Staging of Locally Advanced Rectal Cancer Beyond TME

Abstract

The management of rectal cancer is complex and continually evolving. With advancements in technology and the use of multidisciplinary teams to guide the treatment decision making, staging, oncologic, and functional outcomes are improving, and the management is moving toward personalized treatment strategies to optimize each individual patient's outcomes. Key in this evolution is imaging. Magnetic resonance imaging (MRI) has emerged as the dominant method of pelvic imaging in rectal cancer, and use of MRI for staging is best practice in multiple international guidelines. MRI allows a noninvasive assessment of the tumor site, relationship to surrounding structures, and provides highly accurate rectal cancer staging, which is necessary for determining the appropriate treatment strategy. However, the applications of MRI extend far beyond pretreatment staging. MRI can be used to predict outcomes in locally advanced rectal cancer and guide the surgical or nonsurgical plan, serving as a predictive and prognostic biomarker. With continued MRI hardware improvement and new sequence development, MRI may offer new perspectives in the assessment of treatment response and new innovations that could provide better insight into the staging, restaging, and outcomes with rectal cancer.

A major advance in the modern management of rectal cancer was the widespread application of the total mesorectal excision (TME). The TME, popularized by professor Heald in the early 1980s, entails a sharp resection in the avascular embryologic “holy plane” for en bloc removal of the cancer and surrounding mesorectum, allowing an intact mesorectal fascia (MRF) and preservation of the autonomic nerves. ¹ Application of the TME has resulted in reproducible dramatic reductions in local recurrence and improvement in disease-free and overall survival rates. ^{2 3 4} Performance of a proper TME remains the main prognostic factor for disease recurrence. ^{2 3} However, TME alone is not sufficient for locally advanced tumors—the addition of neoadjuvant chemoradiotherapy (nCRT) can reduce local recurrence rates in curable rectal cancers and allow patients initially deemed unresectable to undergo curative resection. ⁵ nCRT will downstage 50 to 60% of patients, with ∼10 to 30% of patients demonstrating a pathologic complete response (pCR). ^{6 7 8} There is a need to identify the locally advanced cancer patients appropriate for nCRT and good responders to therapy, to avoid over- and undertreatment. In this current era of precision medicine, technology and multidisciplinary team (MDT) input further optimize the staging, treatment strategy, and outcomes of locally advanced rectal cancer. A critical element in this evolution is imaging.

The Current Clinical Staging for Locally Advanced Rectal Cancer

The current staging for rectal cancer follows the TNM (tumor, node, metastases) system from the AJCC (American Joint Committee on Cancer) Cancer Staging Manual, 8th edition, in conjunction with pathological assessment. ⁹ Magnetic resonance imaging (MRI) has advanced the clinical staging of locally advanced rectal cancer and has emerged as the dominant method of rectal cancer imaging. High-resolution T2-weighted imaging (T2WI) MRI is the mainstay due to its superior tissue contrast resolution. It entails use of a thin-section (3-mm) T2-weighted fast spin-echo sequence performed orthogonal to the tumor in the sagittal, axial, and coronal plane. ¹⁰ The high-resolution T2WI sequences and principles for staging advanced rectal cancers were developed with sufficient resolution to assess the layers of the rectal wall and the mesorectal plane. ¹¹ In low rectal cancers, additional sequences can be added to optimally depict the levator muscles, sphincter complex, intersphincteric plane, and the relationship to the rectal wall, improving visualization and operative planning. There is also value in the posttreatment MRI, as restaging posttreatment can impact surgical planning, the timing of surgery, sphincter preservation, deferral of surgery for good responders, and intensified treatment before surgery for poor responders. ¹²

Key features are well defined on MRI, allowing for noninvasive, highly accurate rectal cancer staging. Measurement of the distance from the most caudal aspect of the tumor to the anal verge and the relationship of the tumor to the upper margin of the puborectalis sling are clearly seen, assisting in determination of sphincter preservation. The location of the tumor relative to the anal sphincter complex in locally advanced low rectal cancer is also clearly seen, to help identify patients who will benefit from nCRT, potentially improving sphincter preservation and disease-free survival, and reducing abdominoperineal resection rates ^{13 14} ( Fig. 1 ). T-staging is based mainly on differences in T2 signal intensity among the tumor, submucosa, muscular layer, and mesorectum. The accuracy of MRI for T-staging has been reported as up to 95%. ^{10 15 16} Most staging failures occur in the differentiation between T2 and borderline T3 lesions or distinguishing MRF invasion from desmoplastic reactions. ^{10 17 18}

Fig. 1.

