KEYNOTE-B49 (NCT04895358) is Merck’s phase 3 trial of pembrolizumab (KEYTRUDA) or placebo plus chemotherapy in 342 patients with PD-L1-positive (CPS ≥1) HR+/HER2− advanced breast cancer after endocrine therapy. At ESMO 2026 (LBA51), PFS improved (6.7 vs 6.0 months; HR 0.74) but overall survival did not (HR 1.04). This use is investigational.
ESMO 2026 · LBA51 · Final analysisPhase III · NCT04895358PD-L1+ (CPS ≥1) HR+/HER2− advanced breast cancerPembrolizumab vs placebo, each + chemotherapyAnnals of Oncology · Oct 24, 2026⚠ Investigational · not FDA approved in HR+/HER2− breast cancer
At the interim analysis (data cutoff Sep 22, 2025; median follow-up 29.1 months), PFS by blinded independent central review was 6.7 months (95% CI 6.2–8.3) with pembrolizumab plus chemotherapy vs 6.0 months (4.2–8.1) with placebo plus chemotherapy: HR 0.74 (0.58–0.95), P = 0.0089, meeting the prespecified one-sided α of 0.025. Events: 124/171 vs 139/171. The 12-month PFS rate was 26.3% vs 20.3% and the 24-month rate 15.0% vs 4.3% (LBA51 abstract and presentation slides, IA). At the descriptive final analysis (data cutoff Mar 16, 2026; median follow-up 34.9 months) the HR was unchanged at 0.74 (0.58–0.95), with 24-month PFS of 14.1% vs 3.5% (LBA51 slides, FA).
Key secondary endpoint not met: overall survival (PD-L1 CPS ≥1)
At the final analysis (data cutoff Mar 16, 2026), median OS was 21.0 months (18.9–22.9) vs 22.2 months (17.6–26.3): HR 1.04 (0.81–1.33), P = 0.6185, which did not meet the prespecified criterion. Events: 126/171 vs 124/171 (LBA51 abstract and slides, FA). In a post hoc exploratory analysis adjusting for subsequent therapy (received by 73.1% vs 86.0%; ADCs by 29.2% vs 38.0%), the OS HR was 0.94 (0.74–1.21) (LBA51 slides, FA). Per the presentation, reducing enrollment from a planned 800 to 340 participants lowered the power for OS to about 65%.
PFS: 7.5 months (4.2–10.3) vs 4.4 months (4.0–8.3), HR 0.59 (0.35–0.98); 12-month PFS 23.6% vs 9.3% (LBA51 abstract and slides, IA). OS: 22.9 months (18.2–33.4) vs 22.5 months (12.7–30.4), HR 0.82 (0.47–1.43) (LBA51 slides, FA). The investigators describe a numerical effect favoring pembrolizumab plus chemotherapy in this group. 21.1% of patients in each arm had CPS ≥10; 78.9% had CPS 1–9.
Treatment-related AEs, n (%) (FA, DCO Mar 16, 2026)
Pembro + chemo (n=170)
Placebo + chemo (n=168)
Any
155 (91.2)
152 (90.5)
Grade ≥3
64 (37.6)
55 (32.7)
Serious
23 (13.5)
13 (7.7)
Led to any treatment discontinuation
19 (11.2)
9 (5.4)
Led to death
3 (1.8)
0
Hypothyroidism (any grade, %)
18.2
3.0
Rash (any grade, %)
10.6
5.4
The three treatment-related deaths were autoimmune hepatitis, gastrointestinal inflammation and pneumonitis. The investigators reported no new safety signals (LBA51 slides and abstract, FA).
On the forest plot (data cutoff Sep 22, 2025), PFS hazard ratios were 0.80 (0.60–1.05) for CPS 1–9 and 0.59 (0.35–0.98) for CPS ≥10; 0.53 (0.35–0.81) after 1 prior line of endocrine therapy for metastatic disease and 0.90 (0.67–1.23) after ≥2; 0.55 (0.40–0.77) with and 0.98 (0.68–1.42) without prior (neo)adjuvant taxanes; 0.75 (0.54–1.05) with capecitabine as the on-study chemotherapy (LBA51 slides, IA).
