Serum cholesterol-riched apolipoprotein-B (apoB) containing lipoproteins predict the response to immune checkpoints inhibitors (ICIs) in non-small cell lung cancer (NSCLC) : A retrospective analysis

doi:10.21203/rs.3.rs-1749981/v2

Objectives: Lipid metabolism alternations reprogrammed tumor microenvironment (TME) to influence the response to immune checkpoint inhibitors (ICIs) in non-small cell lung cancer (NSCLC). Body mass index (BMI) and apolipoproteins B (apoB) including low-density lipoprotein cholesterol (LDL-C) and remnant cholesterol (RC) reflect the lipid metabolism status. The associations between BMI, lipoproteins, and clinical outcome of ICIs are far from clear.

Methods: A retrospective case-control analysis of 99 NSCLC patients treated with chemotherapy plus ICIs regimens in the department of oncology and lung cancer center of China-Japan Friendship Hospital was performed. The prognostic value of the BMI, serum lipids, lipoproteins, and apolipoproteins were assessed. Efﬁcacy was evaluated by the ICIs response (durable clinical beneﬁt [DCB] / non-durable benefit [NDB]), best response (active/ non-active) and progression-free survival (PFS). The validation test and the mechanism exploration were performed with public data using bioinformatic methods.

Results: BMI ≥ 25 kg/m² was a risk indicator for NDB, non-active response to ICIs, shorter PFS with the confounders adjusted (OR= 6.06, 95% CI 2.05 - 19.89, p.val = 0.002; OR = 4.37, 95% CI 1.64 - 12.43, p.val = 0.004; HR = 3.08, 95% CI 1.63 - 5.82, p.val < 0.001). High apoB containing lipoproteins were risk factors for poor ICIs response. Low RC predicted the better ICIs response and the best response with the satisfied discriminative ability and the well-fitting calibration curves (OR = 0.12, 95%CI 0.02 - 0.63, p.val = 0.017; OR = 0.22. 95%CI 0.05 - 0.96, p.val = 0.047). Serum LDL-C predicted the longer PFS of ICIs with the added value to the BMI containing model (HR = 0.43, 95%CI 0.22 - 0.86, p.val = 0.016). Bioinformatic exploration showed that the LUAD samples with high APOB were infiltrated by elevated fibroblasts and other immune-suppressive cells. Functional enrichment analysis indicated that the acquired ICIs resistance might be through the LRP6/5-Wnt-TGF-β axis which required laboratory experiments to verify.

Conclusions: BMI ≥ 25 kg/m², the elevated apoB containing lipoproteins indicated poor response to the combination of chemotherapy and ICIs in the in the NSCLC population. Further large-scale validations and experimental explorations are needed.

immune checkpoint inhibitor

non-small cell lung cancer

body mass index

low-density lipoprotein cholesterol

remnant cholesterol

Lung cancer accounts for the leading cause of cancer-associated mortality, about 85% of which are non-small cell lung cancers (NSCLC) including squamous cell carcinoma, adenocarcinoma, and large cell carcinomas (¹). Immune checkpoint inhibitors (ICIs) targeting the programmed cell death protein 1 (PD-1)/ programmed death-ligand 1 (PD‐L1) axis have markedly led to a dramatic change in the treatment of NSCLC especially those without molecular targets. Chemotherapy can enhance tumor antigenicity to improve the efficacy of ICIs and lower the incidence of immune-related adverse events (irAEs). Thereby, the combination regimen of chemotherapy and ICIs has been considered the standard first-line treatment for NSCLC patients without sensitive mutations (²).

Lipids participate in energy source and storage and play critical roles in the constitution of the structural basis of biological membranes and signaling molecules. Altered lipid metabolism is among the most prominent metabolic alterations in cancer. Enhanced synthesis or uptake of lipids contributes to tumorigenesis (³). The crosstalk between altered lipid metabolisms and the adaptable tumor microenvironment (TME) can strongly reprogram tumor initiation, growth, invasion, metastasis, and response to therapies. Lipid metabolism was also plastic and shaped by TME, which allows tumors to thrive (⁴, ⁵).

Obesity-associated inflammation dysregulates immune response, potentially leading profound effects on the toxicity and efficacy of ICIs. Body mass index (BMI) is clinically used to assess body fat in human subjects (⁶, ⁷). Past studies declared a positive prognostic role of high BMI in NSCLC patients treated by first-line single agent ICIs-based regimens, whereas conflict result was obtained in the chemoimmunotherapy combinations (6-⁸). Consistently, high BMI was demonstrated to be associated with high occurrence of immune-related adverse events (irAEs) (⁹, ¹⁰).

Circulating lipids including cholesterol and triglycerides are packaged into apolipoproteins and transported in the form of lipoproteins. Lipoproteins are classiﬁed by their density and in a state of continuous ﬂux due to the shifting of the composition of lipids within the lipoproteins in the blood circulation (¹¹). The cholesterol carried by apolipoprotein B (apoB) containing lipoproteins, including low-density lipoprotein cholesterol (LDL-C) and remnant cholesterol (RC) are the predominant atherogenic agent that feeds the development of arterial wall plaques (¹²). Researchers revealed that statins showed a positive impact on the ICIs treatment outcome (¹³). The positive association of cancer risk and apoB level has been declared, while high LDL-C was suggested to be associated with high incidence of cancer and metastasis in most cancers (¹⁴, ¹⁵). Recent studies indicated a positive correlation between LDL-C, HDL-C and serum cholesterol and the ICIs monotherapy response, whereas the contrary results were obtained in the chemotherapy groups (¹⁶). Since the apoB containing lipoproteins have been shown as the risk factors for tumorigenesis, the results remained to be controversial (¹⁷). In the light of the available evidence, the roles of lipoproteins in the ICIs response are far from clear.

To elucidate the association between lipoproteins and clinical outcome of ICIs in NSCLC, we performed a retrospective case-control analysis of NSCLC patients treated with chemotherapy plus ICIs regimens in the department of oncology and lung cancer center of China-Japan Friendship Hospital.

2.1 Patient cohorts:

We retrospectively collected the clinical data of NSCLC who were administrated with the regimens of chemotherapy plus anti-PD-1/PD-L1 from January 2019 to December 2021 and followed up in the department of oncology and lung cancer center at China-Japan Friendship Hospital. For the inclusion criteria: (1) Patients had a diagnosis of Stage III-IV NSCLC at the ﬁrst ICIs initiation regardless of the treatment line. The radiographic data or medical records used to evaluate the efficacy of ICIs were required. The disease was staged according to the 7th and 8th editions of the tumor, node, metastasis system (TNM) classiﬁcation (¹⁸). (2) Patients with Stage I - II NSCLC who pre-operatively received ICIs and had the post-operative pathological data used to confirm the efficacy of the neoadjuvant therapy. For the exclusion criteria: (1) Patients whose lung or mediastinal tumor lesions received radiotherapy or interventional ablation therapy during the ICIs treatment. (2) Patients who were not evaluable radiologically. (3) Patients discontinued ICIs for the detected sensible Epidermal Growth Factor Receptor (EGFR) alterations to receive target agents.

The clinical response was evaluated by the Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 (¹⁹). Computed tomography (CT) studies were independently read by two radiologists. The primary outcome measured was progressive disease (PD). Progression-free survival (PFS) was assessed from the start date of chemoimmunotherapy to the date of progression. Patients who discontinued ICIs for the serious adverse events or had not progressed at the termination of data collection were censored at the date of their last scan. Cases retrospectively adjudicated to not be PD per RECIST but determined in real-time by the treating clinician as PD were considered as events. In the study, the evaluation was stable disease (SD) with insufﬁcient shrinkage to qualify for PR, then the response was considered as SD-a. SD-b was defined as SD without sufﬁcient increase to qualify for PD. If the target lesion had no changes radiologically, the response was determined as SD.

The clinical response to ICIs was defined with two methods. Firstly, durable clinical benefit (DCB) was defined as partial response (PR) or SD that lasted more than 6 months, whereas non-durable benefit (NDB) was considered as progressive disease (PD) or SD that lasted less than 6 months (²⁰). Secondly, the clinical response was also evaluated as the best response to ICIs. The best response to ICIs was PR or SD-a was regarded as the active response to ICIs. Patients with SD lasting more than 6 months were also classified as active in ICIs. The non-active response to ICIs was defined by meeting any of the following conditions: (1) The evaluation of PD in the first following-up; (2) The best response to ICIs was SD-b, indicating the disease progressed slowly during the treatment of ICIs; (3) SD lasted less than 6 months.

