Immunotherapy for hormone receptor‒positive HER2-negative breast cancer

Breast cancer is the most common cancer worldwide and the fifth leading cause of cancer death¹. In 2020, there were ∼2.3 million new global cases of breast cancer among women and 682,000 associated deaths¹. Hormone receptor–positive (HR +)/human epidermal growth factor receptor 2–negative (HER2 −) is the most common subtype, accounting for ∼70% of all breast cancer cases^2,3,4,5,6. Notably, since 2016 in the United States, HR+ breast cancer incidence rates have increased dramatically among young women between 20 and 49 years of age⁷. This highlights the need for further research to identify specific risk factors of breast cancer and possible strategies for early detection and prevention in the younger population⁷. Early-stage HR + /HER2− breast cancer is considered potentially curable⁸. Mainstays of treatment include breast surgery with or without radiation and endocrine therapy with or without adjuvant CDK4/6 inhibitor for node-positive disease^9,10,11. Patients with high-risk, early-stage HR + /HER2− disease, defined by clinicopathologic features and/or genomic assays, are good candidates for (neo)adjuvant chemotherapy followed by adjuvant endocrine therapy for ≥5 years with or without adjuvant abemaciclib for 2 years^9,10,11. Genomic assays are used to identify patients who can forego chemotherapy without sacrificing long-term outcomes¹². In contrast, advanced or metastatic (hereafter referred to as metastatic) breast cancers are almost always incurable, whereby the main goals of therapy are to prolong survival and maintain or improve patient quality of life^9,13. For the majority of metastatic HR + /HER2− disease, initial treatment is a combination of endocrine therapy plus a cyclin-dependent kinase (CDK) 4/6 inhibitor^9,13,14. Sequential mono-chemotherapy or antibody-drug conjugates (ADCs) are reserved for patients who have exhausted and/or are refractory to all available endocrine treatment options and for patients with visceral crisis^9,13,14. The RIGHT Choice study compared outcomes in patients with clinically aggressive HR + /HER2− advanced breast cancer who were treated with the combination of endocrine therapy plus the CDK4/6 inhibitor ribociclib versus combination chemotherapy¹⁵. In a subgroup analysis, the hazard ratio (HR) for progression-free survival (PFS) was 0.953 (95% CI, 0.574−1.582) in patients with investigator-assessed visceral crisis, although symptomatic adverse event (AE) rates were lower in those receiving endocrine therapy plus CDK4/6 inhibitor¹⁵.

The prognosis among women with HR + /HER2− breast cancer varies by stage^2,6 and histological grade¹⁶. Five-year survival rates are estimated at >90% for stage I/II disease, 82% for stage III disease, and 36% for stage IV disease⁶, and patients with grade 3 versus grade 1/2 tumors have a poorer prognosis¹⁶. Unfortunately, despite available treatment options, disease recurrence is common^17,18 and increases with disease stage and tumor grade^19,20. These findings underscore the need for additional therapies to improve outcomes in HR + /HER2− breast cancer. Several treatment strategies have been investigated with variable success²¹, while others are being evaluated in ongoing studies^22,23. A key area of focus has been the evaluation of monoclonal antibodies targeting the programmed death 1 pathway by binding to programmed cell death protein 1 (PD-1) or its ligand programmed cell death ligand 1 (PD-L1) and activating an antitumor immune response. The addition of the anti−PD-1 monoclonal antibody pembrolizumab to chemotherapy significantly improved outcomes (pathological complete response [pCR], event-free survival [EFS], and overall survival [OS]) in patients with triple-negative breast cancer (TNBC) in the early-stage setting^24,25,26 and PFS and OS in patients with PD-L1 combined positive score (CPS) ≥ 10 in the metastatic setting^27,28. Mixed results have been observed with the addition of the anti−PD-L1 monoclonal antibody atezolizumab to chemotherapy in the early-stage TNBC setting, with significant improvement in pCR when added to neoadjuvant chemotherapy²⁹ but no improvement in invasive disease-free survival when added to adjuvant chemotherapy³⁰. Agents targeting PD-(L)1 are currently being assessed for the treatment of HR + /HER2− breast cancer³¹. This article reviews the rationale for immunotherapy in HR + /HER2− breast cancer, current knowledge and ongoing studies on the use of immunotherapy in this population, including the potential role of biomarkers, and discusses how these agents might be implemented into clinical practice in the future if supported by clinical data.

The identification of tumor-infiltrating lymphocytes (TILs) and PD-1/PD-L1 expression in breast cancer tissue has suggested the potential for benefit with immunotherapy in this setting³². Rates of TILs and lymphocytic cell subsets vary considerably within and between breast cancer subtypes. In a pooled analysis of patients with previously untreated breast cancer, intermediate (defined as 11%−59%) to high levels (defined as ≥60%) of TILs were found in 44% (607/1366) of patients with HR + /HER2− breast cancer compared with 56% (774/1379) and 71% (646/906) of patients with HER2+ and TNBC, respectively³³. A systematic review of 15 studies in which chemotherapy was administered in the (neo)adjuvant setting found a similar pattern of TILs across breast cancer subtypes³⁴. Among HR + /HER2− tumors, it was reported that a median of 43% had cytotoxic CD8⁺ T-cell infiltration and a median of 38% had high FOXP3⁺ T-cell infiltration. Corresponding results for the other breast cancer subtypes ranged from 60−61% and 67−70%, respectively³⁴. Another analysis of 458 tumor tissue samples from patients with HR + /HER2− breast cancer found that in the adjuvant setting, high levels of TILs (defined as ≥40%) were associated with larger tumors, higher tumor grade, and high proliferation ( ≥ 15% Ki67 expression)³⁵. Within this subtype, tumors with high levels of TILs often had higher gene expression levels of CD8, CD19, PD-1, and PD-L1³⁵.

