Genomic Profiling Reveals Novel Predictive Biomarkers for Chemo-Radiotherapy Efficacy and Thoracic Toxicity in Non-Small-Cell Lung Cancer

Kewen He, Shaotong Zhang, Jiaohui Pang, Jiani C. Yin, Dianbin Mu, Jun Wang, Ge Hong, Jie Ma, Zhe Yang, Xiaoli Zheng, Lihua Dong, Junli Zhang, Pengyu Chang, Li Li, Shanshan Tang, Hua Bao, Xue Wu, Xiaonan Wang, Yang Shao, Jinming Yu, Shuanghu YuanFrontiers in Oncology 2022 · DOI

TL;DR

This study uses comprehensive genomic profiling of non-small cell lung cancer (NSCLC) tumors to identify genetic markers that predict patient response to chemo-radiotherapy (CRT) and the risk of developing treatment-induced toxicity. They found that alterations in pathways like KEAP1-NRF2, MET, PTEN, and NOTCH2 are associated with poorer survival, while specific SNPs in DNA repair genes can predict toxicity, paving the way for more personalized CRT strategies.

What Problem Does This Paper Address?

Chemo-radiotherapy (CRT) is a cornerstone treatment for unresectable stage III non-small cell lung cancer (NSCLC). However, its effectiveness is highly variable among patients, with some experiencing disease recurrence while others suffer severe side effects like pneumonitis and esophagitis. This variability is attributed to individual differences in radio-sensitivity and the body's response to radiation. Currently, there's a critical need for reliable biomarkers that can predict both treatment efficacy (survival outcomes) and toxicity risk before treatment begins. This would allow clinicians to tailor CRT regimens, potentially optimizing treatment intensity and minimizing harm for individual patients.

Methods

The researchers employed a targeted next-generation sequencing (NGS) approach, utilizing a panel of 474 genes known to be relevant to cancer and radiotherapy response. They analyzed pre-treatment biopsy samples from 122 patients with unresectable stage III NSCLC who underwent definitive CRT. The genomic data, including somatic mutations, copy number variations, and single nucleotide polymorphisms (SNPs), were correlated with clinical outcomes such as progression-free survival (PFS), overall survival (OS), and the incidence of radiation-induced pneumonitis and esophagitis. Statistical analyses included Kaplan-Meier survival curves, log-rank tests, and multivariate Cox proportional hazards models to identify independent predictive factors.

Datasets

The study utilized pre-treatment tumor biopsy samples from 122 patients with unresectable stage III NSCLC. These samples were collected between October 2014 and March 2019 from multiple centers. The genomic data generated from these samples formed the primary dataset for analysis. No new benchmark datasets were introduced.

Key Insights and Ideas

  • KEAP1-NRF2 Pathway as a Prognostic Indicator: The study strongly reinforces the role of the KEAP1-NRF2 pathway in predicting poor outcomes following CRT. This pathway is crucial for cellular defense against oxidative stress, and its dysregulation appears to significantly impair treatment response.
  • Novel Predictive Biomarkers Identified: Beyond the well-established KEAP1-NRF2 pathway, the research highlights FGFR family genes, MET, PTEN, and NOTCH2 as novel, independent predictors of poor post-CRT survival. This expands the repertoire of genetic markers for risk stratification.
  • Combined Genetic Signatures Improve Stratification: The power of combining alterations in these key genes (KEAP1, MET, PTEN, NOTCH2, FGFR) was demonstrated to significantly improve the stratification of patients into high- and low-risk groups for both PFS and OS.
  • SNPs Predict Radiotherapy Toxicity: The study moves beyond tumor genetics to identify specific SNPs in DNA repair genes (XRCC5, XRCC1) and metabolic genes (MTHFR, NQO1) that are associated with an increased risk of developing high-grade pneumonitis and esophagitis. This is a crucial step towards predicting and potentially mitigating treatment toxicity.
  • Somatic Mutations in Tumor Cells Can Influence Toxicity: A particularly novel finding is the association of somatic alterations in ZNF217 (amplification) and POLD1 (mutations) within tumor cells with an increased likelihood of developing severe radiation-induced toxicity. This suggests that the tumor's own genetic landscape can influence the normal tissue's response to radiation.

Key Results

  • Survival Outcomes: Patients with alterations in the KEAP1-NRF2 pathway had significantly shorter median PFS (6.7 months vs. 12.2 months) and OS (18.8 vs. 37.8 months) compared to those with wild-type genes.
  • Novel Gene Associations:
    • MET alterations were associated with shorter PFS (HR = 2.33, P=0.01) but not OS.
  • Deleterious PTEN mutations were linked to a higher progression risk (PFS, HR = 2.19, P=0.02).
  • NOTCH2 mutations correlated with unfavorable survival outcomes (PFS, HR = 2.0, P=0.02; OS, HR = 3.12, P=0.0002).
  • FGFR family gene alterations were associated with earlier progression (PFS, HR = 1.72, P=0.03) and poorer OS (HR = 2.04, P=0.01).
  • EGFR Mutations and OS: While EGFR activating mutations did not impact PFS, they were associated with improved OS (HR = 0.42, P=0.02), potentially due to subsequent targeted therapies.
  • Toxicity Prediction:
    • SNPs in XRCC1 (rs25487) and XRCC5 (rs3835) conferred increased risks of radiation-induced pneumonitis and esophagitis.
  • ZNF217 amplification was strongly associated with increased risks of developing severe pneumonitis and esophagitis (OR = 16.7, P=0.0007).
  • POLD1 mutations were associated with an increased overall toxicity risk (OR = 5.4, P=0.03).
  • Multivariate Analysis: MET, PTEN loss-of-function, and FGFR pathway alterations remained independent predictors of reduced PFS. NOTCH2 alterations and FGFR pathway alterations were independent predictors of unfavorable OS.

