{"id":{"repo_id":"edinburgh","oai_identifier":"oai:era.ed.ac.uk:1842/43843"},"canonical_url":"https://search.dev.ndltd.org/etd/edinburgh/oai:era.ed.ac.uk:1842/43843","repository":{"repo_id":"edinburgh","name":"University of Edinburgh","base_url":"https://era.ed.ac.uk/server/oai/request"},"display":{"title":"Search for valid surrogate endpoints: review of current evidence and development of practical strategies for surrogate evaluation","abstract":"In light of the ongoing global pursuit of more quickly available treatments, surrogate endpoint evaluation has become a vital part of drug development. Despite the increasing uptake of surrogate endpoints and the large number of statistical methods proposed to evaluate a surrogate, there is still no consensus on how to determine when a surrogate endpoint is acceptable. Therefore, this work aimed to address three questions. PROJECT 1: How well do recently proposed statistical methods perform? PROJECT 2: What is the validation status of surrogate endpoints currently in use? PROJECT 3: Can a new framework for surrogate evaluation be developed by way of consensus? PROJECT 1. A scoping review was conducted to identify recent, novel statistical methods of surrogate evaluation. The method with the greatest research activity was the proportion of treatment effect explained (PTE). The performances of two fundamental methods based on the PTE were compared through simulations. Neither method was considered to be sufficient on its own. PROJECT 2. A systematic review was conducted to identify both validated and non-validated surrogate endpoints in use, and the statistical methods applied to evaluate the validated surrogate endpoints. Based on the combination of three surrogate evaluation frameworks, many surrogate endpoints accepted by regulatory agencies were not statistically valid. Correlation-based methods were the most frequently applied statistical approaches. PROJECT 3. A new framework was developed from a two-round Delphi survey followed by an international, multi-stakeholder consensus meeting. The survey included questions on the prioritisation of non-statistical factors in, use of numerical thresholds for and preferred statistical methods of surrogate evaluation. Four non-statistical factors reached consensus by the end of the second round. However, opinions were divided on the issues of thresholds and methods. A consensus meeting was subsequently held to ratify the factors, thresholds and methods to be included in a framework. Five non-statistical factors were deemed necessary, with two non-statistical factors for possible consideration, and correlation thresholds for possible consideration. In addition, the framework deemed trial-level analysis as necessary, and provided general guidance on the choice of statistical methods, rather than recommending specific methods. As found in all three projects, surrogate endpoint evaluation remains a challenging, multi-stakeholder task. Nonetheless, the newly-developed framework has partly resolved the longstanding lack of agreement on which factors to consider when evaluating a surrogate endpoint, and may catalyse further improvements to come in this cornerstone of drug development.","abstract_html":"In light of the ongoing global pursuit of more quickly available treatments, surrogate endpoint evaluation has become a vital part of drug development. Despite the increasing uptake of surrogate endpoints and the large number of statistical methods proposed to evaluate a surrogate, there is still no consensus on how to determine when a surrogate endpoint is acceptable. Therefore, this work aimed to address three questions. PROJECT 1: How well do recently proposed statistical methods perform? PROJECT 2: What is the validation status of surrogate endpoints currently in use? PROJECT 3: Can a new framework for surrogate evaluation be developed by way of consensus? PROJECT 1. A scoping review was conducted to identify recent, novel statistical methods of surrogate evaluation. The method with the greatest research activity was the proportion of treatment effect explained (PTE). The performances of two fundamental methods based on the PTE were compared through simulations. Neither method was considered to be sufficient on its own. PROJECT 2. A systematic review was conducted to identify both validated and non-validated surrogate endpoints in use, and the statistical methods applied to evaluate the validated surrogate endpoints. Based on the combination of three surrogate evaluation frameworks, many surrogate endpoints accepted by regulatory agencies were not statistically valid. Correlation-based methods were the most frequently applied statistical approaches. PROJECT 3. A new framework was developed from a two-round Delphi survey followed by an international, multi-stakeholder consensus meeting. The survey included questions on the prioritisation of non-statistical factors in, use of numerical thresholds for and preferred statistical methods of surrogate evaluation. Four non-statistical factors reached