Abstract
Background
Recurrence of colorectal cancer has been linked to the presence of methylated circulating tumour DNA (ctDNA) in patient plasma after surgery. The prognostic significance of ctDNA before treatment remains unclear. This study investigated the correlation between pretreatment ctDNA and current radiological [magnetic resonance imaging (MRI)] prognostic markers in patients with rectal cancer and its association with recurrence-free survival and overall survival (OS).
Patients and methods
A total of 42 patients with rectal cancer were enrolled. All patients had staging MRI before treatment. Blood was taken at diagnosis for ctDNA analysis for the presence of either methylated branched chain amino acid transaminase 1 (BCAT1) or IKAROS family zinc finger 1 (IKZF1). The correlation of MRI prognostic indicators and ctDNA test results was assessed with chi-square tests. Univariable and multivariate Cox regression analyses were carried out to determine variables associated with recurrence-free survival and OS.
Results
The mean age of patients was 64.4 years (standard deviation 12.5 years), and the majority were male (30/42, 71.4%). A total of 11, 13, 9 and 9 patients were in stages I, II, III and IV, respectively. Patients were followed up for a minimum of 36 months unless disease recurrence or death occurred earlier. A total of 36 (85.7%) patients received neoadjuvant chemoradiotherapy, and 30 (71.4%) underwent surgical resection. The 3-year survival rate was 64%. About 67% (28/42) of patients were positive for the methylated ctDNA at diagnosis. Further, 11 out of 12 patients with a positive circumferential resection margin (CRM+) were ctDNA positive; univariable analysis showed that prognostic indicators for OS were presence of extramural venous invasion [EMVI; hazard ratio (HR) 2.63, 95% confidence interval (CI) 0.95-7.31], CRM+ (HR 10.69, 95% CI 3.51-32.56), metastatic disease (HR 7.7, 95% CI 2.79-21.67) and ctDNA% methylation (HR 1.04, 95% CI 1.02-1.06). The presence of CRM+ and a positive ctDNA had an HR of 19.57 (95% CI 3.47-110.49). In the multivariate analysis, including adjustment for age and EMVI, only the CRM+/ctDNA+ variable was an independent predictor for poor survival (HR 19.57, 95% CI 3.47-110.49).
Conclusions
In rectal cancer, almost all patients with CRM involvement have ctDNA, and these patients had the worst prognosis. Future studies with longitudinal ctDNA assessment before and after treatment may potentially inform prognosis and help tailor patients’ treatment.
Key words: prognosis, survival, MRI, methylated ctDNA, BCAT1, IKZF1, rectal cancer
Highlights
-
•
Methylated ctDNA was detected in 67% of patients at diagnosis of rectal cancer.
-
•
MRI-defined CRM involvement and the percentage methylation at diagnosis were both associated with worse OS outcomes.
-
•
Methylated ctDNA and MRI assessment are complementary to inform prognosis and therapy decisions in rectal cancer.
Introduction
Management of rectal cancer is stage dependent and requires a multidisciplinary approach. In most patients with locally advanced rectal cancer, neoadjuvant therapy that involves concurrent long-course chemoradiotherapy (LCCRT) followed by surgery is the standard of care.1 However, there has been a paradigm shift in the management of locally advanced rectal cancer. The shift is towards a strategy of total neoadjuvant therapy (TNT) that incorporates neoadjuvant systemic chemotherapy and CRT and an emphasis on avoiding total mesorectal excision altogether if there is a complete clinical response to neoadjuvant therapy (referred to as the ‘watch and wait’ approach).2, 3, 4, 5 After neoadjuvant therapy with either LCCRT or TNT and surgery, ∼30% of cases can achieve complete pathologic response. TNT can also lead to complete clinical response, leading to surgery-free survival in ∼50% of patients.2,4,6 Although the risk of recurrence is reduced in individuals with a pathological complete response,7 distant recurrences to lung, bone and liver occur at a relatively high frequency.8 It is therefore important to have sensitive tools to assess prognosis and to be able to guide the best treatment approach.
Magnetic resonance imaging (MRI) plays a crucial role in the staging and response assessment to treatment in rectal cancer.9,10 High-resolution MRI not only can provide accurate staging but also detect tumour-specific high-risk features that are associated with worse cancer-related outcomes.11 MRI-detected circumferential resection margin (CRM) involvement, extramural tumour deposits and the presence of extramural venous invasion (EMVI) are high-risk features that predict worse outcomes.12,13 Based on this, MRI can aid towards a personalised multimodal treatment that can be tailored for patients with rectal cancer in the preoperative setting. Importantly, MRI is required for accurate treatment response assessment and forms the backbone of ongoing surveillance for patients who are deemed suitable for a non-operative approach.12,14 However, MRI has limitations in staging, particularly with detecting nodal involvement, with a study showing that 43% of patients were incorrectly staged for nodal involvement (overstaged) and 24% were understaged.15 Incorrect staging has consequences for treatment decisions. Clinical care would benefit from additional approaches, such as blood biomarkers, to help ascertain tumour burden for guiding treatment.
