TITLE

Rectal cancer MRI: imaging beyond morphology

ABSTRACT

Magnetic resonance imaging (MRI) has in recent years progressively established itself as one

of the most valuable modalities for the diagnosis, staging and response assessment of rectal

cancer and its use has largely focused on accurate morphological assessment.

The potential of MRI, however, extends beyond detailed anatomical depiction: aspects of

tissue physiology such as perfusion, oxygenation and water molecule diffusivity can be

assessed indirectly. Functional MRI is rapidly evolving as a promising non-invasive

assessment tool for tumour phenotyping and assessment of response to new therapeutic

agents.

In spite of promising experimental data, the evidence base for the application of functional

MRI techniques in rectal cancer remains modest, reflecting the relatively poor agreement on

technical protocols, image processing techniques and quantitative methodology to date,

hampering routine integration into clinical management.

This review article outlines the established strengths and the critical limitations of

anatomical MRI in rectal cancer; it then introduces some of the functional MRI techniques

and quantitative analysis methods that are currently available, describing their applicability

in rectal cancer and reviewing the relevant literature; finally, it introduces the concept a

multi-parametric quantitative approach to rectal cancer.

KEYWORDS

rectal cancer; magnetic resonance imaging; diffusion MRI; functional imaging;

multiparametric MRI; whole body MRI.

MANUSCRIPT

INTRODUCTION

The past fifteen years have witnessed a progressive affirmation of MRI as a valuable imaging

modality for rectal cancer. Specialist and multidisciplinary international guidelines now

recommend MRI as the technique of choice for primary staging (with the exception of T1

tumours) and for restaging after neoadjuvant chemoradiation therapy (CRT) (1-3).

The fundamental strength of MRI, most evident since the use of thin-section high-resolution

imaging sequences, has been the ability to depict in great detail the anatomy of the rectal

wall, perirectal tissues and pelvic organs, thanks to its excellent high-contrast soft tissue

resolution.

The potential of MRI, however, extends beyond detailed anatomical depiction: aspects of

tissue physiology such as perfusion, oxygenation and water molecule diffusivity can be

indirectly assessed; metabolic information can also be derived by means of MRI alone or by

the acquisition of simultaneous PET data on state-of-the-art PET/MRI scanners.

The additional value of functional and metabolic MRI techniques in the clinical contexts of

rectal cancer staging, tumour phenotyping and therapy response assessment is only partly

understood and remains the focus of current research.

This review article outlines the established strengths and the critical limitations of

anatomical MRI in rectal cancer; it then illustrates some of the currently available functional

MRI techniques and quantitative imaging methods applicable to MRI, summarising the

salient research findings that foreshadow their potential role in rectal cancer; finally, it

introduces the concept of an integrated, multi-modality and multi-sequence quantitative

imaging approach (multi-parametric), made practicable in rectal cancer by recent

technological advances and already in use for other pelvic malignancies.

STRENGTHS AND LIMITATIONS OF ANATOMICAL MRI

Primary tumour staging

Seminal studies published at the turn of this century have demonstrated how MRI is a

reliable technique for measuring the extent of mural and extramural tumour penetration,

key for T (tumour) staging, and for determining the tumour distance from the mesorectal

fascia, or circumferential resection margin (CRM), thus identifying patients who are likely to

have a clear CRM and those who may benefit from preoperative (chemo)radiotherapy to

increase the likelihood of a R0 resection (clear surgical margin) subsequently (4, 5).

Since then, multicentre prospective research has reinforced these findings, by showing that

CRM status assessed preoperatively by MRI is a significant predictor of overall survival,

disease free survival and local recurrence and that it is possible to predict negative

pathological CRM using a 1 mm cut-off on MRI (6, 7).

The international MERCURY II study has recently investigated MRI in the assessment of low

rectal tumours and their relationship with the anal sphincter complex, crucial to determine

the feasibility of a restorative surgical resection: primary surgical management with

intersphincteric resection in patients deemed to have a ‘safe’ low rectal plane assessed by

MRI led to a clear pathological CRM margin in 98% of cases (8).

Important local features such as the tumour relation to the peritoneal reflection and the

presence of extramural macroscopic venous invasion (EMVI) can be identified with accuracy

on high-resolution MRI and are prognostically significant (9-11) .

Nodal staging

Accurate nodal staging remains problematic with standard anatomical sequences. State-of-

the-art high resolution T2 sequences allow the assessment of nodal morphological features

such as shape, border irregularity and signal heterogeneity in addition to size, increasing the

accuracy of MRI over size criteria alone (12, 13). These morphological features, however, are

challenging or impossible to assess in small nodes (typically < 5 mm), which have been

shown to represent over half of the nodal metastases in rectal cancer (14). Currently, cases

where no nodes are visible on MRI are considered N0; mesorectal nodes ≥8 mm showing

two or more of the mentioned morphological features are considered N+; nodes ≤7 mm

should be viewed with uncertainty and only be called positive when their features are

strongly suggestive of malignancy (15).

