model-based parameter from CA 125 kinetics in ovarian cancer: Data from
CALYPSO trial (a GINECO-GCIG study). Gynecol Oncol 130:289-294, 2013
12. Lauby A, Colomban O, Corbaux P, et al: The increasing prognostic and predictive roles of the tumor primary chemosensitivity assessed by CA-125 elimination
rate constant K (KELIM) in ovarian cancer: A narrative review. Cancers (Basel) 14:98, 2021
13. Dartois C, Brendel K, Comets E, et al: Overview of model-building strategies in population PK/PD analyses: 2002-2004 literature survey. Br J Clin Pharmacol 64:
603-612, 2007
14. Biomarker Kinetics: Modeled CA-125 KELIMÔ in patients with stage III-IV high grade serous ovarian carcinomas treated with first line adjuvant chemotherapy.
https://www.biomarker-kinetics.org/CA-125
15. Van Wagensveld L, Colomban O, Van der AA MA, et al: The prognostic value of chemosensitivity, estimated by the modeled CA-125 KELIM, in ovarian cancer
patients treated with neo-adjuvant chemotherapy in the Netherlands. Ann Oncol 31, 2020 (suppl; abstr 847P)
16. You B, Robelin P, Tod M, et al: CA-125 ELIMination rate constant K (KELIM) is A marker of chemosensitivity in patients with ovarian cancer: Results from the
phase II CHIVA trial. Clin Cancer Res 26:4625-4632, 2020
17. Heller G, McCormack R, Kheoh T, et al: Circulating tumor cell number as a response measure of prolonged survival for metastatic castration-resistant prostate
cancer: A comparison with prostate-specific antigen across five randomized phase III clinical trials. J Clin Oncol 36:572-580, 2018
18. Beal SL, Boeckmann A, Bauer R, et al: NONMEM User’s Guides (1989–2009). Ellicott City, MD, Icon, 2009
19. Jonsson EN, Karlsson MO: Xpose—An S-PLUS based population pharmacokinetic/pharmacodynamic model building aid for NONMEM. Comput Methods
Programs Biomed 58:51-64, 1999
20. van Driel WJ, Koole SN, Sikorska K, et al: Hyperthermic intraperitoneal chemotherapy in ovarian cancer. N Engl J Med 378:230-240, 2018
21. Colomban O, Tod M, Peron J, et al: Bevacizumab for newly diagnosed ovarian cancers: Best candidates among high-risk disease patients (ICON-7). JNCI
Cancer Spectr 4:pkaa026, 2020
22. You B, Clamp A, Cook A, et al: Differential benefit from fractionated dose-dense first-line chemotherapy for epithelial ovarian cancer (EOC) according to KELIM-
evaluated tumor primary chemosensitivity: Exploratory analyses of ICON-8 trial. J Clin Oncol 39, 2021 (suppl 15; abstr 5530)
23. Clamp AR, James EC, McNeish IA, et al: Weekly dose-dense chemotherapy in first-line epithelial ovarian, fallopian tube, or primary peritoneal carcinoma
treatment (ICON8): Primary progression free survival analysis results from a GCIG phase 3 randomised controlled trial. Lancet 394:2084-2095, 2019
24. Printezi MI, Kilgallen AB, Bond MJG, et al: Toxicity and efficacy of chronomodulated chemotherapy: A systematic review. Lancet Oncol 23:e129-e143, 2022
25. Wright AA, Bohlke K, Armstrong DK, et al: Neoadjuvant chemotherapy for newly diagnosed, advanced ovarian cancer: Society of Gynecologic Oncology and
American Society of Clinical Oncology clinical practice guideline. J Clin Oncol 34:3460-3473, 2016
26. Tew WP, Lacchetti C, Ellis A, et al: PARP inhibitors in the management of ovarian cancer: ASCO Guideline. J Clin Oncol 38:3468-3493, 2020
27. Norquist BM, Brady MF, Harrell MI, et al: Mutations in homologous recombination Genes and outcomes in ovarian carcinoma patients in GOG 218: An NRG
Oncology/Gynecologic Oncology Group study. Clin Cancer Res 24:777-783, 2018
28. You B, Sehgal V, Hosmane B, et al: The CA-125 KELIM as a potential complementary tool for predicting veliparib benefit: An exploratory analysis from the VELIA/
GOG-3005 Study. J Clin Oncol 10.1200/JCO.22.00430 [epub ahead of print on July 22, 2022]
29. Biomarker Kinetics: Modeled CA-125 KELIMÔ in patients with stage III-IV high grade serous ovarian carcinomas treated with first line neo-adjuvant che-
motherapy. https://www.biomarker-kinetics.org/CA-125-neo

n n n

3974 © 2022 by American Society of Clinical Oncology Volume 40, Issue 34

 KELIM and Bevacizumab in GOG-0218

AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Identification of Patients With Ovarian Cancer Experiencing the Highest Benefit From Bevacizumab in the First-Line Setting on the Basis of Their Tumor-Intrinsic
Chemosensitivity (KELIM): The GOG-0218 Validation Study
The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted.
Relationships are self-held unless noted. I 5 Immediate Family Member, Inst 5 My Institution. Relationships may not relate to the subject matter of this manuscript.
For more information about ASCO’s conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/jco/authors/author-center.
Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).