Although perhaps the least immunogenic of the breast cancer subtypes, as mentioned previously, HR + /HER2− is by far the most common. The population with HR + /HER2− breast cancer and TILs therefore exceeds the combined population with HER2+ breast cancer or TNBC and TILs, thereby representing a large group of patients who could potentially benefit from immunotherapy³². Additionally, it may be possible to combine immunotherapy with other agents to enhance the antitumor response. There is increasing evidence, for example, that chemotherapy may increase the antitumor activity of immune-checkpoint inhibitors. In addition to causing tumor cell death, certain conventional chemotherapies may also have immunogenic effects³⁶. These chemotherapy agents can lead to “immunogenic cell death” provided they are given in an immunocompetent environment³⁶. Furthermore, perioperative treatment with taxanes has been shown to enhance T-cell and natural killer cell functions in patients with early breast cancer, contravening previous assumptions that taxanes suppress patients’ immune cell functions³⁷. Chemotherapy has also been shown to upregulate PD-L1 expression on some tumor cells, potentially sensitizing tumors to treatment with immune-checkpoint inhibitors³⁸. Additionally, the cytotoxic effects of chemotherapy are believed to expose the immune system to high levels of tumor antigens, which might also lead to an enhanced immune response³⁶. CDK4/6 inhibitors have also demonstrated immunomodulatory effects in multiple tumor types^39,40, including reductions in circulating regulatory T cells and myeloid-derived suppressor cells and increases in CD4 + T cells in patients with HR + /HER2− metastatic breast cancer³⁹.

Completed studies in early-stage HR + /HER2− breast cancer

Preliminary evidence for immunotherapy in early-stage HR + /HER2− breast cancer is summarized in the following sections and Table 1.

Pembrolizumab

The earliest indication of the potential benefit of adding pembrolizumab to neoadjuvant chemotherapy was derived from the phase 2, adaptively randomized, open-label I-SPY2 study in patients with previously untreated, early-stage, high-risk HER2− breast cancer⁴¹. The HR + /HER2− population comprised 40 patients who received neoadjuvant pembrolizumab plus chemotherapy and 96 patients who received neoadjuvant chemotherapy alone (see Table 1 for details on regimens). The primary endpoint of pCR (ypT0/Tis ypN0) in this population was estimated to be 30% and 13%, respectively. Based on these results, the probability of combination therapy being superior to chemotherapy alone was >99%. In I-SPY2, enrollment in a particular experimental group was stopped when Bayesian predictive probability of success reached a prespecified efficacy threshold (85% in this case) for any biomarker signature (HR + /HER2− with high risk of relapse per MammaPrint in this case) in a hypothetical confirmatory phase 3 study; at that time, the experimental arm was deemed to have “graduated” from the study. The predictive probability of success with pembrolizumab plus chemotherapy was >99%, thereby meeting the prespecified efficacy threshold. The safety profile of the combination was consistent with that of the individual components⁴¹. These results helped provide the rationale for further investigation of neoadjuvant and adjuvant pembrolizumab combination therapy in the ongoing phase 3 KEYNOTE-756 study (see Ongoing studies in early-stage HR + /HER2− breast cancer).

Nivolumab

The phase 2, single-arm, open-label GIADA study was conducted in patients with previously untreated, early-stage HR + /HER2− breast cancer (histologically confirmed luminal B-like only)⁴². All patients received neoadjuvant chemotherapy followed by nivolumab; triptorelin was started concomitant to chemotherapy, and exemestane was started concomitant to nivolumab. The primary endpoint of pCR was achieved in 16.3% of patients (7/43). The statistical hypothesis of ≥8 pCR of 43 patients was not met⁴². The phase 1b/2, open-label CheckMate 7A8 study examined neoadjuvant nivolumab plus palbociclib and anastrozole in estrogen receptor–positive (ER + )/HER2− breast cancer^43,44. The study was closed due to a higher than expected incidence of grade ≥3 treatment-related hepatic AEs^43,44. Neoadjuvant and adjuvant nivolumab combination therapy is being evaluated in the phase 3 CheckMate 7FL study (see Ongoing studies in early-stage HR + /HER2− breast cancer).