Restaging after preoperative CRT

Anatomical MRI assessment of tumour regression grade (TRG), performed 6-8 weeks after

completion of neoadjuvant CRT, has been shown to correlate with disease-free survival and

overall survival in the MERCURY study (16). MRI can accurately assess clearance from an

initially threatened or invaded mesorectal fascia, with negative predictive values for

involvement up to 90% (17, 18), and potentially justify an alteration to the initial

management plan in good responders.

The main limitation of anatomical MRI in the restaging setting is its inability to distinguish

between small remaining tumour foci and post treatment fibrosis; this impacts negatively

on its sensitivity for ypT stage (tumour pathological staging following neoadjuvant therapy),

as low as 50% (19).

In contrast to primary staging, restaging MRI performs well in the assessment of nodal

disease, demonstrating a negative predictive value of around 95%; this means that ypN0

patients can be accurately identified (20, 21).

Detection of local relapse

MRI has been shown to be more accurate than CT for the detection of local relapse and is

certainly valuable in assessing whether a local relapse is surgically resectable (22, 23). The

requirement to exclude distant metastases favours whole-body imaging techniques in this

setting, such as CT and PET/CT. Technological advances, however, allow the use of MRI for

whole-body imaging in clinically acceptable times (<60 minutes); the performance of whole-

body MRI for colorectal cancer staging is currently under investigation in the NHS (Figure 1).

Staging with hybrid PET/MRI with 18F-fluorodeoxyglucose (18F-FDG) is also being explored,

combining the sensitivity of FDG with the high contrast and spatial resolution of MRI.

FUNCTIONAL MRI TECHNIQUES

A number of ‘functional’ MRI techniques are now available to assess several aspects of

tumour physiology in clinical practice: these include water molecule diffusivity, perfusion,

oxygenation and metabolism. Of these techniques diffusion-weighted MRI has been

implemented most widely.

Diffusion-weighted MRI

Diffusion-weighted imaging (DWI) probes the random movement of water molecules,

occurring largely in the extracellular space. The technique is simple on most modern

scanners and does not require the injection of intravenous contrast agent: water molecules

in a volume of tissue are ‘labelled’ by applying a radiofrequency pulse; the same tissue

volume is then resampled at sequential time intervals to determine the proportion of

‘labelled’ water molecules that remain present; where the motion of water molecules is

hindered, for example by densely packed cell membranes (restricted diffusion), more

‘labelled’ water molecules remain present in the volume when the MRI signal is resampled

compared with conditions where water has greater freedom (free diffusion) (24). Most

cancers are densely cellular in comparison to normal tissue and, as a general rule, display

restricted diffusion; rectal cancer is no exception, with reported sensitivities above 90% for

tumour detection on high b-value DWI (Figure 2) (25); the water-diffusion properties of

tumours are nevertheless complex and reflect the coexistence of dense cellularity, fibrosis,

necrosis, neovascularization and haemorrhage. If sequences are acquired using multiple b-

values (corresponding to the time interval after which the signal intensity from labelled

water molecules is measured), it is also possible to calculate a quantitative measure of

diffusion over time, known as the apparent diffusion coefficient (ADC), complementing

qualitative signal intensity assessment. An ADC value can be assigned to each unit volume

(voxel) to provide a coloured parametric ADC ‘map’.

With regards to nodal staging, subjective visual assessment cannot discriminate between

benign and malignant nodes as both display high DWI signal with increasing b-values. The

ADC of malignant nodes has been shown to be slightly lower than that of benign nodes but

not enough to allow the identification of a reliable cut-off (26).

The most promising application of DWI is in response assessment. Several studies have

attempted to establish a relationship between pre-treatment ADC values and treatment

response: in 2002, Dzik-Jurasz et al. were among the first to report a strong negative

correlation between mean pre-treatment tumour ADC and tumour volume reduction after

CRT (27), a finding confirmed by other investigators since. Such negative correlation can be

accounted for by the known relationship between tumour necrosis, which increases ADC

values, and poorer response to treatment. A recent retrospective study on 76 patients with

locally advanced rectal cancer, however, failed to reproduce the same results (28),

reinforcing the concept that multiple factors other than necrosis, such as histological grade,

differentiation and mucinous/non-mucinous type have an impact on ADC. Evidence from

large scale studies is missing in this regard.

The effect of therapy on DWI signal intensity is more straightforward: within days of

initiating therapy ADC increases as a consequence of cell death, cellular membrane

disruption, and decreased cellularity which contribute to increased water diffusion. This

change in ADC has been shown in single-centre clinical studies as early as one week into CRT

(29, 30).

Stronger evidence supports the value of DWI in assessing response after completion of

neoadjuvant CRT. A retrospective multicentric study by Lambregts et al. (31) evaluated the

accuracy of DWI in addition to standard rectal MRI for the identification of complete

responders before surgery in 120 patients; diagnostic performance improved (AUC 0.78–

0.8) compared to standard MRI only (AUC 0.58–0.76), resulting in a substantial reduction in

the number of equivocal cases and improved inter-observer agreement; specificity was

>90% thanks to the ability of DWI to distinguish small areas of residual tumour (high signal

due to high cellular density) from fibrosis (low-signal on high b-value sequences); sensitivity,

nevertheless, was 64% at the most due to the erroneous interpretation of high signal in

‘normal’ post treatment rectal wall as residual tumour.