⚠ Pembrolizumab plus chemotherapy in HR-positive, HER2-negative breast cancer is investigational; it is not FDA approved for this use. ✓ KEYTRUDA’s FDA-approved breast cancer indications are in triple-negative breast cancer: high-risk early-stage TNBC with neoadjuvant chemotherapy, continued as adjuvant monotherapy, and locally recurrent unresectable or metastatic TNBC with PD-L1 CPS ≥10 in combination with chemotherapy. The presentation notes that the ongoing TroFuse-010 study (NCT06312176) is further investigating pembrolizumab with sacituzumab tirumotecan (sac-TMT) in an overall population and a PD-L1-defined subgroup.
Sources: ESMO 2026 abstract LBA51 and presentation (Hope S. Rugo, MD, City of Hope, October 24, 2026); Rugo HS et al., Annals of Oncology 2026 (doi:10.1016/j.annonc.2026.10.1285); ClinicalTrials.gov NCT04895358; KEYTRUDA US prescribing information. IA = interim (primary PFS) analysis, data cutoff Sep 22, 2025; FA = final analysis, data cutoff Mar 16, 2026. Slides were shared on X by the physicians credited in Key Slides.
Free Access · KOL Pulse Intelligence
Track oncology’s top KOLs in your specialty
Create a free account, pick the tumor types you cover, and go beyond the tweet — the intelligence pharma teams use to map influence and prepare for engagement:
Choose the tumor types you follow. No cost, unsubscribe anytime.
Influence Leaders
Top KOLs Discussing KEYNOTE-B49
The presenter, Hope S. Rugo, MD (City of Hope), first, followed by the physicians whose KEYNOTE-B49 posts drew the most views. Their posts are in Physician Reactions below.
The LBA51 presentation slides (presenter: Hope S. Rugo, MD) as photographed and shared on X by physicians during the October 24 session, plus the abstract and the Annals of Oncology publication. Survival curves are summarized in number tables under the slide. Click any image to view it full size, and open the OCR text under each group. All ESMO 2026 KEYNOTE-B49 slides in the KOL Pulse slide library →
Overall survival, final analysis (DCO Mar 16, 2026), as printed on the LBA51 slides
Population
Pembro + chemo
Placebo + chemo
HR (95% CI)
PD-L1 CPS ≥1 (key secondary; N=171 per arm)
21.0 mo (18.9–22.9)
22.2 mo (17.6–26.3)
1.04 (0.81–1.33), P = 0.6185 (not met)
PD-L1 CPS ≥10 (secondary; N=36 per arm)
22.9 mo (18.2–33.4)
22.5 mo (12.7–30.4)
0.82 (0.47–1.43)
Slide text (transcribed from the slide images)
OS in Participants with PD-L1 CPS ≥1 (Key Secondary Endpoint)
FA | Events | Median (95% CI), mo | HR (95% CI)
Pembro + chemo | 73.7% | 21.0 (18.9–22.9) | 1.04 (0.81–1.33), P = 0.6185(a)
Placebo + chemo | 72.5% | 22.2 (17.6–26.3)
No. at risk (0–54 mo, every 6 mo)
Pembro + chemo: 171 154 131 103 69 33 12 5 2 0
Placebo + chemo: 171 151 124 96 76 44 16 9 2 0
Median follow-up, 34.9 months (range, 23.9–55.1). Median OS based on the Kaplan-Meier method. HRs (95% CIs) based on a Cox regression model with treatment as a covariate and the Efron method of tie handling stratified by the randomization stratification factors. (a)Did not meet the prespecified statistical criterion of α = 0.025. Data cutoff, Mar 16, 2026.
