Abstract
Background
Circulating tumor DNA (ctDNA) has emerged as a potential novel biomarker to predict molecular residual disease (MRD) in lung cancer after definitive treatment. Herein, we investigated the value of ctDNA in prognosing risk of relapse and monitoring the effect of adjuvant therapy in surgical non‐small cell lung cancer (NSCLC).
Methods
We enrolled 58 NSCLC patients in a real‐world setting, and 58 tumor tissues and 325 plasma samples were analyzed. Tumor tissues and plasma samples were subjected to targeted next‐generation sequencing (NGS) of 1021 cancer‐related and ultra‐deep targeted NGS covering 338 genes, respectively.
Results
ctDNA was detected in 31.0% of cases at the first postoperative time, which was associated with advanced tumor stage, T stage and KEAP1 or GRIN2A mutations in tissues. ctDNA positivity at landmark and longitudinal indicated the shorter disease‐free survival. For patients with ctDNA positivity at the first postoperative time, regardless of adjuvant therapy, all patients who were persistently ctDNA positive during postoperative surveillance had disease recurrence. Among the patients who were ctDNA negative, only two patients (15.4%, 2/13) receiving adjuvant therapy relapsed, while one patient (50.0%, 1/2) without adjuvant therapy relapsed. For the first postoperative ctDNA negative patients, the recurrence rate of patients with adjuvant therapy was and higher than without adjuvant therapy (22.6% [7/31] vs. 11.1% [1/9]). The patients who became ctDNA positive may also benefit from intervention therapy.
Conclusion
Postoperative ctDNA is a prognostic marker, and ctDNA‐detection may facilitate personalized adjuvant therapy, and applying adjuvant therapy to the patients with detectable ctDNA could bring clinical benefits for them.
Keywords: adjuvant therapy, ctDNA, disease relapse, molecular residue disease, non‐small cell lung cancer
Here, we aimed to investigate the value of ctDNA in prognosing risk of relapse and monitoring the effect of adjuvant therapy in surgical NSCLC and we found that pTNM stage were independent predictors of ctDNA detection at postoperative one month. Postoperative ctDNA is a prognostic marker, and ctDNA‐detecting may facilitate personalized adjuvant therapy, and applying adjuvant therapy to the patients with detectable ctDNA could bring clinical benefits for them.
INTRODUCTION
Non‐small cell lung cancer (NSCLC) accounts for 85%–90% of all malignant lung cancers and is one of the most common types of cancer with high morbidity and mortality worldwide. 1 , 2 The surgical resection of NSCLC stage I–IIIA is the preferred curative treatment for 30% of patients. 3 , 4 It is common for patients with NSCLC who are initially free of metastatic disease to experience relapse within 5 years of primary tumor resection, depending on the tumor type and stage. 5 , 6 A high percentage of patients with early‐stage NSCLC experience recurrence, so they require further treatment, but their outcomes are poor. 7 , 8 , 9 Clinical studies have shown only minimal benefit of adjuvant chemotherapy on overall survival. 7 , 8 , 9
In recent decades, adjuvant treatment models including immunotherapy and targeted therapy have made great progress in NSCLC to increase the cure rate of surgical NSCLC after radical resection. 10 , 11 Clinical surveillance of these patients after treatment is recommended to identify the patients who would benefit from adjuvant therapy and those who are at risk of relapse, and avoiding overtreatment of patients who have been successfully cured. 12 Molecular residual disease (MRD) that may persist after initial therapy as a potential source of subsequent metastatic relapse at distant sites cannot be detected by radiological imaging. 13 Therefore, an identifying reliable biomarker to detect MRD to identify patients at high risk of relapse is critical.
ctDNA reveals a comprehensive view of the tumor, as it was shed by tumor cells throughout the tumor. 13 As a noninvasive biomarker that can be used in several cancers to determine molecular profiles and predict MRD after definitive treatment, it has shown promise. 14 , 15 , 16 , 17 In lung cancer, several studies have also proved that the detection of ctDNA may reveal MRD and identify patients who may be at higher risk of relapse in patients with localized disease. 12 , 18 , 19 , 20 , 21 , 22 Recent studies have confirmed that ctDNA‐based MRD detection plays an important prognostic role in patients with NSCLC after definitive surgery. 12 , 18 , 19 , 20 , 21 , 22 Patients with detectable ctDNA have a worse prognosis than those with undetectable ctDNA. 12 , 18 , 19 , 20 , 21 , 22 The studies have also shed light on the ctDNA predictive value for adjuvant therapy. 12 , 21 , 22 It suggested that adjuvant therapy may be unnecessary for patients with undetectable ctDNA. 12 However, the potential of ctDNA in indicating the recurrence risk and monitoring the effect of adjuvant therapy remains to be examined in a real‐world setting in clinical practice, which has more diverse characteristics of patients and treatment features.
Examining the clinical utility of ctDNA in real‐world settings can further confirm the potential of ctDNA in predicting MRD in clinical practice. To assess the clinical value of ctDNA in prognosing risk of relapse and monitoring the effect of adjuvant therapy for patients with surgical NSCLC, we investigated 58 NSCLC patients, who underwent NGS in a real‐world setting using an optimized tumor‐informed ctDNA detection method.
METHODS
Clinical cohort
In this study, 60 patients with stage I to III NSCLC (no malignant tumor history within the past 5 years) treated with surgery definitive were enrolled from January 2019 to December 2021. Clinicopathological data, demography, tumor histopathological results were collected from each patient. This study was approved by the institutional review board of Xuanwu hospital capital medical university. All subjects provided informed, written consent before study‐related procedures. This study was conducted in accordance with the Declaration of Helsinki.