Quantitative DWI assessment with tumour volumetry has been recently reported to achieve

sensitivities of 65-70% and specificities of 76-98% for the identification of complete

responders using pre-defined volume thresholds (32-33).

Finally, DWI has the potential to become extremely valuable in the follow up of clinical

complete responders where a wait-and-see-policy (omission of surgery with follow-up) is

opted for due to its high sensitivity for small areas of recurrent disease.

Perfusion MRI

Perfusion MRI is an attractive technique for assessing the vascular physiology of tumours, by

combining good anatomical detail with the ability to quantify vascular parameters. It has

been used increasingly for primary tumour characterisation and for the prediction and

evaluation of treatment response, as an indirect measure of tumour angiogenesis. MRI has

the advantage of a good intrinsic signal-to-noise ratio but quantification of contrast agent

concentrations is challenging because of the complex relationship between signal intensity

and contrast medium concentration: this is dependent on many factors, including native

tissue-relaxation rates, contrast agent dose, rate of injection, chosen imaging sequence and

machine parameters.

The most frequently applied perfusion technique in body oncology is dynamic contrast-

enhanced MRI (DCE-MRI), a T1-weighted sequence sensitive to the T1 relaxation effect of

gadolinium contrast agent. Quantification using pharmacokinetic modelling is possible, in

addition to more qualitative assessments looking at maximal enhancement, slope of the

enhancement curve and area under the enhancement curve (AUC). A widely used model is

the Tofts model (34); this provides information on the rate of contrast extraction (transfer

constant, Ktrans), fractional extracellular leakage space (ve) and rate of contrast return from

the extravascular-extracellular space to the vascular compartment (rate constant, kep). Ktrans

describes the trans-endothelial transport of the contrast medium and is affected by both

plasma flow and vascular permeability; its spatial distribution has been shown to be

heterogeneous in primary and metastatic colorectal cancer, adding further complexity to

quantification (Figure 3).

In rectal cancer, most published studies to date have focused on pre-treatment

flow/permeability measures and their changes during treatment, generating conflicting

results. Prospective research by DeVries et al. in patients with T3 disease has shown that the

pre-treatment mean tumour perfusion index (PI), a quantitative parameter directly related

to Ktrans, is significantly higher in patients who fail to respond to neoadjuvant chemoradiation

and is associated with a significantly worse disease-free and overall survival (35, 36). A

possible explanation for these results is that high PI values are secondary to the presence of

arteriovenous shunts, with a high perfusion rate but low exchange of

nutrients/chemotherapy, and/or reflect higher vessel permeability based on increased neo-

angiogenesis.

Similarly, using a blood pool contrast agent, Martens et al. recently reported that the

semiquantitative kinetic parameter ‘late slope’ of the DCE-MRI enhancement curve could

accurately predict response (Mandard TRG 1 or 2 (37)) before CRT with an AUC 0.90 (38).

The pre-treatment mean value of this perfusion parameter was significantly higher in poor

responders and was thought to reflect the higher vascular permeability of angiogenic

tumours, again supporting the theory that tumour angiogenic activity has a negative impact

on therapy outcome.

With regards to DCE MRI post neoadjuvant therapy, a prospective study by Gollub et al.,

evaluating 23 patients with locally advanced rectal cancer before and after induction

chemotherapy with FOLFOX and bevacizumab, found that patients with pathological

complete response (pCR) had statistically significantly lower Ktrans values post-treatment

versus patients with residual disease; moreover, post-treatment Ktrans correlated with

percentage tumour response and final tumour size at histopathology (39). A marked

decrease in perfusion was found in cases of good response despite residual morphological

abnormality on MRI, stressing the added value of DCE MRI to standard anatomical

sequences.

Intrinsic Susceptibility-weighted MRI

Intrinsic Susceptibility-weighted MRI (ISW), also known as Blood Oxygenation Level

Dependent (BOLD) MRI, can be used to assess indirectly the level of vascular oxygenation in

tissue. Image contrast is provided by two components: firstly the endogenous paramagnetic

deoxyhaemoglobin, which increases the transverse relaxation rate (T2*) of water in blood

and surrounding tissues; secondly, static tissue components such as tissue collagen, present

in fibrosis or ligamentous structures, as well as iron contents. The measure R2*

(representing the decay of signal intensity over echo-time) can be quantified, normally by

performing gradient-echo, echo-planar imaging sequences (Figure 4).

BOLD MRI images are more likely to reflect acute (perfusion related) tissue hypoxia than

chronic (diffusion related) hypoxia. The ability of baseline R2* to reflect tumour hypoxia

therefore requires simultaneous assessments of the accompanying functional vasculature,

for example by means of DCE MRI. Only hypoxic tumours with a functional vasculature have

been shown respond to therapy by decreasing R2*, which coincides with improvements in

measured tissue pO2. Another possible approach to single out the vascular contribution to

R2* is to measure it before and after inhalation of oxygen (R2*).

