ABSTRACT
Background
Combination therapy comprising immune checkpoint inhibitors (ICIs) and chemotherapy is standard treatment for many cancer types. However, the management of adverse events (AEs) induced by combination therapy is more complicated than that of AEs induced by ICI monotherapy. Although immune‐related AEs (irAEs) induced by ICIs and AEs induced by chemotherapy require distinct therapeutic approaches, distinguishing between them can be challenging. Therefore, in this exploratory study, we aimed to investigate whether lymphocyte proportion could serve as a candidate biomarker associated with the occurrence of irAEs in patients receiving ICI plus chemotherapy combination therapy.
Patients and Methods
We categorized 122 patients into the irAE group (those who experienced irAEs), chemo‐AE group (those with chemotherapy‐induced AEs), and non‐AE group (those without AEs) and analyzed their lymphocyte proportions before one course (baseline sample) and two courses (two‐course sample) of treatment.
Results
Lymphocyte proportions of the baseline samples did not differ between the non‐AE, chemo‐AE, and irAE groups. The lymphocyte proportion of the two‐course sample of the irAE group was significantly higher than that of the non‐AE group (25.8% vs. 21.9%; p < 0.05). The lymphocyte proportion of the two‐course sample of the irAE group was significantly higher than that of the chemo‐AE group (25.8% vs. 19.6%; p < 0.05). The multivariate logistic regression analysis showed that a lymphocyte proportion ≥ 23.9% in the two‐course sample significantly increased the risk of irAEs (odds ratio = 3.04; p < 0.05).
Conclusions
An increased lymphocyte proportion before the two courses of treatment was associated with the occurrence of irAEs and may serve as a candidate biomarker.
Keywords: biomarker, chemotherapy, immune checkpoint inhibitor, immune‐related adverse event, lymphocyte
1. Introduction
Immune checkpoint inhibitors (ICIs) are effective against many cancer types; therefore, they are an established innovative treatment modality in the oncology field [1, 2, 3, 4]. Additionally, ICI plus chemotherapy combination therapy as well as ICI monotherapy are standard treatments for many cancer types. ICI plus chemotherapy combination therapy can improve progression‐free survival (PFS) and overall survival (OS) of patients with several malignancy types [5, 6, 7, 8, 9, 10, 11, 12, 13]. However, compared to chemotherapy alone or ICI monotherapy, ICI plus chemotherapy combination therapy has been reported to increase the incidence of any grade and grade ≥ 3 adverse events (AEs) including immune‐related AEs (irAEs) and AEs caused by chemotherapy [14, 15].
ICI plus chemotherapy combination therapy causes two types of AEs: irAEs and AEs caused by chemotherapy [16, 17]. ICIs can result in irAEs caused by excessive immune activation, which affect various organs, such as the skin, gastrointestinal system, and endocrine system [15, 18]. Because irAEs and AEs induced by chemotherapy have different mechanisms and characteristics, interventions for these AEs also differ. Therefore, it is important for healthcare professionals to determine the cause of AEs, such as chemotherapy or ICIs, and diagnose and treat them as early as possible.
In particular, irAEs are associated with a wide variety of organs related to immune activation; however, because of their characteristics, it is difficult to predict when and in which organ they will occur. Additionally, irAEs can have fatal consequences because they can affect all organs [15, 18]. Therefore, identifying biomarkers associated with the occurrence of irAEs in patients receiving ICI plus chemotherapy combination therapy can help differentiate between irAEs and chemotherapy‐induced AEs, thus enabling the early diagnosis and treatment of irAEs. However, reports of biomarkers associated with ICI plus chemotherapy combination therapy remain limited. Therefore, in this exploratory study, we aimed to investigate candidate biomarkers of irAEs in patients undergoing ICI plus chemotherapy combination therapy by analyzing various blood parameters.
2. Materials and Methods
2.1. Study Design and Treatment
We enrolled 190 patients who received ICI plus chemotherapy combination therapy at Nagoya City University Hospital between January 2019 and September 2023 (Figure S1). We excluded 68 patients who had received ICI monotherapy before they received ICI plus chemotherapy combination therapy and whose adverse events could not be classified as irAEs or chemotherapy‐induced AEs based on medical records. Therefore, 122 patients were included in the analysis. These 122 patients were divided into two groups according to whether they experienced irAEs of any grade (irAE group) or no irAEs (non‐irAE group). We further subdivided the non‐irAE group into two groups according to whether the patients had not experienced any AEs (non‐AE group) or had experienced chemotherapy‐induced AEs (chemo‐AE group). Patients who had experienced irAEs or both irAEs and chemotherapy‐induced AEs were defined as the irAE group. All patients were followed up until death or loss of contact. The diagnosis of irAEs was based on a comprehensive clinical assessment, including clinical manifestations, laboratory findings, imaging studies, and pathological findings when available, after excluding alternative causes such as chemotherapy‐induced AEs, infection, and disease progression. When necessary, the diagnosis was made through multidisciplinary discussion involving the treating physician and relevant organ specialists. The severity of adverse events was graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events (CTCAE), version 5.0. This study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of Nagoya City University Graduate School of Medical Science (approval number 60‐22‐0076).
2.2. Data Collection
We analyzed the following 27 peripheral blood parameters before one course (baseline sample) and two courses (two‐course sample) of treatment with ICI plus chemotherapy combination therapy: white blood cell count, red blood cell count, hemoglobin level, neutrophil proportion and count, monocyte proportion and count, lymphocyte proportion and count, eosinophil proportion and count, basophil proportion and count, platelet count, total protein concentration, albumin level, C‐reactive protein level, glutamic pyruvic transaminase level, glutamic oxaloacetic transaminase level, lactate dehydrogenase level, serum creatinine level, blood urea nitrogen level, sodium level, potassium level, chloride level, calcium level, and total bilirubin level. Because of the retrospective nature of this study, blood parameters were unavailable for some patients at the two‐course sample. Therefore, analyses were performed using the available data. The median time from the start of ICI plus chemotherapy combination therapy to two courses of treatment was 27 days (range, 15–62 days).
2.3. Least Absolute Shrinkage and Selection Operator (LASSO) Regression
LASSO regression was performed to identify candidate variables associated with the occurrence of irAEs. Among the 27 laboratory parameters that were evaluated in the two‐course sample, variables that were considered likely to exhibit multicollinearity were excluded before model construction. The final candidate variables entered into the LASSO model included white blood cell count, hemoglobin, neutrophil proportion, monocyte proportion, eosinophil proportion, basophil proportion, platelet count, albumin, C‐reactive protein, glutamic pyruvic transaminase, glutamic oxaloacetic transaminase, lactate dehydrogenase, serum creatinine, blood urea nitrogen, sodium, potassium, chloride, calcium, bilirubin, a dichotomized lymphocyte proportion ≥ 23.9%, age, sex, PS, tumor type, and ICI regimen. The optimal penalty parameter (λ) was determined using 10‐fold cross‐validation, and variables with non‐zero coefficients were included in the multivariate logistic regression analysis.
2.4. Statistical Analysis
Statistical significance was considered when p < 0.05. Differences in the quantified data of the groups were compared using Mann–Whitney U tests. Fisher's exact test and one‐way analysis of variance were used to assess differences in patient characteristics. The optimal cutoff points for potential peripheral blood biomarkers to predict irAE onset were determined by analyzing the receiver‐operating characteristic curves.
To assess the internal validity and potential optimism of the multivariate logistic regression model, internal validation was performed using bootstrap resampling with 1000 repetitions. The model included lymphocyte proportion ≥ 23.9% in the two‐course sample, breast cancer, ipilimumab plus nivolumab treatment, neutropenia, and neutrophil count in the two‐course sample. The optimism‐corrected area under the receiver operating characteristic curve (AUC) was calculated by subtracting the mean optimism from the apparent AUC. The median OS and PFS were calculated using the Kaplan–Meier method and log‐rank tests. Statistical analyses were performed using GraphPad Prism 9 software and EZR (Saitama Medical Center, Jichi Medical University) [19].
