Statistical recommendations for count, binary, and ordinal data in rare disease cross-over trials.
Cross-over
Epidermolysis bullosa simplex
GEE-like semiparametric model
Generalized pairwise comparison (GPC)
Guidance
Model averaging
NparLD
Rare Diseases
Recommendation
Repeated measures
Small sample size
Journal
Orphanet journal of rare diseases
ISSN: 1750-1172
Titre abrégé: Orphanet J Rare Dis
Pays: England
ID NLM: 101266602
Informations de publication
Date de publication:
19 Dec 2023
19 Dec 2023
Historique:
received:
07
08
2023
accepted:
19
11
2023
medline:
20
12
2023
pubmed:
20
12
2023
entrez:
20
12
2023
Statut:
epublish
Résumé
Recommendations for statistical methods in rare disease trials are scarce, especially for cross-over designs. As a result various state-of-the-art methodologies were compared as neutrally as possible using an illustrative data set from epidermolysis bullosa research to build recommendations for count, binary, and ordinal outcome variables. For this purpose, parametric (model averaging), semiparametric (generalized estimating equations type [GEE-like]) and nonparametric (generalized pairwise comparisons [GPC] and a marginal model implemented in the R package nparLD) methods were chosen by an international consortium of statisticians. It was found that there is no uniformly best method for the aforementioned types of outcome variables, but in particular situations, there are methods that perform better than others. Especially if maximizing power is the primary goal, the prioritized unmatched GPC method was able to achieve particularly good results, besides being appropriate for prioritizing clinically relevant time points. Model averaging led to favorable results in some scenarios especially within the binary outcome setting and, like the GEE-like semiparametric method, also allows for considering period and carry-over effects properly. Inference based on the nonparametric marginal model was able to achieve high power, especially in the ordinal outcome scenario, despite small sample sizes due to separate testing of treatment periods, and is suitable when longitudinal and interaction effects have to be considered. Overall, a balance has to be found between achieving high power, accounting for cross-over, period, or carry-over effects, and prioritizing clinically relevant time points.
Sections du résumé
BACKGROUND
BACKGROUND
Recommendations for statistical methods in rare disease trials are scarce, especially for cross-over designs. As a result various state-of-the-art methodologies were compared as neutrally as possible using an illustrative data set from epidermolysis bullosa research to build recommendations for count, binary, and ordinal outcome variables. For this purpose, parametric (model averaging), semiparametric (generalized estimating equations type [GEE-like]) and nonparametric (generalized pairwise comparisons [GPC] and a marginal model implemented in the R package nparLD) methods were chosen by an international consortium of statisticians.
RESULTS
RESULTS
It was found that there is no uniformly best method for the aforementioned types of outcome variables, but in particular situations, there are methods that perform better than others. Especially if maximizing power is the primary goal, the prioritized unmatched GPC method was able to achieve particularly good results, besides being appropriate for prioritizing clinically relevant time points. Model averaging led to favorable results in some scenarios especially within the binary outcome setting and, like the GEE-like semiparametric method, also allows for considering period and carry-over effects properly. Inference based on the nonparametric marginal model was able to achieve high power, especially in the ordinal outcome scenario, despite small sample sizes due to separate testing of treatment periods, and is suitable when longitudinal and interaction effects have to be considered.
CONCLUSION
CONCLUSIONS
Overall, a balance has to be found between achieving high power, accounting for cross-over, period, or carry-over effects, and prioritizing clinically relevant time points.
Identifiants
pubmed: 38115074
doi: 10.1186/s13023-023-02990-1
pii: 10.1186/s13023-023-02990-1
doi:
Types de publication
Journal Article
Langues
eng
Sous-ensembles de citation
IM
Pagination
391Subventions
Organisme : H2020 Societal Challenges
ID : 825575
Informations de copyright
© 2023. The Author(s).
Références
Aoki Y, Röshammar D, Hamrén B, Hooker AC. Model selection and averaging of nonlinear mixed-effect models for robust phase III dose selection. J Pharmacokinet Pharmacodyn. 2017;44(6):581–97. https://doi.org/10.1007/s10928-017-9550-0 .
doi: 10.1007/s10928-017-9550-0
pubmed: 29103208
pmcid: 5686275
Ast (n.d.) ASTERIX: advances in small trials design for regulatory innovation and excellence. FP7-HEALTH-2013-INNOVATION-1 Grant-Agreement No. 603160. https://www.asterix-fp7.eu/
Bartlett MS. Properties of sufficiency and statistical tests. Proc R Soc Lond Ser A Math Phys Sci. 1937;160(901):268–82.