Correlations between pre-treatment baseline BOLD/DCE MRI and immunohistochemical

markers of hypoxia and angiogenesis have been examined in histological tumour sections

from 12 patients with rectal adenocarcinoma in a study performed by Atkin et al. (40). No

correlation between BOLD MRI measurements and the expression of CA-IX, a validated

intrinsic hypoxic marker, was observed, likely because CA-IX co-localizes histologically to

regions adjacent to necrosis, normally distant from nutrient vessels. On the other hand, in a

cohort of 33 patients with prostate cancer, Hoskin et al. demonstrated a high sensitivity

(88%, improving with the addition of low regional blood volume information to 95%) but

low specificity (36% and 29% respectively) of R2* maps for the detection of prostatic

tumour hypoxia when compared to pimonidazole staining (41).

In the era of personalised cancer treatment, non-invasive imaging of tumour hypoxia is

certainly attractive and has the potential to improve early therapeutic triage, enable the

effects of novel hypoxia-modifying agents to be monitored and improve radiotherapy field

contouring through delineation of a biological tumour volume; more research is currently

required to investigate whether BOLD MRI can establish itself as a reliable hypoxia

biomarker; the technique is prone to motion and susceptibility artefacts, which make its

clinical application in the bowel particularly arduous.

MR spectroscopy

MR spectroscopy (MRS) allows for a number of cellular biochemical processes to be studied.

MR-active nuclei have magnetic properties (a magnetic moment) that can be exploited to

generate information about their underlying chemical properties. 1H (proton) spectroscopy

is the easiest to perform and has the highest sensitivity; other magnetic nuclei in descending

order of sensitivity include 19F (fluorine), 31P (phosphorus) and 13C (carbon). In the presence

of an external magnetic field (B0), the magnetic moments from these nuclei will align with or

against B0 and precess at a resonant frequency (ω0). Detection of the magnetic moments is

possible after excitation with a radiofrequency pulse at the resonant frequency of the nuclei

of interest and subsequent relaxation. Relaxation generates RF signals whose intensities

depend on the concentration of the nuclei, and T1-T2 rate constants. A frequency spectrum

is generated containing peaks from different metabolites. In a 1H spectrum at 1.5-tesla, the

metabolites are spread out between 63 and 64 MHz. Resonant frequencies of metabolites

are expressed in parts per million (ppm) and each metabolite reflects a specific cellular

process. Metabolites that have been assessed in clinical practice include choline (related to

phospholipid metabolism), creatine (related to energy metabolism), citrate (a prominent

metabolite in normal prostate tissue) and lactate (a product of anaerobic glycolysis).

Outside of the central nervous system, 1H-MR spectroscopy has been performed most

widely for the detection and evaluation of primary prostate cancer (42), where it has been

incorporated into the PIRADS reporting system (43). Here the ratio of choline, creatine and

citrate are used to provide a score that reflects likelihood of a clinically significant tumour.

1H-MR spectroscopy is technically challenging in the bowel due to presence of bowel gas,

peristalsis and the relative thinness of the normal bowel wall; in clinical practice, a single

voxel technique yields the most reliable results in rectal cancer. A small study evaluating 1H-

MR spectroscopy in patients with rectal cancer receiving neoadjuvant chemoradiation has

shown that the choline peak at 3.2 ppm may disappear following therapy (44).

QUANTITATIVE IMAGE ANALYSIS METHODS

Image Feature Analysis

Image feature analysis is emerging as a new non-invasive tool for assessing intratumoural

heterogeneity, which plays an important role in cancer progression and therapeutic

resistance. It refers to a variety of mathematical methods that may be applied to describe

the relationships between the grey level intensity of pixels or voxels and their position

within an image. It is a post-processing technique that can be easily applied to standard

clinical data, on any imaging modality, with the use of dedicated software. Statistics-based

techniques are most commonly applied and derive parameters from the spatial distribution

of grey levels among pixels or voxels; they are categorized into first-order (one pixel),

second-order (two pixels) and higher order (three or more pixels) statistics (45).

Very limited data has been published to date on rectal cancer MRI. De Cecco et al. recently

investigated whether feature analysis of T2-weighted MRI images acquired at 3 Tesla can

predict tumoural response in 15 patients treated with neoadjuvant CRT (46). Pre-treatment

kurtosis was found to be significantly lower in pathological complete responders (n = 6)

versus partial or non-responders, possibly reflecting lesser intratumoural heterogeneity.

Confirmatory evidence from larger scale studies is awaited. Fine-texture features such as

entropy, uniformity, kurtosis, skewness and standard deviation of the pixel distribution on

contrast-enhanced CT have been shown to be associated with poorer 5-year overall survival

in patients with primary colorectal cancer (47).