3. Results
3.1. Patient Characteristics
Table 1 shows the patient characteristics at the beginning of ICI plus chemotherapy combination therapy. The median age of the patients was 69 years. Of the 122 patients, 73% (n = 89) were male, and 27% (n = 33) were female. Of the enrolled patients, 82% (n = 100) had a performance status of 0 or 1, and 8.2% (n = 10) had a performance status of 2 to 4. The primary tumor types were lung cancer (58.2%; n = 71), esophageal cancer (14.8%; n = 18), gastric cancer (10.7%; n = 13), head and neck cancer (7.4%; n = 9), breast cancer (6.6%; n = 8), and cholangiocarcinoma (2.5%; n = 3). Most patients received pembrolizumab (49.2%; n = 60), followed by atezolizumab (18.9%; n = 23), nivolumab (13.1%; n = 16), durvalumab (10.7%; n = 13), and ipilimumab plus nivolumab (8.2%; n = 10). All patients were administered steroids as supportive therapy, such as nausea suppressant, allergy prophylaxis, and edema prevention. The median observation duration was 149 days (range: 21–1446 days).
TABLE 1.
Clinical features of non‐irAE and irAE groups.
| Characteristics | All patients | Non‐irAE group | irAE group | p |
|---|---|---|---|---|
| Total, n (%) | 122 | 66 (54.1) | 56 (45.9) | |
| Age (range) | 69 (28–85) | 68 (28–85) | 70 (48–85) | 0.33 |
| Sex, n (%) | ||||
| Male | 89 (73.0) | 46 (69.7) | 43 (76.8) | 0.42 |
| Female | 33 (27.0) | 20 (30.3) | 13 (23.2) | |
| Performance status, n (%) | ||||
| 0–1 | 100 (82.0) | 53 (80.3) | 47 (83.9) | 1.00 |
| 2–4 | 10 (8.2) | 5 (7.6) | 5 (8.9) | |
| Unknown | 12 (9.8) | 8 (12.1) | 4 (7.1) | |
| Primary tumor type, n (%) | ||||
| Lung cancer | 71 (58.2) | 37 (56.1) | 34 (60.7) | 0.86 |
| Esophageal cancer | 18 (14.8) | 9 (13.6) | 9 (16.1) | 0.80 |
| Gastric cancer | 13 (10.7) | 7 (10.6) | 6 (10.7) | 1.00 |
| Head and neck cancer | 9 (7.4) | 4 (6.1) | 5 (8.9) | 0.41 |
| Breast cancer | 8 (6.6) | 7 (10.6) | 1 (1.8) | 0.07 |
| Cholangio carcinoma | 3 (2.5) | 2 (3.0) | 1 (1.8) | 1.00 |
| Immune checkpoint inhibitor type, n (%) | ||||
| Pembrolizumab | 60 (49.2) | 32 (48.5) | 28 (50.0) | 1.00 |
| Pemetrexed + Pembrolizumab | 1 | 0 | 1 | |
| Carboplatin + Pemetrexed + Pembrolizumab | 18 | 8 | 10 | |
| Carboplatin + nab‐Paclitaxel + Pembrolizumab | 8 | 4 | 4 | |
| Carboplatin + Paclitaxel + Pembrolizumab | 6 | 5 | 1 | |
| Cisplatin + Pemetrexed + Pembrolizumab | 4 | 3 | 1 | |
| Fluorouracil + Cisplatin + Pembrolizumab | 21 | 10 | 11 | |
| Gemcitabine + Carboplatin + Pembrolizumab | 2 | 2 | 0 | |
| Atezolizumab | 23 (18.9) | 16 (24.2) | 7 (12.5) | 0.11 |
| Carboplatin + Etoposide + Atezolizumab | 5 | 3 | 2 | |
| Carboplatin + nab‐Paclitaxel + Atezolizumab | 7 | 5 | 2 | |
| Carboplatin + Pemetrexed + Atezolizumab | 1 | 0 | 1 | |
| Carboplatin + Paclitaxel + Bevacizumab + Atezolizumab | 10 | 8 | 2 | |
| Nivolumab | 16 (13.1) | 7 (10.6) | 9 (16.1) | 0.43 |
| Fluorouracil + Cisplatin + Nivolumab | 5 | 3 | 2 | |
| Folinic acid + Fluorouracil + Oxaliplatin + Nivolumab | 4 | 1 | 3 | |
| S‐1 + Oxaliplatin + Nivolumab | 7 | 3 | 4 | |
| Durvalumab | 13 (10.7) | 8 (12.1) | 5 (8.9) | 0.77 |
| Carboplatin + Etoposide + Durvalumab | 9 | 6 | 3 | |
| Gemcitabine + Cisplatin + Durvalumab | 4 | 2 | 2 | |
| Ipilimumab plus Nivolumab | 10 (8.2) | 3 (4.5) | 7 (12.5) | 0.18 |
| Carboplatin + Pemetrexed + Ipilimumab + Nivolumab | 6 | 2 | 4 | |
| Carboplatin + Paclitaxel + Ipilimumab + Nivolumab | 1 | 0 | 1 | |
| Cisplatin + Pemetrexed + Ipilimumab + Nivolumab | 3 | 1 | 2 | |
| Use of steroid as supportive therapy, n (%) | 122 (100.0) | 66 (100.0) | 56 (100.0) | 1.00 |
| Duration of observation, median (day, range) | 149 (21–1446) | 174 (21–915) | 225 (21–1446) | 0.13 |
Abbreviations: irAE, immune‐related adverse event; nab, nanoparticle albumin‐bound.
Of the 122 patients, 54.1% (n = 66) comprised the non‐irAE group, and 45.9% (n = 56) comprised the irAE group. Among the patients in the irAE group (n = 56), 16.1% (n = 9) experienced irAEs during one course of treatment, whereas the remaining 83.9% (n = 47) developed irAEs after two courses of treatment. The distributions of age, sex, performance status, primary tumor type, ICI type, steroid use, and observation duration did not differ between the irAE and non‐irAE groups.
3.2. AEs Induced by ICIs or Chemotherapy
Fifty‐six patients experienced irAEs of any grade (82 events), including grade ≥ 3 (14 events) (Table S1). Among irAEs of any grade, endocrine disorders (36.6%; 30 events) were the most common, followed by skin disorders (25.6%; 21 events), gastrointestinal disorders (22.0%; 18 events), pulmonary disorders (11.0%; 9 events), and other disorders (4.9%; 4 events). Grade ≥ 3 irAEs included endocrine disorders (2.4%; 2 events), skin disorders (2.4%; 2 events), gastrointestinal disorders (7.2%; 6 events), pulmonary disorders (2.4%; 2 events), and other disorders (2.4%; 2 events).
The proportions of irAEs of any grade were 6.1% (5 events) at 0 to < 2 weeks, 14.6% (12 events) at ≥ 2 to < 4 weeks, 17.1% (14 events) at ≥ 4 to < 8 weeks, 14.6% (12 events) at ≥ 8 to < 12 weeks, and 47.6% (39 events) at ≥ 12 weeks after treatment (Table S2). The proportions of grade ≥ 3 irAEs were 1.2% (1 event) from 0 to < 2 weeks, 1.2% (1 event) from ≥ 2 to < 4 weeks, 1.2% (1 event) from ≥ 4 to < 8 weeks, 4.9% (4 events) from ≥ 8 to < 12 weeks, and 8.5% (7 events) ≥ 12 weeks after treatment.
