TL;DR
This study introduces a novel "TMR score," combining tumor mutational burden (TMB) and T-cell receptor (TCR) repertoire diversity, to predict immunotherapy response in advanced non-small cell lung cancer (NSCLC). The TMR score demonstrated superior predictive power compared to individual biomarkers and was validated across multiple cohorts, offering a promising tool for personalized cancer treatment.
What Problem Does This Paper Address?
Immunotherapy has revolutionized cancer treatment, particularly for NSCLC, but predicting which patients will respond remains a significant challenge. Current biomarkers like PD-L1 expression and tumor mutational burden (TMB) have limitations; TMB, while indicative of neoantigen generation, doesn't fully capture the immune system's capacity to act on these antigens. The tumor microenvironment, specifically the diversity and activity of T cells, plays a crucial role in immunotherapy efficacy. However, integrating these distinct aspects—tumor genomics (TMB) and immune cell repertoire (TCR)—into a single predictive model for NSCLC immunotherapy response was not well-established. This paper aims to bridge this gap by developing and validating a combined biomarker that leverages both tumor intrinsic features and the adaptive immune response.
Methods
The researchers developed a "TMB-and-TCR" (TMR) score by combining TMB and TCR diversity using logistic regression. TMB was assessed through targeted DNA sequencing of tumor biopsy specimens, focusing on somatic single-nucleotide variants and indels. TCR repertoire analysis involved high-throughput sequencing of CDR3 beta regions from FFPE biopsy samples. Clonality was quantified using a formula that considers the proportion of rearranged TCR beta sequences. The TMR score was then calculated using TMB and TCR clonality values, with immunotherapy response (partial response/stable disease vs. progressive disease) as the categorical outcome. The predictive performance of the TMR score was evaluated using receiver operating characteristic (ROC) curves and compared against individual biomarkers like TMB and PD-L1. Progression-free survival (PFS) was analyzed using Kaplan-Meier curves and Cox hazard models.
Datasets
The study utilized several datasets:
- Discovery Set: 17 patients with advanced NSCLC who received immunotherapy.
- Validation Set: 21 patients with advanced NSCLC who received immunotherapy.
- External Validation Cohort 1: 225 early-stage NSCLC patients from a previously published study (MDACC cohort).
- External Validation Cohort 2: 306 NSCLC patients from the S2UC-MARK cohort.
The study also referenced data from other studies for comparative analysis of biomarkers like TIDE, TIGS, and IFNG.
Key Insights and Ideas
- Synergy of TMB and TCR: The core insight is that combining TMB (a measure of tumor's potential to generate neoantigens) with TCR diversity (a measure of the adaptive immune system's readiness) creates a more powerful predictive biomarker than either alone.
- TCR Clonality as a Key Component: TCR clonality, representing the expansion of specific T-cell clones, was found to be an independent factor and a crucial partner to TMB in predicting response.
- LRP1B Mutation Linkage: Tumors with higher TMR scores were more frequently associated with LRP1B gene mutations. This suggests a potential interplay between genomic alterations and immune response that is captured by the TMR score.
- Immune Activation and LRP1B: The study hints that LRP1B loss-of-function mutations might activate anti-tumor immune cells and pathways, potentially explaining the positive correlation between LRP1B mutations and better immunotherapy outcomes.
- Pathogen-Associated TCRs: A higher TMR score was linked to a greater abundance of pathogen-associated TCRs, suggesting that immunotherapy might prime T cells against common antigens, which can then cross-react with tumor neoantigens.
Key Results
- Superior Predictive Power: The TMR score showed significantly better predictive performance for immunotherapy response (distinguishing disease control from progression) compared to TMB or PD-L1 alone, with higher AUC values in both discovery and validation sets.
- Improved PFS Prediction: Patients with high TMR scores exhibited significantly longer progression-free survival (PFS) in both the discovery and validation cohorts, and this association remained independent in multivariate analysis.
- External Validation Success: The TMR score's predictive capability for PFS was confirmed in two independent external validation cohorts (225 and 306 NSCLC patients), demonstrating its robustness.
- Correlation with Tumor Shrinkage: While the TMR score itself was negatively correlated with tumor shrinkage, the LRP1B gene mutation frequency was positively correlated with tumor shrinkage (r = 0.68, p = 0.007), and LRP1B allele frequency was negatively correlated with the top 1% TCR sequences (r = -0.55, p = 0.033).
- TCR Expansion in High-TMR Group: Patients with high TMR scores had a higher total clone count and a greater frequency of the most frequent (top 1%) TCR clones, indicating a more robust T-cell response.
Limitations and Future Work
The authors acknowledge several limitations:
- Small Sample Size: The initial discovery and validation sets were relatively small (n=17 and n=21, respectively).
- Therapy Heterogeneity: The study included patients receiving both mono-immunotherapy and combination therapies (chemoimmunotherapy, antiangiogenic therapy), which could introduce variability.
- Lack of Experimental Evidence: The study is observational and lacks direct experimental evidence linking gene mutation dose to TCR expansion.
Future work could focus on:
- Larger, Prospective Studies: Validating the TMR score in larger, prospective clinical trials to solidify its role as a predictive biomarker.
- Mechanistic Investigations: Further exploring the molecular mechanisms underlying the interaction between LRP1B mutations, TCR repertoire, and immunotherapy response.
- Cost-Effectiveness: Investigating the feasibility and cost-effectiveness of integrating TMB and TCR sequencing into routine clinical practice.
- Broader Application: Assessing the TMR score's utility in other cancer types or for predicting other treatment outcomes.
Summary updated