A Study of Pembrolizumab (MK-3475) Versus Paclitaxel for Participants With Advanced Gastric/Gastroesophageal Junction Adenocarcinoma That Progressed After Therapy With Platinum and Fluoropyrimidine (MK-3475-061/KEYNOTE-061)

• Sponsor: Merck Sharp & Dohme LLC (Industry)
• Overall Status: Completed
• CT.gov ID: NCT02370498
• Collaborator: (none)
• Enrollment: 592
• Arms: 2
• Actual Duration (Months): 73

Study Details

Study Description

Brief Summary

This is a study for participants with advanced gastric or gastroesophageal junction adenocarcinoma who have had tumor progression after first-line treatment with platinum and fluoropyrimidine doublet therapy. The primary study hypotheses are that pembrolizumab (MK-3475) prolongs progression free survival (PFS) and overall survival (OS) for participants with tumors that show positive programmed cell death ligand 1 (PD-L1) expression.

As of 20-March-2016, enrollment will be limited to PD-L1 positive participants.

Condition or Disease	Intervention/Treatment	Phase
Gastric Adenocarcinoma; Gastroesophageal Junction Adenocarcinoma	Biological: pembrolizumab; Drug: paclitaxel	Phase 3

Study Design

• Study Type: Interventional
• Actual Enrollment: 592 participants
• Allocation: Randomized
• Intervention Model: Parallel Assignment
• Masking: None (Open Label)
• Primary Purpose: Treatment
• Official Title: A Phase III, Randomized, Open-label Clinical Trial of Pembrolizumab (MK-3475) Versus Paclitaxel in Subjects With Advanced Gastric or Gastroesophageal Junction Adenocarcinoma Who Progressed After First-Line Therapy With Platinum and Fluoropyrimidine
• Actual Study Start Date: May 11, 2015
• Actual Primary Completion Date: Oct 26, 2017
• Actual Study Completion Date: Jun 10, 2021

Arms and Interventions

Arm	Intervention/Treatment
Experimental: Pembrolizumab Participants receive 200 mg intravenous (IV) pembrolizumab on Day 1 of each 21-day cycle, for up to 35 administrations (approximately 2 years).	Biological: pembrolizumab; 200 mg administered as IV infusion on Day 1 of each 21-day cycle.; Other Names: MK-3475
Active Comparator: Paclitaxel Participants receive 80 mg/m^2 IV paclitaxel on Days 1, 8, and 15 of each 28-day cycle, until disease progression or unacceptable toxicity.	Drug: paclitaxel; 80 mg/m^2 administered as IV infusion on Days 1, 8, and 15 of each 28-day cycle.; Other Names: TAXOL®

Outcome Measures

Primary Outcome Measures

1. Progression-free Survival (PFS) According to Response Criteria in Solid Tumors Version 1.1 (RECIST 1.1) Based on Blinded Independent Central Review (BICR) in Programmed Death-Ligand 1 (PD-L1) Positive Participants [Up to 30 months (through database cut-off date of 26 Oct 2017)]

   PFS was defined as the time from randomization to the first documented disease progression per RECIST 1.1 based on BICR, or death due to any cause, whichever occurs first. According to RECIST 1.1, progressive disease (PD) was defined as a 20% relative increase in the sum of diameters (SOD) of target lesions, taking as reference the nadir SOD and an absolute increase of >5 mm in the SOD, or the appearance of new lesions. PFS was analyzed using the Kaplan-Meier method and median PFS (95% confidence interval [CI]) in months was reported for PD-L1 positive participants by treatment group.

2. Overall Survival (OS) in PD-L1 Positive Participants [Up to 30 months (through database cut-off date of 26 Oct 2017)]

   OS was defined as the time from randomization to death due to any cause. Participants without documented death at the time of the final analysis were censored at the date of the last follow-up. OS was analyzed using the Kaplan-Meier method and median OS (95% CI) in months was reported for PD-L1 positive participants by treatment group.

Secondary Outcome Measures

1. PFS According to RECIST 1.1 Based on BICR in All Participants [Up to 30 months (through database cut-off date of 26 Oct 2017)]

   PFS was defined as the time from randomization to the first documented disease progression per RECIST 1.1 based on BICR, or death due to any cause, whichever occurs first. According to RECIST 1.1, PD was defined as a 20% relative increase in the SOD of target lesions, taking as reference the nadir SOD and an absolute increase of >5 mm in the SOD, or the appearance of new lesions. PFS was analyzed using the Kaplan-Meier method and median PFS (95% CI) in months was reported for all participants by treatment group.

2. OS in All Participants [Up to 30 months (through database cut-off date of 26 Oct 2017)]

   OS was defined as the time from randomization to death due to any cause. Participants without documented death at the time of the final analysis were censored at the date of the last follow-up. OS was analyzed using the Kaplan-Meier method and median OS (95% CI) in months was reported for all participants by treatment group.

3. PFS According to RECIST 1.1 Based on Investigator Assessment in PD-L1 Positive Participants [Up to 30 months (through database cut-off date of 26 Oct 2017)]

   PFS was defined as the time from randomization to the first documented disease progression per RECIST 1.1 based on investigator assessment, or death due to any cause, whichever occurs first. According to RECIST 1.1, PD was defined as a 20% relative increase in the SOD of target lesions, taking as reference the nadir SOD and an absolute increase of >5 mm in the SOD, or the appearance of new lesions. PFS was analyzed using the Kaplan-Meier method and median PFS (95% CI) in months was reported for PD-L1 positive participants by treatment group.

4. PFS According to RECIST 1.1 Based on Investigator Assessment in All Participants [Up to 30 months (through database cut-off date of 26 Oct 2017)]

   PFS was defined as the time from randomization to the first documented disease progression per RECIST 1.1 based on investigator assessment, or death due to any cause, whichever occurs first. According to RECIST 1.1, PD was defined as a 20% relative increase in the SOD of target lesions, taking as reference the nadir SOD and an absolute increase of >5 mm in the SOD, or the appearance of new lesions. PFS was analyzed using the Kaplan-Meier method and median PFS (95% CI) in months was reported for all participants by treatment group.

5. PFS According to Immune-Related Response Evaluation Criteria in Solid Tumors (irRECIST) Based on BICR in PD-L1 Positive Participants [Up to 30 months (through database cut-off date of 26 Oct 2017)]

   PFS defined as time from randomization to first documented PD per irRECIST based on BICR, or death due to any cause, whichever occurs first. Following initial PD by RECIST 1.1 (20% relative increase in SOD of target lesions), participants were assessed according to irRECIST: tumor assessment was repeated ≥4 weeks later to confirm PD with the option of continuing treatment until this scan was obtained for clinically stable participants. If PD confirmed, participant was discontinued from treatment unless investigator determined benefit. If repeat scan indicated stable disease (SD; neither sufficient shrinkage or increase of target lesion), partial response (PR; ≥30% decrease in the SOD of target lesions), or complete response (CR; disappearance of all non-nodal target lesions), participant could continue treatment at investigator's discretion. PFS analyzed using Kaplan-Meier method and median PFS (95% CI) in months was reported for PD-L1 positive participants by treatment group.

6. PFS According to irRECIST Based on BICR in All Participants [Up to 30 months (through database cut-off date of 26 Oct 2017)]

   PFS defined as time from randomization to first documented PD per irRECIST based on BICR, or death due to any cause, whichever occurs first. Following initial PD by RECIST 1.1 (20% relative increase in SOD of target lesions), participants were assessed according to irRECIST: tumor assessment was repeated ≥4 weeks later to confirm PD with the option of continuing treatment until this scan was obtained for clinically stable participants. If PD confirmed, participant was discontinued from treatment unless investigator determined benefit. If repeat scan indicated SD (neither sufficient shrinkage or increase of target lesion), PR (≥30% decrease in the SOD of target lesions), or CR (disappearance of all non-nodal target lesions), participant could continue treatment at investigator's discretion. PFS analyzed using Kaplan-Meier method and median PFS (95% CI) in months was reported for all participants by treatment group.

7. Time to Tumor Progression (TTP) According to RECIST 1.1 Based on BICR in PD-L1 Positive Participants [Up to 30 months (through database cut-off date of 26 Oct 2017)]

   TTP was defined as the time from randomization to the first documented disease progression per RECIST 1.1 based on BICR. Using RECIST 1.1, progressive disease was defined as a 20% relative increase in the SOD of target lesions, taking as reference the nadir SOD and an absolute increase of >5 mm in the SOD, or the appearance of new lesions. If there was no documented disease progression, TTP was censored at last tumor assessment date. TTP was analyzed using the Kaplan-Meier method and median TTP (95% CI) in months was reported for PD-L1 positive participants by treatment group.

8. TTP According to RECIST 1.1 Based on BICR in All Participants [Up to 30 months (through database cut-off date of 26 Oct 2017)]

   TTP was defined as the time from randomization to the first documented disease progression per RECIST 1.1 based on BICR. Using RECIST 1.1, progressive disease was defined as a 20% relative increase in the SOD of target lesions, taking as reference the nadir SOD and an absolute increase of >5 mm in the SOD, or the appearance of new lesions. If there was no documented disease progression, TTP was censored at last tumor assessment date. TTP was analyzed using the Kaplan-Meier method and median TTP (95% CI) in months was reported for all participants by treatment group.

9. TTP According to RECIST 1.1 Based on Investigator Assessment in PD-L1 Positive Participants [Up to 30 months (through database cut-off date of 26 Oct 2017)]

   TTP was defined as the time from randomization to the first documented disease progression per RECIST 1.1 based on investigator assessment. Using RECIST 1.1, progressive disease was defined as a 20% relative increase in the SOD of target lesions, taking as reference the nadir SOD and an absolute increase of >5 mm in the SOD, or the appearance of new lesions. If there was no documented disease progression, TTP was censored at last tumor assessment date. TTP was analyzed using the Kaplan-Meier method and median TTP (95% CI) in months was reported for PD-L1 positive participants by treatment group.

10. TTP According to RECIST 1.1 Based on Investigator Assessment in All Participants [Up to 30 months (through database cut-off date of 26 Oct 2017)]

    TTP was defined as the time from randomization to the first documented disease progression per RECIST 1.1 based on investigator assessment. Using RECIST 1.1, progressive disease was defined as a 20% relative increase in the SOD of target lesions, taking as reference the nadir SOD and an absolute increase of >5 mm in the SOD, or the appearance of new lesions. If there was no documented disease progression, TTP was censored at last tumor assessment date. TTP was analyzed using the Kaplan-Meier method and median TTP (95% CI) in months was reported for all participants by treatment group.

11. Objective Response Rate (ORR) According to RECIST 1.1 Based on BICR in PD-L1 Positive Participants [Up to 30 months (through database cut-off date of 26 Oct 2017)]

    ORR was defined as the percentage of the participants in the analysis population who had a confirmed CR (disappearance of all non-nodal target lesions and any pathological lymph nodes must have become normal) or PR (at least a 30% decrease in the SOD of target lesions, taking as reference the baseline SOD) according to RECIST 1.1 and based on BICR. ORR was analyzed using the stratified Miettinen and Nurminen method, and reported with 95% CI for PD-L1 positive participants by treatment group.