Low rectal tumor in relation to the sphincter complex. Solid line depicts the distal margin of the tumor. Dashed line depicts the top of the internal sphincter.

The number and status of lymph nodes examined have been long described as one of the most important criteria in determining the management and outcomes in rectal cancer. ¹⁹ Historic studies described the number of lymph nodes retrieved to correlate with outcome, prognosis, and survival. ²⁰ However, these studies did not consider the influence of other factors that can impact prognosis, such as the quality of the surgery. In addition, for locally advanced rectal cancers, where preoperative radiotherapy decreases both the mean number of lymph nodes retrieved and the mean number of positive lymph nodes, studies have shown that survival is not influenced by the number of lymph nodes retrieved. ¹⁹ Determining the presence of nodal involvement on MRI has traditionally relied on size assessment. However, there is no single size cutoff to distinguish benign from malignant nodes with both high sensitivity and specificity. Furthermore, micrometastases in normal-sized lymph nodes is common. ^{21 22} Criteria based on the shape and morphological features, such as heterogeneous signal intensity and lymph node irregular capsule border may be more reliable and accurate predictors of metastatic spread within the nodes. ^{21 23} High-resolution MRI enables highly accurate characterization of nodes based on these features rather than size, which has been shown to be a poor predictors of nodal status. ^{21 22} Using the border contour and signal intensity, characteristics instead of size criteria improved MRI prediction of nodal involvement in rectal cancer, and potentially prevents the overtreating of patients based on nodal size criteria. ²¹

Limitations of the TNM Staging System

The TNM staging system provides a universal language, but it may not be sufficient to base treatment decisions on the image-predicted tumor (T) and node (N) stages alone, as the clinical T and N stages with standard imaging can be inaccurate, risking overstaging and overtreatment. ^{24 25 26} Nodal status has no prognostic importance for local recurrence if good quality TME surgery is performed. ^{27 28} MRI-identified lymph nodes involving the MRF are rarely a true cause of pathologic circumferential resection margin (CRM) infiltration, and involvement of the CRM by lymph node metastases alone is uncommon. ²⁹ The TNM system also does not consider major pelvic vessels, the pelvic parietal fascia, or the MRF as “organs,” which can affect the staging. In addition, up to 30% of patients with locally advanced rectal cancers at baseline staging undergo a pCR after nCRT ³⁰ ; these patients who develop complete tumor regression could potentially have avoided radical surgery and its associated morbidity, mortality, functional consequences, and need for stomas if identified appropriately. The TNM system also lacks measures of prognosis, and may no longer be sufficient as the sole basis for treatment. ³¹

Using High-Resolution MRI to Determine Prognosis and Predict Outcomes

High-resolution MRI addresses many limitations of the TNM system, as it has the ability to define rectal cancer in the traditional TMN staging system, as well as other relevant imaging parameters not included in the TNM classification that describe prognostic signs to help stage, restage, and plan the best treatment course. ³¹ For preoperative staging, MRI prognostic factors enhance stratification of patient's treatment and prognosis. ³¹ Included in these measures are the extramural depth (EMD) spread or tumor infiltration, the MRI CRM for defining the TME plane and resectability, and the presence or absence of tumor signal into extramural venous invasion (EMVI). ^{4 10 18 27 32 33 34 35} Using these imaging biomarkers, MRI can identify patients who may benefit from more intensive treatment, without subjecting those who will not benefit to unnecessary treatment. ^{4 17}