Currently there are no validated blood biomarkers that are routinely used to accurately predict prognosis and/or treatment response to neoadjuvant treatment in locally advanced curable rectal cancer. Circulating tumour DNA (ctDNA) could be a putative candidate in this regard,16, 17, 18 with selected ctDNA biomarkers being either tumour agnostic or tumour informed. The latter biomarker class requires a two-step process of first sequencing of the primary tumour to identify the mutation profile, with an assay then developed to detect the same mutations in the circulating cell-free DNA.17,18 Incorporation of methylated ctDNA (tumour agnostic) or mutation ctDNA (tumour informed) as a noninvasive test pretreatment can improve treatment selection by potentially providing accurate prognostic information.16,19,20 Several studies have assessed whether ctDNA levels before treatment are associated with prognosis and treatment response, but results vary between studies, with only a few finding that baseline ctDNA was significantly associated with disease-free survival and/or overall survival (OS).21 However, a combination of preoperative ctDNA levels with other clinically used variables may improve prognostic measures.
MRI-based tumour regression grade (mrTRG) provides an assessment of disease response to neoadjuvant treatment in rectal cancer and has been shown to provide prognostic information for progression-free survival and OS.22 However, it is a subjective measure that relies on radiologist experience. Hence it is intuitive to evaluate a blood-based noninvasive biomarker along with MRI parameters that can support treatment strategy decisions. Our prior research has shown that aberrant methylation of BCAT1 (branched chain amino acid transaminase 1) and IKZF1 (IKAROS family zinc finger 1) within colorectal tumours is a more frequent event compared with mutations, with the methylated DNA detected in tumour tissue throughout the stages of development.23 It is a tumour-agnostic biomarker that can be used to detect ctDNA, with preoperative blood levels significantly associated with the stage of disease and tumour burden,23,24 and post-treatment levels associated with minimal residual disease and risk for recurrence.19,20,25 However, the association of ctDNA methylated in BCAT1/IKZF1 and MRI-based factors with prognosis is unknown in patients with rectal cancer. The aim of this study was to investigate the correlation of ctDNA methylated in BCAT1/IKZF1 with MRI-detected risk factors in a cohort of rectal cancer and if these measures could be used to identify individuals with poorer outcomes.
Materials and methods
Overview
Cell-free DNA was extracted from plasma samples of 42 patients at diagnosis of rectal cancer (all stages) and assayed for ctDNA with a minimum follow-up of 3 years. The presence of either methylated BCAT1 or IKZF1 in plasma was deemed a positive ctDNA result with levels reported as % methylation relative to total DNA. All patients had a staging rectal MRI before treatment. Clinical measures, ctDNA results and MRI variables were assessed for association with recurrence-free survival and OS.
Study population and clinical procedures
This was a retrospective analysis of a prospective cohort, single-centre study. The study was approved by the Southern Adelaide Clinical Human Research Ethics Committee (ethics approval number 134.045) and written informed consent was obtained before study enrolment. Study registration is through the Australian and New Zealand Clinical Trials Registry (registration number 12611000318987).
Patients aged ≥18 years, with a diagnosis of rectal adenocarcinoma confirmed on histology (all stages) were invited to the study at Flinders Medical Centre (South Australia) between 2011 and 2017. Exclusion criteria were patients without invasive malignancy, nonadenocarcinoma histology, absence of pretreatment MRI staging, inadequate staging or failure to meet blood sampling requirements.
Venous blood (18 ml; K3-EDTA VACUETTE tubes; Greiner Bio-One, Frickenhausen, Germany) was collected before initial therapy for analysis for methylated BCAT1 and IKZF1 DNA. Patient demographics, histopathology, imaging and treatment details were collected from case notes. Radiological staging of rectal cancer was based on pretreatment staging MRI in addition to computed tomography staging. A review of the staging MRI scans was carried out by one author (KG) with an assessment of tumour size and location (distance from the anal verge, craniocaudal length) and potential prognostic indicators (T stage, depth of extramural extension, mucinous component, continuous EMVI, N stage, number of mesorectal lymph nodes, pelvic side wall nodes, tumour deposits and CRM involvement).
