EFS Improved With Neoadjuvant and Adjuvant Pembrolizumab Across Subgroups of Patients With Early-Stage TNBC

Patients with early triple-negative breast cancer across prespecified subgroups who were given neoadjuvant pembrolizumab plus chemotherapy followed by adjuvant single-agent pembrolizumab in a phase 3 trial saw an improvement in event-free survival.

The phase 3 KEYNOTE-522 study (NCT03036488) showed that patients with early-stage triple-negative breast cancer (TNBC) had an event-free survival benefit when treated with neoadjuvant pembrolizumab (Keytruda) plus chemotherapy compared with placebo and these results were seen across prespecified subgroups, according to a presentation from the 2021 San Antonio Breast Cancer Symposium.

Results from the primary EFS analysis showed that the 784 patients treated with pembrolizumab plus chemotherapy followed by pembrolizumab EFS events occurred in 15.7% of patients compared with 23.8% among 390 patients in the placebo plus chemotherapy followed by placebo arm (HR 0.63; 95% CI, 0.48-0.82; P = .00031). Both cohorts had a median follow-up of 39.1 months (range, 30.0-48.0). The 36-month EFS rates were 84.5% vs 76.8%, respectively.

Investigators performed 5 additional EFS sensitivity analyses (SAs) in addition to the primary EFS analysis. Each additional analysis included the same 5 definitions of an event as the primary analysis: local disease progression precluding definitive surgery, local recurrence, distant progressive disease, distant recurrence, and death from any cause. The primary analysis and SA 1 also considered positive margin at last surgery and a second primary cancer as EFS events; the 2 analyses had identical EFS event criteria with alternate censoring rules.

Additional criteria added to each SA were as follows: SA 2 included new anticancer therapy for metastatic disease as an EFS event, SA 3 included positive margin at last surgery, SA 4 included only the 5 EFS events shared by all the analyses, and SA 5 added positive margin at last surgery, second primary cancer (nonbreast), and second primary breast cancer.

In SA 1, 14.3% of patients experienced an event in the experimental arm vs 21.5% in the control arm (HR 0.64; 95% CI, 0.48-0.84). The 36-month EFS rates were 85.3% vs 77.9%, respectively.

Patients in the experimental arm of SA 2 had an EFS event occurrence of 15.7% compared with 23.8% in the control group (HR 0.63; 95% CI, 0.48-0.82). In SA 3 these rates were 15.6% vs 23.1%, respectively (HR 0.65; 95% CI, 0.50-0.85) with 36-month EFS rates of 84.6% vs 77.5%, respectively. EFS events occurred in 14.8% vs 22.6% of patients in the experimental vs control arms, respectively, in SA 4 (HR 0.63; 95% CI, 0.48-0.84). Finally, in SA 5 events occurred in 16.1% of patients in the experimental cohort and 24.4% of patients in the control arm (HR 0.63; 95% CI, 0.48-0.82).

“The treatment benefit with pembrolizumab in each SA was consistent with the primary analysis,” said Peter Schmid, MD, PhD, a professor of medicine at Cancer Research UK Barts Centre in London, England. “This shows the robustness of the primary EFS results. These results support pembrolizumab plus platinum-containing neoadjuvant chemotherapy followed by adjuvant pembrolizumab after surgery as a new standard-of-care treatment regimen for patients with high-risk, early-stage TNBC.”

Pembrolizumab plus chemotherapy followed by adjuvant pembrolizumab also improved EFS results regardless of nodal status. In patients with node-negative disease, events reported for 11.4% of patients experimental group compared with 18.6% in the control arm (HR 0.58; 95% CI, 0.37-0.91). The 36-month EFS rate was 88.6% in the experimental arm (n = 376) vs 82.2% in the control arm (n = 194). Patients with node-positive disease had events at a rate of 19.6% vs 29.1% in the experimental and control arms, respectively (HR 0.65; 95% CI, 0.46-0.91). The 36-month EFS rates were 80.7% in the experimental arm (n = 408) vs 71.5% in the control arm (n = 196).

