Overview of ctDNA

The earliest clinical application of ctDNA testing as CDx was epidermal growth factor receptor (EGFR) mutation detection, mainly used to identify late stage NSCLC patients who may benefit from EGFR TKI. In the ENSURE study, the specificity and sensitivity of ctDNA detection for EGFR 19del and L858R mutations were 98.2% and 76.7%. According to the results of ctDNA testing, patients receiving Erlotinib treatment had significantly improved PFS compared to chemotherapy patients. Based on this research result, the FDA approved the first liquid biopsy CDx product Cobas EGFR Mutation Test v2 in 2016. Subsequently, the National Drug Administration (NMPA) approved the CDx product for domestic marketing in 2019. In the FLAURA study, the objective response rate (ORR) of Oxitinib in ctDNA T790M positive patients was consistent with that in tumor tissue biopsy positive patients. Based on this, in 2020, the FDA approved Guardant360 CDx for plasma ctDNA testing to identify NSCLC patients with EGFR L858R/19del/T790 Mmutations that can benefit from treatment with Oxitinib, EGFR 20ins mutations corresponding to Amivantamab (JNJ-6372), and KRAS G12C mutations corresponding to Sotorasib. Later, Foundation One Liquid CDx, a CDx product with a wider scope of application was approved, which was mainly used to identify NSCLC EGFR sensitive mutation (L858R/19del), ALK rearrangement, MET exon 14 hopping, prostate cancer BRCA1/2, ATM mutations, epithelial ovarian cancer/fallopiantube cancer or primary peritoneal cancer BRCA1/2 mutations, and breast cancer PIK3CA mutations to guide the medication treatment.

Currently, first-line targeted therapeutic drugs for driving gene-positive advanced NSCLC, such as EGFR, ALK, ROS1, BRAF, MET, RET, NTRK, and KRAS (G12C mutation), have been successively approved. Clinical treatment choices based on ctDNA testing results are increasingly supported by evidence from registered clinical trials and real-world studies. Studies have shown that in metastatic NSCLC, ctDNA testing can increase the detection rate by an additional 48% compared to patients who only use tissue biopsy for the recommended biomarkers in the guidelines, and can significantly reduce the detection report TAT (9 d vs15 d, P<0.001). The NSCLC NCCN Guidelines 2022 V1 recommend conducting ctDNA NGS testing, including EGFR, ALK, ROS1, BRAF, MET, RET, and KRAS (G12C mutation), for late stage patients who are not suitable for invasive biopsy, or when the quality of tissue samples obtained is poor, as well as when the sample amount is insufficient to carry out the necessary multi gene mutation testing. However, it also reminds that ctDNA NGS testing has a false negative rate of about 30%, and the corresponding analysis will be affected by clonal hematopoiesis. In 2021, the International Association for the Study of Lung Cancer (IASLC) consensusstated that ctDNA can be an effective tool for genotyping in newly diagnosed advanced NSCLC patients, of which the results can usually be complementary to tissue analysis. It can also be used as the preferred strategy for evaluating and monitoring the efficacy of targeted therapies using diagnostic biomarkers (with plasma as the priority).

Based on the results of SOLAR-1 research, FDA approved the second liquid biopsy CDx product Therascreen PIK3CA RGQ PCR Kit in 2019. This product mainly identifies patients with PIK3CA mutation advanced HR+/HER2 - breast cancer who benefit from the treatment of PIK3CA inhibitor Alpelisib (combined with Fulvestrant) through plasma ctDNA detection. A meta-analysis of breast cancer showed that the sensitivity and specificity of ctDNA for detecting PIK3CA mutations were 86% and 98%. NCCN Guide for breast cancer Version 2022 V1 recommends that recurrent unresectable or metastatic HR+/HER2 breast cancer should be subject to PIK3CA mutation analysis based on puncture tissue or ctDNA. If ctDNA detection is preferred but the result is negative, tissue detection should be conducted again for confirmation.

In gastric esophageal adenocarcinoma, there is often heterogeneity between the primary tumor and metastatic lesion, while ctDNA testing can comprehensively reflect the overall tumor situation to guide precise treatment more effectively. Research has shown that combining tissue and plasma ctDNA HER2 amplification detection can improve the predictive value of anti HER2 therapy in patients with esophageal cancer and gastroesophageal junction carcinoma. The NCCN Guidelines for Gastric and Esophageal Cancer and gastroesophageal junction carcinoma 2022 V1 suggest that patients with gastric and esophageal cancer who are not suitable for tissue biopsy may consider plasma ctDNA NGS testing. However, it is emphasized that negative ctDNA testing does not rule out the presence of genetic mutations through tissue analysis. In addition, due to the lack of effective treatment methods, NCCN Guide 2022 V1 suggests that ctDNA NGS detection should be considered when metastatic pancreatic cancer cannot obtain enough tissue through biopsy, and at least includes ALK, NRG1, NTRK, ROS1 fusion gene mutation and BRAF, BRCA1/2, HER2, KRAS, PALB2 gene mutation analysis. The NCCN guideline 2022 V1 for malignant melanoma of the skin also mentions ctDNA NGS detection as an alternative molecular typing method.