The ICIs response (NDB/DCB) was regarded as the primary outcome. The secondary outcome was the best response to ICIs and the incidence of irAEs of any grade. IrAEs were graded according to the US National Cancer Institute Common Toxicity Criteria for Adverse Events (NCI-CTCAE; version 5.0).

We complied with all relevant ethical regulations for working with human participants. Informed consent was obtained. This study was approved by the institutional review board at China-Japan Friendship Hospital (2022-KY-051).

2.2 Collection of clinical data

Baseline demographic and clinical data were collected from patients' medical records on the first administration of ICIs or during the ICIs treatment. Self-reported cigarette smoking and physical activity characteristics were extracted from the medical records. Diabetes status (yes/no) at baseline was based on either one of the following criteria: fasting plasma glucose levels greater than 126 mg/dL, non-fasting plasma glucose >200 mg/dL, glycated hemoglobin (HbA1c) ≥ 6.5%, taking anti-diabetic medication within 2 weeks of baseline data collection, or self-report of a diagnosis of diabetes. Smoking history (yes/no) was defined as either active smoking or having a previous smoking history. The patients’ pre-treatment height and body weight were recorded. BMI was calculated by dividing a patient’s weight in kilograms (kg) by the square of the patient’s height in meters (m). BMI was classiﬁed according to the World Health Organization (WHO) categories: underweight, BMI < 18.5 kg/m²; normal, 18.5 kg/m² ≤ BMI ≤ 24.9 kg/m²; overweight, 25 kg/m² ≤ BMI ≤ 29.9 kg/m²; obesity, BMI ≥ 30 kg/m².

Blood samples were obtained after 8-hour of overnight fasting before the pre-treatment or during the treatment of ICIs. Serum concentrations of total cholesterol (TC) (mmol/L), total triglycerides (TG) (mmol/ L), high-density lipoprotein cholesterol (HDL-C) (mmol/L), LDL-C (mmol/L), apoA1 (g/L), apoB (g/L) levels were measured and recorded. TG, TC, LDL-C and HDL-C were measured with Beckman Coulter kit. The apolipoproteins were measured by Nittbo Medical Co.LTD kit. The baseline apoB/(apoB+ApoA1) ratio was calculated as apoB level divided by the sum of apoB and apoA1 levels. RC is the cholesterol content of all non-LDL-C and non-HDL-C and was calculated as TC minus LDL-C and HDL-C. Patients were divided into two groups based on the threshold of serum LDL-C and RC levels. The thresholds of LDL-C and RC were selected by the “surv-cutpoint” function implemented in the survminer R package since non-reports about the cut points of the markers in NSCLC were available.

2.3 Bioinformatic validation

Public data including GSE126044 and GSE135222 were downloaded from Gene Expression Omnibus (GEO) datasets (https://www.ncbi.nlm.nih.gov/) used for the validation of the correlation between apoB level and the ICIs response (²¹, ²²). The Cancer Genome Atlas (TCGA) lung adenocarcinoma (LUAD) and lung squamous cell carcinoma (LUSC) cohorts’ level 3 RNA-seq data (HTSeq-Counts) were directly downloaded by GDC data transfer tool (https://portal.gdc.cancer.gov/). Minus germ-line somatic copy number alternations (sCNA) and merged somatic simple nucleotide variations (sSNV) segmented data of TCGA LUAD cohort were downloaded from GDAC Firehose (Broad Institute TCGA Genome Data Analysis Center, https://gdac.broadinstitute.org/) for oncogene status analysis. The corresponding DNA methylation data (IDATs) of TCGA-LUAD and TCGA-LUSC methylation beta value was obtained. The Illumina Human Methylation450 bead chip raw data were processed by minfi and ChAMP package (²³, ²⁴).

LUAD samples with EGFR mutations were excluded from the analysis for the acknowledged resistance to ICIs. The immunophenotypes of TCGA NSCLC samples were classified with the method previously introduced by single-sample gene set enrichment analysis (ssGSEA) (²⁵, ²⁶). The immunologically cold tumors, which were characterized by fewer infiltrated immune cytotoxic cells and low expression of immune regulators in the tumor microenvironment (TME), were excluded from the analysis for the insensitivity to the ICIs. The consistency of different clusters was verified by cytolytic activity (CYT), which represented the ultimate effective mechanism in the cancer immunity cycle and was calculated as the geometric mean of granzyme A (GZMA) and perforin (PRF1) expression levels as previously defined (²⁷). Another two methods were applied to measure the TME, including MCPcounter and EpiDISH, which were used to estimate the population abundance of tissue-infiltrating immune and stromal cell populations in heterogeneous tissues from transcriptomic data and methylation data respectively (²⁸, ²⁹). The validation cohorts were divided into two groups respectively according to the median expression of APOB genes of every cohort. Functional enrichment analysis between groups was realized by GSVA based on gene expression data matrices with (³⁰). Differentially expressed genes (DEGs) were identified by DESeq2 between NDB and DCB patients in the LUAD subgroup of GSE126044 and among the immunophenotype clusters in the TCGA-LUAD cohort (³¹).

2.4 Statistical analysis

Statistical analysis was conducted with R software (version 4, 4.0.4). The study size was calculated according to the promary endpoint (NDB/DCB) with the expectant OR = 2, α = 0.05 and β = 0.10. Baseline characteristics are presented as means and standard deviations, or as medians and interquartile ranges (IQR) for continuous variables, and as frequencies and percentages for categorical variables. BMI was regarded as a continuous and categorical variable respectively. Patients with BMI ≥ 25mg/m² were combined as the overweight and obesity group for analysis and BMI < 25 kg/m² was treated as the reference. T-test or Mann Whitney U test was used for comparison between the two groups. Distributions of the groups were estimated by the Chi test. The unadjusted univariable and multivariable logistic regression models were used to estimate crude and adjusted odds ratios (OR) for BMI, circulating apolipoprotein, and lipids respectively. BMI was also regarded as a confounder to influence the serum lipids. The other confounders including age, gender (female or male), DM (yes or no), and smoking history (yes or no) were used for the multivariate regression. The missing values were handled with the listwise deletion which was the default method in R. Time-of-follow-up was calculated from the date of the first initiation of ICIs to the date of the PD, ICIs discontinuation for the severe AE, or end of follow-up (May 1, 2022). Cox proportional hazards models were applied for the analyses of circulating apolipoprotein and lipids and the duration of ICIs and to compute the hazard ratios (HRs) with 95% CIs. Multivariate analyses were adjusted for potential confounders mentioned above. PFS was evaluated and compared using the Kaplan-Meier (KM) method and the log-rank test for trends. Receiver operating characteristic (ROC) curves were used to evaluate the discriminative ability of the markers by pROC package. The time-dependent ROC curves at 3-year of the Cox model were pictured by the timeROC R package (³²). The efficiency of models was assessed by comparing the decision curve analyses (DCA) curves of models (³³). All p values were two-sided, and confidence intervals (CIs) were set at the 95% level, with significance predefined to be at < 0.05.

3.1 Patients’ characteristics

The workflow of this study was showed in Figure 1. From January 2019 to December 2021, a total of 102 patients with NSCLC treated with combination of chemotherapy and ICIs in China-Japan Friendship Hospital were enrolled in this study. 3 patients didn't complete the first evaluation for the first administration of ICIs starting recently. Therefore, a total of 99 patients were enrolled for analysis. Patient baseline clinical characteristics are summarized in Table 1. Of these, there were 55 patients (55.6%) with LUAD, 37 patients (37.4%) with LUSC, the others were large cell carcinomas or mixed NSCLC. The median age was 64 years (58 - 69). 15 patients (15.2%) were female. 4 patients (4.04%) suffered from Stage I to IIIa disease and received ICIs as pre-operative adjuvant therapy. The 99 patients were all in good condition (the range of Eastern Cooperative Oncology Group Performance Status [ECOG-PS] was 0–1 score). 18 patients (18.2%) with LUAD had the recorded KRAS mutations. 58 patients (58.6%) were normal weight, 36 patients (36.4%) were overweight, and only 5 patients (5.05%) were obese. No patients were underweight. The overweight and obese groups were thereby combined as the BMI ≥ 25kg/m² group. No significant difference between the BMI < 25kg/m² and the BMI ≥ 25kg/m² group in the baseline circulating lipids, lipoproteins, and apolipoproteins (Table S1). The mean follow-up period for this study was 9.56 months. 75 patients which had been followed up for more than 6 months and had the primary endpoint (DCB or NDB). The best response to ICIs (non-active or active) of 89 patients was available. irAEs of 99 patients were recorded. Patient demographics and disease characteristics including age, gender, smoking history, and diabetes status were generally balanced between the NDB and DCB arms (Table S2), what’s more, between the non-active and active response arms (Table S3).

