Interpretation of Rectal Cancer MRI (Initial Stage) D Kim 1-10-21

1. Baseline study- INITIAL STAGE MR

a. Imaging is a required component of initial rectal cancer staging which is clinically-based
(denoted cTNM). Clinical staging includes diagnostic biopsy, physical exam and imaging.
i. cTNM staging determines whether patients receive neoadjuvant
chemoradiation (or TNT-total neoadjuvant therapy) prior to surgery
ii. After surgery, staging is based on pathology (denoted pTNM). If neoadjuvant
chemoradiation was given, the staging is denoted with a prefix ‘y’ (ypTNM)

b. Rectal MR has replaced endoscopic rectal ultrasound (ERUS) as the imaging modality of
choice for rectal cancer staging due to rectal MR’s superior ability in overall assessment
of local extent of disease (ie, extension past the muscularis propria, relationship to the
mesorectal fascia, regional lymphadenopathy). ERUS remains the modality of choice for
a minority of cases with the specific question of T1 versus T2 status (cancer limited to
the submucosa versus extension into the muscularis propria) to determine if an
endoscopic resection is possible. 1 Rectal MR should not be used for this assessment. At
 rectal MR staging, T1 and T2 are lumped into a single category (denoted ‘T1/T2’ or ‘T1-
T2’).
c. Other helpful clinical information to be aware of to aid Rectal MR interpretation
includes:
i. Digital rectal exam (DRE) results for MR interpretation of low tumors. DRE can
help assess for fixation to sphincters, relationship to anorectal ring and pelvic
floor musculature
ii. Rigid proctoscope or flexible sigmoidoscope findings which help to localize
tumor and determine potential distal margin (see ‘Tumor location and position’
section)
d. Tumor location and position
i. It is important to localize and describe the relationship to several anatomic
landmarks which help guide surgical approach and predict surgical margins.
May impact need for neoadjuvant chemoradiation.
1. Distance between anal verge and tumor to localize as a upper, mid, or
lower rectal tumor
2. Distance between top of anal canal (anorectal junction) and tumor to
help define distal margin of the surgical resection
3. Distance to the mesorectal fascia to determine the potential
circumferential resection margin (CRM) at surgical resection
4. Relationship to the anterior peritoneal reflection
5. The proximal margin is typically not an issue which is determined by
level of ligation of supplying arterial vasculature. Typically is well away
from the location of the tumor
e. Distance from anal verge localizes tumor into upper, mid, and lower rectum
i. Anal verge (definition): external end of the anal canal; the transitional zone
between the epithelium of the anal canal and the perianal skin.
a. Upper rectum >10-15 cm from anal verge
b. Mid rectum >5-10 cm from anal verge
c. Low rectum <5 cm from anal verge
ii. How this distance is measured:
1. Distance classically determined by rigid proctoscopy (gold standard)
2. Flexible endoscopy may not correlate well (often overestimates
distance from anal verge compared to rigid proctoscopy which
straightens bends)
3. Measured at MR along the along the long axis of the anal canal from the
distal end of the internal sphincter to the top of the anal canal and the
optimized distance from top of canal to inferior edge of tumor. The two
distances should be summed.
 iii. Importance of this measurement:
1. Categorizes tumor as a high, mid, or low rectal tumor which holds
implications regarding probable surgical approach
2. Classically, high and mid rectal tumors are amenable to a sphincter
sparing surgery (Low Anterior Resection; LAR) as opposed to low rectal
tumors are more likely to go on to an Abdominal-Peritoneal Resection
(APR) which leads to a loss of the anal sphincter and a permanent
colostomy.
3. Increasingly, LAR is a potential option for low rectal tumors— see
section ‘Relationship to the anal canal’
4. High rectal tumors may be treated like distal sigmoid cancers in select
cases. Discussion at multi-disciplinary conference is crucial.
f. Distance between the top of the anal canal (level of the puborectalis at the anorectal
junction) and the inferior edge of tumor.
i. Gives information regarding potential distal surgical margin. Historically,
surgeons preferred a distal margin of 5 cm. This has decreased to 2 cm. 2
Evidence that a 1 cm margin may be acceptable after neoadjuvant
chemoradiation. 3
ii. How the distance is measured
1. Shortest linear distance between the top of the anal canal at the level of
the puborectalis and the most inferior portion of the tumor
Distal margin measurement (yellow highlighted blue line)