Sample collection and DNA extraction
Resected tissues were the source of tumor tissues for next‐generation sequencing (NGS), and were collected at surgery. Peripheral blood samples were collected for ctDNA analysis, and 20 mL plasma samples were collected at postoperative 1 month and then at every third or sixth month until recurrence. The ctDNA analysis was performed by Geneplus Inc. in a blinded fashion. Separating peripheral blood lymphocytes (PBLs) and plasma from blood samples was done by sequential centrifugation. As per the manufacturer's instructions, the genomic DNA from frozen tissues, PBLs and formalin‐fixed, paraffin‐embedded specimens (FFPE) was extracted using the DNeasy Blood & Tissue Kit (Qiagen) and Maxwell16 FFPE Plus LEV DNA purification kit (Qiagen), respectively. ctDNA was isolated from plasma using the QIAamp circulating nucleic acid kit (Qiagen). DNA concentration was determined using Invitrogen's Qubit fluorometer and the Qubit dsDNA HS kit (high sensitivity).
Next‐generation sequencing
In accordance with the manufacturer's instructions, the sequencing libraries were prepared using the KAPA DNA library preparation kit (Kapa Biosystems). Hybridization of barcoded libraries was performed on a previously reported customized 1021 panel (Table S1) for tissue samples and a 338 panel with ultra‐deep (Table S2) for plasma samples. After hybrid selection, DNA fragments were amplified and pooled into multiple multiplexed libraries. DNA sequencing was performed with the Illumina Nextseq CN 500 (Illumina) or the Gene+Seq‐2000 Sequencing System (GenePlus‐Suzhou).
Statistical analysis
Mann‐Whitney U test, Chi‐square and Fisher's exact tests were used for continuous variables and categorical variables, respectively. Disease‐free survival (DFS) assessed by standard radiological criteria was defined as the time from the data of surgery to the first radiological recurrence (local or distant) or death and was censored at last follow‐up. The Kaplan–Meier method and the log‐rank test was used to assess survival analysis. Multivariate Cox proportional hazards regression analysis was performed to evaluate the association of ctDNA and clinical variables. All data analyses and presentations were performed using R software (version 3.6.3) and GraphPad software (version 8.0.2). All statistical tests were performed with two‐sided methods, and statistical significance was defined as p < 0.05.
RESULTS
Patient characteristics
In this study, 60 patients with surgical NSCLC were enrolled from January 2019 to December 2021. Two patients with multiple primary lung cancer or noncancer causes of death were excluded (Figure 1). A total of 58 eligible patients were included in the analysis of the performance of postoperative ctDNA, whose clinicopathological characteristics and demographics are listed in Table 1. With a median follow‐up of 31.1 months (range: 1.4–36), 14 patients (24.1%) relapsed with disease progression, while 44 (75.9%) remained disease‐free.
FIGURE 1.
Flow chart of study design and patient enrollment. NSCLC, non‐small cell lung cancer; NGS, next‐generation sequencing; ctDNA, circulating tumor DNA.
TABLE 1.
Patient characteristics.
| Characteristic | Number | Percentage (%) |
|---|---|---|
| Age (years) | ||
| Median (range) | 60.5 (35,82) | |
| Sex | ||
| Male | 31 | 53.4 |
| Female | 27 | 46.6 |
| Family history | 21 | |
| Smoking history | ||
| Current/former | 28 | 48.3 |
| Never | 30 | 51.7 |
| Histology | ||
| Adenocarcinoma | 49 | 84.5 |
| Squamous cell carcinoma | 7 | 12.1 |
| Adenosquamous carcinoma | 2 | 3.4 |
| pTNM stage | ||
| I | 28 | 48.3 |
| II | 9 | 15.5 |
| III | 21 | 36.2 |
| T stage | ||
| T1–T3 | 54 | 93.1 |
| T4 | 4 | 6.9 |
| Lymph node metastasis | ||
| No | 32 | 55.2 |
| Yes | 26 | 44.8 |
| Adjuvant therapy | ||
| No | 12 | 20.7 |
| Chemotherapy | 19 | 32.8 |
| Targeted therapy | 25 | 43.1 |
| ICI‐combined chemotherapy | 2 | 3.4 |
| Recurrent disease | ||
| No | 44 | 75.9 |
| Yes | 14 | 24.1 |
| Median follow‐up time, months | 31.1 |
Abbreviations: ICI, immune checkpoint inhibitor; pTNM, postsurgical tumor‐mode‐metastasis; T, tumor.
Mutational profile of tumor tissues and ctDNA detection
Genomic DNA were analyzed to identify tumor variants after excluding clonal hematopoietic and germline mutations. Somatic mutations were detected in all tumor tissues and the mutation landscape of the 58 patients is shown in Figure 2. Overall, 503 somatic variations, including 368 SNVs, 56 indels, nine fusions, and 70 somatic CNVs, were identified, with a median of seven (range: 1–42) gene variations detected in each patient. EGFR was the most frequently mutated driver gene in tumor tissues (62.1%), followed by TP53 (60.3%), LRP1B (19.0%) and KRAS (13.8%). Driver events in EGFR most frequently occurred in patients with adenocarcinoma, whereas more frequent TP53 mutations were observed in other patients. The mutation landscape was consistent with that previously reported in NSCLC patients. 12 , 18 , 19 , 23 , 24
FIGURE 2.
Summary of patient characteristics and somatic mutations in tumor tissues of the 58 evaluable non‐small cell lung cancer (NSCLC) patients.
ctDNA positivity was defined as at least one mutation detected in the plasma sample. The specific analysis method is consistent with the previous research. 12 The first postoperative ctDNA positive was found in 18 of 58 patients (31.0%), which was associated with the postsurgical tumor‐mode‐metastasis (pTMN) stage; and ctDNA positive rate in stage I, stage II and stage III cases was 17.2% (4/28), 44.40% (4/9) and 47.60% (10/21), respectively. For all plasma samples which were collected at follow‐up, ctDNA positivity was found in 61 of 325 cases (18.8%). The most common mutations detected in plasma samples included mutations in TP53 and EGFR. Activating mutations such as EGFR mutation and ALK fusion were also detected in the plasma sample, suggesting the potential treatment options.
