few_shot.json

{
    "1": {
        "query": "Is anything in the article shared (data, code)? Look for the words like Supporting, Additional, supplementary  information/code/material/datashar -ing/-ed/-e, available, reproducibility and similar  + links, appendix",
        "score": 1,
        "reasoning": "Supplementary materials\n Supplementary material associated with this article can be found in the online version at doi:10.1016/j. ebiom.2022.103945.\n \n Data sharing\n All relevant data are within the paper and its supplementary files. The raw data used and/or analysed during the study are available in the Genome Sequence Archive for Human repository [HRA001933 in https://bigd.big.ac.cn/gsa-human/]."
    },
    "2": {
        "query": "Has anything in the article been registered (in advance)?",
        "score": 1,
        "reasoning": "This study was registered and the inclusion criteria for patients were presented on ClinicalTrials.gov with the number NCT02533271, STELLAR. The primary endpoint was 3-year relapse-free survival, defined as the time from the date of randomization to the first occurrence of local-regional failure or distant metastasis. The secondary objectives were 3-year local relapse-free survival, distant metastasis-free survival, and overall survival."
    },
    "3": {
        "query": "Does the article follow any reporting guidelines? To answer this question, follow the 3 steps sequentially. If any of the steps is true, assign a score of 1 and if all the steps are false, give a score of 0. STEP 1. Look for ISRCTN registry. STEP 2. Look if it is published in either The Lancet, The New England Journal of Medicine (NEJM), Journal of the American Medical Association (JAMA), British Medical Journal (BMJ), Annals of Internal Medicine, Nature Medicine, or Journal of Clinical Oncology. STEP 3. Look for one of the following guidelines, CONSORT for randomized controlled trials, PRISMA for meta\u2010analyses or systematic reviews, MOOSE for Meta-analyses of observational studies, STARD for diagnostic/prognostic studies, ARRIVE for animal pre-clinical studies, STROBE observational studies,SPIRIT for study protocols, CARE for case reports, AGREE for clinical practice guidelines, SRQR for qualitative researches,SQUIRE for quality improvement studies, SPIRIT Statement: Standard Protocol Items: Recommendations for Interventional Trials, PRIMER Collaboration: PRESENTATION AND INTERPRETATION OF MEDICAL RESEARCH, MIBBI: Minimum Information for Biological and Biomedical Investigations, COREQ: Consolidated Criteria for Reporting Qualitative Research, MDAR (Materials Design Analysis Reporting) reproducibility checklist is not a traditional reporting guideline like CONSORT or PRISMA. Instead, it's a tool designed to enhance the reproducibility and transparency of scientific research, REMARK (Reporting Recommendations for Tumor Marker Prognostic Studies).",
        "score": 1,
        "reasoning": "The Lancet"
    },
    "4": {
        "query": "Is the methodology described in detail (where, when, how, what, who)?",
        "score": 1,
        "reasoning": "Methods Sixty patients with LARC from a multicentre, phase II/III randomized trial were included, with tissue and blood samples collected. For each cfDNA sample, we profiled MRD using 3 approaches: personalized assay targeting tumour-informed mutations, universal panel of genes frequently mutated in colorectal cancer (CRC), and low depth sequencing for copy number alterations (CNAs).\n \n Patients enrolled were randomly assigned in a 1:1 ratio\n to short-course preoperative radiotherapy (SCPRT, 5 Gy\n x 5 alone) with neoadjuvant chemotherapy (NCT) (4\n cycles of capecitabine plus oxaliplatin regimen) and preoperative long-course chemoradiotherapy (2 Gy x 25\n with capecitabine). The treatment strategies in these\n two groups were described in detail in STELLAR registration file. \n \n For each patient, we selected up to 22 somatic mutations from the tumour tissue. We designed customized\n primers targeting the mutations and used the primers to profile the matched cfDNA with Mutation Capsule\n technology as previously described. Briefly, the cfDNA was ligated to a customized adaptor and amplified to\n produce a whole genome library that was subsequently used as a template and amplified with customized primers. Multiplex PCR primer pairs for the two rounds of nested amplification were designed using Oligo software (v7.53) and their uniqueness were verified in the human genome (http://genome.ucsc.edu/) to ensure amplification efficiency. In the first round of amplification, the whole genome library was amplified in 9 cycles of PCR using a target-specific primer and a primer matching the adapter sequence. A second round of 14 cycles of amplification was performed with one pair of nested primers matching the adapter and the target region to further enrich the target region and add the Illumina adapter sequences to the construct. The final libraries were sequenced using the Illumina NovaSeq 6000 platform at a median depth of 6835\u00a3 after removing duplicate molecules. The median on-target ratio of reads mapped to the target region was 80%. The clean reads were mapped to the human reference hg19 genome using 'BWA (v0.7.15) mem' with the default parameters. Samtools mpileup was used to identify somatic mutations, including SNVs and INDELs, across the targeted regions of interest. Each uniquely labelled template was amplified, resulting in a certain number of daughter molecules with the same sequence (defined as a UID family). If a mutation is pre-existing in the template molecule (original cfDNA) used for amplification, the mutation should be present in each daughter molecule containing the UID (barring any subsequent replication or sequencing errors). A UID family in which at least 80% of the family members have the same mutation is called the EUID family, indicating that it harbours a mutation that should be true instead of a false-positive mutation due to amplification or sequencing error. The mutant allelic fraction was calculated by dividing the number of alternative EUID families by the sum of alternative and reference families. Tissue-specific mutations with at least one distinct paired duplex EUID family or four distinct EUID families were subsequently manually checked in IGV and verified using a cross-validation method. The candidate mutations were annotated with Ensemble Variant Effect Predictor (VEP)."
