Assessing Outcomes in NSCLC: Radiomic analysis, kinetic analysis and circulating tumor DNA

Maniar, Arpita; Wei, Alexander Z.; Dercle, Laurent; Bien, Harold; Fojo, Tito; Bates, Susan E.; Schwartz, Lawrence H.

doi:10.1053/j.seminoncol.2022.06.002

Cited by 3 publications

(3 citation statements)

References 67 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“… 4 Circulating tumor DNA (ctDNA) refers to the nucleic acids derived from the tumor that is present in the cell-free DNA fraction of a patient’s blood plasma. 7 According to several studies, ctDNA has great clinical potential as a noninvasive biomarker to monitor MRD of cancers. 3 , 6 – 8 Increasing evidence showed that positive ctDNA after curative treatment may predict a higher risk for relapse.…”

Section: Introductionmentioning

confidence: 99%

“… 7 According to several studies, ctDNA has great clinical potential as a noninvasive biomarker to monitor MRD of cancers. 3 , 6 – 8 Increasing evidence showed that positive ctDNA after curative treatment may predict a higher risk for relapse. Methods for detecting ctDNA include polymerase chain reaction (PCR)-based testing and next-generation sequencing (NGS).…”

Section: Introductionmentioning

confidence: 99%

“…Methods for detecting ctDNA include polymerase chain reaction (PCR)-based testing and next-generation sequencing (NGS). 7 However, there is currently no gold standard for the detection of ctDNA in ENSCLC.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Prognostic value of postoperative ctDNA detection in patients with early non-small cell lung cancer: a systematic review and meta-analysis

Guo

Lou

et al. 2023

Ther Adv Med Oncol

View full text Add to dashboard Cite

Background: Circulating tumor DNA (ctDNA) has emerged as a potential biomarker for monitoring early non-small cell lung cancer (ENSCLC), particularly after radical surgery. However, the prognostic value of postoperative ctDNA is still being investigated due to the small sample size and heterogeneity of patients with ENSCLC in current trials. Moreover, the potential clinical utility of ctDNA assessment for administering adjuvant therapy (AT) in patients with ENSCLC is also an important area of active research. Objectives: We aimed to identify the prognostic value of postoperative ctDNA detection in ENSCLC patients with stages I–III. Design: This study type is a systematic review and meta-analysis. Data sources and methods: We conducted a search in the Cochrane Library, Embase, PubMed, and ScienceDirect for prospective or retrospective investigations involving patients with ENSCLC, gathering outcomes based on predefined end points. The literature review adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, and the Newcastle–Ottawa scale was employed to carry out a quality evaluation of the included studies. The primary end point of the study was to evaluate the association of ctDNA status in two time points (within 1 month after surgery and long-term postoperative monitoring with more than 3 months) with relapse-free survival (RFS) and overall survival (OS). In addition, the study investigated the role of ctDNA in predicting the response to AT. The secondary end points of the study were to determine the impact of ctDNA on RFS and OS in different subgroups of ENSCLC patients based on pathological subtypes and TNM staging. Results: In total, 2149 studies were screened, and 11 studies met the inclusion criteria for the analysis. The presence of ctDNA within 1 month after surgery as well as long-term postoperative ctDNA were both associated with poorer RFS [hazard ratio (HR) = 4.43; 95% CI: 3.23–6.07 and HR = 7.99; 95% CI: 3.28–19.44, respectively] and worse OS (HR = 5.07; 95% CI: 2.80–9.19 and HR = 7.49; 95% CI: 3.42–16.43, respectively). Most subgroup analyses yielded similar results. Moreover, ctDNA-positive patients could acquire survival benefits from AT (HR = 0.30; 95% CI: 0.16–0.54), while ctDNA-negative patients that received AT did not show significant improvement in RFS (HR = 1.18; 95% CI: 0.67–2.09). Conclusion: The postoperative ctDNA assessment is a promising approach to stratify the risk of relapse and death in ENSCLC patients. Our data suggest that patients with negative ctDNA in the postoperative setting may not benefit from AT, which warrants further investigation. This finding, if validated in prospective trials with a larger sample size, could aid in better-individualized treatment for patients and avoid potential side effects of AT. Registration: This study was designed in accordance with PRISMA and registered with PROSPERO (CRD42022311615).

