Abstract
Introduction
Identifying risk factors and biomarkers of immune-related adverse events (irAE) in patients with early-stage breast cancer could help prevent and mitigate irAE and improve disease outcomes. This retrospective cohort study aims to identify predictive biomarkers of irAE in patients with early-stage breast cancer treated with ICI.
Methods
Patients with early-stage breast cancer treated with ICI between 2014 and 2024 from four institutions were included. Demographics, baseline laboratory values, and ICI treatment information were collected. Chi-square and Wilcoxon rank-sum tests were used to compare patients with and without any-grade irAE and grade ≥3 irAE. Uni- and multivariate logistic regression was performed to determine factors predictive of grade ≥3 irAE.
Results
Of the 423 patients included, 307 (72.6%) developed any-grade irAE and 81 (19.1%) developed grade ≥3 irAE. Hispanic ethnicity was associated with lower rates of any-grade and grade ≥3 irAE compared to non-Hispanic/unknown ethnicity (p < 0.0001 and p = 0.04, respectively). Patients with higher absolute eosinophil counts were more likely to have any-grade irAE (p = 0.0005). Grade ≥3 irAE were more common in non-smokers (p = 0.05) and those with chronic kidney disease (CKD; p = 0.001). Multivariate logistic regression demonstrated that older age (≥50y vs. <50y) [OR 1.70; 95% CI, 1.02-2.85; p = 0.04] and CKD (OR 4.97; 95% CI, 1.59-15.5; p = 0.006) were predictive of grade ≥3 irAE.
Conclusion
In patients with early-stage breast cancer treated with ICI, older age and the presence of CKD were risk factors of grade ≥3 irAE. Further studies to identify biomarkers of irAE are needed to prevent irAE and reduce the long-term complications.
Keywords: Immune-related adverse events, Immune checkpoint inhibitors, Breast cancer, Biomarker, Blood cell counts
Graphical abstract
Highlights
-
•
In our cohort, the incidence of any-grade irAEand grade 3+ irAE was 72.6% and 19.1%, respectively.
-
•
Higher baseline absolute eosinophil counts were predictive of any-grade irAE.
-
•
Older age and chronic kidney disease were associated with increased rates of grade 3 or higher irAE.
-
•
Further studies to identify biomarkers of irAE are needed in patients with early-stage breast cancer.
1. Introduction
The inclusion of immune checkpoint inhibitors (ICI) to the treatment of early-stage triple-negative breast cancer (TNBC) can result in increased toxicity, specifically immune-related adverse events (irAE) [1]. Initially approved in metastatic TNBC whose tumors express PD-L1 (combined positive score [CPS] ≥10), the indication for ICI was expanded to early-stage TNBC based on the results of the phase III KEYNOTE-522 trial which demonstrated that the addition of ICI to chemotherapy resulted in improved breast cancer outcomes in this patient population [2,3].
ICI produce side effects which are distinct from those observed with conventional chemotherapy alone and can result in permanent and life-threatening consequences. In KEYNOTE-522, more than one-third (35.0%) of patients experienced any-grade irAE, despite the exclusion of patients with significant comorbid disease and active autoimmune disease [3]. In addition, 13.0% of patients experienced grade ≥3 irAE, and 2 (0.3%) patients died due to irAE. Real-world studies of patients with early TNBC treated with ICI report even higher rates of any-grade irAE from 31% to 65% and grade ≥3 irAE from 11% to 30% [[4], [5], [6], [7], [8], [9], [10]]. Given the potentially expanding indication of ICI to other breast cancer subtypes, there is an urgent need to identify strategies to predict irAE risk.
Currently, there is no validated biomarker available to predict which patients will develop irAE or to identify those will benefit from ICI in early-stage breast cancer. Not all early-stage TNBC tumors respond to ICI, yet ICI is approved for all patients with early-stage TNBC. As a result, some patients may be subjected to overtreatment with ICI, resulting in potentially avoidable toxicity. The development of irAE can also preclude the ability to give the remaining chemotherapy course, which can result in worse breast cancer outcomes. Ultimately, oncologists need to find the balance between patients who benefit from the addition of ICI to chemotherapy and those for whom the risk of irAE may outweigh that benefit.
In other solid tumors, such as melanoma and non-small cell lung cancer, clinical factors such as age, sex, comorbidities, and cancer-related treatment variables have been shown to increase the risk of irAE [[11], [12], [13], [14], [15]]. Biomarker strategies including gene expression profiling, human leucocyte antigen (HLA) genotyping, micro-RNAs, and the gastrointestinal microbiome are being explored to predict irAE [16]. Clinically available biomarkers such as white blood cell (WBC) counts and blood cell ratios have also been identified as potential predictors of irAE, representing markers of inflammation [16]. For example, higher baseline absolute eosinophil counts (AEC) and absolute lymphocyte counts (ALC) and lower neutrophil-to-lymphocyte (NLR) ratios have been associated with higher incidence of irAE, while higher derived NLR (calculated as absolute neutrophil count [ANC]/[WBC–ANC]) may be protective of irAE [14,[17], [18], [19], [20], [21], [22], [23], [24], [25], [26], [27], [28], [29], [30]].
Despite the high rate of irAE in patients with early-stage breast cancer treated with ICI, no studies to date have established predictive biomarkers of irAE in this patient population. This multi-institutional, retrospective cohort study evaluates clinical factors and laboratory-based biomarkers to predict irAE in patients with early-stage breast cancer.
2. Methods
This was a retrospective cohort study of patients with stage I-III breast cancer treated with ICI across four NCI-designated comprehensive cancer centers. Patients were identified through pharmacy records and institutional databases. Inclusion criteria included any breast cancer subtype, the receipt of any type of ICI, and any combination of ICI therapy with other therapeutic agents. Patients treated on clinical trials were included. A documented start and stop date of ICI treatment was required. Medical records were reviewed through June 2024, at which point data abstraction was cut off. Follow-up data was collected including date of last encounter or date of death. Institutional review board approval was obtained by all four institutions, and the need for written consent from patients was waived per these protocols.
Data was collected via manual chart review and secured in the REDCap database. Demographic information and co-morbidities at the start of ICI were collected including age at start of ICI, gender, race, ethnicity, menopausal status, body mass index (BMI), smoking history, underlying autoimmune disease, type 2 diabetes mellitus, hypertension, hyperlipidemia, chronic kidney disease (CKD), coronary artery disease, and pre-existing lung disease. Baseline laboratory data prior to ICI was collected including the baseline complete blood count with differential, which included WBC, hemoglobin (HGB), platelet count (PLT), absolute neutrophil count (ANC), absolute eosinophil count (AEC), absolute lymphocyte count (ALC), and absolute monocyte count (AMC). Baseline albumin and thyroid-stimulating hormone (TSH) were also collected prior to ICI. Ratios were calculated for lymphocyte-to-monocyte ratio (LMR), neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), and derived neutrophil-to-lymphocyte ratio (dNLR) as defined by ANC/WBC-ANC.
Clinical tumor information including breast cancer stage and subtype and treatment information including type of ICI, number of ICI cycles, and combination agents were recorded. irAE was identified through manual chart review and was defined as any toxicity that was determined by the treating oncology provider to be possibly or definitely related to ICI treatment, or toxicity treated with high dose steroids where the causative agent was not clearly identified. Date of onset and grade of each irAE were recorded based on documentation in the medical record using the Common Terminology Criteria for Adverse Events (CTCAE) v5.