3.2 BMI ≥ 25 kg/m² predicts poor ICIs response in NSCLC

Comparations between groups divided according to the duration of responding to ICIs (NDB or DCB) and best response to ICIs (non-active or active) were performed. NDB patients were more likely to have higher BMI (T-test, 25.71 kg/m² vs 23.29 kg/m², p.val = 0.034, Figure 2A). The prevalence of BMI was higher in the cases with the non-active response, but the difference was not statistically signiﬁcant (T-test, 25.64 kg/m² vs 24.40 kg/m², p.val = 0.065, Figure 2B).

BMI ≥ 25 kg/m² was a negative predictor for DCB and active response to ICIs. The OR for NDB was 6.06 (95% CI 2.05 - 19.89, p.val = 0.002) for patients with BMI ≥ 25 kg/m² by the adjusted for the confounders including age, gender, diabetes, and smoking history. The OR for non-active response to ICIs was 4.37 (95% CI 1.64 - 12.43, p.val = 0.004) for patients with BMI ≥ 25 kg/m² with the confounders adjusted. The area under the ROC curve (AUC) was 0.72 (95% CI 0.60 - 0.85, Figure 2C) for the model used to predict DCB. AUC was 0.70 (95% CI 0.57 - 0.83, Figure 2D) for predicting active response to ICIs. The calibration curves showed that the multivariate models containing the BMI group tended to underestimate the probability of NDB and non-active response (Figure 2E-F).

Patients with normal weight tended to have a longer duration of ICIs response (Log-rank test, median PFS: 17.8 months vs 4.83 months, p.val = 0.001, Figure 2G). The HR for PFS of BMI ≥ 25 kg/m² without confounding factors adjusted was 2.54 (95% CI 1.41 - 4.60, p.val = 0.002), the HR was 3.08 (95% CI 1.63 - 5.82, p.val < 0.001) by multivariate Cox model, indicating that BMI ≥ 25 kg/m² was a negative predictor for the duration of ICIs response. The C-index of the multivariate Cox model was 0.68 (95% CI 0.61 - 0.76). The 3-year AUC was 0.76 and the calibration curve was shown in Figure 2H-I.

The subgroup analyses showed that BMI ≥ 25 kg/m² was an active indicator of DCB and active response to ICIs according to the univariate logistic regression in all subpopulations (Figure S1-2). What’s more, patients with BMI ≥ 25 kg/m² tended to have a shorter PFS of ICIs in all subgroups (Figure S3).

3.3 High levels of remnant cholesterol predict poor ICIs response in NSCLC

Compared with DCB patients, the prevalence of serum cholesterol, RC, and LDL-C were higher in NDB patients, but the difference was statistically signiﬁcant in RC only (T-test, 0.64 mmol/L vs 0.49 mmol/L, p.val = 0.047, Table S2). Serum cholesterol, triglyceride, phospholipid, and HDL-C had no difference between NDB and DCB patients. Circulating apolipoproteins including apoA1, apoE, apoC2, and apoC3 did not significantly differ between patients with NDB and DCB. Circulating lipids, lipoproteins, and apolipoproteins had no significant difference in the active and non-active groups base on the best response to ICIs in the total NSCLC cohort (Table S3).

Based on the PFS of ICIs, the total group was divided into RC high group and RC low group according to the serum RC levels. Similarly, the total cohort was also grouped by LDL-C level and apoB level. The cut points of RC and LDL-C were selected as 0.82 mmol/L and 2.58 mmol/L respectively. The analogous analyses of the independent and joint effects of apoB containing lipoproteins cholesterol (RC and LDL-C) and apoB levels and BMI were performed. Both continuous and categorical for of each marker were respectively brought into univariate and multivariate logistic regression model with the confounders including age, gender, diabetes, smoking history, and BMI adjusted.

The OR of RC for NDB and the non-active response was shown in Table 2, indicating that high serum RC potentially promoted ICIs resistance which led to NDB and non-active response to ICIs. The AUC of the multivariate model containing RC level (continuous) was 0.78 (95% CI 0.67 - 0.89, Figure 3A) for predicting DCB and 0.74 (95%CI 0.63 - 0.86, Figure 3B) for predicting active response to ICIs. The AUC of the model containing low RC was 0.79 (95%CI 0.68 - 0.91, Figure 3C) and 0.75 (95%CI 0.63 - 0.86, Figure 3D) respectively. The calibration curves showed that the multivariate models RC level and low RC all presented satisfied coherence in predicting NDB probability (Figure 3E-F). When it comes to predicting non-active response, the models tended to overestimate the rates when the incidence of non-active response was low and underestimate the rates when the incidence was high (Figure 3G-H).

The OR of low LDL-C for NDB calculated by univariate logist model was 0.31 (95% CI 0.10 - 0.89, p.val = 0.036), however the OR with the confounders including age, gender, diabetes, smoking history, and BMI adjusted was not statistically signiﬁcant (OR = 0.32, 95%CI 0.09 - 1.03, p.val = 0.065). Low serum LDL-C was a protective indicator for active ICIs response both without and with the confounders adjusted (OR = 0.29, 95%CI 0.09 - 0.81, p.val = 0.026; OR = 0.24, 95%CI 0.06 - 0.77, p.val = 0.023). The AUC was 0.76 (95%CI 0.66 - 0.86), but the calibration ability was unsatisfied (Figure S4A-B).

3.4 Low serum LDL-C predicts long PFS of ICIs in NSCLC

Patients with high RC and LDL-C were more like to respond to ICIs with shorter PFS (Log-rank test, median PFS: 14.53 months vs 3.23 months, p.val = 0.005; median PFS: 16.00 months vs 7.40 months, p.val = 0.010; Figure 4A-B). Considering BMI was a potentially important confounding factor for ICIs response and serum apoB containing lipoproteins level, the multivariate Cox regression of RC and LDL-C was performed (Table S4). Low LDL-C was a statistically significant protective factor for the long PFS with the BMI and other confounders adjusted (HR = 0.43, 95%CI 0.22 - 0.86, p.val = 0.016). The serum LDL-C level in continuous form was a negative indicator for ICIs PFS (HR = 1.50, 95%CI 1.01 - 2.24, p.val = 0.045). AUC of the multivariate model with LDL-C (continuous and category) at 3-year was 0.76 and 0.79 respectively (Figure 4C-D). The calibration curves were pictured in Figure 4E-F. The comparison of Cox models used to predict PFS was performed. Compared with only including the BMI category, the adding of serum LDL-C or LDL-C group was more efficient (Anova-test, p.val = 0.049, p.val < 2e-16). The DCA curves showed that the model with the LDL-C group was superior to the model with serum LDL-C in the continuous form and the model without LDL-C (Figure 4G). Serum RC and RC category did not add the extra value to the multivariate Cox model with the BMI category.

Subgroup analysis was performed according to the BMI group. Patients with low RC and LDL-C showed a trend to respond to ICIs with longer PFS in both groups but were statistically different in the BMI < 25 kg/m² group (Figure 5A-E). The multivariate Cox models of RC and LDL-C (continuous and category) were performed by adjusting confounders including gender, age, diabetes status, and smoking history. Low levels of LDL-C and RC were protective factors in the BMI < 25 kg/m² subgroup but not in the BMI ≥ 25 kg/m² group (Figure 5F).

3.5 ApoB containing lipoproteins predict better response to ICIs in LUAD cohort

In the LUAD cohort, the NDB group seemed to have higher serum cholesterol, LDL-C, and RC without a statistical difference (Figure S5A-C). Patients with non-active response to ICIs had higher serum LDL-C and CHO (T-test, 3.45 mmol/L vs 2.84 mmol/L, p.val = 0.010; 5.38 mmol/L vs 4.46 mmol/L, p.val = 0.017, Figure 6A-B). The prevalence of RC tended to be higher in the non-active group but was not statistically significant (Figure 6C). The univariate and the multivariate models including RC (both continuous and categorical), and LDL-C (both continuous and categorical) in the LUAD subgroup demonstrated that low RC and low LDL-C were protective factors for active ICIs response (Figure 6D). However, similar analyses were performed in the LUSC subgroup, serum apoB containing lipoproteins did not show satisfactory ability in predicting the ICIs response (Table S5).