Durvalumab

The I-SPY2 study also evaluated the potential benefit of adding neoadjuvant durvalumab plus the poly(ADP-ribose) polymerase inhibitor olaparib to paclitaxel followed by treatment with doxorubicin plus cyclophosphamide⁴⁵. In the HR + /HER2− population, 52 patients received neoadjuvant durvalumab, olaparib, and chemotherapy and 157 patients received neoadjuvant chemotherapy alone. Although combination therapy initially graduated, the final estimated pCR rates after all patients completed surgery were 28% and 14%, respectively. This corresponded to 75% predictive probability of success in a hypothetical confirmatory phase 3 study, which failed to meet the prespecified efficacy threshold of 85%. No unexpected safety findings were observed with triple therapy⁴⁵. Neoadjuvant durvalumab plus chemotherapy is being evaluated in the ongoing phase 1/2 B-IMMUNE study in 57 patients with either localized luminal B HER2− breast cancer or TNBC⁴⁶, but no phase 3 studies are in progress with durvalumab plus standard chemotherapy in HR + /HER2− disease.

Biomarkers

Biomarker analysis from I-SPY2 revealed that the benefit of adding pembrolizumab⁴⁷ or durvalumab plus olaparib⁴⁵ to neoadjuvant chemotherapy was only apparent in the MammaPrint (panel of 70 genes signature associated with breast cancer recurrence) ultra-high risk (MP2, MP score > −0.154) subset but not in the MammaPrint high-risk (MP1, MP score ≤ −0.154) subset. Specifically, in the MP2 subset, the estimated pCR rates were ∼3-fold greater with combination therapy versus chemotherapy alone. These results suggest that, in addition to its prognostic value, MammaPrint may predict response to these combination therapies in patients with early-stage HR + /HER2− breast cancer. A range of gene expression signatures associated with pCR were identified in the I-SPY2 study⁴⁸. The 53-gene signature ImPrint was developed retrospectively from these initial findings and features genes with immune-related functions known to be involved in immune response (eg, PD-1 and PD-L1), and was identified as having high sensitivity and specificity for predicting pCR to immunotherapy in patients with HR + /HER2− breast cancer⁴⁹. Further assessment of this population in I-SPY2 showed pCR rates of 76% for ImPrint+ and 16% for ImPrint− in patients treated with any PD-1 or PD-L1 inhibitor (alone or as part of combination therapy) compared with 33% and 21%, respectively, in the control arm⁵⁰. Although promising, results from randomized trials (eg, I-SPY2.2) are needed to validate its clinical utility as well as its predictive and prognostic value. Additionally, in I-SPY2, major histocompatibility complex class II molecules on tumor cells were associated with improved outcomes with the addition of immunotherapy to neoadjuvant chemotherapy⁵¹. In GIADA⁴², a combined score of TILs and PAM50 basal molecular subtype showed an area under the curve of 0.95 for pCR prediction. No data were reported on PD-L1 as a biomarker in any of these studies.

Ongoing studies in early-stage HR + /HER2− breast cancer

There are 4 ongoing phase 3 studies, which are described below and in Table 2.

KEYNOTE-756

KEYNOTE-756 (NCT03725059) is a phase 3, randomized, double-blind study of neoadjuvant pembrolizumab (vs placebo) plus chemotherapy followed by adjuvant pembrolizumab (vs placebo) plus endocrine therapy in adult patients with newly diagnosed, previously untreated, high-risk, early-stage ER + /HER2− breast cancer regardless of PD-L1 status⁵². Patients with invasive ductal histology of grade 3 and T1c−T2 (tumor size ≥2 cm)/cN1−cN2 or T3−T4/cN0−cN2 ER + /HER2− breast cancer were randomized 1:1 to receive pembrolizumab or placebo⁵². Randomization was stratified by geographic region (Eastern Europe/China/all other countries), PD-L1 expression (CPS ≥ 1/ < 1), lymph node involvement (positive/negative), ER positivity ( ≥ 10%/ < 10%), and neoadjuvant chemotherapy dosing schedule (every 2 weeks/every 3 weeks). In the neoadjuvant phase, patients received pembrolizumab or placebo in combination with paclitaxel from cycles 1 to 4 and in combination with doxorubicin (or epirubicin) plus cyclophosphamide from cycles 5 to 8 (see Table 2 for doses). After definitive surgery, patients were treated with adjuvant pembrolizumab or placebo in combination with investigator’s choice of endocrine therapy for 9 cycles followed by investigator’s choice of endocrine therapy for up to 10 years. At the investigator’s discretion, adjuvant treatment could be started either concurrently with radiation therapy or sequentially following radiation therapy. The primary endpoints were pCR (ypT0/Tis ypN0) and EFS in the overall population; key secondary endpoints included pCR and EFS in patients with PD-L1−positive tumors and OS in the overall population and in patients with PD-L1−positive tumors⁵². The rationale for pCR as an endpoint is that pCR is a surrogate marker for improved long-term outcomes (EFS and OS) following neoadjuvant chemotherapy in patients with breast cancer^53,54, and among patients with HR + /HER2− breast cancer, this correlation is stronger in patients with grade 3 versus lower-grade tumors⁵³.