Univariate analysis was performed to determine clinical factors and treatment variables predictive of any-grade and grade ≥3 irAE. Given the high rate of any-grade irAE, univariate and multivariate logistic regression was performed only for grade ≥3 irAE. Age, BMI, race, menopausal status, smoking status, presence of type 2 diabetes mellitus, CKD, and pre-existing lung disease, and baseline laboratory values including AEC, LMR, NLR, MLR, PLR, and dNLR were included in the multivariate model. Treatment data including cycles of ICI were excluded from the multivariate logistic regression given the dependent nature of this variable on incidence of grade ≥3 irAE. Variables that were significant in the univariate results with a p-value of 0.3 or lower were entered into the multivariate model, which included age, BMI, race, menopausal status, smoking status, type 2 diabetes mellitus, CKD, and pre-existing lung disease. A significance threshold of 0.05 for p-values was used.
An exploratory analysis was performed to investigate the co-occurrence patterns of multi-organ irAE events. Co-occurring events were tabulated for all 423 patients, with all any-grade irAE included. Co-occurring irAE were categorized into factors across organ systems, including endocrine, cutaneous, respiratory, gastrointestinal, hepatic, musculoskeletal, neurological, renal, hematologic, cardiac, and other as demonstrated by Wan et al. [31] The irAE identified in our cohort were classified as the following endocrine (thyroiditis, adrenal insufficiency/hypophysitis, diabetes), cutaneous (rash, vitiligo, mucositis, angioedema), respiratory (pneumonitis), gastrointestinal (colitis, duodenitis/gastritis, pancreatitis, xerostomia, sialadenitis), hepatic (hepatitis), musculoskeletal (arthralgia, myalgia, myositis), and neurologic (neuropathy, encephalitis/encephalopathy, Bell's palsy), renal (nephritis), hematologic (thrombocytopenia, cytopenia, eosinophilia, neutropenia), cardiac (QT prolongation), and other (fatigue, fever, insomnia). Patients with multiple irAE within the same organ system (e.g., endocrine) were counted once as each unique event within the same organ system was not counted.
3. Results
3.1. Clinical and treatment characteristics
A total of 423 patients with early-stage breast cancer treated with ICI were included in the study. Their characteristics and ICI treatment data are summarized in Table 1 by degree of toxicity. All but 1 patient was female. The median age at diagnosis was 51 years (interquartile range [IQR], 41.0-61.0). Most (86.3%) patients had TNBC and were treated with pembrolizumab (90.8%).
Table 1.
Patient demographics and clinical characteristics according to onset of any-grade and grade ≥3 irAE.
| All patients (n = 423) | Any-grade irAE |
Grade ≥3 irAE |
|||||
|---|---|---|---|---|---|---|---|
| No (n = 116) | Yes (n = 307) | P-value | No (n = 342) | Yes (n = 81) | P-value | ||
| Age at diagnosis, years (IQR) | 51 (41 - 61) | 50.5 (43 - 60) | 51 (41 - 62) | 0.8 | 50 (41 - 61) | 53 (46 - 64) | 0.07 |
| Gender, female (%) | 422 (99.8%) | 115 (99.1%) | 307 (100.0%) | 0.3 | 341 (99.7%) | 81 (100.0%) | 0.6 |
| Menopausal status at diagnosis | 0.6 | 0.2 | |||||
| Pre- or peri-menopausal | 216 (51.0%) | 61 (52.6%) | 155 (50.5%) | 181 (52.9%) | 35 (43.2%) | ||
| Post-menopausal | 200 (47.3%) | 52 (44.8%) | 148 (48.2%) | 155 (45.3%) | 45 (55.6%) | ||
| Unknown/not applicable | 7 (1.7%) | 3 (2.6%) | 4 (1.3%) | 6 (1.8%) | 1 (1.2%) | ||
| BMI, median (kg/m2) (IQR) | 27.1 (23.6 - 31.3) | 27.3 (23.7 - 31.5) | 27.0 (23.4 - 31.3) | 0.7 | 27.4 (23.7 - 31.5) | 26.1 (23.1 - 30.6) | 0.13 |
| Race | 0.6 | 0.3 | |||||
| White | 299 (70.7%) | 79 (68.1%) | 220 (71.7%) | 239 (69.9%) | 60 (74.1%) | ||
| Asian/Pacific Islander | 42 (9.9%) | 12 (10.3%) | 30 (9.8%) | 31 (9.1%) | 11 (13.6%) | ||
| Black | 36 (8.5%) | 10 (8.6%) | 26 (8.5%) | 33 (9.6%) | 3 (3.7%) | ||
| American Indian/Alaska Native | 3 (0.7%) | 2 (1.7%) | 1 (0.3%) | 2 (0.6%) | 1 (1.2%) | ||
| Other/unknown | 43 (10.2%) | 13 (11.2%) | 30 (9.8%) | 37 (10.8%) | 6 (7.4%) | ||
| Ethnicity | <0.0001 | 0.04 | |||||
| Hispanic | 55 (13.0%) | 29 (25.0%) | 26 (8.5%) | 50 (14.6%) | 5 (6.2%) | ||
| Non-Hispanic or Unknown | 368 (87.0%) | 87 (75.0%) | 281 (81.5%) | 292 (85.4%) | 76 (93.8%) | ||
| Smoking status | 0.6 | 0.05 | |||||
| Current smoker | 16 (3.8%) | 4 (3.4%) | 12 (3.9%) | 13 (3.8%) | 3 (3.7%) | ||
| Former smoker | 133 (31.4%) | 32 (27.6%) | 101 (32.9%) | 115 (33.6%) | 18 (22.2%) | ||
| Never smoker | 273 (64.5%) | 80 (69.0%) | 193 (62.9%) | 214 (62.6%) | 59 (72.8%) | ||
| Unknown | 1 (0.2%) | 0 (0.0%) | 1 (0.3%) | 0 (0.0%) | 1 (1.2%) | ||
| Comorbidities | |||||||
| Autoimmune disease | 35 (8.3%) | 5 (4.3%) | 30 (9.8%) | 0.07 | 27 (7.9%) | 8 (9.9%) | 0.6 |
| Diabetes mellitus | 47 (11.1%) | 12 (10.3%) | 35 (11.4%) | 0.8 | 35 (10.2%) | 12 (14.8%) | 0.2 |
| Hypertension | 119 (28.1%) | 34 (29.3%) | 85 (27.7%) | 0.7 | 97 (28.4%) | 22 (27.2%) | 0.8 |
| Hyperlipidemia | 66 (15.6%) | 15 (12.9%) | 51 (16.6%) | 0.4 | 56 (16.4%) | 10 (12.3%) | 0.4 |
| Chronic kidney disease | 13 (3.1%) | 1 (0.9%) | 12 (3.9%) | 0.1 | 6 (1.8%) | 7 (8.6%) | 0.001 |
| Pre-existing lung disease | 137 (32.4%) | 35 (30.2%) | 102 (33.2%) | 0.6 | 105 (30.7%) | 32 (39.5%) | 0.1 |
| Coronary artery disease | 14 (3.3%) | 2 (1.7%) | 12 (3.9%) | 0.3 | 9 (2.6%) | 5 (6.2%) | 0.1 |
| Clinical stage of breast cancer | 0.1 | 0.2 | |||||
| I | 42 (9.9%) | 9 (7.8%) | 33 (10.7%) | 36 (10.5%) | 6 (7.4%) | ||
| II | 233 (55.1%) | 58 (50.0%) | 175 (57.0%) | 182 (53.2%) | 51 (63.0%) | ||
