Gastrointestinal Oncology
Read Time: 23 mins

Use of Circulating Tumour DNA to Assess Minimal Residual Disease in Gastrointestinal Cancers

Published Online: May 10th 2022 touchREVIEWS in Oncology & Haematology. 2022;18(1):26–39 DOI:
Authors: Reetu Mukherji, Ali Alqahtani, Harrison D Winters, Benjamin A Weinberg
Quick Links:
Article Information

Despite our modern perioperative therapies, many patients with gastrointestinal cancer relapse after surgery. Novel strategies to identify and treat patients at high risk of relapse are needed to improve cancer outcomes. Circulating tumour DNA (ctDNA) is a promising, non-invasive biomarker with the potential to identify the earliest signs of cancer relapse. The presence of tumour-specific DNA in the blood in the absence of visualized tumour is suggestive of minimal residual disease and forebodes measurable relapse. Genomic sequencing techniques have advanced over the past few decades, and we have become better able to detect significantly low levels of DNA circulating in the blood from low-volume disease. Numerous studies using various technologies have established ctDNA as a powerful prognostic biomarker for relapse and survival in gastrointestinal cancers. ctDNA has the potential to risk-stratify patients in the postoperative, post-adjuvant and longitudinal settings for therapeutic escalation or de-escalation strategies. It may also capture early tumour dynamics in response to therapeutic intervention. As the multifaceted potential of ctDNA is attracting the attention of researchers, clinicians and patients, many questions remain regarding its use, interpretation and limitations. Here, we discuss the current understanding of ctDNA for minimal residual disease evaluation in gastrointestinal cancers and potential future directions.


Circulating tumour DNA, liquid biopsy, gastrointestinal neoplasms, minimal residual disease


To date, the cornerstone of curative treatment in early-stage gastrointestinal (GI) cancers includes surgery and sometimes radiation therapy. Depending on the stage and tumour type, these approaches are often combined with systemic neoadjuvant or adjuvant therapies. After all definitive therapy has been administered, physicians and patients rely on imaging and non-specific blood tests to monitor for recurrent disease while hoping for a cure. Unfortunately, many patients relapse and have more advanced, non-curable disease by the time recurrent cancer is measurable using imaging. One proposed strategy to potentially improve survival outcomes is to identify and eradicate cancer at its earliest sign of relapse. Liquid biopsy, or blood-based testing to identify circulating tumour components, is an increasingly studied tool for detecting minimal residual disease (MRD) – that is, persistent cancer at the microscopic level that is not yet large enough to be visualized directly or radiographically. It would be useful to be able to confirm both the presence or absence of MRD; the former could guide the augmentation of therapy, while the latter could help identify cured patients, reassuring them and sparing them toxicity from unnecessary further therapy (Figure 1).

For example, the decision to offer adjuvant chemotherapy for patients with resected stage II/III colorectal cancer (CRC) is complicated, requiring a careful balancing of the risk and benefit for the individual patient. Adjuvant chemotherapy is associated with an overall survival (OS) benefit in patients with stage III CRC and some patients with stage II CRC, but it is difficult to predict the benefit for an individual patient.1,2 Features associated with high-risk stage II disease include poorly differentiated tumours, the presence of lymphovascular or perineural invasion, obstruction, perforation, positive margins, tumour budding and fewer than 12 lymph nodes examined.3 However, these criteria are more prognostic than predictive of benefit from adjuvant chemotherapy. Recommendations for adjuvant therapy in stage II disease vary, ranging all the way from observation only to 6 months of fluorouracil, leucovorin and oxaliplatin.3 As medical oncologists, we are generally biased towards over- rather than undertreating patients so as not to miss the opportunity to deliver potentially curative therapy. Clearly, we need to do better in terms of risk-stratifying patients with early-stage CRC and predicting who needs chemotherapy, how much and for how long. The same need for improved prognostic and predictive biomarkers can be seen across other GI tumour types and in other clinical settings, such as risk-stratifying patients with upper GI and rectal cancers managed with non-surgical wait-and-watch strategies. Until now, we have relied on what is essentially 19th-century technology (the pathology report) to guide therapy decisions, when what we really need is a reliable test for MRD. Circulating tumour DNA (ctDNA) is a promising tool to meet this need.

Cell-free DNA (cfDNA) refers to acellular fragments of double-stranded DNA released from cells that circulate in body fluids, including peripheral blood. While cfDNA was initially described in the 1940s, it was only after technological advances in the 1990s that we saw an exponential growth in its study.4,5 These fragmented DNA strands vary in length from 80 to 200 base pairs and can be released during cell death through apoptosis and necrosis, or active transport and spontaneous release from intact cells.6,7 The fragments circulate in the blood as free DNA, protein-bound DNA or DNA encapsulated by extracellular vesicles; they are present in higher levels in patients with conditions associated with more cell death, such as inflammation, trauma, autoimmune diseases and malignancy.5,8

ctDNA refers to cfDNA specifically released from tumour cells and is typically shorter than normal cfDNA. ctDNA accounts for a wide range – <0.1% to >90% – of cfDNA and has a half-life of 16 minutes to 2 hours.9–12 It is identified by detecting tumour-specific mutations or epigenetic changes in the blood through DNA sequencing and profiling. This DNA can be quantified, analysed for single or broad gene profiling, and utilized for real-time assessments of tumour dynamics based on its short half-life. It is increasingly being studied as a powerful, non-invasive tool for detecting early cancer and providing a prognosis. It might also reveal resistance mutations and provide an understanding of patient response to treatment.

ctDNA is emerging as a powerful tool in detecting MRD, the earliest sign of residual or recurrent cancer in low-volume, microscopic states, after curative-intent treatments such as surgery and adjuvant therapy. The concept of identifying MRD to prognosticate patients and guide standard-of-care treatments has already been established in haematological malignancies.13–17 For example, malignant haematologists use quantitative polymerase chain reaction (qPCR) to detect persistent chronic myeloid leukaemia using BCR-ABL1 transcript levels.18 Unfortunately, solid tumours manifest a diverse array of mutational events with significant inter- and intra-tumour heterogeneity that is not amenable to simple qPCR testing of one transcript. Ideally, an MRD test would be so sensitive and specific as to predict who can safely be observed without adjuvant chemotherapy and who needs chemotherapy due to a predicted survival benefit. With advancing genomic technologies, we are finally seeing MRD testing make its way into the solid tumour space.

While variable across solid tumour studies, ctDNA detection after definitive therapy often mirrors the expected recurrence rates, correlates with other prognostic clinicopathological features, and strongly prognosticates for relapse and survival while outperforming standard clinicopathological risk factors in observational studies.19–23 Emerging data also suggest that ctDNA clearance with adjuvant therapy is associated with improved survival, possibly representing micro-metastatic disease eradication. Still, prospective, randomized trial data are needed to validate the predictive value of this biomarker to guide adjuvant therapies.21,24

Commercialized ctDNA MRD assays are becoming more commonly used to provide patient prognoses in clinical practice. These assays are also being integrated into clinical trials to evaluate their role in adjuvant therapy guidance (Table 1).25–27 Using a ctDNA approach in the neoadjuvant and adjuvant settings has the potential to risk-stratify patients and identify high-risk patients who may truly benefit from adjuvant treatment or intensification, and those who are at low risk or unlikely to benefit and therefore should be spared treatment toxicities. ctDNA has the potential to provide early readouts of therapeutic efficacy or inefficacy on a microscopic level, supported by data correlating ctDNA kinetics with radiographic responses and survival outcomes.20,21,24,28–30

It has also essentially identified a ‘new clinical stage’ of disease post-treatment; one that involves no radiographic evidence of disease, but with detectable MRD. This new disease stage currently lacks any treatment guidelines. This previously undefined patient population identifies a high-risk group that may benefit from the development of novel treatment strategies to improve cure rates and cancer outcomes. However, most of the data today are derived from observational cohort studies, and many questions remain unanswered regarding the interpretation of different test results, strengths and limitations of various assays, the ability of ctDNA to guide adjuvant or post-adjuvant therapy, the clinical significance of rise-and-fall rates or stability of ctDNA, and the ability to individualize follow-up treatment. While we may not be there yet, in this review we describe the current state-of-the-art of blood-based MRD testing using ctDNA in CRC and other GI malignancies.

Circulating tumour DNA detection methods

While tumour tissue yields micrograms of DNA for genomic analysis, blood-based biopsies yield significantly lower amounts: reportedly in the range of nanograms cfDNA per 10 mL blood.31 A major challenge in blood-based testing for MRD relates to assay limits of detection; mean mutant allele frequencies (MAFs) or variant allele frequencies (VAFs) may be present at very low levels in relation to background cfDNA (<0.01%). One major approach to ctDNA detection is to use targeted assays to identify single or limited pre-specified tumour mutations. A second approach is to use broader, non-targeted, genome-wide sequencing to capture multiple genes.

