On July 26, 2021, the FDA granted approval to pembrolizumab in combination with chemotherapy for neoadjuvant treatment and then continued as a single agent for adjuvant treatment following surgery for patients with high-risk, early-stage triple-negative breast cancer. Approval was based on results from KEYNOTE-522, an ongoing randomized (2:1) trial evaluating pembrolizumab or placebo in combination with chemotherapy for neoadjuvant treatment and then as a single agent for adjuvant treatment. The co-primary endpoints were pathological complete response (pCR) rate and event-free survival (EFS). The trial demonstrated an improvement in pCR and EFS in the pembrolizumab arm compared with the control arm. The number of patients who experienced an EFS event was 123 (16%) and 93 (24%), respectively [HR: 0.63, 95% confidence interval (CI), 0.48–0.82, P = 0.00031]. Patients on the pembrolizumab arm experienced EFS benefit regardless of tumor PD-L1 status. The absolute pCR rate improvement with the addition of pembrolizumab was 7.5% (95% CI, 1.6–13.4). Among patients receiving pembrolizumab, 44% experienced an immune-related adverse reaction. This article summarizes FDA's review of pembrolizumab and the data supporting the favorable benefit–risk assessment.

Breast cancer remains the second leading cause of cancer-related death in women in the United States (1). Approximately 10%–15% of patients with breast cancer are diagnosed with triple-negative breast cancer (TNBC), a histopathologic subtype defined by absence of estrogen receptor expression, progesterone receptor expression, and HER2 overexpression (2). Although TNBC may be curable if detected at an early stage, even when detected early it is characterized by an increased risk of recurrence and death compared with other breast cancer subtypes.

Despite treatment with multiagent chemotherapy in the neoadjuvant and/or adjuvant setting, 5-year event-free survival (EFS) or disease-free survival (DFS) is approximately 70% and 5-year overall survival (OS) is approximately 77% (3–5), demonstrating continued need for therapies to improve long-term outcomes. FDA will consider pathologic complete response (pCR) rate in certain circumstances to be used as an endpoint reasonably likely to predict clinical benefit and support approval (6). However, there are limitations to interpretation of a pCR endpoint. A large, pooled analysis of patients receiving neoadjuvant treatment for early-stage breast cancer showed that at the individual patient level, patients experiencing a pCR had an improvement in long-term outcomes like EFS and OS, regardless of treatment received, compared with patients with residual disease at time of surgery. In contrast, at the trial level, an improvement in pCR rate in the experimental arm did not always yield an improvement in EFS or OS over the control arm (7). A more recent meta-analysis suggested that there may be a trial-level association between pCR and EFS in TNBC, but this analysis had certain limitations and further work is still needed to understand the association between pCR and EFS and OS at the trial level (8).

Because of uncertainty regarding the relationship between pCR rate and long-term outcomes at the trial level, FDA considers pCR an endpoint that can only support accelerated approval. Demonstration of improvement in EFS, DFS, or OS, which are established measures of clinical benefit, is needed to convert to traditional approval. Furthermore, to grant an accelerated approval based on pCR rate, FDA considers information beyond whether the endpoint reached statistical significance. FDA examines the magnitude of the pCR rate benefit, the toxicity of the experimental treatment, and whether data from another treatment setting demonstrate clinical benefit. Some of the uncertainty associated with the pCR endpoint may be overcome by a large pCR rate difference and compelling data from another setting.

As described in the FDA guidance to industry on the use of pCR in breast cancer, there are two trial design options if seeking accelerated approval for neoadjuvant treatment: the single trial or multiple trial model. In the single trial model, pCR and EFS endpoints are examined in the same trial; earlier pCR results support a potential accelerated approval with later EFS results supporting conversion to a traditional approval (9). As this trial is powered for the EFS endpoint, it may be overpowered for the pCR endpoint and detect a statistically significant pCR rate difference that is too small to be clinically meaningful. In the multiple trial model, pCR results from an initial neoadjuvant trial can support accelerated approval. A subsequent trial in either the neoadjuvant or adjuvant setting with a EFS or DFS endpoint can support conversion to a traditional approval.