iii. Importance of the distal margin status

1. A positive distal margin leads to local recurrence rates of 40%4 and
decreases in 5 year survival5 despite TME and neoadjuvant
chemoradiation
g. Relationship to the mesorectal fascia (MRF) helps to determine potential circumferential
resection margin (CRM)
i. MRF (mesorectal fascia): Fascial layer that encloses the rectum, mesorectal fat,
lymph nodes and lymphatics to form an anatomic unit. Distance of tumor to the
MRF used to predict CRM involvement.
1. MRF fuses with the presacral fascia posteriorly then encircles the
rectum laterally. Anteriorly, it thickens at the level of the prostate and
seminal vesicles in males (called Denonvilliers fascia). The correlate in
females is fascial thickening at the rectovaginal septum. The superior
extent anteriorly is demarcated by the anterior peritoneal reflection.
2. MRF completely encircles the rectum from the anterior peritoneal
reflection superiorly to the level of the pelvic floor (ie, levator ani
musculature) inferiorly
3. MRF is incomplete where the upper and mid rectum is partially
peritonealized (see ‘anterior peritoneal reflection’ section)
 ii. CRM (circumferential resection margin) represents the distance of tumor closest
to the radial edge of the surgical TME specimen.
1. Refers to the non-peritonealized surfaces of the rectum; thus not
applicable peritonealized surfaces which include the anterior and lateral
walls for upper rectum and anterior walls for mid rectum. A helpful
landmark is the anterior peritoneal reflection— to be discussed later).
2. Not applicable if tumor is T1 or T2 (ie, not past the muscularis propria)
3. Positive CRM from primary tumor extension leads to increased
recurrence rates6
4. Estimated by the distance of primary tumor extension to the MRF (<
1mm distance to MRF predicts CRM involvement, 1-2 mm equivocal
involvement, >2 mm not involved). 7 CRM is a pathologic term and not
directly assessed at imaging.
5. Do not measure distance between a positive mesorectal lymph node
and the MRF predict CRM status. Knowledge of these implants may be
helpful to the surgeon to be careful in these areas during the TME
resection if close to the MRF but does not impact CRM status. As EVMI
is a direct extension of tumor (see below), can measure between EMVI
and MRF for this assessment.
iii. TME (total mesorectal excision): Surgical technique characterized by sharp
dissection in the avascular plane between the parietal and visceral pelvic fascia
where the rectum and surrounding structures encapsulated by the mesorectal
fascia are removed en bloc
1. Should extend 5 cm distal to the level of the tumor. Thus, for high rectal
cancers, the TME does not need to extend to the pelvic floor whereas
for mid and lower tumors, the TME is extended to the pelvic floor
 2. Has led to a marked decrease in local recurrence rates8 (ie, less than
10%)

iv. How distance to the MRF is measured:

1. Shortest linear distance between either the tumor and the MRF;
typically done on the high resolution FSE T2 weighted images. It may be
helpful to view from different planes.
a. <2 mm equates to a threatened CRM
b. <1 mm equates to an involved CRM
c. High specificity (92%) in predicting a negative CRM 9

<1 mm tumor to MRF so predicts ‘involved CRM’

 2. Distance to MRF is not applicable to peritonealized portions of the
upper and mid rectum. Thus, it cannot be measured at these levels.
v. Clinical importance of MRF distance:
1. A threatened or involved CRM margin reinforces need for neoadjuvant
chemoradiation to decrease local recurrence rates.
h. Relationship to anterior peritoneal reflection:
i. Anterior peritoneal reflection is a landmark seen on MR which helps to
determine the peritonealized and nonperitonealized portions of the rectum.
Important to delineate relationship for high and mid rectal tumors.
1. Below the anterior reflection, the rectum is completely extraperitoneal
in location and the MRF completely encircles the rectum. The boundary
between the peritoneal and extraperitoneal rectum courses in an
obliquely superior course from anterior to posterior.

2. CRM (and MRF) does not exist for the peritonealized portions of the
high and mid rectum
3. The anterior peritoneal reflection is variable in location but typically at
the level of the seminal vesicles in males and at the cervical-vaginal
junction in females
4. Seagull sign: the anterior peritoneal reflection on axial images is
reminiscent of a seagull
See ‘seagull’ sign on axial image to the right.

ii. Clinical importance

1. Tumor involvement of the anterior peritoneal reflection represents a
T4a tumor and may be at risk for seeding and intraperitoneal recurrence
in rectovesical space

T4a cancer

2. T category
a. Based on depth of invasion. Rectal cancer begins as a mucosal process (either as an
adenomatous or serrated polyp). After transformation to cancer, tumor can extend into
deeper layers of the bowel wall and beyond.