Individuals were followed up according to routine institutional practice. Follow-up was for a minimum of 3 years unless recurrence or death occurred at an earlier time point. Clinical information including flexible sigmoidoscopy and radiology (computed tomography, MRI and positron emission tomography if available) was reviewed to establish recurrence status. Recurrence was defined as locoregional when present at the site of anastomosis or in draining lymph nodes. Evidence of recurrence in the liver, lung or other distant organs was defined as distant recurrence, with this being the principal diagnosis when locoregional recurrence was also present.
Analysis of plasma samples for methylated ctDNA
Samples were analysed as previously described.26 In brief, after centrifuging two times (1500g for 10 min at 4°C with deceleration at the lowest setting), plasma was isolated from whole blood samples and stored at –80°C until analysis.
Analysis was carried out at the Clinical Genomics Laboratory (New South Wales, Australia) in 3.9 ml aliquots of plasma, by staff blinded to clinical results. The ctDNA test method uses commercially available reagents and is based on an automated procedure for DNA extraction, bisulphite conversion and purification, PCR setup and calculation of levels of methylated BCAT1 and IKZF1, as previously described.26 Samples were analysed in triplicate, and a simultaneous assay for ACTB was carried out as a measure for DNA quantity and sample quality. A sample was deemed positive if at least one PCR replicate was positive for either BCAT1 or IKZF1 DNA methylation. The quantitative methylation level was calculated for each sample as the percentage of BCAT1 and IKZF1 methylated DNA relative to the reference gene ACTB (average pg of DNA per well).
Statistical analysis
Descriptive variables and ctDNA levels were expressed as either mean with standard deviation (SD) or median with interquartile range (IQR). Chi-square analysis was used to compare differences between categorical variables. Recurrence-free survival and OS were assessed. Recurrence-free survival was measured from the end of CRT or surgery to local or distant tumour recurrence or death from any cause and was censored at the last follow-up. OS was measured from the end of treatment to death from any cause or censored by the last follow-up if alive. The Kaplan–Meier method was used for the survival estimates while Cox proportional hazard models were used to compare the survival rates between groups and to estimate the hazard ratios (HRs) with a 95% confidence interval (CI). Univariable and multivariate Cox regression analyses were carried out to determine variables associated with OS. The association between prognostic indicators measured at MRI and the ctDNA results was assessed with chi-square tests. Statistical analyses were carried out using STATA (version 13; StataCorp, College Station, TX), with a P value <0.05 considered statistically significant.
Results
Patient characteristics
A total of 42 patients were assessable for the ctDNA levels and MRI staging at diagnosis. Baseline characteristics are summarised in Table 1. The mean age was 64.4 years (SD 12.5 years). Most patients were male (30/42, 71.4%). The stage of disease at diagnosis was 55% locally advanced (stage II-III). As many as 36 (85.7%) patients received neoadjuvant therapy, 29 (69.0%) underwent surgical resection and 24 (57.1%) received adjuvant chemotherapy.
Table 1.
Patient demographics and treatment methods
| Characteristic | Eligible for baseline analysis (n = 42) | Eligible for survival analysis (n = 34) |
|---|---|---|
| Age, mean (standard deviation), years | 64.4 (12.5) | 68.4 (13.4) |
| Male sex, n (%) | 30 (71.4) | 25 (73.5) |
| Stage of rectal cancer, n (%) | ||
| I | 10 (23.8) | 9 (26.5) |
| II | 11 (26.2) | 11 (32.4) |
| III | 12 (28.6) | 12 (35.3) |
| IV | 9 (21.4) | 2 (5.9) |
| Neoadjuvant chemoradiotherapy or radiotherapy, surgical resection, n (%) | 9 (21.4) | 8 (23.5) |
| Neoadjuvant chemoradiotherapy or radiotherapy, surgical resection, adjuvant chemotherapy, n (%) | 15 (35.7) | 14 (41.2) |
| Neoadjuvant chemoradiotherapy or radiotherapy, with or without adjuvant chemotherapy, but no surgery, n (%) | 12 (28.6) | 6 (17.6) |
| Surgical resection, n (%) | 4 (9.5) | 4 (11.8) |
| Surgical resection, adjuvant chemotherapy, n (%) | 2 (4.8) | 2 (5.9) |
A total of 34 patients (81.0%) were considered cancer free or in remission after treatment, with at least 36 months of follow-up unless a recurrence of mortality occurred earlier. This subgroup had a lower proportion of stage IV patients (Table 1). In this cohort, 28 (82.4%) received neoadjuvant therapy, 28 (82.4%) had surgical resection and 19 (55.9%) underwent adjuvant chemotherapy.