The current concept of biological target volumes has prompted the exploration of image

feature analysis as a tool to direct radiotherapy planning. Yu et al. found that a combination

of PET and CT features, particularly second-order and higher-order statistics, can

discriminate tumour from normal tissue; in a small study they found that automated

texture-based segmentation correlated better with tumour delineation by oncologists

compared to PET segmentation (48).

Tumour Volumetry

Gross tumour volume quantification can be easily obtained on modern workstations.

Volumes can be automatically displayed in a 3D format and are calculated by multiplying

each cross-sectional region of interest, manually drawn around the tumour contour, by the

slice thickness and number of slices. Volumetric analysis is less affected by movement,

visceral tortuosity and tumour irregularity than bi-dimensional assessment and has proven

to be reproducible before and after neoadjuvant therapy (32, 49).

Rectal cancer volume reductions ≥70%, measured on high-resolution T2 images, have been

shown to improve response assessment accuracy compared to T down-staging on standard

morphologic sequences alone after 6-8 weeks of neoadjuvant CRT (21). Volume reduction

measurements on both T2-weighted and high b-value DWI have demonstrated high

diagnostic performance (AUC 0.84 and 0.92 respectively) for the assessment of a complete

response (32).

A study by Tan et al. comparing the volumetric and spatial relationships of gross tumour

volume (GTV) derived from CT and MRI found reasonable correlation between the two

modalities for the majority of tumours located in the rectum (n = 15; T3 tumours). In two

patients where there was tumour extension along the sigmoid colon or invading the anal

canal, however, CT-based volume coverage would have resulted in geographic misses,

stressing the importance of reviewing the diagnostic MRI before planning (50).

MULTIPARAMETRIC ASSESSMENT: THE NEXT STEP?

The evidence to date indicates that anatomical MRI has established itself as the imaging

modality of choice for loco-regional staging in rectal cancer, providing excellent anatomical

depiction to guide surgery and to guide decisions towards pre-operative chemoradiation in

locally advanced cancers, where tumour shrinkage is required to increase the likelihood of

successful resection with a clear margin. A blanket approach of chemoradiation for all locally

advanced cancers has been questioned recently related to the increased perioperative

complication rate and long term morbidity of pre-operative radiotherapy (51). Alternative

therapies have shown promise, providing new impetus for a more tailored imaging

approach to further select locally advanced rectal cancer patients who would benefit from

radiotherapy or chemotherapy alone, or a combined chemoradiation approach (52-54).

Anatomical and physiological MRI sequences can be combined in a single examination

within time constraints acceptable to patients (45 minutes or less), providing a more

comprehensive assessment of tumour biology (55); this should help improve risk

stratification, as well as benefiting radiotherapy planning through the definition of a

biological target volume.

The recent availability of state-of-the-art PET/MRI scanners brings further new horizons to

rectal cancer imaging. A recent small pilot study has reported a high diagnostic accuracy of

PET/MRI in T staging of rectal cancer compared with PET/CT in the re-staging setting, and at

least comparable accuracy in N and M staging (56). An example of an integrated PET/MRI

rectal cancer protocol on a hybrid scanner is illustrated in Figure 5. Whole-body MRI,

including diffusion sequences and PET imaging, is supplemented by high-resolution

anatomical sequences of the pelvis. Further locoregional functional sequences may be

added including DCE, BOLD MRI and MRS within a 60-minute examination.

In summary, MRI appears to be the best suited imaging modality to provide a comprehesive,

multi-faceted assessment of rectal cancer: whereas high-resolution anatomical sequences

have strongly affirmed themselves in clinical practice, functional techniques, with the

exception of qualitatively-assessed DWI, have remained largely in the research domain.

Multi-parametric quantitative assessment is emerging as a promising strategy for tumour

phenotyping and response assessment to novel targeted therapies and could have a role in

radiotherapy planning to define a biological target volume. Further clinical evidence is

nevertheless required, best within the trial setting. The use of functional imaging biomarkers

in clinical trials requires strict adherence to technical protocols and the adoption of

stringent quality control programmes; further steps should be undertaken by the imaging

community towards a standardisation of data processing techniques and by researchers

towards biomarker validation in the setting of large-scale clinical studies.

ACKNOWLEDGEMENTS

The authors acknowledge financial support from the National Institute for Health Research

via the Health Technology Assessment Programme; from the Department of Health via the

National Institute for Health Research Comprehensive Biomedical Research Centre award to

Guy’s and St Thomas’ NHS Foundation Trust, in partnership with King’s College London and

King’s College Hospital NHS Foundation Trust; and from the King’s College London /

University College London Comprehensive Cancer Imaging Centre funded by Cancer

Research UK and Engineering and Physical Sciences Research Council, in association with the

Medical Research Council and Department of Health.

REFERENCES

1. Beets-Tan RG, Lambregts DM, Maas M, et al. Magnetic resonance imaging for the clinical

management of rectal cancer patients: recommendations from the 2012 European Society of

Gastrointestinal and Abdominal Radiology (ESGAR) consensus meeting. Eur Radiol. 2013;23(9):2522-

31.

2. Tudyka V, Blomqvist L, Beets-Tan RG, et al. EURECCA consensus conference highlights about

colon & rectal cancer multidisciplinary management: the radiology experts review. Eur J Surg Oncol.

