Currently, colorectal cancer is still a leading cause of cancer-related death worldwide. Surgery remains mainstay of treatment for colorectal cancer, but for non-resectable tumors, chemotherapy, and targeted therapy are mostly used. An example of the targeted therapy for colorectal cancer is anti-epithelial growth factor receptor (EGFR) therapy, e.g., cetuximab and panitumumab, which have been used for the treatment of metastatic colorectal cancer for more than 15 years. However, those targeted therapies were plaqued by drug resistance. For example, somatic mutations of KRAS gene in tumor can cause resistance to anti-EGFR therapy, which makes it necessary to test KRAS mutation status before the therapy is given.
The detection of KRAS mutation in colorectal cancer is mostly performed on tumor tissue sample, but for recurrent or metastatic colorectal cancer patients whose tumor tissue samples are not available, liquid biopsy sample (e.g., plasma, urine, etc.) serves as an alternative. In addition, liquid biopsy is a non-invasive approach in cancer genotyping and also could better indicate tumor heterogeneity.[6,7] Using cell-free DNA extracted from liquid biopsy samples, KRAS mutation status can be determined using several techniques, including digital PCR, amplification refractory mutation system (ARMS), and next generation sequencing (NGS).[8–11]
The primary objective of this study is to assess the accuracy of detecting KRAS mutation status using cell-free DNA in liquid biopsy samples compared to tissue samples. In addition, we also plan to compare the diagnostic accuracy between different detecting methods, including PCR, ARMS, and NGS. The results could guide the use of liquid biopsy in KRAS mutation detection in colorectal cancer. We have performed a thorough search on Pubmed, Embase, Cochrane Library, and PROSPERO, and did not find any other meta-analysis performed on this topic.
2 Methods and analysis
2.1 Study registration
This study protocol has been registered on PROSPERO (Registration number: CRD42020176682).
2.2 Research question development
Research questions were developed following the PICO framework. Please find details in Table 1.
2.3 Eligibility criteria
All original studies describing accuracy of KRAS mutation detection in cell-free DNA of patients with colorectal cancer using digital PCR, ARMS, or NGS, or a comparison among those techniques, with tissue samples as reference (gold standard).
- 1. not a human study;
- 2. not describing KRAS mutation;
- 3. no liquid biopsy samples or tissue samples included;
- 4. did not use any techniques among digital PCR, ARMS, and NGS;
- 5. not colorectal cancer;
- 6. reviews, abstracts, letter to the editor, comments, case reports, or studies with uninterpretable data.
2.4 Information source
Pubmed, Embase, and Cochrane Library databases will be searched for eligible studies. No limitation will be applied.
2.5 Searching strategy
Searching will be performed using keywords “KRAS”, “digital PCR”, “NGS”, “next generation sequencing”, “ARMS”, “amplification refractory mutation system”, “circulating tumour DNA”, “cell-free DNA”, “liquid biopsy” and “colorectal cancer”. Please see Table 2 for details of searching strategy.
2.6 Study selection
Eligible studies will be independently searched and screened by 2 researchers (PY and PC). Any disagreement between the 2 researchers will be resolved by a third researcher (YW). Number of excluded studies will be shown in PRISMA flowchart and reasons of exclusion will be provided, as indicated in Figure 1.
2.7 Data management
After literature search in online databases, list of the searching results will be recorded by the 2 researchers (PY and PC) and sent to a third researcher (YW). After eligible studies are finalized, full-text of the studies will be downloaded. Data will be extracted using a data extraction table which will be uploaded to Systematic Review Data Repository (SRDR) for record.
2.8 Data extraction and collection
Full text of eligible studies will be downloaded and information will be independently extracted by PY and PC using a data extraction table prepared before the information extraction.
2.9 Collected data items
After list of eligible studies is finalized, the following information will be collected: author information (name of first author), publication year, characteristics of patients (age, race), testing platform for KRAS mutation in liquid biopsy, and tissue samples (digital PCR, ARMS or NGS), type of liquid biopsy samples (plasma, serum, urine, cerebrospinal fluid, and etc.), sample size, numbers of true positive, false positive, false negative, and true negative.
2.10 Study outcomes
The primary study outcome will be diagnostic accuracy of detecting KRAS mutation in cell-free DNA, with KRAS mutation status in the paired tissue biopsy as control. The parameters of diagnostic accuracy evaluated in this meta-analysis will include sensitivity, specificity, positive likelihood ratio (PLR), negative likelihood ratio (NLR), diagnostic odds ratio (DOR), the summary receiver operating characteristic (SROC) curve, and area under curve (AUC). The secondary study outcome will be a comparison between the diagnostic accuracy of digital PCR, ARMS, and NGS in detecting KRAS mutation in cell-free DNA.
2.11 Incomplete information and missing data
During the data extraction step, if we find any incomplete or missing information, we will try to contact the author for help. If we fail to obtain those data, the study will be excluded from the final data synthesis.
2.12 Risk of bias in individual study
Quality assessment of diagnostic accuracy studies 2 (QUADAS-2) will be used to evaluate each eligible study, which will be independently performed by 2 researchers (PY and PC). Disagreement between the 2 researchers will be resolved by YW.
2.13 Statistical analysis and data synthesis
Statistical analysis will be performed using STATA software with MIDAS module and Meta-Disc software version 1.4. Pooled values will be calculated for sensitivity, specificity, PLR, and NLR. DOR will be calculated by PLR divided by NLR. The SROC curve will be generated and AUC will be calculated. Cochrans Q and Thompson I2 test will be used to examine inter-study heterogeneity. Based on the results of heterogeneity test, fixed-effects model will be used if no significant heterogeneity is detected (I2 ≤ 50%); otherwise, random-effects model will be used (I2 > 50%).
