FormalPara Key Summary Points

Immune checkpoint inhibitors (ICIs) changed the treatment landscape of breast cancer.

However, ICIs are associated with specific side effects known as immune-mediated adverse events that are unique and potentially fatal.

Immune-mediated adverse events have a specific presentation, differ from other therapies’ side effects, and the treatment is based on immunosuppressive treatment and needs a multidisciplinary approach,

Oncologists must know how to identify these adverse events early and lead the multidisciplinary team that will provide patient care.

Introduction

The use of immune checkpoint inhibitors (ICIs) has revolutionized the way cancer is treated in the past 10 years. Recent estimates state that approximately 46% of all patients in the USA shall be eligible to receive this kind of therapy. [1] These medications are monoclonal antibodies targeting either the cytotoxic T lymphocyte antigen 4 (CTLA-4)–CD28 or the programmed cell death 1 (PD-1)/programmed cell death 1 ligand 1 (PD-L1) axes. Immunotherapy is currently the standard of care for the systemic treatment of several malignancies, from solid tumors to hematological diseases, and is used in a diversity of settings, from curative to palliative [2].

Despite their known therapeutic benefit, ICIs are associated with a unique profile of side effects caused by an off-target effect due to excessive immune system activation [3]. These toxicities, also known as immune-related adverse events (irAEs), are completely different from standard cytotoxic chemotherapy side effects with which clinicians are more familiar. The incidence of irAEs is variable, depending upon the agent used, the cancer type, and patient characteristics. Fatal events are quite rare, ranging from 0.3% to 1.3% of patients, with a median time to fatal irAEs onset of about 14.5 days [4].

Breast cancer is the most frequently diagnosed malignancy in female patients worldwide, accounting for 2.3 million new cases annually [5]. Even though only 5% of these patients are diagnosed with stage IV disease, a significant portion of them with early stage will recur; thus, metastatic breast cancer is the most frequent cause of cancer-related mortality worldwide, chiefly in Western countries, with approximately 600,000 deaths yearly [5, 6]. Though still incurable, metastatic breast cancer’s prognosis has been considerably improved in the past 10 years due to the introduction of new targeted agents, including ICIs [7].

This article aims to provide a brief overview of the role of immunotherapy, mainly ICI, in the treatment landscape of breast cancer as well as review the toxicities associated with this therapy and its management. This article is based on previously conducted studies and does not contain any new studies with human participants or animals performed by any of the authors. As such, we reviewed published literature on the topics that complied with good ethical standards and guaranteed patient safety.

Overview of the Clinical Evidence

Immunotherapy alone or in combination with chemotherapy has shown clinical value in several trials and has promising efficacy in different types of cancers, including triple-negative breast cancer, with improved progression-free survival (PFS) and overall survival (OS).

Among breast malignancies, triple-negative breast cancer (TNBC) is the subtype with the worst prognosis, with a median OS in the metastatic setting of fewer than 15 months when treated with chemotherapy alone. Additionally, approximately 25% of patients with early TNBC who receive neoadjuvant chemotherapy (NACT) will experience disease recurrence, with a 5-year disease-free survival rate (DFS) of 57% for patients who do not achieve pathological complete response (pCR), compared with 90% for those who do [8]. It was in the attempt to overcome this unmet medical need that immunotherapy studies were developed in these tumors.

The phase II GeparNUEVO trial conducted in Germany evaluated the addition of durvalumab to standard NACT in 174 patients with TNBC [9, 10]. Patients were randomized to durvalumab plus NACT with nab-paclitaxel, epirubicin, and cyclophosphamide or placebo plus the same chemotherapy backbone. Durvalumab was started as monotherapy 2 weeks before chemotherapy, but this “window phase” was prematurely closed on the basis of the deliberation of the study’s Independent Data Monitoring Committee (IDMC), which considered the waiting time to start chemotherapy to be too long. The study showed a numerical increase in the pCR rates (53.4% versus 44.2%) with the addition of durvalumab, but without statistical significance. In the subgroup of patients who received the induction with durvalumab before chemotherapy, better outcomes were observed, with a more prominent gain in pCR compared with the placebo (61% versus 41.4%, p = 0.052). A subsequent report with 42 months of follow-up showed that the combination improved the 3-year invasive DFS (85.6% versus 77.2%), reduced the risk of distant recurrence, and increased 3-year OS (95.2% versus 83.5%). Despite being a small study with no direct relevance to the use of checkpoint inhibitors in the adjuvant setting, these promising results paved the way for a more in-depth investigation of ICIs in this population.

