Rationale and Objectives
The aim of this study was to optimize treatment decisions for patients with suspected stage T2 rectal cancer on the basis of mesorectal lymph node size at magnetic resonance imaging.
Materials and Methods
A decision-analytic model was developed to predict outcomes for patients with stage T2 rectal cancer at magnetic resonance imaging. Node-positive patients were assumed to benefit from chemoradiation prior to surgery. Imperfect magnetic resonance imaging performance for primary cancer and mesorectal nodal staging was incorporated. Five triage strategies were considered for administering preoperative chemoradiation: treat all patients; treat for any mesorectal node >3, >5, and >7 mm in size; and treat no patients. If nodal metastases or unsuspected stage T3 disease went untreated preoperatively, postoperative chemoradiation was needed, resulting in poorer outcomes. For each strategy, rates of acute and long-term chemoradiation toxicity and of 5-year local recurrence were computed. Effects of input parameter uncertainty were evaluated in sensitivity analysis.
Results
The optimal strategy depended on the outcome prioritized. Acute and long-term chemoradiation toxicity rates were minimized by triaging only patients with nodes >7 mm to preoperative chemoradiation (18.9% and 10.8%, respectively). A treat-all strategy minimized the 5-year local recurrence rate (5.6%). A 7-mm nodal triage threshold increased the 5-year local recurrence rate to 8.0%; when no patients were treated preoperatively, the local recurrence rate was 10.1%. With improved primary tumor staging, all outcomes could be further optimized.
Conclusions
Mesorectal nodal size thresholds for preoperative chemoradiation should depend on the outcome prioritized: higher size thresholds reduce chemoradiation toxicity but increase recurrence rates. Improvements in nodal staging will have greater impact if primary tumor staging can be improved.
Rectal cancer remains a leading cause of malignancy in the United States, with 40,290 new cases predicted in 2012 . For patients with locally advanced disease, designated as stage T3 or node-positive disease, surgery with adjuvant chemoradiation has recently emerged as the standard of care . Primarily because of increased patient compliance, enhanced tumor radiosensitivity, and radiation sparing of the colonic segment that will be used as a future reservoir for stool, patient outcomes are optimized with preoperative (as opposed to postoperative) chemoradiation . As a result, optimal therapeutic choices hinge on the accuracy of initial staging; whereas patients with advanced disease benefit from preoperative adjuvant therapy, those with stage T2 (or less) node-negative disease may incur unnecessary treatment toxicities if overstaged and overtreated .
Magnetic resonance imaging (MRI) has been increasingly used for preoperative staging and treatment planning in patients with rectal cancer . To date, most studies of MRI performance in this setting have focused on its ability to discern stage T2 and T3 tumors . Several authors have proposed MRI criteria for determining primary tumor stage . Meaningful criteria for determining lymph node positivity at MRI, particularly for commonly visible mesorectal nodes, have been difficult to establish . With increasing nodal size, the likelihood of nodal malignancy is higher, but there is considerable overlap between the sizes of benign and malignant lymph nodes . Other criteria for detecting mesorectal lymph node metastases at MRI, on the basis of border or signal characteristics, can be subject to interpretation variability . Accurate determination of lymph node status is particularly critical for patients with tumors designated as stage T2 preoperatively, because suspected nodal involvement in these patients necessitates preoperative chemoradiation despite the primary tumor’s T2 designation.
Get Radiology Tree app to read full this article<
Materials and methods
Decision Analysis Overview
Get Radiology Tree app to read full this article<
Decision Model and Structure
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Decision Model Input Data
Get Radiology Tree app to read full this article<
Table 1
Input Parameter Estimates for the Decision Model
Parameter BCE Sensitivity Analysis Range Source Sensitivity of each model strategy for detecting mesorectal nodal metastases (75 patients total, 22/75 with nodal metastases) Option 1: assume all patients have mesorectal nodal metastases (even if none visualized at MRI) 1 – Option 2: assume patients with any mesorectal node >3 mm have metastases 0.91 0.71–0.99 ∗ Option 3: assume patients with any mesorectal node >5 mm have metastases 0.73 0.50–0.89 ∗ Option 4: assume patients with any mesorectal node >7 mm have metastases 0.55 0.32–0.76 ∗ Option 5: assume no patients have metastases 0 – Specificity of each model strategy for detecting mesorectal nodal metastases Option 1 0 – Option 2 0.43 0.30–0.58 ∗ Option 3 0.75 0.62–0.86 ∗ Option 4 0.91 0.79–0.97 ∗ Option 5 1 – Probability of stage T2 primary tumor at pathology, given T2 appearance at MRI ( P trueT2 ) 0.65 (0.5–1.5) × BCE † Probability of stage T3 primary tumor at pathology, given T2 appearance at MRI ( P falseT2 ) 0.35 – 1− p trueT2 † Probability of mesorectal nodal metastasis given stage T2 primary tumor at pathology 0.25 (0.5–1.5) × BCE Probability of mesorectal nodal metastasis given stage T3 primary tumor at pathology 0.62 (0.5–1.5) × BCE
BCE, base-case estimate; MRI, magnetic resonance imaging.