Table S3 shows the profiles of chemotherapy‐induced AEs. Ninety‐four AEs of any grade induced by chemotherapy were observed, including 15 grade ≥ 3 AEs (Table S3). The following chemotherapy‐induced AEs of any grade were observed: neutropenia (41.5%; 39 events), nausea (14.9%; 14 events), peripheral neuropathy (12.8%; 12 events), fatigue (8.5%; 8 events), mouth ulcers (6.4%; 6 events), diarrhea (3.2%; 3 events), hiccupping (3.2%; 3 events), rash (3.2%; 3 events), edema (2.1%; 2 events), hearing loss (1.1%; 1 event), liver injury (1.1%; 1 event), muscle pain (1.1%; 1 event), and renal disorders (1.1%; 1 event). Grade ≥ 3 AEs were neutropenia (10.6%; 10 events), nausea (2.1%; 2 events), peripheral neuropathy (2.4%; 2 events), and renal disorders (1.1%; 1 event).
3.3. Analysis of the Variations of 27 Peripheral Blood Parameters Over Time
To identify a candidate biomarker for irAE, we comprehensively examined the variations of 27 peripheral blood parameters over time. There were no statistically significant differences in the 27 peripheral blood parameters between the non‐irAE and irAE groups at baseline (Figure 1A and Figure S2). The proportion of lymphocytes in the two‐course sample of the irAE group was significantly higher than that of the non‐irAE group (25.8% vs. 21.0%; p < 0.05) (Figure 1B); however, other peripheral blood parameters did not differ between the irAE and non‐irAE groups (Figure S3). After excluding 9 patients who developed irAEs during the one course of treatment, the lymphocyte proportion in the two‐course sample remained significantly higher in the irAE group than in the non‐irAE group (27.3% vs. 21.0%; p < 0.05) (Figure S4).
FIGURE 1.
An increased lymphocyte proportion reflects the onset of immune‐related adverse events (irAEs) of any grade. (A) Boxplot showing the lymphocyte proportion in the baseline sample of the non‐irAE group (n = 66) and in that of the irAE group (n = 56). (B) Boxplot showing the lymphocyte proportion in the two‐course sample of the non‐irAE group (n = 55) and in that of the irAE group (n = 47). Mann–Whitney U tests were performed.
3.4. Increased Lymphocyte Proportions and irAE Occurrences
We focused on the increased lymphocyte proportion over time (Figure 1A,B) and investigated the differences in irAEs and chemotherapy‐induced AEs. Statistically significant differences in patient characteristics such as age, sex, performance status, primary tumor type, ICI type, steroid use, and observation duration of the non‐AE, chemo‐AE, and irAE groups were not observed (Table S4). The lymphocyte proportion in the baseline sample of the non‐AE group and that in the baseline sample of the irAE group did not differ significantly (20.4% vs. 20.1%; p = 0.94) (Figure 2A). Additionally, the lymphocyte proportion in the baseline sample of the chemo‐AE group and that in the baseline sample of the irAE group did not differ significantly (18.2% vs. 20.1%; p = 0.16) (Figure 2B). The proportion of lymphocytes in the two‐course sample of the irAE group was significantly higher than that of the non‐AE group (25.8% vs. 21.9%; p < 0.05) (Figure 2C). The lymphocyte proportion in the two‐course sample of the irAE group was significantly higher than that of the chemo‐AE group (25.8% vs. 19.6%; p < 0.05) (Figure 2D), indicating that an increased lymphocyte proportion over time may be associated with the occurrence of irAEs.
FIGURE 2.
Increased lymphocyte proportions and the development of immune‐related adverse events (irAEs). (A) Boxplot showing the proportion of lymphocytes in the baseline sample of the group without adverse events (AEs) (non‐AE group; n = 27) and in that of the irAE group (n = 56). (B) Boxplot showing the proportion of lymphocytes in the baseline sample of the group with chemotherapy‐induced AEs (chemo‐AE group; n = 39) and in that of the irAE group (n = 56). (C) Boxplot showing the proportion of lymphocytes in the two‐course sample of the non‐AE group (n = 22) and in that of the irAE group (n = 47). (D) Boxplot showing the proportion of lymphocytes in the two‐course sample of the chemo‐AE group (n = 33) and in that of the irAE group (n = 47). Mann–Whitney U tests were performed.
3.5. Risk Factors Associated With irAEs of Any Grade
Finally, we investigated whether associations between increased lymphocyte proportions and the occurrence of irAEs of any grade existed. Based on the receiver‐operating characteristic curve, we determined that the optimal cutoff value for the lymphocyte proportion in the two‐course sample to predict the occurrence of irAEs of any grade was 23.9% (area under the curve = 0.64; 95% confidence interval [CI], 0.53–0.75; sensitivity, 0.62; specificity, 0.66) (Figure 3). The univariate and multivariate logistic regression analyses showed that a lymphocyte proportion ≥ 23.9% in the two‐course sample was associated with the occurrence of irAEs of any grade (univariate analysis: odds ratio [OR] 2.63; 95% CI, 1.19–5.97; p < 0.05; multivariate analysis: OR, 3.04; 95% CI, 1.40–6.62; p < 0.05) (Table 2). Internal validation of the multivariate logistic regression model using bootstrap resampling showed an apparent AUC of 0.680. The mean optimism was 0.047, resulting in an optimism‐corrected AUC of 0.633. In the LASSO logistic regression model, three variables were retained at λmin: lymphocyte proportion ≥ 23.9% in the two‐course sample (coefficient = 0.359), breast cancer (coefficient = −0.228), and ipilimumab plus nivolumab therapy (coefficient = 0.040). Among these, lymphocyte proportion had the largest coefficient (0.359), suggesting that it was the variable associated with irAE occurrence (Figure S5).
FIGURE 3.
Optimal cutoff value of the lymphocyte proportion before two courses of treatment with immune checkpoint inhibitors (ICIs) plus chemotherapy combination therapy. Receiver‐operating characteristic curve analysis of the lymphocyte proportion associated with the occurrence of immune‐related adverse events (irAEs) of any grade.
TABLE 2.
Univariate and multivariate logistic regression analysis of risk factors for the occurrence of any grade irAE.
| Univariate | Multivariate | |||||
|---|---|---|---|---|---|---|
| OR | 95% CI | p | OR | 95% CI | p | |
| Age: ≥ 65 years | 1.27 | 0.51–3.27 | 0.67 | |||
| Sex: male | 1.43 | 0.59–3.55 | 0.42 | |||
| Performance status (2–4): yes | 0.84 | 0.18–3.86 | 1.00 | |||
| Lung cancer: yes | 1.11 | 0.51–2.42 | 0.86 | |||
| Esophageal cancer: yes | 1.21 | 0.39–3.75 | 0.80 | |||
| Gastric cancer: yes | 1.01 | 0.26–3.77 | 1.00 | |||
| Head and neck cancer: yes | 1.51 | 0.31–8.05 | 0.73 | |||
| Breast cancer: yes | 0.16 | 0.03–1.27 | 0.07 | 0.16 | 0.02–1.41 | 0.09 |
| Cholangio carcinoma: yes | 0.58 | 0.01–11.51 | 1.00 | |||
| Pembrolizumab: yes | 1.06 | 0.49–2.30 | 1.00 | |||
| Atezolizumab: yes | 0.45 | 0.14–1.28 | 0.11 | |||
| Nivolumab: yes | 1.61 | 0.49–5.49 | 0.43 | |||
| Durvalumab: yes | 0.71 | 0.17–2.66 | 0.77 | |||
| Ipilimumab + Nivolumab: yes | 2.97 | 0.64–18.73 | 0.18 | 3.71 | 0.87–15.90 | 0.26 |
| Neutropenia: yes | 0.64 | 0.30–1.39 | 0.26 | 0.60 | 0.23–1.60 | 0.31 |
| White blood cell count in two‐course sample | 0.91 | 0.79–1.05 | 0.20 | |||
| Neutrophil count in two‐course sample | 0.94 | 0.81–1.09 | 0.40 | 1.02 | 0.88–1.19 | 0.75 |
| Absolute lymphocyte count in two‐course sample | 0.99 | 0.50–1.96 | 0.98 | |||
| Proportion of lymphocyte in two course sample: ≥ 23.9% | 2.63 | 1.19–5.97 | < 0.05 | 3.04 | 1.40–6.62 | < 0.05 |
Abbreviations: ECOG PS, eastern cooperative oncology performance status; irAE, immune‐related adverse event.