For targeted approaches, digital PCR (dPCR) techniques have emerged as more robust and precise methods for detecting low rates of mutations compared with qPCR. dPCR includes digital droplet PCR (ddPCR) and beads, emulsion, amplification and magnets (BEAMing) techniques, both of which can identify pre-specified, known mutations with high accuracy in the plasma. ddPCR uses a water–oil emulsion technology and distributes DNA fragments into thousands of droplets, each of which contains a single DNA strand, and uses PCR and validated probes to detect and quantify mutations. It is a highly sensitive test for low concentrations of nucleic acids, but requires defining pre-specified mutations for interrogation and can only evaluate a limited number of genes at a time.32,33 This method has successfully detected MRD and provided GI cancer patient prognoses.34–36 While more expensive and involved, the BEAMing method combines dPCR with magnetic beads and flow cytometry to deliver an even more sensitive test for detecting and quantifying a small number of targeted mutations that may be present at very low frequencies (0.01%).37

In contrast, broader applications, such as next-generation sequencing (NGS), can more completely analyse cfDNA and uncover multiple mutations across a broad panel of genes. However, unlike targeted PCR-based assays, which can detect VAFs with deep sequencing in the range of 0.001–0.01%, non-targeted NGS assays achieve broad interrogation at the cost of lower sensitivity and detect VAFs in the range of 0.02–10%.31,32,38–40 Multiple NGS platforms exist, and their strengths and limitations have been well described in the literature.32,41 For example, the upgraded multiplex PCR-based NGS TAm-Seq technology, which utilizes tagged-amplicon deep sequencing to target multiple pre-specified genes on a panel, can detect single-nucleotide variants, short insertions and deletions, and copy number variants with high sensitivity at 0.25–0.33% allele frequencies, and can detect mutations at even lower levels at a 0.02% allele frequency.42,43 Although this upgraded technology can interrogate multiple genes and is relatively inexpensive, it also requires prior knowledge of targeted mutations.41 Massive parallel sequencing platforms are prone to error during the sequencing process. Other assays, such as Safe-SeqS, in which unique identifiers are added to template DNA molecules prior to amplification, incorporate error-reducing strategies to improve sensitivity.44 Another platform, CAPP-Seq, uses hybrid-capture to detect all major mutation classes and includes rearrangements that are not always detected on other amplicon-based platforms.45

Whole-exome sequencing (WES) and whole-genome sequencing allow for the most comprehensive profiling, interrogate the whole exome or genome, and do not limit sequencing to predetermined genes. However, this breadth of coverage comes at the cost of higher expense, lower sequencing depth and lower sensitivity.41 These differences are essential to consider when choosing an optimal assay for MRD detection, where robust sensitivity is critical to reducing false-negative results. Methods such as template tagging, duplex sequencing, single-molecule sequencing, error-correcting algorithms, incorporating epigenetic marker evaluation and enriching for highly fragmented ctDNA, among others, are being used to improve the ability to detect low-level MAFs with NGS.38,40,46

Minimal residual disease detection methods

While the techniques to detect ctDNA may vary, MRD assays can be broadly characterized as either tumour agnostic or tumour informed. Tumour-agnostic assays, also known as plasma-only assays, do not require a prior tissue biopsy or knowledge of a patient’s tumour genomic profile; they either narrowly or broadly interrogate cfDNA for commonly mutated cancer genes. Although tumour-agnostic assays may be less sensitive than tumour-informed assays due to their use of broad panels, which compromises the depth of sequencing, they are quick (7- to 10-day turnaround), relatively inexpensive and more comprehensive in characterizing mutations and identifying emerging resistance mechanisms on serial analysis. Tumour-agnostic assays have been applied in CRC MRD studies and have demonstrated prognostic value in predicting relapse and survival.19,35,46 For example, Parikh et al. recently reported a recurrence sensitivity and specificity of 55.6% and 100%, respectively, using the Guardant Reveal/LUNAR-1 plasma-only test in prospectively studied patients with CRC who underwent definitive surgery.46 In this study, the sensitivities improved with longitudinal testing and integration of epigenomic signatures in testing. Tumour-agnostic assays have also demonstrated prognostic value in other prospective and retrospective studies in resected upper GI and pancreatic cancers.23,47–49 When choosing MRD tests, it is important to realize that tumour-agnostic, NGS-based assays for MRD may differ from those frequently used to evaluate mutational profiles in metastatic disease. For instance, MRD assays may specifically combine DNA with epigenetic marker analyses or other methods to increase sensitivity in detecting minuscule ctDNA levels.46

In contrast, tumour-informed assays require prior knowledge of a tumour’s mutational profile through tissue analysis to design patient- and tumour-specific assays targeting these mutations in the blood. For example, the Signatera ctDNA test, which is commercialized and has received US Food and Drug Administration (FDA) Breakthrough Device Designation, requires WES of a patient’s primary tumour and the buffy coat to identify 16 clonal tumour-specific somatic mutations that are not driver mutations under selection pressure and that are not present in the germline.50,51 Sixteen PCR primers are designed for the identified single-nucleotide variants, and a patient’s peripheral blood is drawn at pre-specified times and then tested against the multiplex PCR with ultra-deep sequencing. If at least two mutations are detected, this results in a positive read. This assay can detect very low fractions of tumour DNA (<0.01%) with >95% sensitivity. While other tumour-informed assays using Safe-SeqS, ddPCR and other designs exist, the Signatera test has been widely studied in many GI and non-GI tumour types, demonstrating impressive sensitivities (88–100%) and specificities (93–100%) for MRD, positive predictive values for relapse (>98%) and median lead times from ctDNA detection to relapse (8–9 months).20–22,24,29,34,36,50,52–65 These highly specific and sensitive assays are particularly attractive for MRD detection and are being increasingly studied in this space. However, due to the inherent process of tumour, germline and then peripheral blood sequencing, tumour-informed testing may be more expensive than the plasma-only approach, as it first requires tumour tissue acquisition and testing, and takes longer to obtain results (4–6 weeks for the first and then 7–10 days for subsequent test results). In addition, if there is insufficient tumour tissue, these tests may not be feasible. For example, creating the commercialized Signatera test requires WES on tumour samples with a surface area >25 mm,2 volume ≥60 microns in thickness and tumour content ≥30% when submitted as formalin-fixed paraffin-embedded tumour blocks or unstained slides.66 Fine-needle aspirations, resected tumour with significant treatment effect and scant cytology samples may not always yield adequate tissue for these tests.

Considerations in circulating tumour DNA interpretation

Although the data today increasingly support ctDNA as a powerful tool in the MRD setting, we must remember it is not a perfect test. We have already described some of the inherent sensitivity and technical limitations of various assays. It is important to counsel patients on the potential for false negatives when there may, in fact, be residual disease that is below an assay’s detection threshold. Some studies have demonstrated that testing sensitivities improve with serial monitoring.29 Therefore, it is important to consider longitudinal testing in the adjuvant or post-adjuvant setting in case a negative test turns positive after the cancer has been given enough time to grow. In addition, assays that target one or a few genes may miss subclones of heterogeneous tumours. Testing multiple genes or selecting for early clonal mutations that are not under selection pressure, as seen in the Signatera assay, may overcome these limitations.67 The exact testing intervals and total duration for which we should follow patients with negative tests are currently unknown.

ctDNA detection and quantification can also be influenced by multiple factors such as the primary tumour type, staging, assays used, mutations analysed and organs involved. For example, Bettegowda et al. demonstrated that the frequency and concentration of detected ctDNA in metastatic cancers varied based on tumour type, with relatively higher frequencies of detection in bladder, colorectal, gastroesophageal and ovarian cancers; and lower frequencies of detection in central nervous system, renal, thyroid and prostate cancers.10 Although present at lower concentrations, ctDNA can also be detected in localized, early-stage cancers, and frequencies and concentrations trend higher with higher clinical stages.

Others have shown that the sites of metastatic disease and size of lesions may influence ctDNA levels, with higher detection and concentrations in patients with liver metastases compared with those with lung- and peritoneal-only metastases.68,69 For example, Van’t Erve et al. reported that ctDNA was detected in 93% of patients with CRC with liver metastases compared with 20% of those with peritoneal metastases.68 Bando et al. reported that ctDNA median VAFs were highest in patients with CRC liver metastases (23.1%), lower in those with lymph node metastases (6.0%) and lowest in those with lung and peritoneal metastases (0.4% in both).69 These findings suggest that different tumours have different potentials for ‘DNA shedding’ into the bloodstream, theoretically due to variations in tumour biology, tumour microenvironment and degree of vascularity, for example. Currently, we cannot use ctDNA to predict recurrence location and patterns, since ctDNA shedding may vary based on organ involvement. Therefore, it is possible that patients may develop recurrence in sites associated with ‘low shedding’ with false-negative ctDNA values. We also currently do not know how to interpret different rates of ctDNA changes or ctDNA stability, and how they translate to clinical management.

On the other hand, while many tests are highly specific, we must also be aware of possible false positives. For example, clonal haematopoiesis of indeterminate potential (CHIP) is a process that can be seen with normal aging and even in healthy individuals where there is clonal expansion of haematopoietic stem cells with accumulating somatic mutations in the absence of malignancy or haematological disorders.70 Some of these CHIP-related aberrations may involve genes often associated with GI malignancies, such as TP53 and RAS, among others, and may lead to false-positive results in MRD testing.71–73 To address this, some platforms use bioinformatics to filter out likely benign clonal variants, while others sequence matched tumour samples and peripheral blood leukocytes and perform ctDNA interrogation only for tumour-specific mutations that are not present in the buffy coat.57,74

With ctDNA tests such as Signatera, which has FDA Breakthrough Designation and Medicare coverage for patients with stage II/III CRC,75,76 and many other tests in the pipeline, clinicians are increasingly using ctDNA for MRD monitoring in routine practice. However, many clinicians, especially in the community, are understandably hesitant to incorporate this into routine management in the absence of prospective, randomized data to help guide treatment escalation or de-escalation. While these tests are demonstrating powerful prognostic value, questions remain regarding their predictive capabilities and what to do with the results. Several clinical trials using different assays have been developed to answer these questions, especially in CRC (Table 1). We need large patient numbers to validate survival endpoints when using ctDNA-guided strategies, and increasing community and academic trial enrolment and awareness will hopefully facilitate this research.