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Table 2
Outcome Estimates by Pathologic Stage and Treatment Received
Outcome Model Estimate Source T2N0 ∗ : 5-y probability of local cancer recurrence 0.037 T2N+ † treated with preoperative chemoradiation: 5-y probability of local cancer recurrence 0.070 ( ‡ , ) T2N+ treated with postoperative chemoradiation: 5-y probability of local cancer recurrence 0.15 ( ‡ , ) T3N0 treated with preoperative chemoradiation: 5-y probability of local cancer recurrence 0.046 ( ‡ , ) T3N0 treated with postoperative chemoradiation: 5-y probability of local cancer recurrence 0.10 ( ‡ , ) T3N+ treated with preoperative chemoradiation: 5-y probability of local cancer recurrence 0.096 ( ‡ , ) T3N+ treated with postoperative chemoradiation: 5-y probability of local cancer recurrence 0.21 ( ‡ , ) Acute toxicity rate, if received preoperative chemoradiation 0.27 Acute toxicity rate, if received postoperative chemoradiation 0.40 Long-term toxicity rate, if received preoperative chemoradiation 0.14 Long-term toxicity rate, if received postoperative chemoradiation 0.24
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Table 3
Additional Parameter Estimates for Sensitivity and Secondary Analyses
Parameter Parameter Estimate Source Sensitivity and specificity of each model strategy for detecting mesorectal nodal metastases Sensitivity, specificity ∗ (pooled data included 416 nodes, 78/416 with nodal metastases) Option 1: assume all patients have mesorectal nodal metastases (even if none visualized at magnetic resonance imaging) 1, 0 Option 2: assume patients with any mesorectal node >3 mm have metastases 0.78, 0.71 Option 3: assume patients with any mesorectal node >5 mm have metastases 0.46, 0.91 Option 4: assume patients with any mesorectal node >7 mm have metastases 0.27, 0.96 Option 5: assume no patients have metastases 0, 1 Sensitivity and specificity of irregular nodal border for detecting mesorectal nodal metastases 0.75, 0.98 Sensitivity and specificity of irregular nodal border or mixed signal intensity for detecting mesorectal nodal metastases 0.85, 0.98
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
MRI performance of each nodal size threshold for detecting nodal metastases
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
MRI uncertainty in stage T2 designation of the primary tumor
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Underlying prevalence of lymph node metastases
Get Radiology Tree app to read full this article<
Outcomes: chemoradiation toxicity and cancer recurrence
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Base-case and Sensitivity Analyses
Base-case analysis
Get Radiology Tree app to read full this article<
Sensitivity analysis
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Secondary Analysis: Morphologic Criteria for Positive Mesorectal Nodes
Get Radiology Tree app to read full this article<
Results
Base-case Analysis
Get Radiology Tree app to read full this article<
Table 4
Base-case and Sensitivity Analyses Results
