Pathological complete response with neoadjuvant pembrolizumab and chemotherapy in non-metastatic triple-negative inflammatory breast cancer.

Introduction
Inflammatory breast cancer (IBC) is an aggressive presentation of breast cancer that accounts for 2%-4% of breast cancer cases.1 Patients with non-metastatic IBC are treated with neoadjuvant systemic therapy, modified radical mastectomy, and post-mastectomy radiotherapy.1 Despite the unique biological features of IBC, systemic therapy options are based on standards of care established for patients with non-IBC.
Most patients with non-metastatic triple-negative IBC (TN-IBC) currently receive neoadjuvant pembrolizumab in combination with multi-agent chemotherapy, which is standard of care for patients with high-risk, triple-negative breast cancer (TNBC), according to the results of the KEYNOTE-522 trial.2 However, this pivotal trial only included 17 patients with IBC (1.4%)2,3 and no data are available for this subgroup of patients.
The aim of this study was to evaluate the activity of neoadjuvant chemoimmunotherapy in patients with TN-IBC.

Patients and methods
We conducted a retrospective/prospective, international, multi-cohort study of patients enrolled in three retrospective cohorts [Dana-Farber Cancer Institute (DFCI), MD Anderson Cancer Center (MDACC), and INCORPORATE cohorts] and one prospective clinical trial (PELICAN-IPC 2015-16/Oncodistinct-003, NCT03515798), which randomly assigned patients with human epidermal growth factor receptor 2 (HER2)-negative IBC to neoadjuvant chemotherapy [dose-dense epirubicin plus cyclophosphamide (EC), followed by weekly paclitaxel] with or without pembrolizumab every 3 weeks.4
Patients aged ≥18 years, with a clinical diagnosis of non-metastatic (stage III) TN-IBC, treated with at least one cycle of neoadjuvant chemotherapy plus pembrolizumab were included. TN-IBC was defined as cT4d per American Joint Committee on Cancer (AJCC) TNM (tumor–node–metastasis) (eighth edition),1 estrogen receptor and progesterone receptor-negative (<1% by immunohistochemistry in DFCI/MDACC/INCORPORATE and <10% in PELICAN), and HER2-negative per American Society of Clinical Oncology (ASCO)/College of American Pathologists (CAP) guidelines. Patients with secondary IBC were excluded.
The study was approved by each institution’s institutional review board and was conducted in accordance with the principles stated in the Declaration of Helsinki and with the principles of good clinical practice.
The primary endpoint was pathological complete response (pCR; i.e. ypT0/ypTis ypN0), and the secondary endpoints were residual cancer burden (RCB), the number of locoregional or distant recurrences, and deaths. For our sample size of 63 patients, the half-width of an asymptotic 95% confidence interval (CI) would be ±11.3% for a pCR rate of 30%. This provides adequate precision in the context of an assumed pCR rate of 18% for patients with non-metastatic TN-IBC treated with neoadjuvant chemotherapy, based on a recently published study using data from the National Cancer Database,5 and an assumed improvement of 13% with the addition of pembrolizumab, consistent with the results observed among all patients enrolled in the KEYNOTE-522 trial.6
Categorical variables were expressed as numbers and proportions. pCR was expressed as a proportion with exact binomial 95% CI. Data for each cohort were collected under different circumstances, such that the duration of follow-up is different in each cohort. Median follow-up was estimated from the overall survival censoring distribution (inverse censoring method). Statistical analyses were performed using SAS Enterprise Guide version 7.15 (SAS Institute Inc., Cary, NC).

Results
Sixty-three patients were included (Table 1). Patients initiated neoadjuvant treatment between September 2018 and April 2023. The most common regimen received was pembrolizumab, taxane/carboplatin, and anthracycline/cyclophosphamide (61.9%), followed by pembrolizumab plus taxane/carboplatin (25.4%). The median number of doses of neoadjuvant pembrolizumab (200 mg every 3 weeks or 400 mg every 6 weeks) was 4 (interquartile range 4-6).
Overall, 59 patients (94%) underwent surgery and 4 (6.4%) experienced local and/or distant disease progression during the neoadjuvant treatment. The pCR rate was 31.7% (20/63; 95% CI 20.6% to 44.7%) (Table 2). Among 46 patients who had RCB available, 1 additional patient had RCB 1. Among patients who received the KEYNOTE-522 regimen [n = 39; i.e. EC/adriamycin/cyclophosphamide (AC) followed by carboplatin and paclitaxel], the pCR rate was 35.9% (14/39; 95% CI 21.2% to 52.8%), as compared with 18.8% (3/16; 95% CI 4.1% to 45.7%) among those who received the same chemotherapy regimen without carboplatin (n = 16).
After a median follow-up of 1.4 years (95% CI 1.2-1.7 years), local/regional recurrence occurred in 6 patients and distant recurrence occurred in 16 patients. Twelve patients have died.