12. ORR According to RECIST 1.1 Based on BICR in All Participants [Up to 30 months (through database cut-off date of 26 Oct 2017)]

    ORR was defined as the percentage of the participants in the analysis population who had a confirmed CR (disappearance of all non-nodal target lesions and any pathological lymph nodes must have become normal) or PR (at least a 30% decrease in the SOD of target lesions, taking as reference the baseline SOD) according to RECIST 1.1 and based on BICR. ORR was analyzed using the stratified Miettinen and Nurminen method, and reported with 95% CI for all participants by treatment group.

13. ORR According to RECIST 1.1 Based on Investigator Assessment in PD-L1 Positive Participants [Up to 30 months (through database cut-off date of 26 Oct 2017)]

    ORR was defined as the percentage of the participants in the analysis population who had a confirmed CR (disappearance of all non-nodal target lesions and any pathological lymph nodes must have become normal) or PR (at least a 30% decrease in the SOD of target lesions, taking as reference the baseline SOD) according to RECIST 1.1 and based on investigator assessment. ORR was analyzed using the stratified Miettinen and Nurminen method, and reported with 95% CI for PD-L1 positive participants by treatment group.

14. ORR According to RECIST 1.1 Based on Investigator Assessment in All Participants [Up to 30 months (through database cut-off date of 26 Oct 2017)]

    ORR was defined as the percentage of the participants in the analysis population who had a confirmed CR (disappearance of all non-nodal target lesions and any pathological lymph nodes must have become normal) or PR (at least a 30% decrease in the SOD of target lesions, taking as reference the baseline SOD) according to RECIST 1.1 and based on investigator assessment. ORR was analyzed using the stratified Miettinen and Nurminen method, and reported with 95% CI for all participants by treatment group.

15. Duration of Response (DOR) According to RECIST 1.1 Based on BICR in PD-L1 Positive Participants [Up to 30 months (through database cut-off date of 26 Oct 2017)]

    For PD-L1 positive participants who demonstrated CR (disappearance of all non-nodal target lesions and any pathological lymph nodes must have become normal) or PR (at least a 30% decrease in the SOD of target lesions, taking as reference the baseline SOD) according to RECIST 1.1 and based on BICR, DOR was defined as the time from first documented evidence of CR or PR until disease progression or death due to any cause, whichever occurs first. DOR for participants who had not progressed or died at the time of analysis was censored at the date of their last tumor assessment. DOR was analyzed using the Kaplan-Meier method and median DOR (range) in months was reported for PD-L1 positive participants with response by treatment group.

16. DOR According to RECIST 1.1 Based on BICR in All Participants [Up to 30 months (through database cut-off date of 26 Oct 2017)]

    For participants who demonstrated CR (disappearance of all non-nodal target lesions and any pathological lymph nodes must have become normal) or PR (at least a 30% decrease in the SOD of target lesions, taking as reference the baseline SOD) according to RECIST 1.1 and based on BICR, DOR was defined as the time from first documented evidence of CR or PR until disease progression or death due to any cause, whichever occurs first. DOR for participants who had not progressed or died at the time of analysis was censored at the date of their last tumor assessment. DOR was analyzed using the Kaplan-Meier method and median DOR (range) in months was reported for all participants with response by treatment group.

17. DOR According to RECIST 1.1 Based on Investigator Assessment in PD-L1 Positive Participants [Up to 30 months (through database cut-off date of 26 Oct 2017)]

    For PD-L1 positive participants who demonstrated CR (disappearance of all non-nodal target lesions and any pathological lymph nodes must have become normal) or PR (at least a 30% decrease in the SOD of target lesions, taking as reference the baseline SOD) according to RECIST 1.1 and based on investigator assessment, DOR was defined as the time from first documented evidence of CR or PR until disease progression or death due to any cause, whichever occurs first. DOR for participants who had not progressed or died at the time of analysis was censored at the date of their last tumor assessment. DOR was analyzed using the Kaplan-Meier method and median DOR (range) in months was reported for PD-L1 positive participants with response by treatment group.

18. DOR According to RECIST 1.1 Based on Investigator Assessment in All Participants [Up to 30 months (through database cut-off date of 26 Oct 2017)]

    For participants who demonstrated CR (disappearance of all non-nodal target lesions and any pathological lymph nodes must have become normal) or PR (at least a 30% decrease in the SOD of target lesions, taking as reference the baseline SOD) according to RECIST 1.1 and based on investigator assessment, DOR was defined as the time from first documented evidence of CR or PR until disease progression or death due to any cause, whichever occurs first. DOR for participants who had not progressed or died at the time of analysis was censored at the date of their last tumor assessment. DOR was analyzed using the Kaplan-Meier method and median DOR (range) in months was reported for all participants with response by treatment group.

19. Percentage of PD-L1 Positive Participants Who Experienced an Adverse Event (AE) [Up to 71 months (through database cut-off date of 10 Jun 2021)]

    An AE was defined as any untoward medical occurrence in a participant administered a pharmaceutical product and which does not necessarily have to have a causal relationship with this treatment. An AE could therefore be any unfavourable and unintended sign, symptom, or disease temporally associated with the use of a medicinal product or protocol-specified procedure, whether or not considered related to the medicinal product or protocol-specified procedure. Any worsening of a pre-existing condition that was temporally associated with the use of the Sponsor's product was also an AE. The percentage of participants with at least one AE was reported for PD-L1 positive participants by treatment group.

20. Percentage of All Participants Who Experienced an AE [Up to 71 months (through database cut-off date of 10 Jun 2021)]

    An AE was defined as any untoward medical occurrence in a participant administered a pharmaceutical product and which does not necessarily have to have a causal relationship with this treatment. An AE could therefore be any unfavourable and unintended sign, symptom, or disease temporally associated with the use of a medicinal product or protocol-specified procedure, whether or not considered related to the medicinal product or protocol-specified procedure. Any worsening of a pre-existing condition that was temporally associated with the use of the Sponsor's product was also an AE. The percentage of participants with at least one AE was reported for all participants by treatment group.

21. Percentage of PD-L1 Positive Participants That Discontinued Study Treatment Due to AE [Up to approximately 26.4 months]

    An AE was defined as any untoward medical occurrence in a participant administered a pharmaceutical product and which does not necessarily have to have a causal relationship with this treatment. An AE could therefore be any unfavourable and unintended sign, symptom, or disease temporally associated with the use of a medicinal product or protocol-specified procedure, whether or not considered related to the medicinal product or protocol-specified procedure. Any worsening of a pre-existing condition that was temporally associated with the use of the Sponsor's product was also an AE. The percentage of participants that discontinued study treatment due to an AE was reported for PD-L1 positive participants by treatment group.

22. Percentage of All Participants That Discontinued Study Treatment Due to AE [Up to approximately 26.4 months]

    An AE was defined as any untoward medical occurrence in a participant administered a pharmaceutical product and which does not necessarily have to have a causal relationship with this treatment. An AE could therefore be any unfavourable and unintended sign, symptom, or disease temporally associated with the use of a medicinal product or protocol-specified procedure, whether or not considered related to the medicinal product or protocol-specified procedure. Any worsening of a pre-existing condition that was temporally associated with the use of the Sponsor's product was also an AE. The percentage of participants that discontinued study treatment due to an AE was reported for all participants by treatment group.

Eligibility Criteria

Criteria

• Ages Eligible for Study: 18 Years and Older
• Sexes Eligible for Study: All
• Accepts Healthy Volunteers: No

Inclusion Criteria:

• Histologically- or cytologically-confirmed diagnosis of gastric or gastroesophageal junction adenocarcinoma

• Confirmed metastatic or locally advanced, unresectable disease (by computed tomography [CT] scan or clinical evidence)

• Eastern Cooperative Oncology Group (ECOG) performance status of 0 or 1

• Progression on or after prior first-line therapy containing any platinum/fluoropyrimidine doublet

• Willing to provide tumor tissue for PD-L1 biomarker analysis (new or archived specimens with agreement of Sponsor). As of 20 March 2016, participants must be PD-L1 positive to be enrolled.

• Human epidermal growth factor receptor 2 (HER-2/neu) status known and participants with HER2/neu positive tumors show documentation of disease progression on treatment containing trastuzumab

• Female participants of childbearing potential should be willing to use 2 methods of birth control or be surgically sterile, or abstain from heterosexual activity for the course of the study through 120 days after the last dose of pembrolizumab or through 180 days after the last dose of paclitaxel.

• Male participants should agree to use an adequate method of contraception starting with the first dose of study therapy through 120 days after the last dose of pembrolizumab or through 180 days after the last dose of paclitaxel.

• Adequate organ function

Exclusion Criteria:

• Currently participating and receiving study therapy, or participated in a study of an investigational agent and received study therapy or used an investigation device within 4 weeks of the first dose of medication

• Squamous cell or undifferentiated gastric cancer

• Active autoimmune disease that has required systemic treatment in past 2 years (replacement therapy is not considered a form of systemic treatment

• Diagnosis of immunodeficiency or receiving systemic steroid therapy or any other form of immunosuppressive therapy within 7 days prior to the first dose of study medication

• Prior anti-cancer monoclonal antibody (mAb) within 4 weeks prior to study Day 1 or not recovered from AEs due to agents administered more than 4 weeks earlier

• Prior chemotherapy, targeted small molecule therapy, or radiation therapy within 2 weeks prior to study Day 1 or not recovered from adverse events due to a previously administered agent or surgery

• Known additional malignancy that is progressing or requires active treatment (with the exception of basal cell carcinoma of the skin, squamous cell carcinoma of the skin, or in situ cervical cancer that has undergone potentially curative therapy)

• Known active central nervous system (CNS) metastases and/or carcinomatous meningitis

• History of (noninfectious) pneumonitis that required steroids or current pneumonitis

• Active infection requiring systemic therapy

• Known psychiatric or substance abuse disorders that would interfere with cooperation with the requirements of the trial

• Pregnant or breastfeeding, or expecting to conceive or father children within the projected duration of the trial, starting with the screening visit through 120 days after the last dose of pembrolizumab or through 180 days after the last dose of paclitaxel.

• Prior immunotherapy including anti-PD-1, anti-PD-L1, or anti-PD-L2 agent, or previously participated in Merck pembrolizumab (MK-3475) clinical trial

• Known history of human immunodeficiency virus (HIV)

• Known active Hepatitis B or Hepatitis C

• Live vaccine within 30 days of planned start of study therapy

• Known allergy or hypersensitivity to paclitaxel or any components used in the paclitaxel preparation or other contraindication for taxane therapy

Contacts and Locations

Locations

No locations specified.

Sponsors and Collaborators

• Merck Sharp & Dohme LLC

Investigators

• Study Director: Medical Director, Merck Sharp & Dohme LLC

Study Documents (Full-Text)

• Study Protocol and Statistical Analysis Plan - Mar 9, 2020

More Information

Additional Information:

• Merck Oncology Clinical Trials Information

Publications

None provided.