Correlation between ctDNA status and clinicopathological and molecular features
The relationship between clinicopathological factors and the first postoperative ctDNA detection is shown in Table 2. Postoperative ctDNA positivity at 1 month was associated with more advanced tumor stage and T stage (Table 2). Multivariable analysis indicated that pTNM stage were independent predictors of postoperative ctDNA detection at 1 month (Figure 3a). Regarding gene mutations, we compared the differences in the genes mutated in the tumor tissues of the ctDNA positive and negative groups. ctDNA positivity was more frequent in patients with KEAP1 (16.67% [3/18] vs. 0% [0/40], p = 0.0264) or GRIN2A mutations (16.67% [3/18] vs. 0% [0/40] p = 0.0264) (Figure 3b).
TABLE 2.
Relationship between clinicopathological factors and first postoperative ctDNA status.
| Characteristic | ctDNA positivity (n = 18) | ctDNA negative (n = 40) | p‐value |
|---|---|---|---|
| Age (years) | 0.323 | ||
| Median (min–max) | 62.0 (44,82) | 59.5 (35,79) | |
| Mean (SD) | 62.1 (9.6) | 58.9 (11.5) | |
| Sex | 0.256 | ||
| Male | 12 (66.7%) | 19 (52.5%) | |
| Female | 6 (33.3%) | 21 (47.5%) | |
| Smoking history | 0.189 | ||
| Current/former | 11 (61.1%) | 17 (42.5%) | |
| Never | 7 (38.9%) | 23 (57.5%) | |
| Histology | 0.818 | ||
| Adenocarcinoma | 15 (83.3%) | 34 (85.0%) | |
| Nonadenocarcinoma | 3 (16.7%) | 6 (15.0%) | |
| pTNM stage | 0.039 | ||
| I/II | 8 (44.4%) | 29 (72.5%) | |
| III | 10 (55.6%) | 11 (27.5%) | |
| T stage | 0.048 | ||
| T1–T3 | 15 (83.3%) | 39 (97.5%) | |
| T4 | 3 (16.7%) | 1 (2.5%) | |
| Lymph node metastasis | 0.094 | ||
| No | 7 (38.9%) | 25 (62.5%) | |
| Yes | 11 (61.1%) | 15 (37.5%) | |
| TMB | 0.129 | ||
| Median (min–max) | 5.4 (0.9, 13.4) | 2.9 (0, 15.3) | |
| Mean (SD) | 5.5 (4.1) | 3.9 (3.5) | |
| Number of somatic mutations | 0.767 | ||
| Median (min–max) | 7.5 (1,42) | 7.0 (3,30) | |
| Mean (SD) | 9.7 (9.3) | 8.2 (5.6) |
Note: Chi‐square test, Fisher's exact and Wilcoxon rank sum tests were used for categorical and continuous variables, respectively. p‐values shown reflect comparison between patients with positive first‐postoperative ctDNA and patients with negative first‐postoperative ctDNA. p‐values < 0.05 were considered statistically significant.
Abbreviations: pTNM, pathological tumor‐node‐metastasis; T, tumor; TMB, tumor mutation burden.
FIGURE 3.
Clinicopathological factors and gene mutations affecting first postoperative ctDNA status. (a) Multivariate analysis for disease‐recurrence prediction with multiple clinicopathological variables and ctDNA positivity at postoperative 1 month. (b) The differences in the genes mutated in the tumor tissues of the ctDNA positive and negative groups.
The ctDNA positivity rate after surgery in stage I/II NSCLC was lower than the late stage. In order to explore the features of ctDNA‐positive patients in stage I/II, we analyzed the difference of malignant pathological features and tumor tissues molecular features between the ctDNA‐positive group and the ctDNA‐negative group (Table 3). In the ctDNA‐positive group, patients were more likely to have micropapillary structures. However, the malignant pathological features, such as bronchial wall invasion, pleural invasion, vessel carcinoma embolus in these cases, were unrelated to the ctDNA status. For tumor tissues molecular factors, tumor mutation burden (TMB) was higher in the ctDNA‐positive group and somatic variation had no significant difference.
TABLE 3.
Relationship between malignant pathological and tumor tissues molecular factors and ctDNA status in stage I/II patients.
| Characteristic | ctDNA positivity (n = 8) | ctDNA negative (n = 29) | p‐value |
|---|---|---|---|
| Bronchial wall invasion | 0.655 | ||
| No | 7 (87.5%) | 22 (75.9%) | |
| Yes | 1 (12.5%) | 7 (24.1%) | |
| Pleural invasion | 0.655 | ||
| No | 1 (12.5%) | 7 (24.1%) | |
| Yes | 7 (87.5%) | 22 (75.9%) | |
| Micropapillary structure | 0.041 | ||
| No | 2 (25.0%) | 19 (65.5%) | |
| Yes | 6 (75.0%) | 10 (34.5%) | |
| Vessel carcinoma embolus | 0.925 | ||
| No | 7 (87.5%) | 25 (86.2%) | |
| Yes | 1 (12.5%) | 4 (13.8%) | |
| TMB | 0.036 | ||
| Median (min–max) | 6.24 (0.96–13.44) | 2.88 (0.00–15.36) | |
| Mean (SD) | 6.91 (4.49) | 3.58 (3.43) | |
| Number of somatic mutations | 0.257 | ||
| Median (min–max) | 8 (1–42) | 6 (3–30) | |
| Mean (SD) | 13.0 (12.94) | 7.9 (5.89) |
Note: Chi‐square test, Fisher's exact and Wilcoxon rank sum tests were used for categorical and continuous variables, respectively. p‐values shown reflect comparison between stage I/II patients with positive first‐postoperative ctDNA and patients with negative first‐postoperative ctDNA. p‐values < 0.05 were considered statistically significant.