Minimal residual disease in colorectal cancer

For patients with stage II/III CRC following surgical resection, the presence of ctDNA is clearly a negative prognostic factor. In an analysis of the IDEA-France trial by Taieb et al., 1,017 patients with stage III CRC and ctDNA suitable for analysis collected prior to adjuvant chemotherapy were followed for disease-free survival (DFS) and OS. Overall, 14% of patients had positive postoperative ctDNA, and positive ctDNA was associated with inferior DFS (3-year DFS rate 66.4% versus 76.7%; hazard ratio [HR] 1.46, 95% confidence interval [CI] 1.08–1.97; p=0.015) and OS (5-year OS rate 81% versus 87%; HR 1.56, 95% CI 1.08–2.26; p=0.018).35 Similar findings have been reported for carcinoembryonic antigen (CEA). Sonoda et al. looked at postoperative CEA in 482 patients with stage II/III CRC, and elevated postoperative CEA correlated with significantly inferior DFS and OS.77 However, only 7.6% of patients with stage II disease and 10.6% of those with stage III disease had elevated postoperative CEA, implying that CEA may be less sensitive than ctDNA. Indeed, a study of 110 Japanese patients with metastatic CRC showed a concordance rate of only 75.5% between CEA and ctDNA.78 Moreover, many tumours do not make CEA at baseline preoperatively. The question is: how do we adjust our adjuvant treatment strategies for positive or negative ctDNA? Stated another way, can we rely on negative ctDNA to predict a lack of benefit from adjuvant chemotherapy and positive ctDNA to predict a benefit? We need a test to predict who needs adjuvant chemotherapy and who does not.

Even if ctDNA is not yet able to predict who needs chemotherapy, a growing body of data suggests it may aid in the decision-making process, and ctDNA is the focus of several ongoing, prospective clinical trials to better answer these questions. While positive postoperative ctDNA is negative prognostically, clearance of ctDNA post-adjuvant chemotherapy may highlight a cohort that truly benefits from adjuvant chemotherapy.21,24,57

The most relevant clinical use of ctDNA in the current paradigm is serial monitoring. Even with the advent of ctDNA, standard surveillance includes imaging, laboratory work and visits through 5 years, in addition to surveillance colonoscopies.3 If nothing else, persistently negative longitudinal ctDNA may be the best early predictor of cure: in a Danish study of 125 patients with stage I–III CRC by Reinert et al., only two of 60 (3.3%) persistently ctDNA-negative patients experienced disease recurrence.57 Thus, patient anxiety, especially around the timing of surveillance scans, can hopefully be lessened somewhat as their odds of recurrence with persistently negative ctDNA results drop to relatively very low levels.

Stage II CRC may be the best setting for using ctDNA to guide therapeutic selection. Patients with low-risk stage II disease are typically suitable for observation alone, but a minority of these patients will be ctDNA positive postoperatively and have a higher risk of recurrence, and will potentially benefit from adjuvant chemotherapy.3 This strategy is being actively studied in the COBRA trial (NRG-GI005; NCT04068103).79 Conversely, the benefit of adjuvant chemotherapy in high-risk stage II disease is often overstated, and many patients are overtreated. Thus, ctDNA may be useful to de-escalate adjuvant treatment in this population, where the guidelines recommend a wide range of choices: from 6 months of doublet chemotherapy to observation alone.3 Again, this strategy needs to be evaluated in robust, prospective clinical trials. While we are still learning how to harness the power of this new technology, anecdotal reports of how ctDNA has played a vital role in clinical decision-making in patients with stage II disease have been published.80

Another clinical quandary is whether to offer adjuvant chemotherapy to patients after metastasectomy for stage IV oligometastatic CRC. In the Japanese JCOG0603 trial, 300 patients with resectable liver metastases were randomized 1:1 to 6 months of adjuvant fluorouracil, leucovorin and oxaliplatin or surgery alone.81 While DFS was longer with chemotherapy (5-year rate 49.8% versus 38.7%; HR 0.67, 95% CI 0.50–0.92; p=0.006), there was no OS benefit (5-year rate 83.1% versus 71.2%; HR 1.25, 95% CI 0.78–2.00; p=0.42), and patients who had already received prior adjuvant chemotherapy after primary tumour resection tended to do worse in DFS and OS with more chemotherapy in subgroup analyses. In this setting, negative ctDNA post-metastasectomy may provide some extra solace that chemotherapy may not benefit this group, especially for less fit patients. Again, a prospective study to validate this would be preferred to determine whether positive ctDNA post-metastasectomy correlates with benefit from chemotherapy.

The IDEA collaboration established 3 months of adjuvant doublet chemotherapy as an appropriate alternative to the traditional 6 months of adjuvant therapy for patients with stage III CRC.82 What can ctDNA add to this decision-making algorithm? In the Taieb et al. analysis mentioned above, patients who were ctDNA positive postoperatively and received 3 months of adjuvant chemotherapy had significantly shorter DFS compared with ctDNA-positive patients treated with 6 months of therapy and ctDNA-negative patients treated with 3 or 6 months of therapy (p=0.0021). These findings were especially notable among patients with high-risk disease (T4 and/or N2, p<0.001), and while this trend was noted in low-risk disease, it was not statistically significant (p=0.47).35 It is important to recognize that these findings were retrospective, and future studies should prospectively address whether escalation from 3 to 6 months of adjuvant therapy is beneficial in the ctDNA-positive population. In practice, we are justifiably nervous when someone is postoperatively ctDNA positive, and it is very reasonable to continue adjuvant therapy to 6 months if the patient remains ctDNA positive after 3 months of adjuvant therapy. What is less clear is whether someone who clears ctDNA after 3 months will benefit from an additional 3 months of chemotherapy.

Emerging data from Japan highlight the potential predictive abilities of ctDNA. The observational GALAXY study in CIRCULATE-Japan analysed Signatera ctDNA testing in 1,040 patients with resected stage I–IV CRC, and demonstrated that positive ctDNA 4 weeks after surgery was negatively prognostic for DFS (HR 13.3, 95% CI 8.0–22.2; p<0.001) and that ctDNA trends at 12 weeks post-surgery were also prognostic with a worse DFS reported in patients who remained ctDNA positive compared with those who converted from positive to negative (HR 15.8, 95% CI 5.7–44.2; p<0.001).24 Most interesting was that patients with high-risk stage II or any stage III CRC with negative 4-week postoperative ctDNA received no benefit from adjuvant chemotherapy (adjusted HR 1.3, 95% CI 0.5–3.6; p=0.63), implying we are likely overtreating patients with negative ctDNA. With the important caveats that the decision to give chemotherapy was not randomized based on ctDNA and that the median follow-up time of 11.4 months is relatively short, GALAXY adds more evidence that we will likely be using ctDNA to guide adjuvant therapy decision-making in the near future.

ctDNA may also have important roles in the management of locally advanced rectal cancer (LARC), where curative-intent therapy is multimodal with chemotherapy, radiation and surgery. For instance, although adjuvant chemotherapy after neoadjuvant chemoradiation and surgery is recommended, the survival benefits are controversial, and questions remain regarding which patients truly benefit and what chemotherapy regimens are best used.83 Improved biomarkers to risk-stratify patients would be helpful in this space. Tie et al. prospectively collected blood samples for tumour-informed ctDNA analysis at baseline, after chemoradiation and after surgery in 159 patients with LARC, and reported worse recurrence-free survival (RFS) in patients with ctDNA detected after chemoradiation (HR 6.6, p<0.001) and after surgery (HR 13.0, p<0.001), with a more prominent prognostic impact in patients not treated with adjuvant therapy.53 In addition, a recent systematic review of 21 publications highlighted the promising prognostic potential of ctDNA in LARC.84 Another systematic review that reported on 25 published studies (eight of which specifically evaluated prognosis) also summarized the prognostic value of ctDNA detection at baseline, during or post-neoadjuvant therapy and post-surgery.85 These evolving data indicate that using ctDNA as a prognostic biomarker may help to identify high-risk patients in need of adjuvant treatment.

Another potential use for ctDNA is in patients with rectal cancers who are considering a watch-and-wait approach following neoadjuvant therapy. For patients with low-lying rectal cancers, the prospect of a permanent ostomy is frequently a non-starter when considering surgical resection. Given that some patients may have a pathological complete response and may not have required surgery, ctDNA could help risk-stratify this population and potentially tilt the scales in favour of surgery for those who are ctDNA positive. Recent studies summarized in two systematic reviews have demonstrated conflicting results on the ability of ctDNA to predict tumour responses and pathological complete response after chemoradiation, although the trial designs and ctDNA detection methods utilized were highly variable across studies.84,85 Combining ctDNA data such as present mutations, baseline levels and dynamics on therapy, along with other diagnostics findings such as tumour regression grading on MRI, has the potential to improve our ability to predict a pathological complete response or residual disease.86 While these approaches are encouraging, a lot of work needs to be done to elucidate the prognostic and predictive roles of ctDNA, especially when considering organ-preserving strategies or treatment de-escalation in LARC.

