Transcript No Slide Title

Optimizing Triage to Preoperative
Chemoradiation in T2 Rectal
Cancer Based on Mesorectal Lymph
Node Size: A Decision Analysis
Informed by Patient Outcomes
Chang, Connie Y., M.D., Pandharipande, Pari, M.D., M.P.H.,
Harisinghani, Mukesh, M.D., Gazelle, G. Scott, M.D., M.P.H., Ph.D.
HARVARD
MEDICAL SCHOOL
Background: Rectal Cancer
 MRI
has had increasing role in
preoperative planning for rectal
cancer (T-stage)
 Large
degree of overlap of size of
normal/reactive and cancercontaining peri-rectal lymph
nodes.
Lymph Node Staging in Rectal Cancer
Stage T2 rectal cancer
Perirectal lymph nodes
Purpose
 To
optimize key patient
outcomes in T2 rectal cancer
by identifying mesorectal
lymph node size criteria for
triage to preoperative
chemoradiation.
Methods
 Decision-Analytic
Model
 Model inputs derived from
literature
 T2 rectal cancer
Methods
Treat All Patients with
Pre-operative Chemoradiation
Treat If any Mesorectal Lymph
Nodes are > 3 mm
Stage T2 Rectal
Cancer
Treat If any Mesorectal Lymph
Nodes are > 5 mm
Treat If any Mesorectal Lymph
Nodes are > 7 mm
No Preoperative Chemoradiation
for Any Patients
Methods: Four Disease Scenarios
True Positive
(TP)
False Negative
(FN)
False Positive
(FP)
True Negative
(TN)
Methods: Four Disease Scenarios
Preoperative
chemoradiation
No preoperative
chemoradiation
True Positive
(TP)
False Negative
(FN)
False Positive
(FP)
True Negative
(TN)
Methods: Four Disease Scenarios
Lymph node
metastases at
pathology
Preoperative
chemoradiation
No preoperative
chemoradiation
No lymph node
metastases at
pathology
True Positive
(TP)
False Negative
(FN)
False Positive
(FP)
True Negative
(TN)
Methods: Four Disease Scenarios
Lymph node
metastases at
pathology
No lymph node
metastases at
pathology
TP Appropriate
FP
Preoperative
treatment –
chemoradiation expected morbidity
of preoperative
chemoradiation
TN Appropriate
No preoperative FN
treatment
chemoradiation
Methods: Four Disease Scenarios
Lymph node
metastases at
pathology
No lymph node
metastases at
pathology
TP Appropriate
FP Unnecessary
Preoperative
treatment –
chemoradiation
chemoradiation expected morbidity
of preoperative
chemoradiation
TN Appropriate
No preoperative FN
treatment
chemoradiation
Methods: Four Disease Scenarios
Lymph node
metastases at
pathology
No lymph node
metastases at
pathology
TP Appropriate
FP Unnecessary
Preoperative
treatment –
chemoradiation
chemoradiation expected morbidity
of preoperative
chemoradiation
TN Appropriate
No preoperative FN Increased
morbidity of
treatment
chemoradiation
post-operative
chemoradiation,
increased likelihood
of local recurrence
Base Case Analysis
All nodes considered
malignant
Nodes considered malignant
if any node > 3 mm
Nodes considered malignant
if any node > 5 mm
Nodes considered malignant
if any node > 7 mm
No nodes considered
malignant
* From Kim, et al (2004)
Sensitivity
1
Specificity
0
0.91
0.43
0.73
0.75
0.55
0.91
0
1
Base Case Analysis
Pre-operative
Post-operative
Chemoradiation Chemoradiation
Acute
Chemoradiation
Toxic Effects
Long-term
Chemoradiation
Toxic Effects
5-year Probability
of Local
Recurrence
* Sauer, et al (2004)
27%
40%
14%
24%
6%
13%
Secondary Analysis

Individual node radiology-pathology
correlation
 Schnall et al (1994), Brown et al (2003)
 Expanded data (318 nodes from 78
patients)
 Subject to “clustering bias”

USPIO lymph node contrast agent
 Lahaye et al (2008)
Sensitivity Analysis
Performed to assess the impact of
uncertainty in key model parameter
estimates upon clinical outcomes
 Calculated 95% confidence intervals for
sensitivity and specificity of each
strategy