Survival table
Endpoint: OS, PD-L1 CPS ≥1 (key secondary)
Arm | N | Median | Median CI
Pembro + chemo | 171 | 21.0 mo | 18.9–22.9
Placebo + chemo | 171 | 22.2 mo | 17.6–26.3
HR: 1.04
HR CI: 0.81–1.33
P: 0.6185 (did not meet α=0.025)
Slide text (transcribed from the slide images)
[Slide 1]
[Session banner] 08:30 - 10:00 Proffered paper 2: Breast cancer, early and advanced stage
CHAIRS: SARA HURVITZ, SEOCK-AH IM
OS in Participants with PD-L1 CPS >=10 (Secondary Endpoint)
FA | Events | Median (95% CI), mo | HR (95% CI)
Pembro + chemo | 69.4% | 22.9 (18.2-33.4) | 0.82 (0.47-1.43)
Placebo + chemo | 77.8% | 22.5 (12.7-30.4) |
Y-axis: OS, %; X-axis: Time, mo
No. at risk (0/6/12/18/24/30/36/42/48/54):
Pembro + chemo: 36 / 32 / 29 / 26 / 17 / 9 / 3 / 2 / 1 / 0
Placebo + chemo: 36 / 32 / 26 / 21 / 15 / 11 / 4 / 3 / 1 / 0
Median follow-up, 34.9 months (range, 23.9-55.1). Median OS based on the Kaplan-Meier method. HRs (95% CIs) based on a Cox regression model with treatment as a covariate and the Efron method of tie handling stratified by the randomization stratification factors. Data cutoff, Mar 16, 2026.
[Speaker video] Hope Rugo
Results from KEYNOTE-B49: A phase 3, randomized, double-blind study of pembrolizumab (pembro) vs placebo (pbo) plus chemotherapy (CT) in participants (pts) with HR+/HER2- advanced breast cancer (ABC)
ALICANTE AUDITORIUM - HALL 6
Survival table
Endpoint: OS, PD-L1 CPS >=10 (final analysis)
Arm | N | Median | Median CI
Pembro + chemo | 36 | 22.9 | 18.2-33.4
Placebo + chemo | 36 | 22.5 | 12.7-30.4
HR: 0.82
HR CI: 0.47-1.43
[Slide 2]
[Session banner] 08:30 - 10:00 Proffered paper 2: Breast cancer, early and advanced stage
CHAIRS: SARA HURVITZ, SEOCK-AH IM
OS in Participants with PD-L1 CPS ≥1 (Key Secondary Endpoint)
FA | Events | Median (95% CI), mo | HR (95% CI)
Pembro + chemo | 73.7% | 21.0 (18.9–22.9) | 1.04 (0.81–1.33) P = 0.6185^a
Placebo + chemo | 72.5% | 22.2 (17.6–26.3) |
OS, % vs Time, mo (0-54)
No. at risk
Pembro + chemo: 171 154 131 103 69 33 12 5 2 0
Placebo + chemo: 171 151 124 96 76 44 16 9 2 0
Median follow-up, 34.9 months (range, 23.9–55.1). Median OS based on the Kaplan-Meier method. HRs (95% CIs) based on a Cox regression model with treatment as a covariate and the Efron method of tie handling stratified by the randomization stratification factors. ^aDid not meet the prespecified statistical criterion of α = 0.025. Data cutoff, Mar 16, 2026.
[Speaker panel] Hope Rugo
Results from KEYNOTE-B49: A phase 3, randomized, double-blind study of pembrolizumab (pembro) vs placebo (pbo) plus chemotherapy (CT) in participants (pts) with HR+/HER2- advanced breast cancer (ABC)
ALICANTE AUDITORIUM - HALL 6
Survival table
Endpoint: OS (key secondary), PD-L1 CPS >=1, final analysis
Arm | N | Median | Median CI
Pembro + chemo | 171 | 21.0 mo | 18.9–22.9
Placebo + chemo | 171 | 22.2 mo | 17.6–26.3
HR: 1.04
HR CI: 0.81–1.33
P: 0.6185 (did not meet prespecified alpha 0.025)
[Slide 3]
[Session banner] 08:30 - 10:00 Proffered paper 2: Breast cancer, early and advanced stage
CHAIRS: SARA HURVITZ, SEOCK-AH IM
Adjusted Analysis of OS^a (Post Hoc Exploratory Analysis)
• 73.1% of participants in the pembrolizumab plus chemotherapy arm and 86.0% in the placebo plus chemotherapy arm received subsequent therapy (difference of 12.9%)
– 50 (29.2%) and 65 (38.0%) participants received ADCs, respectively
| Pembro + chemo (n = 171) | Placebo + chemo (n = 171)
Subsequent therapy use, n (%) | 125 (73.1) | 147 (86.0)
OS
Unadjusted HR (95% CI) | 1.04 (0.81–1.33)
Adjusted^a HR (95% CI) | 0.94 (0.74–1.21)
• The adjusted HR for OS suggests the observed OS treatment effect may have been influenced by the confounding effect of subsequent therapy
Median follow-up, 34.9 months (range, 23.9–55.1). ^aA 2-stage model was used to adjust the estimated effect on OS for the potential confounding effect of the difference in subsequent therapy use between arms. Data cutoff, Mar 16, 2026.