Abbreviation: TMB, tumor mutation burden.
Association of MRD detection with patient outcome
During a median 31.1 months follow‐up, 14 patients experienced recurrence. Among these patients, 33.3% (6/18) patients with detectable ctDNA in the first postoperative blood sample experienced recurrence, while 20.0% (8/40) of patients with undetectable ctDNA ultimately experienced recurrence. Patients with postoperative ctDNA detected at 1 month had higher recurrence rate than those in patients without ctDNA detection, but there was no significant difference (p = 0.294) (Figure 4a). During postoperative surveillance, longitudinal collected plasma samples had been used for ctDNA analysis. We detected ctDNA at least one time point after surgery in 31 patients (53.4%), and 13 of them (41.9%) experienced recurrence. Out of a total of 27 patients without ctDNA detection during postoperative surveillance, one (3.7%) of them ultimately recurred. Patients with ctDNA detection during postoperative surveillance had a significantly higher recurrence rate than those without ctDNA detection (p = 0.001) (Figure 4b). Serial ctDNA detection revealed disease recurrence ahead of radiological imaging by a median of 10.3 months (range: 0.98–34.4).
FIGURE 4.
ctDNA monitoring after surgery. (a) Recurrence rate in patients with positive or negative postoperative ctDNA at 1 month. (b) Recurrence rate in patients with positive ctDNA (detected ctDNA at least one time point) or negative ctDNA during postoperative surveillance. (c) Kaplan–Meier estimates of disease‐free survival (DFS) for patients with resectable lung cancer stratified by molecular residual disease (MRD) detection at landmark. (d) Kaplan–Meier estimates of DFS for patients with resectable lung cancer stratified by MRD detection at longitudinal. p‐value was calculated by the log‐rank test and HR by the Cox exp (beta) method. MRD positive was defined as patients with detectable ctDNA, whereas patients without detectable ctDNA were defined as MRD negative.
Based on these findings, we further analyzed the values of landmark and longitudinal MRD relapse prediction. Landmark time point was defined as the time of phlebotomy collection after the last cycle of chemotherapy for patients who received adjuvant chemotherapy, and the time of first phlebotomy collection after surgery for others. Longitudinal time points were defined as every 3 to 6 months since the landmark detection. Only 12.5% (5/40) of patients in the MRD negative group at landmark had disease recurrence, but 50.0% (9/18) of patients in the MRD positive group had relapsed. Compared with patients without MRD detection at landmark, MRD positive patients had a markedly reduced DFS (p = 0.006; hazard ratio [HR], 4.60; 95% CI: 1.54–13.74) (Figure 4c).
During postoperative surveillance, longitudinal collected plasma samples had been used for MRD analysis. We detected longitudinal MRD positive at least one time point during surveillance in 29 patients (50.0%), and 13 of them (44.8%) experienced recurrence. A total of 29 patients (50.0%) remained longitudinal MRD negative during surveillance; however, one of them (3.4%) ultimately recurred. The DFS can be stratified by longitudinal MRD status, with MRD positive groups having significantly high recurrence ratio and reduced DFS (p = 0.007; hazard ratio [HR], 16.07; 95% CI: 2.10–123.05) (Figure 4d).
Association of ctDNA status with outcomes of adjuvant therapies
We explored the possibility of ctDNA detection in aiding the selection of eligible NSCLC patients for adjuvant therapies. Among the 58 patients, 12 patients had surgery alone and 46 patients received adjuvant therapy. Of these, 19 patients received chemotherapy, 25 received targeted therapy, and two received immunotherapy combined chemotherapy. Among the 18 patients who were postoperative ctDNA positive at 1 month, four out of 15 patients who received adjuvant therapies experienced relapse, while two of three patients not receiving adjuvant therapy experienced relapse. Among the 15 patients who received adjuvant therapy and were first postoperative ctDNA positive, 13 patients became ctDNA negative during follow‐up (at least once negative), and two of them relapsed. Another two patients who did not become negative relapsed. For three first postoperative ctDNA positive patients without adjuvant therapy, two of them became negative, and one of them relapsed. Another patient who was positive at all time points relapsed.
A total of 31 patients who were first postoperative ctDNA negative received adjuvant therapies, and 24 remained relapse‐free. A total of eight of nine patients without adjuvant therapy remained relapse‐free. A total of 22 patients were persistently negative, and only one patient who might have harbored nonshedder clinicopathological features for ctDNA detection relapsed.
For patients with surgery only, the recurrence ratio was significantly higher in the ctDNA positive groups than in the ctDNA negative groups (42.9% [3/7] vs. 0% [0/5], p = 0.2045) (Figure 5a). For patients received surgery and adjuvant therapy, the recurrence ratio was significantly higher in the ctDNA positive (detected ctDNA at least one time point) groups than in the ctDNA negative groups (45.5% [10/22] vs. 4.0% [1/24], p = 0.0014) (Figure 5a). For the 46 patients who received surgery and adjuvant therapy, ctDNA analysis could stratify patients before and after adjuvant therapy. A total of 43 patients who received adjuvant therapy had their ctDNA analyzed before adjuvant therapy. The recurrence ratio was 28.8% (4/14) in patients with positive ctDNA and 20.7% (6/29) in patients with negative ctDNA before adjuvant therapy (Figure 5b). For ctDNA analysis after adjuvant therapy, no patients (0/15) with negative ctDNA had disease‐recurrence while 80.0% (4/5) patients with positive ctDNA experienced recurrence (Figure 5b). All five patients with clearance ctDNA by adjuvant therapy remained disease free, and the ctDNA clearance was associated with clinical benefits of adjuvant therapy.