Approval for pembrolizumab for neoadjuvant and adjuvant treatment was initially sought based on pCR results from a trial designed according to the single trial model. FDA requested input at an Oncology Drugs Advisory Committee meeting and following deliberations at this meeting, chose to await EFS and OS results rather than grant approval (10). In this article, we discuss the rationale for FDA's initial deferral and later approval of pembrolizumab for treatment of high-risk, early-stage TNBC in combination with chemotherapy as neoadjuvant treatment and then continued as a single agent for adjuvant treatment following surgery (11).

This approval of pembrolizumab is based on results from KEYNOTE-522 (NCT03036488), an ongoing, randomized (2:1), double-blind, placebo-controlled trial in patients with previously untreated, high-risk, early-stage TNBC (12). Enrolled patients had either (i) a primary tumor >1 cm but ≤2 cm in diameter with lymph node involvement or (ii) a primary tumor >2 cm in diameter regardless of lymph node involvement. Patients were eligible irrespective of tumor PD-L1 expression. Stratification factors for randomization were lymph node status (positive vs. negative), tumor size (T1/T2: <5 cm vs. T3/T4: ≥5 cm), and choice of carboplatin regimen (dosed every 3 weeks vs. dosed weekly).

Patients were randomized to receive either neoadjuvant pembrolizumab 200 mg or placebo i.v. every 3 weeks for eight cycles in combination with neoadjuvant chemotherapy, followed by adjuvant pembrolizumab 200 mg or placebo i.v. every 3 weeks for nine cycles as monotherapy following surgery. Patients received the same experimental treatment (pembrolizumab or placebo) in the neoadjuvant and adjuvant portions of the trial. Neoadjuvant chemotherapy consisted of four cycles of paclitaxel (80 mg/m2 weekly) and carboplatin (either carboplatin AUC 5 mg/mL/minute every 3 weeks or carboplatin AUC 1.5 mg/mL/minute weekly), followed by four cycles of an anthracycline (doxorubicin 60 mg/m2 or epirubicin 90 mg/m2 every 3 weeks) and cyclophosphamide (600 mg/m2 every 3 weeks).

The dual primary endpoints were pCR rate and EFS. The key secondary endpoint was OS. The pCR endpoint, measured at time of surgery, reflected the neoadjuvant treatment effect only, whereas the EFS and OS endpoints capture the effect of the entire neoadjuvant and adjuvant treatment regimen. The entire distribution of EFS was evaluated over time rather than at a landmark timepoint (e.g., 3-year EFS), which would require the analysis to be performed at a specific clinically relevant timepoint. Formal testing of landmark timepoints did not occur. Preplanned analyses included an interim analysis and a final analysis for pCR rate, and up to seven interim analyses and a final analysis for EFS and OS. For EFS, the interim analyses were calendar driven, and the final analysis would occur when 327 EFS events had occurred. The OS analyses followed the same schedule as EFS. OS would only be formally tested whether the EFS endpoint was met. The Lan-DeMets O'Brien–Fleming stopping boundary was used to control a potential type I error inflation due to multiple analyses (10). The trial controlled overall type 1 error at a two-sided level of 0.05 and controlled for multiplicity with a graphical approach.

Efficacy

There were 1,174 patients randomized in KEYNOTE-522, 784 patients to the pembrolizumab arm and 390 patients to the control arm. Patient demographic and disease characteristics were generally balanced between the two arms (Table 1). Approximately 74% of patients had a T1 or T2 primary tumor and 52% of patients had node positive disease. Regarding tumor PD-L1 expression, 83% of patients had tumor PD-L1 combined positive score (CPS) ≥1 and 49% of patients had tumor PD-L1 CPS ≥10.

Table 1.