2014;40(4):469-75.

3. Glimelius B, Tiret E, Cervantes A, Arnold D, Group EGW. Rectal cancer: ESMO Clinical Practice

Guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2013;24 Suppl 6:vi81-8.

4. Brown G, Richards CJ, Newcombe RG, et al. Rectal carcinoma: thin-section MR imaging for

staging in 28 patients. Radiology. 1999;211(1):215-22.

5. Beets-Tan RG, Beets GL, Vliegen RF, et al. Accuracy of magnetic resonance imaging in

prediction of tumour-free resection margin in rectal cancer surgery. Lancet. 2001;357(9255):497-

504.

6. Taylor FG, Quirke P, Heald RJ, et al. Preoperative high-resolution magnetic resonance

imaging can identify good prognosis stage I, II, and III rectal cancer best managed by surgery alone: a

prospective, multicenter, European study. Ann Surg. 2011;253(4):711-9.

7. Taylor FG, Quirke P, Heald RJ, 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.

8. Battersby NJ, How P, Moran B, et al. Prospective Validation of a Low Rectal Cancer Magnetic

Resonance Imaging Staging System and Development of a Local Recurrence Risk Stratification

Model: The MERCURY II Study. Ann Surg. 2015.

9. Brown G, Radcliffe AG, Newcombe RG, Dallimore NS, Bourne MW, Williams GT. Preoperative

assessment of prognostic factors in rectal cancer using high-resolution magnetic resonance imaging.

Br J Surg. 2003;90(3):355-64.

10. Smith NJ, Barbachano Y, Norman AR, Swift RI, Abulafi AM, Brown G. Prognostic significance

of magnetic resonance imaging-detected extramural vascular invasion in rectal cancer. Br J Surg.

2008;95(2):229-36.

11. Gollub MJ, Maas M, Weiser M, et al. Recognition of the anterior peritoneal reflection at

rectal MRI. AJR Am J Roentgenol. 2013;200(1):97-101.

12. Brown G, Richards CJ, Bourne MW, et al. Morphologic predictors of lymph node status in

rectal cancer with use of high-spatial-resolution MR imaging with histopathologic comparison.

Radiology. 2003;227(2):371-7.

13. Kim JH, Beets GL, Kim MJ, Kessels AG, Beets-Tan RG. High-resolution MR imaging for nodal

staging in rectal cancer: are there any criteria in addition to the size? Eur J Radiol. 2004;52(1):78-83.

14. Wang C, Zhou Z, Wang Z, et al. Patterns of neoplastic foci and lymph node micrometastasis

within the mesorectum. Langenbecks Arch Surg. 2005;390(4):312-8.

15. Beets-Tan R. How can we identify nodal involvement? In: Valentini V, Schmoll H, van de

Velde C, eds. Multidisciplinary management of rectal cancer: Springer, 2012; p. 73-5.

16. Patel UB, Blomqvist LK, Taylor F, et al. MRI after treatment of locally advanced rectal cancer:

how to report tumor response--the MERCURY experience. AJR Am J Roentgenol. 2012;199(4):W486-

95.

17. Vliegen RF, Beets GL, Lammering G, et al. Mesorectal fascia invasion after neoadjuvant

chemotherapy and radiation therapy for locally advanced rectal cancer: accuracy of MR imaging for

prediction. Radiology. 2008;246(2):454-62.

18. Kulkarni T, Gollins S, Maw A, Hobson P, Byrne R, Widdowson D. Magnetic resonance imaging

in rectal cancer downstaged using neoadjuvant chemoradiation: accuracy of prediction of tumour

stage and circumferential resection margin status. Colorectal Dis. 2008;10(5):479-89.

19. van der Paardt MP, Zagers MB, Beets-Tan RG, Stoker J, Bipat S. Patients who undergo

preoperative chemoradiotherapy for locally advanced rectal cancer restaged by using diagnostic MR

imaging: a systematic review and meta-analysis. Radiology. 2013;269(1):101-12.

20. Lahaye MJ, Beets GL, Engelen SM, et al. Locally advanced rectal cancer: MR imaging for

restaging after neoadjuvant radiation therapy with concomitant chemotherapy. Part II. What are the

criteria to predict involved lymph nodes? Radiology. 2009;252(1):81-91.

21. Barbaro B, Fiorucci C, Tebala C, et al. Locally advanced rectal cancer: MR imaging in

prediction of response after preoperative chemotherapy and radiation therapy. Radiology.

2009;250(3):730-9.

22. Lambregts DM, Cappendijk VC, Maas M, Beets GL, Beets-Tan RG. Value of MRI and diffusion-

weighted MRI for the diagnosis of locally recurrent rectal cancer. Eur Radiol. 2011;21(6):1250-8.

23. Chew MH, Brown WE, Masya L, Harrison JD, Myers E, Solomon MJ. Clinical, MRI, and PET-CT

criteria used by surgeons to determine suitability for pelvic exenteration surgery for recurrent rectal

cancers: a Delphi study. Dis Colon Rectum. 2013;56(6):717-25.

