Appendices



AppendicesAppendix 1: Agreement on domains and questions Current versionOriginal version as per protocolConsensus in first roundConsensus in second roundConsensus in third roundDomainsa,b,c1. Clinical translatability of results to human disease or condition (construct validity)1. Generalisability of results to human disease or condition (construct validity)-73.3%86.7%2. Reproducibility of results in a range of clinically relevant conditions (external validity)2. External validity-66.7%86.7%3. Bias (internal validity)3. Internal validity-80.0%93.3%4. Experimental design and data analysis7. Appropriate analysis and interpretation-93.3%100.0%5. Reproducibility and replicability of methods and results in the same model5. Reproducibility of results-80.0%85.7%6. Implications of the study findings (study conclusions)d--91.7%92.3%7. Research integrity6. Research integrity-85.7%91.7%8. Research transparencyd--100.0%100.0%-4. Measurement errorse-100.0%100.0%Signalling questions1.1 Did the authors use a model that adequately represents the human disease?1.1 Are the results from the model transferable to humans?68.4%76.5%76.5%1.2 Did the authors sufficiently identify and characterise the model?2.1 Is the model used in the preclinical research really the model that the researcher wants to study?55.6%64.7%70.6%1.3 Were the method and timing of the intervention in the specific model relevant to humans?1.2 Are the dose, route of administration, and timing of the intervention transferable to humans?84.2%100.0%100.0%1.4 If the study used an surrogate outcome, was there a clear and reproducible relationship between an intervention effect on the surrogate outcome (measured at the time chosen in the preclinical research) and that on the clinical outcome?1.3 Is the correlation between the surrogate outcome measured at the appropriate time (chosen in the preclinical research) and clinical outcome strong, consistent and independent?89.5%82.4%82.4%1.5 If the study used an surrogate outcome, did previous experimental studies consistently demonstrate that change in surrogate outcome(s) by a treatment led to a comparable change in clinical outcomes?1.4 Have randomised trials involving the same or different class of drugs consistently demonstrated that improvement in surrogate outcome(s) led to improvement in clinical outcomes?83.3%87.5%87.5%1.6 Did a systematic review with or without meta-analysis demonstrate that the effect of an intervention or a similar intervention on a preclinical model was similar to that in humans?1.5 Did a systematic review demonstrate that the effect of an intervention on a preclinical model is similar to that in humans?61.1%66.7%73.3%2.1 Did the authors describe sample size calculations?7.1 Was an appropriate level of random error chosen for sample size calculations?84.2%88.2%88.2%2.2 Did the authors plan and perform statistical tests taking the type of data, the distribution of data, and the number of groups into account?7.3 Were appropriate statistical tests performed? This depends upon the type of data, the distribution of data, and the type of comparison.95.0%100.0%100.0%-7.5 In the case of multivariate analysis, were the assumptions of analysis true?f89.5%93.8%81.3%2.3 Did the authors make adjustment for multiple hypothesis testing?7.4 Was adjustment made for multiple testing?84.2%93.8%94.1%2.4 If a dose-response analysis was conducted, did the authors describe the results?7.6 If a dose-response relation is feasible, was the dose-response relationship reported which strengthens the association and conclusions?77.8%86.7%88.2%2.5 Did the authors assess and report accuracy?2.6 Did the authors assess and report precision?2.7 Did the authors assess and report sampling error?4.1 Has the measurement error been assessed appropriately?g85.0%82.4%82.4%2.8 Was the measurement error low or was the measurement error adjusted in statistical analysis?4.2 Has the measurement error been adjusted appropriately?65.0%70.6%70.6%3.1 Did the authors minimise the risks of bias such as selection bias, confounding bias, performance bias, detection bias, attrition bias, and selective outcome reporting bias?3.1 SYRCLE's risk of bias tool for animal research (in vivo research) should be used to assess the risk of bias in animal research68.4%88.2%88.2%4.1 Were the results reproduced with alternative preclinical models of the disease/condition being investigated?2.3 Can the results be reproduced with a different pre-clinical model of the disease condition investigated?78.9%87.5%93.8%4.2 Were the results consistent across a range of clinically relevant variations in the model?2.2 Are the results generalisable across different types of individuals of the model?88.9%93.8%93.8%4.3 Did the authors report take existing evidence into account when choosing the comparators?2.4 Has an appropriate comparator been chosen?73.7%93.8%87.5%5.1 Did the authors describe the experimental protocols/methods sufficiently to allow their replication?5.1 Are the experimental procedures described sufficiently to allow reproduction of the results?94.7%100.0%100.0%5.2 Did an independent group of researchers replicate the experimentald protocols/methods?