Recent studies have preliminarily shown that screening patients participating inclinical trials based on ctDNA NGS detection has sufficient accuracy, also can improve the trial enrollment rate and shorten screening time without affecting the efficacy. These research results suggest that ctDNA NGS detection has certain advantages and application prospects in clinical trials. However, in addition to the approved CDx products, the effectiveness and practicality of most ctDNA in other targeted treatment guidelines are still insufficient, which needs to be supported by large amount of experimental results and clinical trials. Subsequent research should incorporate the expected target population as much as possible, and the clinical significance should be fully verified through strictly designed clinical studies. For different cancer species, the consistency between plasma ctDNA oncogene mutation detection and tumor tissue gene mutation detection still exist significant differences. For cancer species with large differences in detection results, it should be comprehensively considered the clinical application risks and the feasibility of ctDNA NGS detection as an alternative to tissue gene detection.

2. The value of ctDNA NGS detection in evaluating the efficacy of tumor treatment and stratification of prognostic risk

Traditional methods such as tumor markers and imaging cannot dynamically reflect the molecular evolution of tumor characteristics when evaluating the therapeutic efficacy of molecular targeting and immune checkpoint inhibitors. CtDNA NGS detection can monitor the abundance changes of ctDNA related to tumor driving genes and determine tumor treatment response. Among them, in EGFR targeted therapy of NSCLC, early clearance of EGFR sensitive mutations can be used to predict the therapeutic efficacy of EGFR TKI. A real-world study on dynamically monitoring ctDNA to predict the clinical treatment efficacy of NSCLC found that a higher baseline ctDNA content or greater number of variants indicates a shorter OS (PFS and OS are longer in patients with ctDNA clearance after treatment), and ctDNA is also an independent prognostic factor independent of treatment type and evaluation time point setting. In patients with advanced NSCLC treated with immune checkpoint inhibitors, the ctDNA response (decreased by>50%) was consistent with the efficacy evaluated by imaging, and was associated with better PFS and OS. Among the patients with stable lesions evaluated by early imaging at 5-9weeks after treatment, median OS significantly prolonged in those with ctDNA evaluatedas responsive (31.2 months vs 18.4 months, HR=0.36). Proteintyrosine phosphatase receptor type D (PTPRD) is inactivated in various tumors. When there is a deletion mutation in the PTPRD phosphatase domain in nonsquamous NSCLC, second-line treatment can benefit more from atezolizumab (median OS:24.04 months vs 9.69 months, HR=0.16, P=0.0273), and PTPRD is a prognosticfactor independent of TMB, PD-L1 expression or TP53, EGFR, KRAS gene mutation status. In metastatic hormone resistant prostate cancer, a decrease in ctDNA levels is associated with a decrease of over 30% in PSA. The TOPARP-A study also found that a decrease in ctDNA levels is associated with OS in advanced prostate cancer patients treated with PARP inhibitors. However, in a recent study on the treatment of melanoma combined with brain metastasis immune checkpoint inhibitor, it was found that plasma ctDNA analysis could not monitor the efficacy of intracranial lesions very well, which showed the adverse effect of blood brain barrier on ctDNA detection.

The concept of minimal residual disease (MRD) originated from hematological tumors. The MRD concept of solid tumors is more commonly referred to as molecular residual lesions (MRD), which refers to the tumor-derived molecular abnormalities that cannot be detected by traditional imaging (including PET/CT) or other laboratory methods after treatment, but are detected through liquid biopsy, indicating the persistence and clinical progression of the tumor. The study of TracerX confirmed the feasibility of ctDNA as detection of MRD and evaluated the predictive value of MRD detection for postoperative recurrence of NSCLC. The ctDNA dynamic analysis results of a retrospective study, which included 22 patients with stage IB-IIIA NSCLC who received neoadjuvant therapy (mostlyneoadjuvant immunotherapy), were highly consistent with postoperative pathological evaluation, with a sensitivity of 100.00% and an accuracy of 91.67%. Among the study, patients who tested positive for ctDNA at 3-8 days after surgery had a higher risk of recurrence (HR=5.37), and the median time for predicting molecular recurrence events based on ctDNA NGS detection was 6.83 months earlier than the imaging assessment.

In 2021, ‘Expert Consensus of Molecular Residual Disease for Non-Small Cell Lung Cancer’ defines lung cancer molecular abnormalities as ctDNA with a stable detection abundance of ≥ 0.02% in peripheral blood, including lung cancer driving genes or other Class I/II genemutations. Results of a clinical study using ctDNA detection to monitor MRD included 261 patients with NSCLC that can be cured by surgery, showed that 36.4% of patients were positive for preoperative ctDNA testing, indicating the exist of no-shedding tumors, but did not affect postoperative MRD monitoring. The study also found that the prognosis of patients with negative MRD detection at a single postoperative node was significantly better than that of positive patients (HR=0.08, 95% CI: 0.02-0.33), indicating the prediction accuracy of dynamic monitoring can be further improved. This result defines the potential cure population for the first time, and it is found that the postoperative MRD negative population cannot benefit from adjuvant therapy. Only the proportion of MRD detection in patients with brain recurrence significantly decreased (20%, 1/5).