As discussed above, ctDNA clearly has multiple shortcomings, and there is a lot that we do not know. These include the fact that a negative ctDNA test, especially at a single postoperative time point, does not rule out MRD as it could be below the level of detection. We also do not know whether ctDNA can be used to influence decisions regarding the use and duration of adjuvant chemotherapy. However, there is a fair amount of retrospective evidence, and prospective trials are ongoing. Finally, ctDNA may allow us to alter our surveillance strategy; for example, intensifying surveillance for those who are ctDNA positive and spreading out surveillance for those who are ctDNA negative. However, we do not have the evidence to make these changes yet.

Minimal residual disease in pancreatic ductal adenocarcinoma

While ctDNA is best described in CRC, it is also gaining traction in other GI tumour types. Utilizing this technology to improve outcomes is particularly attractive in pancreatic ductal adenocarcinoma (PDAC), an aggressive and lethal disease with dismal survival rates.87 Despite advances in oncology, therapeutic options for patients with PDAC continued to be limited, making surgical resection the only hope for a cure. Although surgical techniques have improved and different treatment methods (including chemotherapy and radiotherapy) have been incorporated, the prognosis is still poor, with 5-year OS rates of 39% and 13% in localized and locally advanced disease, respectively.88 Due to the high rates of recurrence, both systemically (>80%) and locally (>20%) following curative-intent surgery, there is a pressing need for a prognostic marker that measures MRD.89

In PDAC, multiple studies have proved the prognostic value of serum or plasma ctDNA to detect MRD.49 Several techniques have been used to analyse ctDNA, including whole-genome sequencing, targeted gene sequencing and ddPCR.90 WES has not been studied widely, given its high cost and low sensitivity. As previously discussed, in contrast to WES, targeted gene sequencing and ddPCR can identify mutant ctDNA with MAFs as low as <0.2% and <0.1%, respectively, making them more sensitive tools for detecting mutant ctDNA and better methods for monitoring MRD.91,92 As PDAC has a high prevalence of KRAS proto-oncogene mutations, molecular profiling of resectable PDAC ctDNA samples has been largely limited to targeted gene sequencing and ddPCR to detect KRAS mutations.93

A prospective, multicentre biomarker study enrolled 42 patients with early-stage, operable PDAC and matched blood and tissue samples.54 PCR-based Safe-SeqS assays were used to obtain pre- and postoperative samples for ctDNA analysis. After a median follow-up of 38.4 months, the study found that detection of ctDNA preoperatively was associated with poorer RFS (HR 4.1, 95% CI 1.8–9.0; p=0.002) and OS (HR 4.1, 95% CI 1.6–10.5; p=0.015). Furthermore, the presence of ctDNA after curative-intent resection was also associated with worse RFS (HR 5.4, 95% CI 1.9–15.2; p<0.0001) and OS (HR 4.0, 95% CI 1.2–13.6; p=0.003). Notably, 13/13 (100%) patients with detectable ctDNA postoperatively experienced disease recurrence, regardless of adjuvant treatment, compared with 10/22 (45%) patients with undetectable postoperative ctDNA.

Another study has looked at the role of peri-operative KRAS-mutant ctDNA in resectable PDAC by collecting plasma samples from 31 patients and analysing ctDNA using ddPCR.94 Six of the 31 patients who had surgery tested positive for postoperative ctDNA. The presence of ctDNA after surgery was linked with significantly poorer RFS (4.6 versus 17.6 months; p=0.03) and OS (19.3 versus 32.2 months; p=0.027).

Our group recently conducted a comprehensive review and meta-analysis to assess the prognostic significance of KRAS-mutated ctDNA in resectable PDAC.95 A total of 13 studies and 954 patients were included in the final analysis. Compared with negative ctDNA, positive ctDNA in the preoperative setting was associated with worse RFS (HR 2.067, 95% CI 1.346–3.174; p<0.001) and OS (HR 2.170, 95% CI, 1.451–3.245; p<0.001). In the postoperative setting, positive ctDNA was again associated with worse RFS (HR 2.986, 95% CI 1.897–4.699; p<0.001) and OS (HR 5.812, 95% CI 1.757–19.228; p=0.004) compared with negative ctDNA.

At the American Society of Clinical Oncology GI 2022 conference, a prospective cohort study was presented that examined the prognostic value of a personalized and tumour-informed ctDNA test (Signatera bespoke multiplex PCR NGS assay) in 93 patients with PDAC.96 Postoperative ctDNA positivity at any time was associated with poorer RFS compared with ctDNA negativity (HR 8.0, 95% CI 3.4–18.7; p=1.6×10).Elevated cancer antigen (CA) 19-9, on the other hand, was not linked with RFS (p=0.35).

At the time of writing, there is growing evidence supporting the use of ctDNA as a marker of MRD in resectable PDAC (Table 2),54,94,97–108 which might have implications for the aggressiveness of monitoring and length of adjuvant therapy in PDAC. However, more prospective trials are needed to assess this utility.

Minimal residual disease in other gastrointestinal cancers

A limited number of studies have investigated whether ctDNA can be used as a predictive or prognostic tool in other cancers such as upper GI malignancies. One prospective study enrolled 97 patients with oesophageal adenocarcinoma treated with neoadjuvant chemotherapy and surgery to determine whether ctDNA was correlated with survival.65 Out of 77 patients with available postoperative ctDNA samples and a known cause of death, 36 (47%) had recurrent disease within 60 months of follow-up. Sixteen out of 77 patients (21%) were ctDNA positive following resection, and recurrence was observed in 12 of these patients (75%). Furthermore, the mean cancer-specific survival for ctDNA-positive patients was half that of ctDNA-negative patients (14.9 versus 29.5 months; HR 2.32, 95% CI 1.14–4.73; p=0.03), and ctDNA-positive patients also had worse DFS (HR 2.35, 95% CI 1.18–4.72; p=0.01). In another study of oesophageal cancers, Azad et al. tested for ctDNA before and after neoadjuvant chemoradiation, and reported that ctDNA detection after chemoradiation was associated with tumour progression, distant metastases and inferior survival.109 For patients who did not proceed with surgery, ctDNA combined with metabolic imaging analyses after chemoradiation was more sensitive than ctDNA alone or positron emission tomography–computed tomography alone in predicting recurrence (sensitivities of 100%, 71.4% and 57.1%, respectively).

In gastric cancer, one study has reported on 46 patients with stage I–III disease who underwent surgical resection with curative intent.23 The sensitivity and specificity of a positive, post-treatment ctDNA result predicting recurrence at 30 months were 39% and 100%, respectively. Additionally, ctDNA positivity after surgical resection was associated with statistically significantly worse DFS (HR 6.56, 95% CI 8.316–208.5; p<0.0001) and OS (HR 5.959, 95% CI 3.765–138.1; p=0.0007). Serial, longitudinal testing was also prognostic and identified emerging mutations during and after systemic therapy. Interestingly, this study also reported that ctDNA was detected a median of 179 days prior to radiographic evidence of recurrence. This opens the possibility for ctDNA to be used as a marker to intensify therapy earlier in these types of cancer. Recently, Huffman et al. used the Signatera tumour-informed MRD assay prospectively in 254 patients with oesophageal, gastro-oesophageal junction and gastric carcinomas. These researchers reported ctDNA positivity rates of 22.6–46.9% for resected localized disease and 62.5–100% for stage IV disease.110 MRD detection was similarly associated with worse survival outcomes. While these studies are beginning to demonstrate the prognostic potential of ctDNA in upper GI cancers, further data are needed to validate ctDNA as a tool to risk-stratify patients in controversial clinical scenarios in addition to detecting MRD after all definitive therapy. For example, ctDNA in combination with other diagnostics could potentially be applied to predict complete clinical or pathological responses after neoadjuvant therapies and select patients who may safely
be offered a wait-and-watch approach instead of proceeding with major oesophagogastrectomies.

ctDNA also has potential applications in hepatobiliary cancers, where we see inconsistent production of the non-specific protein tumour markers (CA 19-9 and alpha-fetoprotein) used currently to monitor disease. Kasi et al. were among the first to use a tumour-specific assay to study MRD in these cancers.111 Using Signatera, they reported 20–29% MRD rates across patients with hepatocellular and biliary carcinomas, with detection of MRD being significantly associated with disease stage. Additional data on longitudinal testing and associations with clinical response are pending. Another group had previously used a tumour-agnostic ddPCR method to detect hotspot mutations in patients with resected hepatocellular carcinoma, and also reported that increased postoperative ctDNA MAFs were associated with relapse and shorter survival.112

Smaller trials have broadened the application of ctDNA to other, less common types of GI cancers. In GI stromal tumours, early data suggest that ctDNA detection intraoperatively may correlate with tumour risk, ctDNA kinetics during systemic therapy and surgery may assess response, and ctDNA analysis may identify emerging tyrosine kinase inhibitor resistance mechanisms.113,114 However, robust studies evaluating the role of ctDNA in MRD in GI stromal tumours are currently lacking. In anal cancer, where concomitant human papillomavirus infection is highly prevalent, human papillomavirus ctDNA was recently studied as a biomarker, and detection after definitive chemoradiation was strongly associated with inferior DFS.115 This highlights the possibility of using ctDNA to measure MRD in other types of viral-induced cancers. Further research is needed to understand the applicability and feasibility of measuring residual disease in a wide variety of cancers.