Patient demographics and disease characteristics in KEYNOTE-522.a

Pembrolizumab + chemotherapy/pembrolizumabPlacebo + chemotherapy/placebo
n = 784n = 390
n (%)n (%)
Age, median (range) 49 (22–80) 48 (24–79) 
Female 783 (99.9) 390 (100) 
Race 
 American Indian or Alaska Native 14 (2) 7 (2) 
 Asian 149 (19) 89 (23) 
 Black or African American 38 (5) 15 (4) 
 Native Hawaiian or Pacific Islander 1 (0.1) 0 (0) 
 White 504 (64) 242 (62) 
 Multiple 13 (2) 6 (2) 
 Missing 65 (8) 31 (8) 
Ethnicity 
 Hispanic or Latino 86 (11) 39 (10) 
 Not Hispanic or Latino 615 (78) 307 (79) 
 Missing 46 (6) 28 (7) 
ECOG performance status 
 0 678 (87) 341 (87) 
 1 106 (13) 49 (13) 
Region 
 North America 166 (21) 78 (20) 
 Rest of World 618 (79) 312 (80) 
Primary tumor size 
 T1/T2 580 (74) 290 (74) 
 T3/T4 204 (26) 100 (26) 
Nodal status 
 Positive 405 (52) 200 (51) 
 Negative 379 (48) 190 (49) 
Disease stage 
 Stage II 590 (75) 291 (75) 
 Stage III 194 (25) 98 (25) 
Tumor PD-L1 status 
 CPS ≥ 1 656 (84) 317 (81) 
 CPS ≥ 10 393 (50) 177 (45) 
Pembrolizumab + chemotherapy/pembrolizumabPlacebo + chemotherapy/placebo
n = 784n = 390
n (%)n (%)
Age, median (range) 49 (22–80) 48 (24–79) 
Female 783 (99.9) 390 (100) 
Race 
 American Indian or Alaska Native 14 (2) 7 (2) 
 Asian 149 (19) 89 (23) 
 Black or African American 38 (5) 15 (4) 
 Native Hawaiian or Pacific Islander 1 (0.1) 0 (0) 
 White 504 (64) 242 (62) 
 Multiple 13 (2) 6 (2) 
 Missing 65 (8) 31 (8) 
Ethnicity 
 Hispanic or Latino 86 (11) 39 (10) 
 Not Hispanic or Latino 615 (78) 307 (79) 
 Missing 46 (6) 28 (7) 
ECOG performance status 
 0 678 (87) 341 (87) 
 1 106 (13) 49 (13) 
Region 
 North America 166 (21) 78 (20) 
 Rest of World 618 (79) 312 (80) 
Primary tumor size 
 T1/T2 580 (74) 290 (74) 
 T3/T4 204 (26) 100 (26) 
Nodal status 
 Positive 405 (52) 200 (51) 
 Negative 379 (48) 190 (49) 
Disease stage 
 Stage II 590 (75) 291 (75) 
 Stage III 194 (25) 98 (25) 
Tumor PD-L1 status 
 CPS ≥ 1 656 (84) 317 (81) 
 CPS ≥ 10 393 (50) 177 (45) 

aData from U.S. FDA (10).

Initially, results from interim analysis 1, 2, and 3 (IA1, IA2, and IA3) were submitted to support an accelerated approval. At IA3, the pCR rate endpoint could be assessed in the entire randomized population. The pCR rate on the pembrolizumab arm was 63.0% and the pCR rate on the control arm was 55.6%, for a pCR rate difference of 7.5% [95% confidence interval (CI), 1.6–13.4]. At IA3, EFS and OS data were immature. Only 53% of EFS events and only 32% of OS events needed for the final analyses had occurred and neither endpoint had reached statistical significance. FDA declined granting accelerated approval based on IA1, IA2, and IA3 results.