A total of 1278 patients were randomized in KEYNOTE-756 (pembrolizumab arm, n = 635; placebo arm, n = 643)^55,56. At the first interim analysis (final pCR analysis), median follow-up was 33.2 months (range, 9.7–51.8 months). Neoadjuvant pembrolizumab plus chemotherapy showed a statistically significant improvement in pCR compared with neoadjuvant placebo plus chemotherapy (24.3% vs 15.6%, P = 0.00005). Safety profiles were consistent with the known profiles of each regimen. Results for EFS are immature and continue to be evaluated^55,56.

CheckMate 7FL

CheckMate 7FL (NCT04109066) is a phase 3, randomized, double-blind study of neoadjuvant nivolumab (vs placebo) plus chemotherapy followed by adjuvant nivolumab (vs placebo) plus endocrine therapy in adult patients with newly diagnosed, previously untreated, high-risk, early-stage ER + /HER2− breast cancer regardless of PD-L1 status^57,58. Patients had localized invasive ductal carcinoma of stages T1c–T2 (tumor size ≥2 cm)/cN1–N2 or T3–T4/cN0–cN2, grade 2 (ER 1–10%) or 3 (ER ≥ 1%) ER + /HER2− breast cancer and were eligible for definitive surgery. Of 830 patients screened, 521 were randomized 1:1 to receive either nivolumab 360 mg every 3 weeks or 240 mg every 2 weeks (arm A, n = 263) or placebo (arm B, n = 258) combined with neoadjuvant chemotherapy followed by adjuvant nivolumab 480 mg every 4 weeks or placebo plus adjuvant endocrine therapy (see Table 2 for dosing details). Patients were stratified by PD-L1 expression, tumor grade (2/3), axillary nodal status (positive/negative), and chemotherapy schedule (every 3 weeks/every 2 weeks)^57,58.

The study was closed to new enrollment in April 2022 based on external data^58,59. The primary endpoint was amended to evaluate only pCR (ypT0/Tis ypN0) in the modified intent-to-treat (ITT) population (arm A, n = 257 and arm B, n = 253). Neoadjuvant nivolumab plus chemotherapy showed a statistically significant improvement in pCR versus neoadjuvant placebo plus chemotherapy (24.5% vs 13.8%; odds ratio, 2.05; 95% CI, 1.29–3.27; P = 0.0021). The benefit of neoadjuvant nivolumab plus chemotherapy on pCR was greater in patients with PD-L1–positive tumors and stromal TILs >1%, whereas Ki67 index was not associated with benefit. In the neoadjuvant phase, treatment-related grade 3 or 4 AEs were similar between the 2 groups (35% vs 32%). There were 2 treatment-related AEs (pneumonitis, hepatitis) leading to death in the nivolumab plus chemotherapy group. As expected, the incidence of immune-mediated AEs was higher in the nivolumab plus chemotherapy group^58,59. The study was further amended to evaluate EFS as an exploratory endpoint with about 1 year of follow-up with adjuvant open-label design⁵⁸.

Southwest Oncology Group (SWOG) S2206

SWOG S2206 (NCT06058377) is a phase 3, randomized, open-label study of neoadjuvant durvalumab plus chemotherapy compared with neoadjuvant chemotherapy alone in patients with ultra-high risk (MammaPrint MP2), early-stage HR + / HER2− breast cancer⁶⁰. Approximately 3680 patients are expected to be enrolled. The primary endpoint is breast cancer EFS up to 10 years after completion of study treatment. Secondary endpoints include pCR, residual cancer burden, distant relapse-free survival, and OS. The estimated study completion date is May 2026⁶⁰.

TROPION-Breast04

TROPION-Breast04 (NCT06112379) is a phase 3, randomized, open-label study of neoadjuvant durvalumab plus datopotamab deruxtecan (an anti-trophoblast cell surface antigen 2 [TROP2] ADC) followed by adjuvant durvalumab with or without chemotherapy compared with neoadjuvant pembrolizumab plus chemotherapy followed by adjuvant pembrolizumab with or without chemotherapy in patients with previously untreated, stage II or III TNBC or HR-low (ER and/or progesterone receptor levels 1% to <10%)/HER2− breast cancer⁶¹. Approximately 1728 patients are expected to be enrolled. The primary endpoints are pCR at the time of definitive surgery and EFS; OS is a key secondary endpoint. The estimated study completion date is August 2030⁶¹.

Apart from the 4 phase 3 studies in HR + /HER2− breast cancer, the ongoing phase 2 Neo-CheckRay study (NCT03875573) is evaluating whether the addition of durvalumab or durvalumab plus oleclumab (adenosine pathway inhibitor) to stereotactic body radiation therapy plus chemotherapy can improve tumor response in patients with early-stage ER + /HER2− breast cancer⁶². Preliminary results from the safety run-in demonstrated that this novel combination was feasible⁶². The ongoing phase 2 I-SPY2.2 study (NCT01042379) is evaluating neoadjuvant datopotamab deruxtecan in patients with breast cancer, including HR + /HER2– disease⁶³. After 4 cycles of treatment, 32% of patients were able to skip traditional chemotherapy and progress to surgery, although no subtype met the statistical threshold for further assessment in this setting⁶³.

Completed studies in metastatic HR + /HER2− breast cancer

Preliminary evidence for immunotherapy in metastatic HR + /HER2− breast cancer is summarized in the following sections and Table 3.