Circulating tumour cells to assess minimal residual disease

In addition to ctDNA and cancer-derived protein markers, other liquid-biopsy components such as circulating tumour cells (CTCs) have the potential to detect residual disease. CTCs are tumour cells that have been shed from a primary or metastatic tumour into the bloodstream, and it is believed that subsets of these released from a primary tumour are responsible for metastases.116 Studies have demonstrated that the presence of CTCs is prognostic in advanced CRC and correlates with stage of disease.117–120 Studies in advanced CRC have also demonstrated decent concordance of PCR-based sequencing of CTCs, tumour tissue and ctDNA in hotspot mutations.121 Therefore, isolating and studying CTCs from peripheral blood may be another non-invasive way of detecting residual cancer after curative therapies and studying a tumour’s heterogeneous mutational landscape. In non-metastatic CRC, the presence of CTCs has been reported as a negative prognostic factor for progression-free survival and OS across many studies.119,122–124 In the adjuvant setting in CRC, the presence of postoperative CTCs has been reported as a poor prognostic factor for DFS and OS in some studies125,126 but as being non-prognostic in others.127

However, a notable limitation in utilizing CTCs today stems from the major challenge of isolating intact, rare CTCs from peripheral blood. CTC isolation techniques largely rely on antibodies for tumour markers and cell properties (e.g. size and density).128,129 For example, the FDA-approved CellSearch technology harvests epithelial cell adhesion molecule-positive cells, while also utilizing a leukocyte common antigen (CD45)-negative selection process to remove leukocytes.130 CellSearch technology and other methods that isolate cells based on physical properties have success rates of ≤25% in harvesting CTCs, although new strategies improving upon this are under way.129 In addition, genomic discordance has been reported among CTCs, tumour and ctDNA, and may reflect either heterogeneous tumour subclones or limited sensitivities of CTC genomic testing in samples with low CTC yield and purity.121

Fortunately, in the past decade, we have seen a rise in CTC studies and technological advances. Identifying and studying CTCs in the MRD setting could theoretically help us to identify genotypic and phenotypic metastatic subclones that could be targeted therapeutically and studied for resistance mechanisms. However, given the challenges in isolating CTCs in advanced disease settings, the development, validation and distribution of rapid, sensitive tests are important steps to increasing the success of routinely using CTCs in low tumour burden settings.


Following considerable scientific efforts to enhance genomic-profiling technologies, treatment strategies in GI cancers are increasingly relying on precision medicine. Whether it be through detecting MRD, risk-stratifying patients for adjuvant treatment strategies, guiding surveillance, characterizing baseline and emerging mutations, or predicting responses to therapy, ctDNA is a powerful tool with which we can individualize treatments for patients. While most data to date describe ctDNA in CRC, we are seeing the technology evolving across other GI cancers. Many questions regarding its interpretation in the MRD setting remain, but with emerging prospective studies and initiatives to incorporate biomarker analyses into clinical trials, we are hopeful the current uncertainties will be clarified in the future. We anticipate ctDNA will continue to be increasingly adopted in clinical practice to deliver personalized care and ultimately improve cancer outcomes.

Article Information:

Reetu Mukherji, Ali Alqahtani, Harrison D Winters and Benjamin A Weinberg have no financial or non-financial relationships or activities to declare in relation to this article.

Compliance With Ethics

This article involves a review of the literature and did not involve any studies with human or animal subjects performed by any of the authors.

Review Process

Double-blind peer review.


All named authors meet the International Committee of Medical Journal Editors (ICMJE) criteria for authorship of this manuscript, take responsibility for the integrity of the work as a whole, and have given final approval for the version to be published.


Reetu Mukherji, Division of Hematology and Oncology, Medstar Georgetown University Hospital, 3800 Reservoir Rd NW, Washington, DC 20007, USA. E:


No funding was received in the publication of this article.


This article is freely accessible at © Touch Medical Media 2022


The authors would like to thank Marion L Hartley, PhD, for her valuable edits and suggestions during the composition of this manuscript.

Data Availability

Data sharing is not applicable to this article as no datasets were generated or analysed during the current study/during the writing of this article.