| III | 146 (34.5%) | 49 (42.2%) | 97 (31.6%) | 123 (36.0%) | 23 (28.4%) | ||
| Unknown | 2 (0.5%) | 0 (0.0%) | 2 (0.7%) | 1 (0.3%) | 1 (1.2%) | ||
| Breast cancer subtype | |||||||
| TNBC | 365 (86.3%) | 110 (94.8%) | 255 (83.1%) | 0.002 | 293 (85.7%) | 72 (88.9%) | 0.4 |
| HR+/HER2- | 52 (12.3%) | 6 (5.2%) | 46 (15.0%) | 0.006 | 44 (12.9%) | 8 (9.9%) | 0.5 |
| HER2+ | 7 (1.7%) | 0 (0.0%) | 7 (2.3%) | 0.1 | 6 (1.8%) | 1 (1.2%) | 0.7 |
| Clinical trial | 0.07 | 0.12 | |||||
| Yes | 118 (27.9%) | 25 (21.6%) | 93 (30.3%) | 101 (29.5%) | 17 (21.0%) | ||
| No | 305 (72.1%) | 91 (78.4%) | 214 (69.7%) | 241 (70.3%) | 64 (80.0%) | ||
| ICI name | 0.2 | 0.6 | |||||
| Pembrolizumab | 384 (90.8%) | 0 (0.0%) | 12 (3.9%) | 310 (90.6%) | 74 (91.4%) | ||
| Dostarlimab | 12 (2.8%) | 3 (2.6%) | 9 (2.9%) | 9 (2.6%) | 3 (3.7%) | ||
| Durvalumab | 12 (2.8%) | 109 (94.0%) | 275 (89.6%) | 9 (2.6%) | 3 (3.7%) | ||
| Other/Unknown | 15 (3.5%) | 4 (3.4%) | 11 (3.6%) | 14 (4.1%) | 1 (1.2%) | ||
| Median number of ICI cycles (IQR) | 8 (4 - 16) | 11 (4 - 17) | 7 (4 - 15) | 0.006 | 9 (4 - 17) | 4 (3 - 9) | <0.0001 |
| Median duration of ICI, months (IQR) | 5.5 (2.1, 12.0) | 10.6 (2.7, 12.5) | 4.9 (2.1, 11.6) | 0.02 | 7.0 (2.3 - 12.4) | 3.0 (1.7 - 7.3) | <0.0001 |
| irAE-free months from start of treatment, median (95% CI) | 4.5 (3.7, 5.2) | NR (NR, NR) | 3.0 (2.3,3.4) | <0.0001 | 4.9 (4.0, 6.9) | 3.7 (2.5, 4.7) | <0.0001 |
| Median follow-up months (95% CI) | 22.6 (20.5, 24.5) | 19.7 (18.0, 24.8) | 23.9 (21.1, 25.3) | 0.02 | 23.9 (20.8, 25.2) | 19.4 (17.4, 24.2) | 0.5 |
| Median OS months (95% CI) | NR (81.8, NR) | NR (NR, NR) | NR (81.8, NR) | 0.049 | NR (NR, NR) | 81.8 (43.4, NR) | 0.07 |
Abbreviations: HR+/HER2-, hormone receptor-positive HER2-negative; HER2+, HER2-positive; TNBC, triple-negative breast cancer; ICI, immune checkpoint inhibitor; IQR, interquartile range; irAE, immune-related adverse event.
3.2. Incidence of irAE
After a median follow-up of 22.6 months (95% confidence interval [CI], 20.5-24.5), 307 (72.6%) patients developed any-grade irAE and 81 (19.1%) of patients developed grade ≥3 irAE. No deaths (grade 5) due to irAE occurred. Among the 423 patients, 180 (42.6%) developed one any-grade irAE, 97 (22.9%) developed two irAE, and 30 (7.1%) developed 3 or more unique irAE. 81 patients developed at least 1 grade ≥ 3 irAE, with 74 (17.0%) who developed one grade ≥3 irAE and 7 (1.7%) who developed two unique grade ≥3 irAE. Fig. 1 illustrates the number of any-grade and grade ≥3 irAE per patient.
Fig. 1.
Distribution of number of all-grade and grade ≥3 irAE per patient.
Abbreviations: irAE, immune-related adverse event.
Fig. 2 depicts a bar plot of the number of irAE by subtype and grade. Among the 471 unique any-grade irAE, the most common were thyroiditis (24.6%), rash (22.7%), colitis (14.9%), hepatitis (10.4%), and adrenal insufficiency/hypophysitis (7.9%). Among the 88 grade ≥ 3 irAE, the most common were adrenal insufficiency/hypophysitis (21.6%), rash (19.3%), hepatitis (12.5%), colitis (11.4%), and autoimmune diabetes mellitus (8.0%). The median time to onset of irAE by irAE type and organ system are shown in Supplementary Tables 2 and 3.
Fig. 2.
Bar plot of the number of irAE by subtype and grade.
Abbreviations: irAE, immune-related adverse event.
3.3. Predictive clinical factors and treatment variables associated with any-grade and grade ≥3 irAE
3.3.1. Any-grade irAE
Patients of Hispanic ethnicity had lower rates of any-grade irAE compared to non-Hispanic or unknown ethnicity (47.3% vs. 76.4%; p < 0.0001; Table 1). Though not statistically significant, patients with underlying autoimmune disease had numerically higher rates of any-grade irAE compared to those without underlying autoimmune disease (85.7% vs. 71.4%; p = 0.07). Age at diagnosis, gender, menopausal status, race, BMI, smoking status, the presence of type 2 diabetes mellitus, hypertension, hyperlipidemia, pre-existing lung disease, CKD, and coronary artery disease were not different according to onset of any-grade irAE.
Patients with any-grade irAE received fewer cycles of ICI compared to patients without any-grade irAE, with a median number of ICI cycles of 7 (IQR, 4-15) and 11 (IQR, 4-17), respectively (p = 0.006). Similarly, patients with any-grade irAE received ICI for a shorter duration of 4.9 months (IQR, 2.9-11.6) compared to 10.6 months (IQR, 2.7-12.5) for those without any-grade irAE (p = 0.02). Patients on clinical trial tended to have a higher incidence of any-grade irAE (78.8%) compared to those receiving standard of care therapy (70.2%), although not significant (p = 0.07).
Among breast cancer subtypes, any-grade irAE occurred in 69.9% (255/365) of TNBC, 88.2% (45/51) of HR+/HER2-, and 100% (7/7) of HER2+ (p = 0.006). Most (95.9%) patients with TNBC received pembrolizumab, whereas nearly half of patients with HR+ and HER2+ breast cancer received a different ICI. Median cycles of ICI were higher among patients with TNBC vs. the other two subtypes (9 vs. 4 vs. 4, respectively), as was the duration (7.7 vs. 2.1 vs. 3.0 months, respectively), likely reflecting the longer duration of ICI administration in the KEYNOTE-522 regimen compared to other clinical trial regimens.
3.3.2. Grade ≥3 irAE
Former smokers had a lower rate of grade ≥3 irAE (13.5%) compared to those who never smoked (21.6%) and current smokers (18.8%; p = 0.05; Table 1), although patients with a history of smoking (former or current smokers) had a non-significant trend toward increased risk of respiratory irAE (e.g., pneumonitis) (odds ratio [OR] 3.32, 95% CI, 0.96-11.6, p = 0.059).