b. Depth of tumor wall involvement impacts treatment options. The major distinctions
involve differentiation between:
i. T1 versus T2. T1 tumors limited to within the submucosa (T1,N0,M0) are
amenable to local endoscopic excision by TEMS (transanal endoscopic micro-
 surgery) or TAMIS (transanal minimally invasive surgery) as opposed to more
invasive surgeries (LAR- low anterior resection, APR- abdominoperineal
resection).
1. Evaluated best by endoscopic rectal ultrasound which can directly
visualize the layers of the bowel wall.
2. Rectal MR with a phased array coil typically cannot distinguish between
T1 and T2 status as the layers of bowel wall are not resolved. Rectal MR
should not be used if this is the clinical question.
a. At MR, T1 and T2 are grouped together (T1/T2)

T1 tumor on left and T2 tumor on right on endoscopic ultrasound

ii. T2 versus T3. T2 tumors involve the muscularis propria but are contained by
this muscle layer whereas T3 tumors extend past the muscular propria.
1. Major treatment decision point whether neoadjuvant chemoradiation is
given (for T3) or not (for T2) prior to surgery. (more detail to follow in
next sections)
iii. Low risk T3 (≤ 5mm past the muscularis propria) v. high risk T3 (> 5mm past
muscularis propria).
1. Survival data suggest a difference in survival for T3 tumors defined by
extramural extension. 10 See section on ‘Extramural depth of invasion
(EMD)’.
2. Some suggest that decision for neoadjuvant chemoradiation should be
made at this boundary as opposed to T2 versus T3 status. 11 Currently,
decisions for neoadjuvant chemoradiation are not made as standard
care in the United States for this T-category subdivision.
c. T2 versus T3 status
i. Rectal MR and ERUS essentially equivalent for evaluation of T2 versus T3.
However, there is increasing consensus that MR is superior to ERUS (aside from
 the specific clinical question of T1 v T2 status) in assessing other features of
rectal cancer staging (ie, tumor location, relationship to mesorectal fascia for
CRM prediction, assessment for pelvic side wall lymph nodes, etc).
ii. Why assessment of tumor wall involvement important?
1. Major determinant for need for neoadjuvant chemoradiation prior to
surgery (positive regional lymphadenopathy and threatened
circumferential resection margin are other indications). German rectal
trial (n=823) showed that although neoadjuvant chemoradiation did not
change 5 year survival (76% for neoadjuvant versus 74% for adjuvant
therapy), it decreased local recurrence rates from 13% to 6% for clinical
stage T3 or T4 or node positive disease. 12
iii. How to make this determination
1. High resolution, thin slice T2-weighted images without fat saturation
a. Other sequences have less contrast resolution (post contrast T1,
T2 with fat saturation or DWI) which impacts tumor wall
assessment
2. Including series angled to the short axis and long axis of the tumor to
minimize volume averaging artifact. Otherwise, may lead to blurring
and difficulty in assessment.
3. Avoid motion (bowel and patient). Glucagon is helpful. Motion can
cause blurring of the muscularis propria.
4. Rectal gel may be helpful. Gel can minimally separate colonic walls and
locate the edges of the tumor to assess the wall status
iv. Imaging characteristics:
1. For a T2 tumor where tumor involves but is contained by the muscularis
propria (MP):
a. Smooth curving outer border of MP without lobulated outer
coutour
b. At MR, because we cannot resolve the layers of the bowel wall,
we cannot differentiate between T1 and T2. Tumors should be
noted T1/T2.
T1/T2 tumor. Note smoothly curving outer contour without break on sagittal and oblique axial

2. For a T3 where tumor extends past the MP

a. Discontinuity/disruption of the MP (presents as a curving low
signal band if intact) secondary to tumor
b. Broad-based bulge or focal lobulation beyond expected location
of MP; helpful to confirm on orthogonal planes
c. Small vessels can pierce MP at level of tumor. Does not
necessarily mean tumor extension. If the vessel is expanded,
should consider extramural vessel invasion (see EMVI section)
T3 cancer. Yellow line is expected location of the MP/rectal wall which has been obliterated by tumor
growth past it.