  1. André T, Boni C, Mounedji-Boudiaf L, et al. Oxaliplatin, fluorouracil, and leucovorin as adjuvant treatment for colon cancer. N Engl J Med. 2004;350:2343–51.
  2. André T, de Gramont A, Vernerey D, et al. Adjuvant fluorouracil, leucovorin, and oxaliplatin in stage II to III colon cancer: updated 10-year survival and outcomes according to BRAF mutation and mismatch repair status of the MOSAIC study. J Clin Oncol. 2015;33:4176–87.
  3. 2021. NCCN Guidelines Version 3.2021 Colon Cancer. Available at: (accessed 7 January 2022).
  4. Mandel P, Metais P. Nuclear acids in human blood plasma. C R Seances Soc Biol Fil. 1948;142:241–3.
  5. Volik S, Alcaide M, Morin RD, Collins C. Cell-free DNA (cfDNA): clinical significance and utility in cancer shaped by emerging technologies. Mol Cancer Res. 2016;14:898–908.
  6. Duque-Afonso J, Waterhouse M, Pfeifer D, et al. Cell-free DNA characteristics and chimerism analysis in patients after allogeneic cell transplantation. Clin Biochem. 2018;52:137–41.
  7. Szilágyi M, Pös O, Márton É, et al. Circulating cell-free nucleic acids: main characteristics and clinical application. Int J Mol Sci. 2020;21:6827.
  8. Mondelo-Macía P, Castro-Santos P, Castillo-García A, et al. Circulating free DNA and its emerging role in autoimmune diseases. J Pers Med. 2021;11:151.
  9. Underhill HR. Leveraging the fragment length of circulating tumour DNA to improve molecular profiling of solid tumour malignancies with next-generation sequencing: a pathway to advanced noninvasive diagnostics in precision oncology? Mol Diagn Ther. 2021;25:389–408.
  10. Bettegowda C, Sausen M, Leary RJ, et al. Detection of circulating tumor DNA in early- and late-stage human malignancies. Sci Transl Med. 2014;6:224ra24.
  11. Bachet JB, Bouché O, Taieb J, et al. RAS mutation analysis in circulating tumor DNA from patients with metastatic colorectal cancer: the AGEO RASANC prospective multicenter study. Ann Oncol. 2018;29:1211–9.
  12. Diehl F, Schmidt K, Choti MA, et al. Circulating mutant DNA to assess tumor dynamics. Nat Med. 2008;14:985–90.
  13. Venditti A, Piciocchi A, Candoni A, et al. GIMEMA AML1310 trial of risk-adapted, MRD-directed therapy for young adults with newly diagnosed acute myeloid leukemia. Blood. 2019;134:935–45.
  14. Schuurhuis GJ, Heuser M, Freeman S, et al. Minimal/measurable residual disease in AML: a consensus document from the European LeukemiaNet MRD Working Party. Blood. 2018;131:1275–91.
  15. Gökbuget N, Dombret H, Bonifacio M, et al. Blinatumomab for minimal residual disease in adults with B-cell precursor acute lymphoblastic leukemia. Blood. 2018;131:1522–31.
  16. Cumbo C, Anelli L, Specchia G, Albano F. Monitoring of minimal residual disease (MRD) in chronic myeloid leukemia: recent advances. Cancer Manag Res. 2020;12:3175–89.
  17. Munshi NC, Avet-Loiseau H, Rawstron AC, et al. Minimal residual disease predicts superior survival in patients with multiple myeloma: a meta-analysis. JAMA Oncol. 2017;3:28–35.
  18. Hughes T, Deininger M, Hochhaus A, et al. Monitoring CML patients responding to treatment with tyrosine kinase inhibitors: review and recommendations for harmonizing current methodology for detecting BCR-ABL transcripts and kinase domain mutations and for expressing results. Blood. 2006;108:28–37.
  19. Overman MJ, Vauthey JN, Aloia TA, et al. Circulating tumor DNA (ctDNA) utilizing a high-sensitivity panel to detect minimal residual disease post liver hepatectomy and predict disease recurrence. J Clin Oncol. 2017;35(Suppl.):3522.
  20. Tie J, Wang Y, Tomasetti C, et al. Circulating tumor DNA analysis detects minimal residual disease and predicts recurrence in patients with stage II colon cancer. Sci Transl Med. 2016;8:346ra92.
  21. Tie J, Cohen JD, Wang Y, et al. Circulating tumor DNA analyses as markers of recurrence risk and benefit of adjuvant therapy for stage III colon cancer. JAMA Oncol. 2019;5:1710–7.
  22. Tie J, Cohen J, Wang Y, et al. A pooled analysis of multicenter cohort studies of post-surgery circulating tumor DNA (ctDNA) in early stage colorectal cancer (CRC). J Clin Oncol. 2019;37(Suppl.):3518.
  23. Yang J, Gong Y, Lam VK, et al. Deep sequencing of circulating tumor DNA detects molecular residual disease and predicts recurrence in gastric cancer. Cell Death Dis. 2020;11:1–9.
  24. Kotaka M, Shirasu H, Watanabe J, et al. Association of circulating tumor DNA dynamics with clinical outcomes in the adjuvant setting for patients with colorectal cancer from an observational GALAXY study in CIRCULATE-Japan. J Clin Oncol. 2022;40(Suppl.):9.
  25. Kopetz S. ctDNA as a tool to detect minimal residual disease after surgery. Oral presentation at: 2022 ASCO Gastrointestinal Cancers Symposium; 20 January 2022; San Francisco, CA.
  26. Hall W. Emerging roles of ctDNA on the horizon of gastrointestinal cancers. Oral presentation at: 2022 ASCO Gastrointestinal Cancers Symposium; 20 January 2022; San Francisco, CA.
  27. Naidoo M, Gibbs P, Tie J. ctDNA and adjuvant therapy for colorectal cancer: time to re-invent our treatment paradigm. 2021;13:346.
  28. Henriksen TV, Tarazona N, Frydendahl A, et al. Circulating tumor DNA in stage III colorectal cancer, beyond minimal residual disease detection, toward assessment of adjuvant therapy efficacy and clinical behavior of recurrences. Clin Cancer Res. 2022;28:507–17.
  29. Loupakis F, Sharma S, Derouazi M, et al. Detection of molecular residual disease using personalized circulating tumor DNA assay in patients with colorectal cancer undergoing resection of metastases. JCO Precis Oncol. 2021;5:PO.21.00101.
  30. Bratman SV, Yang SYC, Iafolla MAJ, et al. Personalized circulating tumor DNA analysis as a predictive biomarker in solid tumor patients treated with pembrolizumab. Nat Cancer. 2020;1:873–81.
  31. Odegaard JI, Vincent JJ, Mortimer S, et al. Validation of a plasma-based comprehensive cancer genotyping assay utilizing orthogonal tissue- and plasma-based methodologies. Clin Cancer Res. 2018;24:3539–49.
  32. Elazezy M, Joosse SA. Techniques of using circulating tumor DNA as a liquid biopsy component in cancer management. Comput Struct Biotechnol J. 2018;16:370–8.
  33. Taylor SC, Laperriere G, Germain H. Droplet digital PCR versus qPCR for gene expression analysis with low abundant targets: from variable nonsense to publication quality data. Sci Rep. 2017;7:2409.
  34. Schøler LV, Reinert T, Ørntoft MBW, et al. Clinical implications of monitoring circulating tumor DNA in patients with colorectal cancer. Clin Cancer Res. 2017;23:5437–45.
  35. Taieb J, Taly V, Henriques J, et al. Prognostic value and relation with adjuvant treatment duration of ctDNA in stage III colon cancer: a post hoc analysis of the PRODIGE-GERCOR IDEA-France trial. Clin Cancer Res. 2021;27:5638–46.
  36. Tarazona N, Gimeno-Valiente F, Gambardella V, et al. Targeted next-generation sequencing of circulating-tumor DNA for tracking minimal residual disease in localized colon cancer. Ann Oncol. 2019;30:1804–12.
  37. Denis JA, Guillerm E, Coulet F, et al. The role of BEAMing and digital PCR for multiplexed analysis in molecular oncology in the era of next-generation sequencing. Mol Diagn Ther. 2017;21:587–600.
  38. Moati E, Taly V, Garinet S, et al. Role of circulating tumor DNA in gastrointestinal cancers: current knowledge and perspectives. Cancers (Basel). 2021;13:4743.
  39. Woodhouse R, Li M, Hughes J, et al. Clinical and analytical validation of FoundationOne Liquid CDx, a novel 324-Gene cfDNA-based comprehensive genomic profiling assay for cancers of solid tumor origin. PLoS One. 2020;15:e0237802.
  40. Singh RR. Next-generation sequencing in high-sensitive detection of mutations in tumors: challenges, advances, and applications. J Mol Diagn. 2020;22:994–1007.
  41. Li H, Jing C, Wu J, et al. Circulating tumor DNA detection: a potential tool for colorectal cancer management. Oncol Lett. 2019;17:1409–16.
  42. Gale D, Lawson ARJ, Howarth K, et al. Development of a highly sensitive liquid biopsy platform to detect clinically-relevant cancer mutations at low allele fractions in cell-free DNA. PLoS One. 2018;13:e0194630.
  43. Forshew T, Murtaza M, Parkinson C, et al. Noninvasive identification and monitoring of cancer mutations by targeted deep sequencing of plasma DNA. Sci Transl Med. 2012;4:136ra68.
  44. Kinde I, Wu J, Papadopoulos N, et al. Detection and quantification of rare mutations with massively parallel sequencing. Proc Natl Acad Sci U S A. 2011;108:9530–5.
  45. Bratman SV, Newman AM, Alizadeh AA, Diehn M. Potential clinical utility of ultrasensitive circulating tumor DNA detection with CAPP-Seq.Expert Rev Mol Diagn. 2015;15:715–9.
  46. Parikh AR, Seventer EEV, Siravegna G, et al. Minimal residual disease detection using a plasma-only circulating tumor DNA assay in patients with colorectal cancer. Clin Cancer Res. 2021;27:5586–94.
  47. Ococks E, Frankell AM, Masque Soler N, et al. Longitudinal tracking of 97 esophageal adenocarcinomas using liquid biopsy sampling. Ann Oncol. 2021;32:522–32.
  48. Yang L, Jiang J, Ye S, et al. Circulating tumor DNA analysis to detect minimal residual disease and predict recurrence in patients with resectable pancreatic cancer. J Clin Oncol. 2020;38(Suppl.):e16799.
  49. Lee JS, Rhee TM, Pietrasz D, et al. Circulating tumor DNA as a prognostic indicator in resectable pancreatic ductal adenocarcinoma: a systematic review and meta-analysis. Sci Rep. 2019;9:16971.
  50. Coombes RC, Page K, Salari R, et al. Personalized detection of circulating tumor DNA antedates breast cancer metastatic recurrence. Clin Cancer Res. 2019;25:4255–63.
  51. Sethi H, Salari R, Navarro S, et al. Analytical validation of the Signatera™ RUO assay, a highly sensitive patient-specific multiplex PCR NGS-based noninvasive cancer recurrence detection and therapy monitoring assay. Cancer Res. 2018;78(13 Suppl.):4542.
  52. Powles T, Assaf ZJ, Davarpanah N, et al. ctDNA guiding adjuvant immunotherapy in urothelial carcinoma. Nature. 2021;595:432–7.
  53. Tie J, Cohen JD, Wang Y, et al. Serial circulating tumour DNA analysis during multimodality treatment of locally advanced rectal cancer: a prospective biomarker study. Gut. 2019;68:663–71.
  54. Lee B, Lipton L, Cohen J, et al. Circulating tumor DNA as a potential marker of adjuvant chemotherapy benefit following surgery for localized pancreatic cancer. Ann Oncol. 2019;30:1472–8.
  55. Khakoo S, Carter PD, Brown G, et al. MRI tumor regression grade and circulating tumor DNA as complementary tools to assess response and guide therapy adaptation in rectal cancer. Clin Cancer Res. 2020;26:183–92.
  56. Diehn M, Alizadeh AA, Adams HP, et al. Early prediction of clinical outcomes in resected stage II and III colorectal cancer (CRC) through deep sequencing of circulating tumor DNA (ctDNA). J Clin Oncol. 2017;35(Suppl.):3591.
  57. Reinert T, Henriksen TV, Christensen E, et al. Analysis of plasma cell-free DNA by ultradeep sequencing in patients with stages I to III colorectal cancer. JAMA Oncol. 2019;5:1124.
  58. Henriksen TV, Tarazona N, Reinert T, et al. Circulating tumor DNA analysis for assessment of recurrence risk, benefit of adjuvant therapy, and early relapse detection after treatment in colorectal cancer patients. J Clin Oncol. 2021;39(Suppl.):11.
  59. Henriksen TVV, Tarazona N, Frydendahl A, et al. Serial circulating tumor DNA analysis to assess recurrence risk, benefit of adjuvant therapy, growth rate and early relapse detection in stage III colorectal cancer patients. J Clin Oncol. 2021;39(Suppl.):3540.
  60. Yukami H, Nakamura Y, Watanabe J, et al. Minimal residual disease by circulating tumor DNA analysis for colorectal cancer patients receiving radical surgery: an initial report from CIRCULATE-Japan. J Clin Oncol. 2021;39(Suppl.):3608.
  61. Abbosh C, Birkbak NJ, Wilson GA, et al. Phylogenetic ctDNA analysis depicts early-stage lung cancer evolution. Nature. 2017;545:446–51.
  62. Christensen E, Birkenkamp-Demtröder K, Sethi H, 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. J Clin Oncol. 2019;37:1547–57.
  63. Chapman JS, Pierson WE, Smith-McCune K, et al. Circulating tumor DNA predicts disease recurrence in ovarian cancer patients. Cancer Res. 2021;81(Suppl.):552.
  64. Dhakal B, Sharma S, Shchegrova S, et al. Personalized, ctDNA analysis to detect minimal residual disease and identify patients at high risk of relapse with multiple myeloma. J Clin Oncol. 2021;39(Suppl.):8029.
  65. Ococks E, Frankell AM, Masque Soler N, et al. Longitudinal tracking of 97 esophageal adenocarcinomas using liquid biopsy sampling. Ann Oncol. 2021;32:522–32.