FIGURE 5.
Analysis of ctDNA during adjuvant therapy. (a) Comparison of overall relapse proportion between ctDNA negative and ctDNA positive groups during postoperative surveillance. (b) Recurrence ratio in patients with negative and positive ctDNA before and after adjuvant therapy.
In a standard clinical setting, patients with stage IB NSCLC who have high risk factors based on their clinicopathological characteristics, as well as stage II–III NSCLC patients, would commonly receive adjuvant therapies following radical surgery, which only yielded improved survival in a small fraction. 25 This study suggests that the combination of clinical and pathological features with ctDNA detection can further refine the perioperative treatment of stage IB–III NSCLC. We explored the possibility of patients benefiting from the intervention therapy. Among the 58 patients, 14 patients who underwent surgery and adjuvant chemotherapy in combination with or without immunotherapy, showed negative ctDNA tests at baseline, 1 month (±7 days) after surgery. During the MRD monitoring process, we analyzed the recurrent events in these patients. Four patients with adjuvant chemotherapy after surgery showed ctDNA positive during or after adjuvant therapy, and all of them relapsed. However, only one of the patients remaining negative ctDNA experienced recurrence. The two patients with adjuvant chemotherapy in combination with immunotherapy showed positive ctDNA tests at baseline, and showed ctDNA clearance during the maintenance therapy. One died of immune pneumonia, and the other patient did not relapse after 36 months of follow‐up. In addition, among the 25 patients who received adjuvant targeted therapy, 17 patients showed negative ctDNA at baseline, 1 month (±7 days) after surgery, and none of the 12 patients with persistent negative MRD recurred. One of the patients was treated with Zadaxin after ctDNA conversion to positive during the targeted therapy. The patient subsequently achieved ctDNA clearance, and no recurrence was found up to follow‐up.
Above all, it predicted that the patients with detectable ctDNA may benefit from the intervention therapy. The ctDNA status after surgery may provide predictive value for adjuvant therapy. It is also possible to determine the response to adjuvant therapy by the dynamic changes in ctDNA status during or after treatment.
DISCUSSION
Application value of ctDNA‐based MRD detection in early‐stage NSCLC after radical surgery has been a research hotspot in recent years. 12 , 18 , 19 , 20 , 21 , 22 Researchers have demonstrated that ctDNA is a promising biomarker for noninvasive molecular profiling and can identify MRD and monitor recurrence in early‐stage NSCLC in several studies. 18 , 19 , 20 ctDNA‐based MRD detection can predict the recurrence of NSCLC earlier than post‐treatment imaging monitoring, which provides a valuable basis for the formulation of precise adjuvant therapy. 12 , 21 , 22 As of now, the decision to use adjuvant therapy largely depends on pre‐existing clinical factors and risk stratification. 25 In studies of adjuvant therapy after surgery for stage IB NSCLC, the guidelines differ on whether adjuvant therapy should be given, 25 , 26 , 27 , 28 , 29 while ctDNA can serve as a more accurate predictor to help guide adjuvant therapy after surgery and enable the patient to benefit from the treatment of the disease. 12 At present, ctDNA, a molecular marker as prognostic indicator, is gradually moving from research to clinic. 30 The potential method of predicting recurrence risk and monitoring the effect of adjuvant therapy for early NSCLC based on ctDNA during perioperative times remains to be examined in a real‐world setting in clinical practice. In this study, we assessed the clinical value of ctDNA in prognosing risk of relapse and monitoring the effect of adjuvant therapy for patients with surgical NSCLC in a real‐world setting.
To explore the correlation between ctDNA status and clinicopathological and molecular features, we analyzed parameters including the age, sex, smoking history, histology, pTNM stage, T stage, lymph node metastasis, TMB, as well as number of somatic mutations. Postoperative ctDNA positivity at 1 month was associated with more advanced tumor stage and T stage. Multivariable analysis indicated that pTNM stage were independent predictors of postoperative ctDNA detection at 1 month. Results showed that the other clinicopathological factors at baseline were not associated with the postoperative ctDNA status at 1 month. None of the stage III patients remaining negative ctDNA relapsed. It suggested that ctDNA may be a better predictor of recurrence than stage. By analyzing the significantly mutated genes in this cohort, we found that harboring KEAP1 or GRIN2A mutations was significantly associated with the ctDNA status. Previous studies have shown that KEAP1 mutations are associated with poor survival in metastatic NSCLC and TMB was significantly associated with mutations in genes like KEAP1 and GRIN2A. 31 , 32 Whether these two ctDNA positivity‐related gene mutations associations with higher TMB make postoperative ctDNA positivity more likely is a question worth exploring further. Further identification of which gene mutations or molecular features are more likely to result in ctDNA positivity after surgery could benefit patients with different molecular features. For patients with stage I and II NSCLC, the malignant pathology was also analyzed in our study. The results showed that bronchial wall invasion, pleural invasion and vessel carcinoma embolus were unrelated to the ctDNA status, and micropapillary structures were more common in ctDNA positive patients. It has been hypothesized that the polar‐inverted micropapillary structure contributes to the secretion by tumor cells of cytokines that act on the stroma and blood vessels, such as metalloproteinase, and predispose tumor cells to infiltrate and spread. 33 , 34 The potentiality of micropapillary pattern to occur lymph node metastasis and to infiltrate lymphatic vessels and small veins may also be related to the polar disappearance of tumor cells. 33 , 34 The characteristic of being prone to metastasis of micropapillary pattern may also predispose NSCLC patients with micropapillary structures to MRD residues leading to ctDNA positivity after surgery. In patients with stage IB NSCLC, whether presence of micropapillary structures or ctDNA positivity after surgery is an independent risk factor for predicting prognosis and guiding adjuvant therapy should be further explored. More prospective clinical studies are also needed to confirm whether patients with stage IB NSCLC, or even stage IA, with only micropapillary components can benefit from adjuvant therapy.