v. Peritumoral stranding into mesorectal fat

1. Represents desmoplastic response to tumor which may or may not
contain tumor.
2. Fine low signal spicules less likely to represent tumor than thick
irregular bands
3. Controversy whether to call tumor extension past wall or not. For fine
spicules, may consider fibrosis and not tumor.
d. Extramural depth of invasion
i. Independent predictor for survival; may correlate better than wall stage
1. Marked decrease in survival greater than 5 mm of EMD past the
muscularis propria (from 85% to 59%)10
ii. Not considered in AJCC staging system (7th ed) but incorporated in various MR
reporting templates. T3 is subdivided into:
1. T3a < 1mm
2. T3b 1- <5 mm
3. T3c 5-15 mm
4. T3d > 15 mm
iii. How to measure:
1. Measure distance between outermost edge of tumor and outer edge of
MP (or expected outer edge of the MP if not present due to
replacement by tumor). Do not include spicules.
2. HR resolution MR is equivalent to histology (within 0.5 mm against
histology). 13 Other studies show more variability. 14
Thin blue line is the expected level of the outer rectal wall. Yellow line represents the EMD

e. Involvement of pelvic structures

i. Potential structures include:
1. GU: bladder, left ureter, right ureter, cervix, uterus, vagina, prostate,
seminal vesicle, urethra)
2. Pelvic sidewall: obturator internus, piriformis, ischiococcygeus
3. Pelvic floor: pubococcygeus, ileococcygeus, puborectalis
4. Sacrum: Sacral involvement above S2 may be unresectable
5. Vessels: left internal iliac vessels, right internal iliac vessels, left external
iliac vessels, right external iliac vessels
6. Nerves: lumbosacral nerve roots, sciatic nerve
ii. Classified as a T4b tumor
iii. How to make imaging diagnosis
1. Loss of fat plane should be noted as abutment
2. Signal abnormality extending into structure is highly suspicious for
invasion
f. Relationship to the anal canal
i. Anal canal anatomy
1. Surgical anal canal which includes the puborectalis (see diagram) is
longer than the ‘anatomic’ anal canal which extends from the anal verge
to the dentate line (boundary between squamous and columnar
 epithelium- not visualized at MR but estimated by the junction between
the deep and superficial portions of the external sphincter)
2. Internal sphincter is a concentric thickening of the circular muscle layer
of rectal wall
3. External sphincter is a continuation of the pelvic floor
musculature/levator ani.
4. Intersphincteric space is a potential space between the external
sphincter and internal sphincter which may be seen as fatty intensity
cleft (The intersphincteric space is actually between the external
sphincter and the conjoint longitudinal muscle—this muscle may be
difficult to resolve from the outer border of the internal sphincter).

Anal canal anatomy coronal

ii. Traditionally tumor involvement of anal canal led to APR with loss of the
sphincter and permanent colostomy.
iii. Newer surgical techniques to extend the distal margin such as a LAR with
intersphincteric resection (ISR) where the tumor is limited to the internal
sphincter may be possible in select patient groups (here the internal and
external sphincters separated and the internal sphincter partially or completely
removed.
Blue line- ultralow LAR resection line

iv. How best to evaluate

1. Coronal high resolution thin slice T2-weighted images oriented to the
long axis of the anal canal
2. Assess for intact fatty intersphincteric space free of tumor. Extension to
involve intersphincteric plane and puborectalis/external sphincter leads
to APR

Arrow- extension past the intersphincteric fissure to involve the puborectalis on the left. Would not be
a candidate for ultralow LAR
 g. Mucinous subtype
i. More likely to be invasive or present with metastatic disease.
ii. >50% of the tumor/tumor pool has very high T2 signal intensity compared to
perirectal fat/muscle

Mucinous tumor

h. Extramural venous invasion (EMVI)

i. Presence of tumor in a mesorectal vessel extending from the primary tumor;
typically a vein
ii. By definition, EMVI makes the rectal tumor a T3 lesion.
iii. Imaging:
1. Assess any vessels (tubular or serpiginous structures) in the meso-rectal
fat extending from the tumor
a. A nodular contour, expansion of the lumen with tumor signal
should raise suspicion for vessel involvement and EMVI
iv. EMVI is an independent poor prognostic factor15
v. MR-detected EMVI predictive of higher rates of relapse 16
No EMVI9 normal veins without expansion EMVI

Lymph nodes

1. Sensitivity and specificity moderate (ie, 70%/70%) with available imaging biomarkers (size,
internal attenuation, outer border). So make best assessment using expected drainage
patterns.
a. Lymphatic flow in mesorectum is cephalad. (90% superior back to IMA takeoff; 10%
lateral to pelvic side wall lymphatics following branches of middle rectal from internal
iliac.
2. Use SAR criteria to maximize sensitivity and specificity

SAR criteria for positive LN

9 mm or greater

5-8 mm + 2 morphologic criteria

<5 mm + 3 morphologic criteria

Morphologic criteria: Round (first on left), internal heterogenous architecture (second from left),
irregular border (last two on right)

3. Regional lymph nodes (which constitute positive ‘N’ status on TNM system) represent malignant
involvement from direct lymphatic drainage from the tumor. These include mesorectal lymph
nodes extending back to the IMA take off from the aorta and lateral pelvic lymph nodes
(internal obturator and adjacent to branches of the internal iliac). Positive regional lymph nodes
lead to Stage 3 disease and patients typically undergo neoadjuvant chemoradiation or TNT.