--100.0%100.0%5.3 Did the authors or an independent group of researchers reproduce the results in similar and different laboratory conditions?5.2 Were the results reproduced under the same laboratory conditions?h90.0%94.1%100.0%-5.3 Were the results reproduced under different laboratory conditions?h75.0%82.4%88.2%-5.5 Were the results reproduced by a different group of researchers or was this a study that showed reproducibility of results of experiments by a different group of researchers?h80.0%82.4%82.4%6.1 Did the authors’ conclusions represent the study findings, taking its limitations into account?7.7 Were appropriate conclusions reached?80.0%88.2%88.2%6.2 Did the authors provide details on additional research required to conduct first-in-human studies?d--83.3%76.9%7.1 Did the research team obtain ethical approvals and any other regulatory approvals required to perform the research prior to the start of the study?d--84.6%84.6%7.2 Did the authors take steps to prevent unintentional changes to data?6.3 Were appropriate steps taken to prevent inadvertent errors in data collection?60.0%76.5%70.6%8.1 Did the authors describe the experimental procedures sufficiently in a protocol that was registered prior to the start of the research?8.2 Did the authors describe any deviations from the registered protocol?6.1 Were the experimental procedures described sufficiently in a registered protocol and any deviations from the protocol described appropriately?i85.0%94.1%94.1%8.3 Did the authors provide the individual subject data along with explanation for any numerical codes/substitutions or abbreviations used in the data to allow other groups of researchers to analyse?5.4 Is the raw data available for other groups of researchers to analyse?85.0%88.2%88.2%- If computer codes were used for analysis, did the authors make them available to allow reanalysis of the data?d,j--66.7%71.4%Signalling questions for which consensus was not reached-3.2 A new risk of bias tool for in-vitro research should be developed and used to assess the risk of bias in in-vitro research.57.9%53.3%60.0%k-6.2 Was the research overseen by an independent steering committee?55.0%58.8%56.3%-7.2 Were the observed differences and variation in the outcome similar to the difference used for sample size calculations?73.7%76.5%25.0%aThe first number indicates the domain number. The second number (after the decimal point) indicates the number of the signalling question under the domain.bWe have ordered the tool according to the finally agreed version.cIn the first round, the agreement on overall structure of the domain was sought. So, there are no figures available for individual domains. In subsequent rounds, agreement on each domain was sought.dThis is a new domain or signalling question eThe most frequently preferred option of the Delphi panel was to include the questions under this domain under ‘Experimental design and analysis’. Therefore, this domain does not feature in the final tool despite consensus agreement that the questions under domain were important.fThe most frequently preferred option of the panel members was to include this question in the explanation for question 6.3 current version, 7.3 initial version. Therefore, this question was removed despite achieving consensus.gThe most frequently preferred option of the panel members was to split this signalling question into three different signalling questions.hThree questions (original protocol version 5.2, 5.3, and 5.5) were combined into a single signalling question as this was the most frequently preferred option by the paneliThe most frequently preferred option of the panel members was to split this signalling question into two different signalling questions.jAlthough, this was added as a new signalling question and a consensus reached on inclusion, it was decided by the panel members that this can be combined with 8.3.kThis reflects the lack of consensus about including in vitro research for this tool rather than the lack of consensus about the necessity for a tool to assess the reliability of in vitro research.Appendix 2 Determining association between surrogate outcome and clinical outcomeThere are many ways of determining the association between two variables ADDIN EN.CITE <EndNote><Cite><Author>Lee Rodgers</Author><Year>1988</Year><RecNum>74</RecNum><DisplayText>(Lee Rodgers and Nicewander 1988)</DisplayText><record><rec-number>74</rec-number><foreign-keys><key app="EN" db-id="2xx9d5ww02zawsefx9l59592dxxass0af05d" timestamp="1535084289">74</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Lee Rodgers, Joseph</author><author>Nicewander, W. Alan</author></authors></contributors><titles><title>Thirteen Ways to Look at the Correlation Coefficient</title><secondary-title>The American Statistician</secondary-title></titles><periodical><full-title>The American Statistician</full-title></periodical><pages>59-66</pages><volume>42</volume><number>1</number><dates><year>1988</year><pub-dates><date>1988/02/01</date></pub-dates></dates><publisher>Taylor &amp; Francis</publisher><isbn>0003-1305</isbn><urls><related-urls><url>;(Lee Rodgers and Nicewander 1988). Two common ways of determining the association between a surrogate outcome and a clinical outcome (or its animal equivalent) are correlation coefficient and relative risk ratio (or relative odds ratio) of the treatment effect of the surrogate outcome compared with that of the clinical outcome PEVuZE5vdGU+PENpdGU+PEF1dGhvcj5JbnN0aXR1dGUgZm9yIFF1YWxpdHkgYW5kIEVmZmljaWVu