Brunner E, Bathke AC, Konietschke F. Rank and pseudo-rank procedures for independent observations in factorial designs, using R and SAS. Berlin Heidelberg: Springer; 2019.
Brunner E, Munzel U. Nichtparametrische Datenanalyse: Unverbundene Stichproben. Berlin, Heidelberg: Springer; 2013.
doi: 10.1007/978-3-642-37184-4
Buyse M. Generalized pairwise comparisons of prioritized outcomes in the two-sample problem. Stat Med. 2010;29:3245–57.
doi: 10.1002/sim.3923
pubmed: 21170918
Coakley CW, Heise MA. Versions of the sign test in the presence of ties. Biometrics. 1996;52:1242–51.
doi: 10.2307/2532840
der Elst WV, Alonso AA, Geys H, Meyvisch P, Bijnens L, Sengupta R, Molenberghs G. Univariate versus multivariate surrogates in the single-trial setting. Stat Biopharm Res. 2019;11(3):301–10.
doi: 10.1080/19466315.2019.1575276
Dixon W, Massey F. An introduction to statistical analysis. New York: McGraw-Hill; 1951.
European-Commission Collaboration: a key to unlock the challenges of rare diseases research, Publications Office of the European Union (2021)
Fagerland M, Lydersen S, Laake P. The McNemar test for binary matched-pairs data: mid-p and asymptotic are better than exact conditional. BMC Med Res Methodol. 2013;13:91.
doi: 10.1186/1471-2288-13-91
pubmed: 23848987
pmcid: 3716987
Fagerland MW. T-tests, non-parametric tests, and large studies—a paradox of statistical practice? BMC Med Res Methodol. 2012;12:78.
doi: 10.1186/1471-2288-12-78
pubmed: 22697476
pmcid: 3445820
Finkelstein D, Schoenfeld D. Combining mortality and longitudinal measures in clinical trials. Stat Med. 1999;18:1341–54.
doi: 10.1002/(SICI)1097-0258(19990615)18:11<1341::AID-SIM129>3.0.CO;2-7
pubmed: 10399200
Geroldinger M, Verbeeck J, Thiel KE, Molenberghs G, Bathke AC, Laimer M, Zimmermann G. A neutral comparison of statistical methods for analyzing longitudinally measured ordinal outcomes in rare diseases. Biom J. 2023. https://doi.org/10.1002/bimj.202200236 .
doi: 10.1002/bimj.202200236
pubmed: 36890631
Graf AC, Posch M, Koenig F. Adaptive designs for subpopulation analysis optimizing utility functions. Biom J. 2015;57(1):76–89.
doi: 10.1002/bimj.201300257
pubmed: 25399844
Heller GZ, Manuguerra M, Chow R. How to analyze the visual analogue scale: myths, truths and clinical relevance. Scand J Pain. 2016;13(1):67–75.
doi: 10.1016/j.sjpain.2016.06.012
pubmed: 28850536
Hilgers R-D, Bogdan M, Burman C-F, Dette H, Karlsson M, König F, Male C, Mentré F, Molenberghs G, Senn S. Lessons learned from ideal—33 recommendations from the ideal-net about design and analysis of small population clinical trials. Orphanet J Rare Dis. 2018;13(1):1–17.
doi: 10.1186/s13023-018-0820-8
Hurvich C, Tsai C. Regression and time series model selection in small samples. Biometrika. 1989;76(2):297–307.
doi: 10.1093/biomet/76.2.297
Ide (n.d.) IDEAL: integrated design and analysis of small population group trials. EU grant agreement no. 602552. https://www.ideal.rwth-aachen.de/
Ins (n.d.) INSPIRE: innovative methodology for small populations research. EU grant agreement no. 602144. https://warwick.ac.uk/fac/sci/med/research/hscience/stats/completedprojects/inspire/
Jones B, Kenward M. Design and analysis of cross-over trials. New York: Chapman and Hall/CRC; 2014.