    },
    "5": {
        "query": "Is the data collection processes described in detail (where, when, how, what, who)?",
        "score": 1,
        "reasoning": "The tumour tissues were collected at the diagnostic stage by biopsy sampling, and peripheral blood was collected in EDTA Vacutainer tubes (BD Diagnostics; Franklin Lakes, NJ, USA) and centrifuged within 2 h of collection at 4000 \u00a3 g for 10 min to separate plasma and blood cells. Plasma was centrifuged a second time at 12,000 \u00a3 g for 10 min at 4\u00b0C to remove any remaining cellular debris and stored at -80\u00b0C.\n \n Clinical serum levels of the biomarkers carcinoembryonic antigen (CEA) and carbohydrate antigen 19-9 (CA 19-9) were monitored at baseline, before surgery and after surgery. CEA and CA19-9 levels were measured with immunoelectrochemiluminescence, with CEA concentrations of < 5.0 ng/mL and CA19-9 concentrations of < 27.0 U/mL considered within the reference range. Chest/abdominal/pelvic CT scans were performed every 3 months during the first two years and then every 6 months for a total of 5 years. Clinicians were blinded to the ctDNA results during the courses of neoadjuvant therapy.\n \n Genomic DNA (gDNA) was extracted from fresh frozen tumour biopsies and WBCs with the QIAamp DNA Mini Kit (Qiagen; Germantown, MD, USA), and cfDNA was extracted from 1.5-4.5 mL of plasma with the Apostle MiniMax cfDNA isolation kit (C40605, Apostle; San Jose, CA, USA). Targeted sequencing of a panel of 509 genes or exomes was performed using genomic DNA obtained from tumour tissue and WBCs as previously described.\n \n Briefly, the raw data (FASTQ file) were aligned to the UCSC human reference genome hg19 using BurrowsWheeler aligner software (BWA, v0.7.15). Basic processing, marking duplicates, local realignments and score recalibration were performed using The Genome Analysis Toolkit (GATK, v3.6), Picard (v2.7.1) and Samtools (v1.3.1). Candidate somatic mutations were detected by comparing sequencing data from tumour tissue samples with MuTect1 and Strelka. All selected mutations were further validated by performing a manual inspection using Integrated Genome Viewer (IGV).\n \n The raw sequencing data were treated as described above, and the next segmentation analysis was performed using QDNASeq (v1.14.0). The resulting output files were summarized using R software (v4.0.3). Overlap analysis was performed using bedtools (v2.17.0) and plotted with UpSetR (v1.4.0) within the R package (v4.0.3). Chromosome arm-level alterations show cancer-specific patterns. For example, a hierarchical clustering analysis of mean arm-level calls performed across 3,000 TCGA samples revealed that gastrointestinal tumours clustered with gains of chromosomes 8q, 13q, and 20. Some of these CNAs, including gains of chromosomes 1q, 8q, 7,12q, 13q, and 20q and loss of chromosomes 1p, 20p, and 22q, were also recurrently identified in our cohort as hot CNAs (34 baseline plasma samples from patients with LARC compared with 70 plasma samples from healthy controls). Therefore, we defined the CNA number as the sum of hot chromosome arms altered (|Z| > 2) to represent the level of copy number variation."
