Purpose. Comparative diagnostic accuracy (CDA) studies are typically small retrospective studies supporting a higher accuracy for one modality over another for either staging a particular disease or assessing response to therapy, and they are used to generate hypotheses for larger prospective trials. The purpose of this article is to introduce the group sequential design (GSD) approach in planning these larger trials. Methods. Methodology needed for using GSD in the CDA studies is recently developed. In this article, GSD with the O’Brien and Fleming (OBF) stopping rule is described and guidelines for sample size calculation are provided. Simulated data is used to demonstrate the application of GSD in the design/analysis of a clinical trial in theCDAstudy setting. Results. The expected sample size needed for planning a trial with GSD (under the OBF stopping rule) is slightly inflated but may ultimately result in greater savings of patient resources. Conclusion. GSD is a specialized statistical method that is helpful in balancing the ethical and financial advantages of stopping a study early against the risk of an incorrect conclusion and should be adopted for planning CDAstudies.
1. Schiepers C, Hoh CK. Positron emission tomography as a diagnostic tool in oncology. Eur Radiol. 1998;8(8):1481–1494.
2.
2. Pepe MS. The Statistical Evaluation ofMedical Tests for Classification and Prediction. Oxford (UK): Oxford Statistical Science Series; 2004.
3.
3. Begg CB. Advances in statistical methodology for diagnostic medicine in the 1980s. Stat Med. 1991;10(12):1887–1895.
4.
4. Schirrmeister H, Guhlmann A, Elsner K, et al. Sensitivity in detecting osseous lesions depends on anatomic localization: planar bone scintigraphy versus 18FPET. JNuclMed. 1999;40(10):1623–1629.
5.
5. Ichikawa T, SouH, Araki T, et al. Duct-penetrating sign atMRCP: usefulness for differentiating inflammatory pancreatic mass from pancreatic carcinomas. Radiology. 2001;221(1):107–116.
6.
6. Enya M, Goto H, Nandate Y, et al. Usefulness of dynamic MR mammography for diagnosis of axillary lymph node status in breast cancer patient. Nippon Igaku Hoshasen Gakkai Zasshi. 2000;60(13): 764–770.
7.
7. Butzen J, Prost R, Chetty V, et al. Discrimination between neoplastic and nonneoplastic brain lesions by use of protonMRspectroscopy: the limits of accuracy with a logistic regression model. Am J Neuroradiol. 2000;21(7):1213–1219.
8.
8. Reimer P, JahnkeN, FiebichM, et al. Hepatic lesion detection and characterization: value of nonenhanced MR imaging, superparamagnetic iron oxide-enhancedMR imaging, and spiral CT-ROC analysis. Radiology. 2000;217(1):152–158.
9.
9. YangWT, LamWW, YuMY, et al. Comparison of dynamic helical CTand dynamicMRimaging in the evaluation of pelvic lymphnodes in cervical carcinoma. Am J Roentgenol. 2000;175(3):759–766.
10.
10. Powell KA, Mallasch PG, Obuchowski NA, et al. Clinical evaluation of wavelet-compressed digitized screen-film mammography. Acad Radiol. 2000;7(5):311–316.
11.
11. Strotzer M, Volk M, Wild T, et al. Simulated bone erosions in a hand phantom: detection with conventional screen-film technology versus cesium iodide-amorphous silicon flat-panel detector. Radiology. 2000;215(2):512–515.
12.
12. Vansteenkiste JF, Stroobants SG, De Leyn PR, et al. Lymph node staging in non-small-cell lung cancer with FDG-PET scan: a prospective study on 690 lymph node stations from 68 patients. J Clin Oncol. 1998;16(6):2142–2149.
13.
13. Spaepen K, Stroobants S, Dupont P, et al. Prognostic value of positron emission tomography (PET) with [18F]FDG after first-line chemotherapy in non-Hodgkin’s lymphoma: is PET a valid alternative to conventional diagnostic methods?J Clin Oncol. 2001;19(2):414–419.
14.
14. Kao CH, ChangLai SP, Chieng PU, et al. Detection of recurrent or persistent nasopharyngeal carcinomas after radiotherapy with 18 fluoro-2-deoxyglucose positron emission tomography and comparison with computed tomography. J Clin Oncol. 1998;16(11):3550–3555.
15.
15. Geller NL. Planned interim analysis and its role in cancer clinical trials. J Clin Oncol. 1987;5:1485–1490.
16.