3.6. Relationship Between Lymphocyte Proportion, irAE Type/Severity, Primary Tumor Type, and ICI Regimen
To investigate whether the association between lymphocyte proportion and irAE occurrence differed according to the affected organ, we performed organ‐specific analyses. The lymphocyte proportion in the two‐course sample was significantly higher in patients who developed gastrointestinal (27.7% vs. 20.1%, p < 0.05) and endocrine (26.6% vs. 20.1%, p < 0.05) irAEs than in those who did not develop irAEs. No significant differences were observed in patients with skin or pulmonary irAEs (Figure S6). We also evaluated the association between lymphocyte proportion and severe (grade ≥ 3) irAEs. No significant difference in lymphocyte proportion was observed between patients with grade ≥ 3 irAEs and those without irAEs (26.6% vs. 20.1%, p = 0.17) (Figure S7). Subgroup analyses according to tumor type and ICI regimen revealed no consistent subgroup‐specific associations (Figure S8).
3.7. Relationship Between Lymphocyte Proportion and OS and PFS
Patients with a lymphocyte proportion ≥ 23.9% in the two‐course sample tended to have longer OS (median OS, 14.9 vs. 12.3 months; log‐rank p = 0.07), whereas no significant association was observed with PFS (median PFS, 5.31 vs. 3.84 months; log‐rank p = 0.42) (Figure S9).
4. Discussion
In the current study, we found that an increased lymphocyte proportion before two courses of treatment for patients treated with ICI plus chemotherapy combination therapy was associated with irAE occurrence. Two studies of irAE predictors that included a cohort of patients who underwent ICI plus chemotherapy combination therapy have been published [20, 21]. Although those studies reported an association between eosinophils or autoantibodies and the occurrence of irAEs, they analyzed patients who underwent both ICI monotherapy and ICI plus chemotherapy or combination therapy comprising an anti‐angiogenic agent [20, 21]. Therefore, there is a limited number of studies on a biomarker that can determine whether AEs are attributable to ICIs or chemotherapy for patients undergoing ICI plus chemotherapy combination therapy. To our knowledge, this is one of the few studies to explore biomarkers associated with irAEs of any grade induced by ICI plus chemotherapy combination therapy.
Many studies have reported predictors of irAEs for patients treated with ICI monotherapy, such as cytokines, autoantibodies, blood parameters, microbiome variants, and genetic variants [22, 23]. We previously found that an increased eosinophil count is a predictor of irAEs for patients treated with ICI monotherapy [24, 25, 26]. However, unlike the results of our previous studies [24, 25, 26], the eosinophil variations between the baseline and two‐course samples in the current study were not significantly different (Figures S2H,I and S3H,I). Differences in blood cell variations observed during our previous and current studies may have been attributable to the suppression of bone marrow function caused by chemotherapy and administration of steroids as supportive therapy, such as nausea suppressant, allergy prophylaxis, and edema prevention. Therefore, patients undergoing ICI monotherapy and those undergoing ICI plus chemotherapy combination therapy should be analyzed separately to determine predictors of irAEs. In this study, we found that an increased lymphocyte proportion before two courses of treatment may be associated with the development of irAEs but not with that of chemotherapy‐induced AEs in patients undergoing ICI plus chemotherapy combination therapy. Additionally, we analyzed a cohort of patients receiving ICI plus chemotherapy combination therapy; therefore, our approach using the lymphocyte proportion before two courses of treatment may enable the early diagnosis and treatment of irAEs or chemotherapy‐induced AEs in such patients.
Several studies have reported that ICIs that target programmed death‐1 (PD‐1) or cytotoxic T‐lymphocyte‐associated antigen 4 (CTLA‐4) exert anti‐tumor effects by activating and increasing lymphocytes [27, 28, 29, 30, 31]. For instance, Iwai et al. revealed that the percentages of CD4 and CD8 T cells increased in mice treated with PD‐1 antibody or PD‐1 knockout mice compared to those in wild‐type mice (CD4 T cells: wild‐type, 17%; PD‐1 antibody, 24%; PD‐1 knockout, 42%; CD8 T cells: wild‐type, 3%; PD‐1 antibody, 6%; PD‐1 knockout, 16%) [29]. Additionally, recent studies have shown that CTLA‐4 blockade‐associated colitis is driven by excessive activation of IFNγ‐producing CD4 T cells together with depletion of peripherally induced regulatory T cells, resulting in the loss of immune tolerance [32, 33]. Therefore, ICI administration may be closely related to an increase in lymphocytes, which is associated with the occurrence of irAEs. These findings support our findings of increased lymphocyte proportions.
Steroids have been reported to suppress the proliferation and differentiation of naive T cells, suggesting a potential inhibitory effect on immune activation [34]. By contrast, Shen et al. reported that although steroids are routinely administered as supportive therapy to prevent chemotherapy‐related toxicities in patients receiving ICI plus chemotherapy combination therapy, no reduction in the antitumor efficacy of ICIs was observed, suggesting that the effects of steroids on immune responses depend on the clinical context in which they are used [35]. In the present study, because all patients received steroids as supportive therapy during ICI plus chemotherapy combination therapy, differences in steroid exposure between the irAE and non‐irAE groups were minimal (Table 1). Nevertheless, we cannot completely rule out the possibility that corticosteroids influenced peripheral lymphocyte dynamics, which should be considered when interpreting our findings.
Several clinical trials that involved large cohorts have reported various incidence rates of irAEs induced by ICI plus chemotherapy combination therapy, including 40% reported by KEYNOTE‐189 [6], 35% reported by KEYNOTE‐407 [7], 26% reported by KEYNOTE‐859 [8], 26% reported by KEYNOTE‐355 [9], 45% reported by IMpower130 [10], 28% reported by IMpower133 [11], 20% reported by CASPIAN [12], and 36% reported by Govindan et al. [13] In the current study, the incidence of irAEs was 45.9%, which was consistent with the results of previous clinical trials.
Studies of the onset of irAEs caused by ICIs have shown that skin disorders occur within 2 to 3 weeks, gastrointestinal disorders occur within 6 to 7 weeks, and endocrine disorders occur after 9 weeks of treatment [18]. Daban et al. reported that the median time to onset of irAEs of any grade in patients treated with combination therapy with ICIs and chemotherapy was 13 weeks [36]. In the present study, the median onset time of irAEs was 13.3 weeks; additionally, skin disorders occurred at 2 weeks, gastrointestinal disorders occurred at 6.5 weeks, and endocrine disorders occurred at 9.7 weeks. Our results are consistent with those of previous studies of the onset of ICI‐induced irAEs. Moreover, of the 82 irAEs of any grade in the current study, 20.7% (17 events) occurred within 4 weeks of treatment initiation, whereas 79.3% (65 events) developed more than 4 weeks after treatment initiation. Because approximately 80% of patients developed irAEs more than 4 weeks after treatment initiation, observing lymphocyte dynamics before two courses of treatment are administered may be appropriate for predicting the onset of irAEs.
This study had some limitations. First, it was a retrospective study; therefore, patient selection bias could not be controlled. Additionally, patients who developed both irAEs and chemotherapy‐induced AEs were included in the irAE group, which may have introduced classification bias and reduced the specificity of lymphocyte proportion as a biomarker for irAEs. Second, some laboratory data were missing during follow‐up, resulting in differences in sample sizes between the baseline and two‐course samples. Third, the optimal cutoff value was derived and evaluated within the same dataset, which may have introduced overfitting. Moreover, the discriminatory performance of the ROC analysis was modest; therefore, the proposed cutoff value should be interpreted with caution. Fourth, the study population was heterogeneous in terms of cancer types and treatment regimens. Additionally, subgroup analyses according to tumor type and ICI regimen were limited by the small sample size within each subgroup, which may have reduced statistical power and increased the uncertainty of the estimated effects, as reflected by the wide CIs. This heterogeneity may have also contributed to the lack of a significant association between lymphocyte proportion and OS or PFS. Given these limitations, validation in larger prospective, independent cohorts is required to confirm the robustness and generalizability of our findings.