  66. Signatera Residual Disease Test (MRD). Tissue sample collection instructions [Brochure]. San Carlos, CA: Natera, Inc, 2019.
  67. Signatera Residual Disease Test (MRD). A personalized, tumor-informed approach to detect molecular residual disease with high sensitivity and specificity. Design and analytical validation of Signatera™, the first ctDNA assay custom-built for detecting MRD and assessing treatment response. [Brochure]. San Carlos, CA: Natera, Inc, 2019.
  68. Van’t Erve I, Rovers KP, Constantinides A, et al. Detection of tumor-derived cell-free DNA from colorectal cancer peritoneal metastases in plasma and peritoneal fluid. J Pathol Clin Res. 2021;7:203–8.
  69. Bando H, Nakamura Y, Taniguchi H, et al. Impact of a metastatic site on circulating tumor DNA (ctDNA) analysis in patients (pts) with metastatic colorectal cancer (mCRC). J Clin Oncol. 2021;39(Suppl.):3554.
  70. von Bonin M, Jambor HK, Teipel R, et al. Clonal hematopoiesis and its emerging effects on cellular therapies. Leukemia. 2021;35:2752–8.
  71. Huang F, Yang Y, Chen X, et al. Chemotherapy-associated clonal hematopoiesis mutations should be taken seriously in plasma cell-free DNA KRAS/NRAS/BRAF genotyping for metastatic colorectal cancer. Clin Biochem. 2021;92:46–53.
  72. Gibson CJ, Steensma DP. New insights from studies of clonal hematopoiesis. Clin Cancer Res. 2018;24:4633–42.
  73. Hu Y, Ulrich BC, Supplee J, et al. False-positive plasma genotyping due to clonal hematopoiesis. Clin Cancer Res. 2018;24:4437–43.
  74. Dasari A, Morris VK, Allegra CJ, et al. ctDNA applications and integration in colorectal cancer: an NCI Colon and Rectal–Anal Task Forces whitepaper. Nat Rev Clin Oncol.2020;17:757–70.
  75. 2021. FDA grants two new breakthrough device designations for Natera’s Signatera™ MRD test. Available at: 1 February 2022).
  76. 2020. Natera receives final medicare coverage for its Signatera™ MRD test in stage II-III colorectal cancer. Available at: 1 February 2022).
  77. Sonoda H, Yamada T, Matsuda A, et al. Elevated serum carcinoembryonic antigen level after curative surgery is a prognostic biomarker of stage II-III colorectal cancer. Eur J Surg Oncol. 2021;47:2880–7.
  78. Osumi H, Shinozaki E, Ooki A, et al. Correlation between circulating tumor DNA and carcinoembryonic antigen levels in patients with metastatic colorectal cancer. Cancer Med. 2021;10:8820–8.
  79. Morris VK, Yothers G, Kopetz S, et al. NRG-GI005 (COBRA): phase II/III study of circulating tumor DNA as a predictive biomarker in adjuvant chemotherapy in patients with stage II colon cancer. J Clin Oncol. 2020;38(Suppl.):TPS261.
  80. Weinberg BA, Winslow ER, Bayasi M, et al. Early detection of circulating tumor DNA postoperatively enables discovery of resectable metastatic disease in a patient with colon cancer. Case Rep Oncol. 2021;14:1748–53.
  81. Kanemitsu Y, Shimizu Y, Mizusawa J, et al. Hepatectomy followed by mFOLFOX6 versus hepatectomy alone for liver-only metastatic colorectal cancer (JCOG0603): A phase II or III randomized controlled trial. J Clin Oncol. 2021;39:3789–99.
  82. Grothey A, Sobrero AF, Shields AF, et al. Duration of adjuvant chemotherapy for stage III colon cancer. N Engl J Med. 2018;378:1177–88.
  83. Hong TS, Ryan DP. Adjuvant chemotherapy for locally advanced rectal cancer: is it a given? J Clin Oncol. 2015;33:1878–80.
  84. Morais M, Pinto DM, Machado JC, Carneiro S. ctDNA on liquid biopsy for predicting response and prognosis in locally advanced rectal cancer: A systematic review. Eur J Surg Oncol. 2022;48:218–27.
  85. Massihnia D, Pizzutilo EG, Amatu A, et al. Liquid biopsy for rectal cancer: A systematic review. Cancer Treat Rev. 2019;79:101893.
  86. Wang Y, Yang L, Bao H, et al. Utility of ctDNA in predicting response to neoadjuvant chemoradiotherapy and prognosis assessment in locally advanced rectal cancer: a prospective cohort study. PLoS Med. 2021;18:e1003741.
  87. Sarantis P, Koustas E, Papadimitropoulou A, et al. Pancreatic ductal adenocarcinoma: treatment hurdles, tumor microenvironment and immunotherapy. World J Gastrointest Oncol. 2020;12:173–81.
  88. American Cancer Society. 2022. Cancer statistics center. Pancreas. Available at:!/cancer-site/Pancreas(accessed 24 January 2022).
  89. Allen PJ, Kuk D, Castillo CFD, et al. Multi-institutional validation study of the American Joint Commission on Cancer (8th Edition) changes for T and N staging in patients with pancreatic adenocarcinoma. Ann Surg. 2017;265:185–91.
  90. Grunvald MW, Jacobson RA, Kuzel TM, et al. Current status of circulating tumor DNA liquid biopsy in pancreatic cancer. Int J Mol Sci. 2020;21:E7651.
  91. Park G, Park JK, Son DS, et al. Utility of targeted deep sequencing for detecting circulating tumor DNA in pancreatic cancer patients. Sci Rep. 2018;8:11631.
  92. Sivapalan L, Kocher HM, Ross-Adams H, Chelala C. Molecular profiling of ctDNA in pancreatic cancer: opportunities and challenges for clinical application. Pancreatology. 2021;21:363–78.
  93. Perets R, Greenberg O, Shentzer T, et al. Mutant KRAS circulating tumor DNA is an accurate tool for pancreatic cancer monitoring. Oncologist. 2018;23:566–72.
  94. Pietrasz D, Pécuchet N, Garlan F, et al. Plasma circulating tumor DNA in pancreatic cancer patients is a prognostic marker. Clin Cancer Res. 2017;23:116–23.
  95. Alqahtani A, Alloghbi A, Yin C, et al. Prognostic utility of preoperative and postoperative circulating tumor DNA (ctDNA) in resected pancreatic ductal adenocarcinoma: A systematic review and meta-analysis. J Clin Oncol. 2022;40(Suppl.):595.
  96. Botta GP, Abdelrahim M, Aushev VN, et al. Association of personalized and tumor-informed ctDNA with patient survival outcomes in pancreatic adenocarcinoma. J Clin Oncol. 2022;40(Suppl.):517.
  97. Takai E, Totoki Y, Nakamura H, et al. Clinical utility of circulating tumor DNA for molecular assessment in pancreatic cancer. Sci Rep. 2015;5:18425.
  98. Hadano N, Murakami Y, Uemura K, et al. Prognostic value of circulating tumour DNA in patients undergoing curative resection for pancreatic cancer. Br J Cancer. 2016;115:59–65.
  99. Nakano Y, Kitago M, Matsuda S, et al. KRAS mutations in cell-free DNA from preoperative and postoperative sera as a pancreatic cancer marker: a retrospective study. Br J Cancer. 2018;118:662–9.
  100. Yang X, Xu W, Tian X, et al. Diagnostic and prognostic value of KRAS mutations in circulating pancreatic ductal adenocarcinoma tumor DNA. Transl Cancer Res. 2018;7.
  101. Kim MK, Woo SM, Park B, et al. Prognostic implications of multiplex detection of KRAS mutations in cell-free DNA from patients with pancreatic ductal adenocarcinoma. Clin Chem. 2018;64:726–34.
  102. Groot VP, Mosier S, Javed AA, et al. Circulating tumor DNA as a clinical test in resected pancreatic cancer. Clin Cancer Res. 2019;25:4973–84.
  103. Watanabe F, Suzuki K, Tamaki S, et al. Longitudinal monitoring of KRAS-mutated circulating tumor DNA enables the prediction of prognosis and therapeutic responses in patients with pancreatic cancer. PLoS One. 2019;14:e0227366.
  104. Guo S, Shi X, Shen J, et al. Preoperative detection of KRAS G12D mutation in ctDNA is a powerful predictor for early recurrence of resectable PDAC patients. Br J Cancer. 2020;122:857–67.
  105. Okada T, Mizukami Y, Ono Y, et al. Digital PCR-based plasma cell-free DNA mutation analysis for early-stage pancreatic tumor diagnosis and surveillance. J Gastroenterol. 2020;55:1183–93.
  106. Jiang J, Ye S, Xu Y, et al. Circulating tumor DNA as a potential marker to detect minimal residual disease and predict recurrence in pancreatic cancer. Front Oncol. 2020;10:1220.
  107. Yamaguchi T, Uemura K, Murakami Y, et al. Clinical implications of pre- and postoperative circulating tumor DNA in patients with resected pancreatic ductal adenocarcinoma. Ann Surg Oncol. 2021;28:3135–44.
  108. Botta GP, Abdelrahim M, Aushev VN, et al. Association of personalized and tumor-informed ctDNA with patient survival outcomes in pancreatic adenocarcinoma. J Clin Oncol. 2022;40(Suppl.):517.
  109. Azad TD, Chaudhuri AA, Fang P, et al. Circulating tumor DNA analysis for detection of minimal residual disease after chemoradiotherapy for localized esophageal cancer. Gastroenterology. 2020;158:494–505.e6.
  110. Huffman BM, Budde G, Chao J, et al. Performance of a tumor-informed circulating tumor DNA assay from over 250 patients with over 600 plasma time points in esophageal and gastric cancer. Ann Oncol. 2021;32(Suppl.):S1040–75.
  111. Kasi PM, Budde G, Dayyani F, et al. Tumor-informed assessment of circulating tumor DNA and its incorporation into practice for patients with hepatobiliary cancers. J Clin Oncol. 2021;39(Suppl.):4103.
  112. Wang J, Huang A, Wang YP, et al. Circulating tumor DNA correlates with microvascular invasion and predicts tumor recurrence of hepatocellular carcinoma. Ann Transl Med. 2020;8:237.
  113. Yoo C, Ryu MH, Na YS, et al. Analysis of serum protein biomarkers, circulating tumor DNA, and dovitinib activity in patients with tyrosine kinase inhibitor-refractory gastrointestinal stromal tumors. Ann Oncol. 2014;25:2272–7.
  114. Johansson G, Berndsen M, Lindskog S, et al. Monitoring circulating tumor DNA during surgical treatment in patients with gastrointestinal stromal tumors. Mol Cancer Ther. 2021;20:2568–76.
  115. Cabel L, Jeannot E, Bieche I, et al. Prognostic impact of residual HPV ctDNA detection after chemoradiotherapy for anal squamous cell carcinoma. Clin Cancer Res. 2018;24:5767–71.
  116. Baccelli I, Schneeweiss A, Riethdorf S, et al. Identification of a population of blood circulating tumor cells from breast cancer patients that initiates metastasis in a xenograft assay. Nat Biotechnol. 2013;31:539–44.
  117. Zhang Y, Zarrabi K, Hou W, et al. Assessing clinical outcomes in colorectal cancer with assays for invasive circulating tumor cells. Biomedicines. 2018;6:69.
  118. Gorges TM, Stein A, Quidde J, et al. Improved detection of circulating tumor cells in metastatic colorectal cancer by the combination of the CellSearch®system and the AdnaTest®PLoS One. 2016;11:e0155126.
  119. Bork U, Rahbari NN, Schölch S, et al. Circulating tumour cells and outcome in non-metastatic colorectal cancer: a prospective study. Br J Cancer. 2015;112:1306–13.
  120. Cohen SJ, Punt CJA, Iannotti N, et al. Relationship of circulating tumor cells to tumor response, progression-free survival, and overall survival in patients with metastatic colorectal cancer. J Clin Oncol. 2008;26:3213–21.
  121. Kidess-Sigal E, Liu HE, Triboulet MM, et al. Enumeration and targeted analysis of KRAS, BRAF and PIK3CA mutations in CTCs captured by a label-free platform: comparison to ctDNA and tissue in metastatic colorectal cancer. Oncotarget. 2016;7:85349–64.
  122. Bahnassy AA, Salem SE, Mohanad M, et al. Prognostic significance of circulating tumor cells (CTCs) in Egyptian non-metastatic colorectal cancer patients: a comparative study for four different techniques of detection (Flowcytometry, CellSearch, Quantitative Real-time PCR and Cytomorphology). Exp Mol Pathol. 2019;106:90–101.
  123. Iinuma H, Watanabe T, Mimori K, et al. Clinical significance of circulating tumor cells, including cancer stem-like cells, in peripheral blood for recurrence and prognosis in patients with Dukes’ stage B and C colorectal cancer. J Clin Oncol. 2011;29:1547–55.
  124. Yang C, Chen F, Wang S, Xiong B. Circulating tumor cells in gastrointestinal cancers: current status and future perspectives. Front Oncol. 2019;9:1427.
  125. Lu CY, Uen YH, Tsai HL, et al. Molecular detection of persistent postoperative circulating tumour cells in stages II and III colon cancer patients via multiple blood sampling: prognostic significance of detection for early relapse. Br J Cancer. 2011;104:1178–84.
  126. Galizia G, Gemei M, Orditura M, et al. Postoperative detection of circulating tumor cells predicts tumor recurrence in colorectal cancer patients. J Gastrointest Surg. 2013;17:1809–18.
  127. Sotelo MJ, Sastre J, Maestro ML, et al. Role of circulating tumor cells as prognostic marker in resected stage III colorectal cancer. Ann Oncol. 2015;26:535–41.
  128. Haber DA, Velculescu VE. Blood-based analyses of cancer: circulating tumor cells and circulating tumor DNA.Cancer Discov. 2014;4:650–61.
  129. Xiao J, McGill JR, Stanton K, et al. Efficient propagation of circulating tumor cells: a first step for probing tumor metastasis. Cancers (Basel). 2020;12:2784.
  130. Riethdorf S, Fritsche H, Müller V, et al. Detection of circulating tumor cells in peripheral blood of patients with metastatic breast cancer: a validation study of the CellSearch system. Clin Cancer Res. 2007;13:920–8.

