File Download

There are no files associated with this item.

  Links for fulltext
     (May Require Subscription)
Supplementary

Article: Robust EM continual reassessment method in oncology dose finding

TitleRobust EM continual reassessment method in oncology dose finding
Authors
KeywordsAdaptive design
Expectation-maximization algorithm
Late-onset toxicity
Maximum tolerated dose
Missing data
Model averaging
Model selection
Issue Date2011
PublisherAmerican Statistical Association. The Journal's web site is located at http://www.amstat.org/publications/jasa/index.cfm?fuseaction=main
Citation
Journal Of The American Statistical Association, 2011, v. 106 n. 495, p. 818-831 How to Cite?
AbstractThe continual reassessment method (CRM) is a commonly used dose-finding design for phase I clinical trials. Practical applications of this method have been restricted by two limitations: (1) the requirement that the toxicity outcome needs to be observed shortly after the initiation of the treatment; and (2) the potential sensitivity to the prespecified toxicity probability at each dose. To overcome these limitations, we naturally treat the unobserved toxicity outcomes as missing data, and use the expectation-maximization (EM) algorithm to estimate the dose toxicity probabilities based on the incomplete data to direct dose assignment. To enhance the robustness of the design, we propose prespecifying multiple sets of toxicity probabilities, each set corresponding to an individual CRM model. We carry out these multiple CRMs in parallel, across which model selection and model averaging procedures are used to make more robust inference. We evaluate the operating characteristics of the proposed robust EM-CRM designs through simulation studies and show that the proposed methods satisfactorily resolve both limitations of the CRM. Besides improving the MTD selection percentage, the new designs dramatically shorten the duration of the trial, and are robust to the prespecification of the toxicity probabilities. © 2011 American Statistical Association.
Persistent Identifierhttp://hdl.handle.net/10722/139716
ISSN
2015 Impact Factor: 1.725
2015 SCImago Journal Rankings: 3.447
ISI Accession Number ID
Funding AgencyGrant Number
NCI, United StatesR01CA154591-01A1
Research Grants Council of Hong Kong
Funding Information:

The authors thank Dr. Xiudong Lei at the University of Texas MD Anderson Cancer Center for her help in the simulation studies. We gratefully acknowledge the editor, the associate editor, and two anonymous referees for their insightful and constructive comments which substantially improved the article. The research is partially supported by NCI grant R01CA154591-01A1, United States, and a grant from the Research Grants Council of Hong Kong.

References

 

DC FieldValueLanguage
dc.contributor.authorYuan, Yen_HK
dc.contributor.authorYin, Gen_HK
dc.date.accessioned2011-09-23T05:54:45Z-
dc.date.available2011-09-23T05:54:45Z-
dc.date.issued2011en_HK
dc.identifier.citationJournal Of The American Statistical Association, 2011, v. 106 n. 495, p. 818-831en_HK
dc.identifier.issn0162-1459en_HK
dc.identifier.urihttp://hdl.handle.net/10722/139716-
dc.description.abstractThe continual reassessment method (CRM) is a commonly used dose-finding design for phase I clinical trials. Practical applications of this method have been restricted by two limitations: (1) the requirement that the toxicity outcome needs to be observed shortly after the initiation of the treatment; and (2) the potential sensitivity to the prespecified toxicity probability at each dose. To overcome these limitations, we naturally treat the unobserved toxicity outcomes as missing data, and use the expectation-maximization (EM) algorithm to estimate the dose toxicity probabilities based on the incomplete data to direct dose assignment. To enhance the robustness of the design, we propose prespecifying multiple sets of toxicity probabilities, each set corresponding to an individual CRM model. We carry out these multiple CRMs in parallel, across which model selection and model averaging procedures are used to make more robust inference. We evaluate the operating characteristics of the proposed robust EM-CRM designs through simulation studies and show that the proposed methods satisfactorily resolve both limitations of the CRM. Besides improving the MTD selection percentage, the new designs dramatically shorten the duration of the trial, and are robust to the prespecification of the toxicity probabilities. © 2011 American Statistical Association.en_HK
dc.languageengen_US
dc.publisherAmerican Statistical Association. The Journal's web site is located at http://www.amstat.org/publications/jasa/index.cfm?fuseaction=mainen_HK
dc.relation.ispartofJournal of the American Statistical Associationen_HK
dc.subjectAdaptive designen_HK
dc.subjectExpectation-maximization algorithmen_HK
dc.subjectLate-onset toxicityen_HK
dc.subjectMaximum tolerated doseen_HK
dc.subjectMissing dataen_HK
dc.subjectModel averagingen_HK
dc.subjectModel selectionen_HK
dc.titleRobust EM continual reassessment method in oncology dose findingen_HK
dc.typeArticleen_HK
dc.identifier.emailYin, G: gyin@hku.hken_HK
dc.identifier.authorityYin, G=rp00831en_HK
dc.description.naturelink_to_subscribed_fulltext-
dc.identifier.doi10.1198/jasa.2011.ap09476en_HK
dc.identifier.scopuseid_2-s2.0-80054707561en_HK
dc.identifier.hkuros195637en_US
dc.relation.referenceshttp://www.scopus.com/mlt/select.url?eid=2-s2.0-80054707561&selection=ref&src=s&origin=recordpageen_HK
dc.identifier.volume106en_HK
dc.identifier.issue495en_HK
dc.identifier.spage818en_HK
dc.identifier.epage831en_HK
dc.identifier.eissn1537-274X-
dc.identifier.isiWOS:000296224200008-
dc.publisher.placeUnited Statesen_HK
dc.identifier.scopusauthoridYuan, Y=7402709174en_HK
dc.identifier.scopusauthoridYin, G=8725807500en_HK

Export via OAI-PMH Interface in XML Formats


OR


Export to Other Non-XML Formats