[Speaker panel] Hope Rugo
Results from KEYNOTE-B49: A phase 3, randomized, double-blind study of pembrolizumab (pembro) vs placebo (pbo) plus chemotherapy (CT) in participants (pts) with HR+/HER2- advanced breast cancer (ABC)
ALICANTE AUDITORIUM - HALL 6
[Slide 4]
[Session banner] 08:30 - 10:00 Proffered paper 2: Breast cancer, early and advanced stage
CHAIRS: SARA HURVITZ, SEOCK-AH IM
Conclusions
• The phase 3 KEYNOTE-B49 study (NCT04895358) is the largest study evaluating immunotherapy plus chemotherapy in endocrine-resistant HR+/HER2−, PD-L1–positive (CPS ≥1) advanced breast cancer
• Pembrolizumab plus chemotherapy significantly improved PFS vs placebo plus chemotherapy in participants with PD-L1 CPS ≥1 tumors, with a numerically greater benefit observed in those with PD-L1 CPS ≥10 tumors and a generally consistent treatment effect across other prespecified subgroups
• No significant improvement in OS was observed in participants with PD-L1 CPS ≥1 tumors. Interpretation of OS results is limited, as a protocol amendment reduced enrollment from 800 to 340 participants due to low prevalence of PD-L1–positive (CPS ≥1) disease; reducing statistical power for OS to ~65%
• Safety was consistent with the known profiles of each regimen with no new safety signals
• KEYNOTE-B49 met its primary endpoint and demonstrated improved PFS in a PD-L1–defined subset of participants with HR+/HER2− advanced breast cancer. The ongoing TroFuse-010 study (NCT06312176) is further investigating the efficacy of pembrolizumab in combination with sacituzumab tirumotecan (sac-TMT) in both the overall population and a PD-L1–defined subgroup
[Speaker panel] Hope Rugo
Results from KEYNOTE-B49: A phase 3, randomized, double-blind study of pembrolizumab (pembro) vs placebo (pbo) plus chemotherapy (CT) in participants (pts) with HR+/HER2- advanced breast cancer (ABC)
ALICANTE AUDITORIUM - HALL 6
Slide text (transcribed from the slide images)
[Slide 1]
[Session banner] 08:30 - 10:00 Proffered paper 2: Breast cancer, early and advanced stage
CHAIRS: SARA HURVITZ, SEOCK-AH IM
Tumor Response Outcomes at Final Analysis (Secondary Endpoints)
PD-L1 CPS >=1: ORR, % (95% CI) Pembro + chemo 31.6 (54/171); Placebo + chemo 28.1 (48/171)
PD-L1 CPS >=10: ORR, % (95% CI) Pembro + chemo 44.4 (16/36); Placebo + chemo 30.6 (11/36)
PD-L1 CPS >=1 | Pembro + chemo | Placebo + chemo
DOR,(a) median (range), mo | 8.3 (1.0+ to 40.0+) | 8.4 (2.2+ to 27.3)
PD-L1 CPS >=10 | Pembro + chemo | Placebo + chemo
DOR,(a) median (range), mo | 8.3 (4.3-29.7+) | 6.2 (4.2-12.5)
Median follow-up, 34.9 months (range, 23.9-55.1). "+" indicates no PD by the time of the participant's last disease assessment. (a)In participants with a confirmed objective response. Data cutoff, Mar 16, 2026.