Limitations and Future Work

  • Retrospective Study: The study is retrospective, which inherently limits the ability to establish definitive causality.
  • Sample Size: While 122 patients is a reasonable cohort for genomic analysis, larger, prospective studies are needed to fully validate these findings and assess their predictive power across diverse patient populations.
  • Treatment Heterogeneity: The study included a mix of concurrent and sequential CRT, as well as varied radiation doses, which could introduce variability. Future work could focus on more homogeneous patient groups or incorporate these factors more explicitly into predictive models.
  • Subsequent Treatments: Limited information on subsequent lines of therapy for patients with specific alterations (e.g., MET) makes it challenging to fully disentangle the direct impact of CRT from later interventions.
  • Mechanistic Understanding: While associations are identified, the precise biological mechanisms by which these genetic variants influence CRT response and toxicity require further elucidation, potentially through in vitro or in vivo studies.
  • Clinical Validation: The authors emphasize the need for further validation of these biomarkers in larger cohorts and the development of multifactorial prediction models to integrate these genetic insights into clinical decision-making for personalized CRT.

Summary updated

Abstract

Chemo-radiotherapy (CRT) remains the main treatment modality for non-small-cell lung cancer (NSCLC). However, its clinical efficacy is largely limited by individual variations in radio-sensitivity and radiotherapy-associated toxicity. There is an urgent need to identify genetic determinants that can explain patients' likelihood to develop recurrence and radiotherapy-associated toxicity following CRT. In this study, we performed comprehensive genomic profiling, using a 474-cancer- and radiotherapy-related gene panel, on pretreatment biopsy samples from patients with unresectable stage III NSCLCs who underwent definitive CRT. Patients' baseline clinical characteristics and genomic features, including tumor genetic, genomic and molecular pathway alterations, as well as single nucleotide polymorphisms (SNPs), were correlated with progression-free survival (PFS), overall survival (OS), and radiotherapy-associated pneumonitis and/or esophagitis development after CRT. A total of 122 patients were enrolled between 2014 and 2019, with 84 (69%) squamous cell carcinomas and 38 (31%) adenocarcinomas. Genetic analysis confirmed the association between the KEAP1-NRF2 pathway gene alterations and unfavorable survival outcome, and revealed alterations in <i>FGFR</i> family genes, <i>MET</i>, <i>PTEN</i>, and <i>NOTCH2</i> as potential novel and independent risk factors of poor post-CRT survival. Combined analysis of such alterations led to improved stratification of the risk populations. In addition, patients with <i>EGFR</i> activating mutations or any oncogenic driver mutations exhibited improved OS. On the other hand, we also identified genetic markers in relation to radiotherapy-associated thoracic toxicity. SNPs in the DNA repair-associated <i>XRCC5</i> (rs3835) and <i>XRCC1</i> (rs25487) were associated with an increased risk of high-grade esophagitis and pneumonitis respectively. <i>MTHFR</i> (rs1801133) and <i>NQO1</i> (rs1800566) were additional risk alleles related to higher susceptibility to pneumonitis and esophagitis overall. Moreover, through their roles in genome integrity and replicative fidelity, somatic alterations in <i>ZNF217</i> and <i>POLD1</i> might also serve as risk predictors of high-grade pneumonitis and esophagitis. Taken together, leveraging targeted next-generating sequencing, we identified a set of novel clinically applicable biomarkers that might enable prediction of survival outcomes and risk of radiotherapy-associated thoracic toxicities. Our findings highlight the value of pre-treatment genetic testing to better inform CRT outcomes and clinical actions in stage III unresectable NSCLCs.
BibTeX
@inproceedings{he2022genomic,
  author = {Kewen He and Shaotong Zhang and Jiaohui Pang and Jiani C. Yin and Dianbin Mu and Jun Wang and Ge Hong and Jie Ma and Zhe Yang and Xiaoli Zheng and Lihua Dong and Junli Zhang and Pengyu Chang and Li Li and Shanshan Tang and Hua Bao and Xue Wu and Xiaonan Wang and Yang Shao and Jinming Yu and Shuanghu Yuan},
  title = {{Genomic Profiling Reveals Novel Predictive Biomarkers for Chemo-Radiotherapy Efficacy and Thoracic Toxicity in Non-Small-Cell Lung Cancer}},
  booktitle = {Frontiers in Oncology},
  year = {2022},
  doi = {10.3389/fonc.2022.928605},
}
Figure