Discussion
In this multi-cohort study, we evaluated the activity of neoadjuvant chemoimmunotherapy in patients with stage III TN-IBC from three retrospective cohorts and one phase II trial.
We found a pCR rate of 31.7%, an improvement of 13.7% over the historical pCR rate of 18.0% observed among patients with non-metastatic TN-IBC who received neoadjuvant chemotherapy without carboplatin and pembrolizumab.5 Despite the limitations of cross-study comparisons, this result was similar to the absolute difference in pCR rates of 13.6% reported in KEYNOTE-522, favoring the pembrolizumab-containing arm (64.8% versus 51.2%).6
A possible explanation for the lower pCR rate of 17.6% observed in the PELICAN study is the absence of carboplatin in the chemotherapy backbone. In contrast, most patients from the other three cohorts (84.8%) received carboplatin, which was previously shown to significantly improve the pCR rate by ∼15% versus non-platinum-containing chemotherapy in patients with early-stage non-inflammatory TNBC.7 Importantly, the historical pCR rate of 18%5 did not include carboplatin, confounding the attribution of the increased pCR rate to pembrolizumab.
Prospective clinical data regarding the safety and efficacy of immune checkpoint inhibitors (ICIs) in patients with IBC are limited.8 There are few trials focusing on patients with IBC, and many existing trials have been hampered by slow accrual.8,9 Similarly, the phase II PELICAN trial closed accrual when chemoimmunotherapy became standard of care for patients with TNBC. In general, reasons for slow accrual include the following: the rarity of the disease; lack of referral of patients in centers with active clinical trials; the aggressive presentation of the disease; and the finding of N3 or limited ‘regional M1’ nodal disease (e.g. cervical or contralateral axillary nodal involvement). Given the dearth of prospective clinical trials in IBC, multicenter, international retrospective studies are necessary to learn more about the efficacy of novel treatment strategies in this rare and clinically challenging setting.
The pCR rate observed in this study for patients with stage III IBC remains substantially lower than the pCR rate of the KEYNOTE-522 trial (65%), due to the intrinsically lower responsiveness of IBC to standard regimens and possibly also due to the higher tumor burden at baseline compared with early-stage non-inflammatory TNBC.6 Similarly, in a multicenter cohort including 41 patients with metastatic TN-IBC, we did not observe a higher magnitude of clinical benefit provided by the addition of ICI to first-line chemotherapy.10 This seems to be in contrast with the preclinical evidence of a higher immune vulnerability of IBC compared with non-IBC. Indeed, the IBC immunogram suggests the existence of a preexisting immune tumor microenvironment that is suppressed by mechanisms of immune escape, but that might be restored by ICIs.11 A collaborative effort is required to address the potential immune vulnerability of IBC, targeting pathways driving the resistance to ICIs or bolstering the activity of ICIs, for example through epidermal growth factor receptor inhibitors (e.g. panitumumab) or antibody–drug conjugates.12 In this regard, three ongoing clinical trials are assessing the combination of panitumumab (NCT05177796), sacituzumab govitecan (NeoSTAR; NCT04230109), or trastuzumab deruxtecan (TRUDI; NCT05795101) with neoadjuvant immunotherapy in patients with IBC.13,14
The limitations of our analysis include the retrospective observational design of the study, lack of a direct comparison arm, small sample size, difficulty in dissecting the specific contribution of immunotherapy versus carboplatin to the pCR, and lack of data on baseline cN status and post-neoadjuvant treatments. However, this study is the first analysis assessing the addition of an ICI to neoadjuvant chemotherapy in patients with non-metastatic TN-IBC in international referral centers.
In conclusion, pembrolizumab with multi-agent chemotherapy might provide additional benefit beyond chemotherapy alone for patients with stage III TN-IBC.