Other initial studies also evaluated the safety and antitumor activity of immunotherapy in the metastatic setting. The KEYNOTE-012, presented at the American Society of Clinical Oncology (ASCO) 2016, was a phase Ib trial that used pembrolizumab as monotherapy in several metastatic solid tumors. It enrolled overall 32 patients with TNBC, all with positive PD-L1 status, and a median of two prior lines of therapy for metastatic disease, of which 27 cases were evaluable for efficacy. The overall response rate (ORR) was 18.5% and the disease control rate was 25.9% (of note, one patient with eight previous lines of therapy achieved a complete response). Median OS was 11.2 months, and the 1-year OS rate was 43.1%. These results suggested that patients with higher scores of PD-L1 had an increased probability of response (p = 0.28 for ORR) and reduced hazard (p = 0.12) for PFS [11]. The KEYNOTE-086 was a phase II trial that also evaluated pembrolizumab monotherapy in TNBC in two cohorts, with cohort A being patients on second or later lines expressing any level of PD-L1, and cohort B being patients in first-line therapy with PD-L1-positive tumors. Cohort A was composed of 170 patients, the majority with PD-L1-positive tumors (61.8%). ORR was 5.3%, with a median PFS of 2 months and OS of 9 months (PD-L1 status did not significantly change it). At the time of the analysis, the median duration of response was not yet reached, demonstrating the durable activity of pembrolizumab in previously treated patients [12]. On the other hand, cohort B enrolled 84 patients showing a higher ORR of 21.4%, with a median duration of response of 10.4 months, a median PFS of 2.1 months, and a median OS of 18 months [13].

These initial data culminated in the KEYNOTE-119, a phase III trial comparing pembrolizumab monotherapy versus chemotherapy (gemcitabine, eribulin, or capecitabine) in patients with metastatic TNBC who had received one or two prior lines of therapy and at least one anthracycline- and taxane-based treatment. The majority of patients (61%) had a combined positive score (CPS) > 1. Pembrolizumab did not improve OS in patients with CPS > 10 nor CPS > 1, with a median OS of 9.6 months for pembrolizumab versus 10.6 for chemotherapy. Although it was a negative trial, it showed benefits in exploratory analysis for female patients with CPS > 20. In this subgroup, there was an increase in median OS from 12.5 months to 14.9 months [hazard ratio (HR) 0.58; 95% CI 0.38–0.88] [14]. At ASCO 2020, it was reported that patients in this trial with tumor mutational burden (TMB) > 10 tended to derive better benefits from pembrolizumab [15].

These findings paved the way for prembrolizumab to be tested in the first line combined with chemotherapy in the KEYNOTE-355 study. The chemotherapy backbone in both arms was paclitaxel, nab-paclitaxel, or the combination of carboplatin and gemcitabine; physicians could choose freely among these schemes [16]. PD-L1 status was analyzed by IHC using Dako antibody 22C3 and stratified according to CPS. Patients with CPS ≥ 10 who were treated with pembrolizumab had a median PFS of 9.7 months compared with 5.6 months for the placebo group, ORR of 53.2% versus 39.8% for the placebo group, and OS of 23 months versus 16.1 months. Although there were also improvements in PFS among those with CPS ≥ 1, there was no statistically significant OS improvement in this subgroup. That led the FDA to approve pembrolizumab in combination with chemotherapy for the treatment of patients with locally recurrent unresectable or metastatic TNBC whose tumors express PD-L1 [Combined Positive Score (CPS) ≥ 10] as determined by an FDA-approved test.

Additionally, patients with metastatic breast cancer (mBC) can also have access to pembrolizumab based on other tumors characteristics such as microsatellite instability, mismatch repair deficiency, and tumor mutational burden, as this drug has been approved by the FDA based on the agency’s first tumor-agnostic approval after the results of the KEYNOTE-158 trial. [17, 18]. Although the frequency of MSI‐H is reportedly to be extremely low in breast cancer, ranging from 0% to 1.5%, these patients seem to markedly benefit from pembrolizumab. [19].