Pembrolizumab

Results from the phase 1b, nonrandomized, open-label KEYNOTE-028 study provided the earliest evidence of a potential benefit from immunotherapy in patients with metastatic ER + /HER2− breast cancer⁶⁴. In that study, the breast cancer cohort had heavily pretreated, locally advanced or metastatic disease and PD-L1 CPS ≥ 1. Twenty-five patients were treated with pembrolizumab monotherapy. In this population, all patients had previously received chemotherapy (32%, 3 or 4 lines; 48%, ≥5 lines) and 88% had previously received endocrine therapy. The objective response rate (ORR; primary endpoint) was 12%, with no complete responses (CRs) and 3 partial responses (PRs). The 3 patients who achieved PR had previously received ≥3 lines of chemotherapy for metastatic disease. An additional 4 patients had stable disease (SD), which lasted for ≥24 weeks in 2 patients. The results demonstrated that pembrolizumab provided durable overall responses in some patients. Most treatment-related AEs were grade 1 or 2 and manageable⁶⁴.

The modest responses observed with pembrolizumab monotherapy in KEYNOTE-028 led to further assessment of pembrolizumab in combination with other agents^{65,66,67,68,69,70,71,72}. The only study with a comparator arm was a phase 2, randomized, open-label study in patients with metastatic HR + /HER2− breast cancer⁶⁵. In that study, patients were treated with either the microtubule inhibitor eribulin alone (n = 44) or in combination with pembrolizumab (n = 44). The majority of patients had previously received endocrine therapy (81%), CDK4/6 inhibitor therapy (76%), and/or chemotherapy (61%). The primary endpoint of PFS was not different between the eribulin and combination therapy arms (median, 4.2 vs 4.1 months; P = 0.33). The study did, however, have several limitations, including the fact that PD-L1 testing was performed mostly on primary rather than metastatic tissue (69% vs 31%), only 37% of tested tumors were PD-L1 positive, and exploratory analysis related to TILs and PD-L1 status was not possible because of missing tumor samples⁶⁵.

Other phase 1/2, single-arm, open-label studies evaluated various pembrolizumab combination therapies in patients with metastatic HR + /HER2− breast cancer, including 3 studies with chemotherapy (eribulin⁶⁶, capecitabine⁶⁷, and paclitaxel⁶⁸), 2 with a CDK4/6 inhibitor (palbociclib plus hormone therapy⁶⁹ and abemaciclib with or without hormone therapy⁷⁰), 1 with hormone therapy alone (aromatase inhibitors)⁷¹, and 1 with radiotherapy⁷². Although the lack of a control arm precludes any definitive conclusions, some patients derived benefit in these small studies. In the largest⁷⁰, 26 patients who had not received prior systemic anticancer therapy in the metastatic setting were treated with pembrolizumab, abemaciclib, and anastrozole and 28 patients with heavily pretreated disease received pembrolizumab plus abemaciclib. ORRs in these 2 cohorts were 23.1% and 28.6%, respectively. Although antitumor activity was demonstrated, the combination of pembrolizumab plus abemaciclib with or without anastrozole led to higher than anticipated rates of interstitial lung disease/pneumonitis (13.0%) and grade ≥3 aspartate or alanine aminotransferase elevations (each 25.9%)⁷⁰. In contrast, no safety concerns were identified with the combination of pembrolizumab, palbociclib, and letrozole⁶⁹ or any of the other combination therapies^{65,66,67,68,71,72}. Based on the collective evidence with pembrolizumab and nivolumab (CheckMate 7A8)^43,44, combination therapy with PD-(L)1 and CDK 4/6 inhibitors is not being pursued in patients with HR + /HER2− breast cancer.

Further investigation of pembrolizumab combination therapy is being evaluated in the ongoing phase 3 KEYNOTE-B49 study (see Ongoing studies in metastatic HR + /HER2− breast cancer).

Nivolumab

The NEWBEAT study was a phase 2, single-arm, open-label trial in patients with metastatic HER2− breast cancer, including 39 patients with HR + /HER2− disease⁷³. In the HR + /HER2− population, most patients (67%) had previously received chemotherapy, endocrine therapy, or both in the adjuvant or neoadjuvant setting. All patients were treated with nivolumab, paclitaxel, and bevacizumab. The primary endpoint of ORR was 74%, with no CRs and 29 PRs; the remaining 10 patients achieved SD. The safety profile of combination therapy was consistent with those of the individual components. The authors concluded that the results warrant further evaluation of this triple therapy in a phase 3 study⁷³, although none appear to have been initiated based on a search of ClinicalTrials.gov.

In a phase 1, single-arm, open-label study of nivolumab plus trastuzumab deruxtecan in HER2-expressing metastatic breast cancer, the HER2-low cohort (n = 16 with immunohistochemistry score of 1+ [or 2+ without gene amplification by in situ hybridization]) had an ORR (primary endpoint) of 50%⁷⁴. None of the 16 patients were reported to have experienced treatment-related interstitial lung disease⁷⁴. In the phase 3 DESTINY-Breast04 study, trastuzumab deruxtecan monotherapy resulted in treatment-related interstitial lung disease/pneumonitis in 12.1% (45/373) of patients with HER2-low metastatic breast cancer⁷⁵.