There can be difficulty distinguishing between T2 and borderline T3 lesions on MRI, which can change a patient's clinical stage from I to II, and the associated recommendation for nCRT. With this limitation, studies have highlighted the importance of measuring the EMD spread, defined at histopathologic analysis as the distance from the outer edge of the longitudinal muscularis propria to the outer edge of the tumor, rather than just the T stage, as the depth of spread is prognostic of outcomes. MRI can accurately indicate the tumor stage and EMD of invasion, providing valuable information for differentiating T2 and borderline T3 tumors from more invasive T3 tumors in patients who are at high risk of failure of complete excision for preoperative therapy ¹⁰ ( Fig. 2 ). The better assessment of inhomogeneous pT3 rectal cancers allows improved prediction of outcomes and identification of stage II patients who might not benefit from adjuvant treatment. ³⁶ The MERCURY Study Group found MRI was accurate in depicting the EMD of tumor invasion, with equivalent MRI and histopathologic results (within ± 0.5 mm). ³³ The accurate measurement of EMD of tumor spread enables precise preoperative prognostication. Studies have shown that T2 tumors compare with T3 tumor with less than 1 mm spread, and T3 tumors with >5 mm of mesorectal invasion have a cancer-specific 5-year survival rate of ∼50%, while those with <5 mm of mesorectal invasion have a cancer-specific survival rate of more than 85%. ^{36 37} EMD can be a key factor in stratifying patients for preoperative therapy, as it is not the 1-mm distinction between T2 and T3 that may potentially impact treatment decisions, but the identification of high-risk patients whose risk of metastatic disease increases steadily with each millimeter of tumor invasion beyond 5 mm. ¹⁷ The better assessment of inhomogeneous pT3 rectal cancers allows an improved prediction of outcomes and identification of stage II patients who might not benefit from adjuvant treatment. ³⁶ The maximal EMD may also be a more reproducible imaging biomarker than the distance from the tumor to the anticipated CRM, ³⁸ which could have implications for training and implementing guidelines for neoadjuvant therapy.

Fig. 2.

MRI comparison of (a) T2 mass with no sphincter involvement; (b) T3a mass (solid line depicts mesorectal fat invasion, and dashed line depicts the intact muscularis propria); and (c) T3 mass with CRM involvement. CRM, circumferential resection margin; MRI, magnetic resonance imaging.

On high-resolution MRI, the MRF is seen as a fine, low-signal-intensity layer enveloping the perirectal fat and rectum. Preoperative MRI can determine the relationship between the tumor and the MRF, the surgical excision plane in TME resections. ³² Involvement of the CRM is an independent prognostic factor for local recurrence, poor survival, and a variable to help identify patients appropriate for neoadjuvant therapy. ^{39 40} In the original MERCURY trial, preoperative MRI was found to accurately predict the surgical CRM and ability to perform curative resection in all stages of rectal cancer. The accuracy was reproducible, and showed the possibility of using imaging to identify patients likely to fail curative resection and guide further preoperative treatment. ³² An MRI-directed strategy based on the MRI CRM has been shown to significantly reduce positive CRM in rectal cancer by identifying patients with an involved CRM likely to achieve tumor-free resection margins after long-course nCRT. ⁴¹ One study defined a tumor distance > 1 mm from the CRM for selecting patients with a low risk of local cancer recurrence, who could potentially omit neoadjuvant radiotherapy and its associated morbidity. ⁴² The need for routine assessment of the CRM to assess prognosis has also been established. A potentially involved CRM on MRI has a significantly higher risk of recurrence and cancer-related death, showing the prognostic value of preoperative MRI to select patients for nCRT and/or more radical surgery. ⁴³ Further, involvement of the CRM on pathology (pCRM) has been proven an independent prognostic factor for local recurrence and distant metastases. ^{39 40 44} The prognostic relevance of MRI CRM assessment was further confirmed in the 5-year follow-up of the MERCURY Study Group trial, which found MRI-involved CRM was superior to AJCC TNM-based criteria for assessing overall survival, disease-free survival, and time to local recurrence. MRI CRM involvement was also significantly associated with metastatic disease, allowing changes to adjuvant treatment and follow-up accordingly to improve survival outcomes. ²² In low rectal cancers (6 cm or less from the anal verge), the CRM on restaging MRI after neoadjuvant therapy could accurately identify pCRM tumors, appropriately selecting patients for sphincter-preserving surgery, and avoiding overtreatment. ⁴⁵ Thus, MRI assessment of CRM as an imaging biomarker, both pre- and posttreatments, predicts survival outcomes for good and poor responders and provides an opportunity for the MDT to guide treatment options before planning definitive surgery. ^{12 46}