Consistent with the results of previous studies, we observed that patients with ctDNA detection during postoperative surveillance had a significantly higher recurrence rate than those in patients without ctDNA detection. The negative predictive value and positive predictive value of dynamic monitoring are higher than that of single point. Serial ctDNA detection revealed disease recurrence ahead of radiologic imaging by a median of 10.30 months. Compared with patients without MRD detection at landmark and longitudinal, MRD positive patients had a markedly reduced DFS. Patients with postoperative ctDNA detection at 1 month had a higher recurrence rate than those in patients without ctDNA detection, but there was no significant difference. This may be due to postoperative treatment and dynamic changes in ctDNA will affect the prognosis. Patients with positive postoperative ctDNA at 1 month became ctDNA‐negative after adjuvant therapy, with no subsequent recurrence. In patients with postoperative negative ctDNA at 1 month who experienced disease recurrence, the majority (86%, 6/7) had ctDNA detected in subsequent surveillance. It shows that continuous multiple ctDNA monitoring is very important for predicting the prognosis of patients.
Currently, risk stratification of stage and other clinical factors are vital in the decision making for adjuvant therapy in NSCLC. However, a large proportion of patients will be adversely affected by the toxic side effects of long‐term treatment and feel psychologically burdened. There is an urgent need for an effective biomarker to distinguish people who need adjuvant therapy. The predictive value of ctDNA status as an independent risk factor for recurrence in adjuvant therapy has been explored in recent studies. The predictive value of ctDNA for adjuvant therapy has also been preliminarily explored in this real‐world study. We found that longitudinal ctDNA monitoring during or after adjuvant therapy provides a measurement of the response to adjuvant therapy. For patients with detectable ctDNA, application of adjuvant therapy has a certain degree of clinical benefit. For patients with undetectable ctDNA, 95.5% (21/22) patients who received adjuvant therapies remained relapse‐free, while all five patients not receiving adjuvant therapy remained relapse‐free at the end of follow‐up. It may indicate that it is nonessential to receive adjuvant therapy for patients with undetectable ctDNA. The results of this study show the ctDNA status changes during postoperative adjuvant therapy or follow‐up. Regardless of adjuvant therapy, all patients who were persistently positive for ctDNA during postoperative surveillance had disease recurrence, and none of the patients with negative ctDNA during postoperative surveillance had recurrence, except one who might have harbored nonshedder clinicopathological features for ctDNA detection. For patients who became negative during the course of adjuvant therapy, only one patient with stage IIIB relapsed, and of patients who had more than two negative episodes and did not revert to positivity, none relapsed. Adjuvant therapy can eliminate MRD and potentially improve clinical outcomes for patients with detectable ctDNA. All patients with eliminated ctDNA after adjuvant therapy and remained negative after adjuvant therapy remained disease free. In the course of treatment, ctDNA monitoring may provide more dynamic prognostic value than single ctDNA detection. The case of P050 illustrates the importance of dynamic monitoring. In patients who become positive during ctDNA monitoring, intensive treatment should be considered. More prospective clinical trials should elucidate the best clinical options of improving clinical outcomes for patients with worsening ctDNA profiles during adjuvant therapy.
The limitations of the present study included the small number of samples. Although the value of postoperative ctDNA in indicating the recurrence risk and monitoring the effect of adjuvant therapy in surgical NSCLC patients was confirmed with outstanding statistical power, only a total of 14 patients experienced recurrence in this study. The conclusion of this study is that more studies with larger sample sizes need to be performed to further validate the results of the present study. Furthermore, adjuvant therapy can potentially improve the clinical outcomes for patients with detectable ctDNA obtained in this study. However, this is just an observation. More prospective clinical trials are needed to further elucidate the predictive value of ctDNA for adjuvant therapy.
In conclusion, the results of this real‐world study confirmed that postoperative ctDNA is a prognostic marker for patients with stage I to IIIA NSCLC, which stratify patients with high risk of recurrence and reveals disease recurrence ahead of radiographic examination. Detection of ctDNA can be used as predictive factors to aid the selection of eligible NSCLC patients for adjuvant therapies. In addition, ctDNA monitoring in eligible NSCLC patients after definite surgical resection can reveals the effectiveness of adjuvant therapy.
AUTHOR CONTRIBUTIONS
Yi Zhang, Xiaoru Tian and Xingsheng Liu contributed to the study design. Yi Zhang, Xiaoru Tian, Xingsheng Liu, Ruotian Wang, Yuanbo Li, Kun Qian, Tengteng Wang, Xin Zhao, Lei Liu and Pei Long Zhang contributed to the collection and assembly of data. Kai Wang, Yuanyuan Xiong, Rongrong Chen and Jinqiu Rui performed the statistical analyses and interpretation and drafted the manuscript. Yi Zhang and Xiaoru Tian revised the manuscript. All authors contributed to the critical revision of the manuscript and approved its final version. Financial support and study supervision were provided by Yi Zhang, Xiaoru Tian and Xingsheng Liu.
CONFLICT OF INTEREST STATEMENT
K.W., Y.X., R.C. and J.R. are employees of Geneplus‐Beijing. The other authors have no conflicts of interest to declare.
Supporting information
Table S1. List of target regions of 1021 cancer‐related genes.
Table S2. List of target regions of 338 cancer‐related genes.
ACKNOWLEDGMENTS
The authors thank the patients who participated in the trial and their families, as well as the investigators, caregivers, study coordinators, and operations staff.