Patients with low LDL-C and low RC tended to respond to ICIs with longer PFS (Figure 6E-F). The Kaplan-Meier curves of the LDL-C and RC groups were crossed in the LUSC cohort (Figure S5D-E). The univariate Cox analysis diagnosed the low LDL-C and low RC as protective factors for long PFS (Figure 6E-F), but only the HR of low LDL-C was statistically significant with the confounders adjusted (HR = 0.34, 95%CI 0.12 - 0.99, p.val= 0.047).

3.6 Correlation between serum lipids, lipoproteins and apolipoproteins and ICIs related adverse events (irAEs)

Until May 1st, 2022, 38/99 patients (38.4%) occurred irAEs, 9 patients stopped ICIs for irAEs. Patients were grouped according to the occurrence of irAEs. Difference of BMI between the two groups were not significant. The prevalence of circulating lipids, lipoproteins and apolipoproteins was not statistically different between patients with irAEs and those without irAEs. Patients who had suffered from irAEs tended to have higher apoA1/(apoA1+apoB) ratio (0.58 vs 0.54, p.val = 0.040). However, the apoA1 ratio was not an ideal biomarker for predicting irAEs for the diagnosis of the univariate and multivariate models containing apoA1 ratio was unsatisfied.

3.7 High expression of APOB potentially predicted poor ICIs response in the validation cohorts

Expression of APOB which encodes apoB proteins was tested for validation in GSE126044, and GSE135222 cohorts. 16 advanced NSCLC patients including 7 LUAD and 9 LUSC treated by Nivolumab were included in GSE126044. Expression of APOB tended to be higher in the NDB group in the LUAD group despite without statistical difference (Wilcox test, p.val = 0.191, Figure S6A). The 16 patients were divided into two groups according to the expression of APOB. LUAD patients with high expressed APOB tended to have shorter PFS (Log-rank test, 2.22 months vs 3.5 months, p.val = 0.018, Figure S6B). High APOB in LUAD tended to predict a short PFS of ICIs (HR = 2.86, 95%CI 0.82 - 10.02, p.val = 0.100). The tendency did not exist in the total cohort and the LUSC cohort (The total group: HR = 1.05, 95%CI 0.76 - 1.46, p.val = 0.77). 27 advanced NSCLC patients treated with anti-PD-1/PD-L1 were included in GSE135222. Expression of APOB was more likely higher in the NDB group (Wilcox-test, p.val = 0.356, Figure S6C). The low APOB group tended to have a longer PFS in the NSCLC patients without a statistical difference (Log-rank test, 2.95 months vs 1.47 months, p.val = 0.239, Figure S6D). Patients with higher APOB potentially indicated shorter duration of response to ICIs (HR = 1.08, 95%CI 0.99 - 1.18, p.val = 0.073).

3.8 High expression of APOB corresponded to suppressed TME in LUAD

391 LUAD patients without EGFR sensitive mutations and 494 LUSC patients were respectively classified into three clusters in the TCGA cohorts (Figure S7A-D). CYT represented the immune response of the tumor, which decreased from C2 to C3 (C2>C1>C3, Figure S7 B, D). C3 cluster was more likely to be an immunologically cold or non-inflamed tumor and was excluded since it potentially belongs to ICIs resistant subset. C2 cluster presented as an immunologically hot phenotype and was more likely to benefit from immunotherapies. C1 cluster couldn’t be classified in C2 or C3 cluster for the high immune heterogeneity which tended to be the most frequent immunophenotype in routine clinic conditions.

277 patients with LUAD samples and 479 LUSC patients in C1 and C2 clusters were divided into APOB^high and APOB^low groups according to the expression of APOB. According to the classification, APOB^low expressed more major histocompatibility complex class I (MHC-I) molecules (Figure 7A). The infiltration of T helper 2 cells (Th2) cells and effect memory CD4+ T cells (emCD4) and NK cells were also higher in the APOB^high group (Figure 7A). TME estimated by MCPcounter showed that the fibroblasts and monocytic lineage cells were highly infiltrated in APOB^high group, which is more likely to lead to immunosuppression by forming a barrier around the tumor cells (Figure 7B). Likewise, according to the cell component derived from the methylation data, despite the total immune cells showing no statistical difference, the fibroblasts were higher in the APOB^high group (Figure 7C).

When it comes to the LUSC cohort, APOB^high group tended to have higher infiltrated immune cells which potentially explain the reason why the difference of ICIs response in LUSC according to serum apoB containing proteins level was not consistent with LUAD (Figure S8).

Functional enrichment analysis between the APOB^high and APOB^low groups in the TCGA-LUAD cohort showed that apart from metabolic syndrome-associated pathways, many types of cancer-related pathways were upregulated in APOB^high group. TGF beta signaling pathway, Wnt signaling pathway, and adipocytokine signaling pathway were upregulated in APOB^high patients, which indicated the potential crosstalk of metabolism and cancer (Figure 8A). Differential gene analysis demonstrated that the apoB containing lipoproteins receptors including low-density lipoprotein receptor-related protein 5 (LPR5) and LPR6 which involve in the endocytosis of apoB containing lipoproteins and mediate the downstream Wnt signaling were upregulated in the non-inflamed TME and NDB patients. LDLR gene was not highly expressed in the NDB patients and patients with non-inflamed TME (Figure 8B-C).

This was the first study to evaluate the apoB containing lipoproteins and the efficacy of the combination regimens of ICIs and chemotherapy in NSCLC patients. In this study, we investigated the association between BMI, serum lipids, serum lipoproteins levels and the efficacy of ICIs in patients with NSCLC who received the combination of ICIs and chemotherapy for the first time regardless of the PD-L1 status. We demonstrated that BMI ≥ 25 kg/m² was a risk factor for the poor ICIs response and shorter PFS. Moreover, the elevated apoB containing lipoproteins including LDL-C and RC independently predicted the poor response to ICIs especially in BMI < 25 kg/m² and LUAD subgroup. Public data including GSE126044 and GSE135222 verified that patients with high APOB expression tended to have a worse ICIs response. The bioinformatic validation in TCGA cohorts suggested that patients with high expression of APOB manifested as suppressed TME in LUAD which potentially explained the reason of high levels of apoB containing lipoproteins led to poor ICIs response. Our study suggested the crosstalk between cholesterol metabolism and ICIs response and indicated the potential utility of the noninvasively measured biomarkers including LDL-C and RC for predicting the efficacy of ICIs and chemotherapy regimens.

With the lifestyle changes, the issues of excess weight and dyslipidemia worldwide are prominent. High BMI has been linked to a proinflammatory state, leading to a high risk of cancer, which lends further credence to the view that obesity may exert an immune-modulatory rather than a simply prognostic role (6, 7). A prior study suggested the positive effect of BMI on the treatment outcome in patients with advanced NSCLC treated with ICIs monotherapy. However, the results were not reproduced in patients treated with the combination of ICIs and chemotherapy (8).

In this study, we obtained the opposite results: high BMI is a risk indicator for the poor ICIs response in NSCLC patients who are treated with the combination regimen. Practically, the study about BMI and ICIs monotherapy conducted by Cortellini et al demonstrated that obesity was a positive indicator for the longer PFS and better objective response rate (ORR) in the ICIs cohort, while it was a conversely negative indicator in the chemotherapy cohort (6). The polarized results suggested that the obesity-associated immune regulation was radically distinct in the chemotherapy and immunotherapy, which might explain the converse result about the association of BMI and ICIs response obtained in the study. Additionally, the prevalence of the obesity in Asia is lower than that in American and European countries, only 5 patients with BMI ≥ 30 kg/m² were included in the study, most patients were overweight and normal weight, and a similar condition existed in another Japanese researcher (6-8). In most studies, the overweight group showed no significant difference compared with the normal weight. Another Japanese researcher chose BMI ≥ 21.4 kg/m² as the cut point and distinguished this cohort as a potential ICIs monotherapy benefited subgroup, which indicated a race-fitted BMI threshold might be necessary (³⁴). Despite the prevalence of BMI was not significantly different according to the irAEs in this study, past studies demonstrated that higher pre-treatment BMI elevated the occurrence of any grade of irAEs (9-10). Summarily, different from ICIs monotherapy, the overweight subgroup might not gain the benefit from the chemotherapy combining regimen in the Chinese NSCLC patients, whereas patients with a normal weight would be the lucky one.