5. Conclusion
In conclusion, an increased lymphocyte proportion was associated with the occurrence of irAEs in patients treated with ICI plus chemotherapy combination therapy and may represent a candidate biomarker. These findings suggest that monitoring lymphocyte proportion can help differentiate between irAEs and chemotherapy‐induced AEs and facilitate their earlier identification; however, further validation in larger prospective studies is warranted.
Author Contributions
Nanami Ito: conceptualization, methodology, data curation, investigation, formal analysis, writing – original draft, writing – review and editing. Yoshihiko Tasaki: conceptualization, methodology, data curation, formal analysis, investigation, writing – original draft, writing – review and editing, funding acquisition. Yosuke Sugiyama: conceptualization, methodology, data curation, investigation, formal analysis, writing – original draft, writing – review and editing, funding acquisition. Takehiro Uemura: data curation, formal analysis, writing – review and editing. Keisuke Yokota: writing – review and editing, data curation, investigation. Ryo Ogawa: data curation, writing – review and editing, investigation. Takaya Shimura: data curation, writing – review and editing, investigation. Daisuke Kawakita: investigation, data curation, writing – review and editing. Yumi Wanifuchi‐Endo: investigation, writing – review and editing, data curation. Yoshihisa Mimura: data curation, investigation, writing – review and editing. Moeko Iida: data curation, investigation, writing – review and editing. Kunihiro Odagiri: investigation, writing – review and editing, data curation. Yuka Kimura: investigation, writing – review and editing, data curation. Misato Tomita: data curation, writing – review and editing, investigation. Yuji Hotta: investigation, writing – review and editing, data curation. Tatsuya Toyama: writing – review and editing, supervision, data curation, investigation. Shinichi Iwasaki: data curation, supervision, writing – review and editing, investigation. Hiromi Kataoka: investigation, writing – review and editing, data curation, supervision. Shuji Takiguchi: data curation, supervision, writing – review and editing, investigation. Katsuhiro Okuda: investigation, writing – review and editing, supervision, data curation. Akio Niimi: investigation, writing – review and editing, supervision, data curation. Yoko Furukawa‐Hibi: supervision, writing – review and editing, conceptualization.
Funding
This work was supported by JSPS KAKENHI Grant Number 25K18657 (Y. T.) and the Nitto Foundation (Y.S.).
Ethics Statement
This study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of Nagoya City University Graduate School of Medical Science (approval number 60‐22‐0076).
Consent
As this was a retrospective study, patient consent was not required. Patients could also choose to opt out of the study by using the authors' institutional websites.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Figure S1: Patient enrollment flowchart. ICI: immune checkpoint inhibitors, irAE, immune‐related adverse event.
Figure S2: Examination of peripheral blood parameters of baseline sample. (A–Z), Boxplot showing values in the baseline sample for: (A) white blood cell count (non‐irAE (n = 66) and irAE (n = 56) groups), (B) red blood cell count (non‐irAE (n = 66) and irAE (n = 56) groups), (C) hemoglobin level (non‐irAE (n = 66) and irAE (n = 56) groups), (D) neutrophil proportion (non‐irAE (n = 66) and irAE (n = 56) groups), (E) neutrophil count (non‐irAE (n = 66) and irAE (n = 56) groups), (F) monocyte proportion (non‐irAE (n = 66) and irAE (n = 56) groups), (G) monocyte count (non‐irAE (n = 66) and irAE (n = 56) groups), (H) lymphocyte count (non‐irAE (n = 66) and irAE (n = 56) groups), (I) eosinophil proportion (non‐irAE (n = 66) and irAE (n = 56) groups), (J) eosinophil count (non‐irAE (n = 66) and irAE (n = 56) groups), (K) basophil proportion (non‐irAE (n = 66) and irAE (n = 56) groups), (L) basophil count (non‐irAE (n = 66) and irAE (n = 56) groups), (M) platelet count (non‐irAE (n = 66) and irAE (n = 56) groups), (N) total protein concentration (non‐irAE (n = 66) and irAE (n = 56) groups), (O) albumin level (non‐irAE (n = 66) and irAE (n = 56) groups), (P) C‐reactive protein level (non‐irAE (n = 66) and irAE (n = 56) groups), (Q) glutamic pyruvic transaminase level (non‐irAE (n = 66) and irAE (n = 56) groups), (R) glutamic oxaloacetic transaminase level (non‐irAE (n = 66) and irAE (n = 56) groups), (S) lactate dehydrogenase level (non‐irAE (n = 66) and irAE (n = 56) groups), (T) serum creatinine level (non‐irAE (n = 66) and irAE (n = 56) groups), (U) blood urea nitrogen level (non‐irAE (n = 66) and irAE (n = 56) groups), (V) sodium level (non‐irAE (n = 66) and irAE (n = 56) groups), (W) potassium level (non‐irAE (n = 66) and irAE (n = 56) groups), (X) chloride level (non‐irAE (n = 66) and irAE (n = 56) groups), (Y) calcium level (non‐irAE (n = 66) and irAE (n = 56) groups), and (Z) bilirubin level (non‐irAE (n = 66) and irAE (n = 56) groups). (A–Z) Mann–Whitney U test. irAEs, immune‐related adverse events.
Figure S3: Variation of peripheral blood parameters before two course treatment with ICI plus chemotherapy combination therapy. (A–Z), Boxplot showing values in the two‐course sample for: (A) white blood cell count (non‐irAE (n = 55) and irAE (n = 47) groups), (B) red blood cell count (non‐irAE (n = 55) and irAE (n = 47) groups), (C) hemoglobin level (non‐irAE (n = 55) and irAE (n = 47) groups), (D) neutrophil proportion (non‐irAE (n = 55) and irAE (n = 47) groups), (E) neutrophil count (non‐irAE (n = 55) and irAE (n = 47) groups), (F) monocyte proportion (non‐irAE (n = 55) and irAE (n = 47) groups), (G) monocyte count (non‐irAE (n = 55) and irAE (n = 47) groups), (H) lymphocyte count (non‐irAE (n = 55) and irAE (n = 47) groups), (I) eosinophil proportion (non‐irAE (n = 55) and irAE (n = 47) groups), (J) eosinophil count (non‐irAE (n = 55) and irAE (n = 47) groups), (K) basophil proportion (non‐irAE (n = 55) and irAE (n = 47) groups), (L) basophil count (non‐irAE (n = 55) and irAE (n = 47) groups), (M) platelet count (non‐irAE (n = 55) and irAE (n = 47) groups), (N) total protein concentration (non‐irAE (n = 55) and irAE (n = 47) groups), (O) albumin level (non‐irAE (n = 55) and irAE (n = 47) groups), (P) C‐reactive protein level (non‐irAE (n = 55) and irAE (n = 47) groups), (Q) glutamic pyruvic transaminase level (non‐irAE (n = 54) and irAE (n = 47) groups), (R) glutamic oxaloacetic transaminase level (non‐irAE (n = 54) and irAE (n = 47) groups), (S) lactate dehydrogenase level (non‐irAE (n = 44) and irAE (n = 38) groups), (T) serum creatinine level (non‐irAE (n = 54) and irAE (n = 46) groups), (U) blood urea nitrogen level (non‐irAE (n = 54) and irAE (n = 46) groups), (V) sodium level (non‐irAE (n = 55) and irAE (n = 47) groups), (W) potassium level (non‐irAE (n = 55) and irAE (n = 47) groups), (X) chloride level (non‐irAE (n = 55) and irAE (n = 47) groups), (Y) calcium level (non‐irAE (n = 55) and irAE (n = 47) groups), and (Z) bilirubin level (non‐irAE (n = 53) and irAE (n = 47) groups). (A–Z) Mann–Whitney U test. irAEs, immune‐related adverse events.