Durvalumab

The SAFIR02-BREAST IMMUNO study was a phase 2, randomized, open-label trial in patients with metastatic HER2− breast cancer, including 108 patients with HR + /HER2− disease⁷⁶. Patients with HR+ disease were eligible if they experienced relapse or disease progression during endocrine therapy or <12 months after the end of endocrine therapy in the adjuvant setting. All patients were treated with 4–8 cycles of chemotherapy (unspecified). Those without disease progression and without actionable genomic alterations were then treated with maintenance durvalumab or the same chemotherapy regimen. In the HR + /HER2− population, PFS (primary endpoint) and OS were longer with maintenance chemotherapy than maintenance durvalumab. AEs were presented for the overall HER2− population and were consistent with previous reports. In the durvalumab group, the only grade 3 or 4 treatment-related AEs occurring in >1 patient were hepatocellular injury (2.3%) and hypothyroidism (1.6%)⁷⁶.

A phase 2, single-arm, open-label study in patients with advanced solid tumors, including 9 patients with HR + /HER2− breast cancer, evaluated the combination of durvalumab and the oral hypomethylating agent azacitidine (also known as CC-486)⁷⁷. No objective responses were observed. Additionally, the higher dose of azacitidine was poorly tolerated (gastrointestinal and hematologic toxicity), resulting in a protocol amendment for reduced dose⁷⁷. Based on a search of ClinicalTrials.gov, no additional studies with this combination are planned in breast cancer.

Atezolizumab

The phase 2, single-arm GELATO study (NCT03147040) evaluated treatment with atezolizumab after immune induction with carboplatin in patients with metastatic lobular breast cancer⁷⁸. Two of 18 patients with ER + /HER2− breast cancer achieved a PR, with 1 patient having a durable response lasting >1 year⁷⁸.

Biomarkers

Exploratory analyses from several of the pembrolizumab combination studies showed no association between efficacy and baseline PD-L1 status or baseline TILs^{65,66,67,69,70,71}. In the only comparative study (eribulin alone vs eribulin plus pembrolizumab), the HR for PFS was 0.8 (95% CI, 0.3‒2) in patients with PD-L1−positive tumors and 1.1 (95% CI, 0.6‒2.1) in patients with PD-L1−negative tumors, with similar results observed in patients with TILs of 0 to 10 (HR, 0.9 [95% CI, 0.5‒1.6]) and >10 (HR, 1.0 [95% CI, 0.3‒3.5])⁶⁵. In the largest single-arm study of pembrolizumab plus eribulin, patients with PD-L1−positive tumors had an ORR of 38.1% (8/21) and those with PD-L1−negative tumors had an ORR of 40.9% (9/22) (P = 0.85)⁶⁶. In the same study, the clinical benefit rate was greater in patients with PD-L1−negative tumors than PD-L1−positive tumors (68.2% vs 42.9%), although the difference did not reach statistical significance (P = 0.095)⁶⁶. Similar results were reported in the NEWBEAT⁷³ and SAFIR02-BREAST IMMUNO⁷⁶ studies of nivolumab and durvalumab, respectively; however, these analyzes were done for the overall HER2− population.

Ongoing studies in metastatic HR + /HER2− breast cancer

There are 3 ongoing phase 3 studies, which are described below and in Table 4.

KEYNOTE-B49

KEYNOTE-B49 (NCT04895358) is a phase 3, randomized, double-blind study of pembrolizumab (vs placebo) plus chemotherapy in adult patients with PD-L1−positive (CPS ≥ 1), HR + /HER2 − , locally recurrent inoperable or metastatic breast cancer^79,80. In patients with breast cancer, increased PD-L1 expression has demonstrated significant correlations with aggressive clinicopathological features, such as grade 3 tumors and Ki67 > 14%, and a greater likelihood of responding to chemotherapy⁸¹. In high-risk node-positive breast cancer, increased PD-L1 expression has been shown to correlate with the presence of TILs⁸², an adverse prognostic factor for survival in HR + /HER2− breast cancer³³. Selection of PD-L1−positive tumors in KEYNOTE-B49 may therefore serve as a surrogate for those HR + /HER2− breast cancers most likely to benefit from the combination of pembrolizumab plus chemotherapy. Because PD-L1 expression is greater in primary versus metastatic tumors⁸³, biopsy samples in KEYNOTE-B49 are required from the locally recurrent inoperable or metastatic site. Patients are candidates for chemotherapy (no prior chemotherapy for metastatic disease) and have experienced disease progression on prior endocrine-based therapy. Patients must have received (a) ≥2 lines of endocrine therapy for metastatic disease with ≥1 line given in combination with a CDK4/6 inhibitor, (b) 1 line of endocrine therapy for metastatic disease if relapse occurred within 24 months of definitive surgery for primary tumor while on adjuvant endocrine therapy (prior treatment with a CDK4/6 inhibitor in the metastatic and/or adjuvant setting was required), or (c) 1 line of endocrine therapy for metastatic disease with relapse within 12 months (prior treatment with a CDK4/6 inhibitor in the metastatic setting was required). Patients without prior CDK4/6 inhibitor treatment must have had disease progression within 6 months of starting 1 line of endocrine therapy for metastatic disease and must have had a relapse within 24 months of definitive surgery for the primary tumor while on adjuvant endocrine therapy. Approximately 800 patients are randomized 1:1 to receive pembrolizumab or placebo for up to 35 cycles in combination with chemotherapy. Randomization is stratified by PD-L1 expression (CPS 1−9/ ≥ 10), visceral metastases (yes/no), and chemotherapy on study (taxanes/liposomal doxorubicin/capecitabine). The primary endpoints are PFS by blinded independent central review and OS. Endpoints will be evaluated separately in patients with PD-L1 CPS ≥ 10 and CPS ≥ 1, which may help identify patients who derive the greatest survival benefit from adding pembrolizumab to chemotherapy. The KEYNOTE-B49 study was initiated in 2021 and has an estimated completion date of July 2028^79,80.