24. Boone D, Taylor SA, Halligan S. Diffusion weighted MRI: overview and implications for rectal

cancer management. Colorectal Dis. 2013;15(6):655-61.

25. Soyer P, Lagadec M, Sirol M, et al. Free-breathing diffusion-weighted single-shot echo-planar

MR imaging using parallel imaging (GRAPPA 2) and high b value for the detection of primary rectal

adenocarcinoma. Cancer Imaging. 2010;10:32-9.

26. Lambregts DM, Maas M, Riedl RG, et al. Value of ADC measurements for nodal staging after

chemoradiation in locally advanced rectal cancer-a per lesion validation study. Eur Radiol.

2011;21(2):265-73.

27. Dzik-Jurasz A, Domenig C, George M, et al. Diffusion MRI for prediction of response of rectal

cancer to chemoradiation. Lancet. 2002;360(9329):307-8.

28. Kim SH, Lee JY, Lee JM, Han JK, Choi BI. Apparent diffusion coefficient for evaluating tumour

response to neoadjuvant chemoradiation therapy for locally advanced rectal cancer. Eur Radiol.

2011;21(5):987-95.

29. Sun YS, Zhang XP, Tang L, et al. Locally advanced rectal carcinoma treated with preoperative

chemotherapy and radiation therapy: preliminary analysis of diffusion-weighted MR imaging for

early detection of tumor histopathologic downstaging. Radiology. 2010;254(1):170-8.

30. Lambrecht M, Vandecaveye V, De Keyzer F, et al. Value of diffusion-weighted magnetic

resonance imaging for prediction and early assessment of response to neoadjuvant

radiochemotherapy in rectal cancer: preliminary results. Int J Radiat Oncol Biol Phys. 2012;82(2):863-

70.

31. Lambregts DM, Vandecaveye V, Barbaro B, et al. Diffusion-weighted MRI for selection of

complete responders after chemoradiation for locally advanced rectal cancer: a multicenter study.

Ann Surg Oncol. 2011;18(8):2224-31.

32. Curvo-Semedo L, Lambregts DM, Maas M, et al. Rectal cancer: assessment of complete

response to preoperative combined radiation therapy with chemotherapy--conventional MR

volumetry versus diffusion-weighted MR imaging. Radiology. 2011;260(3):734-43.

33. Lambregts DM, Rao SX, Sassen S, et al. MRI and Diffusion-Weighted MRI Volumetry for

Identification of Complete Tumor Responders After Preoperative Chemoradiotherapy in Patients

With Rectal Cancer: A Bi-institutional Validation Study. Ann Surg. 2014.

34. Tofts PS, Kermode AG. Measurement of the blood-brain barrier permeability and leakage

space using dynamic MR imaging. 1. Fundamental concepts. Magn Reson Med. 1991;17(2):357-67.

35. Devries AF, Griebel J, Kremser C, et al. Tumor microcirculation evaluated by dynamic

magnetic resonance imaging predicts therapy outcome for primary rectal carcinoma. Cancer Res.

2001;61(6):2513-6.

36. DeVries AF, Piringer G, Kremser C, et al. Pretreatment evaluation of microcirculation by

dynamic contrast-enhanced magnetic resonance imaging predicts survival in primary rectal cancer

patients. Int J Radiat Oncol Biol Phys. 2014;90(5):1161-7.

37. Mandard AM, Dalibard F, Mandard JC, et al. Pathologic assessment of tumor regression after

preoperative chemoradiotherapy of esophageal carcinoma. Clinicopathologic correlations. Cancer.

1994;73(11):2680-6.

38. Martens MH, Subhani S, Heijnen LA, et al. Can perfusion MRI predict response to

preoperative treatment in rectal cancer? Radiother Oncol. 2015;114(2):218-23.

39. Gollub MJ, Gultekin DH, Akin O, et al. Dynamic contrast enhanced-MRI for the detection of

pathological complete response to neoadjuvant chemotherapy for locally advanced rectal cancer.

Eur Radiol. 2012;22(4):821-31.

40. Atkin G, Taylor NJ, Daley FM, et al. Dynamic contrast-enhanced magnetic resonance imaging

is a poor measure of rectal cancer angiogenesis. Br J Surg. 2006;93(8):992-1000.

41. Hoskin PJ, Carnell DM, Taylor NJ, et al. Hypoxia in prostate cancer: correlation of BOLD-MRI

with pimonidazole immunohistochemistry-initial observations. Int J Radiat Oncol Biol Phys.

2007;68(4):1065-71.

42. Scheenen TW, Klomp DW, Röll SA, Fütterer JJ, Barentsz JO, Heerschap A. Fast acquisition-

weighted three-dimensional proton MR spectroscopic imaging of the human prostate. Magn Reson

Med. 2004;52(1):80-8.

43. Barentsz JO, Richenberg J, Clements R, et al. ESUR prostate MR guidelines 2012. Eur Radiol.

2012;22(4):746-57.