• Responsible Party: Merck Sharp & Dohme LLC
• ClinicalTrials.gov Identifier: NCT02370498
• Other Study ID Numbers: 3475-061; 152988; MK-3475-061; 2014-005241-45
• First Posted: Feb 25, 2015
• Last Update Posted: Jun 6, 2022
• Last Verified: May 1, 2022

• Individual Participant Data (IPD) Sharing Statement: Yes
• Plan to Share IPD: Yes
• Studies a U.S. FDA-regulated Drug Product: Yes
• Studies a U.S. FDA-regulated Device Product: No
• Keywords provided by Merck Sharp & Dohme LLC: Gastric cancer; Gastroesophageal junction cancer; PD1; PD-1; PDL1; PD-L1
• Additional relevant MeSH terms: Adenocarcinoma; Esophageal Neoplasms; Paclitaxel; Pembrolizumab

Study Results

Participant Flow

Recruitment Details	After 20 March 2016, enrollment was limited to programmed cell death ligand 1 (PD-L1) positive participants.
Pre-assignment Detail	Of 592 participants that were randomized to trial, 570 received treatment. At the time of the primary analysis data cut-off, 89 participants are ongoing in the study.

Arm/Group Title	Pembrolizumab	Paclitaxel
Arm/Group Description	Participants receive 200 mg intravenous (IV) pembrolizumab on Day 1 of each 21-day cycle, for up to 35 administrations (approximately 2 years).	Participants receive paclitaxel 80 mg/m^2 IV, on Days 1, 8, and 15 of each 28-day cycle until disease progression or unacceptable toxicity.
Period Title: Overall Study
STARTED	296	296
Treated	294	276
PD-L1 Positive Participants	196	199
COMPLETED	0	0
NOT COMPLETED	296	296

Baseline Characteristics

Arm/Group Title	Pembrolizumab	Paclitaxel	Total
Arm/Group Description	Participants receive 200 mg IV pembrolizumab on Day 1 of each 21-day cycle, for up to 35 administrations (approximately 2 years).	Participants receive paclitaxel 80 mg/m^2 IV, on Days 1, 8, and 15 of each 28-day cycle until disease progression or unacceptable toxicity.	Total of all reporting groups
Overall Participants	296	296	592
Age (Years) [Mean (Standard Deviation) ]
Mean (Standard Deviation) [Years]	60.7 (12.0)	59.6 (11.7)	60.2 (11.9)
Sex: Female, Male (Count of Participants)
Female	94 31.8%	88 29.7%	182 30.7%
Male	202 68.2%	208 70.3%	410 69.3%
Ethnicity (NIH/OMB) (Count of Participants)
Hispanic or Latino	27 9.1%	24 8.1%	51 8.6%
Not Hispanic or Latino	269 90.9%	272 91.9%	541 91.4%
Unknown or Not Reported	0 0%	0 0%	0 0%
Race (NIH/OMB) (Count of Participants)
American Indian or Alaska Native	4 1.4%	3 1%	7 1.2%
Asian	93 31.4%	91 30.7%	184 31.1%
Native Hawaiian or Other Pacific Islander	0 0%	0 0%	0 0%
Black or African American	4 1.4%	2 0.7%	6 1%
White	193 65.2%	198 66.9%	391 66%
More than one race	2 0.7%	2 0.7%	4 0.7%
Unknown or Not Reported	0 0%	0 0%	0 0%
Region of Enrollment (Number) [Number]
Europe/Israel/North America/Australia	190 64.2%	187 63.2%	377 63.7%
Asia	88 29.7%	89 30.1%	177 29.9%
Rest of World	18 6.1%	20 6.8%	38 6.4%
PD-L1 Status (Count of Participants)
Positive	196 66.2%	199 67.2%	395 66.7%
Negative	99 33.4%	96 32.4%	195 32.9%
Unknown	1 0.3%	1 0.3%	2 0.3%
Time To Progression (TTP) on first-line therapy (Count of Participants)
<6 months	186 62.8%	182 61.5%	368 62.2%
≥6 months	110 37.2%	114 38.5%	224 37.8%

Outcome Measures

1. Primary Outcome

Title	Progression-free Survival (PFS) According to Response Criteria in Solid Tumors Version 1.1 (RECIST 1.1) Based on Blinded Independent Central Review (BICR) in Programmed Death-Ligand 1 (PD-L1) Positive Participants
Description	PFS was defined as the time from randomization to the first documented disease progression per RECIST 1.1 based on BICR, or death due to any cause, whichever occurs first. According to RECIST 1.1, progressive disease (PD) was defined as a 20% relative increase in the sum of diameters (SOD) of target lesions, taking as reference the nadir SOD and an absolute increase of >5 mm in the SOD, or the appearance of new lesions. PFS was analyzed using the Kaplan-Meier method and median PFS (95% confidence interval [CI]) in months was reported for PD-L1 positive participants by treatment group.
Time Frame	Up to 30 months (through database cut-off date of 26 Oct 2017)

Outcome Measure Data

Analysis Population Description

All randomized PD-L1 positive participants. Participants were included in the treatment group to which they were randomized.

Arm/Group Title	Pembrolizumab	Paclitaxel
Arm/Group Description	Participants receive 200 mg IV pembrolizumab on Day 1 of each 21-day cycle, for up to 35 administrations (approximately 2 years).	Participants receive paclitaxel 80 mg/m^2 IV, on Days 1, 8, and 15 of each 28-day cycle until disease progression or unacceptable toxicity.
Measure Participants	196	199
Median (95% Confidence Interval) [months]	1.5	4.1

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Pembrolizumab, Paclitaxel
	Comments	Treatment difference in PFS was assessed by the Stratified Log-rank test, and a stratified Cox proportional hazard model with Efron's method of tie handling was used to assess the magnitude of the treatment difference (Hazard Ratio [HR]) between the treatment arms.
	Type of Statistical Test	Other
	Comments
Statistical Test of Hypothesis	p-Value	0.98358
	Comments	One-sided p-value based on log-rank test stratified by geographic region (Europe/Israel/North America/Australia vs. Asia vs. Rest of World) and TTP on first-line therapy (< 6 months vs. >= 6 months).
	Method	Regression, Cox
	Comments	P-value rounded to 5 decimal places.
Method of Estimation	Estimation Parameter	Hazard Ratio (HR)
	Estimated Value	1.27
	Confidence Interval	(2-Sided) 95%; 1.03 to 1.57
	Parameter Dispersion	Type: Value:
	Estimation Comments

2. Primary Outcome

Title	Overall Survival (OS) in PD-L1 Positive Participants
Description	OS was defined as the time from randomization to death due to any cause. Participants without documented death at the time of the final analysis were censored at the date of the last follow-up. OS was analyzed using the Kaplan-Meier method and median OS (95% CI) in months was reported for PD-L1 positive participants by treatment group.
Time Frame	Up to 30 months (through database cut-off date of 26 Oct 2017)

Outcome Measure Data

Analysis Population Description

All randomized PD-L1 positive participants. Participants were included in the treatment group to which they were randomized.

Arm/Group Title	Pembrolizumab	Paclitaxel
Arm/Group Description	Participants receive 200 mg IV pembrolizumab on Day 1 of each 21-day cycle, for up to 35 administrations (approximately 2 years).	Participants receive paclitaxel 80 mg/m^2 IV, on Days 1, 8, and 15 of each 28-day cycle until disease progression or unacceptable toxicity.
Measure Participants	196	199
Median (95% Confidence Interval) [months]	9.1	8.3

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Pembrolizumab, Paclitaxel
	Comments	Treatment difference in OS was assessed by the Stratified Log-rank test, and a stratified Cox proportional hazard model with Efron's method of tie handling was used to assess the magnitude of the treatment difference (HR) between the treatment arms.
	Type of Statistical Test	Other
	Comments
Statistical Test of Hypothesis	p-Value	0.04205
	Comments	One-sided p-value based on log-rank test stratified by geographic region (Europe/Israel/North America/Australia vs. Asia vs. Rest of World) and TTP on first-line therapy (< 6 months vs. >= 6 months).
	Method	Regression, Cox
	Comments	P-value rounded to 5 decimal places.
Method of Estimation	Estimation Parameter	Hazard Ratio (HR)
	Estimated Value	0.82
	Confidence Interval	(2-Sided) 95%; 0.66 to 1.03
	Parameter Dispersion	Type: Value:
	Estimation Comments

3. Secondary Outcome

Title	PFS According to RECIST 1.1 Based on BICR in All Participants
Description	PFS was defined as the time from randomization to the first documented disease progression per RECIST 1.1 based on BICR, or death due to any cause, whichever occurs first. According to RECIST 1.1, PD was defined as a 20% relative increase in the SOD of target lesions, taking as reference the nadir SOD and an absolute increase of >5 mm in the SOD, or the appearance of new lesions. PFS was analyzed using the Kaplan-Meier method and median PFS (95% CI) in months was reported for all participants by treatment group.
Time Frame	Up to 30 months (through database cut-off date of 26 Oct 2017)

Outcome Measure Data

Analysis Population Description

All randomized participants. Participants were included in the treatment group to which they were randomized.

Arm/Group Title	Pembrolizumab	Paclitaxel
Arm/Group Description	Participants receive 200 mg IV pembrolizumab on Day 1 of each 21-day cycle, for up to 35 administrations (approximately 2 years).	Participants receive paclitaxel 80 mg/m^2 IV, on Days 1, 8, and 15 of each 28-day cycle until disease progression or unacceptable toxicity.
Measure Participants	296	296
Median (95% Confidence Interval) [months]	1.5	4.1

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Pembrolizumab, Paclitaxel
	Comments	Treatment difference in PFS was assessed by the Stratified Log-rank test, and a stratified Cox proportional hazard model with Efron's method of tie handling was used to assess the magnitude of the treatment difference (Hazard Ratio [HR]) between the treatment arms.
	Type of Statistical Test	Other
	Comments
Statistical Test of Hypothesis	p-Value	0.99999
	Comments	One-sided p-value based on log-rank test stratified by geographic region (Europe/Israel/North America/Australia vs. Asia vs. Rest of World), TTP on first-line therapy (< 6 months vs. >= 6 months), and PD-L1 status (positive vs. negative).
	Method	Regression, Cox
	Comments	P-value rounded to 5 decimal places.
Method of Estimation	Estimation Parameter	Hazard Ratio (HR)
	Estimated Value	1.49
	Confidence Interval	(2-Sided) 95%; 1.25 to 1.77
	Parameter Dispersion	Type: Value:
	Estimation Comments

4. Secondary Outcome

Title	OS in All Participants
Description	OS was defined as the time from randomization to death due to any cause. Participants without documented death at the time of the final analysis were censored at the date of the last follow-up. OS was analyzed using the Kaplan-Meier method and median OS (95% CI) in months was reported for all participants by treatment group.
Time Frame	Up to 30 months (through database cut-off date of 26 Oct 2017)

Outcome Measure Data

Analysis Population Description

All randomized participants. Participants were included in the treatment group to which they were randomized.