MRI characteristics
A review of the staging MRI scans of the 42 patients resulted in 11 (26.2%) cases being downstaged at diagnosis and 3 (7.1%) being upstaged. Fifteen (35.7%) patients had low rectal tumours, with a mean craniocaudal length of 47.9 mm (SD 16.5 mm). All features found at the MRI review are summarised in Table 2, with the main prognostic indicators of the presence of continuous EMVI and CRM involvement being found in 13 (31.0%) and 12 (28.6%) of patients, respectively. Logistic regression analysis showed that the finding of continuous EMVI was significantly associated with the presence of tumour deposits [odds ratio (OR) 13.3, 95% CI 2.6-68.6, P = 0.002]. The involvement of CRM was significantly associated with the presence of metastatic disease (OR 9.0, 95% CI 1.7-46.6, P = 0.01), T4 stage (OR 20.7, 95% CI 2.1-206.6, P = 0.01), craniocaudal length (OR 1.07, 95% CI 1.01-1.12, P = 0.01) and maximum extramural spread (OR 1.5, 95% CI 1.2-1.9, P = 0.001).
Table 2.
Features identified with staging MRI of 42 rectal cancer cases
| MRI feature | Values |
|---|---|
| Distal from the anal verge (mm) | 70.1 (SD 37.3) |
| Location, n (%) | |
| Upper or mid-rectum | 27 (64.3) |
| Low rectum | 15 (35.7) |
| Craniocaudal length (mm) | 47.9 (SD 16.5) |
| Mucinous component, n (%) | 5 (11.9) |
| TNM stage, n (%) | |
| T1 | 1 (2.4) |
| T2 | 10 (23.8) |
| T3a/b | 14 (33.3) |
| T3c/d | 11 (26.1) |
| T4 | 6 (14.3) |
| N1/2 | 14 (33.3) |
| Continuous extramural venous invasion, n (%) | 13 (31.0) |
| Maximum extramural spread of tumour (mm) | 3.5 (IQR 0-7) |
| Number of mesorectal lymph nodes | 0-6 (IQR 0-1) |
| Presence of pelvic side wall nodes, n (%) | 3 (7.1) |
| Tumour deposits, n (%) | 11 (26.2) |
| CRM involved, n (%) | 12 (28.6) |
CRM, circumferential resection margin; IQR, interquartile range; MRI, magnetic resonance imaging; SD, standard deviation; TNM, tumour–node–metastasis.
Detection of methylated ctDNA
A total of 28 patients (66.7%) had a positive ctDNA at the time of diagnosis. This included 7 (16.7%) with a blood result positive for only methylated BCAT1, 4 (9.5%) with a blood result positive for only methylated IKZF1 and 17 (40.5%) positive for both biomarkers. A positive ctDNA result was associated with craniocaudal length (OR 1.1, 95% CI 1.01-1.11, P = 0.02) and stage at diagnosis (stage III: OR 7.0, 95% CI 1.07-45.90, P = 0.04; stage IV: OR 18.7, 95% CI 1.56-222.93, P = 0.02). Of the 28 patients with a pretreatment-positive ctDNA result, 39.3% had CRM involvement, which was a greater proportion than those with a negative ctDNA (7.1%, P = 0.03). There were no differences in the proportion of EMVI detected between ctDNA-positive cases (28.6%) and ctDNA-negative cases (35.7%, P = 0.64).
Methylated ctDNA value, MRI risk correlation and overall survival outcomes
At the time of analysis, the median time of follow-up after treatment was 51.0 months (IQR 40.4-56.7 months). Fifteen patients died during the follow-up time with a median time until the death of 30.8 months (IQR 17.8-46.3 months). Overall, the 3-year survival rate was 81.0%. Univariate analysis revealed extramural spread, involvement of CRM, metastatic disease and ctDNA percentage methylation as poor prognostic indicators for OS (Table 3). Of the individuals that died during the follow-up time, 80% had a positive ctDNA result and 66.7% had CRM involvement, with 9 out of 15 individuals having both features. CRM involvement and a positive ctDNA at the time of diagnosis correlated with worse survival outcomes (HR 19.57, 95 CI 3.47-110.49; Figure 1). In the multivariate analysis, including adjustment for the significant features shown in Table 3, M stage (HR 8.92, 95% CI 2.05-38.76) and the combined CRM/ctDNA presence (HR 20.42, 95% CI 1.95-213.54) were independent predictors for poor survival.
Table 3.