Atezolizumab is another anti-PDL1 antibody that turned out to be the first approved immunotherapy for the upfront treatment of mTNBC on the basis of the results of the IMpassion130 trial [20]. The study randomized 902 patients regardless of PD-L1 status to receive atezolizumab (or placebo) plus nab-paclitaxel in the first line and showed a statistically significant gain in terms of PFS in the intention-to-treat population (ITT) favoring atezolizumab (7.2 months versus 5.5 months, HR 0.80; 95% CI 0.69–0.92; p = 0.002). The study, however, failed to demonstrate benefit on its co-primary endpoint, overall survival (OS) in the ITT population (21.3 months versus 17.6 months; HR 0.84; 95% CI 0.69–1.02; p = 0.08). The subgroup analysis of the patients who were PD-L1 positive revealed an apparently meaningful benefit favoring atezolizumab of 25.0 months versus 15.5 months (HR 0.62; 95% CI 0.45–0.86; no p-value), but no formal statistical inference could be drawn due to the hierarchical method of analysis of the trial. On the basis of these results, the European Medicines Agency granted full approval for the drug in first-line PD-L1 positive mTNBC, while the FDA granted accelerated approval for atezolizumab conditional to a confirmatory study in March 2019. With a similar design but using paclitaxel as the backbone chemotherapy, the IMpassion131 trial failed to demonstrate the benefit of atezolizumab in terms of PFS or OS hierarchically tested in PD-L1-positive and then ITT populations (PFS 6 months versus 5.7 months and OS 22.1 months versus 28.3 months in patients who were PD-L1 positive) [21]. After these results, the FDA approval was withdrawn in September 2021.

On the curative setting, the KEYNOTE-522 was a phase III trial that randomized 1174 patients with previously untreated stage II or III TNBC to receive standard NACT concurrent with pembrolizumab or placebo followed by surgery, and adjuvant pembrolizumab or placebo for 27 weeks independent of pCR. The addition of pembrolizumab to NACT raised the overall pCR rate from 51.2% to 64.8% (p < 0.01), independently of PD-L1 and lymph node status. Additionally, the study also showed improvement in the 3-year event-free survival (85% versus 77% overall and 67.4% versus 56.8% in those patients who failed to achieve pCR) [22, 23]. The KEYNOTE-522 was the first phase III study that showed long-term benefits with the combination of chemotherapy and immunotherapy in the neoadjuvant plus immunotherapy in the adjuvant settings. In July 2022, the FDA approved the use of pembrolizumab in that indication for high-risk TNBC.

Furthermore, IMpassion031 is a phase III trial with a smaller sample that evaluated the addition of atezolizumab to platinum-free NACT and showed a significant increase in pCR rate (41% to 57.6%, p < 0.01), regardless of PD-L1 status. Results are still immature (and possibly underpowered) for EFS, DFS, and OS [24]. In contrast, the NeoTRIPaPDL1 trial randomized 280 patients with TNBC to receive NACT with nab-paclitaxel and carboplatin, without anthracycline or cyclophosphamide, associated with atezolizumab/placebo. The results showed a lack of improvement in pCR rates (48.6% versus 44.4% p = 0.48) overall or in either the PD-L1-positive or PD-L1-negative subsets. It is important to note, however, that caution must be taken when interpreting these results as pCR is a secondary endpoint of the study, while the primary endpoint of event-free survival (EFS) is still awaited [25]. With a very similar design compared with KEYNOTE-522, the GeparDouze/NSABP B-59 (ClinicalTrials.gov identifier: NCT02008227) trial has finished its recruitment and results are expected to provide the definitive answer concerning atezolizumab activity in the curative setting soon.

There are many new immunotherapeutic strategies being developed in the preclinical stage to overcome mechanisms of resistance to ICI. One of the most promising therapeutic areas is the manipulation of the tumor microenvironment (TME). Recent studies show that depending on the cell and cytokine configuration present in the TME, cancer cells are more or less responsive to immunotherapy, leading to the definitions of “hot” and “cold” tumors [26]. There are many targets directed toward tumor-associated immune and stromal compartments and of these, we highlight tumor growth factor (TGF)-beta, which has many immunosuppressive effects in the TME, by affecting the activities of tumor-infiltrating lymphocytes, macrophages, and regulatory T and dendritic cells and bispecific antibodies (targeting TGF-beta and PD-L1), such as YM101, BiTP, and M7824 [27,28,29].