Arm/Group Title	Pembrolizumab	Paclitaxel
Arm/Group Description	Participants receive 200 mg IV pembrolizumab on Day 1 of each 21-day cycle, for up to 35 administrations (approximately 2 years).	Participants receive paclitaxel 80 mg/m^2 IV, on Days 1, 8, and 15 of each 28-day cycle until disease progression or unacceptable toxicity.
Measure Participants	296	296
Median (95% Confidence Interval) [months]	6.7	8.3

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Pembrolizumab, Paclitaxel
	Comments	Treatment difference in OS was assessed by the Stratified Log-rank test, and a stratified Cox proportional hazard model with Efron's method of tie handling was used to assess the magnitude of the treatment difference (Hazard Ratio [HR]) between the treatment arms.
	Type of Statistical Test	Other
	Comments
Statistical Test of Hypothesis	p-Value	0.24463
	Comments	One-sided p-value based on log-rank test stratified by geographic region (Europe/Israel/North America/Australia vs. Asia vs. Rest of World), TTP on first-line therapy (<6 months vs. ≥6 months), and PD-L1 status (positive vs. negative).
	Method	Regression, Cox
	Comments	P-value rounded to 5 decimal places.
Method of Estimation	Estimation Parameter	Hazard Ratio (HR)
	Estimated Value	0.94
	Confidence Interval	(2-Sided) 95%; 0.79 to 1.12
	Parameter Dispersion	Type: Value:
	Estimation Comments

5. Secondary Outcome

Title	PFS According to RECIST 1.1 Based on Investigator Assessment in PD-L1 Positive Participants
Description	PFS was defined as the time from randomization to the first documented disease progression per RECIST 1.1 based on investigator assessment, or death due to any cause, whichever occurs first. According to RECIST 1.1, PD was defined as a 20% relative increase in the SOD of target lesions, taking as reference the nadir SOD and an absolute increase of >5 mm in the SOD, or the appearance of new lesions. PFS was analyzed using the Kaplan-Meier method and median PFS (95% CI) in months was reported for PD-L1 positive participants by treatment group.
Time Frame	Up to 30 months (through database cut-off date of 26 Oct 2017)

Outcome Measure Data

Analysis Population Description

All randomized PD-L1 positive participants. Participants were included in the treatment group to which they were randomized.

Arm/Group Title	Pembrolizumab	Paclitaxel
Arm/Group Description	Participants receive 200 mg IV pembrolizumab on Day 1 of each 21-day cycle, for up to 35 administrations (approximately 2 years).	Participants receive paclitaxel 80 mg/m^2 IV, on Days 1, 8, and 15 of each 28-day cycle until disease progression or unacceptable toxicity.
Measure Participants	196	199
Median (95% Confidence Interval) [months]	1.6	3.1

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Pembrolizumab, Paclitaxel
	Comments	Treatment difference in PFS was assessed by the Stratified Log-rank test, and a stratified Cox proportional hazard model with Efron's method of tie handling was used to assess the magnitude of the treatment difference (HR) between the treatment arms.
	Type of Statistical Test	Other
	Comments
Statistical Test of Hypothesis	p-Value	0.41331
	Comments	One-sided p-value based on log-rank test stratified by geographic region (Europe/Israel/North America/Australia vs. Asia vs. Rest of World) and TTP on first-line therapy (<6 months vs. ≥6 months).
	Method	Regression, Cox
	Comments	P-value rounded to 5 decimal places.
Method of Estimation	Estimation Parameter	Hazard Ratio (HR)
	Estimated Value	0.98
	Confidence Interval	(2-Sided) 95%; 0.79 to 1.21
	Parameter Dispersion	Type: Value:
	Estimation Comments

6. Secondary Outcome

Title	PFS According to RECIST 1.1 Based on Investigator Assessment in All Participants
Description	PFS was defined as the time from randomization to the first documented disease progression per RECIST 1.1 based on investigator assessment, or death due to any cause, whichever occurs first. According to RECIST 1.1, PD was defined as a 20% relative increase in the SOD of target lesions, taking as reference the nadir SOD and an absolute increase of >5 mm in the SOD, or the appearance of new lesions. PFS was analyzed using the Kaplan-Meier method and median PFS (95% CI) in months was reported for all participants by treatment group.
Time Frame	Up to 30 months (through database cut-off date of 26 Oct 2017)

Outcome Measure Data

Analysis Population Description

All randomized participants. Participants were included in the treatment group to which they were randomized.

Arm/Group Title	Pembrolizumab	Paclitaxel
Arm/Group Description	Participants receive 200 mg IV pembrolizumab on Day 1 of each 21-day cycle, for up to 35 administrations (approximately 2 years).	Participants receive paclitaxel 80 mg/m^2 IV, on Days 1, 8, and 15 of each 28-day cycle until disease progression or unacceptable toxicity.
Measure Participants	296	296
Median (95% Confidence Interval) [months]	1.6	3.2

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Pembrolizumab, Paclitaxel
	Comments	Treatment difference in PFS was assessed by the Stratified Log-rank test, and a stratified Cox proportional hazard model with Efron's method of tie handling was used to assess the magnitude of the treatment difference (HR) between the treatment arms.
	Type of Statistical Test	Other
	Comments
Statistical Test of Hypothesis	p-Value	0.97481
	Comments	One-sided p-value based on log-rank test stratified by geographic region (Europe/Israel/North America/Australia vs. Asia vs. Rest of World), TTP on first-line therapy (<6 months vs. ≥6 months), and PD-L1 status (positive vs. negative).
	Method	Regression, Cox
	Comments	P-value rounded to 5 decimal places.
Method of Estimation	Estimation Parameter	Hazard Ratio (HR)
	Estimated Value	1.19
	Confidence Interval	(2-Sided) 95%; 1.00 to 1.42
	Parameter Dispersion	Type: Value:
	Estimation Comments

7. Secondary Outcome

Title	PFS According to Immune-Related Response Evaluation Criteria in Solid Tumors (irRECIST) Based on BICR in PD-L1 Positive Participants
Description	PFS defined as time from randomization to first documented PD per irRECIST based on BICR, or death due to any cause, whichever occurs first. Following initial PD by RECIST 1.1 (20% relative increase in SOD of target lesions), participants were assessed according to irRECIST: tumor assessment was repeated ≥4 weeks later to confirm PD with the option of continuing treatment until this scan was obtained for clinically stable participants. If PD confirmed, participant was discontinued from treatment unless investigator determined benefit. If repeat scan indicated stable disease (SD; neither sufficient shrinkage or increase of target lesion), partial response (PR; ≥30% decrease in the SOD of target lesions), or complete response (CR; disappearance of all non-nodal target lesions), participant could continue treatment at investigator's discretion. PFS analyzed using Kaplan-Meier method and median PFS (95% CI) in months was reported for PD-L1 positive participants by treatment group.
Time Frame	Up to 30 months (through database cut-off date of 26 Oct 2017)

Outcome Measure Data

Analysis Population Description

All randomized PD-L1 positive participants. Participants were included in the treatment group to which they were randomized.

Arm/Group Title	Pembrolizumab	Paclitaxel
Arm/Group Description	Participants receive 200 mg IV pembrolizumab on Day 1 of each 21-day cycle, for up to 35 administrations (approximately 2 years).	Participants receive paclitaxel 80 mg/m^2 IV, on Days 1, 8, and 15 of each 28-day cycle until disease progression or unacceptable toxicity.
Measure Participants	196	199
Median (95% Confidence Interval) [months]	1.9	4.2

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Pembrolizumab, Paclitaxel
	Comments	Treatment difference in PFS was assessed by the Stratified Log-rank test, and a stratified Cox proportional hazard model with Efron's method of tie handling was used to assess the magnitude of the treatment difference (HR) between the treatment arms.
	Type of Statistical Test	Other
	Comments
Statistical Test of Hypothesis	p-Value	0.80696
	Comments	One-sided p-value based on log-rank test stratified by geographic region (Europe/Israel/North America/Australia vs. Asia vs. Rest of World) and TTP on first-line therapy (<6 months vs. ≥6 months).
	Method	Regression, Cox
	Comments	P-value rounded to 5 decimal places.
Method of Estimation	Estimation Parameter	Hazard Ratio (HR)
	Estimated Value	1.11
	Confidence Interval	(2-Sided) 95%; 0.89 to 1.38
	Parameter Dispersion	Type: Value:
	Estimation Comments

8. Secondary Outcome

Title	PFS According to irRECIST Based on BICR in All Participants
Description	PFS defined as time from randomization to first documented PD per irRECIST based on BICR, or death due to any cause, whichever occurs first. Following initial PD by RECIST 1.1 (20% relative increase in SOD of target lesions), participants were assessed according to irRECIST: tumor assessment was repeated ≥4 weeks later to confirm PD with the option of continuing treatment until this scan was obtained for clinically stable participants. If PD confirmed, participant was discontinued from treatment unless investigator determined benefit. If repeat scan indicated SD (neither sufficient shrinkage or increase of target lesion), PR (≥30% decrease in the SOD of target lesions), or CR (disappearance of all non-nodal target lesions), participant could continue treatment at investigator's discretion. PFS analyzed using Kaplan-Meier method and median PFS (95% CI) in months was reported for all participants by treatment group.
Time Frame	Up to 30 months (through database cut-off date of 26 Oct 2017)

Outcome Measure Data

Analysis Population Description

All randomized participants. Participants were included in the treatment group to which they were randomized.

Arm/Group Title	Pembrolizumab	Paclitaxel
Arm/Group Description	Participants receive 200 mg IV pembrolizumab on Day 1 of each 21-day cycle, for up to 35 administrations (approximately 2 years).	Participants receive paclitaxel 80 mg/m^2 IV, on Days 1, 8, and 15 of each 28-day cycle until disease progression or unacceptable toxicity.
Measure Participants	296	296
Median (95% Confidence Interval) [months]	1.6	4.2

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Pembrolizumab, Paclitaxel
	Comments	Treatment difference in PFS was assessed by the Stratified Log-rank test, and a stratified Cox proportional hazard model with Efron's method of tie handling was used to assess the magnitude of the treatment difference (HR) between the treatment arms.
	Type of Statistical Test	Other
	Comments
Statistical Test of Hypothesis	p-Value	0.99932
	Comments	One-sided p-value based on log-rank test stratified by geographic region (Europe/Israel/North America/Australia vs. Asia vs. Rest of World), TTP on first-line therapy (< 6 months vs. >= 6 months), and PD-L1 status (positive vs. negative).
	Method	Regression, Cox
	Comments	P-value rounded to 5 decimal places.
Method of Estimation	Estimation Parameter	Hazard Ratio (HR)
	Estimated Value	1.34
	Confidence Interval	(2-Sided) 95%; 1.12 to 1.60
	Parameter Dispersion	Type: Value:
	Estimation Comments

9. Secondary Outcome

Title	Time to Tumor Progression (TTP) According to RECIST 1.1 Based on BICR in PD-L1 Positive Participants
Description	TTP was defined as the time from randomization to the first documented disease progression per RECIST 1.1 based on BICR. Using RECIST 1.1, progressive disease was defined as a 20% relative increase in the SOD of target lesions, taking as reference the nadir SOD and an absolute increase of >5 mm in the SOD, or the appearance of new lesions. If there was no documented disease progression, TTP was censored at last tumor assessment date. TTP was analyzed using the Kaplan-Meier method and median TTP (95% CI) in months was reported for PD-L1 positive participants by treatment group.
Time Frame	Up to 30 months (through database cut-off date of 26 Oct 2017)

Outcome Measure Data

Analysis Population Description

All randomized PD-L1 positive participants. Participants were included in the treatment group to which they were randomized.