TroFuse-010

TroFuse-010 is a phase 3, randomized, open-label study of the anti-TROP2 ADC sacituzumab tirumotecan (sac-TMT; also known as MK-2870/SKB264). The study is evaluating sac-TMT alone and in combination with pembrolizumab versus treatment of physician’s choice chemotherapy in patients with HR + /HER2− unresectable locally advanced or metastatic breast cancer⁸⁴. Patients are candidates for chemotherapy and have experienced disease progression on ≥1 line of endocrine therapy (1 in combination with a CDK4/6 inhibitor). The primary endpoint is PFS by blinded independent central review; OS is a key secondary endpoint. The study, which was initiated in 2024 and has an estimated completion date of April 2031, is currently recruiting patients⁸⁴.

AMBITION

AMBITION (NCT04732598) is a phase 3, randomized, open-label study of atezolizumab plus chemotherapy plus bevacizumab versus chemotherapy plus bevacizumab in patients ≥20 years of age with HR + /HER2 − , locally recurrent inoperable or metastatic breast cancer^85,86. The study is being conducted exclusively in Japan. Patients are candidates for chemotherapy (no prior chemotherapy for recurrent or advanced disease) or have life-threatening metastases and the disease is hormone refractory. Approximately 280 patients are randomized 1:1 to receive paclitaxel plus bevacizumab with or without atezolizumab. Randomization is stratified by disease status (recurrent/de novo), liver metastasis (yes/no), and PD-L1 status (positive/negative). The primary endpoint is investigator-assessed PFS; OS is a key secondary endpoint. The study was initiated in 2021 and has an estimated completion date of June 2025^85,86.

A strong rationale exists for investigating immunotherapy in patients with HR + /HER2− breast cancer, which is the most common breast cancer subtype. Although less immunogenic than other subtypes, TILs are present in a larger absolute number of patients with HR + /HER2− disease than the other breast cancer subtypes combined. Thus, the population of patients with HR + /HER2− breast cancer that could potentially derive benefit from immunotherapy is substantial. Additional evidence for examining immunotherapy in HR + /HER2− breast cancer stems from the demonstrated efficacy and safety of anti−PD-1 or anti−PD-L1 antibodies in patients with TNBC^{24,25,27,28,29}. One of these agents (pembrolizumab) is approved for use in combination with chemotherapy in both the early-stage and metastatic TNBC settings, and, in some countries, another (atezolizumab) is approved for use in combination with nab-paclitaxel in the metastatic TNBC setting.

Preliminary evidence has begun to accumulate on the use of anti−PD-1 or anti−PD-L1 antibodies as part of combination therapy in patients with HR + /HER2− disease. In the early-stage setting, pCR has generally been evaluated as a key efficacy endpoint because it may serve as a surrogate for predicting long-term clinical benefits⁵³. The value of pCR is stronger in TNBC and HER2+ breast cancer. Nevertheless, in a pooled analysis of neoadjuvant clinical studies, achievement of pCR was also positively associated with both EFS (HR, 0.49 [95% CI, 0.33–0.71]) and OS (HR, 0.43 [95% CI, 0.23–0.71]) in patients with HR + /HER2− disease. The association was more pronounced in patients with high-grade tumors (HR [95% CI] of 0.27 [0.14–0.50] for EFS and 0.29 [0.13−0.65] for OS) than low-grade tumors (HR [95% CI] of 0.63 [0.38–1.04] for EFS and 0.47 [0.21−1.07] for OS)⁵³. Incorporation of anti−PD-1 or anti−PD-L1 antibodies into the early-stage treatment setting for patients with HR + /HER2− breast cancer has generally led to favorable results, particularly for the subgroup of patients at high risk for recurrence based on clinicopathologic features and tumor genomic profile^41,42,45,47. However, results have been reported for only 2 phase 2 studies (I-SPY2, GIADA), which had limited sample sizes. The ongoing phase 3 KEYNOTE-756 and CheckMate 7FL studies are more definitively assessing the potential role of adding a PD-1 inhibitor to neoadjuvant chemotherapy in patients with high-risk, early-stage ER + /HER2− breast cancer. Initial promising results from KEYNOTE-756 and CheckMate 7FL were presented at the European Society for Medical Oncology Congress and San Antonio Breast Cancer Symposium in 2023. In those studies, statistically significant improvements in pCR were seen when anti−PD-1 therapy was added to neoadjuvant chemotherapy. Of note, while no association between clinical outcomes and PD-L1 expression was observed in patients with early-stage TNBC (KEYNOTE-522, Impassion031), preliminary results from KEYNOTE-756 and CheckMate 7FL suggest this might not be the case in ER + /HER2− breast cancer. Full results from both KEYNOTE-756 and CheckMate 7FL are expected to be published in 2024.

