NeXT Personal identifies circulating tumor DNA in 81% of early-stage lung adenocarcinoma cases, offering potential for better disease stratification and outcome prediction.
Study: Ultrasensitive ctDNA detection for preoperative disease stratification in early-stage lung adenocarcinoma. Image Credit: Rawpixel.com/Shutterstock.com
In a recent study published in Nature Medicine, a group of researchers investigated the use of ultrasensitive circulating tumor DNA (ctDNA) detection to assess clinical risk and predict outcomes before surgery in early-stage lung adenocarcinoma (LUAD), a type of non-small-cell lung cancer originating from glandular cells.
Background
Liquid biopsy for ctDNA shows great promise for personalized management of early-stage cancers. Preoperative ctDNA status serves as a potential biomarker, while postoperative detection can guide adjuvant therapy and enable early identification of molecular residual disease (MRD).
Tumor-informed ctDNA detection, leveraging genomic profiling of tumor tissues, enhances sensitivity compared to tumor-agnostic approaches. However, challenges persist due to low ctDNA levels in early-stage cancers, interference from non-malignant cell DNA, sequencing errors, and clonal hematopoiesis. Further research is required to refine detection platforms for improved sensitivity, specificity, and clinical utility.
About the study
TRACERx patient recruitment and sample collection followed ethical regulations, with protocols approved by an independent research ethics committee. Eligible patients had histopathologically confirmed stage I-IIIB non-small-cell lung cancer (NSCLC) and were enrolled in the TRACERx study after providing informed consent.
Patient identifiers were anonymized and centrally managed. DNA degradation in archived formalin-fixed paraffin-embedded (FFPE) samples presented challenges for ctDNA panel quality, prompting the use of fresh frozen (FF) tissue for some samples.
Of the 204 patients, 62 had low-quality FFPE samples, and updated panels were generated for 31 using FF tissue, ensuring consistent quality with recently collected FFPE samples.
Tumor and normal samples underwent whole-genome sequencing (WGS) with a minimum tumor cellularity threshold of 20%. Libraries were prepared using standardized workflows and sequenced to 30× coverage.
Hybrid capture probe panels for ctDNA detection were designed using proprietary algorithms, prioritizing somatic variants from tumor-normal WGS data. Panels included approximately 1,800 high-quality variants to ensure ultrasensitive detection while excluding regions prone to sequencing errors or clonal hematopoiesis of indeterminate potential (CHIP).
Plasma samples, processed and stored at −80°C, were analyzed with a consistent pipeline. ctDNA levels were measured using Poisson models with a specificity threshold of 99.9%. The assay design minimized confounding factors, ensuring precise ctDNA detection and clinical relevance for survival analyses.
Statistical assessments conducted in the R statistical environment included survival and multivariable Cox regression models to evaluate the prognostic significance of ctDNA.
Study results
The study demonstrates the utility of an ultrasensitive ctDNA detection platform, NeXT Personal, for preoperative clinical risk stratification in early-stage LUAD within the TRACERx cohort.
NeXT Personal employs a tumor-informed approach, leveraging WGS to achieve a limit of detection (LOD) as low as 1-3 parts per million (ppm) with 99.9% specificity, even with low DNA input volumes. This platform addresses the challenge of detecting ctDNA at exceedingly low levels, which has limited the prognostic value of previous assays in early-stage LUAD.
In a cohort of 171 patients with non-small-cell lung cancer (NSCLC), including 94 LUADs and 77 non-LUADs, ctDNA was detected preoperatively in 81% of LUAD patients and 100% of non-LUAD patients. This represents a significant improvement compared to prior studies, which struggled to detect ctDNA in patients with stage I LUAD, where levels often fall below 80 ppm.
Using NeXT Personal, ctDNA was detected in 57% of patients with pathological stage I LUAD, compared to detection rates of 14% and 13% in earlier studies. Notably, ctDNA levels below 80 ppm were still prognostic of worse outcomes, underscoring the clinical significance of detecting ctDNA at such low levels.
Preoperative ctDNA status correlated strongly with clinical outcomes. Patients with ctDNA-negative status exhibited markedly better overall survival (OS) and recurrence-free survival (RFS) than those with detectable ctDNA, even at low levels. Specifically, 5-year OS for ctDNA-negative LUAD patients was 100%, compared to 61.4% for ctDNA-low and 48.8% for ctDNA-high patients.
Multivariable analyses confirmed ctDNA as an independent prognostic factor for OS and RFS in LUAD but not in non-LUAD, highlighting distinct biological differences between these NSCLC subtypes.
The findings suggest that highly sensitive ctDNA assays can identify a subset of LUAD patients at very low risk of recurrence, potentially guiding personalized therapeutic strategies.
While the study used retrospective data, the results demonstrate the feasibility of ctDNA as a biomarker for preoperative risk stratification, minimal residual disease monitoring, and early recurrence detection.
Conclusions
To summarize, the study concludes that the NeXT Personal platform offers unparalleled sensitivity and specificity for preoperative ctDNA detection in early-stage LUAD, with a detection limit as low as 1-3 ppm
This allows for accurate prognostic stratification, particularly in stage I LUAD, identifying high-risk patients even at ctDNA levels below the thresholds of prior methods.
While the assay demonstrated potential for predicting outcomes and tailoring therapies, prospective validation is necessary to confirm its clinical utility. Overall, the platform’s capability to detect ultra-low ctDNA levels holds promise for advancing personalized medicine in early-stage lung cancer.
AI could revolutionize cervical cancer screening and detection