Arm/Group Title	Pembrolizumab	Paclitaxel
Arm/Group Description	Participants receive 200 mg IV pembrolizumab on Day 1 of each 21-day cycle, for up to 35 administrations (approximately 2 years).	Participants receive paclitaxel 80 mg/m^2 IV, on Days 1, 8, and 15 of each 28-day cycle until disease progression or unacceptable toxicity.
Measure Participants	196	199
Median (95% Confidence Interval) [months]	1.6	4.0

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Pembrolizumab, Paclitaxel
	Comments	Treatment difference in TTP was assessed by the Stratified Log-rank test, and a stratified Cox proportional hazard model with Efron's method of tie handling was used to assess the magnitude of the treatment difference (HR) between the treatment arms.
	Type of Statistical Test	Other
	Comments
Statistical Test of Hypothesis	p-Value	0.99661
	Comments	One-sided p-value based on log-rank test stratified by geographic region (Europe/Israel/North America/Australia vs. Asia vs. Rest of World) and TTP on first-line therapy (<6 months vs. ≥6 months).
	Method	Regression, Cox
	Comments	P-value rounded to 5 decimal places.
Method of Estimation	Estimation Parameter	Hazard Ratio (HR)
	Estimated Value	1.45
	Confidence Interval	(2-Sided) 95%; 1.11 to 1.89
	Parameter Dispersion	Type: Value:
	Estimation Comments

10. Secondary Outcome

Title	TTP According to RECIST 1.1 Based on BICR in All Participants
Description	TTP was defined as the time from randomization to the first documented disease progression per RECIST 1.1 based on BICR. Using RECIST 1.1, progressive disease was defined as a 20% relative increase in the SOD of target lesions, taking as reference the nadir SOD and an absolute increase of >5 mm in the SOD, or the appearance of new lesions. If there was no documented disease progression, TTP was censored at last tumor assessment date. TTP was analyzed using the Kaplan-Meier method and median TTP (95% CI) in months was reported for all participants by treatment group.
Time Frame	Up to 30 months (through database cut-off date of 26 Oct 2017)

Outcome Measure Data

Analysis Population Description

All randomized participants. Participants were included in the treatment group to which they were randomized.

Arm/Group Title	Pembrolizumab	Paclitaxel
Arm/Group Description	Participants receive 200 mg IV pembrolizumab on Day 1 of each 21-day cycle, for up to 35 administrations (approximately 2 years).	Participants receive paclitaxel 80 mg/m^2 IV, on Days 1, 8, and 15 of each 28-day cycle until disease progression or unacceptable toxicity.
Measure Participants	296	296
Median (95% Confidence Interval) [months]	1.5	4.1

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Pembrolizumab, Paclitaxel
	Comments	Treatment difference in TTP was assessed by the Stratified Log-rank test, and a stratified Cox proportional hazard model with Efron's method of tie handling was used to assess the magnitude of the treatment difference (HR) between the treatment arms.
	Type of Statistical Test	Other
	Comments
Statistical Test of Hypothesis	p-Value	1.00000
	Comments	One-sided p-value based on log-rank test stratified by geographic region (Europe/Israel/North America/Australia vs. Asia vs. Rest of World), TTP on first-line therapy (< 6 months vs. >= 6 months), and PD-L1 status (positive vs. negative).
	Method	Regression, Cox
	Comments	P-value rounded to 5 decimal places.
Method of Estimation	Estimation Parameter	Hazard Ratio (HR)
	Estimated Value	1.77
	Confidence Interval	(2-Sided) 95%; 1.42 to 2.20
	Parameter Dispersion	Type: Value:
	Estimation Comments

11. Secondary Outcome

Title	TTP According to RECIST 1.1 Based on Investigator Assessment in PD-L1 Positive Participants
Description	TTP was defined as the time from randomization to the first documented disease progression per RECIST 1.1 based on investigator assessment. Using RECIST 1.1, progressive disease was defined as a 20% relative increase in the SOD of target lesions, taking as reference the nadir SOD and an absolute increase of >5 mm in the SOD, or the appearance of new lesions. If there was no documented disease progression, TTP was censored at last tumor assessment date. TTP was analyzed using the Kaplan-Meier method and median TTP (95% CI) in months was reported for PD-L1 positive participants by treatment group.
Time Frame	Up to 30 months (through database cut-off date of 26 Oct 2017)

Outcome Measure Data

Analysis Population Description

All randomized PD-L1 positive participants. Participants were included in the treatment group to which they were randomized.

Arm/Group Title	Pembrolizumab	Paclitaxel
Arm/Group Description	Participants receive 200 mg IV pembrolizumab on Day 1 of each 21-day cycle, for up to 35 administrations (approximately 2 years).	Participants receive paclitaxel 80 mg/m^2 IV, on Days 1, 8, and 15 of each 28-day cycle until disease progression or unacceptable toxicity.
Measure Participants	196	199
Median (95% Confidence Interval) [months]	2.1	3.3

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Pembrolizumab, Paclitaxel
	Comments	Treatment difference in TTP was assessed by the Stratified Log-rank test, and a stratified Cox proportional hazard model with Efron's method of tie handling was used to assess the magnitude of the treatment difference (HR) between the treatment arms.
	Type of Statistical Test	Other
	Comments
Statistical Test of Hypothesis	p-Value	0.39280
	Comments	One-sided p-value based on log-rank test stratified by geographic region (Europe/Israel/North America/Australia vs. Asia vs. Rest of World) and TTP on first-line therapy (<6 months vs. ≥6 months).
	Method	Regression, Cox
	Comments	P-value rounded to 5 decimal places.
Method of Estimation	Estimation Parameter	Hazard Ratio (HR)
	Estimated Value	0.97
	Confidence Interval	(2-Sided) 95%; 0.77 to 1.23
	Parameter Dispersion	Type: Value:
	Estimation Comments

12. Secondary Outcome

Title	TTP According to RECIST 1.1 Based on Investigator Assessment in All Participants
Description	TTP was defined as the time from randomization to the first documented disease progression per RECIST 1.1 based on investigator assessment. Using RECIST 1.1, progressive disease was defined as a 20% relative increase in the SOD of target lesions, taking as reference the nadir SOD and an absolute increase of >5 mm in the SOD, or the appearance of new lesions. If there was no documented disease progression, TTP was censored at last tumor assessment date. TTP was analyzed using the Kaplan-Meier method and median TTP (95% CI) in months was reported for all participants by treatment group.
Time Frame	Up to 30 months (through database cut-off date of 26 Oct 2017)

Outcome Measure Data

Analysis Population Description

All randomized participants. Participants were included in the treatment group to which they were randomized.

Arm/Group Title	Pembrolizumab	Paclitaxel
Arm/Group Description	Participants receive 200 mg IV pembrolizumab on Day 1 of each 21-day cycle, for up to 35 administrations (approximately 2 years).	Participants receive paclitaxel 80 mg/m^2 IV, on Days 1, 8, and 15 of each 28-day cycle until disease progression or unacceptable toxicity.
Measure Participants	296	296
Median (95% Confidence Interval) [months]	1.6	3.8

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Pembrolizumab, Paclitaxel
	Comments	Treatment difference in TTP was assessed by the Stratified Log-rank test, and a stratified Cox proportional hazard model with Efron's method of tie handling was used to assess the magnitude of the treatment difference (HR) between the treatment arms.
	Type of Statistical Test	Other
	Comments
Statistical Test of Hypothesis	p-Value	0.97033
	Comments	One-sided p-value based on log-rank test stratified by geographic region (Europe/Israel/North America/Australia vs. Asia vs. Rest of World), TTP on first-line therapy (< 6 months vs. >= 6 months), and PD-L1 status (positive vs. negative).
	Method	Regression, Cox
	Comments	P-value rounded to 5 decimal places.
Method of Estimation	Estimation Parameter	Hazard Ratio (HR)
	Estimated Value	1.21
	Confidence Interval	(2-Sided) 95%; 1.00 to 1.47
	Parameter Dispersion	Type: Value:
	Estimation Comments

13. Secondary Outcome

Title	Objective Response Rate (ORR) According to RECIST 1.1 Based on BICR in PD-L1 Positive Participants
Description	ORR was defined as the percentage of the participants in the analysis population who had a confirmed CR (disappearance of all non-nodal target lesions and any pathological lymph nodes must have become normal) or PR (at least a 30% decrease in the SOD of target lesions, taking as reference the baseline SOD) according to RECIST 1.1 and based on BICR. ORR was analyzed using the stratified Miettinen and Nurminen method, and reported with 95% CI for PD-L1 positive participants by treatment group.
Time Frame	Up to 30 months (through database cut-off date of 26 Oct 2017)

Outcome Measure Data

Analysis Population Description

All randomized PD-L1 positive participants. Participants were included in the treatment group to which they were randomized.

Arm/Group Title	Pembrolizumab	Paclitaxel
Arm/Group Description	Participants receive 200 mg IV pembrolizumab on Day 1 of each 21-day cycle, for up to 35 administrations (approximately 2 years).	Participants receive paclitaxel 80 mg/m^2 IV, on Days 1, 8, and 15 of each 28-day cycle until disease progression or unacceptable toxicity.
Measure Participants	196	199
Number (95% Confidence Interval) [Percentage of participants]	15.8 5.3%	13.6 4.6%

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Pembrolizumab, Paclitaxel
	Comments	Stratified Miettinen and Nurminen's (MN) method was used for comparison of the ORR between the treatment arms. A 95% CI for the difference in response rates between the pembrolizumab arm and paclitaxel arm was provided.
	Type of Statistical Test	Other
	Comments
Statistical Test of Hypothesis	p-Value	0.28967
	Comments	Stratification factors for MN method included geographic region (Europe/Israel/North America/Australia vs. Asia vs. Rest of World) and TTP on first-line therapy (<6 months vs. ≥6 months) weighting by sample size.
	Method	Miettinen and Nurminen method
	Comments	P-value rounded to 5 decimal places.
Method of Estimation	Estimation Parameter	Difference in Percentage
	Estimated Value	2.0
	Confidence Interval	(2-Sided) 95%; -5.0 to 9.1
	Parameter Dispersion	Type: Value:
	Estimation Comments

14. Secondary Outcome

Title	ORR According to RECIST 1.1 Based on BICR in All Participants
Description	ORR was defined as the percentage of the participants in the analysis population who had a confirmed CR (disappearance of all non-nodal target lesions and any pathological lymph nodes must have become normal) or PR (at least a 30% decrease in the SOD of target lesions, taking as reference the baseline SOD) according to RECIST 1.1 and based on BICR. ORR was analyzed using the stratified Miettinen and Nurminen method, and reported with 95% CI for all participants by treatment group.
Time Frame	Up to 30 months (through database cut-off date of 26 Oct 2017)

Outcome Measure Data

Analysis Population Description

All randomized participants. Participants were included in the treatment group to which they were randomized.