Subsequently, at IA4, the trial demonstrated a statistically significant improvement in EFS for the pembrolizumab arm compared with the control arm (HR: 0.63, 95% CI, 0.48–0.82, P = 0.0003). EFS results across key patient subgroups defined by stratification factors and tumor PD-L1 status also favored the pembrolizumab arm. Although OS data remained immature with only 45% of events needed for the final analysis having occurred, they did not indicate a detriment with pembrolizumab treatment. FDA granted traditional approval based on IA4 results. Key efficacy results are shown in Table 2.

Table 2.

Efficacy results from KEYNOTE-522.a

Pembrolizumab + chemotherapy/pembrolizumabPlacebo + chemotherapy/placebo
n = 784n = 390
pCRb 
 Number of patients with pCR 494 217 
 pCR rate (%), (95% CI) 63.0 (59.5–66.4) 55.6 (50.6–60.6) 
 Treatment difference (%), (95% CI) 7.5 (1.6–13.4) 
EFSc 
 Number of patients with event (%) 123 (16%) 93 (24%) 
 HR (95% CI) 0.63 (0.48–0.82) 
P 0.00031 
Pembrolizumab + chemotherapy/pembrolizumabPlacebo + chemotherapy/placebo
n = 784n = 390
pCRb 
 Number of patients with pCR 494 217 
 pCR rate (%), (95% CI) 63.0 (59.5–66.4) 55.6 (50.6–60.6) 
 Treatment difference (%), (95% CI) 7.5 (1.6–13.4) 
EFSc 
 Number of patients with event (%) 123 (16%) 93 (24%) 
 HR (95% CI) 0.63 (0.48–0.82) 
P 0.00031 

aData from U.S. FDA (11).

bThe pCR rate difference reached statistical significance based on a prior prespecified interim pCR analysis in n = 602 patients, with P = 0.00055 compared with a significance level of 0.003.

cEFS reached statistical significance at a prespecified interim analysis (compared with a significance level of 0.0052).

Safety

The safety evaluation of pembrolizumab included 1,167 patients who received at least one dose of study treatment on KEYNOTE-522. Fatal adverse reactions (AR) occurred in 7 (0.9%) of patients who received pembrolizumab and included adrenal crisis, autoimmune encephalitis, hepatitis, pneumonia, pneumonitis, pulmonary embolism, and sepsis. Grade 3–5 ARs occurred in 83% of patients and serious ARs occurred in 44% of patients who received pembrolizumab.

The added toxicity of neoadjuvant and adjuvant pembrolizumab when combined with neoadjuvant chemotherapy was predominantly characterized by immune-mediated ARs (imAR). imARs occurred in 44% of patients who received pembrolizumab compared with 22% of patients who received placebo. The most common imARs in patients who received pembrolizumab were infusion reactions (18%), hypothyroidism (15%), severe skin reactions (6%), hyperthyroidism (5%), and adrenal insufficiency (AI; 3%). A total of 16% of all patients who received pembrolizumab initiated thyroid replacement, and for some patients, the need for thyroid replacement may be lifelong (10).

AI is an important safety signal for patients with high-risk, early-stage TNBC receiving pembrolizumab for neoadjuvant treatment. Almost all patients in this population will receive breast surgery, and undiagnosed AI can cause significant perioperative morbidity and mortality. AI occurred in 3% of patients who received pembrolizumab, including one patient who died from adrenal crisis in the perioperative setting. The Warnings and Precautions section of the pembrolizumab label was updated to recommend serum cortisol monitoring at baseline, prior to surgery, and as clinically indicated for patients receiving neoadjuvant pembrolizumab.