44. Kim MJ, Lee SJ, Lee JH, et al. Detection of rectal cancer and response to concurrent

chemoradiotherapy by proton magnetic resonance spectroscopy. Magn Reson Imaging.

2012;30(6):848-53.

45. Chicklore S, Goh V, Siddique M, Roy A, Marsden PK, Cook GJ. Quantifying tumour

heterogeneity in 18F-FDG PET/CT imaging by texture analysis. Eur J Nucl Med Mol Imaging.

2013;40(1):133-40.

46. De Cecco CN, Ganeshan B, Ciolina M, et al. Texture analysis as imaging biomarker of tumoral

response to neoadjuvant chemoradiotherapy in rectal cancer patients studied with 3-T magnetic

resonance. Invest Radiol. 2015;50(4):239-45.

47. Ng F, Ganeshan B, Kozarski R, Miles KA, Goh V. Assessment of primary colorectal cancer

heterogeneity by using whole-tumor texture analysis: contrast-enhanced CT texture as a biomarker

of 5-year survival. Radiology. 2013;266(1):177-84.

48. Yu H, Caldwell C, Mah K, et al. Automated radiation targeting in head-and-neck cancer using

region-based texture analysis of PET and CT images. Int J Radiat Oncol Biol Phys. 2009;75(2):618-25.

49. Regini F, Gourtsoyianni S, Cardoso De Melo R, et al. Rectal tumour volume (GTV) delineation

using T2-weighted and diffusion-weighted MRI: Implications for radiotherapy planning. Eur J Radiol.

2014;83(5):768-72.

50. Tan J, Lim Joon D, Fitt G, et al. The utility of multimodality imaging with CT and MRI in

defining rectal tumour volumes for radiotherapy treatment planning: a pilot study. J Med Imaging

Radiat Oncol. 2010;54(6):562-8.

51. Glynne-Jones R. Do t3 rectal cancers always need radiochemotherapy? Recent Results

Cancer Res. 2014;203:95-115.

52. Schrag D, Weiser MR, Goodman KA, et al. Neoadjuvant chemotherapy without routine use of

radiation therapy for patients with locally advanced rectal cancer: a pilot trial. J Clin Oncol.

2014;32(6):513-8.

53. Ngan SY, Burmeister B, Fisher RJ, et al. Randomized trial of short-course radiotherapy versus

long-course chemoradiation comparing rates of local recurrence in patients with T3 rectal cancer:

Trans-Tasman Radiation Oncology Group trial 01.04. J Clin Oncol. 2012;30(31):3827-33.

54. Pettersson D, Glimelius B, Iversen H, Johansson H, Holm T, Martling A. Impaired

postoperative leucocyte counts after preoperative radiotherapy for rectal cancer in the Stockholm III

Trial. Br J Surg. 2013;100(7):969-75.

55. García-Figueiras R, Goh VJ, Padhani AR, et al. CT perfusion in oncologic imaging: a useful

tool? AJR Am J Roentgenol. 2013;200(1):8-19.

56. Paspulati RM, Partovi S, Herrmann KA, Krishnamurthi S, Delaney CP, Nguyen NC. Comparison

of hybrid FDG PET/MRI compared with PET/CT in colorectal cancer staging and restaging: a pilot

study. Abdom Imaging. 2015.

FIGURE CAPTIONS

Figure 1. Whole-body MRI for primary staging of rectal cancer (white arrow). Post contrast

coronal T1 Dixon in phase and water and diffusion b-900 MIP (maximum intensity

projection) images demonstrate the rectal tumour with no evidence of metastatic disease.

Figure 2. Free Brownian motion of water molecules in the extracellular-extravascular space

(EES) (A) is restricted in a highly cellular environment (B). High-resolution, axial-oblique T2

acquisition demonstrates a semi-circumferential lower rectal tumour (white arrow); the

tumour returns high signal on DWI (b0 and b800) and low signal on ADC, corresponding to

restricted diffusion.

Figure 3. In DCE MRI, tissue T1 signal intensity alters over time as contrast travels from the

intravascular to the extravascular-extracellular space and back. Signal changes may be

plotted as signal intensity/time curve. Following conversion of signal intensity to gadolinium

concentration, applied mathematical models allow pixel-by-pixel calculation of quantitative

parameters, such as Ktrans, the transfer constant, that are visually expressed in colour

parametric maps. Anatomical T2 acquisition, post contrast subtracted T1 image and Ktrans

parametric map of a semi-circumferential mid-rectal tumour are shown.

Figure 4. R2* values are calculated on a pixel-by-pixel basis from a straight line fitted to a

logarithmic plot (ln [signal intensity]) against echo-time (TE) using a least squares approach;

the line gradient is -R2*. The R2* map of the same mid-rectal tumour (white arrow)

demonstrated in Figure 3 is shown.

Figure 5. Coronal ‘fused’ PET/MRI acquisition (C), as well as axial (A) and sagittal (S)

reconstructions, of an FDG-avid mid-rectal tumour. Courtesy of Dr Geoff Charles-Edwards.