Tian X, Liu X, Wang K, Wang R, Li Y, Qian K, et al. Postoperative ctDNA in indicating the recurrence risk and monitoring the effect of adjuvant therapy in surgical non‐small cell lung cancer. Thorac Cancer. 2024;15(10):797–807. 10.1111/1759-7714.15251
REFERENCES
- 1. Bray F, Ferlay J, Soerjomataram I, Siegel R, Torre L, Jemal A. Global cancer statistics 2018: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA‐Cancer J Clin. 2018;68(6):394–424. 10.3322/caac.21492 [DOI] [PubMed] [Google Scholar]
- 2. Siegel R, Miller K, Jemal A. Cancer statistics, 2020. Ca‐Cancer J Clin. 2020;70(1):7–30. 10.3322/caac.21590 [DOI] [PubMed] [Google Scholar]
- 3. Watanabe K, Tsuboi M, Sakamaki K, Nishii T, Yamamoto T, Nagashima T, et al. Postoperative follow‐up strategy based on recurrence dynamics for non‐small‐cell lung cancer. Eur J Cardiothorac Surg. 2016;49(6):1624–1631. 10.1093/ejcts/ezv462 [DOI] [PubMed] [Google Scholar]
- 4. Postmus P, Kerr K, Oudkerk M, Senan S, Waller DA, Vansteenkiste J, et al. Early and locally advanced non‐small‐cell lung cancer (NSCLC): ESMO clinical practice guidelines for diagnosis, treatment and follow‐up. Ann Oncol. 2017;28:iv1–iv21. 10.1093/annonc/mdx222 [DOI] [PubMed] [Google Scholar]
- 5. Uramoto H, Tanaka F. Recurrence after surgery in patients with NSCLC. Transl Lung Cancer Res. 2014;3(4):242–249. 10.3978/j.issn.2218-6751.2013.12.05 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Carnio S, Novello S, Papotti M, Loiacono M, Scagliotti GV. Prognostic and predictive biomarkers in early stage non‐small cell lung cancer: tumor based approaches including gene signatures. Transl Lung Cancer Res. 2013;2(5):372–381. 10.3978/j.issn.2218-6751.2013.10.05 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Pignon J‐P, Tribodet H, Scagliotti GV, Douillard JY, Shepherd FA, Stephens RJ, et al. Lung adjuvant cisplatin evaluation: a pooled analysis by the LACE collaborative group. J Clin Oncol. 2008;26(21):3552–3559. 10.1200/JCO.2007.13.9030 [DOI] [PubMed] [Google Scholar]
- 8. Burdett S, Rydzewska L, Tierney JF, Fisher DJ. A closer look at the effects of postoperative radiotherapy by stage and nodal status: updated results of an individual participant data meta‐analysis in non‐small‐cell lung cancer. Lung Cancer. 2013;80(3):350–352. 10.1016/j.lungcan.2013.02.005 [DOI] [PubMed] [Google Scholar]
- 9. Le Pechoux C, Pourel N, Barlesi F, et al. LBA3_PR an international randomized trial, comparing post‐operative conformal radiotherapy (PORT) to no PORT, in patients with completely resected non‐small cell lung cancer (NSCLC) and mediastinal N2 involvement: primary end‐point analysis of LungART (IFCT‐0503, UK NCRI, SAKK) NCT00410683. Ann Oncol. 2020;31:S1178. 10.1016/j.annonc.2020.08.2280 [DOI] [Google Scholar]
- 10. Wu Y‐L, Tsuboi M, He J, John T, Grohe C, Majem M, et al. Osimertinib in resected EGFR‐mutated non–small‐cell lung cancer. N Engl J Med. 2020;383(18):1711–1723. 10.1056/NEJMoa2027071 [DOI] [PubMed] [Google Scholar]
- 11. Felip E, Altorki N, Zhou C, Csőszi T, Vynnychenko I, Goloborodko O, et al. Adjuvant atezolizumab after adjuvant chemotherapy in resected stage IB–IIIA non‐small‐cell lung cancer (IMpower010): a randomised, multicentre, open‐label, phase 3 trial. Lancet. 2021;398(10308):1344–1357. 10.1016/S0140-6736(21)02098-5 [DOI] [PubMed] [Google Scholar]
- 12. Zhang J‐T, Liu S‐Y, Gao W, Liu SYM, Yan HH, Ji L, et al. Longitudinal undetectable molecular residual disease defines potentially cured population in localized non‐small cell lung cancer. Cancer Discovery. 2022;12:1690–1701. 10.1158/2159-8290.cd-21-1486 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Pantel K, Alix‐Panabières C. Liquid biopsy and minimal residual disease—latest advances and implications for cure. Nat Rev Clin Oncol. 2019;16(7):409–424. 10.1038/s41571-019-0187-3 [DOI] [PubMed] [Google Scholar]
- 14. Yang J, Gong Y, Lam VK, Shi Y, Guan Y, Zhang Y, et al. Deep sequencing of circulating tumor DNA detects molecular residual disease and predicts recurrence in gastric cancer. Cell Death Dis. 2020;11(5):1–9. 10.1038/s41419-020-2531-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Coombes RC, Page K, Salari R, Hastings RK, Armstrong A, Ahmed S, et al. Personalized detection of circulating tumor DNA antedates breast cancer metastatic recurrence personalized ctDNA detection of breast cancer recurrence. Clin Cancer Res. 2019;25(14):4255–4263. 10.1158/1078-0432.CCR-18-3663 [DOI] [PubMed] [Google Scholar]
- 16. Reinert T, Henriksen TV, Christensen E, Sharma S, Salari R, Sethi H, et al. Analysis of plasma cell‐free DNA by ultradeep sequencing in patients with stages I to III colorectal cancer. JAMA Oncol. 2019;5(8):1124–1131. 10.1001/jamaoncol.2019.0528 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Christensen E, Birkenkamp‐Demtröder K, Sethi H, Shchegrova S, Salari R, Nordentoft I, et al. Early detection of metastatic relapse and monitoring of therapeutic efficacy by ultra‐deep sequencing of plasma cell‐free DNA in patients with urothelial bladder carcinoma. Cancer Res. 2019;79(13):913. 10.1158/1538-7445.AM2019-913 [DOI] [PubMed] [Google Scholar]