Figure S4: An increased lymphocyte proportion reflects the onset of immune‐related adverse events (irAEs) after excluding patients who developed irAEs during one course of treatment. Boxplot showing the lymphocyte proportion in the (A) baseline sample of the non‐irAE group (n = 66) and in that of the irAE group (n = 47). (B) Boxplot showing the lymphocyte proportion in the two‐course sample of the non‐irAE group (n = 55) and in that of irAE group (n = 41). Mann–Whitney U test were performed.
Figure S5: LASSO coefficient profiles of candidate variables associated with immune‐related adverse events. Coefficient profiles of candidate variables generated by LASSO regression are plotted against log (λ). LASSO, least absolute shrinkage and selection operator.
Figure S6: Associations between lymphocyte proportion in the two‐course sample and organ‐specific immune‐related adverse events (irAEs). (A–D), Boxplots showing the lymphocyte proportion in the two‐course sample for: (A) endocrine irAEs (non‐irAE group (n = 55) and irAE group (n = 22)), (B) skin irAEs (non‐irAE group (n = 55) and irAE group (n = 18)), (C) gastrointestinal irAEs (non‐irAE group (n = 55) and irAE group (n = 18)), and (D) pulmonary irAEs (non‐irAE group (n = 55) and irAE group (n = 9)). (A–D) Mann–Whitney U test. irAEs, immune‐related adverse events.
Figure S7: Association between lymphocyte proportion in the two‐course sample and grade ≥ 3 irAEs. Boxplot showing the lymphocyte proportion in the two‐course sample for the non‐irAE group (n = 55) and grade ≥ 3 irAE group (n = 13). Mann–Whitney U test. irAEs, immune‐related adverse events.
Figure S8: Forest plot of subgroup analyses according to ICI regimens and tumor types. Subgroup analyses of the association between lymphocyte proportion ≥ 23.9% in the two‐course sample and irAEs according to ICI regimen and tumor type. ICI: immune checkpoint inhibitor, irAEs: immune‐related adverse events.
Figure S9: OS and PFS according to lymphocyte proportion in the two‐course sample. (A) Kaplan–Meier curves for OS in patients with a lymphocyte proportion < 23.9% (n = 72) and ≥ 23.9% (n = 50) in the two‐course sample. (B) Kaplan–Meier curves for PFS in patients with a lymphocyte proportion < 23.9% (n = 72) and ≥ 23.9% (n = 50) in the two‐course sample. (A, B) Survival distributions were compared using the log‐rank test. OS, Overall survival. PFS, progression‐free survival.
Table S1: Profile of irAEs.
Table S2: Time from initiation of treatment to irAEs occurrences.
Table S3: Profile of adverse events induced by chemotherapy.
Table S4: Clinical features between non‐adverse event, chemotherapy‐adverse event, irAE groups.
Acknowledgments
The authors have nothing to report.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1. Tie Y., Yang H., Zhao R., et al., “Safety and Efficacy of Atezolizumab in the Treatment of Cancers: A Systematic Review and Pooled‐Analysis,” Drug Design, Development and Therapy 13 (2019): 523–538, 10.2147/DDDT.S188893. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Kang Y. K., Boku N., Satoh T., et al., “Nivolumab in Patients With Advanced Gastric or Gastro‐Oesophageal Junction Cancer Refractory to, or Intolerant of, at Least Two Previous Chemotherapy Regimens (ONO‐4538‐12, ATTRACTION‐2): A Randomised, Double‐Blind, Placebo‐Controlled, Phase 3 Trial,” Lancet 390, no. 10111 (2017): 2461–2471, 10.1016/S0140-6736(17)31827-5. [DOI] [PubMed] [Google Scholar]
- 3. O'Byrne K., Popoff E., Badin F., et al., “Long‐Term Comparative Efficacy and Safety of Nivolumab Plus Ipilimumab Relative to Other First‐Line Therapies for Advanced Non‐Small‐Cell Lung Cancer: A Systematic Literature Review and Network Meta‐Analysis,” Lung Cancer 177 (2023): 11–20, 10.1016/j.lungcan.2023.01.006. [DOI] [PubMed] [Google Scholar]
- 4. Wang Y., Zhang T., Huang Y., et al., “Real‐World Safety and Efficacy of Consolidation Durvalumab After Chemoradiation Therapy for Stage III Non‐Small Cell Lung Cancer: A Systematic Review and Meta‐Analysis,” International Journal of Radiation Oncology, Biology, Physics 112, no. 5 (2022): 1154–1164, 10.1016/j.ijrobp.2021.12.150. [DOI] [PubMed] [Google Scholar]
- 5. Inoue T. and Narukawa M., “Anti‐Tumor Efficacy of Anti‐PD‐1/PD‐L1 Antibodies in Combination With Other Anticancer Drugs in Solid Tumors: A Systematic Review and Meta‐Analysis,” Cancer Control 29 (2022): 10732748221140694, 10.1177/10732748221140694. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Horinouchi H., Nogami N., Saka H., et al., “Pembrolizumab Plus Pemetrexed‐Platinum for Metastatic Nonsquamous Non‐Small‐Cell Lung Cancer: KEYNOTE‐189 Japan Study,” Cancer Science 112, no. 8 (2021): 3255–3265, 10.1111/cas.14980. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Novello S., Kowalski D. M., Luft A., et al., “Pembrolizumab Plus Chemotherapy in Squamous Non‐Small‐Cell Lung Cancer: 5‐Year Update of the Phase III KEYNOTE‐407 Study,” Journal of Clinical Oncology 41, no. 11 (2023): 1999–2006, 10.1200/JCO.22.01990. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Rha S. Y., Oh D. Y., Yañez P., et al., “Pembrolizumab Plus Chemotherapy Versus Placebo Plus Chemotherapy for HER2‐Negative Advanced Gastric Cancer (KEYNOTE‐859): A Multicentre, Randomised, Double‐Blind, Phase 3 Trial,” Lancet Oncology 24, no. 11 (2023): 1181–1195, 10.1016/S1470-2045(23)00515-6. [DOI] [PubMed] [Google Scholar]
- 9. 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 396, no. 10265 (2020): 1817–1828, 10.1016/S0140-6736(20)32531-9. [DOI] [PubMed] [Google Scholar]
- 10. West H., McCleod M., Hussein M., et al., “Atezolizumab in Combination With Carboplatin Plus Nab‐Paclitaxel Chemotherapy Compared With Chemotherapy Alone as First‐Line Treatment for Metastatic Non‐Squamous Non‐Small‐Cell Lung Cancer (IMpower130): A Multicentre, Randomised, Open‐Label, Phase 3 Trial,” Lancet Oncology 20, no. 7 (2019): 924–937, 10.1016/S1470-2045(19)30167-6. [DOI] [PubMed] [Google Scholar]
- 11. Mansfield A. S., Każarnowicz A., Karaseva N., et al., “Safety and Patient‐Reported Outcomes of Atezolizumab, Carboplatin, and Etoposide in Extensive‐Stage Small‐Cell Lung Cancer (IMpower133): A Randomized Phase I/III Trial,” Annals of Oncology 31, no. 2 (2020): 310–317, 10.1016/j.annonc.2019.10.021. [DOI] [PubMed] [Google Scholar]
- 12. Paz‐Ares L., Dvorkin M., Chen Y., et al., “Durvalumab Plus Platinum‐Etoposide Versus Platinum‐Etoposide in First‐Line Treatment of Extensive‐Stage Small‐Cell Lung Cancer (CASPIAN): A Randomised, Controlled, Open‐Label, Phase 3 Trial,” Lancet 394, no. 10212 (2019): 1929–1939, 10.1016/S0140-6736(19)32222-6. [DOI] [PubMed] [Google Scholar]