Although the number of patients with CKD was low (n = 13), patients with CKD had higher rates of grade ≥3 irAE compared to those without CKD (53.8% [7/13] vs. 18.0% [74/410]; p = 0.001). Of the 7 patients with CKD and grade ≥3 irAE, 3 developed renal irAE (e.g., nephritis). Logistic regression demonstrated that patients with CKD had a higher rate of renal irAE compared to those without CKD (OR 62.0; 95% CI, 14-275; p < 0.0001).
Patients of Hispanic ethnicity had lower rates of grade ≥3 irAE compared to non-Hispanic or unknown ethnicity (0.1% vs. 20.7%; p = 0.04). Age, gender, menopausal status, BMI, race, and the presence of underlying autoimmune disease, type 2 diabetes mellitus, hypertension, hyperlipidemia, pre-existing lung disease, and coronary artery disease were not associated with onset of grade ≥3 irAE.
Patients with grade ≥3 irAE received fewer ICI cycles compared to patients without grade ≥3 irAE, with a median number of ICI cycles of 4 (IQR 3-9) and 9 (4-17), respectively (p < 0.0001). Similarly, patients with grade ≥3 irAE received ICI for a shorter median duration of 3.0 months (IQR, 1.7-7.3) compared to 7.0 months (IQR, 2.3-12.4) for those without grade ≥3 irAE (p < 0.0001). There was no difference in grade ≥3 irAE according to patients on clinical trial versus those receiving standard of care.
3.3.3. Subgroup analysis of patients who received pembrolizumab-only
A focused analysis of patients who received pembrolizumab-only was performed to identify any differences in irAE from PD-1 versus PD-L1 inhibitors. With the omission of non-pembrolizumab patients, demographics and rate of irAE were comparable to the entire cohort. The rate of any-grade irAE in the pembrolizumab-only cohort was 275/384 (71.6%) compared to 307/423 (72.6%) in the entire cohort (p = 0.09). Similarly, the rate of grade ≥3 irAE in pembrolizumab-only versus the entire cohort were similar at 19.3% and 19.1%, respectively (chi-square p = 0.9). Predictive clinical factors and treatment variables of patients who received pembrolizumab-only are shown in Supplementary Table 1.
3.3.4. Predictive peripheral blood biomarkers associated with any-grade and grade ≥3 irAE
Baseline laboratory values prior to the start of ICI treatment according to onset of any-grade and grade ≥3 irAE are shown in Table 2. Baseline median AEC values were higher for those with any-grade irAE compared to those without any-grade irAE (p = 0.0005), with low AEC (as defined by < 0.03 K/μL) more commonly seen in patients without irAE versus with irAE (26.1% vs. 12.9%; p = 0.0012). Baseline albumin, TSH, WBC, HGB, PLT, ANC, AEC, ALC, AMC, LMR, NLR, MLR, PLR, and dNLR were similar for patients with and without any-grade irAE. There were no significant differences in baseline peripheral blood markers for patients with and without grade ≥3 irAE.
Table 2.
Baseline laboratory values according to onset of any-grade and grade ≥3 irAE.
| All patients (n = 423) | Any-grade irAE |
Grade ≥3 irAE |
|||||
|---|---|---|---|---|---|---|---|
| No (n = 116) | Yes (n = 307) | P-value | No (n = 342) | Yes (n = 81) | P-value | ||
| Albumin (g/dL), median (IQR) | 4.1 (3.9 - 4.4) | 4.2 (3.9 - 4.4) | 4.1 (3.9 - 4.4) | 0.4 | 4.1 (3.9 - 4.4) | 4.1 (3.8 - 4.4) | 0.2 |
| TSH (mIU/L), median (IQR) | 1.5 (1.1 - 2.2) | 1.4 (1.1 - 2.0) | 1.5 (1.1 - 2.3) | 0.4 | 1.5 (1.1 - 2.1) | 1.5 (1.1 - 2.4) | 0.5 |
| WBC (K/μL), median (IQR) | 6.6 (5.3 - 7.9) | 6.6 (5.2 - 7.7) | 6.6 (5.4 - 7.9) | 0.5 | 6.6 (5.2 - 7.9) | 6.7 (5.6 - 7.9) | 0.6 |
| HGB (g/dL), median (IQR) | 13.2 (12.5-13.9) | 13.2 (12.3 - 13.8) | 13.2 (12.5 - 13.9) | 0.2 | 13.3 (12.5 - 13.8) | 13.2 (12.5 - 13.9) | 0.9 |
| PLT (K/μL), median (IQR) | 254 (219 - 302) | 255 (216 - 298) | 254 (219 - 302) | 0.8 | 250 (216 -301) | 270 (238 -306) | 0.13 |
| ANC (K/μL), median (IQR) | 4.0 (3.0 - 5.1) | 4.0 (3.0 - 5.0) | 4.0 (3.1 - 5.1) | 0.8 | 4.0 (3.1 - 5.0) | 3.9 (3.0 - 5.1) | 0.9 |
| AEC (K/μL), median (IQR) | 0.10 (0.06 - 0.20) | 0.10 (0.02 - 0.16) | 0.10 (0.07 - 0.20) | 0.0005 | 0.10 (0.05 - 0.20) | 0.10 (0.10 - 0.20) | 0.12 |
| ALC (K/μL), median (IQR) | 1.7 (1.3 - 2.2) | 1.7 (1.2 - 2.2) | 1.7 (1.4 - 2.3) | 0.2 | 1.7 (1.3 - 2.2) | 1.8 (1.4 - 2.2) | 0.6 |
| AMC, median (IQR) | 0.5 (0.4 - 0.6) | 0.5 (0.3 - 0.6) | 0.5 (0.4 - 0.6) | 0.1 | 0.5 (0.4 - 0.6) | 0.5 (0.4 - 0.6) | 0.7 |
| LMR (K/μL), median (IQR) | 3.7 (2.8 – 4.9) | 3.8 (2.8 – 4.9) | 3.6 (2.8 – 4.7) | 0.3 | 3.7 (2.8 – 4.8) | 3.7 (2.9 – 5.0) | 0.5 |
| NLR, median (IQR) | 2.2 (1.6 – 3.1) | 2.3 (1.7 – 3.4) | 2.2 (1.6 – 3.1) | 0.5 | 2.2 (1.6 – 3.2) | 2.2 (1.6 – 3.0) | 1.0 |
| MLR, median (IQR) | 0.3 (0.2 – 0.4) | 0.3 (0.2 – 0.4) | 0.3 (0.2 – 0.3) | 0.3 | 0.3 (0.2 – 0.4) | 0.3 (0.2 – 0.3) | 0.5 |
| PLR, median (IQR) | 150 (118 – 191) | 153 (117 – 191) | 148 (117 – 186) | 0.2 | 150 (117 – 187) | 147 (122 – 197) | 0.6 |
| dNLR, median (IQR) | 1.6 (1.2 – 2.2) | 1.6 (1.2 – 2.3) | 1.6 (1.2 – 2.2) | 0.3 | 1.6 (1.2 – 2.2) | 1.5 (1.1 – 2.1) | 0.4 |
Abbreviations: TSH, thyroid-stimulating hormone; WBC, white blood cell; HGB, hemoglobin; PLT, platelet count; ANC, absolute neutrophil count; AEC, absolute eosinophil count; ALC, absolute lymphocyte count; AMC, absolute monocyte count, LMR, lymphocyte-to-monocyte ratio; NLR, neutrophil-to-lymphocyte ratio; PLR, platelet-to-lymphocyte ratio; dNLR, derived neutrophil-to-lymphocyte ratio.