Pembrolizumab was granted traditional approval for patients with high-risk, early-stage TNBC in combination with chemotherapy as neoadjuvant treatment and then as a single agent for adjuvant treatment following surgery. Pembrolizumab is the first non-chemotherapy agent approved for the treatment of high-risk, early-stage TNBC, and the first product approved for treatment of early-stage TNBC since 1999. The basis of the favorable benefit-risk assessment was demonstration of a statistically significant and clinically meaningful improvement in EFS, no detriment in OS, and an acceptable safety profile (Table 3). As EFS is a composite endpoint, FDA considers the different types of EFS events which occur on each arm when interpreting the clinical meaningfulness of an EFS result. In KEYNOTE-522, distant recurrence occurred in 7.7% of patients on the pembrolizumab arm compared with 13.1% of patients on the control arm (13). Sensitivity analyses and subgroup analyses supported the robustness of the EFS results. As EFS results were consistent in subgroups regardless of tumor PD-L1 CPS, approval was granted for all patients. FDA considers EFS an established measure of clinical benefit which can support a traditional approval. Although OS data may be immature at the time of an early-stage breast cancer approval, OS is an important measure of both efficacy and safety and FDA always evaluates for a potential detriment, especially in a curative disease setting (14). Future assessments of OS will occur as part of KEYNOTE-522, and FDA issued a postmarketing commitment to receive the final OS results to further characterize efficacy and safety.

Table 3.

FDA benefit–risk assessment for pembrolizumab for high-risk, early-stage TNBC.a

DimensionEvidence and uncertaintiesConclusions and reasons
Analysis of condition 
  • High-risk, early-stage TNBC is potentially curable. However, compared with the other subtypes of breast cancer, there is an increased risk of metastatic recurrence.

  • Once metastatic, TNBC is incurable and will lead to death.

 
High-risk, early-stage TNBC is a serious and life-threatening condition. 
Current treatment options 
  • Even with intensive treatment, which includes surgery, ±radiotherapy, and multiagent chemotherapy, 5-year EFS or DFS is only ∼70%.

 
Therapies to improve long-term outcomes for patients with high-risk, early-stage TNBC are needed. 
Benefit 
  • KEYNOTE-522 met its EFS endpoint with an HR of 0.63 (95% CI, 0.48–0.82, P = 0.00031) favoring the pembrolizumab arm.

  • EFS results in all subgroups defined by tumor PD-L1 CPS were directionally consistent favoring the pembrolizumab arm.

  • Although OS data are immature, there was no OS detriment.

 
  • Neoadjuvant and adjuvant pembrolizumab demonstrated a statistically significant and clinically meaningful EFS benefit compared with placebo. The OS trend favoring the pembrolizumab arm provides supportive evidence of clinical benefit.

  • Tumor testing for PD-L1 status is not required for safe use of pembrolizumab in patients with high-risk, early-stage TNBC.

 
Risk and risk management 
  • Patients receiving pembrolizumab experienced increased all grade imARs compared with patients receiving placebo: 44% vs. 22%.

  • Undiagnosed AI poses a particular risk to the indicated patient population as almost all patients will undergo surgery.

 
  • The added toxicity from pembrolizumab was acceptable given the EFS benefit demonstrated and the serious and life-threatening nature of high-risk, early-stage TNBC.

  • The Warnings and Precautions section of the label was updated to recommend cortisol monitoring at baseline, prior to surgery, and as clinically indicated.

  • The extended dosing regimen for pembrolizumab: 400 mg i.v. every 6 weeks, is under an accelerated approval for all adult indications and can be used in lieu of the 200 mg i.v. every 3 weeks regimen (19).

 
DimensionEvidence and uncertaintiesConclusions and reasons
Analysis of condition 
  • High-risk, early-stage TNBC is potentially curable. However, compared with the other subtypes of breast cancer, there is an increased risk of metastatic recurrence.

  • Once metastatic, TNBC is incurable and will lead to death.

 
High-risk, early-stage TNBC is a serious and life-threatening condition. 
Current treatment options 
  • Even with intensive treatment, which includes surgery, ±radiotherapy, and multiagent chemotherapy, 5-year EFS or DFS is only ∼70%.

 
Therapies to improve long-term outcomes for patients with high-risk, early-stage TNBC are needed. 
Benefit 
  • KEYNOTE-522 met its EFS endpoint with an HR of 0.63 (95% CI, 0.48–0.82, P = 0.00031) favoring the pembrolizumab arm.