Parameter Estimate Option 1: Treat All Option 2: Treat >3 mm Option 3: Treat >5 mm Option 4: Treat >7 mm Option 5: Treat None Rate of acute chemoradiation toxicity Base-case ∗ 27.0% 22.5% 19.7% 18.9% 20.5% Base-case sensitivity, upper limit specificity ∗ 27.0% 20.8% 18.5% 18.1% 20.5% Base-case sensitivity, lower limit specificity ∗ 27.0% 24.0% 21.2% 20.2% 20.5% Upper limit sensitivity, base-case specificity ∗ 27.0% 22.1% 19.0% 17.8% 20.5% Lower limit sensitivity, base-case specificity ∗ 27.0% 23.5% 20.9% 20.0% 20.5% Upper limit sensitivity, upper limit specificity ∗ 27.0% 20.4% 17.7% 17.1% 20.5% Lower limit sensitivity, lower limit specificity ∗ 27.0% 25.0% 22.4% 21.3% 20.5%p trueT2 0.33 (0.5 × base-case estimate) 27.0% 26.2% 26.3% 27.0% 30.3% 0.98 (1.5 × base-case estimate) 27.0% 18.8% 13.1% 10.8% 10.7% Probability of nodal metastasis given T2 tumor at pathology 0.13 (0.5 × base-case estimate) 27.0% 21.4% 17.8% 16.4% 17.3% 0.38 (1.5 × base-case estimate) 27.0% 23.5% 21.7% 21.4% 23.8% Probability of nodal metastasis given T3 tumor at pathology 0.31 (0.5 × base-case estimate) 27.0% 23.0% 20.4% 19.5% 20.5% 0.93 (1.5 × base-case estimate) 27.0% 22.0% 19.0% 18.3% 20.5% Sensitivity and specificity (node-by-node) basis † 27.0% 20.0% 19.3% 19.6% 20.5% Rate of long-term chemoradiation toxicity Base-case ∗ 14.0% 12.0% 10.9% 10.8% 12.3% Base-case sensitivity, upper limit specificity ∗ 14.0% 11.2% 10.3% 10.4% 12.3% Base-case sensitivity, lower limit specificity ∗ 14.0% 12.7% 11.6% 11.4% 12.3% Upper limit sensitivity, base-case specificity ∗ 14.0% 11.7% 10.3% 9.9% 12.3% Lower limit sensitivity, base-case specificity ∗ 14.0% 12.7% 11.8% 11.6% 12.3% Upper limit sensitivity, upper limit specificity ∗ 14.0% 10.9% 9.7% 9.6% 12.3% Lower limit sensitivity, lower limit specificity ∗ 14.0% 13.5% 12.5% 12.2% 12.3%p trueT2 0.33 (0.5 × base-case estimate) 14.0% 14.1% 14.7% 15.5% 18.2% 0.98 (1.5 × base-case estimate) 14.0% 9.8% 7.1% 6.0% 6.5% Probability of nodal metastasis given T2 tumor at pathology 0.13 (0.5 × base-case estimate) 14.0% 11.4% 9.8% 9.3% 10.4% 0.38 (1.5 × base-case estimate) 14.0% 12.5% 12.0% 12.1% 14.3% Probability of nodal metastasis given T3 tumor at pathology 0.31 (0.5 × base-case estimate) 14.0% 12.3% 11.4% 11.2% 12.3% 0.93 (1.5 × base-case estimate) 14.0% 11.6% 10.4% 10.3% 12.3% Sensitivity and specificity (node-by-node) basis † 14.0% 10.9% 11.0% 11.5% 12.3% 5-y rate of local cancer recurrence Base-case ∗ 5.6% 6.3% 7.2% 8.0% 10.1% Base-case sensitivity, upper limit specificity ∗ 5.6% 6.4% 7.3% 8.0% 10.1% Base-case sensitivity, lower limit specificity ∗ 5.6% 6.2% 7.1% 7.9% 10.1% Upper limit sensitivity, base-case specificity ∗ 5.6% 6.0% 6.6% 7.2% 10.1% Lower limit sensitivity, base-case specificity ∗ 5.6% 7.0% 8.1% 8.9% 10.1% Upper limit sensitivity, upper limit specificity ∗ 5.6% 6.1% 6.7% 7.2% 10.1% Lower limit sensitivity, lower limit specificity ∗ 5.6% 6.9% 8.0% 8.8% 10.1%p trueT2 0.33 (0.5 x base-case estimate) 6.7% 7.8% 9.2% 10.4% 13.4% 0.98 (1.5 x base-case estimate) 4.6% 4.8% 5.2% 5.6% 6.8% Probability of nodal metastasis given T2 tumor at pathology 0.13 (0.5 × base-case estimate) 5.4% 6.0% 6.8% 7.4% 9.2% 0.38 (1.5 × base-case estimate) 5.9% 6.6% 7.7% 8.6% 11.1% Probability of nodal metastasis given T3 tumor at pathology 0.31 (0.5 × base-case estimate) 5.1% 5.9% 6.8% 7.4% 8.9% 0.93 (1.5 × base-case estimate) 6.2% 6.7% 7.6% 8.6% 11.3% Sensitivity and specificity (node-by-node) basis † 5.6% 7.0% 8.3% 9.1% 10.1% Reduction factor ‡ 0.23 (0.5 × base-case estimate) 3.7% 4.7% 6.0% 7.1% 10.1% 0.69 (1.5 × base-case estimate) 7.5% 7.9% 8.4% 8.9% 10.1%
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Sensitivity Analysis: Reported by Outcome of Interest
Minimizing acute chemoradiation toxicity
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Minimizing long-term chemoradiation toxicity
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Local recurrence rates achieved by treating preoperatively versus postoperatively