consensus by the end of the second round. However, opinions were divided on the issues of thresholds and methods. A consensus meeting was subsequently held to ratify the factors, thresholds and methods to be included in a framework. Five non-statistical factors were deemed necessary, with two non-statistical factors for possible consideration, and correlation thresholds for possible consideration. In addition, the framework deemed trial-level analysis as necessary, and provided general guidance on the choice of statistical methods, rather than recommending specific methods. As found in all three projects, surrogate endpoint evaluation remains a challenging, multi-stakeholder task. Nonetheless, the newly-developed framework has partly resolved the longstanding lack of agreement on which factors to consider when evaluating a surrogate endpoint, and may catalyse further improvements to come in this cornerstone of drug development.","abstract_has_math":false,"creators":["Lo, Wang Pok"],"institution":"The University of Edinburgh","degree_name":null,"degree_level":null,"degree_discipline":null,"degree_department":null,"school":null,"contributors":[],"advisors":["Weir, Christopher","Ensor, Hannah","Lewis, Steff"],"committee_chairs":[],"committee_members":[],"year":2025,"date_issued":"2025-08-21","date_published":"2025-08-21","updated_at":"2026-07-24T02:14:07Z","subjects":["surrogate endpoints","surrogate evaluation","clinical trials","statistical methodology","expert opinion","Delphi method"],"languages":["en"],"rights":[],"rights_urls":["https://creativecommons.org/licenses/by/4.0/"],"identifier_entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["http://dx.doi.org/10.7488/era/6374"],"render_values":[{"text":"http://dx.doi.org/10.7488/era/6374","href":"http://dx.doi.org/10.7488/era/6374","code":true}]}]},"links":{"outbound_url":"https://hdl.handle.net/1842/43843","outbound_label":"Handle","outbound_source":"dc:identifier.uri"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor.advisor","label":"Advisor","values":["Weir, Christopher","Ensor, Hannah","Lewis, Steff"]},{"key":"dc:creator","label":"Author","values":["Lo, Wang Pok"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:date.accessioned","label":"Dc Date Accessioned","values":["2025-08-21T14:38:49Z"]},{"key":"dc:date.available","label":"Dc Date Available","values":["2025-08-21T14:38:49Z"]},{"key":"dc:date.issued","label":"Date","values":["2025-08-21"]},{"key":"dc:publisher","label":"Institution","values":["The University of Edinburgh"]},{"key":"dc:type","label":"Dc Type","values":["Thesis or Dissertation"]},{"key":"dc:type.qualificationlevel","label":"Dc Type Qualificationlevel","values":["Doctoral"]},{"key":"dc:type.qualificationname","label":"Dc Type Qualificationname","values":["PhD Doctor of Philosophy"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["surrogate endpoints","surrogate evaluation","clinical trials","statistical methodology","expert opinion","Delphi method"]}]},{"id":"language_rights","label":"Language and Rights","entries":[{"key":"dc:language.iso","label":"Language (ISO)","values":["en"]},{"key":"dc:rights.uri","label":"Rights URI","values":["https://creativecommons.org/licenses/by/4.0/"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier.uri","label":"Identifier URI","values":["https://hdl.handle.net/1842/43843","http://dx.doi.org/10.7488/era/6374"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["In light of the ongoing global pursuit of more quickly available treatments, surrogate endpoint evaluation has become a vital part of drug development. Despite the increasing uptake of surrogate endpoints and the large number of statistical methods proposed to evaluate a surrogate, there is still no consensus on how to determine when a surrogate endpoint is acceptable. Therefore, this work aimed to address three questions. PROJECT 1: How well do recently proposed statistical methods perform? PROJECT 2: What is the validation status of surrogate endpoints currently in use? PROJECT 3: Can a new framework for surrogate evaluation be developed by way of consensus? PROJECT 1. A scoping review was conducted to identify recent, novel statistical methods of surrogate evaluation. The method with the greatest research activity was the proportion of treatment effect explained (PTE). The performances of two fundamental methods based on the PTE were compared through simulations. Neither method was considered to be sufficient on its own. PROJECT 2. A systematic review was conducted to identify both validated and non-validated surrogate endpoints in use, and the statistical methods applied to evaluate the validated surrogate endpoints. Based on the combination of three surrogate evaluation frameworks, many surrogate endpoints accepted by regulatory agencies were not statistically valid. Correlation-based methods were the most frequently applied statistical approaches. PROJECT 3. A new framework was developed from a two-round Delphi survey followed by an international, multi-stakeholder consensus meeting. The