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Prognostic value of postoperative ctDNA detection in patients with early non-small cell lung cancer: a systematic review and meta-analysis

Guo

Lou

et al. 2023

Ther Adv Med Oncol

View full text Add to dashboard Cite

show abstract

A distributed feature selection pipeline for survival analysis using radiomics in non-small cell lung cancer patients

Gottardelli,

Gouthamchand,

Masciocchi

et al. 2024

Sci Rep

View full text Add to dashboard Cite

Predictive modelling of cancer outcomes using radiomics faces dimensionality problems and data limitations, as radiomics features often number in the hundreds, and multi-institutional data sharing is ()often unfeasible. Federated learning (FL) and feature selection (FS) techniques combined can help overcome these issues, as one provides the means of training models without exchanging sensitive data, while the other identifies the most informative features, reduces overfitting, and improves model interpretability. Our proposed FS pipeline based on FL principles targets data-driven radiomics FS in a multivariate survival study of non-small cell lung cancer patients. The pipeline was run across datasets from three institutions without patient-level data exchange. It includes two FS techniques, Correlation-based Feature Selection and LASSO regularization, and Cox Proportional-Hazard regression with Overall Survival as endpoint. Trained and validated on 828 patients overall, our pipeline yielded a radiomic signature comprising "intensity-based energy" and "mean discretised intensity". Validation resulted in a mean Harrell C-index of 0.59, showcasing fair efficacy in risk stratification. In conclusion, we suggest a distributed radiomics approach that incorporates preliminary feature selection to systematically decrease the feature set based on data-driven considerations. This aims to address dimensionality challenges beyond those associated with data constraints and interpretability concerns.

show abstract

A Distributed Feature Selection Pipeline for Survival Analysis using Radiomics in Non-Small Cell Lung Cancer Patients

Gottardelli,

Gouthamchand,

Masciocchi

et al. 2023

Preprint

View full text Add to dashboard Cite

Predictive modelling of cancer outcomes using radiomics faces dimensionality problems and data limitations, as radiomics features often number in the hundreds, and multi-institutional data sharing is often unfeasible. Federated learning (FL) and feature selection (FS) techniques combined can help overcome these issues, as one provides the means of training models without exchanging sensitive data, while the other identifies the most informative features, reduces overfitting, and improves model interpretability. Our proposed FS pipeline based on FL principles targets data-driven radiomics FS in a multivariate survival study of non-small cell lung cancer patients. The pipeline was run across datasets from three institutions without patient-level data exchange. It includes two FS techniques, Correlation-based Feature Selection and LASSO regularization, and Cox Proportional-Hazard regression with Overall Survival as endpoint. Trained and validated on 828 patients overall, our pipeline yielded a radiomic signature comprising "intensity-based energy" and "mean discretised intensity". Validation resulted in a mean Harrell C-index of 0.59, showcasing fair efficacy in risk stratification. In conclusion, we propose an approach for distributed radiomics involving preliminary FS to reduce the available feature set to the most important on a data-driven basis and reduce dimensionality problems other than those related to data limitations and interpretability.

show abstract

Assessing Outcomes in NSCLC: Radiomic analysis, kinetic analysis and circulating tumor DNA

Cited by 3 publications

References 67 publications

Prognostic value of postoperative ctDNA detection in patients with early non-small cell lung cancer: a systematic review and meta-analysis

Prognostic value of postoperative ctDNA detection in patients with early non-small cell lung cancer: a systematic review and meta-analysis

A distributed feature selection pipeline for survival analysis using radiomics in non-small cell lung cancer patients

A Distributed Feature Selection Pipeline for Survival Analysis using Radiomics in Non-Small Cell Lung Cancer Patients

Contact Info

Product

Resources

About