1.
www.profginabrown.com
@prof_gina_brown
Should MRI be used to stage
SPECC
rectal lesions?

2.
• 1994
• Endorectal coil tested for staging of 12
rectal cancers
• High resolution T2
• T2-weighted FSE Sequence (4,000/100, 256
x 256 matrix, echo train length of 16). All
images were obtained with an 80-120-mm
field of view and 3-mm-thick sections with
no gap.
• Pixel size =120/256 = 0.46mm x 0.46mm

3.
High resolution MRI
Without endorectal coil
Pixel size
0.6 x 0.6mm

4.
Technique Essential Checklist
• Scan duration = quality
• 4-6 NSA/NEX and T2- FSE / TSE
• 0.6mm x 0.6mm x 3mm = 1.1mm3 voxel
• Adequate coverage – 5cm above top of tumour
• Perpendicular to the rectal wall
• Low rectal cancer – parallel to anal canal
• Ensure discontinuous deposits are covered on high res
• Buscopan
• Saturation Bands
• firm coil placement with secure strapping

5.
With Saturation
band and
buscopan
Without
Saturation band

6.
The submucosal fold pattern

7.
ERC subclassification
• T0/early T1sm1 – no evident disruption of the
submucosa – entire thickness of SM appears preserved
• T1sm2 – at least 1mm of submucosa is preserved
• T1sm3/early T2: full thickness of muscularis propria is
preserved but <1mm submucosa is visible
• T2 early >1mm muscularis is preserved
• T2/T3a – 0mm of muscularis preserved microscopic
invasion beyond muscularis <1mm – prognosis
identical for this subgroup
• T3b 1-5mm – good prognosis

8.
• 68 yo female
• Rectal bleeding, colonscopy malignant 2cm
adenocarcinoma (biopsy proven) mass in
lower rectum
• Performance status 1, no comorbidities

9.
1. APE/ELAPE
2. Ultralow TME
3. Local excision
4. Full thickness TEM
5. Preop Rx

10.
1. APE/ELAPE
2. Ultralow TME
3. Local excision
4. Full thickness TEM
5. Preop Rx
Morphology – flat semi-annular
Diameter 16mm
Thickness of lesion : 7mm
Clockface location of central depression
=4 oclock
Invasive edge = 5mm diameter
Muscularis fully preserved
Submucosa/muscle interface lost over
3mm distance on single slice at 4 o’clock
Lymph nodes show smooth nodal
capsule and no heterogeneity - benign
Assess height of 6.5 cm above anal verge
and 12mm above puborectalis sling
No extramural venous invasion
T1sm3/ with potential focal early T2
invasion on a single slice section

11.
Final pathology
• CLINICAL SUMMARY
TEMS-anorectal polypoid lesion
• HISTOLOGY
The referred sections are of colonic mucosa with a tubulovillous adenoma having
both highly and low-grade dysplasia, with an area of moderately to poorly
differentiated adenocarcinoma with desmoplasia, tumour budding and focal
mucinous differentiation. Invasive tumour area measures 13 x 5.5 mm and just
involve the innermost fibres of the muscularis propria. There is lymphatic vascular
invasion and focal submucosal venous angioinvasion. In therefore constitutes
a higher risk lesion, although the nearest deep excisional or marginal clearance is
3.5 mm. Further deeper sections do not reveal any deeper invasion.
• pT2 NX MX LVI+ R0

12.
I/12 after TEM 3/12 after TEM

13.
4 years post Surveillance
• She has been extremely well since her last review in the clinic. She is able
to manage a normal diet and her weight is stable.
• She refers to mild incontinence for gas and stool but denies any blood or
mucous in the stool.
• On examination today, the abdomen was soft, non tender. There was no
palpable lymphadenopathy. On PR examination, the sphincter tone was
mild. There was no palpable mass in the rectum.
• There was no blood or stool on the gloved examination finger. The CEA
on the 11th February was 3 and the CA19-9 was 3.
• The last MRI showed no evidence of recurrent disease.