  • EFS results in all subgroups defined by tumor PD-L1 CPS were directionally consistent favoring the pembrolizumab arm.

  • Although OS data are immature, there was no OS detriment.

 
  • Neoadjuvant and adjuvant pembrolizumab demonstrated a statistically significant and clinically meaningful EFS benefit compared with placebo. The OS trend favoring the pembrolizumab arm provides supportive evidence of clinical benefit.

  • Tumor testing for PD-L1 status is not required for safe use of pembrolizumab in patients with high-risk, early-stage TNBC.

 
Risk and risk management 
  • Patients receiving pembrolizumab experienced increased all grade imARs compared with patients receiving placebo: 44% vs. 22%.

  • Undiagnosed AI poses a particular risk to the indicated patient population as almost all patients will undergo surgery.

 
  • The added toxicity from pembrolizumab was acceptable given the EFS benefit demonstrated and the serious and life-threatening nature of high-risk, early-stage TNBC.

  • The Warnings and Precautions section of the label was updated to recommend cortisol monitoring at baseline, prior to surgery, and as clinically indicated.

  • The extended dosing regimen for pembrolizumab: 400 mg i.v. every 6 weeks, is under an accelerated approval for all adult indications and can be used in lieu of the 200 mg i.v. every 3 weeks regimen (19).

 

aData from U.S. FDA (10, 11).

Prior to receiving a traditional approval, the sponsor initially sought accelerated approval based on results from KEYNOTE-522 which showed a pCR rate difference at IA3 of 7.5% (95% CI, 1.6–13.4) between treatment arms. Although the pCR endpoint had reached statistical significance at IA1, this analysis included just half of the randomized population. FDA considers the pCR rates in the entire randomized population at IA3 to be a more accurate representation of pCR rates in a TNBC population. EFS and OS data were immature and there was uncertainty as to whether pembrolizumab would demonstrate an EFS or OS advantage over the control arm because the statistical boundary was not crossed and the pCR rate difference was small. Safety was also considered, and imARs were increased with pembrolizumab. In the metastatic setting, at the time, pembrolizumab was only under an accelerated approval for patients with tumor PD-L1 CPS ≥10 based on demonstration of a moderate improvement in progression-free survival (PFS). OS data in the metastatic setting were also immature (15). In addition, as the pCR endpoint only measured the neoadjuvant portion of treatment and the EFS and OS endpoints, which measured neoadjuvant and adjuvant treatment, were immature, the need for adjuvant pembrolizumab was not demonstrated based only on the pCR endpoint. FDA discussed these earlier pCR results at an Oncologic Drugs Advisory Meeting on February 9, 2021. This meeting allowed for external expert advice to be provided on the application. The committee voted 10-0 to defer an approval decision until EFS and OS data from future analyses were available. Because of the small pCR rate difference of questionable clinical meaningfulness, the added toxicity from pembrolizumab, and lack of compelling evidence of clinical benefit from another treatment setting, FDA considered the overall benefit-risk assessment for neoadjuvant and adjuvant pembrolizumab unfavorable and did not approve the application at this timepoint.

KEYNOTE-522 was designed according to the single trial model, in which the same trial can potentially support an accelerated approval based on results from an earlier endpoint and the conversion to a traditional approval with results from an endpoint available later. This type of model may have applicability beyond the early-stage breast cancer setting. For example, in the metastatic breast cancer setting, a single randomized trial could be designed to assess overall response rate to support an accelerated approval, and PFS or OS to support conversion to a traditional approval. Using a single trial to support accelerated approval and traditional approval may circumvent accrual challenges that occur when two separate trials are used for accelerated approval and traditional approval and ensure that confirmation of clinical benefit occurs in a timely manner.