Immunotherapy Adverse Events

Mechanisms of irAEs are not completely understood. Immune-related toxicities are generally related to the inflammatory reactions produced by immune system responses against specific organs and tissues. Translational research provides some evidence that irAEs may result from some combination of autoreactive T cells, autoantibodies, and/or proinflammatory cytokines (e.g., interleukin-17) [30, 31].

Currently, biomarkers capable of predicting which patients will have irAEs are not yet available in clinical practice. However, in retrospective observational studies, some clinical characteristics such as female sex, germline, somatic genetic features, specific microbiome compositions, and circulating biomarkers were associated with a higher risk of developing irAE [32,33,34]. Although rare, fulminant and even fatal toxicities may occur with immune checkpoint inhibitors [4]. Therefore, it is of paramount importance to promptly recognize and manage irAEs.

The current clinical experience with immunotherapies in various tumors shows that the most common irAEs include dermatologic, gastrointestinal, hepatic, endocrine, and pneumonitis, associated with other less common inflammatory events.

Dermatologic Immune-Related Adverse Events

Dermatological irAEs have been observed in up to 44% of patients treated with ICIs, however, less than 2% of these have been considered severe (grade 3 or 4) [35]. Cutaneous toxicity can be divided into inflammatory, neoplastic, severe reactions (Stevens–Johnson syndrome/toxic epidermal necrolysis), connective tissue disease, antibody mediated, hair, and other rarer ones that include exacerbation of preexisting diseases.

Inflammatory eruptions are the most common cutaneous reactions to ICIs. They include nonspecific, maculopapular eruption and psoriasiform, eczematous, and lichenoid dermatoses. Pruritus can occur in association with cutaneous xerosis or inflammatory skin reactions, sometimes disproportionate to visible dermatitis, or may present in the absence of a concurrent skin eruption. Pruritus without rash occurs in approximately 20% of patients treated with ICI after a median of three treatment cycles. It usually involves the trunk and lower extremities, where excoriations and prurigo nodularis-like lesions can be noted [36].

The time to onset of specific cutaneous irAE may provide a clue to the correct diagnosis: [37].

  1. ●

    0 to 3 weeks–eczematous and psoriasiform dermatitis

  2. ●

    3–6 weeks: maculopapular rash and pruritus

  3. ●

    6–12 weeks: lichenoid drug eruption

  4. ●

    7 to ≥ 16 weeks: vitiligo

  5. ●

    13–15 weeks: bullous pemphigoid

  6. ●

    13 to ≥ 16 weeks: alopecia areata

  7. ●

    Any time: Stevens–Johnson syndrome/toxic epidermal necrolysis (SJS/TEN), drug reaction with eosinophilia and systemic symptoms (DRESS), acute generalized exanthematous pustulosis (AGEP).

Gastrointestinal Immune-Related Adverse Events

Diarrhea and colitis are the most frequently reported gastrointestinal immune-related adverse events in patients undergoing treatment with checkpoint inhibitor immunotherapy. Immune checkpoint inhibitors diarrhea and colitis are defined by the ASCO on the basis of symptoms alone rather than by colonic inflammation, demonstrated by endoscopic findings (mucosal inflammation with or without ulcers). Diarrhea is watery bowel movements with any increase in stool frequency over baseline and colitis involves abdominal pain, mucus or blood in the stools, and fever. Diarrhea/colitis is the most common gastrointestinal irAE, usually starting around week 6 or 7 of treatment with recovery occurring around week 10 [38].

Hepatic Immune-Related Adverse Events

Hepatotoxicity related to immunotherapy is a relatively rare event (around 5% of all patients on ICIs). The main manifestation consists in the elevation of transaminases or bilirubin, with fever associated in a minority of patients [35]. In approximately 1% of patients, hepatotoxicity is severe. Elevations in liver enzymes typically occur 8–12 weeks after treatment initiation. However, liver enzyme elevations have been reported as early as 8 days after treatment is commenced or as late as several months after therapy is completed [39].

Endocrine Immune-Related Adverse Events

The most common endocrinopathies are hypothyroidism, hyperthyroidism, and hypophysitis. Inflammation of the pituitary, thyroid, or adrenal glands due to checkpoint blockade often presents nonspecific symptoms such as nausea, headache, fatigue, and vision changes. The incidence of endocrinopathies has been difficult to precisely state due to variable methods of assessment, diagnosis, and monitoring in different clinical trials. In a systematic review and meta-analysis, the overall incidence of clinically significant endocrinopathies is approximately 10% of patients treated with checkpoint inhibitors [40]. Different than other irAEs, most patients do not need either to receive immunosuppressors or stop therapy, as most of the cases can be successfully managed with hormonal replacement [2].