EMVI is defined as the presence of malignant cells within blood vessels located beyond the muscularis propria in the mesorectal fat. EMVI exists in up to 60% of locally advanced colorectal cancers and is relatively common in T3–T4 tumors. ^{27 35 47 48} EMVI is a proven poor prognostic factor, ^{35 49} and MRI is the only imaging modality that has been shown to consistently demonstrate EMVI ²⁷ ( Fig. 3 ). MRI-detected (mrEMVI) at baseline is an independent prognostic factor for local and distant recurrence and poor overall survival outcomes. ^{27 35 50} The severity of mrEMVI is correlated with disease-free survival ^{27 47} and worse oncological outcomes. ⁵¹ The mrEMVI status before CRT is a prognostic factor that can be used to stratify stage T3 patients who might benefit from nCRT ⁴⁹ ; mrEMVI before CRT was more likely to be present in patients with a higher T3 subgroup and MRF involvement, and decreased disease-free survival. ⁴⁹ After neoadjuvant therapy, the response of EMVI (graded by an MRI tumor regression grade [mrTRG]) is associated with disease-free survival, and lack of disease recurrence, supporting use of mrEMVI as an imaging biomarker to direct intensified treatment strategies. ⁴⁷ mrEMVI can help identify high-risk stage II rectal cancer patients, who behave more like stage III patients, and would benefit from adjuvant chemotherapy to improve disease-free survival. ²⁸ EMVI can also be used to guide treatment decisions after surgery. Patients with persistent mrEMVI following CRT saw a significant survival benefit with adjuvant chemotherapy, with improved disease-free survival compared with patients not receiving adjuvant chemotherapy, regardless of age and performance status. ⁵² EMVI status may also be a poor prognostic factor for metastases, as a recent meta-analysis reported mrEMVI positive patients presented five times more frequently with synchronous metastases and developed metastases four times more often after curative resection than mrEMVI negative tumors. ⁵³ The combination of MRI assessment of EMVI, a 1-mm or less distance from the tumor to the surgical resection plane, and the mrTRG in patients after nCRT may be more powerful prognostic predictors than T and N stages. ³¹ With these signs, MRI provides invaluable prognostic information, and is an essential part of the staging for the MDT to make appropriate treatment decisions.

Fig. 3.

Example of extramural venous invasion.

Considering these prognostic features, the European Society of Gastrointestinal and Abdominal Radiology published updated recommendations on the acquisition, interpretation, and reporting of MRI for clinical staging and restaging of locally advanced rectal cancer. ⁵⁴ The minimum reporting standards for MRI rectal cancer staging were also recently published in the St. Gallen consensus ( Table 1 ). ⁵⁵

Table 1. MRI reporting standards.

Site of tumor—upper/mid/lower third
Height from puborectalis sling and anal verge and craniocaudal length
Assessment of the safety of the TME surgical plane a
Relationship to important landmarks, e.g., peritoneal reflection/seminal vesicles
Morphology (annular/semiannular/mucinous infiltrating border [smooth or nodular])
Presence or absence of extramural venous invasion
Presence or absence of vascular mediate tumor deposits (N1c)
Maximum extramural depth spread
Presence or absence of malignant lymph nodes (smooth border/uniform signal = benign irrespective of size)
Minimum distance to mesorectal fascia or intersphincteric plane > 1 mm (MRI CRM clear)
In the final assessment, the TNM stage
In the final assessment, assessment of potential resection margin involvement/safety of the TME plane (classified as potentially involved if tumor < 1 mm to the mesorectal fascia/intersphincteric plane)

Abbreviations: CRM, circumferential resection margin; MRI, magnetic resonance imaging; TME, total mesorectal excision; TNM, tumor, node, metastases.

Source : Adamina et al. ⁵⁵

^aFor tumors arising at or within 2 cm above the level of the puborectalis sling.

Planning Resections Beyond the TME—MRI Anatomic Compartments

In addition to diagnosis and staging, high-resolution T2WI MRI can provide value in the depiction of anatomical structures and compartments relevant to planning surgical resection. In locally advanced rectal cancers, where the cancer invades adjacent structures within the pelvis or there is CRM involvement, it is necessary to extend the surgical planes beyond the TME plane for an R0 resection. MRI is accurate in predicting the extent of tumor involvement within the pelvis and resectability, and can be used to determine the best surgery for curative resection. ^{56 57} MRI detects the planes required for histological clearance, and indicates when extralevator abdominoperineal resections or exenterative surgery should be performed for R0 resection and improved local recurrence-free survival. ^{56 58} The extent of pelvic exenteration surgery depends on radiological criteria; in these advanced cases, studies have shown the accuracy of MRI for detecting invasion of intrapelvic compartments for planning exenterative pelvic surgery and ensuring there is clearance of disease. ⁵⁹ The compartments are: (1) central above peritoneal reflection: any structure above the peritoneal reflection/uterus; (2) anterior below peritoneal reflection: bladder/upper vagina/ovaries/prostate/seminal vesicles/urethra; (3) posterior: bony cortex/periosteum S1–5, coccyx/presacral fascia (S1–5)/sciatic nerve/sacral nerve branches (S1/2); (4) lateral: pelvic fascia/pelvic sidewall/internal/external iliac vessels/sacrotuberous/sacrospinous ligaments/piriformis/obturator internus muscles; (5) infralevator: levator muscles/sphincter complex; and (6) anterior urogenital triangle/perineum: vaginal introitus/urethra/retropubic space. ⁵⁶ MRI can be used to help predict resection margins, with positive margins in the anterior and central compartments accurately identified, but involvement of the lateral/pelvic sidewall and high posterior margins less accurately identified and more likely to be associated with a positive resection margin. ⁵⁹ Further work is needed to address these limitations in exenterative cases.