Arm/Group Title	Pembrolizumab	Paclitaxel
Arm/Group Description	Participants receive 200 mg IV pembrolizumab on Day 1 of each 21-day cycle, for up to 35 administrations (approximately 2 years).	Participants receive paclitaxel 80 mg/m^2 IV, on Days 1, 8, and 15 of each 28-day cycle until disease progression or unacceptable toxicity.
Measure Participants	296	296
Number (95% Confidence Interval) [Percentage of participants]	11.1 3.8%	12.5 4.2%

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Pembrolizumab, Paclitaxel
	Comments	Stratified MN method was used for comparison of the ORR between the treatment arms. A 95% CI for the difference in response rates between the pembrolizumab arm and paclitaxel arm was provided.
	Type of Statistical Test	Other
	Comments
Statistical Test of Hypothesis	p-Value	0.69010
	Comments	Stratification factors for MN method included geographic region (Europe/Israel/North America/Australia vs. Asia vs. Rest of World), TTP on first-line therapy (<6 months vs. ≥6 months), and PD-L1 status (positive vs. negative) weighting by sample size
	Method	Miettinen and Nurminen method
	Comments	P-value rounded to 5 decimal places.
Method of Estimation	Estimation Parameter	Difference in Percentage
	Estimated Value	-1.3
	Confidence Interval	(2-Sided) 95%; -6.5 to 4.0
	Parameter Dispersion	Type: Value:
	Estimation Comments

15. Secondary Outcome

Title	ORR According to RECIST 1.1 Based on Investigator Assessment in PD-L1 Positive Participants
Description	ORR was defined as the percentage of the participants in the analysis population who had a confirmed CR (disappearance of all non-nodal target lesions and any pathological lymph nodes must have become normal) or PR (at least a 30% decrease in the SOD of target lesions, taking as reference the baseline SOD) according to RECIST 1.1 and based on investigator assessment. ORR was analyzed using the stratified Miettinen and Nurminen method, and reported with 95% CI for PD-L1 positive participants by treatment group.
Time Frame	Up to 30 months (through database cut-off date of 26 Oct 2017)

Outcome Measure Data

Analysis Population Description

All randomized PD-L1 positive participants. Participants were included in the treatment group to which they were randomized.

Arm/Group Title	Pembrolizumab	Paclitaxel
Arm/Group Description	Participants receive 200 mg IV pembrolizumab on Day 1 of each 21-day cycle, for up to 35 administrations (approximately 2 years).	Participants receive paclitaxel 80 mg/m^2 IV, on Days 1, 8, and 15 of each 28-day cycle until disease progression or unacceptable toxicity.
Measure Participants	196	199
Number (95% Confidence Interval) [Percentage of participants]	17.3 5.8%	15.6 5.3%

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Pembrolizumab, Paclitaxel
	Comments	Stratified MN method was used for comparison of the ORR between the treatment arms. A 95% CI for the difference in response rates between the pembrolizumab arm and paclitaxel arm was provided.
	Type of Statistical Test	Other
	Comments
Statistical Test of Hypothesis	p-Value	0.33220
	Comments	Stratification factors for MN method included geographic region (Europe/Israel/North America/Australia vs. Asia vs. Rest of World) and TTP on first-line therapy (<6 months vs. ≥6 months), weighting by sample size.
	Method	Miettinen and Nurminen method
	Comments	P-value rounded to 5 decimal places.
Method of Estimation	Estimation Parameter	Difference in Percentage
	Estimated Value	1.6
	Confidence Interval	(2-Sided) 95%; -5.8 to 9.1
	Parameter Dispersion	Type: Value:
	Estimation Comments

16. Secondary Outcome

Title	ORR According to RECIST 1.1 Based on Investigator Assessment in All Participants
Description	ORR was defined as the percentage of the participants in the analysis population who had a confirmed CR (disappearance of all non-nodal target lesions and any pathological lymph nodes must have become normal) or PR (at least a 30% decrease in the SOD of target lesions, taking as reference the baseline SOD) according to RECIST 1.1 and based on investigator assessment. ORR was analyzed using the stratified Miettinen and Nurminen method, and reported with 95% CI for all participants by treatment group.
Time Frame	Up to 30 months (through database cut-off date of 26 Oct 2017)

Outcome Measure Data

Analysis Population Description

All randomized participants. Participants were included in the treatment group to which they were randomized.

Arm/Group Title	Pembrolizumab	Paclitaxel
Arm/Group Description	Participants receive 200 mg IV pembrolizumab on Day 1 of each 21-day cycle, for up to 35 administrations (approximately 2 years).	Participants receive paclitaxel 80 mg/m^2 IV, on Days 1, 8, and 15 of each 28-day cycle until disease progression or unacceptable toxicity.
Measure Participants	296	296
Number (95% Confidence Interval) [Percentage of participants]	12.2 4.1%	15.2 5.1%

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Pembrolizumab, Paclitaxel
	Comments	Stratified Miettinen and Nurminen's method was used for comparison of the ORR between the treatment arms. A 95% CI for the difference in response rates between the pembrolizumab arm and paclitaxel arm is provided.
	Type of Statistical Test	Other
	Comments
Statistical Test of Hypothesis	p-Value	0.85922
	Comments	Stratification factors included geographic region (Europe/Israel/North America/Australia vs. Asia vs. Rest of World), TTP on first-line therapy (<6 months vs. ≥6 months), and PD-L1 status (positive vs. negative) weighting by sample size.
	Method	Miettinen and Nurminen method
	Comments	P-value rounded to 5 decimal places.
Method of Estimation	Estimation Parameter	Difference in Percentage
	Estimated Value	-3.0
	Confidence Interval	(2-Sided) 95%; -8.5 to 2.6
	Parameter Dispersion	Type: Value:
	Estimation Comments

17. Secondary Outcome

Title	Duration of Response (DOR) According to RECIST 1.1 Based on BICR in PD-L1 Positive Participants
Description	For PD-L1 positive participants who demonstrated CR (disappearance of all non-nodal target lesions and any pathological lymph nodes must have become normal) or PR (at least a 30% decrease in the SOD of target lesions, taking as reference the baseline SOD) according to RECIST 1.1 and based on BICR, DOR was defined as the time from first documented evidence of CR or PR until disease progression or death due to any cause, whichever occurs first. DOR for participants who had not progressed or died at the time of analysis was censored at the date of their last tumor assessment. DOR was analyzed using the Kaplan-Meier method and median DOR (range) in months was reported for PD-L1 positive participants with response by treatment group.
Time Frame	Up to 30 months (through database cut-off date of 26 Oct 2017)

Outcome Measure Data

Analysis Population Description

The subset of all randomized PD-L1 positive participants that showed a CR or PR according to RECIST 1.1 and based on BICR. Participants were included in the treatment group to which they were randomized.

Arm/Group Title	Pembrolizumab	Paclitaxel
Arm/Group Description	Participants receive 200 mg IV pembrolizumab on Day 1 of each 21-day cycle, for up to 35 administrations (approximately 2 years).	Participants receive paclitaxel 80 mg/m^2 IV, on Days 1, 8, and 15 of each 28-day cycle until disease progression or unacceptable toxicity.
Measure Participants	31	27
Median (Full Range) [months]	18.0	5.2

18. Secondary Outcome

Title	DOR According to RECIST 1.1 Based on BICR in All Participants
Description	For participants who demonstrated CR (disappearance of all non-nodal target lesions and any pathological lymph nodes must have become normal) or PR (at least a 30% decrease in the SOD of target lesions, taking as reference the baseline SOD) according to RECIST 1.1 and based on BICR, DOR was defined as the time from first documented evidence of CR or PR until disease progression or death due to any cause, whichever occurs first. DOR for participants who had not progressed or died at the time of analysis was censored at the date of their last tumor assessment. DOR was analyzed using the Kaplan-Meier method and median DOR (range) in months was reported for all participants with response by treatment group.
Time Frame	Up to 30 months (through database cut-off date of 26 Oct 2017)

Outcome Measure Data

Analysis Population Description

The subset of all randomized participants that showed a CR or PR according to RECIST 1.1 and based on BICR. Participants were included in the treatment group to which they were randomized.

Arm/Group Title	Pembrolizumab	Paclitaxel
Arm/Group Description	Participants receive 200 mg IV pembrolizumab on Day 1 of each 21-day cycle, for up to 35 administrations (approximately 2 years).	Participants receive paclitaxel 80 mg/m^2 IV, on Days 1, 8, and 15 of each 28-day cycle until disease progression or unacceptable toxicity.
Measure Participants	33	37
Median (Full Range) [months]	18.0	5.5

19. Secondary Outcome

Title	DOR According to RECIST 1.1 Based on Investigator Assessment in PD-L1 Positive Participants
Description	For PD-L1 positive participants who demonstrated CR (disappearance of all non-nodal target lesions and any pathological lymph nodes must have become normal) or PR (at least a 30% decrease in the SOD of target lesions, taking as reference the baseline SOD) according to RECIST 1.1 and based on investigator assessment, DOR was defined as the time from first documented evidence of CR or PR until disease progression or death due to any cause, whichever occurs first. DOR for participants who had not progressed or died at the time of analysis was censored at the date of their last tumor assessment. DOR was analyzed using the Kaplan-Meier method and median DOR (range) in months was reported for PD-L1 positive participants with response by treatment group.
Time Frame	Up to 30 months (through database cut-off date of 26 Oct 2017)

Outcome Measure Data

Analysis Population Description

The subset of all randomized PD-L1 positive participants that showed a CR or PR according to RECIST 1.1 and based on investigator assessment. Participants were included in the treatment group to which they were randomized.

Arm/Group Title	Pembrolizumab	Paclitaxel
Arm/Group Description	Participants receive 200 mg IV pembrolizumab on Day 1 of each 21-day cycle, for up to 35 administrations (approximately 2 years).	Participants receive paclitaxel 80 mg/m^2 IV, on Days 1, 8, and 15 of each 28-day cycle until disease progression or unacceptable toxicity.
Measure Participants	34	31
Median (Full Range) [months]	15.7	4.3

20. Secondary Outcome

Title	DOR According to RECIST 1.1 Based on Investigator Assessment in All Participants
Description	For participants who demonstrated CR (disappearance of all non-nodal target lesions and any pathological lymph nodes must have become normal) or PR (at least a 30% decrease in the SOD of target lesions, taking as reference the baseline SOD) according to RECIST 1.1 and based on investigator assessment, DOR was defined as the time from first documented evidence of CR or PR until disease progression or death due to any cause, whichever occurs first. DOR for participants who had not progressed or died at the time of analysis was censored at the date of their last tumor assessment. DOR was analyzed using the Kaplan-Meier method and median DOR (range) in months was reported for all participants with response by treatment group.
Time Frame	Up to 30 months (through database cut-off date of 26 Oct 2017)

Outcome Measure Data

Analysis Population Description

The subset of all randomized participants that showed a CR or PR according to RECIST 1.1 and based on investigator assessment. Participants were included in the treatment group to which they were randomized.

Arm/Group Title	Pembrolizumab	Paclitaxel
Arm/Group Description	Participants receive 200 mg IV pembrolizumab on Day 1 of each 21-day cycle, for up to 35 administrations (approximately 2 years).	Participants receive paclitaxel 80 mg/m^2 IV, on Days 1, 8, and 15 of each 28-day cycle until disease progression or unacceptable toxicity.
Measure Participants	36	45
Median (Full Range) [months]	15.7	4.3

21. Secondary Outcome

Title	Percentage of PD-L1 Positive Participants Who Experienced an Adverse Event (AE)
Description	An AE was defined as any untoward medical occurrence in a participant administered a pharmaceutical product and which does not necessarily have to have a causal relationship with this treatment. An AE could therefore be any unfavourable and unintended sign, symptom, or disease temporally associated with the use of a medicinal product or protocol-specified procedure, whether or not considered related to the medicinal product or protocol-specified procedure. Any worsening of a pre-existing condition that was temporally associated with the use of the Sponsor's product was also an AE. The percentage of participants with at least one AE was reported for PD-L1 positive participants by treatment group.
Time Frame	Up to 71 months (through database cut-off date of 10 Jun 2021)

Outcome Measure Data

Analysis Population Description

All randomized PD-L1 positive participants who received at least 1 dose of study treatment.