[Speaker video] Hope Rugo
Results from KEYNOTE-B49: A phase 3, randomized, double-blind study of pembrolizumab (pembro) vs placebo (pbo) plus chemotherapy (CT) in participants (pts) with HR+/HER2- advanced breast cancer (ABC)
ALICANTE AUDITORIUM - HALL 6
[Slide 2]
[Session banner] 08:30 - 10:00 Proffered paper 2: Breast cancer, early and advanced stage
CHAIRS: SARA HURVITZ, SEOCK-AH IM
Summary of Treatment-Related(a) AEs
| Pembro + chemo (n = 170) | Placebo + chemo (n = 168)
Treatment-related AEs | 155 (91.2) | 152 (90.5)
Grade >=3 | 64 (37.6) | 55 (32.7)
Serious | 23 (13.5) | 13 (7.7)
Led to any treatment discontinuation | 19 (11.2) | 9 (5.4)
Led to death | 3 (1.8)(b) | 0
Treatment-Related AEs with Incidence >=10% in Either Treatment Arm (Incidence, %; Pembro + chemo / Placebo + chemo; Grade 1/2 and 3-5 stacked):
PPE 30.6 / 33.3; Anemia 29.4 / 24.4; Nausea 25.9 / 20.8; Diarrhea 21.2 / 22.6; Alopecia 18.2 / 20.8; Fatigue 21.2 / 17.9; Neutropenia 17.6 / 13.1; Neutrophil count decreased 17.1 / 12.5; Asthenia 13.5 / 15.5; WBC count decreased 11.8 / 16.1; AST increased 14.7 / 11.9; ALT increased 11.8 / 9.5; Hypothyroidism 18.2 / 3.0; Vomiting 10.0 / 8.9; Rash 10.6 / 5.4
PPE, palmar-plantar erythrodysesthesia syndrome. Data in the table are n (%). (a)Determined by the investigator to be related to any study treatment. (b)Grade 5 events were autoimmune hepatitis, gastrointestinal inflammation, and pneumonitis. Data cutoff, Mar 16, 2026.
[Speaker video] Hope Rugo
Results from KEYNOTE-B49: A phase 3, randomized, double-blind study of pembrolizumab (pembro) vs placebo (pbo) plus chemotherapy (CT) in participants (pts) with HR+/HER2- advanced breast cancer (ABC)
ALICANTE AUDITORIUM - HALL 6
Slide text (transcribed from the slide images)
Annals of Oncology
Available online 24 October 2026
In Press, Journal Pre-proof What's this?
Original Article
Pembrolizumab Plus Chemotherapy for HR+/HER2– PD-L1–Positive Advanced Breast Cancer: Final Analysis of the Phase 3 KEYNOTE-B49 Study
H.S. Rugo 1, M. Kok 2, A. Goncalves 3, D. Loirat 4, G. Bianchini 5 6, P.A. Fasching 7, D.W. Cescon 8, R. Romero Díaz 9, F. Zagouri 10, E. Korbenfeld 11, F. Dalenc 12, Y. Jerez Gilarranz 13, J. Hashimoto 14, Y.H. Park 15, S.-A. Im 16, X. Hu 17, K.M. Hirshfield 18, L. Jia 18, K. Tryfonidis 18, P. Schmid 19
Slide text (transcribed from the slide images)
Proffered Paper session
LBA51 - Results from KEYNOTE-B49: A phase 3, randomized, double-blind study of pembrolizumab (pembro) vs placebo (pbo) plus chemotherapy (CT) in participants (pts) with HR+/HER2− advanced breast cancer (ABC)
Speakers Hope S. Rugo (Duarte, United States of America)
Lecture Time 08:30 - 08:40
Abstract
Background
After failure of endocrine-based therapy (ET), patients with HR+/HER2− ABC have poor outcomes. We report results of the phase 3 KEYNOTE-B49 (NCT04895358) study of pembro + CT vs pbo + CT in pts with PD-L1–positive HR+/HER2− ABC.