14.
Post TEM surveillance
of mesorectum

15.
Local excision
scar

16.
28/34 had 1 or more high risk
features
22/24 patients with low tumours
and high risk features which
would have required APER have
so far avoided radical surgery
and remain disease free at a
median follow up 3.2 years.
Lesions identified on MRI –
roadmap for TEM/TAMIS: LE
Assess mesorectum and TEM
planes
Review histology
Counsel patients: completion
TME surgery versus adjuvant
CRT and surveillance

17.
Recommended reporting structure for staging
early rectal cancer using MRI
• State morphology – flat, polypoidal, mucin containing
• Measure diameter and thickness of lesion
• If polypoidal –state site and diameter of fibromuscular stalk
• If flat – quadrant or clockface location of central depression versus raised rolled edges
• Measure extent/diameter of invasive border
• Assess degree of preservation of the mucosa, submucosa, muscularis propria layers at
the stalk
• Assess lymph nodes for malignant characteristics based on nodal capsule breach or
heterogeneity of signal
• Assess height of lesion in relation to anal verge and puborectalis sling
• Evaluate extramural veins for discontinuous spread

18.
19
PRE-therapeutic MRI assessment of Early StagE Rectal Cancer and significant
Rectal Polyps to aVoid major resectional surgery: A new approach to the
management of Early stage rectal cancer
MRI-PRESERVE/MINSTREL
Current Investigators and Collaborators
Professor Gina Brown Professor Wendy Atkin Mr Chris Cunningham Professor Robert Goldin
Mr Greg Wynne Dr Monica Terlizzo Mr Shahnawaz Rasheed Mr James Kinross
Mr Ian Swift Ms Karen Thomas Ms Annabel Shaw (Croydon Res Fellow)

19.
The Royal Marsden MRI-PRESERVE 3rd
May 201720
Background
– Increase in detection rates of early rectal cancer
(ERC) through bowel cancer screening programme
(BCSP)
– >10% of positive FOBT diagnosed with malignancy
– ~ 30% rectal cancer
– ~ 30% limited to bowel wall without nodal spread

20.
The Royal Marsden
Current problem
Current pathway following diagnosis of polyp / polyp cancer does not
involve pre-op staging
Cancer polyps that are removed endoscopically are mostly subjected
to TME surgery afterwards
At present, 90% of stage I cancers undergo major
resectional surgery when detected by bowel screening
MRI-PRESERVE 3rd May 201721

21.
The Royal Marsden MRI-PRESERVE 3rd May 201722
65 patients undergoing primary surgery/local excision for mrT3b or less tumours
MRI accuracy for <T1sm2 =89%(95%Cl:63-87%) PPV 77%, NPV 92%
MRI accuracy <T2 89% (95%Cl:79%-95%), PPV 93%, NPV 81%
Kappa score 0.7 for radiologists (GB and SB)
21/65(32%) patients underwent local excision or TEM 20/21 were staged as
MR<T2 and confirmed as such by pathology.
On follow up, none had relapse.

22.
The Royal Marsden
MRI staged patients <T2 undergoing radical
surgery
22/44 were <mrT2. MRI accuracy in predicting lymph node status was
84%(95%Cl:70%-92%), PPV 71% and NPV 90%
If the decision had been made to offer local excision on MRI TN staging
rather than clinical assessment a significant increase in organ
preservation surgery from 32% to 60% would have been observed
(difference 23%, 95%Cl:9%-35%)
MRI-PRESERVE 3rd May 201723

23.
The Royal Marsden
Future goal
To improve organ preservation rates in screen
detected / symptomatic colonoscopy identified
ERC by incorporating MRI into the
staging pathway
MRI-PRESERVE 3rd May 201724

24.
The Royal Marsden
MRI Staging - MINSTREL
New MRI staging system
Four surgical planes
• Mucosal – endoscopic mucosal excision (EMR)
• Submucosal – endoscopic submucosal resection / partial thickness
transanal endoscopic microsurgery (TEMS)
• Deep submucosal / muscle invasion – full thickness TEMS
• Deep muscle / extramural disease – full oncological resection
Increase in organ preservation by at least 10%
High-resolution MRI as a method to predict potentially safe surgical planes in early
rectal cancer patients
Balyasnikova S et al.
MRI-PRESERVE 3rd May 201725

25.
The Royal Marsden
MRI-allocated surgical planes
MRI-PRESERVE 3rd May 201726
A
Mucosal
EMR (mucosal) resection could
clear deep margin
No definite SM
(submucosal)
invasion
B
Submucosal
Endoscopic Sub-mucosal
Resection / partial thickness TEMS
Definite SM invasion
seen but <1mm SM
>1mm SM invasion
seen but >1mm SM
preserved to MP
C
Deep submucosal / early muscle invasion
Full thickness TEMS
SM invasion seen,
<1mm if SM
preserved but no
definite muscle
invasion
Definite muscle
invasion seen but
greater than 1mm
muscle preserved to
mesorectum
D
Deep muscle or extramural disease
full oncological resection required
Less that 1mm of
muscle preserved but
no mesorectal
invasion
Definite invasion of
mesorectum