survey included questions on the prioritisation of non-statistical factors in, use of numerical thresholds for and preferred statistical methods of surrogate evaluation. Four non-statistical factors reached consensus by the end of the second round. However, opinions were divided on the issues of thresholds and methods. A consensus meeting was subsequently held to ratify the factors, thresholds and methods to be included in a framework. Five non-statistical factors were deemed necessary, with two non-statistical factors for possible consideration, and correlation thresholds for possible consideration. In addition, the framework deemed trial-level analysis as necessary, and provided general guidance on the choice of statistical methods, rather than recommending specific methods. As found in all three projects, surrogate endpoint evaluation remains a challenging, multi-stakeholder task. Nonetheless, the newly-developed framework has partly resolved the longstanding lack of agreement on which factors to consider when evaluating a surrogate endpoint, and may catalyse further improvements to come in this cornerstone of drug development."]},{"key":"dc:title","label":"Title","values":["Search for valid surrogate endpoints: review of current evidence and development of practical strategies for surrogate evaluation"]}]}],"canonical_facts":{"dc:contributor.advisor":["Weir, Christopher","Ensor, Hannah","Lewis, Steff"],"dc:creator":["Lo, Wang Pok"],"dc:date.accessioned":["2025-08-21T14:38:49Z"],"dc:date.available":["2025-08-21T14:38:49Z"],"dc:date.issued":["2025-08-21"],"dc:description.abstract":["In light of the ongoing global pursuit of more quickly available treatments, surrogate endpoint evaluation has become a vital part of drug development. Despite the increasing uptake of surrogate endpoints and the large number of statistical methods proposed to evaluate a surrogate, there is still no consensus on how to determine when a surrogate endpoint is acceptable. Therefore, this work aimed to address three questions. PROJECT 1: How well do recently proposed statistical methods perform? PROJECT 2: What is the validation status of surrogate endpoints currently in use? PROJECT 3: Can a new framework for surrogate evaluation be developed by way of consensus? PROJECT 1. A scoping review was conducted to identify recent, novel statistical methods of surrogate evaluation. The method with the greatest research activity was the proportion of treatment effect explained (PTE). The performances of two fundamental methods based on the PTE were compared through simulations. Neither method was considered to be sufficient on its own. PROJECT 2. A systematic review was conducted to identify both validated and non-validated surrogate endpoints in use, and the statistical methods applied to evaluate the validated surrogate endpoints. Based on the combination of three surrogate evaluation frameworks, many surrogate endpoints accepted by regulatory agencies were not statistically valid. Correlation-based methods were the most frequently applied statistical approaches. PROJECT 3. A new framework was developed from a two-round Delphi survey followed by an international, multi-stakeholder consensus meeting. The survey included questions on the prioritisation of non-statistical factors in, use of numerical thresholds for and preferred statistical methods of surrogate evaluation. Four non-statistical factors reached consensus by the end of the second round. However, opinions were divided on the issues of thresholds and methods. A consensus meeting was subsequently held to ratify the factors, thresholds and methods to be included in a framework. Five non-statistical factors were deemed necessary, with two non-statistical factors for possible consideration, and correlation thresholds for possible consideration. In addition, the framework deemed trial-level analysis as necessary, and provided general guidance on the choice of statistical methods, rather than recommending specific methods. As found in all three projects, surrogate endpoint evaluation remains a challenging, multi-stakeholder task. Nonetheless, the newly-developed framework has partly resolved the longstanding lack of agreement on which factors to consider when evaluating a surrogate endpoint, and may catalyse further improvements to come in this cornerstone of drug development."],"dc:identifier.uri":["https://hdl.handle.net/1842/43843","http://dx.doi.org/10.7488/era/6374"],"dc:language.iso":["en"],"dc:publisher":["The University of Edinburgh"],"dc:rights.uri":["https://creativecommons.org/licenses/by/4.0/"],"dc:subject":["surrogate endpoints","surrogate evaluation","clinical trials","statistical methodology","expert opinion","Delphi method"],"dc:title":["Search for valid surrogate endpoints: review of current evidence and development of practical strategies for surrogate evaluation"],"dc:type":["Thesis or Dissertation"],"dc:type.qualificationlevel":["Doctoral"],"dc:type.qualificationname":["PhD Doctor of Philosophy"]},"updated_at":"2026-07-24T02:14:07Z"}