- 13. Govindan R., Szczesna A., Ahn M. J., et al., “Phase III Trial of Ipilimumab Combined With Paclitaxel and Carboplatin in Advanced Squamous Non‐Small‐Cell Lung Cancer,” Journal of Clinical Oncology 35, no. 30 (2017): 3449–3457, 10.1200/JCO.2016.71.7629. [DOI] [PubMed] [Google Scholar]
- 14. Rached L., Laparra A., Sakkal M., et al., “Toxicity of Immunotherapy Combinations With Chemotherapy Across Tumor Indications: Current Knowledge and Practical Recommendations,” Cancer Treatment Reviews 127 (2024): 102751, 10.1016/j.ctrv.2024.102751. [DOI] [PubMed] [Google Scholar]
- 15. Wang Y., Zhou S., Yang F., et al., “Treatment‐Related Adverse Events of PD‐1 and PD‐L1 Inhibitors in Clinical Trials: A Systematic Review and Meta‐Analysis,” JAMA Oncology 5, no. 7 (2019): 1008–1019, 10.1001/jamaoncol.2019.0393. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Yamanaka Y., Okuno Y., Kamisako K., et al., “Efficacy and Safety Evaluation of Immune Checkpoint Inhibitors in Combination With Chemotherapy for Extensive Small Cell Lung Cancer: Real‐World Evidence,” Cancer Medicine 13, no. 12 (2024): e70480, 10.1002/cam4.70480. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Sundar R., Chia D. K. A., Zhao J. J., et al., “Phase I PIANO Trial‐PIPAC‐Oxaliplatin and Systemic Nivolumab Combination for Gastric Cancer Peritoneal Metastases: Clinical and Translational Outcomes,” ESMO Open 9, no. 9 (2024): 103681, 10.1016/j.esmoop.2024.103681. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Weber J. S., Kähler K. C., and Hauschild A., “Management of Immune‐Related Adverse Events and Kinetics of Response With Ipilimumab,” Journal of Clinical Oncology 30, no. 21 (2012): 2691–2697, 10.1200/JCO.2012.41.6750. [DOI] [PubMed] [Google Scholar]
- 19. Kanda Y., “Investigation of the Freely Available Easy‐To‐Use Software ‘EZR’ for Medical Statistics,” Bone Marrow Transplantation 48, no. 3 (2013): 452–458, 10.1038/bmt.2012.244. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Daban A., Gonnin C., Phan L., et al., “Preexisting Autoantibodies as Predictor of Immune Related Adverse Events (irAEs) for Advanced Solid Tumors Treated With Immune Checkpoint Inhibitors (ICIs),” Oncoimmunology 12, no. 1 (2023): 2204754, 10.1080/2162402X.2023.2204754. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. 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 150 (2020): 76–82, 10.1016/j.lungcan.2020.08.015. [DOI] [PubMed] [Google Scholar]
- 22. Poto R., Troiani T., Criscuolo G., et al., “Holistic Approach to Immune Checkpoint Inhibitor‐Related Adverse Events,” Frontiers in Immunology 13 (2022): 804597, 10.3389/fimmu.2022.804597. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Les I., Martínez M., Pérez‐Francisco I., et al., “Predictive Biomarkers for Checkpoint Inhibitor Immune‐Related Adverse Events,” Cancers 15, no. 5 (2023): 1629, 10.3390/cancers15051629. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Tasaki Y., Sugiyama Y., Hamamoto S., et al., “Eosinophil May Be a Predictor of Immune‐Related Adverse Events Induced by Different Immune Checkpoint Inhibitor Types: A Retrospective Multidisciplinary Study,” Cancer Medicine 12 (2023): 21666–21679. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Tasaki Y., Hamamoto S., Sugiyama Y., et al., “Elevated Eosinophils Proportion as Predictor of Immune‐Related Adverse Events After Ipilimumab and Nivolumab Treatment of Advanced and Metastatic Renal Cell Carcinoma,” International Journal of Urology 30 (2023): 866–874. [DOI] [PubMed] [Google Scholar]
- 26. Tasaki Y., Hamamoto S., Yamashita S., et al., “Eosinophil Is a Predictor of Severe Immune‐Related Adverse Events Induced by Ipilimumab Plus Nivolumab Therapy in Patients With Renal Cell Carcinoma: A Retrospective Multicenter Cohort Study,” Frontiers in Immunology 15 (2024): 1483956, 10.3389/fimmu.2024.1483956. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Pardoll D. M., “The Blockade of Immune Checkpoints in Cancer Immunotherapy,” Nature Reviews Cancer 12, no. 4 (2012): 252–264, 10.1038/nrc3239. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Kitazawa Y., Fujino M., Wang Q., et al., “Involvement of the Programmed Death‐1/Programmed Death‐1 Ligand Pathway in CD4+CD25+ Regulatory T‐Cell Activity to Suppress Alloimmune Responses,” Transplantation 83, no. 6 (2007): 774–782, 10.1097/01.tp.0000256293.90270.e8. [DOI] [PubMed] [Google Scholar]
- 29. Iwai Y., Terawaki S., and Honjo T., “PD‐1 Blockade Inhibits Hematogenous Spread of Poorly Immunogenic Tumor Cells by Enhanced Recruitment of Effector T Cells,” International Immunology 17, no. 2 (2005): 133–144, 10.1093/intimm/dxh194. [DOI] [PubMed] [Google Scholar]
- 30. Leach D. R., Krummel M. F., and Allison J. P., “Enhancement of Antitumor Immunity by CTLA‐4 Blockade,” Science 271, no. 5256 (1996): 1734–1736, 10.1126/science.271.5256.1734. [DOI] [PubMed] [Google Scholar]
- 31. Geng R., Tang H., You T., et al., “Peripheral CD8+CD28+ T Lymphocytes Predict the Efficacy and Safety of PD‐1/PD‐L1 Inhibitors in Cancer Patients,” Frontiers in Immunology 14 (2023): 1125876, 10.3389/fimmu.2023.1125876. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Lo B. C., Kryczek I., Yu J., et al., “Microbiota‐Dependent Activation of CD4(+) T Cells Induces CTLA‐4 Blockade‐Associated Colitis via Fcγ Receptors,” Science 383, no. 6678 (2024): 62–70, 10.1126/science.adh8342. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Luoma A. M., Suo S., Williams H. L., et al., “Molecular Pathways of Colon Inflammation Induced by Cancer Immunotherapy,” Cell 182, no. 3 (2020): 655–671.e22, 10.1016/j.cell.2020.06.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Giles A. J., Hutchinson M. K. N. D., Sonnemann H. M., et al., “Dexamethasone‐Induced Immunosuppression: Mechanisms and Implications for Immunotherapy,” Journal for Immunotherapy of Cancer 6, no. 1 (2018): 51, 10.1186/s40425-018-0371-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35. Shen Y. C., Pinato D. J., Liu T. H., et al., “Revisiting Prophylactic Corticosteroids to Mitigate Severe Immune‐Related Adverse Events in Hepatocellular Carcinoma,” Journal of Hepatology 84, no. 6 (2026): 1178–1182, 10.1016/j.jhep.2026.01.009. [DOI] [PubMed] [Google Scholar]
- 36. Ng K. Y. Y., Tan S. H., Tan J. J. E., et al., “Impact of Immune‐Related Adverse Events on Efficacy of Immune Checkpoint Inhibitors in Patients With Advanced Hepatocellular Carcinoma,” Liver Cancer 11, no. 1 (2022): 9–21, 10.1159/000518619. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Figure S1: Patient enrollment flowchart. ICI: immune checkpoint inhibitors, irAE, immune‐related adverse event.