Methods
Pts with locally recurrent inoperable or metastatic HR+/HER2− ABC with PD-L1 CPS ≥1 who had prior ET and were candidates for first-line CT were randomized to pembro 200 mg or pbo Q3W for up to 35 cycles + CT (paclitaxel, nab-paclitaxel, liposomal doxorubicin, or capecitabine). The primary endpoint was PFS per RECIST v1.1 by BICR in pts with PD-L1 CPS ≥1. The key secondary endpoint was OS in pts with PD-L1 CPS ≥1. Following a protocol amendment, the sample size was reduced due to low PD-L1 prevalence.
Results
342 pts were randomized. Median age was 57 y; 99.4% of pts in the pembro + CT arm and 98.8% in the pbo + CT arm had prior CDK4/6i. At primary analysis (data cutoff, Sep 22, 2025; median follow-up, 29.1 mo), pembro + CT showed statistically significant improvement in PFS vs pbo + CT (HR, 0.74; 95% CI, 0.58–0.95; P = 0.0089; table). In pts with PD-L1 CPS ≥10, the HR for PFS (0.59; 95% CI, 0.35–0.98) favored pembro + CT (median [95% CI] PFS, 7.5 [4.2–10.3] mo in the pembro + CT arm and 4.4 [4.0–8.3] mo in the pbo + CT arm). At final analysis (data cutoff, Mar 16, 2026; median follow-up, 34.9 mo), PFS benefit was maintained (HR, 0.74; 95% CI, 0.58–0.95), but there was no significant difference in OS (HR, 1.04; 95% CI, 0.81–1.33; P = 0.6185; table). Grade ≥3 treatment-related AEs occurred in 37.6% in the pembro + CT arm and 32.7% in the pbo + CT arm; 11.2% and 5.4% discontinued treatment due to treatment-related AEs, respectively.
Conclusions
In pts with PD-L1 CPS ≥1 HR+/HER2− ABC, pembro + CT demonstrated a statistically significant improvement in PFS, indicating a 26% risk reduction, and a manageable safety profile. A numerical effect favoring pembro + CT for PFS was observed in pts with PD-L1 CPS ≥10. Table: LBA51
| Pembro + CT | Pbo + CT
Primary analysis for PFS
PD-L1 CPS ≥1a
Events, n/N | 124/171 | 139/171
Median (95% CI), mo | 6.7 (6.2–8.3) | 6.0 (4.2–8.1)
HR (95% CI) | 0.74 (0.58–0.95); P = 0.0089
Final analysis for OS
PD-L1 CPS ≥1a
Events, n/N | 126/171 | 124/171
Median (95% CI), mo | 21.0 (18.9–22.9) | 22.2 (17.6–26.3)
HR (95% CI) | 1.04 (0.81–1.33); P = 0.6185
aMultiplicity was controlled at α = 0.025 (1-sided)
KOL-Made Infographics
Summary graphics designed by physicians (not conference slides). Values are as the authors drew them. Click to enlarge.
@ChandrakanthMv
@ChandrakanthMv
@ChandrakanthMv
@ChandrakanthMv
@ozdogan_md
Physician Reactions · ESMO 2026
Physician reactions to KEYNOTE-B49
Full, verbatim posts from 24 physicians (29 posts), newest first, linked to the originals. Where a post’s figure differs from the presentation slide, a note sits under it.
KEYNOTE-B49 asked whether adding the PD-1 inhibitor pembrolizumab to first-line chemotherapy improves outcomes for patients with HR-positive, HER2-negative advanced breast cancer whose disease has progressed on endocrine-based therapy and whose tumors express PD-L1. Almost all participants (99.4% and 98.8%) had received a CDK4/6 inhibitor. The trial is sponsored by Merck Sharp & Dohme LLC. Because PD-L1-positive tumors were less common than expected (19.4% of 1,767 screened patients were randomized; 59% of screen failures were PD-L1-negative), a protocol amendment cut the sample size from 800 to 340. The final analysis was presented by Hope S. Rugo, MD, at ESMO 2026 (LBA51, Proffered Paper session 2, October 24, 2026) and published simultaneously in Annals of Oncology.
Study Design
Randomized 1:1, double-blind, placebo-controlled phase 3 (NCT04895358); 342 randomized (171 per arm). Stratified by PD-L1 CPS (1–9 vs ≥10), visceral disease and planned chemotherapy.