Benoit You Bradley J. Monk

Consulting or Advisory Role: `Roche/Genentech, AstraZeneca, Novartis, LEK, Leadership: US Oncology
TESARO, Bayer, Amgen, Clovis Oncology, GlaxoSmithKline, ECS PROGASTRIN, Honoraria: Agenus, Akeso Biopharma, Amgen, Aravive, AstraZeneca, Clovis
Immunomedics, Daiichi Sankyo Europe GmbH, Myriad Genetics, MSD Oncology, Oncology, Eisai, Genmab/Seattle Genetics, ImmunoGen, Iovance
Seattle Genetics Biotherapeutics, Merck, Mersana, Pfizer, Puma Biotechnology, Regeneron,
Research Funding: Merck Serono (Inst), Roche/Genentech (Inst), Clovis Oncology (Inst) Roche/Genentech, TESARO/GSK, Vascular Biogenics, GOG Foundation, Elevar
Travel, Accommodations, Expenses: Roche/Genentech, AstraZeneca, BMS, Therapeutics, Novocure, Gradalis, Karyopharm Therapeutics, Bayer, EMD
MSD Oncology, Bayer Serono/Merck, Macrogenics, Sorrento Therapeutics, US Oncology, Myriad
Pharmaceuticals, Novartis, OncoC4, Pieris Pharmaceuticals
Larry J. Copeland
Consulting or Advisory Role: Agenus, Akeso Biopharma, Amgen, Aravive,
Consulting or Advisory Role: Myriad Genetics, GlaxoSmithKline, Elevar Therapeutics,
AstraZeneca, Clovis Oncology, Eisai, Genmab/Seattle Genetics, GOG
Toray Industries, Rubius Therapeutics, Sorrento Therapeutics, Celsion, Corcept
Foundation, ImmunoGen, Iovance Biotherapeutics, Merck, Mersana, Myriad
Therapeutics, VBL Therapeutics, Onconova Therapeutics, InxMed, Immunogen
Pharmaceuticals, Pfizer, Puma Biotechnology, Regeneron, Roche/Genentech,
Elizabeth M. Swisher TESARO/GSK, Vascular Biogenics, Gradalis, Karyopharm Therapeutics,
Leadership: IDEAYA Biosciences Sorrento Therapeutics, Novocure, Bayer, Elevar Therapeutics, EMD Serono/
Michael A. Bookman Merck, Gradalis, US Oncology, Novartis, Pieris Pharmaceuticals, OncoC4
Speakers’ Bureau: Roche/Genentech, AstraZeneca, Clovis Oncology, Eisai,
Employment: The Permanente Medical Group
Consulting or Advisory Role: AstraZeneca, AbbVie, Immunogen, Merck Sharp & TESARO/GSK, Merck
Research Funding: Novartis (Inst), Amgen (Inst), Genentech (Inst), Lilly (Inst),
Dohme, Genentech/Roche, Seattle Genetics, Aravive
Janssen (Inst), Array BioPharma (Inst), Tesaro (Inst), Morphotek (Inst), Pfizer (Inst),
Gini Fleming Advaxis (Inst), AstraZeneca (Inst), Immunogen (Inst), Regeneron (Inst), Nucana (Inst)
This author is an Associate Editor for Journal of Clinical Oncology. Journal policy
Robert S. Mannel
recused the author from having any role in the peer review of this manuscript.
Stock and Other Ownership Interests: Edwards Lifesciences, Stryker, Danaher
Honoraria: Curio Science, Physicans’ Education Resource
Research Funding: Corcept Therapeutics (Inst), AbbVie (Inst), Iovance Joan L. Walker
Biotherapeutics (Inst), Syros Pharmaceuticals (Inst), Sermonix Pharmaceuticals Research Funding: us biotest, Genentech, Pieces Tech (Inst)
(Inst), Compugen (Inst), Plexxikon (Inst), Roche (Inst), GlaxoSmithKline (Inst),
Fabio Cappuccini
Celldex (Inst), AstraZeneca (Inst), Molecular Templates (Inst), CytomX
Honoraria: GlaxoSmithKline, AstraZeneca
Therapeutics (Inst), Astellas Pharma (Inst), K-Group Beta (Inst)
Speakers’ Bureau: GlaxoSmithKline, AstraZeneca
Other Relationship: DSI (Inst), Merck (Inst), Caris Life Sciences (Inst), Eisai
Patents, Royalties, Other Intellectual Property: sponsored programs