    },
    "6": {
        "query": "Is there any sample description? eg. size, demographics, recruitment, in-/exclusion criteria",
        "score": 1,
        "reasoning": "Patient characteristics and tissue mutation identification\n Patients with locally advanced rectal cancer (n = 82; cT3- 4N0 or cTanyN1-2) were enrolled in the trial from December 30, 2016, to October 8, 2018. Twenty-two patients were excluded due to the lack of plasma samples obtained after NAT (Figure 1a). Thirty-one patients were treated with long-course neoadjuvant chemoradiotherapy (LCRT), and 29 patients were treated with short-course neoadjuvant radiotherapy (SCPRT) with neoadjuvant chemotherapy (Table 1). The median follow-up period was 33.25 months (range, 9.6342.43 months). Seventeen (28.33%) patients were diagnosed with local relapse or metastasis during follow-up, including 5/17 (29.41%) with local relapse, 6/17 (35.29%) with liver metastasis and 6/17 (35.29%) with lung metastasis (Table S1).\n One hundred ninety-six blood samples were available during the treatment process, including baseline (collected before NAT, n = 42), in-process (collected during NAT, n = 35), post-NAT (collected 2 weeks after SCPRT or LCRT, n = 60) and pre-TME (collected before surgery, n = 59) samples (Figure 1a). We performed targeted sequencing with a panel of 509 genes or exome sequencing on the genomic DNA isolated from the tumour tissue and matched WBCs, and then identified a median of 51 (range, 3-177) somatic mutations in each tumour (Table S2). The mutational landscape of the top 15 most significantly mutated genes in the cohort was shown in Figure 1b. Customized primers were designed to profile up to 22 somatic mutations in the matched cfDNA with Mutation Capsule technology (Table S3) as previously described.\n \n Thirty-five patients with a positive ctDNA fraction at baseline were analysed (35/42 patients) to explore the performance of the ctDNA fraction in monitoring the NAT response. With ctDNA clearance defined as ratio of post-NAT ctDNA fraction to baseline ctDNA fraction below 2% (median value of the ratio), 19 (54.29%) patients showed no clearance at the post-NAT time point relative to baseline ctDNA fraction values (Figures 5, S3b). For patients with or without ctDNA clearance, there were 9/16 (56.25%) and 18/19 (94.74%) exhibited nonpCR/cCR (clinical complete response), respectively."
    },
    "7": {
        "query": "Does the article describe the data analysis process?",
        "score": 1,
        "reasoning": "Statistics\n In this clinical cohort-based investigative study, the primary aim was to test the hypothesis that changes in the ctDNA fraction during treatment dynamically reflect minimal residual disease. Correlation analysis between input and estimated ctDNA in ctDNA fraction model and analysis of variance for the assessment of longitudinal plasma samples were the exploratory studies. Method for hypothesis testing and survival analysis was commonly used by previous researchers. Specifically, correlation analysis used Spearman\u2019s correlation analysis. For continuous variables, differences in ctDNA fractions between recurrence and non-recurrence groups were assessed with MannWhitney (rank sum) test, ctDNA fractions across treatment courses of NAT were assessed by Kruskal-Wallis test and post hoc using Dunn's multiple comparisons test, and the ctDNA fraction was assessed for patients with paired baseline and post-NAT data using Wilcoxon matched-pairs signed rank test. Differences in clinical characteristics between patients with positive and negative ctDNA fractions were evaluated with Fisher\u2019s exact test for categorical variables. These statistical analyses were performed with Prism 8 software (v8.4.3). Relapse-free survival (RFS) was measured from the date of randomization to the first occurrence of local-regional failure or distant metastasis. The univariate analysis was conducted using the KaplanMeier method with the log-rank test. HR values were calculated using univariate Cox proportional hazard models. The multivariate analysis was based on the Cox proportional hazard model in which the common important factors, such as age, sex, and clinical risk (according to the ESMO guidelines) were included. The survival model was evaluated with the C-index. The KaplanMeier curves were verified by performing a time-dependent receiver operating characteristic (ROC) curve analysis, and the area under the curve (AUC) was calculated to evaluate the prognostic performance. These analyses were performed using R software (v4.0.3). P values < 0.05 from a 2-sided test were considered statistically significant in all analyses. A sample of fifty patients was needed to achieve the power of 0.8 in this study as previously described.