2.14 Subgroup analysis
We plan to perform subgroup analysis on the testing platform for KRAS mutation in liquid biopsy (e.g., digital PCR vs ARMS vs NGS), and on metastatic and non-metastatic colorectal cancer, if feasible. In case of significant inter-study heterogeneity, we will try to find possible sources of heterogeneity and perform subgroup analysis if possible.
2.15 Publication bias
Begg funnel plot and Egger test will be used to evaluate publication bias.
2.16 Confidence in cumulative evidence
Confidence in cumulative evidence will be evaluated following GRADE guideline. Imprecision will be evaluated using sample size and confidence interval of outcomes. Inconsistency will be evaluated by Thompson I2 test as described in Section 2.13. Indirectness will be evaluated using the PICO information from the eligible studies. Publication bias will be evaluated as described in Section 2.15.
In the era of precision medicine, precise cancer genotyping is very important for the success of targeted therapies. Cancer genotyping in clinical practice is mostly performed using tumor tissue sample (referred as “gold standard”), which includes surgically-resected and biopsy tumor samples. However, the procedure of obtaining tumor tissue sample is invasive and results based on tumor tissue sample could be biased due to tumor heterogeneity.[13–15] Liquid biopsy sample has been intensively investigated for its use as a surrogate of tissue sample in cancer genotyping since its non-invasiveness and better presentation of tumor heterogeneity.[16–18] However, its accuracy and reliability need to be proven. In this study, we propose a protocol for a systematic review and meta-analysis on the accuracy of KRAS mutation detection in colorectal cancer using liquid biopsy sample, with paired tissue sample as control. We hope the results of this study could be used as a reference for the future use of liquid biopsy in KRAS mutation detection in colorectal cancer by clinicians and researchers.
Conceptualization: Peng Ye, Yuanyuan Wei.
Data curation: Peng Ye, Peiling Cai.
Funding acquisition: Yuanyuan Wei.
Methodology: Peng Ye, Jing Xie.
Project administration: Yuanyuan Wei.
Resources: Peng Ye, Peiling Cai.
Supervision: Yuanyuan Wei.
Writing – original draft: Peng Ye.
Writing – review & editing: Peiling Cai, Jing Xie, Yuanyuan Wei.
. Torre LA, Bray F, Siegel RL, et al. Global cancer statistics, 2012. CA 2015;65:87–108.
. Koulis C, Yap R, Engel R, et al. Personalized medicine-current and emerging predictive and prognostic biomarkers in colorectal cancer
. Cancers 2020;12:812.
. Troiani T, Napolitano S, Della Corte CM, et al. Therapeutic value of EGFR inhibition in CRC and NSCLC: 15 years of clinical evidence. ESMO Open 2016;1:e000088.
. Van Cutsem E, Cervantes A, Adam R, et al. ESMO consensus guidelines for the management of patients with metastatic colorectal cancer
. Ann Oncol 2016;27:1386–422.
. 2020;Harle A. Cell-Free DNA
in the management of colorectal cancer
. recent results in cancer research. 215:253–61.
. Mader S, Pantel K. Liquid biopsy: current status and future perspectives. Oncol Res Treat 2017;40:404–8.
. Poulet G, Massias J, Taly V. Liquid biopsy: general concepts. Acta Cytol 2019;63:449–55.
. Martinelli E, Ciardiello D, Martini G, et al. Implementing anti-epidermal growth factor receptor (EGFR) therapy in metastatic colorectal cancer
: challenges and future perspectives. Ann Oncol 2020;31:30–40.
. Olmedillas Lopez S, Garcia-Olmo DC, Garcia-Arranz M, et al. KRAS
G12 V mutation detection by droplet digital PCR in circulating cell-Free DNA
of colorectal cancer
patients. Int J Molec Sci 2016;17:484.
. Sefrioui D, Mauger F, Leclere L, et al. Comparison of the quantification of KRAS
mutations by digital PCR and E-ice-COLD-PCR in circulating-cell-free DNA
from metastatic colorectal cancer
patients. Clin Chim Acta 2017;465:1–4.
. Yao J, Zang W, Ge Y, et al. RAS/BRAF circulating tumor DNA mutations as a predictor of response to first-line chemotherapy in metastatic colorectal cancer
patients. Can J Gastroenterol Hepatol 2018;2018:4248971.
. Schardt C, Adams MB, Owens T, et al. Utilization of the PICO framework to improve searching PubMed for clinical questions. BMC Med Inform Decis Mak 2007;7:16.
. Ye P, Zhang M, Fan S, et al. Intra-tumoral heterogeneity of HER2, FGFR2, cMET and ATM in Ggstric cancer: optimizing personalized healthcare through innovative pathological and statistical analysis. PloS One 2015;10:e0143207.
. Burrell RA, McGranahan N, Bartek J, et al. The causes and consequences of genetic heterogeneity in cancer evolution. Nature 2013;501:338–45.
. Hiley C, de Bruin EC, McGranahan N, et al. Deciphering intratumor heterogeneity and temporal acquisition of driver events to refine precision medicine. Genome Biol 2014;15:453.
. Yamada T, Matsuda A, Koizumi M, et al. Liquid biopsy for the management of patients with ccolorectal cancer. Digestion 2019;99:39–45.
. Normanno N, Cervantes A, Ciardiello F, et al. The liquid biopsy in the management of colorectal cancer
patients: current applications and future scenarios. Cancer Treat Rev 2018;70:1–8.
. Scripcariu V, Scripcariu DV, Filip B, et al. Liquid Biopsy” - is it a feasible option in colorectal cancer
? Chirurgia (Bucur) 2019;114:162–6.