MRI for Predicting Response to Treatment—Tumor Regression

The ideal way to define the treatment response continues to evolve. Currently, the only validated measure in solid tumors is the response evaluation criteria in solid tumors (RECIST) criteria, which is based on objective change in craniocaudal length of measurable disease. ⁶⁰ Under these guidelines, at least a 30% reduction rate is needed to be considered a good or favorable response. However, with RECIST, there is no established threshold for defining response in a luminal organ such as the rectum. In rectal cancer, assessment of treatment response after nCRT has historically relied on histopathologic assessment of the surgical specimen; here the degree of tumor regression following nCRT is a prognostic factor. ^{61 62} Qualitative assessment of the degree of fibrosis in the pathology specimen can also be used to define a tumor regression grade (TRG), which predicts overall and disease-free survivals. ^{63 64 65} An alternative method of assessing tumor regression uses mrTRG. mrTRG is a validated prognostic tool that follows the principles of the modified TRG system of Mandard et al to evaluate the response to CRT and classifying posttreatment responses using five categories based on the proportion of intermediate tumor signal intensity versus low fibrosis signal intensity within the tumor ( Table 2 ). ^{63 66 67} mrTRG is an independent prognostic factor of disease-free and overall survivals, and thus patient prognosis before definitive surgery. ^{12 46} mrTRG showed statistical correlation with ypT, and assessing the degree of tumor replacement by fibrosis correlated better with survival than the ymrT stage. ⁴⁶ mrTRG has the benefit over pathological TRG (pTRG) of assessing the tumor response before surgery, for a personalized treatment plan, including alteration of the surgical planes, additional use of chemotherapy, or deferral of surgery. ^{12 46 68} mrTRG also has the advantage over pTRG of easy comparison to the pretreatment stage and examining the whole tumor in one scale, reducing heterogeneity and errors in interpretation. mrTRG can also distinguish between good and poor responses in relation to survival better than pTRG. ⁶⁹ While promising, there are challenges with mrTRG. First, there are several regression scales available, with large variations between the scales, and differences in the reliability for consistently identifying good and poor responders. ⁶⁹ Across oncologic outcomes, the most popular are the Mandard and Dworak scales; Mandard 1, 2 and Dworak 3, 4 could be used to define a good response to nCRT, and Mandard 3, 4, 5 and Dworak 0, 1, 2 to define a poor response ⁶⁹ ( Table 2 ) ( Fig. 4 ). However, published work uses these varied scales to define poor response without consistency. ⁷⁰ In addition, the agreement between mrTRG and pTRG is low, precluding mrTRG to be used as a valid surrogate to identify pCR for the adoption of nonoperative management strategies. ⁷¹

Table 2. Commonly used MRI TRG scales.

TRG scale	Mandard a	Modified Mandard b	Dworak c	Modified Dworak d
0	–	–	No regression	No regression
1	Complete regression—no residual cancer or fibrosis	TRG 1 and 2 of the Mandard scale	Dominant tumor mass with obvious fibrosis and/or vasculopathy	Regression ≤ 25% of tumor mass (dominant tumor mass with fibrosis)
2	Presence of rare residual cancer	TRG 3 of the Mandard scale	Dominant fibrotic change with few tumor cells or groups	Regression >25–50% of tumor mass (mainly fibrotic changes with few tumor cells)
3	Increased number of residual cancer cells, but mainly fibrosis	TRG 4 and 5 of the Mandard scale	Very few tumor cells in fibrotic tissue with or without mucin	Regression > 50% of tumor mass (very few tumor cells in fibrotic tissue with or without mucous substance)
4	Residual cancer outgrowing fibrosis	–	No tumor cells, only fibrosis (total response or regression)	Complete (total) regression (or response), no vital tumor cells
5	Absence of regressive changes	–	–	–

Abbreviations: MRI, magnetic resonance imaging; TRG, tumor regression grade.