- 18. Abbosh C, Birkbak NJ, Wilson GA, et al. Phylogenetic ctDNA analysis depicts early‐stage lung cancer evolution. Nature. 2017;545(7655):446–451. 10.1038/nature22364 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Chen K, Zhao H, Shi Y, Yang F, Wang LT, Kang G, et al. Perioperative dynamic changes in circulating tumor DNA in patients with lung cancer (DYNAMIC) perioperative dynamic changes in circulating tumor DNA. Clin Cancer Res. 2019;25(23):7058–7067. 10.1158/1078-0432.CCR-19-1213 [DOI] [PubMed] [Google Scholar]
- 20. Moding EJ, Liu Y, Nabet BY, Chabon JJ, Chaudhuri AA, Hui AB, et al. Circulating tumor DNA dynamics predict benefit from consolidation immunotherapy in locally advanced non‐small‐cell lung cancer. Nat Cancer. 2020;1(2):176–183. 10.1038/s43018-019-0011-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Xia L, Mei J, Kang R, Deng S, Chen Y, Yang Y, et al. Perioperative ctDNA‐based molecular residual disease detection for non–small cell lung cancer: a prospective multicenter cohort study (LUNGCA‐1). Clin Cancer Res. 2021;28:3308–3317. 10.1158/1078-0432.CCR-21-3044 [DOI] [PubMed] [Google Scholar]
- 22. Qiu B, Guo W, Zhang F, Lv F, Ji Y, Peng Y, et al. Dynamic recurrence risk and adjuvant chemotherapy benefit prediction by ctDNA in resected NSCLC. Nat Commun. 2021;12(1):1–11. 10.1038/s41467-021-27022-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Lin G, Li C, Li P, et al. Genomic origin and EGFR‐TKI treatments of pulmonary adenosquamous carcinoma. Ann Oncol. 2020;31(4):517–524. 10.1016/j.annonc.2020.01.014 [DOI] [PubMed] [Google Scholar]
- 24. Chatziandreou I, Tsioli P, Sakellariou S, Mourkioti I, Giannopoulou I, Levidou G, et al. Comprehensive molecular analysis of NSCLC; clinicopathological associations. PLoS One. 2015;10(7):e0133859. 10.1371/journal.pone.0133859 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Ettinger D, Wood D, Aisner D, et al. Non‐small cell lung cancer, version 3.2022, NCCN clinical practice guidelines in oncology. J Natl Compr Cancer Network. 2022;20(5):497–530. 10.6004/jnccn.2022.0025 [DOI] [PubMed] [Google Scholar]
- 26. Burdett S, Pignon JP, Tierney J, Tribodet H, Stewart L, le Pechoux C, et al. Adjuvant chemotherapy for resected early‐stage non‐small cell lung cancer. Cochrane Database Syst Rev. 2015;2015(3):CD011430. 10.1002/14651858.CD011430 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Butts CA, Ding K, Seymour L, Twumasi‐Ankrah P, Graham B, Gandara D, et al. Randomized phase III trial of vinorelbine plus cisplatin compared with observation in completely resected stage IB and II non–small‐cell lung cancer: updated survival analysis of JBR‐10. J Clin Oncol. 2010;28(1):29–34. 10.1200/JCO.2009.24.0333 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Douillard J‐Y, Rosell R, De Lena M, et al. Adjuvant vinorelbine plus cisplatin versus observation in patients with completely resected stage IB–IIIA non‐small‐cell lung cancer (adjuvant Navelbine international Trialist association [ANITA]): a randomised controlled trial. Lancet Oncol. 2006;7(9):719–727. 10.1016/S1470-2045(06)70804-X [DOI] [PubMed] [Google Scholar]
- 29. Tu Z, Tian T, Chen Q, Li C. Overall survival analyses following adjuvant chemotherapy or nonadjuvant chemotherapy in patients with stage IB non‐small‐cell lung cancer. J Oncol. 2021;2021:8052752. 10.1155/2021/8052752 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Sanz‐Garcia E, Zhao E, Bratman SV, Siu LL. Monitoring and adapting cancer treatment using circulating tumor DNA kinetics: current research, opportunities, and challenges. Sci Adv. 2022;8(4):eabi8618. 10.1126/sciadv.abi8618 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. de Lima VCC, Corassa M, Saldanha E, et al. STK11 and KEAP1 mutations in non‐small cell lung cancer patients: descriptive analysis and prognostic value among Hispanics (STRIKE registry‐CLICaP). Lung Cancer. 2022;170:114–121. 10.1016/j.lungcan.2022.06.010 [DOI] [PubMed] [Google Scholar]
- 32. Qiu Y, Liu L, Yang H, Chen H, Deng Q, Xiao D, et al. Integrating histologic and genomic characteristics to predict tumor mutation burden of early‐stage non‐small‐cell lung cancer. Front Oncol. 2021;10:608989. 10.3389/fonc.2020.608989 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Nassar H. Carcinomas with micropapillary morphology: clinical significance and current concepts. Adv Anat Pathol. 2004;11(6):297–303. 10.1097/01.pap.0000138142.26882 [DOI] [PubMed] [Google Scholar]
- 34. Zhang J, Liang Z‐y, Liu T‐h. Advance in pulmonary adenocarcinoma with micropapillary pattern. Chin J Pathol. 2011;40(3):202–205. 10.3760/cma.j.issn.0529-5807.2011.03.020 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1. List of target regions of 1021 cancer‐related genes.
Table S2. List of target regions of 338 cancer‐related genes.