(Inst), AstraZeneca (Inst)
Uncompensated Relationships: AbbVie David Cohn
Research Funding: NRG Oncology (Inst), Advaxis (Inst), Agenus (Inst), Ajinomoto
Robert Coleman
(Inst), Array BioPharma (Inst), AstraZeneca (Inst), Bristol Myers Squibb (Inst),
Employment: US Oncology
Clovis Oncology (Inst), Ergomed (Inst), Exelixis (Inst), Genentech (Inst),
Leadership: Onxeo
GlaxoSmithKline (Inst), Gynecologic Oncology Group (Inst), ImmunoGen (Inst),
Stock and Other Ownership Interests: McKesson
INC Research (Inst), inVentiv Health (Inst), Janssen Research & Development
Consulting or Advisory Role: Clovis Oncology, Genentech/Roche, AstraZeneca/
(Inst), Ludwig Institute for Cancer Research (Inst), PRA International (Inst), EMD
MedImmune, Genmab, Tesaro, OncoMed, Sotio, Oncolytics, AbbVie/
Serono (Inst), Stemcentrx (Inst), Tesaro (Inst), AbbVie (Inst), Henry Jackson
Stemcentrx, Immunogen, AbbVie, Agenus, Novocure, Merck, OncXerna
Foundation (Inst), PharmaMar (Inst), Sanofi (Inst), Eisai (Inst), Pfizer (Inst),
Therapeutics, Alkermes, Gradalis, Regeneron
Novartis (Inst), Regeneron (Inst), Tricon Pharmaceuticals (Inst)
Research Funding: AstraZeneca/MedImmune, Esperance Pharmaceuticals,
Other Relationship: Elsevier, UpToDate
Array BioPharma, Clovis Oncology, Johnson & Johnson, Merck, Roche/
Genentech, Abbott/AbbVie, Immunogen (Inst), Mirati Therapeutics (Inst), David Mutch
Amgen (Inst), Pfizer (Inst), Lilly (Inst), Regeneron (Inst) Consulting or Advisory Role: lilly
Travel, Accommodations, Expenses: Merck, AstraZeneca/MedImmune, Array
Andrea Wahner-Hendrickson
BioPharma, Clovis Oncology, Roche/Genentech, Research to Practice, GOG Consulting or Advisory Role: Oxcia
Foundation, Sotio, Vaniam Group Research Funding: ProLynx (Inst), Amgen (Inst)
Leslie M. Randall Lainie Martin
Honoraria: BluPrint Oncology, PER, Curio Science, Projects in Knowledge, Projects in Consulting or Advisory Role: Sutro Biopharma, Elucida Oncology, GlaxoSmithKline
Knowledge
Research Funding: Agenus (Inst), AstraZeneca (Inst), Sutro bioPharma (Inst),
Consulting or Advisory Role: AstraZeneca, Clovis Oncology, GOG Foundation, Immunogen (Inst)
Merck, Mersana, Agenus, Rubius Therapeutics, Myriad Genetics, EMD Serono,
Travel, Accommodations, Expenses: AstraZeneca
Genentech/Roche, Seattle Genetics, Novartis, Eisai
Speakers’ Bureau: AstraZeneca, Tesaro, Merck Robert A. Burger
Research Funding: Genentech/Roche (Inst), On Target Laboratories (Inst), Employment: Genentech/Roche, Mersana
Pfizer (Inst), Aivita Biomedical (Inst), Tesaro (Inst), AstraZeneca (Inst), Merck Stock and Other Ownership Interests: Genentech/Roche, Mersana
(Inst), Akeso Biopharma (Inst), GEICO (Inst) Consulting or Advisory Role: Myriad Genetics
Krishnansu S. Tewari No other potential conflicts of interest were reported.
Honoraria: Tesaro, Clovis Oncology, Merck, Eisai, AstraZeneca, Genmab
Consulting or Advisory Role: Roche/Genentech, Tesaro, Clovis Oncology,
AstraZeneca
Speakers’ Bureau: Roche/Genentech, AstraZeneca, Merck, Tesaro, Clovis
Oncology, Eisai, Genmab
Research Funding: AbbVie (Inst), Genentech/Roche (Inst), Morphotek (Inst),
Merck (Inst), Regeneron (Inst)
Travel, Accommodations, Expenses: Roche/Genentech

Journal of Clinical Oncology