Pneumonitis Immune-Related Adverse Events

Inflammatory conditions affecting the lungs, such as sarcoidosis or organizing inflammatory pneumonia, have been observed with ICI in less than 10% of patients receiving these agents. Pneumonitis, defined as focal or diffuse inflammation of the lung parenchyma, is an uncommon but potentially severe or fatal complication. In most cases, it is asymptomatic and typically identified on routine computed tomography imaging, although possible presenting symptoms may include new or worsening cough, shortness of breath, increased oxygen requirement, chest pain, and/or fever. The median time to onset of pneumonitis is 34 weeks but can range from 1.5 weeks to 127 weeks and is usually managed with first-line corticosteroids followed by stronger immunosuppressors (such as infliximab and cyclophosphamide) in refractory cases [35].

Other Immune-Related Adverse Events

Other immune-related adverse events also described are renal, pancreatic, neurological, cardiological, hematological, ocular, rheumatological, and musculoskeletal [35].

Medications Used for Immunotherapy Toxicity Management

As discussed above, cancer immunotherapy can cause a remarkable class of side effects that constitutes the Achilles’ heel of these therapies. As irAE may occur in about 50% of all patients treated with ICI, oncologists must be capable of recognizing and providing appropriate care. The medications that are applied to treat irAEs and their mechanisms of action are summarized in Table 1. As these drugs are more commonly used in the field of rheumatology, most clinicians are not familiar with their routine usage in practice [41]. Therefore, it is essential to adopt a multidisciplinary approach to treat irAEs, especially in those cases which are refractory to steroids.

Table 1 Summary of treatments used for immune-related adverse events
Full size table

Corticosteroids are potent suppressors of T cell function and activity and represent the mainstay first-line treatment for irAEs. Although the use of prednisone as prophylaxis before starting ICIs to prevent irAEs is not recommended due to a potentially detrimental effect on the ICIs' expected efficacy, the retrospective analysis did not show any impact on response rate or survival when corticosteroids are used to treat an already established irAE [2, 42].

Even though the upfront treatment of irAEs is quite well established, there is no consensus on how to treat patients who need second-line immunosuppressive therapy, and most guidelines recommend a panoply of medications targeting multiple immune pathways. This led to the development of a more personalized approach based on immunopathological findings of each irAE to guide treatment decision [41].

T cells are believed to be a central piece in the immunopathogenesis of most irAEs, thus representing a good target for treatments. Mycophenolate is an inosine monophosphate dehydrogenase (IMPDH) inhibitor that depletes guanosine nucleotides, preferentially in T cells and B cells, and inhibits their proliferation, thereby suppressing cell-mediated immune responses and antibody production [43]. Other strategies to suppress T cell function are calcineurin inhibitors and methotrexate, even though their use for the treatment of irAEs is limited [41].

Targeting the integrin pathway with specific antibodies (vedolizumab and natalizumab) blocks the extravasation of T cells into peripheral tissue where they cause inflammation. This approach has been successfully used for steroid refractory colitis and encephalitis [44, 45]. Additionally, for the treatment of life-threatening irAEs, T cell depletion can be achieved with either anti-thymocyte globulin (ATG), alemtuzumab, or abatacept. ATG are polyclonal antibodies that induce T cell lysis via complement. Alemtuzumab is a monoclonal antibody targeting CD52, a membrane protein expressed by mature lymphocytes, monocytes, and natural killer cells. Abatacept is a CTLA-4 soluble fusion protein that induces T cell anergy. All of them have been successfully used for the treatment of life-threatening irAEs such as hepatitis and myocarditis [46,47,48,49,50]. These strategies should be used with caution in only extreme cases of irAEs given their potential to counterbalance the beneficial effect of ICIs for cancer treatment.