Arm/Group Title	Pembrolizumab	Paclitaxel
Arm/Group Description	Participants receive 200 mg IV pembrolizumab on Day 1 of each 21-day cycle, for up to 35 administrations (approximately 2 years).	Participants receive paclitaxel 80 mg/m^2 IV, on Days 1, 8, and 15 of each 28-day cycle until disease progression or unacceptable toxicity.
Measure Participants	194	188
Number [Percentage of participants]	93.3 31.5%	97.3 32.9%

22. Secondary Outcome

Title	Percentage of All Participants Who Experienced an AE
Description	An AE was defined as any untoward medical occurrence in a participant administered a pharmaceutical product and which does not necessarily have to have a causal relationship with this treatment. An AE could therefore be any unfavourable and unintended sign, symptom, or disease temporally associated with the use of a medicinal product or protocol-specified procedure, whether or not considered related to the medicinal product or protocol-specified procedure. Any worsening of a pre-existing condition that was temporally associated with the use of the Sponsor's product was also an AE. The percentage of participants with at least one AE was reported for all participants by treatment group.
Time Frame	Up to 71 months (through database cut-off date of 10 Jun 2021)

Outcome Measure Data

Analysis Population Description

All randomized participants who received at least 1 dose of study treatment.

Arm/Group Title	Pembrolizumab	Paclitaxel
Arm/Group Description	Participants receive 200 mg IV pembrolizumab on Day 1 of each 21-day cycle, for up to 35 administrations (approximately 2 years).	Participants receive paclitaxel 80 mg/m^2 IV, on Days 1, 8, and 15 of each 28-day cycle until disease progression or unacceptable toxicity.
Measure Participants	294	276
Number [Percentage of participants]	93.9 31.7%	97.1 32.8%

Statistical Analysis 1

Statistical Analysis Overview	Comparison Group Selection	Pembrolizumab, Paclitaxel
	Comments	Between-treatment differences (Pembrolizumab vs. Paclitaxel) in the percentage of participants with events and accompanying 95% confidence intervals were based on the Miettinen and Nurminen method. Negative values correspond to a greater percentage of events for Paclitaxel.
	Type of Statistical Test	Other
	Comments
Statistical Test of Hypothesis	p-Value
	Comments
	Method
	Comments
Method of Estimation	Estimation Parameter	Difference in Percentage
	Estimated Value	-3.2
	Confidence Interval	(2-Sided) 95%; -6.9 to 0.2
	Parameter Dispersion	Type: Value:
	Estimation Comments

23. Secondary Outcome

Title	Percentage of PD-L1 Positive Participants That Discontinued Study Treatment Due to AE
Description	An AE was defined as any untoward medical occurrence in a participant administered a pharmaceutical product and which does not necessarily have to have a causal relationship with this treatment. An AE could therefore be any unfavourable and unintended sign, symptom, or disease temporally associated with the use of a medicinal product or protocol-specified procedure, whether or not considered related to the medicinal product or protocol-specified procedure. Any worsening of a pre-existing condition that was temporally associated with the use of the Sponsor's product was also an AE. The percentage of participants that discontinued study treatment due to an AE was reported for PD-L1 positive participants by treatment group.
Time Frame	Up to approximately 26.4 months

Outcome Measure Data

Analysis Population Description

All randomized PD-L1 positive participants who received at least 1 dose of study treatment.

Arm/Group Title	Pembrolizumab	Paclitaxel
Arm/Group Description	Participants receive 200 mg IV pembrolizumab on Day 1 of each 21-day cycle, for up to 35 administrations (approximately 2 years).	Participants receive paclitaxel 80 mg/m^2 IV, on Days 1, 8, and 15 of each 28-day cycle until disease progression or unacceptable toxicity.
Measure Participants	194	188
Number [Percentage of participants]	4.1 1.4%	8.0 2.7%

24. Secondary Outcome

Title	Percentage of All Participants That Discontinued Study Treatment Due to AE
Description	An AE was defined as any untoward medical occurrence in a participant administered a pharmaceutical product and which does not necessarily have to have a causal relationship with this treatment. An AE could therefore be any unfavourable and unintended sign, symptom, or disease temporally associated with the use of a medicinal product or protocol-specified procedure, whether or not considered related to the medicinal product or protocol-specified procedure. Any worsening of a pre-existing condition that was temporally associated with the use of the Sponsor's product was also an AE. The percentage of participants that discontinued study treatment due to an AE was reported for all participants by treatment group.
Time Frame	Up to approximately 26.4 months

Outcome Measure Data

Analysis Population Description

All randomized participants who received at least 1 dose of study treatment.

Arm/Group Title	Pembrolizumab	Paclitaxel
Arm/Group Description	Participants receive 200 mg IV pembrolizumab on Day 1 of each 21-day cycle, for up to 35 administrations (approximately 2 years).	Participants receive paclitaxel 80 mg/m^2 IV, on Days 1, 8, and 15 of each 28-day cycle until disease progression or unacceptable toxicity.
Measure Participants	294	276
Number [Percentage of participants]	4.8 1.6%	9.1 3.1%