Mesorectal lymph node

Lateral pelvic lymph node

4. Metastatic lymph nodes are ones that result from hematogenous spread. If these lymph node
locations are involved, this is metastatic spread and confers Stage 4 disease which is treated
palliatively for the most part. So enlarged RTP lymph nodes above the level of the IMA, External
iliac LN and inguinal lymph nodes are metastatic. One caveat- if the rectal cancer is low and
involves the anal canal, peripheral external iliac and inguinal lymph nodes represent regional
adenopathy as lymphatics can drain to these regions with anal canal involvement.

References:

1. Beets-Tan RGH, Lambregts DMJ, 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;23:2522-31.
2. Monson JRT, Weiser MR, Buie WD, Chang GJ, Rafferty JF, Stand Practice Task Force Amer S.
Practice Parameters for the Management of Rectal Cancer (Revised). Dis Colon Rectum 2013;56:535-50.
3. Fitzgerald TL, Brinkley J, Zervos EE. Pushing the Envelope Beyond a Centimeter in Rectal Cancer:
Oncologic Implications of Close, But Negative Margins. J Am Coll Surg 2011;213:589-95.
4. Kim YW, Kim NK, Min BS, et al. Factors Associated With Anastomotic Recurrence After Total
Mesorectal Excision in Rectal Cancer Patients. J Surg Oncol 2009;99:58-64.
5. Leo E, Belli F, Miceli R, et al. Distal clearance margin of 1 cm or less: a safe distance in lower
rectum cancer surgery. Int J Colorectal Dis 2009;24:317-22.
6. Quirke P, Dixon MF, Durdey P, Williams NS. Local Recurrence of rectal adenocarcinoma due to
inadequate surgical resection - histopathological study of lateral tumor spread and surgica excision.
Lancet 1986;2:996-9.
7. Taylor FGM, Quirke P, Heald RJ, et al. One millimetre is the safe cut-off for magnetic resonance
imaging prediction of surgical margin status in rectal cancer. Br J Surg 2011;98:872-9.
8. Quirke P, Steele R, Monson J, et al. Effect of the plane of surgery achieved on local recurrence in
patients with operable rectal cancer: a prospective study using data from the MRC CR07 and NCIC-CTG
CO16 randomised clinical trial. Lancet 2009;373:821-8.
9. Brown G, Daniels IR, Heald RJ, et al. Diagnostic accuracy of preoperative magnetic resonance
imaging in predicting curative resection of rectal cancer: prospective observational study. Br Med J
2006;333:779-82.
10. Merkel S, Mansmann U, Siassi M, Papadopoulos T, Hohenberger W, Hermanek P. The prognostic
inhomogeneity in pT3 rectal carcinomas. Int J Colorectal Dis 2001;16:298-304.
11. Evans J, Patel U, Brown G. Rectal Cancer: Primary Staging and Assessment After
Chemoradiotherapy. Semin Radiat Oncol 2011;21:169-77.
12. Sauer R, Becker H, Hohenberger W, et al. Preoperative versus postoperative chemoradiotherapy
for rectal cancer. N Engl J Med 2004;351:1731-40.
13. Fowler JM, Beagley CE, Blomqvist L, et al. Extramural depth of tumor invasion at thin-section MR
in patients with rectal cancer: Results of the MERCURY Study. Radiology 2007;243:132-9.
14. Pedersen BG, Moran B, Brown G, Blomqvist L, Fenger-Gron M, Laurberg S. Reproducibility of
Depth of Extramural Tumor Spread and Distance to Circumferential Resection Margin at Rectal MRI:
Enhancement of Clinical Guidelines for Neoadjuvant Therapy. Am J Roentgenol 2011;197:1360-6.
15. Heald RJ, Ryall RDH. Recurrence and survival after total mesorectal excision for rectal cancer.
Lancet 1986;1:1479-82.
16. 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;95:229-36.