3.3.5. Univariate and multivariate logistic regression to identify predictors for grade ≥3 irAE
Univariate and multivariate logistic regression was performed to identify clinical factors and peripheral blood biomarkers predictive of grade ≥3 irAE (Fig. 3). On univariate logistic analysis, age (≥50 years vs. <50 years) [OR 1.65; 95% CI, 1.0001-3.85; p = 0.05], Hispanic versus non-Hispanic ethnicity [OR 0.38; 95% CI, 0.15-1.0; p = 0.05], and the presence of CKD (OR 5.30; 95% CI, 1.73-16.2; p = 0.003) were significant. Multivariate logistic regression demonstrated that older age (≥50y vs. <50y) [OR 1.70; 95% CI, 1.02-2.85; p = 0.04] and the presence of CKD (OR 4.97; 95% CI, 1.59-15.5; p = 0.006) were predictive of grade ≥3 irAE. Although not significant in the multivariate model, we explicitly controlled for the confounding effect of BMI on age [OR: 0.96 (95% CI, 0.91-1.00); p = 0.07]. Other clinical factors and peripheral blood biomarkers were not associated with onset of grade ≥3 irAE in the multivariate model.
Fig. 3.
Univariate and multivariate logistic analysis to identify predictors for grade ≥3 irAE.
Abbreviations: BMI, body mass index; AEC, absolute eosinophil count; LMR, lymphocyte-to-monocyte ratio; NLR, neutrophil-to-lymphocyte ratio; PLR, platelet-to-lymphocyte ratio; dNLR, derived neutrophil-to-lymphocyte ratio; irAE, immune-related adverse event.
3.4. Co-occurrence patterns of multi-organ irAE events
Of the total 471 irAE events identified, 161 (34.2%) irAE were within the endocrine system, 111 (23.6%) were cutaneous, 74 (15.7%) were gastrointestinal, 49 (10.4%) were hepatic, 31 (6.6%) were musculoskeletal, 13 (2.8%) were other, 11 (2.3%) were respiratory, 9 (1.9%) were renal, 6 (1.3%) were neurologic, 5 (1.1%) were hematologic, and 1 (0.2%) was cardiac. The most common co-occurrence of endocrine irAE included cutaneous (n = 36), gastrointestinal (n = 19), hepatic (n = 18) and musculoskeletal (n = 11) irAE. Gastrointestinal and cutaneous irAE were also found to commonly co-occur with hepatic irAE (n = 15 and 13, respectively). Fig. 4 demonstrates the co-occurrence patterns of irAE by organ system per patient.
Fig. 4.
Pairwise co-occurrence patterns of irAE by organ system by number of patients.
Abbreviations: GI, gastrointestinal; MSK, musculoskeletal.
3.5. Management of irAE
Of the 471 unique irAE events, 158 (33.5%) were treated with systemic steroids. Oral steroids were administered in 7/149 (4.7%) grade 1 irAE events, 65/234 (27.8%) grade 2 irAE events, 53/83 (63.9%) grade 3 irAE events, and 2/5 (40%) grade 4 irAE events. Intravenous (IV) steroids were administered in 0/149 (0%) grade 1 irAE events, 8/234 (3.4%) grade 2 irAE events, 20/83 (24.1%) grade 3 irAE events, and 3/5 (60%) grade 4 irAE events. Topical steroids were given almost exclusively to treat rash and were administered to 43/51 (84.3%) grade 1 rash events, 26/39 (66.7%) grade 2 events, and 7/17 (41.2%) grade 3 rashes. No grade 4 rashes were noted. The use of steroids by irAE type and grade is shown in Supplementary Table 4.
4. Discussion
In this multi-institutional, retrospective cohort study of patients with early-stage breast cancer treated with ICI – which consisted almost entirely of women, including 35 with pre-existing autoimmune disease and 13 with CKD, neither of which would have been eligible for inclusion in the KEYNOTE-522 trial – the incidence of any-grade and grade ≥3 irAE was 72.6% and 19.1%, respectively. Our study found that older age and CKD were significant predictors of grade ≥3 irAE. Peripheral blood markers were not predictive of grade ≥3 events, although elevated AEC was associated with a higher rate of any-grade irAE.
Identifying risk factors and biomarkers of irAE is urgently needed given the potentially expanding indication of ICI in early-stage breast cancer. While ICI is currently approved in early-stage TNBC, many studies are exploring the benefit of ICI in other breast cancer subtypes [[32], [33], [34], [35], [36], [37]]. Additionally, the benefit of adjuvant pembrolizumab is also being investigated, with the OptimICE-pCR (NCT05812807) and Opt-PEMBRO (NCT06606730) trials currently underway to determine the effect of adjuvant pembrolizumab in patients who achieve pCR following neoadjuvant chemo-immunotherapy. During a time in which the benefit of ICI is being widely explored in early-stage breast cancer, providers must carefully balance ICI benefit against the risk of irAE, which can significantly impact quality of life and, in rare cases, be life-threatening.
The incidence of any-grade and grade ≥3 irAE in our study was higher than that reported in the KEYNOTE-522 trial, which demonstrated rates of any-grade and grade ≥3 irAE of 35.0% and 13.0%, respectively [3]. The higher rate of irAE in our cohort may be attributed to the inclusion of an unselected patient population who were older and had multiple co-morbidities. In addition, the median follow-up time of our cohort was significantly longer (22.6 months) compared to the follow-up time of the KEYNOTE-522 trial (30 days for any-grade and 90 days for serious adverse events after discontinuation of treatment). Due to our study being retrospective in nature, there may also have been issues related to attribution of adverse events to ICI, which may have led to overestimation of non-specific irAE (e.g., rash), while low-grade irAE may have been underreported. Nevertheless, other real-world studies of early-stage breast cancer treated with ICI report higher rates of any-grade irAE between 31% and 65% and grade ≥3 irAE between 11% and 30% [[4], [5], [6], [7], [8], [9], [10]].
We identified that patients aged ≥ 50 years were at increased risk of grade ≥3 irAE. The impact of age on irAE is unclear, with conflicting data in other solid tumors [11]. Some studies suggest that older patients are more prone to develop specific types of irAE such as myocarditis, pneumonitis, arthritis, and dermatitis, while younger patients are more susceptible to develop colitis and hepatitis [38,39]. These differences in irAE onset may be due to the greater presence of comorbidities and underlying organ dysfunction in older patients increasing the risk of certain types of irAE, while other irAE types in younger patients may occur due to a heightened immune response.
In addition, seven of the thirteen patients with CKD developed grade ≥3 irAE, of which three developed renal irAE (eg, nephritis). This suggests that patients with CKD may be at higher risk of severe irAE, particularly renal irAE, although these results must be interpreted with caution given the small sample size. Prior studies have demonstrated that patients with pre-existing CKD and lower baseline estimated glomerular filtration rate (eGFR) rates are at increased risk of ICI-associated acute kidney injury [[40], [41], [42]], although other studies refute this association suggesting that the decrease in serum creatinine may be due to lower renal reserve rather than a heightened immunogenicity [41,[43], [44], [45]]. Due to the retrospective nature of this study, confirmation of renal irAE, such as with a kidney biopsy, could not be performed.
Our study also revealed lower rates of any-grade and grade ≥3 irAE in patients of Hispanic ethnicity compared to non-Hispanic/unknown ethnicity. Several prior studies have similarly reported that Hispanic patients experience lower rates of irAE compared to other racial/ethnic groups across multiple solid tumor types [[46], [47], [48]]. In the I-SPY trial, which evaluates various treatment combinations with immunotherapy in early breast cancer, White patients were shown to be at higher risk of developing hypothyroidism compared to non-White patients [49]. These differences may be due to variations in intrinsic immunology and genetics, epigenetic factors secondary to socioeconomic and cultural differences, higher prevalence of certain comorbidities such as autoimmune disease, and/or disparities in health care reporting, which warrants further evaluation.
