Longitudinal plasma proteomic analysis identifies biomarkers and combinational targets for anti-PD1-resistant cancer patients.

The response rate of anti-PD1 therapy is limited, and the influence of anti-PD1 therapy on cancer patients is unclear. To address these challenges, we conducted a longitudinal analysis of plasma proteomic changes with anti-PD1 therapy in non-small cell lung cancer (NSCLC), alveolar soft part sarcoma (ASPS), and lymphoma patients. We included 339 plasma samples before and after anti-PD1 therapy from 193 patients with NSCLC, ASPS, or lymphoma. The plasma proteins were detected using data-independent acquisition-mass spectrometry and customable antibody microarrays. Differential proteomic characteristics in responders (R) and non-responders (NR) before and after anti-PD1 therapy were elucidated. A total of 1019 proteins were detected using our in-depth proteomics platform and distributed across 10-12 orders of abundance. By comparing the differential plasma proteome expression between R and NR groups, 50, 206, and 268 proteins were identified in NSCLC, ASPS, and lymphoma patients, respectively. Th17, IL-17, and JAK-STAT signal pathways were identified upregulated in NR group, while cellular senescence and transcriptional misregulation pathways were activated in R group. Longitudinal proteomics analysis revealed the IL-17 signaling pathway was downregulated after treatment. Consistently, many proteins were identified as potential combinatorial therapeutic targets (e.g., IL-17A and CD22). Five noninvasive biomarkers (FLT4, SFTPB, GNPTG, F5, and IL-17A) were further validated in an independent lymphoma cohort (n = 39), and another three noninvasive biomarkers (KIT, CCL3, and TNFSF1) were validated in NSCLC cohort (n = 76). Our results provide molecular insights into the anti-PD1 therapy in cancer patients and identify new therapeutic strategies for anti-PD1-resistant patients.

Mutational variant allele frequency profile as a biomarker of response to immune checkpoint blockade in non-small cell lung Cancer.

INTRODUCTION: Identifying new biomarkers for predicting immune checkpoint inhibitors (ICIs) response in non-small cell lung cancer (NSCLC) is crucial. We aimed to assess the variant allele frequency (VAF)-related profile as a novel biomarker for NSCLC personalized therapy. METHODS: We utilized genomic data of 915 NSCLC patients via cBioPortal and a local cohort of 23 patients for model construction and mutational analysis. Genomic, transcriptomic data from 952 TCGA NSCLC patients, and immunofluorescence (IF) assessment with the local cohort supported mechanism analysis. RESULTS: Utilizing the random forest algorithm, a 15-gene VAF-related model was established, differentiating patients with durable clinical benefit (DCB) from no durable benefit (NDB). The model demonstrated robust performance, with ROC-AUC values of 0.905, 0.737, and 0.711 across training (n = 313), internal validation (n = 133), and external validation (n = 157) cohorts. Stratification by the model into high- and low-score groups correlated significantly with both progression-free survival (PFS) (training: P < 0.0001, internal validation: P < 0.0001, external validation: P = 0.0066) and overall survival (OS) (n = 341) (P < 0.0001). Notably, the stratification system was independent of PD-L1 (P < 0.0001) and TMB (P < 0.0001). High-score patients exhibited an increased DCB ratio and longer PFS across both PD-L1 and TMB subgroups. Additionally, the high-score group appeared influenced by tobacco exposure, with activated DNA damage response pathways. Whereas, immune/inflammation-related pathways were enriched in the low-score group. Tumor immune microenvironment analyses revealed higher proportions of exhausted/effector memory CD8 + T cells in the high-score group. CONCLUSIONS: The mutational VAF profile is a promising biomarker for ICI therapy in NSCLC, with enhanced therapeutic stratification and management as a supplement to PD-L1 or TMB.

Longitudinal plasma proteomic profiling of EML4-ALK positive lung cancer receiving ALK-TKIs therapy.