Patients also saw benefit with pembrolizumab plus chemotherapy followed by adjuvant pembrolizumab regardless of disease stage. Patients with stage II disease had EFS events reported at a rate of 11.7% vs 18.6% in the experimental and control arms, respectively (HR 0.60; 95% CI, 0.42-0.86). The 36-monnth EFS rates were 88.6% in the experimental arm (n = 590) vs 81.7% in the control arm (n = 291). EFS events were reported in 27.8% vs 39.8%, respectively, for those with stage III disease; the 36-month EFS rates were 71.8% (n = 194) vs 62.0% (n = 98), respectively.

The benefit extended to patients regardless of menopausal status with HRs of 0.62 (95% CI, 0.42-0.91) and 0.64 (95% CI, 0.44-0.93) favoring adjuvant pembrolizumab for patients who were premenopausal and postmenopausal, respectively. Additionally, those with HER2-positivity of 2+ per immunohistochemistry (HR, 0.73; 95% CI, 0.43-1.24) and those with HER2 status 0 to 1 by immunohistochemistry (HR, 0.60; 95% CI, 0.44-0.82) also saw a benefit with the experimental regimen.

“All subgroups appear to derive a comparable EFS benefit, including those by nodal stages and disease stage,” Schmid added.

KEYNOTE-522 randomized a total of 1174 adult patients in a 2:1 manner into the experimental or placebo groups. To be eligible for the trial, patients must have a new diagnosis of TNBC of either stage T1c N1-2 or T2-4 N0-2, an ECOG performance status no greater than 1, and be able to provide a tissue sample for PD-L1 assessment. Stratification factors included nodal status (positive vs negative), tumor size (T1/T2 vs T3/T4), and carboplatin schedule (weekly vs once every 3 weeks).

Baseline characteristics were well-balanced between the 2 arms. Patients in the experimental arm were node-positive at a rate of 51.7% vs 51.3% in the control group, and most patients in both groups were disease stage II (75.3% vs 74.6%). In terms of menopausal status, 55.9% of patients in the experimental cohort were premenopausal compared with 56.7% in the control group.

Both arms of the trial were treated in 2 neoadjuvant treatment phases, with each phase consisting of 4 cycles over 12 weeks. Patients in the experimental arm received chemotherapy plus pembrolizumab 200 mg once every 3 weeks; those in the control arm received placebo in place of pembrolizumab. Both cohorts went to surgery at 12 weeks. The experimental group continued with pembrolizumab 200 mg in the 27-week adjuvant phase and the other arm continued with placebo.

The primary end points of the trial were pathologic control rate (pCR) assessed by the local pathologist and investigator assessed EFS. Secondary end points included pCR by alternative definitions, overall survival, and safety. The exploratory analyses included EFS sensitivity analyses and EFS in patient subgroups.

Regarding safety, treatment-related adverse events (TRAEs) occurred in 98.9% and 99.7% of patients in the experimental and control groups, respectively. Grade 3 to 5 AEs were present in 77.1% of patients in the experimental group vs 73.3% in the control arm. Serious AEs occurred in 34.1% of patients in the experimental arm, 0.5% of AEs led to death and 27.7% led to discontinuation in this group. AEs leading to death occurred in 0.3% of patients in the control group and led to discontinuation 14.1% of the time.

Common grade 1 to 2 TRAEs occurring in the adjuvant phase of treatment in the experimental arm included arthralgia (8.5%), rash (6.0%), and pruritus (5.1%). TRAEs in the control group of grade 1 to 2 in the control group consisted of arthralgia (6.9%), asthenia (5.1%), and fatigue (4.8%), among others. Immune-mediated AEs of any grade occurred in the adjuvant phase at a rate of 10.2% and 6.0%, in the experimental and control arms, respectively.

“The rate of adverse events (AEs) with pembrolizumab was low, especially in the adjuvant setting,” Schmid said.


Schmid P, Cortes J, Dent R, et al. KEYNOTE-522 study of neoadjuvant pembrolizumab + chemotherapy vs placebo + chemotherapy, followed by adjuvant pembrolizumab vs placebo for early-stage TNBC: event-free survival sensitivity and subgroup analyses. Presented at: 2021 San Antonio Breast Cancer Symposium; December 7-10, 2021; Virtual. Abstract GS1-01.