Univariable and multivariate analyses for overall survival
| Variable | Univariate | Multivariate | ||
|---|---|---|---|---|
| Hazard ratio (95% CI) | P value | Hazard ratio (95% CI) | P value | |
| Age (years) | 1.00 (0.96-1.04) | 0.977 | — | — |
| Male sex (versus female) | 0.62 (0.21-1.84) | 0.393 | — | — |
| Lower rectal (versus mid and upper) | 1.75 (0.63-4.85) | 0.280 | — | — |
| T4 (versus T1, T2 and T3) | 2.58 (0.81-8.21) | 0.109 | — | — |
| N1/2 (versus N0) | 1.03 (0.35-3.01) | 0.959 | — | — |
| M1 (versus M0) | 7.77 (2.79-21.67) | <0.001 | 8.92 (2.05-38.76) | 0.003 |
| Distance from the anal verge (mm) | 0.99 (0.98-1.01) | 0.340 | — | — |
| Craniocaudal length (mm) | 1.04 (1.01-1.07) | 0.014 | 1.04 (0.99-1.09) | 0.109 |
| Tumour deposits (versus no tumour deposits) | 1.77 (0.60-5.22) | 0.299 | — | — |
| Mucinous component (versus no mucinous component) | 2.07 (0.58-7.41) | 0.262 | — | — |
| Continuous EMVI (versus no EMVI) | 2.63 (0.95-7.31) | 0.063 | — | — |
| Maximum extramural spread of tumour (mm) | 1.11 (1.04-1.20) | 0.002 | 0.85 (0.72-1.00) | 0.055 |
| Number of mesorectal lymph nodes | 0.95 (0.61-1.49) | 0.830 | — | — |
| Presence of pelvic side wall nodes | 1.15 (0.15-8.93) | 0.896 | — | — |
| CRM involved (versus not involved) | 10.69 (3.51-32.56) | <0.001 | — | — |
| ctDNA % methylation | 1.04 (1.02-1.06) | <0.001 | — | — |
| CRM or ctDNA negative (versus both negative) | 1.75 (0.31-9.73) | 0.524 | 0.98 (0.13-7.33) | 0.984 |
| CRM and ctDNA positive (versus both negative) | 19.57 (3.47-110.49) | 0.001 | 20.42 (1.95-213.54) | 0.012 |
CI, confidence interval; CRM, circumferential resection margin; ctDNA, circulating tumour DNA; EMVI, extramural venous invasion.
Figure 1.
Kaplan–Meier curves for overall survival, with circumferential resection margin (CRM) and circulating tumour DNA (ctDNA; N = 42).
Factors associated with recurrence-free survival
There were 34 individuals that were considered disease free after treatment that could be assessed for recurrence. The median follow-up time after treatment was 51 months (IQR 40.4-56.7 months). There were 10 individuals that were diagnosed with recurrence with a median of 22.2 months (IQR 14.6-46.9 months) after treatment, with 20.6% (7/34) having a recurrence within 36 months. Nearly 70% of all recurrences were diagnosed in a distant location. MRI measures associated with recurrence were T4 stage (HR 4.04, 95% CI 1.00-16.28, P = 0.049) and involved CRM (HR 5.26, 95% CI 1.36-20.35, P = 0.016); however, these were not independent predictors when multivariate analysis was carried out (P > 0.05). There were 60.0% of the cases with recurrence that had a positive ctDNA before treatment and 40% with CRM involved; however, including ctDNA in the risk prediction model did not cause significant improvement, with an HR of 4.07 (95% CI 0.71-23.16, P = 0.114; Figure 2).
Figure 2.
Kaplan–Meier curves for recurrence-free survival, with circumferential resection margin (CRM) and circulating tumour DNA (ctDNA; N = 34).
Discussion
MRI is integral with regard to the management of rectal cancer, and recently MRI-specific tumour-associated risk factors, MRI-detected mrTRG and ctDNA have garnered a significant amount of interest as potential predictive and prognostic biomarkers in this disease.11,13,16 In this study we report an association between methylated ctDNA and MRI-specific risk factors in rectal cancer. This association is likely to be due to the higher risk of the presence of micro-metastatic disease in patients that have EMVI-positive, extramural tumour deposits and/or CRM-involved disease. Our results also confirm that ctDNA has a positive correlation with the presence of CRM-positive disease, and in conjunction with CRM involvement, ctDNA is a robust surrogate to predict overall poorer outcomes.
Our previous work evaluating a correlation between the levels of ctDNA and tumour burden has demonstrated an association between ctDNA levels and tumour burden as well as an association between persisting ctDNA and disease recurrence.19,20 In this cohort of patients with colorectal cancer, ctDNA was detected in just over 50% of patients following incomplete surgical and/or adjuvant treatment. This quantitative ctDNA was also shown to be of higher sensitivity compared with CEA in detecting the recurrence of colorectal cancer.