Our data also supports the existing literature linking higher AEC values with increased risk of irAE [22,23,25,26,[28], [29], [30]]. Eosinophils may potentiate the risk of irAE through its role in adaptive immunity, whereby eosinophils contribute to antigen presentation to T cells and recruitment of other leukocytes [50]. However, higher AEC values were not significantly different according to onset of grade ≥3 irAE in multivariate analysis. Our study did not identify other peripheral blood markers or ratios as predictive of irAE, which have been associated with irAE in other tumor types [16].
In our series, we found that patients who developed one irAE were at increased risk of developing multiple irAE events. The most common irAE events were within the endocrine system, with cutaneous, gastrointestinal, hepatic, and musculoskeletal systems among the most common irAE events to co-occur. Wan et al. demonstrated similar findings in a retrospective study of more than 15,000 patients, with endocrine irAE most commonly found to have co-occurring irAE in the cutaneous (33%), gastrointestinal (33%), and musculoskeletal (31%) systems [31]. Furthermore, endocrine and cutaneous irAE were associated with improved OS compared to irAE in other organ systems. A meta-analysis of 12 randomized controlled trials also reported the favorable prognostic impact of endocrine irAE on OS [51]. Although endocrine irAE are often permanent and require lifelong treatment, this data suggests that the development of endocrine irAE may indicate a favorable prognosis in patients with breast cancer.
Our study was limited by being retrospective in nature, which hindered our ability to confirm attributions of adverse events to ICI, along with accurate grading of the adverse event. Given that certain types of irAE (e.g., rash or arthralgias) can be caused by chemotherapy, it can also be difficult to distinguish retrospectively whether an adverse event was due to chemotherapy versus ICI. In such cases, adverse events were attributed to ICI if documented by the treating provider to be at least possibly due to ICI treatment or if patients were treated with steroids when the causative agent was unclear. This approach may have led to an overestimation of the rate of irAE since attributions were unable to be prospectively confirmed. On the other hand, low-grade irAE may have also been underreported in medical charts. In addition, the reporting of the management of irAE may have also been underreported in medical charts given the lower than expected number of oral and IV steroids administered to grade 3/4 events.
Additionally, patients included in this study had different ICI therapies, dosing regimens, clinical trial agents, and combination therapies, each of which may influence irAE rates. In addition, the timing of irAE in relation to the neoadjuvant or adjuvant treatment phase was not reported. Our study may also have been underpowered to detect associations between various comorbidities and peripheral blood markers, and did not include a detailed analysis of irAE by subtype due to small sample sizes within each category. Despite these limitations, this study represents the largest real-world cohort of patients with early-stage breast cancer treated with ICI, with an extended follow-up time to assess irAE onset.
5. Conclusion
In this cohort of 423 patients with early-stage breast cancer treated with ICI, the incidence of any- and grade ≥3 irAE was 72.6% and 19.1%, respectively. Older age and CKD were significant predictors of grade ≥3 irAE. Elevated AEC was associated with higher rates of any-grade irAE. As the indication of ICI expands in the treatment of early-stage breast cancer, further studies are urgently needed to predict which individuals will derive significant benefit from ICI compared to those who may be at risk of serious and potentially permanent toxicities.
CRediT authorship contribution statement
Alexis LeVee: Writing – review & editing, Writing – original draft, Visualization, Resources, Project administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Saya Jacob: Writing – review & editing, Project administration, Methodology, Investigation, Data curation, Conceptualization. Samantha Fisch: Writing – review & editing, Investigation. Carolyne Face: Writing – review & editing, Investigation. Madhuri Chengappa: Writing – review & editing, Investigation. Saliha Chaudhry: Writing – review & editing, Investigation. Nikita Baclig: Writing – review & editing, Investigation. Andrew Soliman: Writing – review & editing, Investigation. Nora Ruel: Writing – review & editing, Visualization, Formal analysis, Data curation. Megan Wong: Writing – review & editing, Investigation. Karen Tsai: Writing – review & editing, Investigation. Irene Kang: Writing – review & editing. Austin Kordic: Writing – review & editing. Laura Huppert: Writing – review & editing. Laura Quintal: Writing – review & editing. Michelle Melisko: Writing – review & editing. Melanie Majure: Writing – review & editing. Jo Chien: Writing – review & editing. Melissa Lechner: Writing – review & editing, Resources. Dame Idossa: Writing – review & editing. Anne Blaes: Writing – review & editing, Supervision. Kelly McCann: Writing – review & editing, Supervision. Hope S. Rugo: Writing – review & editing, Supervision, Conceptualization. Joanne Mortimer: Writing – review & editing, Supervision, Conceptualization.
Ethics declaration
The authors have NOT obtained informed consent from participants or their legal representatives. This study was a retrospective chart review with no anticipated risks to participants therefore the need for written informed consent from patients was waived per institutional review board protocols.
This study was performed in compliance with relevant laws, regulatory frameworks and guidelines where the research took place. The University of California San Francisco, University of California Los Angeles, City of Hope, and University of Minnesota granted an exemption. The authors provided the following explanation: This study was a retrospective chart review with no anticipated risks to participants.
Ethical approval
Institutional review board approval was obtained at each institution: University of California, San Francisco IRB #17-22987, University of Minnesota IRB #STUDY00004918, City of Hope IRB #22702, University of California Los Angeles IRB# 24-038. Given this was a retrospective study, written consent from patients was waived per these protocols.
Funding
None.
Declaration of competing interest
Hope Rugo is the social media editor for The Breast. Given her role as the social media editor, she had no involvement in the peer review of this article and had no access to information regarding its peer review. Full responsibility for the editorial process for this article was delegated to another journal editor. The authors declare that they have no other known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
Melissa Lechner: Doris Duke Charitable Foundation, NIH (K08DK129829), Aramont Charitable Foundation.
Footnotes
Supplementary data to this article can be found online at https://doi.org/10.1016/j.breast.2026.104941.
Appendix A. Supplementary data
The following is the Supplementary data to this article:
Data availability
Data that support the findings of this study are available upon reasonable request.