Cholesterol is essential for maintaining the integrity and fluidity of membrane and other lipid microdomains named lipid rafts and is the precursor to support the synthesis of vitamin D, bile acid, and steroid hormones (³⁵). ApoB containing lipoproteins including LDL-C and remnant lipoproteins, such as small VLDL, IDL, and chylomicron remnants are critical carriers of cholesterol (³⁶). Unlike LDL-C, HDL-C is an apoA1 containing lipoproteins and acts as a cholesterol transporter to the major steroidogenic organs (17). It has been acknowledged that elevated apoB containing lipoproteins and decreased HDL-C are prevalent risk factors for atherosclerosis diseases (17).

Preclinical and clinical studies have shown that tumorigenesis and cancer progression occur accompanied by elevated cholesterol requirements manifesting as increased de novo biosynthesis, hydrolysis of cholesterol ester stores, and uptake from the circulation via LDL receptor (LDLR) mediated endocytosis. High cholesterol level was seen in the non-inflamed TME which manifested as CD8 +T cell exhaustion and upregulation of PD-1 expression on CD8 +T cells (³⁷). Increased LDL-C and decreased HDL-C have been reported to be associated with an increased risk of cancer development and poor prognoses in many types of cancer (17, ³⁸). Recent studies demonstrated that the high level of pre-treatment serum cholesterol, LDL-C, and HDL-C indicated better ICIs response and longer PFS in the NSCLC patients treated with ICIs monotherapy, whereas showed no association with the efficacy of cytotoxic chemotherapy (³⁹). However, the results remained controversial for the counteractive effects on anticancer immunity of HDL-C and LDL-C. Additionally, the association between serum cholesterol and the cholesterol-riched lipoproteins and ICIs treatment response in ICIs and chemotherapy combination is far from clear.

In our study, we demonstrated that the high cholesterol-riched apoB containing lipoproteins including LDL-C and RC could be promise indicators for poor ICIs response. Serum RC was superior in predicting the ICIs response (DCB/NDB) and the best response (active/non-active) with the satisfied discriminative ability and the well-fitting calibration curves, whereas serum LDL-C tended to work well in predicting the PFS of ICIs with the favorable diagnosis of DCA. Despite the groups being demographically balanced, the above analyses were all based on the multivariate regression with the confounders including gender, age, diabetes status and smoking history, and BMI category adjusted as dyslipidemia always occurs in patients with obesity and diabetes mellitus. In the subgroup analysis, we found that the predictive ability of apoB containing lipoproteins was more obvious in the LUAD cohort and BMI normal cohort.

We further validated our results in two external cohorts including GSE126044, and GSE135222. As the apoB was the main component of apoB containing lipoproteins, the expression of APOB gene which encodes apoB protein was used as the substitution of serum apoB containing lipoproteins. Despite the sample size of the two cohorts being both small, the tendency of negative effect on ICIs response to apoB expression was prominent. GSE126044 provided the histology details of the 16 samples, whereas the influence of APOB expression on ICIs response seems more obvious in the 7 LUAD samples.

To explore the mechanism of the negative apoB containing lipoproteins on ICIs response, we further performed the bioinformatic exploration in the TCGA cohort. As LUAD patients with EGFR mutations had been widely acknowledged as the ICIs resistant subgroup, we exclude these patients. The measurement of TME of the LUAD and LUSC cohorts was based on both expression and methylation data. Unlike LUAD, LUSC is characterized by a high overall mutation rate and marked genomic complexity, which tended to inflame tumors. In our study, 277/391 LUAD samples and 479/494 LUSC samples were not immunologically cold tumors, which confirmed the previous viewpoint (⁴⁰). The non-inflamed samples were excluded for their potential resistance to ICIs. The APOB high group had more fibroblasts infiltrated which potentially form the acquired stromal barrier to protect the effector cytotoxic immune cells from killing tumor cells. The immunosuppressive cells including Th2 cells and monocytic lineage cells and the decreased MHC I molecules of APOB high tumors might further lead to the immunologically cold TME which might be the potential mechanism of the group showing poor response to ICIs.

The further functional enrichment analysis in the TCGA-LUAD cohort showed that TGF-β signaling and Wnt signaling were upregulated in the APOB high group apart from participating in cancers and metabolism syndrome signalings. TME plays dynamic and sophisticated roles in facilitating tumor growth and drug responses. The elevated cholesterol metabolism reprogrammed immunosuppressive cells such as tumor-associated macrophages and fibroblasts. The cholesterol contained in apoB lipoproteins is delivered into cancer cells by binding to the LDL receptor (LDLR) family. Low-density lipoprotein-related protein 6 (LRP6) and its close homolog LRP5 belong to the LDLR family and are known as important regulatory factors of serum cholesterol by acting as the co-receptors of Wnt pathway (⁴¹, ⁴²). This crosstalk provides the potential possibility to form the non-inflamed TME by high LDL-C cholesterol since the Wnt pathway is the vital upstream of TGF-β signaling which promotes the transformation of fibroblasts and suppresses immune responses.

The mevalonate pathway is a well-known process in cholesterol biosynthesis. 3-hydroxy-3-methylglutaryl-CoA (HMG-CoA) is reduced by its reductase called HMG-CoA (HMGCR) to form mevalonate. HMGCR is the primary rate-limiting enzyme in cholesterol biosynthesis. Statins mainly target HMGCR to exert a cholesterol-lowering effect and decrease the serum apoB containing lipoproteins level to inhibit the cancer cells’ cholesterol acquirement. Past studies suggested that the concomitant use of statins during ICI treatment showed a positive impact on treatment outcomes. Lipophilic statins enable better antigen presentation and tumor-suppressive responses by involving in endosomal trafficking. More recent studies demonstrated that statins could inhibit the recruitment of pro-tumorigenic macrophages by suppression of CC chemokine receptor 1 (CCR1) ligand secretion. However, the statins exposure could elevate the risk of ICIs-associated myositis by evoking the anti-HMGCR antibodies (⁴³). The experimental findings proposed that the prolonged atorvastatin treatment led to drug resistance and progression of lung adenomas into the invasive disease by the acquisition of additional stromal changes and non-inflamed TME. Therefore, despite statins in combination with chemotherapy and ICs regimens can be a promised anti-tumor strategy, if the combination will be used routinely remains controversial.

Our study acknowledges several limitations beyond the single-center study and the consequent central effect bias such as the baseline characteristics and the concomitant anti-cancer medication habits. The sample size in the one center is small for which the effect of apoB containing lipoproteins on many potentially meaningful subgroups was not observed. In our cohort, the BMI of all patients was more than 18.5 kg/m², and the apoB containing lipoproteins effect on the underweight patients could not be evaluated. In addition, as a likely result of the clinicians’ attitude to reserve chemotherapy for fitter patients, the ECOG-PS of patients in our cohort were all within 0 to 1. Moreover, only 5 patients were obese and therefore were combined with the overweight group. More evaluations on the underweight and the obese groups were needed.

Secondly, apart from the selection bias such as the recall bias and investigation bias which could be attenuated by choosing the objective indicators and medical records, the retrospective design also caused the event effect bias. As the ICIs response could be obtained at least the 6-month observation and some patents stopped ICIs for irAEs or other reasons, we added the best response to ICIs as the supplement endpoint. Patients experiencing long duration of ICIs treatment and showing active response to ICIs were more likely to suffer from the recorded irAEs. Patients without irAEs might have two conditions: on the one hand, patients did not occur the irAEs, on the other hand, the patients did not reach the occurrence time of irAEs.

Thirdly, the concomitant medications of statins and metformin were not brought into research despite their roles in controlling dyslipidemia, as the main points of the study were to observe the effects of apoB containing lipoproteins and other serum lipids and lipoproteins on ICIs response no matter how to control and maintain. Additionally, the APOB expression was used as the substitution of apoB containing lipoproteins in the validation cohorts as the serum lipids and lipoproteins data were unavailable and apoB was the major apolipoproteins in the transport of circulating cholesterol. We speculated that the high cholesterol-riched apoB containing lipoproteins mediated the poor ICIs response by promoting the enrichment of fibroblasts and other immunosuppressive cells through LRP6/5-Wnt-TGF-β axis. The potential mechanism required more laboratory experiments to verify.