\n \n We next checked longitudinal status of the ctDNA fraction and its possible association with the disease course, therapeutic effect and survival status of all 60 patients (Figure 4a). Compared with baseline and in-process samples, a clear trend of a reduced post-NAT ctDNA fraction was observed in both the recurrence and nonrecurrence groups (Figure 4b), which highlighted the significant therapeutic effect of NAT. We noticed a more substantial reduction in the ctDNA fraction during baseline, in-process and post-NAT stages within the nonrecurrence group (Dunn\u2019s multiple comparison test, baseline vs. in-process: P = 0.0130; baseline vs. postNAT: P < 0.0001; in-process vs. post-NAT: P = 0.0009) compared to the recurrence group (Dunn\u2019s multiple comparison test, baseline vs. in-process: P > 0.9999; baseline vs. post-NAT: P = 0.1819; in-process vs. post-NAT: P = 0.4114) (KruskalWallis test, nonrecurrence group, P < 0.0001; recurrence group, P = 0.113) (Figure 4b). Moreover, the post-NAT ctDNA fraction status exhibited the strongest association with RFS, followed by the status at the in-process (HR = 3.61; 95% CI, 0.73-17.91; log-rank P = 0.093) and baseline stages (HR = 1.58; 95% CI, 0.20-12.67; log-rank P = 0.66). For the 17 patients experiencing recurrence, the median lead time between the detection of positive post-NAT ctDNA fraction and finding of radiological recurrence was 10.2 months (range, 0.1-33.2 months) (Wilcoxon matched-pairs signed rank test, P = 0.0001) (Figure S3a). We explored whether ctDNA fraction dynamics were linked to RFS by specifically focusing on the 42 patients with both baseline and post-NAT samples and observed a decreased ctDNA fraction in most patients (85.71%, 36/42). For the 9 patients experiencing recurrence, the ctDNA fraction after NAT increased in 4 (44.44%) patients and decreased but was still positive in 4 (44.44%) patients. In the nonrecurrence group (n = 33), the ctDNA fraction decreased to undetectable levels in 30 patients (90.90%) (Figure 4c). These data showed better predictive value of the post-NAT ctDNA fraction status than the ctDNA fraction dynamics (HR = 7.40; 95% CI: 1.97-27.82; log-rank P = 0.00053; sensitivity of 44.44% and specificity of 93.94%) for RFS estimation. The ctDNA fraction (post-NAT) in MRD-positive samples varied significantly from 0.05% to 12.74%. We divided the post-NAT samples into two groups to test if the ctDNA fraction values were correlated with the recurrence status: highly positive ctDNA fraction ( 1%) and moderately positive ctDNA fraction (0.05%-1%). The RFS of the 3 patients with highly positive post-NAT ctDNA fractions was shorter (< 200 days) than that of the moderately positive group (Figure 4d). In patient FL126 with two post-NAT plasma samples, the ctDNA fraction in plasma was moderately positive (0.16%) at 20 days after NAT and highly positive (3.50%) at 141 days, and lung metastases appeared in this patient only 43 days after the second time point (Figure 4e). In patient FL199 with a moderately positive ctDNA fraction (0.23%), local relapse occurred 306 days later (Figure 4e). The dynamic ctDNA fraction in the remaining samples was shown in Figure S4.\n \n The association between ctDNA fraction clearance and response to neoadjuvant therapy was significant (Fisher's exact test, P = 0.013)."
    },
    "8": {
        "query": "Were measures taken to avoid or minimize systematic bias?",
        "score": 1,
        "reasoning": "This study had several limitations. First, the sample size was modest, and a limited number of patients were included in each subgroup, such as longitudinal plasma samples or patients who accepted LCRT/SCPRT. Second, intervention studies are required to explore the potential clinical utility of ctDNA to guide therapeutic decision-making and to determine whether the administration of neoadjuvant chemotherapy under ctDNA guidance may exert a positive effect on survival.\n \n Declaration of interests\n YCJ is one of the cofounders, has owner interest in Genetron Holdings, and receives royalties from Genetron. The other authors have declared that no competing interest exists.\n \n Role of the funding source\n The sponsors did not have any role in the study design, data collection, data analyses, interpretation, or writing of the manuscript.\n \n Funding\n The National Key R&D Program of China, Beijing Municipal Science & Technology Commission, National Natural Science Foundation of China, and CAMS Innovation Fund for Medical Sciences.\n \n Acknowledgements\n The authors would like to thank Ying Zhang for the assistance with sample collection and Pei Wang for the primary technical assistance. This work was supported by financial support were as follows: the National Key R&D Program of China [2021YFC2500900], Beijing Municipal Science & Technology Commission [Z181100001718136], National Natural Science Foundation of China [82073352], and CAMS Innovation Fund for Medical Sciences [2017-I2M-1-006 and 2021-I2M-1- 067]. The sponsors had no role in study design, data collection, data analyses, interpretation, and writing of the manuscript.\n \n Funding The Beijing Municipal Science & Technology Commission, National Natural Science Foundation of China, and CAMS Innovation Fund for Medical Sciences"
    },
    "9": {
        "query": "Has the article been published in a journal?",
        "score": 1,
        "reasoning": "the lancet"
    }
}