References
- 1.Schmid P., Cortes J., Pusztai L., et al. Pembrolizumab for early triple-negative breast cancer. N Engl J Med. 2020;382(9):810–821. doi: 10.1056/NEJMoa1910549. [DOI] [PubMed] [Google Scholar]
- 2.Cortes J., Cescon D.W., Rugo H.S., et al. Pembrolizumab plus chemotherapy versus placebo plus chemotherapy for previously untreated locally recurrent inoperable or metastatic triple-negative breast cancer (KEYNOTE-355): a randomised, placebo-controlled, double-blind, phase 3 clinical trial. Lancet. 2020;396(10265):1817–1828. doi: 10.1016/S0140-6736(20)32531-9. [DOI] [PubMed] [Google Scholar]
- 3.Peter Schmid, Javier Cortes, Rebecca Dent, et al. Overall survival with pembrolizumab in early-stage triple-negative breast cancer. N Engl J Med. 2024;391(21):1981–1991. doi: 10.1056/NEJMoa2409932. [DOI] [PubMed] [Google Scholar]
- 4.Marhold M., Udovica S., Halstead A., et al. Emergence of immune-related adverse events correlates with pathological complete response in patients receiving pembrolizumab for early triple-negative breast cancer. OncoImmunology. 2023;12(1) doi: 10.1080/2162402X.2023.2275846. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Jayan A., Sukumar J.S., Fangman B., et al. Real-World immune-related adverse events in patients with early triple-negative breast cancer who received pembrolizumab. JCO Oncol Pract. 2024;0(0) doi: 10.1200/OP.24.00371. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Andrade M de O, Gutierres IG, Tavares MC, et al. Immune-related adverse events among patients with early-stage triple-negative breast cancer treated with pembrolizumab plus chemotherapy: real-world data from the Neo-Real/GBECAM 0123 study.Breast. Published online April 10, 2025:104473. doi:10.1016/j.breast.2025.104473. [DOI] [PMC free article] [PubMed]
- 7.Barberi V., Adelantado E.Z., Joval L.R., et al. 105P Immune-related adverse events (irAE) and pathological complete response (pCR) in early TNBC (eTNBC) treated with pembrolizumab (P) plus chemotherapy (CT) ESMO Open. 2025;10 doi: 10.1016/j.esmoop.2025.104659. [DOI] [Google Scholar]
- 8.Rached L., Peyre-Pradat F., Spotti M., et al. Real-world safety and effectiveness of neoadjuvant chemotherapy combination with pembrolizumab in triple-negative breast cancer. ESMO Real World Data and Digital Oncology. 2024;5 doi: 10.1016/j.esmorw.2024.100061. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Hughes J.(.F., McCormick C.A., Smith M., Penny P., Maskell D., Hannon C. 262P real-world data on immune-related adverse events with neoadjuvant pembrolizumab in early triple-negative breast cancer at the Norfolk and Norwich University Hospital. ESMO Open. 2025;10 doi: 10.1016/j.esmoop.2025.104816. [DOI] [Google Scholar]
- 10.Connors C., Valente S.A., ElSherif A., et al. Real-World outcomes with the KEYNOTE-522 regimen in early-stage triple-negative breast cancer. Ann Surg Oncol. 2025;32(2):912–921. doi: 10.1245/s10434-024-16390-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Jayathilaka B., Mian F., Cockwill J., Franchini F., Au-Yeung G., Ijzerman M. Analysis of risk factors for immune-related adverse events induced by immune checkpoint inhibitor treatment in cancer: a comprehensive systematic review. Crit Rev Oncol Hematol. 2025;207 doi: 10.1016/j.critrevonc.2024.104601. [DOI] [PubMed] [Google Scholar]
- 12.Rong Y., Bentley J.P., Bhattacharya K., et al. Incidence and risk factors of immune-related adverse events induced by immune checkpoint inhibitors among older adults with non-small cell lung cancer. Cancer Med. 2024;13(1) doi: 10.1002/cam4.6879. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Eun Y., Kim I.Y., Sun J.M., et al. Risk factors for immune-related adverse events associated with anti-PD-1 pembrolizumab. Sci Rep. 2019;9(1) doi: 10.1038/s41598-019-50574-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Suazo-Zepeda E., Bokern M., Vinke P.C., Hiltermann T.J.N., de Bock G.H., Sidorenkov G. Risk factors for adverse events induced by immune checkpoint inhibitors in patients with non-small-cell lung cancer: a systematic review and meta-analysis. Cancer Immunol Immunother. 2021;70(11):3069–3080. doi: 10.1007/s00262-021-02996-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Gougis P., Jochum F., Abbar B., et al. Clinical spectrum and evolution of immune-checkpoint inhibitors toxicities over a decade—a worldwide perspective. eClinicalMedicine. 2024;70 doi: 10.1016/j.eclinm.2024.102536. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Les I., Martínez M., Pérez-Francisco I., et al. Predictive biomarkers for checkpoint inhibitor immune-related adverse events. Cancers. 2023;15(5) doi: 10.3390/cancers15051629. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Les I., Martínez M., Narro A., et al. Association of immune-related adverse events induced by nivolumab with a battery of autoantibodies. Ann Med. 2021;53(1):762–769. doi: 10.1080/07853890.2021.1931956. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Peng L., Wang Y., Liu F., et al. Peripheral blood markers predictive of outcome and immune-related adverse events in advanced non-small cell lung cancer treated with PD-1 inhibitors. Cancer Immunol Immunother. 2020;69(9):1813–1822. doi: 10.1007/s00262-020-02585-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Egami S., Kawazoe H., Hashimoto H., et al. Peripheral blood biomarkers predict immune-related adverse events in non-small cell lung cancer patients treated with pembrolizumab: a multicenter retrospective study. J Cancer. 2021;12(7):2105–2112. doi: 10.7150/jca.53242. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Michailidou D., Khaki A.R., Morelli M.P., Diamantopoulos L., Singh N., Grivas P. Association of blood biomarkers and autoimmunity with immune related adverse events in patients with cancer treated with immune checkpoint inhibitors. Sci Rep. 2021;11(1):9029. doi: 10.1038/s41598-021-88307-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Sugimoto A., Kaneda H., Yoshimoto N., et al. Derived neutrophil-to-lymphocyte ratio has the potential to predict safety and outcomes of durvalumab after chemoradiation in non-small cell lung cancer. Sci Rep. 2024;14(1) doi: 10.1038/s41598-024-70214-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Nakamura Y., Tanaka R., Maruyama H., et al. Correlation between blood cell count and outcome of melanoma patients treated with anti-PD-1 antibodies. Jpn J Clin Oncol. 2019;49(5):431–437. doi: 10.1093/jjco/hyy201. [DOI] [PubMed] [Google Scholar]
- 23.Ma Y., Ma X., Wang J., Wu S., Wang J., Cao B. Absolute eosinophil count may be an optimal peripheral blood marker to identify the risk of immune-related adverse events in advanced malignant tumors treated with PD-1/PD-L1 inhibitors: a retrospective analysis. World J Surg Oncol. 2022;20(1):242. doi: 10.1186/s12957-022-02695-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Ruste V., Goldschmidt V., Laparra A., et al. The determinants of very severe immune-related adverse events associated with immune checkpoint inhibitors: a prospective study of the French REISAMIC registry. Eur J Cancer. 2021;158:217–224. doi: 10.1016/j.ejca.2021.08.048. [DOI] [PubMed] [Google Scholar]
- 25.Bai R., Chen N., Chen X., et al. Analysis of characteristics and predictive factors of immune checkpoint inhibitor-related adverse events. Cancer Biol Med. 2021;18(4):1118–1133. doi: 10.20892/j.issn.2095-3941.2021.0052. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Chu X., Zhao J., Zhou J., et al. Association of baseline peripheral-blood eosinophil count with immune checkpoint inhibitor-related pneumonitis and clinical outcomes in patients with non-small cell lung cancer receiving immune checkpoint inhibitors. Lung Cancer. 2020;150:76–82. doi: 10.1016/j.lungcan.2020.08.015. [DOI] [PubMed] [Google Scholar]