In the NSCLC population treated with the combination of chemotherapy and ICIs regimen, BMI ≥ 25 kg/m2, the elevated cholesterol-riched apoB containing lipoproteins including LDL-C and RC are independent risk indicators for the poor ICIs response. The elevated apoB containing lipoproteins might promote the infiltration of fibroblasts and other immunosuppressive cells through LRP6/5-Wnt-TGF-β axis to acquire resistance to ICIs. BMI and apoB containing lipoproteins are promise biomarkers in predicting the response to the chemotherapy plus ICIs regimen, however further large-scale validations are needed. The mechanisms beyond the apoB containing lipoproteins regulating the ICIs response need experimental exploration.

NSCLC: Non-small cell lung cancer; LUAD: lung adenocarcinoma; LUSC: Lung squamous cell carcinoma; ICIs: Immune checkpoint inhibitors; TME: Tumor microenvironment; PD-L1: Programmed death-ligand 1; CYT: Cytolytic activity; BMI: body mass index; CHO: cholesterol; TG: ApoB: apolipoproteins B; LDL-C: low-density lipoproteins cholesterol; HDL-C: High-density lipoprotein cholesterol; RC: Remnant cholesterol; OS: Overall survival time; PFS: Progression-free survival; DCB: Durable clinical benefit; NDB: Non-durable benefit; PR: Partial response; SD: Stable disease; PD: Progressive disease; DCA: Decision curve analyses; HR: Hazard ratios; OR: Odds ratio; ROC: Receiver operating characteristic; AUC: The area under the ROC curve; TCGA: The Cancer Genome Atlas; GEO: Gene Expression Omnibus; GSEA: Gene set enrichment analysis; DEGs: Differentially expressed genes;

Conflict of Interest

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Author Contributions

ZX-H and HJ-C contributed to the idea and design. YM-Z, YX-Y, XY-L and J-L contributed to the data acquisition. JB-Z, HJ-C evaluated the radiographic data independently. ZX-H, JQ-L, Y-W, J-L contributed to the statistical analyses. ZX-H, JB-Z and HJ-C participated in the manuscript drafting and revision. All authors contributed to the publication according to the ICMJE guidelines for the authorship. All authors read and approved the manuscript and agree to be accountable for all aspects of the research in ensuring that the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Funding

This work was supported by the Capital Health Development Scientific Research Project: Combination of Gegen Qinlian tablets with immune checkpoint inhibitors and chemotherapy in advanced lung squamous cell carcinoma: a clinical study (2022-2-4065).

Acknowledgments

The first author values the period of time studying with Y-W and JQ-L at the department of cardiology. We would like to acknowledge patients and their families. Publicly available datasets were analyzed in this study. TCGA-LUSC transcriptome and DNA methylation data were downloaded from https://portal.gdc.cancer.gov/. GSE126044 and GSE135222 could be respectively accessed via https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE126044 and https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE135222 .

Ethical approval statement

The study was carried out with the ethical approval of the institutional Ethics Committee of the Faculty of Medicine at China-Japan Friendship Hospital approval (2022-KY-051). Patient consents of ICIs and chemotherapy combination treatment and medication information collection were acquired.

ORCID

Zixin Hu https://orcid.org/0000-0002-2013-9451

Yixuan Yu https://orcid.org/0000-0003-1950-1203

Huijuan Cui https://orcid.org/0000-0003-2900-0461

Data Availability Statement

The datasets used during the present study are available from the corresponding authors upon reasonable request.