Historically, postsurgery residual disease in the form of histological ypT and/or ypN stage of the tumour has been linked to the prognosis of locally advanced rectal cancer, influencing adjuvant treatment decision making.27,28 With the evolution of MRI, several observational studies have now associated MRI-detected T stage tumour, lymph node, CRM, EMVI, tumour deposits and post-neoadjuvant treatment mrTRG as prognostic indicators of disease-specific outcomes.22,29,30 MRI has been shown to accurately evaluate mesorectal fascia involvement in locally advanced tumours, and previous studies have shown that MRI-detected CRM involvement is associated with a higher risk of loco-regional failure and worse disease-free survival.11,12 In addition to this, MRI-detected presence of tumour in veins outside the muscular layer (EMVI) has also been associated with disease recurrence and poor survival outcomes.12 The accuracy of MRI restaging remains controversial due to issues surrounding interobserver variability and the differentiation of post-treatment fibrosis, oedema and tumour.
The strategy of having a complementary tool in the form of ctDNA in addition to predictive MRI markers may help in treatment intensification or alternative treatment strategies in patients to reduce the risk of metastases. This is important in the era of TNT for patients with locally advanced disease where up to 40%-50% of patients may benefit from an organ-sparing approach avoiding morbid surgery and have a similar disease-free interval as demonstrated in the OPRA trial.4 Any additional test after TNT that could add to the predictive capabilities of disease recurrence will help in clinical decision making. Most patients with rectal cancer, however, will still require surgery after neoadjuvant therapy, likely due to incomplete clinical response based on clinical and radiological parameters. In a scenario where a wait-and-watch strategy is being contemplated, the persistence of ctDNA may aid in clinical decision making towards a surgical intervention rather than an organ-sparing approach. In a similar biomarker study, Khakoo et al.16 demonstrated that the detection of ctDNA at the end of neoadjuvant CRT predicted local regrowth in patients that would have been suitable for an organ-preserving approach. Similarly, Wang et al.18 in a prospective cohort study of 119 patients used a combined model of ctDNA and MRI features and demonstrated improved predictive utility of the model compared with data derived from individual information. Combining ctDNA with MRI risk features can potentially stratify patients, thereby enabling the prediction of a high risk of recurrence and prognosis. ctDNA can also supplement MRI to better predict neoadjuvant CRT response, and it could potentially help patient selection for non-operative management and guide the treatment and follow-up strategy for those with different recurrence risks. Hence ctDNA could complement imaging and endoscopic evaluation and detect patients at risk of local regrowth or distant metastatic recurrence who would have otherwise been deemed suitable for an organ-preservation strategy.
One of the limitations of our study was the lack of serial post-treatment ctDNA. In the study by Khakoo et al.16 undetectable ctDNA at the end of CRT was seen in 8 of the 13 patients and was unhelpful as these patients developed local regrowth. Gene mutation-informed ctDNA could be undetectable in a proportion of treatment-naive patients with a localised disease or even in metastatic disease. It is hypothesised that difficulty in detecting pretreatment ctDNA could be due to smaller, less aggressive tumours; however, a better anticipated cancer-related outcome in this group is unproven. Our methylated ctDNA analysis is not restricted to the known mutation profile of the primary tumour, and all nine patients that had stage IV disease were positive for ctDNA, therefore, methylated ctDNA may have a better detection rate and clinical relevance. Larger prospective studies with serial ctDNA pre- and post-treatment will be needed to validate these findings.
From a clinical practice point of view, routine analysis of any ctDNA is likely to be beneficial for a target patient population where adequate turn-around times for reporting results and cost-effectiveness will be required for implementation. Our study has several limitations, some of which are inherent to any single-centre retrospective, observational design, as well as relatively small sample size and lack of post-treatment serial samples. However, this study is a continuation and, in some way, validatory of prior published studies exploring the utility of methylated ctDNA in upper and lower gastrointestinal cancers.16, 17, 18 In addition, all analyses were conducted in the same laboratory with uniform methodology. Despite this being a retrospective study, laboratory staff were blinded to the clinical status of patients while radiologists were blinded to the ctDNA results. The results reported here are hypothesis generating and may guide the direction of future studies, particularly given that risk-adapted personalised treatment strategies continue to be a research priority.
In conclusion, our findings support the idea that methylated ctDNA and MRI assessment are potentially complementary tools to both predict and assess local response and systemic disease status to guide therapy adaptation and overall management in rectal cancer. Future trials will address the true value of incorporating such strategies into routine clinical practice.
Disclosure
TP reports payment or honoraria for lectures, presentations, speakers bureaus, manuscript writing or educational events from Servier, MSD, and Amgen; and support for attending meetings and/or travel from Amgen. All other authors have declared no conflicts of interest. AR reports payment or honoraria for lectures, presentations, speakers bureaus, educational events and Ad boards from BeiGene, Takeda, Servier, MSD, BMS and Ipsen.