- 27.Takada S., Murooka H., Tahatsu K., et al. Identifying early predictive markers for immune-related adverse events in nivolumab-treated patients with renal cell carcinoma and gastric cancer. Asian Pac J Cancer Prev APJCP. 2022;23(2):695–701. doi: 10.31557/APJCP.2022.23.2.695. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Caliman E., Fancelli S., Ottanelli C., et al. Absolute eosinophil count predicts clinical outcomes and toxicity in non-small cell lung cancer patients treated with immunotherapy. Cancer Treatment and Research Communications. 2022;32 doi: 10.1016/j.ctarc.2022.100603. [DOI] [PubMed] [Google Scholar]
- 29.Giorgione R., Pillozzi S., Mela M.M., et al. Absolute eosinophil count as predictive biomarker of irAEs in patients with metastatic renal cell carcinoma (mRCC) treated with immune checkpoint inhibitors (ICIs) J Clin Orthod. 2021;39(6_suppl) doi: 10.1200/JCO.2021.39.6_suppl.345. 345-345. [DOI] [Google Scholar]
- 30.Giommoni E., Giorgione R., Paderi A., et al. Eosinophil count as predictive biomarker of immune-related adverse events (irAEs) in immune checkpoint inhibitors (ICIs) therapies in oncological patients. Immuno. 2021;1(3):253–263. doi: 10.3390/immuno1030017. [DOI] [Google Scholar]
- 31.Wan G., Chen W., Khattab S., et al. Multi-organ immune-related adverse events from immune checkpoint inhibitors and their downstream implications: a retrospective multicohort study. Lancet Oncol. 2024;25(8):1053–1069. doi: 10.1016/S1470-2045(24)00278-X. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Wolf D.M., Yau C., Campbell M., et al. Immune subtyping identifies patients with hormone receptor–positive early-stage breast cancer who respond to neoadjuvant immunotherapy (IO): results from five IO arms of the I-SPY2 trial. JCO Precis Oncol. 2025;(9) doi: 10.1200/PO-24-00776. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Loi S., Salgado R., Curigliano G., et al. Neoadjuvant nivolumab and chemotherapy in early estrogen receptor-positive breast cancer: a randomized phase 3 trial. Nat Med. 2025;31(2):433–441. doi: 10.1038/s41591-024-03414-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Cardoso F., O'Shaughnessy J., Liu Z., et al. Pembrolizumab and chemotherapy in high-risk, early-stage, ER+/HER2− breast cancer: a randomized phase 3 trial. Nat Med. 2025;31(2):442–448. doi: 10.1038/s41591-024-03415-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Huober J., Barrios C.H., Niikura N., et al. Atezolizumab with neoadjuvant anti-human epidermal growth factor receptor 2 therapy and chemotherapy in human epidermal growth factor receptor 2-Positive early breast cancer: primary results of the randomized phase III IMpassion050 trial. J Clin Oncol. 2022;40(25):2946–2956. doi: 10.1200/JCO.21.02772. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Gianni L., Huang C.S., Egle D., et al. Pathologic complete response (pCR) to neoadjuvant treatment with or without atezolizumab in triple-negative, early high-risk and locally advanced breast cancer: NeoTRIP Michelangelo randomized study. Ann Oncol. 2022;33(5):534–543. doi: 10.1016/j.annonc.2022.02.004. [DOI] [PubMed] [Google Scholar]
- 37.McArthur H., Chan I., Leal J., et al. Abstract RF3-05: a phase 2 Study of neoadjuvant HER2-targeted therapy +/- immunotherapy with pembrolizumab (neoHIP) Clin Cancer Res. 2025;31(12_Supplement):RF3–RF5. doi: 10.1158/1557-3265.SABCS24-RF3-05. [DOI] [Google Scholar]
- 38.Shah K.P., Song H., Ye F., et al. Demographic factors associated with toxicity in patients treated with anti–programmed cell Death-1 therapy. Cancer Immunol Res. 2020;8(7):851–855. doi: 10.1158/2326-6066.CIR-19-0986. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39.Wong S.K., Blum S.M., Sun X., et al. Efficacy and safety of immune checkpoint inhibitors in young adults with metastatic melanoma. Eur J Cancer. 2023;181:188–197. doi: 10.1016/j.ejca.2022.12.013. [DOI] [PubMed] [Google Scholar]
- 40.Cortazar F.B., Kibbelaar Z.A., Glezerman I.G., et al. Clinical features and outcomes of immune checkpoint inhibitor–associated AKI: a multicenter Study. J Am Soc Nephrol. 2020;31(2) doi: 10.1681/ASN.2019070676. https://journals.lww.com/jasn/fulltext/2020/02000/clinical_features_and_outcomes_of_immune.20.aspx [DOI] [PMC free article] [PubMed] [Google Scholar]
- 41.Gupta S., Short S.A.P., Sise M.E., et al. Acute kidney injury in patients treated with immune checkpoint inhibitors. J Immunother Cancer. 2021;9(10) doi: 10.1136/jitc-2021-003467. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Gérard A.O., Barbosa S., Parassol N., et al. Risk factors associated with immune checkpoint inhibitor-induced acute kidney injury compared with other immune-related adverse events: a case-control study. Clin Kidney J. 2022;15(10):1881–1887. doi: 10.1093/ckj/sfac109. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Seethapathy H., Zhao S., Chute D.F., et al. The incidence, causes, and risk factors of acute kidney injury in patients receiving immune checkpoint inhibitors. Clin J Am Soc Nephrol. 2019;14(12):1692–1700. doi: 10.2215/CJN.00990119. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Meraz-Muñoz A., Amir E., Ng P., et al. Acute kidney injury associated with immune checkpoint inhibitor therapy: incidence, risk factors and outcomes. J Immunother Cancer. 2020;8(1) doi: 10.1136/jitc-2019-000467. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45.Rao Ullur A., Côté G., Pelletier K., Kitchlu A. Immunotherapy in oncology and the kidneys: a clinical review of the evaluation and management of kidney immune-related adverse events. Clin Kidney J. 2023;16(6):939–951. doi: 10.1093/ckj/sfad014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46.Peravali M., Gomes-Lima C., Tefera E., et al. Racial disparities in immune-related adverse events of immune checkpoint inhibitors and association with survival based on clinical and biochemical responses. World J Clin Oncol. 2021;12(2):103–114. doi: 10.5306/wjco.v12.i2.103. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Resnick K., Zang P., Larsen T., et al. Impact of ethnicity and immune-related adverse events (IRAE) on outcomes for non-small cell lung cancer (NSCLC) patients treated with immune checkpoint inhibitors. J Clin Orthod. 2022;40(16_suppl) doi: 10.1200/JCO.2022.40.16_suppl.e21115. e21115-e21115. [DOI] [Google Scholar]
- 48.Florez M.A., Kemnade J.O., Chen N., et al. Persistent ethnicity-associated disparity in Antitumor effectiveness of immune checkpoint inhibitors despite equal access. Cancer Res Commun. 2022;2(8):806–813. doi: 10.1158/2767-9764.CRC-21-0143. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Basu A., Umashankar S., Melisko M., et al. Abstract GS5-04: identification of symptoms that are associated with irAEs in the I-SPY clinical trial. Cancer Res. 2023;83(5_Supplement) doi: 10.1158/1538-7445.SABCS22-GS5-04. [DOI] [Google Scholar]
- 50.Simon S.C.S., Utikal J., Umansky V. Opposing roles of eosinophils in cancer. Cancer Immunol Immunother. 2019;68(5):823–833. doi: 10.1007/s00262-018-2255-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Gomes-Lima C.J., Kwagyan J., King F., Fernandez S.J., Burman K.D., Veytsman I. A comprehensive meta-analysis of endocrine immune-related adverse events of immune checkpoint inhibitors and outcomes in head and neck cancer and lung cancer. J Clin Orthod. 2019;37(15_suppl) doi: 10.1200/JCO.2019.37.15_suppl.e14096. e14096-e14096. [DOI] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Data that support the findings of this study are available upon reasonable request.