Despite the positive KEYNOTE-522 results at IA4, there are several important clinical questions that remain unanswered. First, it is unclear whether all patients must receive both neoadjuvant and adjuvant pembrolizumab to experience EFS benefit. In KEYNOTE-522, all patients randomized to the experimental arm were planned to receive neoadjuvant and adjuvant pembrolizumab, and therefore, due to the trial design, the relative contribution of effect of each treatment phase on the EFS (and OS) endpoints cannot be determined. One particularly relevant clinical question is the extent to which patients who experience a pCR benefit from adjuvant pembrolizumab. An exploratory responder analysis of EFS by pCR status suggested that EFS was similarly favorable in all patients experiencing a pCR, regardless of whether they were on the pembrolizumab arm or control arm (10). However, because pCR is a postrandomization outcome variable, this analysis does not preserve randomization. Any interpretation of the association between pCR and EFS is prone to bias and could be due to differences in baseline prognostic factors. Second, it is unknown what magnitude of pCR improvement could translate to an EFS benefit in trials involving pembrolizumab and other immune checkpoint inhibitors for early-stage TNBC. In KEYNOTE-522, patients on the experimental arm received adjuvant pembrolizumab after assessment of the pCR endpoint. It is impossible to determine whether a small pCR improvement is translating to EFS benefit, or whether the EFS endpoint was further modulated by adjuvant pembrolizumab.

Besides pCR, another measure of neoadjuvant treatment effect in patients with breast cancer is the residual cancer burden (RCB). At the individual patient level, across breast cancer subtypes, RCB class may be associated with prognosis (16). However, the association between the different RCB scores and/or classes and long-term outcomes at the trial level is unknown, and FDA does not currently use this endpoint in regulatory decision-making.

Although FDA determined that the added toxicity from neoadjuvant and adjuvant pembrolizumab was acceptable given the EFS benefit, the toxicity of the five-drug KEYNOTE-522 regimen is high. Among patients who received pembrolizumab, 83% experienced a grade 3–5 AR and 44% experienced an IMAR. Because many patients with high-risk early-stage TNBC will experience a cure with standard-of-care chemotherapy alone, most neoadjuvant trials in this disease setting have an add-on design. If these trials are positive, like KEYNOTE-522, they commit the entire patient population to a regimen with higher toxicity. Similar to other adjuvant therapies, this likely represents overtreatment for some patients. To better tailor treatment and minimize toxicity, an adjuvant trial could be conducted in patients who received standard-of-care neoadjuvant treatment and have residual disease at time of surgery (4, 17). Because the trial population would be enriched for patients at higher risk of recurrence, the time to completion could be shorter than a more traditional adjuvant trial. This would speed the availability of new treatments to those patients with the highest unmet need (18).

Pembrolizumab demonstrated favorable benefit-risk in KEYNOTE-522, which supported traditional approval. EFS results were statistically significant and clinically meaningful, there was no OS detriment, and the added toxicity of pembrolizumab was acceptable given the high risk of recurrence. Follow-up of patients for OS will continue in KEYNOTE-522. Pembrolizumab, in combination with chemotherapy for neoadjuvant treatment and then continued as a single agent for adjuvant treatment, represents a new treatment option for patients with high-risk, early-stage TNBC. This group of patients has a high unmet medical need for therapies that improve long-term outcomes, and this is the first treatment approved for early-stage TNBC in over two decades.

No disclosures were reported.

The Editor handling the peer review and decision-making process for this article has no relevant employment associations to disclose.