Population
HR+/HER2− locally recurrent inoperable or metastatic breast cancer, PD-L1 CPS ≥1, no prior chemotherapy for metastatic disease; progression on ≥1 line of endocrine therapy with a CDK4/6 inhibitor (or early relapse criteria). Visceral metastases in about 85%.
Interventions
Pembrolizumab 200 mg or placebo every 3 weeks for up to 35 cycles, plus physician’s choice of paclitaxel, nab-paclitaxel, liposomal doxorubicin or capecitabine (capecitabine in 53.2% per arm).
Endpoints
Primary: PFS (RECIST v1.1, BICR) in CPS ≥1. Key secondary: OS in CPS ≥1. Secondary: PFS and OS in CPS ≥10; ORR, DCR, DOR; safety. Multiplicity controlled at one-sided α = 0.025.
Presenter
Hope S. Rugo, MD (City of Hope Comprehensive Cancer Center, Duarte, CA); senior author Peter Schmid (Barts Cancer Institute, Queen Mary University of London).
Data Cuts
Interim (primary PFS) analysis: Sep 22, 2025, median follow-up 29.1 months. Final analysis (OS; descriptive PFS update): Mar 16, 2026, median follow-up 34.9 months.
KEYNOTE-B49 (NCT04895358) is a randomized, double-blind phase 3 trial by Merck of pembrolizumab or placebo plus physician's-choice chemotherapy in 342 patients with PD-L1-positive (CPS ≥1), HR-positive, HER2-negative locally recurrent inoperable or metastatic breast cancer after endocrine therapy. Design details: Trial Design.
What did KEYNOTE-B49 show at ESMO 2026?
The primary endpoint was met: progression-free survival improved with pembrolizumab plus chemotherapy (median 6.7 vs 6.0 months; HR 0.74). Overall survival, the key secondary endpoint, did not improve (HR 1.04). Figures and data cuts: At a Glance.
Did patients with PD-L1 CPS 10 or higher benefit more?
In the 72 patients with CPS ≥10, a secondary endpoint, the PFS hazard ratio was 0.59 and the OS hazard ratio 0.82. The investigators describe this as a numerical effect in a small subgroup. Values: At a Glance.
Is pembrolizumab approved for HR-positive, HER2-negative breast cancer?
No. In breast cancer, KEYTRUDA (pembrolizumab) is FDA approved only for triple-negative breast cancer: high-risk early-stage TNBC with neoadjuvant chemotherapy and then as adjuvant monotherapy, and locally recurrent unresectable or metastatic TNBC with PD-L1 CPS ≥10 in combination with chemotherapy. Its use in HR-positive, HER2-negative disease, as in KEYNOTE-B49, is investigational.
What were the safety findings?
Grade ≥3 treatment-related adverse events occurred in 37.6% with pembrolizumab plus chemotherapy and 32.7% with placebo plus chemotherapy, and 3 treatment-related deaths (1.8%) occurred in the pembrolizumab arm. Details: At a Glance.
Why was the trial smaller than planned?
A protocol amendment reduced the planned sample size from 800 to 340 because PD-L1-positive tumors were less common than expected; 59% of screen failures were due to negative PD-L1 expression. Per the presentation, this reduced the statistical power for overall survival to about 65%.
Physician Opinions
Key KOL Sentiments - KEYNOTE-B49
2 positive, 19 neutral and 8 negative posts from physicians, each shown verbatim.
Compiled and reviewed by the KOL Pulse research team, led by Brian Shields, Founder, KOL Pulse. Last updated October 25, 2026. Every quote verbatim; every number checked against the ESMO 2026 LBA51 abstract and presentation slides, the Annals of Oncology publication, ClinicalTrials.gov or the FDA label, with the source and data cut labelled.
Free access — track oncology’s top KOLs in your specialty: profiles, pharma influence, ad-board patterns & 2025 Open Payments analysis.KOL Pulse Insider
@ChandrakanthMv
@ChandrakanthMv
@ChandrakanthMv
@ChandrakanthMv
@ozdogan_md