^a Mandard et al. ⁶⁷

^b Ryan et al. ⁹⁸

^c Dworak et al. ⁶³

^d Wittekind and Tannapfel. ⁹⁹

Fig. 4.

(a) Pretreatment T4b rectal cancer; (b) good response/mrTRG 1 on posttreatment imaging. mrTRG, magnetic resonance imaging tumor regression grade.

Advanced MRI Techniques to Better Assess Treatment Response

While high-resolution MRI is the diagnostic study of choice for restaging after nCRT, the reliability is controversial and its use remains contentious. The posttreatment MRI has value in potentially changing the surgical course, and even avoiding unnecessary surgery in a complete pCR. ^{45 72 73} The major restrictions with current MRI techniques are difficulties differentiating cancerous tissues from fibrosis, determining lymph node status, and predicting a pCR following nCRT. ⁷⁴ It is also inadequate for prediction and assessment of individual response to CRT. ⁷⁵ The signal intensity of post-CRT fibrosis and viable tumors is indistinguishable by visual inspection. ^{76 77} The presence of inflammation and fibrosis makes it difficult to accurately evaluate residual tumor, lymph nodes, the relation of the tumor to the mesorectal fascia, and to measure the response to treatment. ^{78 79} Furthermore, the readings are operator dependent with features that are qualitative, not quantitative, impacting the reproducibility. These limitations have led to new advancements in high-resolution imaging, such as functional MRI.

Functional MRI visualizes the underlying biological characteristics of tumors, in addition to the morphological information from conventional MRI. The combination of these imaging biomarkers may provide a more comprehensive picture of tumor heterogeneity and its changes in response to treatment in locally advanced rectal cancer. ⁷⁵ As technologies continue to emerge, several functional technologies have already entered clinical practice, with the most data supporting diffusion-weighted imaging (DWI). ⁸⁰

DWI provides qualitative and quantitative information on the underlying cellular architecture of the tissue, based on differences in the Brownian motion of extracellular water molecules, so-called diffusivity, in different environments. DWI is dependent on multiple factors such as cell density, vascularity, viscosity of extracellular fluid, and cell membrane integrity. The total diffusivity is quantitatively expressed as the apparent diffusion coefficient (ADC). In tissues with normal cellularity, water molecules can diffuse relatively freely, resulting in a progressive loss of signal with increasing secondary magnetic field strengths ( b -value), while in tissues with increased cellularity, such as tumor, the diffusion is restricted, resulting in preservation of high signal. ⁸¹ DWI is increasingly acknowledged as a valuable adjunct to the standard MRI, particularly in the re-evaluation of the local tumor stage after nCRT, ruling out a pathological complete response, and identifying residual tumor from irradiated fibrotic scar tissue. ^{78 81 82 83} The increase in tumor ADC (increased diffusivity or decreased restricted diffusion) has been shown to correlate with response to treatment, and can serve as an imaging biomarker. ^{84 85 86} Adding a DWI sequence could help identify a pCR, helping the MDT select true complete responders for watchful waiting. ^{87 88} While promising, there are limitations to DWI. The evidence to date is mainly from single-center expert studies and prospective multicenter data are lacking, there is no standard for acquisition or interpretation, and reports for the quantitative diffusion parameter ADC in responsive and nonresponsive tumors are not standard across multiple MRI vendors, magnetic field strengths, and other factors. High DWI signal can be caused by factors other than residual tumor, and air in the rectal lumen can cause artifact, which can affect the staging. ^{80 82 89} Proposed improvements for diagnostic performance include adding additional precision features to the DWI sequence and other new technology, such as image segmentation with histogram analysis, ^{90 91} texture analysis, ^{92 93} diffusion kurtosis imaging, ⁹⁴ or a combined T2WI and DWI pattern-based approach, ⁸² adding volumetry with defined thresholds, ^{95 96} and deep learning methods for automatic localization and segmentation of tumor. ⁹⁷

Conclusion

High-resolution MRI is the gold standard for rectal cancer staging and a key element in the multidisciplinary management. MRI provides information beyond TNM staging, with preoperative identification of important surgical and pathological prognostic biomarkers. Defining these features may help select the most appropriate treatment strategy, potentially avoiding over- and undertreatment. Reporting guidelines using these biomarkers and an mrTRG to assess treatment response could redefine the standard of care and further improve outcomes. Functional MRI, such as DWI, and other emerging technologies may complement the information on conventional high-resolution MRI to increase its predictive and prognostic ability. With great clinical implications for the development and validation of these tools, the future of MRI for staging locally advanced rectal cancer is exciting.