1.
American Cancer Society
.
Cancer facts and figures 2021
;
2021
.
2.
National Cancer Institute: Surveillance Epidemiology, and End Results Program
.
Cancer stat facts: female breast cancer subtypes
;
2021
.
3.
Sikov
WM
,
Polley
M-Y
,
Twohy
E
,
Perou
CM
,
Singh
B
,
Berry
DA
, et al
.
CALGB (Alliance) 40603: Long-term outcomes (LTOs) after neoadjuvant chemotherapy (NACT) ± carboplatin (Cb) and bevacizumab (Bev) in triple-negative breast cancer (TNBC)
.
J Clin Oncol
37: 15s,
2019
(suppl; abstr 591).
4.
Masuda
N
,
Lee
S-J
,
Ohtani
S
,
Im
Y-H
,
Lee
E-S
,
Yokota
I
, et al
.
Adjuvant capecitabine for breast cancer after preoperative chemotherapy
.
N Engl J Med
2017
;
376
:
2147
59
.
5.
National Comprehensive Cancer Network (NCCN)
.
NCCN clinical practice guidelines in oncology: breast cancer
. Version 2.2022;
2021
.
6.
Amiri-Kordestani
L
,
Wedam
S
,
Zhang
L
,
Tang
S
,
Tilley
A
,
Ibrahim
A
, et al
.
First FDA approval of neoadjuvant therapy for breast cancer: pertuzumab for the treatment of patients with HER2-positive breast cancer
.
Clin Cancer Res
2014
;
20
:
5359
64
.
7.
Cortazar
P
,
Zhang
L
,
Untch
M
,
Mehta
K
,
Costantino
JP
,
Wolmark
N
, et al
.
Pathological complete response and long-term clinical benefit in breast cancer: the CTNeoBC pooled analysis
.
Lancet
2014
;
384
:
164
72
.
8.
Huang
M
,
O'Shaughnessy
J
,
Zhao
J
,
Haiderali
A
,
Cortes
J
,
Ramsey
S
, et al
.
Evaluation of pathologic complete response as a surrogate for long-term survival outcomes in triple-negative breast cancer
.
J Natl Compr Canc Netw
2020
;
18
:
1096
104
.
9.
U.S. Food and Drug Administration
.
Pathological complete response in neoadjuvant treatment of high-risk early-stage breast cancer: use as an endpoint to support accelerated approval
;
2020
.
10.
U.S. Food and Drug Administration
.
Meeting of the Oncologic Drugs Advisory Committee Meeting Announcement
;
2021
.
11.
drugs@FDA
.
KEYTRUDA (pembrolizumab)
.
12.
Schmid
P
,
Cortes
J
,
Pusztai
L
,
McArthur
H
,
Kümmel
S
,
Bergh
J
, et al
.
Pembrolizumab for early triple-negative breast cancer
.
N Engl J Med
2020
;
382
:
810
21
.
13.
Schmid
P
,
Cortes
J
,
Dent
R
,
Pusztai
L
,
McArthur
H
,
Kümmel
S
, et al
.
Event-free survival with pembrolizumab in early triple-negative breast cancer
.
N Engl J Med
2022
;
386
:
556
67
.
14.
Royce
M
,
Osgood
C
,
Mulkey
F
,
Bloomquist
E
,
Pierce
WF
,
Roy
A
, et al
.
FDA approval summary: abemaciclib with endocrine therapy for high-risk early breast cancer
.
J Clin Oncol
2022
;
40
:
1155
62
.
15.
U.S. Food and Drug Administration
.
FDA grants accelerated approval to pembrolizumab for locally recurrent unresectable or metastatic triple negative breast cancer
;
2020
.
16.
Yau
C
,
Osdoit
M
,
van der Noordaa
M
,
Shad
S
,
Wei
J
,
de Croze
D
, et al
.
Residual cancer burden after neoadjuvant chemotherapy and long-term survival outcomes in breast cancer: a multicentre pooled analysis of 5161 patients
.
Lancet Oncol
2022
;
23
:
149
60
.
17.
von Minckwitz
G
,
Huang
C-S
,
Mano
MS
,
Loibl
S
,
Mamounas
EP
,
Untch
M
, et al
.
Trastuzumab emtansine for residual invasive HER2-positive breast cancer
.
N Engl J Med
2018
;
380
:
617
28
.
18.
Prowell
TM
,
Beaver
JA
,
Pazdur
R
.
Residual disease after neoadjuvant therapy—developing drugs for high-risk early breast cancer
.
N Engl J Med
2019
;
380
:
612
5
.
19.
U.S. Food and Drug Administration
.
FDA approves new dosing regimen for pembrolizumab
;
2020
.