KEYNOTE-522: A Biomarker Resource for PD-1 Inhibition in Early Triple-Negative Breast Cancer

Get Permission

In the phase III KEYNOTE-522 trial reported in The New England Journal of Medicine and reviewed in this issue of The ASCO Post, Schmid et al1 found that the addition of pembrolizumab to neoadjuvant chemotherapy in stage II or III triple-negative breast cancer significantly improved the pathologic complete response rates at surgery. This is the first report of a phase III study evaluating a programmed cell death protein 1 (PD-1)-targeting agent in the early-stage setting.

Sherene Loi, MD

Sherene Loi, MD

Pathologic complete response rates were significantly increased in the pembrolizumab group in the intention-to-treat population, with the addition of pembrolizumab being associated with an increase to 64.8%, compared with 51.2% in the control arm (placebo plus paclitaxel and once weekly or every-3-week carboplatin, followed by doxorubicin or epirubicin plus cyclophosphamide). This rate is consistent with previous smaller phase II studies (I-SPY2,2 KEYNOTE-173,3 and GeparNuevo4). There is perhaps a suggestion of more benefit for patients with node-positive vs node-negative disease, as well as in the weekly vs every-3-week carboplatin subgroups. These findings warrant further investigation in similar upcoming studies.

In this early-stage population, approximately 80% of the intention-to-treat population was considered programmed cell death ligand 1 (PD-L1)-positive, compared with approximately 40% in the metastatic population of IMpassion130.5 This latter trial found that atezolizumab improved overall and progression-free survival in the PD-L1–positive population. Unlike in advanced disease, increases in pathologic complete response were observed across PD-L1–positive and PD-L1–negative cohorts.

Differences in Assays and Benefit

Notably, different PD-L1 assays were used in these two studies, so it remains to be seen how they compare head-to-head at the specified current cutoffs: the SP142 assay evaluates PD-L1 expression on immune cells, and the assay generating combined positive score evaluates PD-L1 expression on tumor cells, lymphocytes, and macrophages divided by the total number of tumor cells multiplied by 100. Unlike in other solid cancers, in breast cancer PD-L1 is expressed predominantly on immune rather than tumor cells.6 In previously reported pembrolizumab studies performed in breast cancer, enrichment of preexisting immunity was required, and benefit is increased with higher combined positive PD-L1 scores in the advanced triple-negative breast cancer population.

There are a few possible biologic explanations for this difference. First, in early-stage disease, treatment is more effective, and hence, new tumor antigens can be created. Second, the host immune system is likely more robust as the cancer burden is limited and hence can still mount an effective antitumor immune response to new antigens.

“There is perhaps a suggestion of more benefit for patients with node-positive vs node-negative disease, as well as in the weekly vs every-3-week carboplatin subgroups.”
— Sherene Loi, MD, PhD

Tweet this quote

A preliminary look at event-free survival was performed in the current study (at one-third of the planned events), with fewer distant recurrences seen in the pembrolizumab arm. Further follow-up is necessary, but one would expect effects on event-free survival to be more pronounced than that on pathologic complete response rates, as this phenomenon seems to be a class effect. However, it will be interesting to see whether this holds true also for the PD-L1–negative or PD-L1–low subgroup.

The addition of pembrolizumab was well tolerated, with only slightly more grade 3 or higher adverse events (76.8% vs 72.2%) and more serious treatment-related adverse events (32.5% vs 19.5%). Febrile neutropenia rates were only slightly increased (14.6% vs 12.1%). Notably, immune-related adverse events with pembrolizumab included severe adrenal insufficiency in 1.3% as well as any-grade hyperthyroidism or hypothyroidism in 18.3%; these effects can require lifelong replacement therapy. In this younger population (median age of 49 years), effects on fertility will require further study. Discontinuation of pembrolizumab occurred in 23% of patients.

Optimizing Future Treatment

Given that 48% of the intention-to-treat population had clinically node-negative disease at diagnosis, issues of overtreatment are of concern, especially as carboplatin is not a standard in many parts of the world. Treatment optimization in the future is highly possible, given the rich biomarker resource that this trial will provide.

One could speculate that lower stage, higher tumor-infiltrating lymphocyte content, and higher PD-L1 and CD8 expression may permit less traditional cytotoxic chemotherapies in combination with pembrolizumab; removing the anthracycline with cyclophosphamide will mean the potential long-term effects of cardiac dysfunction and secondary cancers will no longer be of concern. Rapidly advancing technologies that can detect minimal residual disease may also facilitate a shorter duration of pembrolizu­mab after surgery and potentially fewer immune-related adverse events. 

Dr. Loi is Head of the Translational Breast Cancer Genomics and Therapeutics Laboratory, Peter MacCallum Cancer Centre, Melbourne.

DISCLOSURE: Dr. Loi reported institutional research funding from Novartis, Bristol Meyers Squibb, Merck, Roche-Genentech, Puma Biotechnology, Pfizer, Eli Lilly, and Seattle Genetics. She has acted as an uncompensated consultant to Seattle Genetics, Pfizer, Novartis, BMS, Merck, AstraZenca, and Roche-Genentech and as a consultant with honoraria paid to her institution for Aduro Biotech, Novartis, and G1 Therapeutics.


1. Schmid P, Cortes J, Pusztai L, et al: Pembrolizumab for early triple-negative breast cancer. N Engl J Med 382:810-821, 2020.

2. Nanda R, Liu MC, Yau C, et al: Effect of pembrolizumab plus neoadjuvant chemotherapy on pathologic complete response in women with early-stage breast cancer: An analysis of the ongoing phase II adaptively randomized I-SPY2 trial. JAMA Oncol. February 13, 2020 (early release online).

3. Schmid P, Park YH, Muñoz-Couselo E, et al: Pembrolizumab + chemotherapy as neoadjuvant treatment for triple-negative breast cancer: Preliminary results from KEYNOTE-173. 2017 ASCO Annual Meeting. Abstract 556. Presented June 4, 2017.

4. Loibl S, Untch M, Burchardi N, et al: A randomised phase II study investigating durvalumab in addition to an anthracycline taxane-based neoadjuvant therapy in early triple-negative breast cancer: Clinical results and biomarker analysis of GeparNuevo study. Ann Oncol 30:1279-1288, 2019.

5. Schmid P, Adams S, Rugo HS, et al: Atezolizumab and nab-paclitaxel in advanced triple-negative breast cancer. N Engl J Med 379:2108-2121, 2018.

6. Savas P, Virassamy B, Ye C, et al: Single-cell profiling of breast cancer T cells reveals a tissue-resident memory subset associated with improved prognosis. Nat Med 24:986-993, 2018.

Related Articles

Neoadjuvant and Adjuvant Pembrolizumab Improves Outcomes in Early Triple-Negative Breast Cancer

As reported in The New England Journal of Medicine by Peter Schmid, MD, PhD, of Barts Cancer Institute, Queen Mary University of London, and colleagues, analyses in the phase III KEYNOTE-522 trial have shown that the addition of pembrolizumab to neoadjuvant chemotherapy and the use of adjuvant...