Rituximab is a monoclonal antibody that targets CD20, therefore targeting B cell function, and has been successfully used for treating ICI-induced hematological toxicities, myasthenia gravis, and Sjögren’s syndrome [51,52,53,54]. Plasmapheresis depletes pathogenic auto-antibodies by extracorporeal removal and was used in the setting of myasthenia gravis-myositis-myocarditis overlap syndrome, with limited success [49, 50]. Intravenous immunoglobulins (IVIG) induce the inactivation of autoreactive T cells, regulation of cytokine production, downregulation of B cell activation and antibody production, interference with complement activation, inhibition of macrophage and dendritic cells, and neutralization of pathogenic autoantibodies by anti-idiotypic antibodies [54]. Its use, however, can cause rash, acute kidney injury, and thromboembolic events, which might represent serious caveats of this strategy and can also be confused with new irAEs [41].

Commensal microbiota is another common pathway that integrates environmental inputs, such as the diet, with genetic and immune signals to affect the immunity of the host. Hematopoietic and non-hematopoietic cells of the innate immune system are often located strategically in the gut at the host–microbiota interface [41, 55]. Although still experimental, there are case reports of two patients with immune-related colitis refractory to corticosteroids, infliximab, and vedolizumab that were treated with fecal microbial transplantation from a single unrelated donor not diagnosed with cancer. Both patients had complete resolution of the clinical symptoms of colitis following treatment [56]. These preliminary results demonstrate that the gut microbiome can be a therapeutic opportunity that can be exploited in the management of irAEs.

Infliximab is a monoclonal antibody that targets the tumor necrosis factor (TNF), an endotoxin-inducible serum factor with the ability to cause necrosis of tumors via immune stimulation (it is considered the underlying factor explaining the effect of Coley’s toxins). Despite its name, TNF has no direct anti-cancer effect, and its inhibition has been shown to be effective in treating primary autoimmune disorders such as rheumatoid arthritis and inflammatory bowel disease. It has also been successfully used in the management of steroid-refractory colitis. Some reference centers are already using it as an upfront treatment instead of corticosteroids for its treatment and as prophylaxis together with anti-PD1 therapy in high-risk patients [41]. Infliximab can rarely cause cardiomyopathy and hepatitis. These conditions represent clinical challenges and clinicians must be aware of them as cardiac and hepatic toxicities can be irAEs themselves.

Tocilizumab is a monoclonal antibody anti-interleukin-6 (IL-6R) receptor and has been successfully used for the treatment of irAEs caused by nivolumab. IL6 is a cytokine with broad-ranging biological effects on both hematopoietic cells and cells of visceral organs. It has been used as a therapeutic target in autoimmune diseases such as rheumatoid arthritis, giant cell arteritis, and Castleman disease [57]. Usually well tolerated, tocilizumab carries a risk of serious opportunistic infections, as well as gastrointestinal perforation, neutropenia and thrombocytopenia, thus hampering its use for GI or hematological irAEs.

IL-17 is a key cytokine in host immunological defense mechanisms and is a mediator of diverse immune-mediated diseases. It is produced by cells of both the adaptive and the innate immune systems and acts as a link between immune and non-immune cellular activities [58]. Secukinumab is an anti-IL17 antibody used for psoriasis treatment that has been used for the treatment of refractory skin irAEs with success and no significant side effects [59].

Rapamycin is an mTOR inhibitor that reduces T cell proliferation by preventing the translation of stress signals into immune system cell differentiation. It was successfully used in patients with solid organ transplantation who required ICIs, in combination with anti-PD-1 therapy, and did indeed maintain allograft tolerance by abating the threshold for innate and adaptive immune response activation. Its use in other irAEs is still considered experimental [41, 60].

Conclusions

Breast cancer oncology has changed dramatically in recent years due to the introduction of checkpoint inhibitors. In both the curative and palliative settings, the emergence of these therapies has changed the natural history of TNBC but also uncovered a somewhat expected downside: immune-mediated adverse events. Due to their pathophysiology, irAEs have one of the most pleomorphic clinical presentations, representing a diagnostic challenge for clinicians. Oncologists should be aware of these toxicities to provide early diagnosis, and therefore be able to act in time, leading a multidisciplinary team that has to work together to take care of these patients. Organ-specific autoimmunity specialists (such as cardiologists, rheumatologists, gastroenterologists, etc.) and basic scientists must be part of these teams. Balancing the treatment benefit of ICI with the potential hampering effect of the immunosuppressors used for the management of the emerging irAEs, as well as being familiar with the potential side effects of these agents (e.g., secondary infections and lung/liver toxicity), is not trivial. Nonetheless, this knowledge should become part of the skill set of any cancer specialist treating patients with breast cancer.