Adverse Events

Time Frame	Up to 71 months (through database cut-off date of 10 Jun 2021)
Adverse Event Reporting Description	All-Cause Mortality was reported for all randomized participants. Serious AEs and Other AEs were reported for the Safety Population: all randomized participants who received at least 1 dose of study treatment. Per protocol, disease progression of cancer on study was not considered an AE unless considered related to study drug. Therefore, MedDRA preferred terms "Neoplasm progression", "Malignant neoplasm progression" and "Disease progression" not related to study drug are excluded as AEs.
Arm/Group Title	Pembrolizumab First Course		Paclitaxel		Pembrolizumab Second Course
Arm/Group Description	Participants receive 200 mg IV pembrolizumab on Day 1 of each 21-day cycle, for up to 35 administrations (approximately 2 years).		Participants receive paclitaxel 80 mg/m^2 IV, on Days 1, 8, and 15 of each 28-day cycle until disease progression or unacceptable toxicity.		Qualified participants who received pembrolizumab as a first course and stopped the first course of pembrolizumab due to complete response (CR) or completed the first course of pembrolizumab and had stable disease but progressed after discontinuation, initiated a second course of pembrolizumab at the investigator's discretion for up to 17 cycles (approximately 1 year additional).
All Cause Mortality
	Pembrolizumab First Course		Paclitaxel		Pembrolizumab Second Course
	Affected / at Risk (%)	# Events	Affected / at Risk (%)	# Events	Affected / at Risk (%)	# Events
Total	278/296 (93.9%)		287/296 (97%)		2/3 (66.7%)
Serious Adverse Events
	Pembrolizumab First Course		Paclitaxel		Pembrolizumab Second Course
	Affected / at Risk (%)	# Events	Affected / at Risk (%)	# Events	Affected / at Risk (%)	# Events
Total	108/294 (36.7%)		68/276 (24.6%)		1/3 (33.3%)
Blood and lymphatic system disorders
Anaemia	11/294 (3.7%)	14	5/276 (1.8%)	6	0/3 (0%)	0
Disseminated intravascular coagulation	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Febrile neutropenia	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Microcytic anaemia	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Neutropenia	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Splenic infarction	0/294 (0%)	0	0/276 (0%)	0	1/3 (33.3%)	1
Cardiac disorders
Atrial fibrillation	1/294 (0.3%)	1	1/276 (0.4%)	1	0/3 (0%)	0
Myocardial infarction	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Endocrine disorders
Addison's disease	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Hypophysitis	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Hypopituitarism	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Hypothyroidism	2/294 (0.7%)	2	0/276 (0%)	0	0/3 (0%)	0
Eye disorders
Cataract	0/294 (0%)	0	1/276 (0.4%)	2	0/3 (0%)	0
Retinal detachment	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Gastrointestinal disorders
Abdominal adhesions	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Abdominal pain	8/294 (2.7%)	9	4/276 (1.4%)	4	1/3 (33.3%)	1
Abdominal pain lower	1/294 (0.3%)	1	1/276 (0.4%)	1	0/3 (0%)	0
Abdominal pain upper	4/294 (1.4%)	4	0/276 (0%)	0	0/3 (0%)	0
Ascites	4/294 (1.4%)	4	2/276 (0.7%)	2	0/3 (0%)	0
Colitis	1/294 (0.3%)	1	3/276 (1.1%)	3	0/3 (0%)	0
Constipation	5/294 (1.7%)	5	4/276 (1.4%)	4	0/3 (0%)	0
Duodenal stenosis	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Dyspepsia	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Dysphagia	2/294 (0.7%)	2	2/276 (0.7%)	2	0/3 (0%)	0
Gastric haemorrhage	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Gastric hypomotility	0/294 (0%)	0	1/276 (0.4%)	2	0/3 (0%)	0
Gastric stenosis	2/294 (0.7%)	2	0/276 (0%)	0	0/3 (0%)	0
Gastrointestinal haemorrhage	2/294 (0.7%)	2	0/276 (0%)	0	0/3 (0%)	0
Gastrointestinal perforation	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Gastrosplenic fistula	0/294 (0%)	0	0/276 (0%)	0	1/3 (33.3%)	1
Haematemesis	0/294 (0%)	0	2/276 (0.7%)	2	0/3 (0%)	0
Ileal perforation	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Ileus	4/294 (1.4%)	4	1/276 (0.4%)	1	0/3 (0%)	0
Impaired gastric emptying	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Intestinal obstruction	1/294 (0.3%)	1	4/276 (1.4%)	7	0/3 (0%)	0
Large intestinal obstruction	1/294 (0.3%)	1	1/276 (0.4%)	1	0/3 (0%)	0
Large intestine perforation	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Mechanical ileus	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Melaena	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Nausea	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Obstruction gastric	0/294 (0%)	0	2/276 (0.7%)	2	0/3 (0%)	0
Oesophageal obstruction	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Oesophageal stenosis	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Pancreatic failure	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Pancreatitis	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Rectal haemorrhage	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Small intestinal obstruction	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Subileus	1/294 (0.3%)	1	1/276 (0.4%)	1	0/3 (0%)	0
Upper gastrointestinal haemorrhage	3/294 (1%)	3	1/276 (0.4%)	1	0/3 (0%)	0
Vomiting	6/294 (2%)	6	2/276 (0.7%)	2	0/3 (0%)	0
General disorders
Asthenia	2/294 (0.7%)	2	1/276 (0.4%)	1	0/3 (0%)	0
Complication associated with device	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Death	1/294 (0.3%)	1	1/276 (0.4%)	1	0/3 (0%)	0
Fatigue	2/294 (0.7%)	2	0/276 (0%)	0	0/3 (0%)	0
General physical health deterioration	3/294 (1%)	3	0/276 (0%)	0	0/3 (0%)	0
Influenza like illness	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Oedema peripheral	2/294 (0.7%)	2	0/276 (0%)	0	0/3 (0%)	0
Pyrexia	5/294 (1.7%)	7	4/276 (1.4%)	4	0/3 (0%)	0
Hepatobiliary disorders
Autoimmune hepatitis	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Bile duct stenosis	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Biliary obstruction	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Cholangitis	1/294 (0.3%)	1	1/276 (0.4%)	1	0/3 (0%)	0
Cholecystitis acute	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Cholelithiasis	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Hepatitis	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Hyperbilirubinaemia	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Jaundice cholestatic	2/294 (0.7%)	2	1/276 (0.4%)	1	0/3 (0%)	0
Immune system disorders
Hypersensitivity	0/294 (0%)	0	1/276 (0.4%)	3	0/3 (0%)	0
Infections and infestations
Aspergillus infection	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Bacteraemia	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Bacterial sepsis	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Biliary sepsis	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Enterobacter infection	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Escherichia infection	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Escherichia urinary tract infection	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Gastroenteritis	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Infection	1/294 (0.3%)	1	1/276 (0.4%)	1	0/3 (0%)	0
Infective exacerbation of chronic obstructive airways disease	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Influenza	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Lower respiratory tract infection	1/294 (0.3%)	1	3/276 (1.1%)	3	0/3 (0%)	0
Medical device site infection	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Otitis media	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Pneumonia	9/294 (3.1%)	9	7/276 (2.5%)	8	0/3 (0%)	0
Pneumonia bacterial	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Pyelonephritis	1/294 (0.3%)	2	0/276 (0%)	0	0/3 (0%)	0
Respiratory tract infection	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Sepsis	2/294 (0.7%)	2	3/276 (1.1%)	3	0/3 (0%)	0
Septic shock	1/294 (0.3%)	1	1/276 (0.4%)	1	0/3 (0%)	0
Urinary tract infection	1/294 (0.3%)	1	1/276 (0.4%)	1	0/3 (0%)	0
Injury, poisoning and procedural complications
Contusion	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Femoral neck fracture	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Hip fracture	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Humerus fracture	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Muscle strain	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Skin wound	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Toxicity to various agents	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Urinary tract stoma complication	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Wrist fracture	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Investigations
Alanine aminotransferase increased	2/294 (0.7%)	2	1/276 (0.4%)	1	0/3 (0%)	0
Aspartate aminotransferase increased	2/294 (0.7%)	2	1/276 (0.4%)	1	0/3 (0%)	0
Blood bilirubin increased	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Weight decreased	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Metabolism and nutrition disorders
Decreased appetite	2/294 (0.7%)	2	0/276 (0%)	0	0/3 (0%)	0
Dehydration	4/294 (1.4%)	4	1/276 (0.4%)	1	0/3 (0%)	0
Hypercalcaemia	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Hyperglycaemia	1/294 (0.3%)	1	1/276 (0.4%)	1	0/3 (0%)	0
Hypoglycaemia	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Hyponatraemia	2/294 (0.7%)	2	0/276 (0%)	0	0/3 (0%)	0
Type 2 diabetes mellitus	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Musculoskeletal and connective tissue disorders
Back pain	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Muscle disorder	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Neoplasms benign, malignant and unspecified (incl cysts and polyps)
Metastases to meninges	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Tumour pain	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Nervous system disorders
Cerebral infarction	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Cerebral ischaemia	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Depressed level of consciousness	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Epilepsy	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Intracranial mass	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Product Issues
Device dislocation	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Device occlusion	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Psychiatric disorders
Completed suicide	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Confusional state	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Panic disorder	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Renal and urinary disorders
Acute kidney injury	3/294 (1%)	3	0/276 (0%)	0	0/3 (0%)	0
Chronic kidney disease	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Dysuria	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Haematuria	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Hydronephrosis	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Nephrolithiasis	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Renal failure	1/294 (0.3%)	1	1/276 (0.4%)	1	0/3 (0%)	0
Renal injury	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Urinary retention	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Urinary tract obstruction	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Respiratory, thoracic and mediastinal disorders
Aspiration	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Cough	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Dyspnoea	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Dyspnoea paroxysmal nocturnal	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Interstitial lung disease	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Pleural effusion	3/294 (1%)	3	1/276 (0.4%)	1	0/3 (0%)	0
Pneumonia aspiration	3/294 (1%)	3	2/276 (0.7%)	3	0/3 (0%)	0
Pneumonitis	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Pneumothorax	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Pulmonary embolism	1/294 (0.3%)	1	4/276 (1.4%)	4	0/3 (0%)	0
Respiratory failure	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Vascular disorders
Deep vein thrombosis	1/294 (0.3%)	1	1/276 (0.4%)	1	0/3 (0%)	0
Shock haemorrhagic	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Subclavian vein thrombosis	1/294 (0.3%)	1	0/276 (0%)	0	0/3 (0%)	0
Vascular compression	0/294 (0%)	0	1/276 (0.4%)	1	0/3 (0%)	0
Other (Not Including Serious) Adverse Events
	Pembrolizumab First Course		Paclitaxel		Pembrolizumab Second Course
	Affected / at Risk (%)	# Events	Affected / at Risk (%)	# Events	Affected / at Risk (%)	# Events
Total	256/294 (87.1%)		260/276 (94.2%)		3/3 (100%)
Blood and lymphatic system disorders
Anaemia	44/294 (15%)	56	68/276 (24.6%)	86	1/3 (33.3%)	1
Neutropenia	5/294 (1.7%)	5	31/276 (11.2%)	62	0/3 (0%)	0
Endocrine disorders
Hypothyroidism	22/294 (7.5%)	25	1/276 (0.4%)	1	0/3 (0%)	0
Gastrointestinal disorders
Abdominal distension	17/294 (5.8%)	17	12/276 (4.3%)	12	0/3 (0%)	0
Abdominal pain	40/294 (13.6%)	42	49/276 (17.8%)	57	0/3 (0%)	0
Abdominal pain upper	17/294 (5.8%)	21	16/276 (5.8%)	17	1/3 (33.3%)	1
Ascites	19/294 (6.5%)	19	12/276 (4.3%)	14	0/3 (0%)	0
Constipation	56/294 (19%)	62	53/276 (19.2%)	71	0/3 (0%)	0
Diarrhoea	40/294 (13.6%)	42	72/276 (26.1%)	104	1/3 (33.3%)	2
Dyspepsia	18/294 (6.1%)	18	11/276 (4%)	13	0/3 (0%)	0
Dysphagia	19/294 (6.5%)	20	17/276 (6.2%)	22	0/3 (0%)	0
Melaena	6/294 (2%)	6	1/276 (0.4%)	1	1/3 (33.3%)	1
Nausea	65/294 (22.1%)	75	77/276 (27.9%)	119	1/3 (33.3%)	1
Oesophageal pain	2/294 (0.7%)	2	1/276 (0.4%)	1	1/3 (33.3%)	1
Stomatitis	7/294 (2.4%)	7	19/276 (6.9%)	22	0/3 (0%)	0
Vomiting	49/294 (16.7%)	59	47/276 (17%)	65	1/3 (33.3%)	1
General disorders
Asthenia	30/294 (10.2%)	41	38/276 (13.8%)	63	1/3 (33.3%)	1
Fatigue	77/294 (26.2%)	95	89/276 (32.2%)	132	1/3 (33.3%)	1
Mucosal inflammation	2/294 (0.7%)	4	16/276 (5.8%)	19	0/3 (0%)	0
Oedema peripheral	24/294 (8.2%)	25	29/276 (10.5%)	37	0/3 (0%)	0
Pyrexia	32/294 (10.9%)	40	30/276 (10.9%)	38	2/3 (66.7%)	2
Immune system disorders
Seasonal allergy	0/294 (0%)	0	0/276 (0%)	0	1/3 (33.3%)	1
Infections and infestations
Nasopharyngitis	7/294 (2.4%)	9	17/276 (6.2%)	21	0/3 (0%)	0
Skin infection	1/294 (0.3%)	1	1/276 (0.4%)	1	1/3 (33.3%)	2
Investigations
Alanine aminotransferase increased	18/294 (6.1%)	19	18/276 (6.5%)	26	0/3 (0%)	0
Aspartate aminotransferase increased	24/294 (8.2%)	26	16/276 (5.8%)	22	0/3 (0%)	0
Blood alkaline phosphatase increased	21/294 (7.1%)	21	9/276 (3.3%)	12	0/3 (0%)	0
Neutrophil count decreased	2/294 (0.7%)	5	36/276 (13%)	70	0/3 (0%)	0
Weight decreased	22/294 (7.5%)	23	17/276 (6.2%)	21	1/3 (33.3%)	1
White blood cell count decreased	3/294 (1%)	6	19/276 (6.9%)	45	0/3 (0%)	0
Metabolism and nutrition disorders
Decreased appetite	76/294 (25.9%)	83	79/276 (28.6%)	99	0/3 (0%)	0
Hyperglycaemia	7/294 (2.4%)	8	2/276 (0.7%)	21	1/3 (33.3%)	1
Hypoalbuminaemia	19/294 (6.5%)	20	6/276 (2.2%)	7	0/3 (0%)	0
Hypokalaemia	15/294 (5.1%)	17	10/276 (3.6%)	11	0/3 (0%)	0
Musculoskeletal and connective tissue disorders
Arthralgia	23/294 (7.8%)	29	24/276 (8.7%)	46	0/3 (0%)	0
Back pain	33/294 (11.2%)	37	23/276 (8.3%)	29	0/3 (0%)	0
Groin pain	1/294 (0.3%)	2	1/276 (0.4%)	1	1/3 (33.3%)	1
Myalgia	9/294 (3.1%)	10	25/276 (9.1%)	31	0/3 (0%)	0
Nervous system disorders
Dizziness	14/294 (4.8%)	15	15/276 (5.4%)	18	0/3 (0%)	0
Headache	10/294 (3.4%)	12	15/276 (5.4%)	20	0/3 (0%)	0
Neuropathy peripheral	8/294 (2.7%)	8	42/276 (15.2%)	45	0/3 (0%)	0
Peripheral sensory neuropathy	5/294 (1.7%)	5	37/276 (13.4%)	43	0/3 (0%)	0
Sciatica	0/294 (0%)	0	0/276 (0%)	0	1/3 (33.3%)	2
Syncope	0/294 (0%)	0	0/276 (0%)	0	1/3 (33.3%)	1
Psychiatric disorders
Insomnia	16/294 (5.4%)	16	24/276 (8.7%)	26	0/3 (0%)	0
Respiratory, thoracic and mediastinal disorders
Cough	18/294 (6.1%)	18	28/276 (10.1%)	34	0/3 (0%)	0
Dyspnoea	25/294 (8.5%)	27	19/276 (6.9%)	22	0/3 (0%)	0
Dyspnoea exertional	1/294 (0.3%)	1	2/276 (0.7%)	2	1/3 (33.3%)	2
Skin and subcutaneous tissue disorders
Alopecia	1/294 (0.3%)	1	116/276 (42%)	120	0/3 (0%)	0
Dermatitis bullous	0/294 (0%)	0	0/276 (0%)	0	1/3 (33.3%)	1
Pruritus	29/294 (9.9%)	29	18/276 (6.5%)	25	0/3 (0%)	0
Rash	29/294 (9.9%)	34	22/276 (8%)	30	0/3 (0%)	0

Limitations/Caveats

[Not Specified]

More Information

Certain Agreements

Principal Investigators are NOT employed by the organization sponsoring the study.

The Sponsor must have the opportunity to review all proposed abstracts, manuscripts or presentations regarding this trial 45 days prior to submission for publication/presentation. Any information identified by the Sponsor as confidential must be deleted prior to submission; this confidentiality does not include efficacy and safety results. Sponsor review can be expedited to meet publication timelines.