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Improved primary tumor staging and the effects on local recurrence
Get Radiology Tree app to read full this article<
Secondary Analysis: Morphologic Criteria for Positive Mesorectal Nodes
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Discussion
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Conclusions
Get Radiology Tree app to read full this article<
Appendix
MRI Uncertainty in Stage T2 Designation of the Primary Tumor
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
Table A1
Treatment Implications for Different Combinations of Imaging and Intraoperative Findings
Clinical Scenario (Tumor at Magnetic Resonance Imaging) Intraoperative Findings (Tumor at Pathology) Preoperative Treatment Given Postoperative Treatment Given T2N− T2N− No No T2N+ Yes (unnecessary) No T2N− T2N+ No Yes T2N+ Yes No T2N− T3N− No Yes T2N+ Yes No T2N− T3N+ No Yes T2N+ Yes No
Get Radiology Tree app to read full this article<
Get Radiology Tree app to read full this article<
References
1. American Cancer Society. Cancer facts and figures 2012. Available at: http://www.cancer.org/acs/groups/content/@epidemiologysurveilance/documents/document/acspc-031941.pdf . Accessed April 28, 2012.
2. Van Cutsem E., Dicato M., Haustermans K., et. al.: The diagnosis and management of rectal cancer: expert discussion and recommendations derived from the 9th World Congress on Gastrointestinal Cancer, Barcelona 2007. Ann Oncol 2008; 19: pp. vi1-vi8.
3. Sauer R., Becker H., Hohenberger W., et. al.: Preoperative versus postoperative chemoradiotherapy for rectal cancer. N Engl J Med 2004; 351: pp. 1731-1740.
4. Gunderson L.L., Haddock M.G., Schild S.E.: Rectal cancer: Preoperative versus postoperative irradiation as a component of adjuvant treatment. Semin Radiat Oncol 2003; 13: pp. 419-432.
5. Valentini V., Aristei C., Glimelius B., et. al.: Multidisciplinary rectal cancer management: 2nd European Rectal Cancer Consensus Conference (EURECA-CC2). Radiother Oncol 2009; 92: pp. 148-163.
6. Kim J.H., Beets G.L., Kim M.J., et. al.: High-resolution MR imaging for nodal staging in rectal cancer: are there any criteria in addition to the size?. Eur J Radiol 2004; 52: pp. 78-83.
7. Bartram C., Brown G.: Endorectal ultrasound and magnetic resonance imaging in rectal cancer staging. Gastroenterol Clin North Am 2002; 31: pp. 827-839.
8. Beets-Tan R.G., Beets G.L.: Rectal cancer: how accurate can imaging predict the T stage and the circumferential resection margin?. Int J Colorectal Dis 2003; 18: pp. 385-391.
9. Brown G., Richards C.J., Newcombe R.G., et. al.: Rectal carcinoma: thin-section MR imaging for staging in 28 patients. Radiology 1999; 211: pp. 215-222.
10. Klessen C., Rogalla P., Taupitz M.: Local staging of rectal cancer: the current role of MRI. Eur Radiol 2007; 17: pp. 379-389.
11. Tytherleigh M.G., Ng V.V., Pittathankal A.A., et. al.: Preoperative staging of rectal cancer by magnetic resonance imaging remains an imprecise tool. ANZ J Surg 2008; 78: pp. 194-198.
12. Skandarajah A.R., Tjandra J.J.: Preoperative loco-regional imaging in rectal cancer. ANZ J Surg 2006; 76: pp. 497-504.
13. Telford J.J., Saltzman J.R., Kuntz K.M., et. al.: Impact of preoperative staging and chemoradiation versus postoperative chemoradiation on outcome in patients with rectal cancer: a decision analysis. J Natl Cancer Inst 2004; 96: pp. 191-201.
14. Schnall M.D., Furth E.E., Rosato E.F., et. al.: Rectal tumor stage: correlation of endorectal MR imaging and pathologic findings. Radiology 1994; 190: pp. 709-714.