Given their role as an Associate Editor, Timothy Price, had no involvement in the peer review of this article and has no access to information regarding its peer review. Full responsibility for the editorial process for this article was delegated to Irit Ben-Aharon, Editor-in-Chief, of the Journal.
References
- 1.Glynne-Jones R., Wyrwicz L., Tiret E., et al. Rectal cancer: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2018;29(suppl 4):iv263. doi: 10.1093/annonc/mdy161. [DOI] [PubMed] [Google Scholar]
- 2.Bahadoer R.R., Dijkstra E.A., van Etten B., et al. Short-course radiotherapy followed by chemotherapy before total mesorectal excision (TME) versus preoperative chemoradiotherapy, TME, and optional adjuvant chemotherapy in locally advanced rectal cancer (RAPIDO): a randomised, open-label, phase 3 trial. Lancet Oncol. 2021;22(1):29–42. doi: 10.1016/S1470-2045(20)30555-6. [DOI] [PubMed] [Google Scholar]
- 3.Conroy T., Bosset J.F., Etienne P.L., et al. Neoadjuvant chemotherapy with FOLFIRINOX and preoperative chemoradiotherapy for patients with locally advanced rectal cancer (UNICANCER-PRODIGE 23): a multicentre, randomised, open-label, phase 3 trial. Lancet Oncol. 2021;22(5):702–715. doi: 10.1016/S1470-2045(21)00079-6. [DOI] [PubMed] [Google Scholar]
- 4.Garcia-Aguilar J., Patil S., Gollub M.J., et al. Organ preservation in patients with rectal adenocarcinoma treated with total neoadjuvant therapy. J Clin Oncol. 2022;40(23):2546–2556. doi: 10.1200/JCO.22.00032. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Alawawdeh A., Krishnan T., Roy A., et al. Curative therapy for rectal cancer. Expert Rev Anticancer Ther. 2021;21(2):193–203. doi: 10.1080/14737140.2021.1845145. [DOI] [PubMed] [Google Scholar]
- 6.Sposato L.A., Lam Y., Karapetis C., et al. Observation of “complete clinical response” in rectal cancer after neoadjuvant chemoradiation: the Flinders experience. Asia Pac J Clin Oncol. 2018;14(6):439–445. doi: 10.1111/ajco.12993. [DOI] [PubMed] [Google Scholar]
- 7.Martin S.T., Heneghan H.M., Winter D.C. Systematic review and meta-analysis of outcomes following pathological complete response to neoadjuvant chemoradiotherapy for rectal cancer. Br J Surg. 2012;99(7):918–928. doi: 10.1002/bjs.8702. [DOI] [PubMed] [Google Scholar]
- 8.Qiu M., Hu J., Yang D., Cosgrove D.P., Xu R. Pattern of distant metastases in colorectal cancer: a SEER based study. Oncotarget. 2015;6(36):38658–38666. doi: 10.18632/oncotarget.6130. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Dieguez A. Rectal cancer staging: focus on the prognostic significance of the findings described by high-resolution magnetic resonance imaging. Cancer Imaging. 2013;13(2):277–297. doi: 10.1102/1470-7330.2013.0028. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Horvat N., Carlos Tavares Rocha C., Clemente Oliveira B., Petkovska I., Gollub M.J. MRI of rectal cancer: tumor staging, imaging techniques, and management. Radiographics. 2019;39(2):367–387. doi: 10.1148/rg.2019180114. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Taylor F.G., Quirke P., Heald R.J., et al. Preoperative magnetic resonance imaging assessment of circumferential resection margin predicts disease-free survival and local recurrence: 5-year follow-up results of the MERCURY study. J Clin Oncol. 2014;32(1):34–43. doi: 10.1200/JCO.2012.45.3258. [DOI] [PubMed] [Google Scholar]
- 12.Bates D.D.B., Homsi M.E., Chang K.J., Lalwani N., Horvat N., Sheedy S.P. MRI for rectal cancer: staging, mrCRM, EMVI, lymph node staging and post-treatment response. Clin Colorectal Cancer. 2022;21(1):10–18. doi: 10.1016/j.clcc.2021.10.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Lord A.C., D’Souza N., Shaw A., et al. MRI-diagnosed tumor deposits and EMVI status have superior prognostic accuracy to current clinical TNM staging in rectal cancer. Ann Surg. 2022;276(2):334–344. doi: 10.1097/SLA.0000000000004499. [DOI] [PubMed] [Google Scholar]
- 14.Fernandes M.C., Gollub M.J., Brown G. The importance of MRI for rectal cancer evaluation. Surg Oncol. 2022;43 doi: 10.1016/j.suronc.2022.101739. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Brouwer N.P.M., Stijns R.C.H., Lemmens V., et al. Clinical lymph node staging in colorectal cancer; a flip of the coin? Eur J Surg Oncol. 2018;44(8):1241–1246. doi: 10.1016/j.ejso.2018.04.008. [DOI] [PubMed] [Google Scholar]
