Purpose: No study has investigated the precise perioperative dynamic changes in circulating tumor DNA (ctDNA) in any patients with early-stage cancer. This study (DYNAMIC) investigated perioperative dynamic changes in ctDNA and determined the appropriate detection time of ctDNA-based surveillance for surgical patients with lung cancer. Experimental Design: Consecutive patients who underwent curative-intent lung resections were enrolled prospectively (NCT02965391). Plasma samples were obtained at multiple prespecified time points including before surgery (time A), during surgery after tumor resection (time B–time D), and after surgery (time P1–time P3). Next-generation sequencing–based detection platform was performed to calculate the plasma mutation allele frequency. The primary endpoint was ctDNA half-life after radical tumor resection. Results: Thirty-six patients showed detectable mutations in time A. The plasma ctDNA concentration showed a rapid decreasing trend after radical tumor resection, with the average mutant allele fraction at times A, B, C, and D being 2.72%, 2.11%, 1.14%, and 0.17%, respectively. The median ctDNA half-life was 35.0 minutes. Patients with minimal residual disease (MRD) detection had a significant slower ctDNA half-life than those with negative MRD (103.2 minutes vs. 29.7 minutes, P = 0.001). The recurrence-free survival of patients with detectable and undetectable ctDNA concentrations at time P1 was 528 days and 543 days, respectively (P = 0.657), whereas at time P2 was 278 days and 637 days, respectively (P = 0.002). Conclusions: ctDNA decays rapidly after radical tumor resection. The ctDNA detection on the third day after R0 resection can be used as the baseline value for postoperative lung cancer surveillance.

Purpose: Nodule evaluation is challenging and critical to diagnose multiple pulmonary nodules (MPNs). We aimed to develop and validate a machine learning–based model to estimate the malignant probability of MPNs to guide decision-making. Experimental Design: A boosted ensemble algorithm (XGBoost) was used to predict malignancy using the clinicoradiologic variables of 1,739 nodules from 520 patients with MPNs at a Chinese center. The model (PKU-M model) was trained using 10-fold cross-validation in which hyperparameters were selected and fine-tuned. The model was validated and compared with solitary pulmonary nodule (SPN) models, clinicians, and a computer-aided diagnosis (CADx) system in an independent transnational cohort and a prospective multicentric cohort. Results: The PKU-M model showed excellent discrimination [area under the curve; AUC (95% confidence interval (95% CI)), 0.909 (0.854–0.946)] and calibration (Brier score, 0.122) in the development cohort. External validation (583 nodules) revealed that the AUC of the PKU-M model was 0.890 (0.859–0.916), higher than those of the Brock model [0.806 (0.771–0.838)], PKU model [0.780 (0.743–0.817)], Mayo model [0.739 (0.697–0.776)], and VA model [0.682 (0.640–0.722)]. Prospective comparison (200 nodules) showed that the AUC of the PKU-M model [0.871 (0.815–0.915)] was higher than that of surgeons [0.790 (0.711–0.852), 0.741 (0.662–0.804), and 0.727 (0.650–0.788)], radiologist [0.748 (0.671–0.814)], and the CADx system [0.757 (0.682–0.818)]. Furthermore, the model outperformed the clinicians with an increase of 14.3% in sensitivity and 7.8% in specificity. Conclusions: After its development using machine learning algorithms, validation using transnational multicentric cohorts, and prospective comparison with clinicians and the CADx system, this novel prediction model for MPNs presented solid performance as a convenient reference to help decision-making.

Background: Bariatric surgery is an effective approach to weight loss, which may also affect thyroid function. However, alteration in thyroid-stimulating hormone (ΔTSH) and thyroid hormones after bariatric surgery and the relationship between thyroid function and postoperative weight loss still remains controversial.Methods: Data were collected from euthyroid patients with obesity who underwent sleeve gastrectomy and Roux-en-Y gastric bypass from 2017 to 2022. The alterations of free thyroxine (FT4), free triiodothyronine (FT3), total thyroxine (TT4), total triiodothyronine (TT3), and TSH were calculated 1 year after surgery. Pearson correlation analysis was used to assess the correlation between the percentage of total weight loss (%TWL) and ΔTSH. Multivariable linear regression was utilized to determine the association between %TWL and ΔTSH.Results: A total of 256 patients were included in our study. The mean %TWL was 28.29% after 1 year. TSH decreased from 2.33 (1.67, 3.04) uIU/mL to 1.82 (1.21, 2.50) uIU/mL (P < 0.001), FT3 decreased from 3.23 ± 0.42 pg/mL to 2.89 ± 0.41 pg/mL (P < 0.001), FT4 decreased from 1.11 ± 0.25 ng/dL to 1.02 ± 0.25 ng/dL (P < 0.001), TT3 decreased from 1.13 (1.00, 1.25) ng/mL to 0.89 (0.78, 1.00) ng/mL (P < 0.001), and TT4 decreased from 8.28 ± 1.69 ug/mL to 7.82 ± 1.68 ug/mL 1 year postoperatively (P < 0.001). %TWL was found to be significantly correlated to ΔTSH by Pearson correlation analysis (Pearson correlation coefficient = 0.184, P = 0.003), indicating that the more weight loss, the more TSH declined. After adjusting for covariates in multivariable linear regression, %TWL was found to be independently associated with ΔTSH (β = 0.180 [95% confidence interval (CI), 0.048 – 0.312], P = 0.008). Moreover, %TWL was divided into 3 categorical groups (%TWL ≤ 25%, 25% < %TWL ≤ 35%, and %TWL > 35%) for further exploration, and was also found to be an independent predictor for ΔTSH after adjusting for covariates in multivariable linear regression (β = 0.153 [95% CI, 0.019 – 0.287], P = 0.025).Conclusion: TSH, FT4, FT3, TT4, and TT3 decrease significantly 1 year after bariatric surgery. The decline in TSH is independently mediated by postoperative weight loss; the more the weight loss, the more the TSH decrease.

Stage I non-small cell lung cancer (NSCLC) presents diverse outcomes. To identify molecular features leading to tumor recurrence in early-stage NSCLC, we perform multiregional whole-exome sequencing (WES), RNA sequencing, and plasma-targeted circulating tumor DNA (ctDNA) detection analysis between recurrent and recurrent-free stage I NSCLC patients (CHN-P cohort) who had undergone R0 resection with a median 5-year follow-up time. Integrated analysis indicates that the multidimensional clinical and genomic model can stratify the prognosis of stage I NSCLC in both CHN-P and EUR-T cohorts and correlates with positive pre-surgical deep next generation sequencing (NGS) ctDNA detection. Increased genomic instability related to DNA interstrand crosslinks and double-strand break repair processes is significantly associated with early tumor relapse. This study reveals important molecular insights into stage I NSCLC and may inform clinical postoperative treatment and follow-up strategies.