15. Brown G., Radcliffe A.G., Newcombe R.G., et. al.: Preoperative assessment of prognostic factors in rectal cancer using high-resolution magnetic resonance imaging. Br J Surg 2003; 90: pp. 355-364.
16. Brown G., Richards C.J., Bourne M.W., 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: pp. 371-377.
17. Kaur H., Choi H., You Y.N., et. al.: MR imaging for preoperative evaluation of primary rectal cancer: practical considerations. Radiographics 2012; 32: pp. 389-409.
18. Akasu T., Iinuma G., Takawa M., et. al.: Accuracy of high-resolution magnetic resonance imaging in preoperative staging of rectal cancer. Ann Surg Oncol 2009; 16: pp. 2787-2794.
19. Blomqvist L., Machado M., Rubio C., et. al.: Rectal tumour staging: MR imaging using pelvic phased-array and endorectal coils vs endoscopic ultrasonography. Eur Radiol 2000; 10: pp. 653-660.
20. Branagan G., Chave H., Fuller C., et. al.: Can magnetic resonance imaging predict circumferential margins and TNM stage in rectal cancer?. Dis Colon Rectum 2004; 47: pp. 1317-1322.
21. Donmez F.Y., Tunaci M., Yekeler E., et. al.: Effect of using endorectal coil in preoperative staging of rectal carcinomas by pelvic MR imaging. Eur J Radiol 2008; 67: pp. 139-145.
22. Kim N.K., Kim M.J., Park J.K., et. al.: Preoperative staging of rectal cancer with MRI: accuracy and clinical usefulness. Ann Surg Oncol 2000; 7: pp. 732-737.
23. Poon F.W., McDonald A., Anderson J.H., et. al.: Accuracy of thin section magnetic resonance using phased-array pelvic coil in predicting the T-staging of rectal cancer. Eur J Radiol 2005; 53: pp. 256-262.
24. Petersen S., Hellmich G., von Mildenstein K., et. al.: Is surgery-only the adequate treatment approach for T2N0 rectal cancer?. J Surg Oncol 2006; 93: pp. 350-354.
25. Stocchi L., Nelson H., Sargent D.J., et. al.: Impact of surgical and pathologic variables in rectal cancer: a United States community and cooperative group report. J Clin Oncol 2001; 19: pp. 3895-3902.
26. Park J.H., Yoon S.M., Yu C.S., et. al.: Randomized phase 3 trial comparing preoperative and postoperative chemoradiotherapy with capecitabine for locally advanced rectal cancer. Cancer 2011; 117: pp. 3703-3712.
27. Roh M.S., Colangelo L.H., O’Connell M.J., et. al.: Preoperative multimodality therapy improves disease-free survival in patients with carcinoma of the rectum: NSABP R-03. J Clin Oncol 2009; 27: pp. 5124-5130.
28. Wibe A., Rendedal P.R., Svensson E., et. al.: Prognostic significance of the circumferential resection margin following total mesorectal excision for rectal cancer. Br J Surg 2002; 89: pp. 327-334.
29. Hermanek P., Hohenberger W., Fietkau R., et. al.: Individualized magnetic resonance imaging-based neoadjuvant chemoradiation for middle and lower rectal carcinoma. Colorectal Dis 2011; 13: pp. 39-47.
30. Frasson M., Garcia-Granero E., Roda D., et. al.: Preoperative chemoradiation may not always be needed for patients with T3 and T2N+ rectal cancer. Cancer 2011; 117: pp. 3118-3125.
31. Chang KH, Smith MJ, McAnena OJ, et al. Increased use of multidisciplinary treatment modalities adds little to the outcome of rectal cancer treated by optimal total mesorectal excision. Int J Colorectal Dis. In press.
32. Kusters M., Marijnen C.A., van de Velde C.J., et. al.: Patterns of local recurrence in rectal cancer; a study of the Dutch TME trial. Eur J Surg Oncol 2010; 36: pp. 470-476.
33. Roels S., Duthoy W., Haustermans K., et. al.: Definition and delineation of the clinical target volume for rectal cancer. Int J Radiat Oncol Biol Phys 2006; 65: pp. 1129-1142.
34. Kozak K.R., Moody J.S.: The impact of T and N stage on long-term survival of rectal cancer patients in the community. J Surg Oncol 2008; 98: pp. 161-166.