In colon cancer patients, continuous ctDNA monitoring after surgery can predict molecular recurrence up to 16.5 months earlier than imaging standard assessment. Multiple studies have shown a strong correlation between ctDNA positivity and disease recurrence after colorectal cancer surgery and adjuvant therapy, and postoperative ctDNA analysis results can serve as a basis for prognosis prediction. Recently, the first randomized controlled trial DYNAMIC based on ctDNA for postoperative adjuvant therapy of colon cancer showed that ctDNA guided adjuvant chemotherapy strategy can reduce the postoperative adjuvantchemo therapy rate by nearly 50% (from 28% to 15%) without affecting the overall relapse free survival of patients, avoiding ineffective chemotherapy in some patients. This study provides a direct basis for postoperative adjuvant treatment decisions in patients with stage II colon cancer.

IMvigor010 is a phase III clinical study of adjuvant immunotherapy for high-risk myometrial invasive urothelium cancer. The results show that the median disease-free survival and OS of ctDNA positive patients receiving Atezolizumab have significantly improved, with a 42% reduction in the risk of recurrence and a 41% reduction in the risk of death. At present, many companies have launched solid tumor MRD detection products, but no approved (FDA/NMPA) products have been launched. It is urgent to verify the clinical efficacy of these products through large-scale, multicenter, and prospective clinical trials.

3. The value of ctDNA NGS detection in analyzing the mechanism of acquired drug resistance in tumors

Acquired drug resistance is an important factor affecting the efficacy of precise tumor treatment, which is also the biggest challenge faced by the overall management of clinical application of anti-tumor drugs. Due to its advantages, ctDNA NGS detection has been widely used in the molecular mechanism analysis of acquired drug resistance in tumors and subsequent drug guidance. For example, when advanced NSCLC patients receiving Osimertinib treatment develop resistance, new mutations in the EGFR gene are often detected, with EGFRG796/C797 mutations accounting for 24.7%, EGFR L792 mutations accounting for10.8%, and EGFR L718/G719 mutations accounting for 9.7%. Based on ctDNA NGS detection, it was found that the mechanism of ALK inhibitor resistance is highly heterogeneous. For example, secondary mutations such as ALK L1196M,I1171T, D1203N, G1269A/F1174L often occur in Crizotinib resistance, and there may be mutations related to bypass activation such as NRAS G12V, EGFR R108K, PIK3CA E545K, etc. Cerinib resistance often results in secondary mutations such as ALK G1128A, G1202R, G1269A, I1171T/E1210K, and may involve bypass activation related mutations such as KIT D820E, MET E1012 *, EGFR P265_ C291del. Alectinib resistance often results in secondary mutations such as ALK G1202R, W1295C, G1202R/L1196M, and may involve bypass activation related mutations such as EGFRP753S. Lorlatinib resistance often results in secondary mutations such as ALKG1202R/G1269A, and may involve bypass activation related mutations such as BRAFV600E and MET D1246N.

Mutations in protein coding genes and/or alterations in the extracellular domain (ECD) of EGFR in the RAS signaling pathway may be the main molecular mechanisms leading to drug resistance in metastatic colorectal cancer patients receiving anti EGFR monoclonal antibody treatment. In the treatment of gastric cancer, the change of HER2 copy number detected by ctDNA monitoring indicates the drug resistance of trastuzumab. Therefore, it is considered that HER2 copy number in ctDNA can be used as an indicator to monitor the efficacy of trastuzumab in the treatmentof advanced gastric cancer.

The mechanism of acquired drug resistance in tumor immune checkpoint inhibitor therapy is complex. The evolution of tumor subclones under treatment selection pressure is only a partial reason. In the most active field of tumor transformation research, ctDNA targeted sequencing, whole exon and/or whole genome detection, as well as many new technologies such as single-cell omics, metagenomics, and spatial genomics, have begun to be applied. In a phase II clinical trial of Avelumab combined with EGFR monoclonal antibody and modified FOLFOX6 regimen for microsatellite stable, RAS/BRAF wild-type metastatic colon cancer, ctDNA NGS detection revealed that selective pressure will cause PD-L1 mutation subclone evolution, with a few patients exhibiting PD-L1 K162fs mutations that led to RNA attenuation and PD-L1 L88S subclones that enhanced protein degradation. However, although current research has shown that ctDNA NGS detection has certain advantages in exploring the molecular mechanisms of tumor resistance and drug guidance, especially providing important references for clinical diagnosis or treatment planning of difficult/complex cases, as an emerging technological means, there is still a lack of sufficient clinical data to support its routine application in clinical practice. It is believed that with the continuous deepening research of clinical efficacy and utility, ctDNA testing will play a greater role in tumor resistance, thereby benefiting more patients.