In the metastatic HR + /HER2− setting, there have been a number of phase 1/2 studies conducted in which patients were treated with an anti−PD-1 or anti−PD-L1 antibody alone⁶⁴ or as part of various combination regimens^{65,66,67,68,69,70,71,72,73,74,76,77}. The 2 studies with a control arm were negative and cited study limitations^65,76, although some patients achieved durable responses in the single-arm studies. The phase 3 KEYNOTE-B49 and AMBITION studies will further assess the potential role of adding an anti−PD-1 or anti−PD-L1 antibody to chemotherapy in patients with hormone-refractory, metastatic HR + /HER2− breast cancer. Although one might anticipate greater benefits from delivering immunotherapy earlier (neoadjuvant phase) as opposed to later in the metastatic setting (eg, more T-cell dysfunction or exhaustion), data from the phase 3 studies are needed to confirm whether this is the case. Additionally, while there was correlation between clinical outcomes and PD-L1 expression in patients with metastatic TNBC (KEYNOTE-355, KEYNOTE-119), preliminary results from phase 1/2 studies have not suggested an association in ER + /HER2− breast cancer; however, no phase 3 data are available to confirm these findings.

At present, it is uncertain whether anti−PD-1 or anti−PD-L1 antibodies provide differing benefit in patients with HR + /HER2− breast cancer. However, in the TNBC setting, the preponderance of positive data from phase 3 studies has come from those evaluating anti−PD-1 antibodies (in particular, pembrolizumab)^{24,25,27,28,87}. The immunoglobulin G (IgG) subtype of a monoclonal antibody may influence the pharmacokinetics of PD-1/PD-L1 antibodies and, thereby, their clinical efficacy and safety. Most antibodies targeting PD-1 or PD-L1 are of the IgG4 and IgG1 subtypes. IgG1 compared with IgG4 subtypes are associated with higher affinities for Fcγ receptors, which may increase their clearance via endocytic degradation by immune cells⁸⁸. However, similarly long circulating half-lives have been reported for anti−PD-1 and anti−PD-L1 antibodies irrespective of IgG isotype, which suggests that any such differences are not clinically relevant^88,89.

Other immunotherapy-based combination regimens may be useful in making the tumor microenvironment of patients with HR + /HER2− breast cancer more immunologically “hot.” Investigating the potential benefits of combining immunotherapy with endocrine therapy and CDK4/6 inhibitors is of interest based on preliminary findings⁶⁹, including the potential role of immunotherapy in delaying or reversing resistance to CDK4/6 inhibitors⁹⁰. Data in patients with early-stage ER + /HER− breast cancer^43,44 and metastatic HR + /HER2− breast cancer⁷⁰, however, suggest that toxicity with some CDK4/6 inhibitors may pose limitations. Another area of interest based on preliminary findings⁷⁴ is whether the addition of immunotherapy to ADCs might further improve outcomes for patients with HER2-low metastatic breast cancer. This is particularly interesting in view of the OS and PFS benefit demonstrated with trastuzumab deruxtecan monotherapy over chemotherapy⁷⁵ and OS and PFS benefit demonstrated with sacituzumab govitecan monotherapy over chemotherapy^91,92 in these patients. The ongoing phase 3 TroFuse-010 study is evaluating the ADC sac-TMT alone or with pembrolizumab compared with chemotherapy in patients with HR + /HER2− unresectable locally advanced or metastatic breast cancer⁸⁴. Preliminary results from the phase 1b/2 BEGONIA study showed that the combination of the ADC datopotamab deruxtecan plus durvalumab led to durable responses and had manageable safety in patients with previously untreated metastatic TNBC^93,94. The ongoing phase 3 TROPION-Breast04 study is further evaluating this combination in the first-line setting for patients with stage II or III TNBC as well as HR-low/HER2− breast cancer⁶¹.

In conclusion, research on the potential role of immunotherapy in the treatment of HR + /HER2− breast cancer is rapidly expanding. Results from several ongoing randomized controlled clinical trials are eagerly awaited and should help define the future role of immunotherapy in both the early and metastatic settings. Given that HR + /HER2− is a very heterogenous subtype of breast cancer, the role of biomarkers to help define which tumors will derive the greatest benefit from immunotherapy or from other types of therapies is fundamental and needs to be further researched and implemented to maximize the clinical benefit of these agents and avoid unnecessary exposure and toxicity. An additional important future consideration is the potential financial impact of any forthcoming approvals of novel therapies for HR + /HER2− breast cancer, which represents 65% to 70% of cases of the cancer with the highest incidence worldwide.