doi: 10.1201/b17537
Karlsson K, Vong C, Bergstrand M, Jonsson E, Karlsson M. Comparisons of analysis methods for proof-of-concept trials. CPT Pharmacomet Syst Pharmacol. 2013;2(1):23. https://doi.org/10.1038/psp.2012.24 .
doi: 10.1038/psp.2012.24
le Cessie S, van Houwelingen J. Logistic regression for correlated binary data. Appl Stat. 1994;43:95–108.
doi: 10.2307/2986114
Li P, Redden D. Small sample performance of bias-corrected sandwich estimators for cluster-randomized trials with binary outcomes. Stat Med. 2015;34(2):281–96.
doi: 10.1002/sim.6344
pubmed: 25345738
Mann H, Whitney D. On a test of whether one of two random variables is stochastically larger than the other. Ann Math Stat. 1947;18(1):50–60.
doi: 10.1214/aoms/1177730491
Molenberghs G, Verbeke G. Models for discrete longitudinal data. New York: Springer; 2005.
Nguengang Wakap S, Lambert DM, Olry A, Rodwell C, Gueydan C, Lanneau V, Murphy D, Le Cam Y, Rath A. Estimating cumulative point prevalence of rare diseases: analysis of the orphanet database. Eur J Hum Genet. 2020;28(2):165–73.
doi: 10.1038/s41431-019-0508-0
pubmed: 31527858
Noguchi K, Gel YR, Brunner E, Konietschke F. nparLD: an R software package for the nonparametric analysis of longitudinal data in factorial experiments. J Stat Softw. 2012;50(12):1–23.
doi: 10.18637/jss.v050.i12
Pocock S, Ariti C, Collier T, Wang D. The win ratio: a new approach to the analysis of composite endpoints in clinical trials based on clinical priorities. Eur Heart J. 2012;33:176–82.
doi: 10.1093/eurheartj/ehr352
pubmed: 21900289
Prodinger C, Diem A, Ude-Schoder K, Pinon-Hofbauer J, Kitzmueller S, Bauer J, Laimer M. Profiling trial burden and patients attitudes to improve clinical research in epidermolysis bullosa. Orphanet J Rare Dis. 2020;15:182.
doi: 10.1186/s13023-020-01443-3
pubmed: 32650809
pmcid: 7350741
Strömberg E, Hooker A. The effect of using a robust optimality criterion in model based adaptive optimization. J Pharmacokinet Pharmacodyn. 2017;44:317–24.
doi: 10.1007/s10928-017-9521-5
pubmed: 28386710
pmcid: 5514236
Verbeeck J, Deltuvaite-Thomas V, Berckmoes B, Burzykowski T, Aerts M, Thas O, Buyse M, Molenberghs G. Unbiasedness and efficiency of non-parametric and UMVUE estimators of the probabilistic index and related statistics. Stat Methods Med Res. 2021;30(3):747–68.
doi: 10.1177/0962280220966629
pubmed: 33256560
Verbeeck J, Geroldinger M, Thiel KE, Molenberghs G, Bathke AC, Johann B, Zimmermann G. How to analyze continuous and discrete repeated measures in small sample cross-over trials? Biometrics. 2023. https://doi.org/10.1111/biom.13920 .
doi: 10.1111/biom.13920
pubmed: 37587671
Verbeeck J, Spitzer E, de Vries T, van Es G, Anderson W, Van Mieghem N, Leon M, Molenberghs G, Tijssen J. Generalized pairwise comparison methods to analyze (non)prioritized composite endpoints. Stat Med. 2019;38(30):5641–56.
doi: 10.1002/sim.8388
pubmed: 31659790
Wally V, Hovnanian A, Ly J, Buckova H, Brunner V, Lettner T, Ablinger M, Felder T, Hofbauer P, Wolkersdorfer M, Lagler F, Hitzl W, Laimer M, Kitzmüller S, Diem A, Bauer J. Diacerein orphan drug development for epidermolysis bullosa simplex: a phase 2/3 randomized, placebo-controlled, double-blind clinical trial. J Am Acad Dermatol. 2018;78(5):892–901.
doi: 10.1016/j.jaad.2018.01.019
pubmed: 29410318
Wittkowski K. Versions of the sign test in the presence of ties. Biometrics. 1998;54(2):789–91.
doi: 10.2307/3109786