Repeated analysis with upper and lower
limits of the confidence intervals.
Results – Base Case Analysis
30.00%
25.00%
20.00%
15.00%
10.00%
5.00%
0.00%
Treat All
Treat > 3mm Treat > 5mm Treat > 7 mm Treat None
% Patients with Acute Chemoradiation Toxicity
% Patients with Long-Term Chemoradiation Toxicity
5-Year Local Recurrence
Results – Base Case Analysis
30.00%
●
25.00%
Lowest Value
●
20.00%
15.00%
●
10.00%
5.00%
0.00%
Treat All
Treat > 3mm Treat > 5mm Treat > 7 mm Treat None
% Patients with Acute Chemoradiation Toxicity
% Patients with Long-Term Chemoradiation Toxicity
5-Year Local Recurrence
Results – Base Case Analysis
30.00%
●
25.00%
Lowest Value
20.00%
15.00%
10.00%
5.00%
*
**
*
●*
*
*
0.00%
Treat All
Treat > 3mm Treat > 5mm Treat > 7 mm Treat None
% Patients with Acute Chemoradiation Toxicity
% Patients with Long-Term Chemoradiation Toxicity
5-Year Local Recurrence
Results – Base Case Analysis
30.00%
●
25.00%
Lowest Value
20.00%
15.00%
10.00%
5.00%
●
0.00%
Treat All
Treat > 3mm Treat > 5mm Treat > 7 mm Treat None
% Patients with Acute Chemoradiation Toxicity
% Patients with Long-Term Chemoradiation Toxicity
5-Year Local Recurrence
Results –
Secondary and Sensitivity Analysis


Individual node analysis – similar pattern of
results to base case analysis
Upper limits of all confidence intervals –
differed for long-term chemoradiation toxicity


Minimized if treat no patients preoperatively
Lower limits of all confidence intervals –
differed only for acute chemoradiation toxicity

Minimized if treat patients with LNs > 7 mm
Results – Sensitivity Analysis
USPIO-Enhancement
30.00%
25.00%
20.00%
15.00%
10.00%
*
5.00%
**
0.00%
Treat All
Treat >
Treat > Treat > 7
Treat
USPIO
3mm
5mm
mm
None
Positivity
% Patients with Acute Chemoradiation Toxicity
% Patients with Long-Term Chemoradiation Toxicity
5-Year Local Recurrence
Limitations
 Reduction
of a complex disease into
a simple decision model.
 Correct identification of stage T2
rectal cancer
Conclusions

Lymph node size criteria used is based
on outcome prioritized at the individual
patient level
 Acute
toxicity – treat no patients
 Long-term toxicity – treat > 7 mm
 Local recurrence – treat all patients

A higher threshold may better balance all
three outcomes.
Conclusions

USPIO-positivity should be better than
all size criteria for triaging patients to
pre-operative chemoradiation.
Thank you!
References


Brown G, Richards, CJ, Bourne, MW, et al.
Morphologic predictors of lymph node
status in rectal cancer with use of highspatial-resolution MR imaging with
histopathologic comparison. Radiology
2003; 227:371-377.
Kim JH, Beets GL, Kim, MJ, et al. High
resolution MR imaging for nodal staging in
rectal cancer: are there any criteria in
addition to the size? EJR 2004; 52:78-83.
References



Lahaye MJ, Engelen SME, Kessels AGH, et al.
USPIO-enhanced MR Imaging for Nodal Staging in
Patients with Primary Rectal Cancer: Predictive
Criteria. Radiology 2008; 246(3), 804-811.
Schnall MD, Furth EE, Rosato EF, Kressel HY.
Rectal tumor stage: Correlation of endorectal MR
imaging and pathologic findings. Radiology 1994;
190:709-714.
Sauer R, Becker H, Hohenberger W, et al.
Preoperative versus postoperative
chemoradiotherapy for rectal cancer. NEJM 2004;
351;17:1731-40.
Secondary Analysis
Sensitivity
Specificity
1
0
0.30
0.94
0.86
0.51
0.57
0.82
0
1
All nodes considered
malignant
Nodes considered
malignant if > 3 mm
Nodes considered
malignant if > 3 mm
Nodes considered
malignant if > 3 mm
No nodes considered
malignant
* Schnall et al (1994) and Brown et al (2003)