Tang S, Qin C, Hu H, et al. Immune Checkpoint Inhibitors in Non-Small Cell Lung Cancer: Progress, Challenges, and Prospects. Cells. 2022;11(3):320. Published 2022 Jan 19. doi:10.3390/cells11030320
Wang M, Liang H, Wang W, et al. Immune-related adverse events of a PD-L1 inhibitor plus chemotherapy versus a PD-L1 inhibitor alone in first-line treatment for advanced non-small cell lung cancer: A meta-analysis of randomized control trials. Cancer. 2021;127(5):777-786. doi:10.1002/cncr.33270
Snaebjornsson MT, Janaki-Raman S, Schulze A. Greasing the Wheels of the Cancer Machine: The Role of Lipid Metabolism in Cancer. Cell Metab. 2020;31(1):62-76. doi:10.1016/j.cmet.2019.11.010
Broadfield LA, Pane AA, Talebi A, Swinnen JV, Fendt SM. Lipid metabolism in cancer: New perspectives and emerging mechanisms. Dev Cell. 2021;56(10):1363-1393. doi:10.1016/j.devcel.2021.04.013
Cao Y. Adipocyte and lipid metabolism in cancer drug resistance. J Clin Invest. 2019;129(8):3006-3017. Published 2019 Jul 2. doi:10.1172/JCI127201
Cortellini A, Ricciuti B, Tiseo M, et al. Baseline BMI and BMI variation during first line pembrolizumab in NSCLC patients with a PD-L1 expression ≥ 50%: a multicenter study with external validation. J Immunother Cancer. 2020;8(2):e001403. doi:10.1136/jitc-2020-001403
Cortellini A, Ricciuti B, Vaz VR, et al. Prognostic effect of body mass index in patients with advanced NSCLC treated with chemoimmunotherapy combinations. J Immunother Cancer. 2022;10(2):e004374. doi:10.1136/jitc-2021-004374
Tateishi A, Horinouchi H, Yoshida T, et al. Correlation between body mass index and efficacy of anti-PD-1 inhibitor in patients with non-small cell lung cancer. Respir Investig. 2022;60(2):234-240. doi:10.1016/j.resinv.2021.11.003
Zhang D, Shah NJ, Cook M, et al. Association between Body Mass Index and Immune-Related Adverse Events (irAEs) among Advanced-Stage Cancer Patients Receiving Immune Checkpoint Inhibitors: A Pan-Cancer Analysis. Cancers (Basel). 2021;13(23):6109. Published 2021 Dec 3. doi:10.3390/cancers13236109
Cortellini A, Bersanelli M, Santini D, et al. Another side of the association between body mass index (BMI) and clinical outcomes of cancer patients receiving programmed cell death protein-1 (PD-1)/ Programmed cell death-ligand 1 (PD-L1) checkpoint inhibitors: A multicentre analysis of immune-related adverse events. Eur J Cancer. 2020;128:17-26. doi:10.1016/j.ejca.2019.12.031
Burnett JR, Hooper AJ, Hegele RA. Remnant Cholesterol and Atherosclerotic Cardiovascular Disease Risk. J Am Coll Cardiol. 2020;76(23):2736-2739. doi:10.1016/j.jacc.2020.10.029
Borén J, Williams KJ. The central role of arterial retention of cholesterol-rich apolipoprotein-B-containing lipoproteins in the pathogenesis of atherosclerosis: a triumph of simplicity. Curr Opin Lipidol. 2016;27(5):473-483. doi:10.1097/MOL.0000000000000330
Zhang Y, Chen H, Chen S, Li Z, Chen J, Li W. The effect of concomitant use of statins, NSAIDs, low-dose aspirin, metformin and beta-blockers on outcomes in patients receiving immune checkpoint inhibitors: a systematic review and meta-analysis. Oncoimmunology. 2021;10(1):1957605. Published 2021 Aug 2. doi:10.1080/2162402X.2021.1957605
Borgquist S, Butt T, Almgren P, et al. Apolipoproteins, lipids and risk of cancer. Int J Cancer. 2016;138(11):2648-2656. doi:10.1002/ijc.30013
Maran L, Hamid A, Hamid SBS. Lipoproteins as Markers for Monitoring Cancer Progression. J Lipids. 2021;2021:8180424. doi:10.1155/2021/8180424
Karayama M, Inui N, Inoue Y, et al. Increased serum cholesterol and long-chain fatty acid levels are associated with the efficacy of nivolumab in patients with non-small cell lung cancer. Cancer Immunol Immunother. 2022;71(1):203-217. doi:10.1007/s00262-021-02979-4
Kucharska-Newton AM, Rosamond WD, Schroeder JC, et al. HDL-cholesterol and the incidence of lung cancer in the Atherosclerosis Risk in Communities (ARIC) study. Lung Cancer. 2008;61(3):292-300. doi:10.1016/j.lungcan.2008.01.015
Goldstraw P, Crowley J, Chansky K, et al. The IASLC Lung Cancer Staging Project: proposals for the revision of the TNM stage groupings in the forthcoming (seventh) edition of the TNM Classification of malignant tumours. J Thorac Oncol. 2007;2(8):706-714. doi:10.1097/JTO.0b013e31812f3c1a
Eisenhauer EA, Therasse P, Bogaerts J, et al. New response evaluation criteria in solid tumours: revised RECIST guideline (version 1.1). Eur J Cancer. 2009;45(2):228-247. doi:10.1016/j.ejca.2008.10.026
Rizvi H, Sanchez-Vega F, La K, et al. Molecular Determinants of Response to Anti-Programmed Cell Death (PD)-1 and Anti-Programmed Death-Ligand 1 (PD-L1) Blockade in Patients With Non-Small-Cell Lung Cancer Profiled With Targeted Next-Generation Sequencing. J Clin Oncol. 2018;36(7):633-641. doi:10.1200/JCO.2017.75.3384
Cho JW, Hong MH, Ha SJ, et al. Genome-wide identification of differentially methylated promoters and enhancers associated with response to anti-PD-1 therapy in non-small cell lung cancer. Exp Mol Med. 2020;52(9):1550-1563. doi:10.1038/s12276-020-00493-8
Kim JY, Choi JK, Jung H. Genome-wide methylation patterns predict clinical benefit of immunotherapy in lung cancer. Clin Epigenetics. 2020;12(1):119. Published 2020 Aug 6. doi:10.1186/s13148-020-00907-4
Aryee MJ, Jaffe AE, Corrada-Bravo H, et al. Minfi: a flexible and comprehensive Bioconductor package for the analysis of Infinium DNA methylation microarrays. Bioinformatics. 2014;30(10):1363-1369. doi:10.1093/bioinformatics/btu049
Tian, Y., Morris, T. J., Webster, A. P., Yang, Z., Beck, S., Feber, A., & Teschendorff, A. E. (2017). ChAMP: updated methylation analysis pipeline for Illumina BeadChips. Bioinformatics (Oxford, England), 33(24), 3982–3984. https://doi.org/10.1093/bioinformatics/btx513
Hu Z, Xue C, Zheng J, et al. Hyper-Methylated Hub Genes of T-Cell Receptor Signaling Predict a Poor Clinical Outcome in Lung Adenocarcinoma. J Oncol. 2022;2022:5426887. doi:10.1155/2022/5426887
Charoentong, P., Finotello, F., Angelova, M., Mayer, C., Efremova, M., Rieder, D., Hackl, H., & Trajanoski, Z. (2017). Pan-cancer Immunogenomic Analyses Reveal Genotype-Immunophenotype Relationships and Predictors of Response to Checkpoint Blockade. Cell reports, 18(1), 248–262. https://doi.org/10.1016/j.celrep.2016.12.019
Rooney, M. S., Shukla, S. A., Wu, C. J., Getz, G., & Hacohen, N. (2015). Molecular and genetic properties of tumors associated with local immune cytolytic activity. Cell, 160(1-2), 48–61. https://doi.org/10.1016/j.cell.2014.12.033
Becht E, Giraldo NA, Lacroix L, et al. Estimating the population abundance of tissue-infiltrating immune and stromal cell populations using gene expression [published correction appears in Genome Biol. 2016 Dec 1;17 (1):249]. Genome Biol. 2016;17(1):218. Published 2016 Oct 20. doi:10.1186/s13059-016-1070-5
Zheng SC, Breeze CE, Beck S, Teschendorff AE. Identification of differentially methylated cell types in epigenome-wide association studies. Nat Methods. 2018;15(12):1059-1066. doi:10.1038/s41592-018-0213-x
Hänzelmann S, Castelo R, Guinney J. GSVA: gene set variation analysis for microarray and RNA-seq data. BMC Bioinformatics. 2013;14:7. Published 2013 Jan 16. doi:10.1186/1471-2105-14-7
Love MI, Huber W, Anders S. Moderated estimation of fold change and dispersion for RNA-seq data with DESeq2. Genome Biol. 2014;15(12):550. doi:10.1186/s13059-014-0550-8
Blanche P, Dartigues JF, Jacqmin-Gadda H. Estimating and comparing time-dependent areas under receiver operating characteristic curves for censored event times with competing risks. Stat Med. 2013;32(30):5381-5397. doi:10.1002/sim.5958
Vickers AJ, Elkin EB. Decision curve analysis: a novel method for evaluating prediction models. Med Decis Making. 2006;26(6):565-574. doi:10.1177/0272989X06295361
Imai H, Kishikawa T, Minemura H, et al. Pretreatment Glasgow prognostic score predicts survival among patients with high PD-L1 expression administered first-line pembrolizumab monotherapy for non-small cell lung cancer. Cancer Med. 2021;10(20):6971-6984. doi:10.1002/cam4.4220
Guillaumond F, Bidaut G, Ouaissi M, et al. Cholesterol uptake disruption, in association with chemotherapy, is a promising combined metabolic therapy for pancreatic adenocarcinoma. Proc Natl Acad Sci U S A. 2015;112(8):2473-2478. doi:10.1073/pnas.1421601112
Deng CF, Zhu N, Zhao TJ, et al. Involvement of LDL and ox-LDL in Cancer Development and Its Therapeutical Potential. Front Oncol. 2022;12:803473. Published 2022 Feb 16. doi:10.3389/fonc.2022.803473
Huang B, Song BL, Xu C. Cholesterol metabolism in cancer: mechanisms and therapeutic opportunities. Nat Metab. 2020;2(2):132-141. doi:10.1038/s42255-020-0174-0
Loosen SH, Kostev K, Luedde M, Luedde T, Roderburg C. Low blood levels of high-density lipoprotein (HDL) cholesterol are positively associated with cancer [published online ahead of print, 2021 Dec 12]. J Cancer Res Clin Oncol. 2021;10.1007/s00432-021-03867-1. doi:10.1007/s00432-021-03867-1
Hofbauer J, Hauck A, Matos C, et al. Immunometabolic Markers in a Small Patient Cohort Undergoing Immunotherapy. Biomolecules. 2022;12(5):716. Published 2022 May 18. doi:10.3390/biom12050716
Heist RS, Sequist LV, Engelman JA. Genetic changes in squamous cell lung cancer: a review. J Thorac Oncol. 2012;7(5):924-933. doi:10.1097/JTO.0b013e31824cc334
Abou Ziki MD, Mani A. The interplay of canonical and noncanonical Wnt signaling in metabolic syndrome. Nutr Res. 2019;70:18-25. doi:10.1016/j.nutres.2018.06.009
Go GW, Mani A. Low-density lipoprotein receptor (LDLR) family orchestrates cholesterol homeostasis. Yale J Biol Med. 2012;85(1):19-28.
von Itzstein MS, Khan S, Popat V, et al. Statin Intolerance, Anti-HMGCR Antibodies, and Immune Checkpoint Inhibitor-Associated Myositis: A "Two-Hit" Autoimmune Toxicity or Clinical Predisposition?. Oncologist. 2020;25(8):e1242-e1245. doi:10.1634/theoncologist.2019-0911

Table1: Baseline clinical characteristics of the 99 NSCLC patients administrated by the combination of chemotherapy and ICIs regimens.