Figure S2: Examination of peripheral blood parameters of baseline sample. (A–Z), Boxplot showing values in the baseline sample for: (A) white blood cell count (non‐irAE (n = 66) and irAE (n = 56) groups), (B) red blood cell count (non‐irAE (n = 66) and irAE (n = 56) groups), (C) hemoglobin level (non‐irAE (n = 66) and irAE (n = 56) groups), (D) neutrophil proportion (non‐irAE (n = 66) and irAE (n = 56) groups), (E) neutrophil count (non‐irAE (n = 66) and irAE (n = 56) groups), (F) monocyte proportion (non‐irAE (n = 66) and irAE (n = 56) groups), (G) monocyte count (non‐irAE (n = 66) and irAE (n = 56) groups), (H) lymphocyte count (non‐irAE (n = 66) and irAE (n = 56) groups), (I) eosinophil proportion (non‐irAE (n = 66) and irAE (n = 56) groups), (J) eosinophil count (non‐irAE (n = 66) and irAE (n = 56) groups), (K) basophil proportion (non‐irAE (n = 66) and irAE (n = 56) groups), (L) basophil count (non‐irAE (n = 66) and irAE (n = 56) groups), (M) platelet count (non‐irAE (n = 66) and irAE (n = 56) groups), (N) total protein concentration (non‐irAE (n = 66) and irAE (n = 56) groups), (O) albumin level (non‐irAE (n = 66) and irAE (n = 56) groups), (P) C‐reactive protein level (non‐irAE (n = 66) and irAE (n = 56) groups), (Q) glutamic pyruvic transaminase level (non‐irAE (n = 66) and irAE (n = 56) groups), (R) glutamic oxaloacetic transaminase level (non‐irAE (n = 66) and irAE (n = 56) groups), (S) lactate dehydrogenase level (non‐irAE (n = 66) and irAE (n = 56) groups), (T) serum creatinine level (non‐irAE (n = 66) and irAE (n = 56) groups), (U) blood urea nitrogen level (non‐irAE (n = 66) and irAE (n = 56) groups), (V) sodium level (non‐irAE (n = 66) and irAE (n = 56) groups), (W) potassium level (non‐irAE (n = 66) and irAE (n = 56) groups), (X) chloride level (non‐irAE (n = 66) and irAE (n = 56) groups), (Y) calcium level (non‐irAE (n = 66) and irAE (n = 56) groups), and (Z) bilirubin level (non‐irAE (n = 66) and irAE (n = 56) groups). (A–Z) Mann–Whitney U test. irAEs, immune‐related adverse events.
Figure S3: Variation of peripheral blood parameters before two course treatment with ICI plus chemotherapy combination therapy. (A–Z), Boxplot showing values in the two‐course sample for: (A) white blood cell count (non‐irAE (n = 55) and irAE (n = 47) groups), (B) red blood cell count (non‐irAE (n = 55) and irAE (n = 47) groups), (C) hemoglobin level (non‐irAE (n = 55) and irAE (n = 47) groups), (D) neutrophil proportion (non‐irAE (n = 55) and irAE (n = 47) groups), (E) neutrophil count (non‐irAE (n = 55) and irAE (n = 47) groups), (F) monocyte proportion (non‐irAE (n = 55) and irAE (n = 47) groups), (G) monocyte count (non‐irAE (n = 55) and irAE (n = 47) groups), (H) lymphocyte count (non‐irAE (n = 55) and irAE (n = 47) groups), (I) eosinophil proportion (non‐irAE (n = 55) and irAE (n = 47) groups), (J) eosinophil count (non‐irAE (n = 55) and irAE (n = 47) groups), (K) basophil proportion (non‐irAE (n = 55) and irAE (n = 47) groups), (L) basophil count (non‐irAE (n = 55) and irAE (n = 47) groups), (M) platelet count (non‐irAE (n = 55) and irAE (n = 47) groups), (N) total protein concentration (non‐irAE (n = 55) and irAE (n = 47) groups), (O) albumin level (non‐irAE (n = 55) and irAE (n = 47) groups), (P) C‐reactive protein level (non‐irAE (n = 55) and irAE (n = 47) groups), (Q) glutamic pyruvic transaminase level (non‐irAE (n = 54) and irAE (n = 47) groups), (R) glutamic oxaloacetic transaminase level (non‐irAE (n = 54) and irAE (n = 47) groups), (S) lactate dehydrogenase level (non‐irAE (n = 44) and irAE (n = 38) groups), (T) serum creatinine level (non‐irAE (n = 54) and irAE (n = 46) groups), (U) blood urea nitrogen level (non‐irAE (n = 54) and irAE (n = 46) groups), (V) sodium level (non‐irAE (n = 55) and irAE (n = 47) groups), (W) potassium level (non‐irAE (n = 55) and irAE (n = 47) groups), (X) chloride level (non‐irAE (n = 55) and irAE (n = 47) groups), (Y) calcium level (non‐irAE (n = 55) and irAE (n = 47) groups), and (Z) bilirubin level (non‐irAE (n = 53) and irAE (n = 47) groups). (A–Z) Mann–Whitney U test. irAEs, immune‐related adverse events.
Figure S4: An increased lymphocyte proportion reflects the onset of immune‐related adverse events (irAEs) after excluding patients who developed irAEs during one course of treatment. Boxplot showing the lymphocyte proportion in the (A) baseline sample of the non‐irAE group (n = 66) and in that of the irAE group (n = 47). (B) Boxplot showing the lymphocyte proportion in the two‐course sample of the non‐irAE group (n = 55) and in that of irAE group (n = 41). Mann–Whitney U test were performed.
Figure S5: LASSO coefficient profiles of candidate variables associated with immune‐related adverse events. Coefficient profiles of candidate variables generated by LASSO regression are plotted against log (λ). LASSO, least absolute shrinkage and selection operator.
Figure S6: Associations between lymphocyte proportion in the two‐course sample and organ‐specific immune‐related adverse events (irAEs). (A–D), Boxplots showing the lymphocyte proportion in the two‐course sample for: (A) endocrine irAEs (non‐irAE group (n = 55) and irAE group (n = 22)), (B) skin irAEs (non‐irAE group (n = 55) and irAE group (n = 18)), (C) gastrointestinal irAEs (non‐irAE group (n = 55) and irAE group (n = 18)), and (D) pulmonary irAEs (non‐irAE group (n = 55) and irAE group (n = 9)). (A–D) Mann–Whitney U test. irAEs, immune‐related adverse events.
Figure S7: Association between lymphocyte proportion in the two‐course sample and grade ≥ 3 irAEs. Boxplot showing the lymphocyte proportion in the two‐course sample for the non‐irAE group (n = 55) and grade ≥ 3 irAE group (n = 13). Mann–Whitney U test. irAEs, immune‐related adverse events.
Figure S8: Forest plot of subgroup analyses according to ICI regimens and tumor types. Subgroup analyses of the association between lymphocyte proportion ≥ 23.9% in the two‐course sample and irAEs according to ICI regimen and tumor type. ICI: immune checkpoint inhibitor, irAEs: immune‐related adverse events.
Figure S9: OS and PFS according to lymphocyte proportion in the two‐course sample. (A) Kaplan–Meier curves for OS in patients with a lymphocyte proportion < 23.9% (n = 72) and ≥ 23.9% (n = 50) in the two‐course sample. (B) Kaplan–Meier curves for PFS in patients with a lymphocyte proportion < 23.9% (n = 72) and ≥ 23.9% (n = 50) in the two‐course sample. (A, B) Survival distributions were compared using the log‐rank test. OS, Overall survival. PFS, progression‐free survival.
Table S1: Profile of irAEs.
Table S2: Time from initiation of treatment to irAEs occurrences.
Table S3: Profile of adverse events induced by chemotherapy.
Table S4: Clinical features between non‐adverse event, chemotherapy‐adverse event, irAE groups.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.