Further Resources

Share this Article
Related Content In Gastrointestinal Oncology
  • Copied to clipboard!
    accredited arrow-down-editablearrow-downarrow_leftarrow-right-bluearrow-right-dark-bluearrow-right-greenarrow-right-greyarrow-right-orangearrow-right-whitearrow-right-bluearrow-up-orangeavatarcalendarchevron-down consultant-pathologist-nurseconsultant-pathologistcrosscrossdownloademailexclaimationfeedbackfiltergraph-arrowinterviewslinkmdt_iconmenumore_dots nurse-consultantpadlock patient-advocate-pathologistpatient-consultantpatientperson pharmacist-nurseplay_buttonplay-colour-tmcplay-colourAsset 1podcastprinter scenerysearch share single-doctor social_facebooksocial_googleplussocial_instagramsocial_linkedin_altsocial_linkedin_altsocial_pinterestlogo-twitter-glyph-32social_youtubeshape-star (1)tick-bluetick-orangetick-red tick-whiteticktimetranscriptup-arrowwebinar Sponsored Department Location NEW TMM Corporate Services Icons-07NEW TMM Corporate Services Icons-08NEW TMM Corporate Services Icons-09NEW TMM Corporate Services Icons-10NEW TMM Corporate Services Icons-11NEW TMM Corporate Services Icons-12Salary £ TMM-Corp-Site-Icons-01TMM-Corp-Site-Icons-02TMM-Corp-Site-Icons-03TMM-Corp-Site-Icons-04TMM-Corp-Site-Icons-05TMM-Corp-Site-Icons-06TMM-Corp-Site-Icons-07TMM-Corp-Site-Icons-08TMM-Corp-Site-Icons-09TMM-Corp-Site-Icons-10TMM-Corp-Site-Icons-11TMM-Corp-Site-Icons-12TMM-Corp-Site-Icons-13TMM-Corp-Site-Icons-14TMM-Corp-Site-Icons-15TMM-Corp-Site-Icons-16TMM-Corp-Site-Icons-17TMM-Corp-Site-Icons-18TMM-Corp-Site-Icons-19TMM-Corp-Site-Icons-20TMM-Corp-Site-Icons-21TMM-Corp-Site-Icons-22TMM-Corp-Site-Icons-23TMM-Corp-Site-Icons-24TMM-Corp-Site-Icons-25TMM-Corp-Site-Icons-26TMM-Corp-Site-Icons-27TMM-Corp-Site-Icons-28TMM-Corp-Site-Icons-29TMM-Corp-Site-Icons-30TMM-Corp-Site-Icons-31TMM-Corp-Site-Icons-32TMM-Corp-Site-Icons-33TMM-Corp-Site-Icons-34TMM-Corp-Site-Icons-35TMM-Corp-Site-Icons-36TMM-Corp-Site-Icons-37TMM-Corp-Site-Icons-38TMM-Corp-Site-Icons-39TMM-Corp-Site-Icons-40TMM-Corp-Site-Icons-41TMM-Corp-Site-Icons-42TMM-Corp-Site-Icons-43TMM-Corp-Site-Icons-44TMM-Corp-Site-Icons-45TMM-Corp-Site-Icons-46TMM-Corp-Site-Icons-47TMM-Corp-Site-Icons-48TMM-Corp-Site-Icons-49TMM-Corp-Site-Icons-50TMM-Corp-Site-Icons-51TMM-Corp-Site-Icons-52TMM-Corp-Site-Icons-53TMM-Corp-Site-Icons-54TMM-Corp-Site-Icons-55TMM-Corp-Site-Icons-56TMM-Corp-Site-Icons-57TMM-Corp-Site-Icons-58TMM-Corp-Site-Icons-59TMM-Corp-Site-Icons-60TMM-Corp-Site-Icons-61TMM-Corp-Site-Icons-62TMM-Corp-Site-Icons-63TMM-Corp-Site-Icons-64TMM-Corp-Site-Icons-65TMM-Corp-Site-Icons-66TMM-Corp-Site-Icons-67TMM-Corp-Site-Icons-68TMM-Corp-Site-Icons-69TMM-Corp-Site-Icons-70TMM-Corp-Site-Icons-71TMM-Corp-Site-Icons-72