26.
The Royal Marsden
Minstrel A – mucosal
EMR (mucosal) resection could clear deep
margin
T1 sm0/1
MRI-PRESERVE 3rd May 201727

27.
The Royal Marsden MRI-PRESERVE 3rd May 201728
Minstrel B – submucosal
Endoscopic Sub-mucosal Resection / partial
thickness TEMS
T1 sm2/3

28.
The Royal Marsden MRI-PRESERVE 3rd May 201729
Minstrel C – deep submucosal / muscle invasion
Full thickness TEM
T1 sm 3

29.
The Royal Marsden
MRI MINSTREL-PRESERVE
Feasibility phase to confirm diagnostic accuracy in
predicting feasibility of TEM/local excision plane
Multi-centre interventional trial – PRESERVE Trial
End points:
>10% 5 year organ preservation rate in screen-detected
T1 / early T2 rectal cancers
>80% disease-free survival / overall survival
Radiologists have been trained in the new MRI staging
system for early rectal cancer (as investigated in the
Minstrel Study)
MRI-PRESERVE 3rd May 201730

30.
The Royal Marsden
Primary objectives
to increase rates of successful local excision of early rectal cancers
MRI-PRESERVE 3rd May 201731
Secondary objectives
• to improve MRI radiological staging of ERC, and increase ability for
surgical resection plane planning based upon initial imaging
• to train radiologists in the use of the new reporting system, and
demonstrate good agreement with the correct surgical plane
• to improve local excision techniques – (R0 resection) through
integration of improved pre-operative imaging protocols with real time
tissue phenotyping technologies - i-Knife
• embedding new technologies developed at Imperial into mainstream
surgical practice for improved therapy of ERC
• identifying biological phenotypes that predict good outcomes following
local excision

31.
The Royal Marsden
Control Comparison
surgical management according to conventional MDT decisions, clinical
assessment, and standard MRI reporting protocols – contemparaneous
matched data to be obtained from Public Health England
MRI-PRESERVE 3rd May 201732
Intervention
Surgical decision determined by extent of deep margin and absence
of EMVI /obvious malignant nodes/ discontinuous deposits
• Mucosal - endoscopic mucosal excision (EMR)
• Submucosal – endoscopic submucosal resection / partial
thickness TEMS
• Deep submucosal / early muscle invasion – full thickness
TEMS
• Deep muscle / extramural disease – full oncological resection

32.
The Royal Marsden
Inclusion / Exclusion Criteria
Inclusion
• Rectal polyps >2cm in diameter
• Suspected malignant rectal polyp
• Biopsy proven rectal malignancy
Exclusion
• Patients requiring neoadjuvant therapy
• Metastatic disease
• Synchronous malignancy
• Contraindications to MRI
MRI-PRESERVE 3rd May 201733

33.
The Royal Marsden
Statistical recommendations
Simon two-stage optimal design
First stage
• Total of 30 patients
• Success in at least 4/30 in order to proceed
• Success defined as organ preservation with R0
Second stage
• Total sample size of 89
• Success in at least 14/89 for trial to reach positive conclusion
• success will be defined from longer follow-up, including
disease-free survival, overall survival and overall stoma-free
survival
MRI-PRESERVE 3rd May 201734

34.
The Royal Marsden MRI-PRESERVE 3rd May 201735
Trial Flowchart

35.
The Royal Marsden
Outcomes for patients
This will all result in improvements in the management of patients with ERC
Cost benefit
• Reduction of TME major resections
• Increased organ-preservation rates
• Reduced length of hospital stay / perioperative complications
Improved quality of life
• Reduced side effect / complication profile for all patients
• Limit the surgical intervention for otherwise well asymptomatic
screening patients /symptomatic patients
MRI-PRESERVE 3rd May 201736

36.
The Royal Marsden37
MINSTREL – Superficial / SM plane/ Full
thickness

37.
• Radiologists at recruting sites are trained
and hold delegated responsibility
• Eligible patients are identified on
colonoscopy if they are found to have a
20mm to 50mm rectal tumour within
150mm of the anal verge (consent and
completion of endoscopy CRF)
• All patients who enter the trial will be
sent for an MRI. The MRI will be
reported using the novel staging
proforma (radiology CRF)
• The patients will proceed to excision or
resection of the tumour as per clinician /
MDT discussion. (MDT CRF)
• The appropriateness of preoperative
stage will be compared against
histopathology gold standard (Pathology
CRF)

38.
The Royal Marsden39

39.
The Royal Marsden
Recruitment
When is the optimal time for surgery after completion of CRT? 13/3/201440