- 16.Khakoo S., Carter P.D., 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(1):183–192. doi: 10.1158/1078-0432.CCR-19-1996. [DOI] [PubMed] [Google Scholar]
- 17.Naidoo M., Piercey O., Tie J. Circulating tumour DNA and colorectal cancer: the next revolutionary biomarker? Curr Oncol Rep. 2021;23(12):140. doi: 10.1007/s11912-021-01137-4. [DOI] [PubMed] [Google Scholar]
- 18.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(8) doi: 10.1371/journal.pmed.1003741. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Symonds E.L., Pedersen S.K., Murray D., et al. Circulating epigenetic biomarkers for detection of recurrent colorectal cancer. Cancer. 2020;126(7):1460–1469. doi: 10.1002/cncr.32695. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Symonds E.L., Pedersen S.K., Yeo B., et al. Assessment of tumor burden and response to therapy in patients with colorectal cancer using a quantitative ctDNA test for methylated BCAT1/IKZF1. Mol Oncol. 2022;16(10):2031–2041. doi: 10.1002/1878-0261.13178. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Piercey O., Tie J. Circulating tumour DNA in the evolving treatment landscape of locally advanced rectal cancer: where does it fit in? Ther Adv Med Oncol. 2023;15 doi: 10.1177/17588359231160138. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Sclafani F., Brown G., Cunningham D., et al. PAN-EX: a pooled analysis of two trials of neoadjuvant chemotherapy followed by chemoradiotherapy in MRI-defined, locally advanced rectal cancer. Ann Oncol. 2016;27(8):1557–1565. doi: 10.1093/annonc/mdw215. [DOI] [PubMed] [Google Scholar]
- 23.Symonds E.L., Pedersen S.K., Murray D.H., et al. Circulating tumour DNA for monitoring colorectal cancer—a prospective cohort study to assess relationship to tissue methylation, cancer characteristics and surgical resection. Clin Epigenetics. 2018;10:63. doi: 10.1186/s13148-018-0500-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Pedersen S.K., Symonds E.L., Baker R.T., et al. Evaluation of an assay for methylated BCAT1 and IKZF1 in plasma for detection of colorectal neoplasia. BMC Cancer. 2015;15:654. doi: 10.1186/s12885-015-1674-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Murray D.H., Symonds E.L., Young G.P., et al. Relationship between post-surgery detection of methylated circulating tumor DNA with risk of residual disease and recurrence-free survival. J Cancer Res Clin Oncol. 2018;144(9):1741–1750. doi: 10.1007/s00432-018-2701-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Symonds E.L., Pedersen S.K., Baker R.T., et al. A blood test for methylated BCAT1 and IKZF1 vs. a fecal immunochemical test for detection of colorectal neoplasia. Clin Transl Gastroenterol. 2016;7(1):e137. doi: 10.1038/ctg.2015.67. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Fischer J., Eglinton T.W., Richards S.J., Frizelle F.A. Predicting pathological response to chemoradiotherapy for rectal cancer: a systematic review. Expert Rev Anticancer Ther. 2021;21(5):489–500. doi: 10.1080/14737140.2021.1868992. [DOI] [PubMed] [Google Scholar]
- 28.On J., Shim J., Mackay C., et al. Pathological response post neoadjuvant therapy for locally advanced rectal cancer is an independent predictor of survival. Colorectal Dis. 2021;23(6):1326–1333. doi: 10.1111/codi.15512. [DOI] [PubMed] [Google Scholar]
- 29.Suzuki C., Halperin S.K., Nilsson P.J., Martling A., Holm T. Initial magnetic resonance imaging tumour regression grade (mrTRG) as response evaluation after neoadjuvant treatment predicts sustained complete response in patients with rectal cancer. Eur J Surg Oncol. 2022;48(7):1643–1649. doi: 10.1016/j.ejso.2022.02.012. [DOI] [PubMed] [Google Scholar]
- 30.Rengo M., Landolfi F., Picchia S., et al. Rectal cancer response to neoadjuvant chemoradiotherapy evaluated with MRI: development and validation of a classification algorithm. Eur J Radiol. 2022;147 doi: 10.1016/j.ejrad.2021.110146. [DOI] [PubMed] [Google Scholar]