16. Fleming TR, Watelet LF. Approaches to monitoring clinical trials. J Natl Cancer Inst. 1989;81:188–193.
17.
17. National Cancer Institute. Policy of the national cancer institute for data and safety monitoring of clinical trials. Available from: URL: http://cancertrials.nci.nih.gov
18.
18. Kelly K, Crowley J, Bunn PA, et al. Randomized phase III trial of paclitaxel plus carboplatin versus vinorelbine plus cisplatin in the treatment of patients with advanced non-small-cell lung cancer: a Southwest Oncology Group trial. J Clin Oncol. 2001;19:3210–3218.
19.
19. Sacco RL, DeRosa JT, Haley, et al. Glycine antagonist in neuroprotection for patients with acute stroke: GAIN Americas—a randomized controlled trial. JAMA. 2001;285(13):1719–1728.
20.
20. Gausche M, Lewis RJ, Stratton SJ, et al. Effect of out-of-hospital pediatric endotracheal intubation on survival and neurological outcome. JAMA. 2000;283(6):783–790.
21.
21. Frustaci S, Gherlinzoni F, De Paoli A, et al. Adjuvant chemotherapy for adult soft tissue sarcomas of the extremities and girdles: results of the Italian Randomized Cooperative Trial. J Clin Oncol. 2001;19:1238–1247.
22.
22. Downs JR, Clearfield M, Weis S, et al. Primary prevention of acute coronary events with lovastatin in men and women with average cholesterol levels: results of AFCAPS/TexCAPS. JAMA. 1998; 279(20):1615–1622.
23.
23. Hanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristic curve. Radiology. 1982;143:29–36.
24.
24. Hanley JA, McNeil BJ. Amethod of comparing the areas under receiver operating characteristic curves derived from the same cases. Radiology. 1983;148:839–843.
25.
25. Metz CE. Codes for ROC analysis fromtheUniversity of Chicago. Available from: URL: http://www-radiology.uchicago.edu/krl/toppage11.htm
26.
26. Obuchowski NA, McClish DK. Sample size determination for diagnostic accuracy studies involving binormal ROC curve indices. Stat Med. 1997;16:1529–1542.
27.
27. Zhou XH, Obuchowski NA, McClish DK. Statistical methods in diagnostic medicine. New York: Wiley-Interscience; 2002.
28.
28. Jennison C, Turnbull BW. Group sequentialmethods with applications to clinical trials. Boca Raton (FL): Chapman & Hall; 2000
29.
29. Green SB, Byar DP. Using observational data from registries to compare treatments: the fallacy of omnimetrics. Stat Med. 1984; 3:361–370.
30.
30. Begg CB. Biases in the assessment of diagnostic tests. Stat Med. 1987;6:411–423.
31.
31. Mazumdar M, Begg CB. Biases in the studies evaluating performance of ELISA kits for the detection of HIV antibodies. Proceedings of Centers for Disease Control Symposium, 1993.
32.
32. Fleming TR, Green SJ, Harrington DP. Considerations for monitoring and evaluating treatment effects in clinical trials. Control Clin Trials. 1984;5:55–66.
33.
33. O’Brien PC, Fleming TR. A multiple testing procedure for clinical trials. Biometrics. 1979;35:549–556.
34.
34. Pocock SJ. Group sequential methods in the design and analysis of clinical trials. Biometrika. 1977;64:191–192.
35.
35. Mazumdar M, Liu A.Group sequential design for diagnostic accuracy studies. Stat Med. 2003;22:727–739.
36.
36. East. UserManual: Software for the Design and InterimMonitoring of Group Sequential Trials. Cambridge (MA): Cytel Software Corporation; 2000.
37.
37. Dorfman RE, Alpern MB, Gross BH, et al. Upper abdominal lymph nodes: criteria for normal size detected with CT. Radiology. 1991;180:319–322.
38.
38. Ramos CD, Erdi YE, Gonen M, et al. FDG-PET standardized uptake values in normal anatomical structures using iterative reconstruction segmented attenuation correction and filtered back projection. Eur J Nucl Med. 2001;28:155–164.
39.
39. Cui L, Hung HMJ, Wang S.Modification of sample size in group sequential trials. Biometrics. 1999;55:853–857.
40.
40. Lan KKG, Demets DL. Discrete sequential boundaries for clinical trials. Biometrika. 1983;70:659–663.
41.
41. Lan KKG, Rosenberger WF, Lachin JM. Use of spending functions for occasional or continuous monitoring of data in clinical trials. Stat Med. 1993;12:2219–2231.