BACKGROUND: Anaplastic lymphoma kinase-tyrosine kinase inhibitors (ALK-TKIs) has demonstrated remarkable therapeutic effects in ALK-positive non-small cell lung cancer (NSCLC) patients. Identifying prognostic biomarkers can enhance the clinical efficacy of relapsed or refractory patients. METHODS: We profiled 737 plasma proteins from 159 pre-treatment and on-treatment plasma samples of 63 ALK-positive NSCLC patients using data-independent acquisition-mass spectrometry (DIA-MS). The consensus clustering algorithm was used to identify subtypes with distinct biological features. A plasma-based prognostic model was constructed using the LASSO-Cox method. We performed the Mfuzz analysis to classify the patterns of longitudinal changes in plasma proteins during treatment. 52 baseline plasma samples from another independent ALK-TKI treatment cohort were collected to validate the potential prognostic markers using ELISA. RESULTS: We identified three subtypes of ALK-positive NSCLC with distinct biological features and clinical efficacy. Patients in subgroup 1 exhibited activated humoral immunity and inflammatory responses, increased expression of positive acute-phase response proteins, and the worst prognosis. Then we constructed and verified a prognostic model that predicts the efficacy of ALK-TKI therapy using the expression levels of five plasma proteins (SERPINA4, ATRN, APOA4, TF, and MYOC) at baseline. Next, we explored the longitudinal changes in plasma protein expression during treatment and identified four distinct change patterns (Clusters 1-4). The longitudinal changes of acute-phase proteins during treatment can reflect the treatment status and tumor progression of patients. Finally, we validated the prognostic efficacy of baseline plasma CRP, SAA1, AHSG, SERPINA4, and TF in another independent NSCLC cohort undergoing ALK-TKI treatment. CONCLUSIONS: This study contributes to the search for prognostic and drug-resistance biomarkers in plasma samples for ALK-TKI therapy and provides new insights into the mechanism of drug resistance and the selection of follow-up treatment.

Development and validation of a deep learning-based model to predict response and survival of T790M mutant non-small cell lung cancer patients in early clinical phase trials using electronic medical record and pharmacokinetic data.

Background: Epidermal growth factor receptor (EGFR) T790M mutation is the standard predictive biomarker for third-generation epidermal growth factor receptor tyrosine kinase inhibitor (EGFR-TKI) treatment. While not all T790M-positive patients respond to third-generation EGFR-TKIs and have a good prognosis, it necessitates novel tools to supplement EGFR genotype detection for predicting efficacy and stratifying EGFR-mutant patients with various prognoses. Mixture-of-experts (MoE) is designed to disassemble a large model into many small models. Meanwhile, it is also a model ensembling method that can better capture multiple patterns of intrinsic subgroups of enrolled patients. Therefore, the combination of MoE and Cox algorithm has the potential to predict efficacy and stratify survival in non-small cell lung cancer (NSCLC) patients with EGFR mutations. Methods: We utilized the electronic medical record (EMR) and pharmacokinetic parameters of 326 T790M-mutated NSCLC patients, including 283 patients treated with Abivertinib in phase I (n=177, for training) and II (n=106, for validation) clinical trials and an additional validation cohort 2 comprising 43 patients treated with BPI-7711. Furthermore, 18 patients underwent whole-exome sequencing for biological interpretation of CoxMoE. We evaluated the predictive performance for therapeutic response using the area under the curve (AUC) and the Concordance index (C-index) for progression-free survival (PFS). Results: CoxMoE exhibited AUCs of 0.73-0.83 for predicting efficacy defined by best overall response (BoR) and achieved C-index values of 0.64-0.65 for PFS prediction in training and validating cohorts. The PFS of 198 patients with a low risk [median, 6.0 (range, 1.0-23.3) months in the abivertinib treated cohort; median 16.5 (range, 1.4-27.4) months in BPI-7711 treated cohort] of being non-responder increased by 43% [hazard ratio (HR), 0.56; 95% confidence interval (CI), 0.40-0.78; P=0.0013] and 50% (HR, 0; 95% CI, 0-0; P=0.01) compared to those at high-risk [median, 4.2 (range, 1.0-35) months in the abivertinib treated cohort; median, 11.0 (range, 1.4-25.1) months in BPI-7711 treated cohort]. Additionally, activated partial thromboplastin time (APTT), creatinine clearance (Ccr), monocyte, and steady-state plasma trough concentration utilited to construct model were found significantly associated with drug resistance and aggressive tumor pathways. A robust correlation was observed between APTT and Ccr with PFS (log-rank test; P<0.01) and treatment response (Wilcoxon test; P<0.05), respectively. Conclusions: CoxMoE offers a valuable approach for patient selection by forecasting therapeutic response and PFS utilizing laboratory tests and pharmacokinetic parameters in the setting of early-phase clinical trials. Simultaneously, CoxMoE could predict the efficacy of third-generation EGFR-TKI non-invasively for T790M-positive NSCLC patients, thereby complementing existing EGFR genotype detection.