*Variable*	*Mean (SD)/Median (95% CI)/Frequency*	*Num*	*Variable*	*Mean (SD)/Median (95% CI)/Frequency*	*Num*
Age	64.0 ( 58.0 - 69.0)	99	ICIs		99
Gender		99	Anti PD1	65 (65.7%)
Female	15 (15.2%)		Anti PDL1	31 (31.3%)
Male	84 (84.8%)		Multi	3 (3.03%)
Stage		99	Line		99
I - II	3 (3.03%)		1	68 (68.7%)
III	24 (24.2%)		2+	27 (27.3%)
IV	72 (72.7%)		PreOC	4 (4.04%)
Smoking History:		97	Best response		89
N	24 (24.7%)		Active	62 (69.7%)
Y	73 (75.3%)		Non active	27 (30.3%)
DM		97	Responsiveness		75
N	79 (81.4%)		DCB	44 (58.7%)
Y	18 (18.6%)		NDB	31 (41.3%)
Histology		99	AE		99
LUAD	55 (55.6%)		N	61 (61.6%)
LUSC	37 (37.4%)		Y	38 (38.4%)
NSCLC	6 (6.06%)		Serum cholestrol (CHO)	4.57 (1.25)	93
NSCLC+SCLC	1 (1.01%)		Total triglyceride (TG)	1.59 (0.84)	93
BMI		99	Low density lipoprotein cholesterol (LDL-C)	2.93 (0.87)	93
normal	58 (58.6%)		High density lipoprotein cholesterol (HDL-C)	1.09 (0.27)	93
overweight	36 (36.4%)		Remnant cholesterol (RC)	0.54 (0.29)	93
obesity	5 (5.05%)		Apolipoprotein B (APOB) (ApoB)	0.96 (0.28)	73
KRAS	18 (100%)	18	ApoA1	1.16 (0.26)	73
TPS_22C3:		27	ApoE	3.50 (1.33)	65
< 1%	4 (14.8%)		ApoC3	8.62 (4.51)	50
1% - 49%	13 (48.1%)		ApoC2	3.82 (2.19)	50
≥ 50%	10 (37.0%)

Table2: Models containing remnant cholesterol (RC) to predict ICIs response (DCB/NDB) and the best response to ICIs (active/non-active). Model 1: the multivariate logistic model containing RC as the continuous form used to predict the ICIs response and the best response. Model 2: the multivariate logistic model containing RC as the category form used to predict the ICIs response and the best response. Model 3: the univariate logistic model containing RC as the continuous form used to predict the ICIs response and the best response. Model 4: the univariate logistic model containing RC as the category form used to predict the ICIs response and the best response.

Variable	*ICIs response*			*The best response to ICIs*
	OR	95%CI	P.val	OR	95%CI	P.val
*Model1*
RC	10.46	1.36 - 103.47	0.031	5.25	0.79 - 41.31	0.100
BMI ≥ 25 kg/m2	6.61	2.08 - 23.92	0.002	5.28	1.86 - 16.35	0.002
*Model2*
Low RC	0.12	0.02 - 0.63	0.017	0.22	0.05 - 0.96	0.047
BMI ≥ 25 kg/m2	6.91	2.12 - 25.85	0.002	5.31	1.85 - 16.76	0.003
*Model 3*
RC	7.34	1.25 - 53.15	0.035	4.69	0.90 - 26.88	0.071
*Model 4*
Low RC	0.19	0.04 - 0.73	0.022	0.29	0.08 - 0.87	0.043

No competing interests reported.

SupplementaryFiguresandTables.zip
Supplementary MaterialTable S1: Comparation of the baseline chacteristics, serum lipids, lipoproteins and apolipoproteins between the BMI < 25kg/m² group and the BMI ≥ 25kg/m² group Table S2: Comparation of the baseline chacteristics, serum lipids, lipoproteins and apolipoproteins according to the ICIs response (DCB/NDB). DCB = durable clinical benefit, NDB = non-durable benefit.Table S3: Comparation of the baseline chacteristics, serum lipids, lipoproteins and apolipoproteins according to the best response to ICIs (active/non-active). Table S4: Hazard ratios (HRs) of apoB containing lipoproteins (RC and LDL-C) in predicting the duration of ICIs response in NSCLC. Table S5: The prognostic ability of apoB containing lipoproteins (RC and LDL-C) in predicting the duration of ICIs response in the LUSC cohort. The multivariate models were used to adjust the confounders including age, gender, smoking history, diabetes status and BMI. Figure S1: Subgroup analysis of the BMI ≥ 25kg/m² in predicting the ICIs response (DCB/NDB).Figure S2: Subgroup analysis of the BMI ≥ 25kg/m² in predicting the best response to ICIs (active/non-active).Figure S3: Subgroup analysis of the BMI ≥ 25kg/m² in predicting the PFS of ICIs.Figure S4: Diagnosis of multivariate models containing low low-density lipoproteins cholesterol (LDL-C) used to predic the best ICIs response by the confounders including age, gender, smoking history, diabetes and BMI adjusted. (A) The receiver operating characteristic (ROC) curve of the multivariate model. (B) The calibration curve of the model. Figure S5: (A) Comparation of serum cholesterol between the durable clinical benefit (DCB) and non-durable benefit (NDB) groups. (B) Comparation of serum low-density lipoproteins cholesterol (LDL-C) between the DCB group and the NDB group. (C) Comparation of serum remnant cholesterol (RC) between the DCB group and the NDB group. (D) The Kaplan-Meier (KM) curves showed LDL-C did not distinct the PFS in the LUSC group (n = 37). (E) The KM analysis showed that low RC did not distinct PFS in the LUSC group.Figure S6: Validation of apoB value in the ICIs treated NSCLC patients. (A) Comparison of APOB gene expression between patients with durable clinical benefit (DCB) and non-durable benefit (NDB) in the LUAD group in GSE126044. (B) The Kaplan-Meier (KM) curves showed patients with low APOB expression tended to have longer PFS in the LUAD group. (C) Comparison of APOB gene expression between patients with durable clinical benefit (DCB) and non-durable benefit (NDB) in the NSCLC group in GSE135222. (D) The KM curves showed that NSCLC patients with low APOB expression tended to have longer PFS in GSE135222.Figure S7: Classification of immunophenotypes in the TCGA-LUAD cohort and TCGA-LUSC chort. (A) Cluster heatmap of TCGA-LUAD cohort. (B) Comparation of CYT among immune related clusters in the TCGA-LUAD cohort. (C) Cluster heatmap of TCGA-LUSC cohort. (D) Comparation of CYT among immune related clusters in the TCGA-LUSC cohort. Figure S8: Heatmap of tumor microenvironment of TCGA-LUSC cohort according to APOB expression. The APOB high group in the LUSC group tended to have higher all types of infiltrated immune cells.

Serum cholesterol-riched apolipoprotein-B (apoB) containing lipoproteins predict the response to immune checkpoints inhibitors (ICIs) in non-small cell lung cancer (NSCLC) : A retrospective analysis

Status:

Version 2

Abstract

Figures

1 Introduction

2 Materials And Methods

2.1 Patient cohorts:

2.2 Collection of clinical data

2.3 Bioinformatic validation

2.4 Statistical analysis

3 Results

3.1 Patients’ characteristics

3.2 BMI ≥ 25 kg/m² predicts poor ICIs response in NSCLC

3.3 High levels of remnant cholesterol predict poor ICIs response in NSCLC

3.4 Low serum LDL-C predicts long PFS of ICIs in NSCLC

3.5 ApoB containing lipoproteins predict better response to ICIs in LUAD cohort

3.6 Correlation between serum lipids, lipoproteins and apolipoproteins and ICIs related adverse events (irAEs)

3.7 High expression of APOB potentially predicted poor ICIs response in the validation cohorts

3.8 High expression of APOB corresponded to suppressed TME in LUAD

4 Discussion

5 Conclusion

Abbreviations

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 2

Serum cholesterol-riched apolipoprotein-B (apoB) containing lipoproteins predict the response to immune checkpoints inhibitors (ICIs) in non-small cell lung cancer (NSCLC) : A retrospective analysis

Status:

Version 2

Abstract

Figures

1 Introduction

2 Materials And Methods

2.1 Patient cohorts:

2.2 Collection of clinical data

2.3 Bioinformatic validation

2.4 Statistical analysis

3 Results

3.1 Patients’ characteristics

3.2 BMI ≥ 25 kg/m2 predicts poor ICIs response in NSCLC

3.3 High levels of remnant cholesterol predict poor ICIs response in NSCLC

3.4 Low serum LDL-C predicts long PFS of ICIs in NSCLC

3.5 ApoB containing lipoproteins predict better response to ICIs in LUAD cohort

3.6 Correlation between serum lipids, lipoproteins and apolipoproteins and ICIs related adverse events (irAEs)

3.7 High expression of APOB potentially predicted poor ICIs response in the validation cohorts

3.8 High expression of APOB corresponded to suppressed TME in LUAD

4 Discussion

5 Conclusion

Abbreviations

Declarations

References

Tables

Additional Declarations

Supplementary Files

Status:

Version 2

3.2 BMI ≥ 25 kg/m² predicts poor ICIs response in NSCLC