Y3kgaW4gSGVhbHRoIENhcmU8L0F1dGhvcj48WWVhcj4yMDExPC9ZZWFyPjxSZWNOdW0+NzU8L1Jl

Y051bT48RGlzcGxheVRleHQ+KEluc3RpdHV0ZSBmb3IgUXVhbGl0eSBhbmQgRWZmaWNpZW5jeSBp

biBIZWFsdGggQ2FyZSAyMDExLCBDaWFuaSwgQnV5c2UgZXQgYWwuIDIwMTMpPC9EaXNwbGF5VGV4

dD48cmVjb3JkPjxyZWMtbnVtYmVyPjc1PC9yZWMtbnVtYmVyPjxmb3JlaWduLWtleXM+PGtleSBh

cHA9IkVOIiBkYi1pZD0iMnh4OWQ1d3cwMnphd3NlZng5bDU5NTkyZHh4YXNzMGFmMDVkIiB0aW1l

c3RhbXA9IjE1MzUwODUzNTEiPjc1PC9rZXk+PC9mb3JlaWduLWtleXM+PHJlZi10eXBlIG5hbWU9

IkpvdXJuYWwgQXJ0aWNsZSI+MTc8L3JlZi10eXBlPjxjb250cmlidXRvcnM+PGF1dGhvcnM+PGF1

dGhvcj5JbnN0aXR1dGUgZm9yIFF1YWxpdHkgYW5kIEVmZmljaWVuY3kgaW4gSGVhbHRoIENhcmUs

PC9hdXRob3I+PC9hdXRob3JzPjwvY29udHJpYnV0b3JzPjx0aXRsZXM+PHRpdGxlPlZhbGlkaXR5

IG9mIHN1cnJvZ2F0ZSBlbmRwb2ludHMgaW4gb25jb2xvZ3k6IEV4ZWN1dGl2ZSBzdW1tYXJ5IG9m

IHJhcGlkIHJlcG9ydCBBMTAtMDUsIFZlcnNpb24gMS4xLjwvdGl0bGU+PHNlY29uZGFyeS10aXRs

ZT5odHRwczovL3d3dy5uY2JpLm5sbS5uaWguZ292L2Jvb2tzL05CSzE5ODc5OS8gKGFjY2Vzc2Vk

IDI0IEF1Z3VzdCAyMDE4KTwvc2Vjb25kYXJ5LXRpdGxlPjwvdGl0bGVzPjxwZXJpb2RpY2FsPjxm

dWxsLXRpdGxlPmh0dHBzOi8vd3d3Lm5jYmkubmxtLm5paC5nb3YvYm9va3MvTkJLMTk4Nzk5LyAo

YWNjZXNzZWQgMjQgQXVndXN0IDIwMTgpPC9mdWxsLXRpdGxlPjwvcGVyaW9kaWNhbD48ZGF0ZXM+

PHllYXI+MjAxMTwveWVhcj48L2RhdGVzPjx1cmxzPjwvdXJscz48L3JlY29yZD48L0NpdGU+PENp

dGU+PEF1dGhvcj5DaWFuaTwvQXV0aG9yPjxZZWFyPjIwMTM8L1llYXI+PFJlY051bT43NjwvUmVj

TnVtPjxyZWNvcmQ+PHJlYy1udW1iZXI+NzY8L3JlYy1udW1iZXI+PGZvcmVpZ24ta2V5cz48a2V5

IGFwcD0iRU4iIGRiLWlkPSIyeHg5ZDV3dzAyemF3c2VmeDlsNTk1OTJkeHhhc3MwYWYwNWQiIHRp

bWVzdGFtcD0iMTUzNTA4NTQxMCI+NzY8L2tleT48L2ZvcmVpZ24ta2V5cz48cmVmLXR5cGUgbmFt

ZT0iSm91cm5hbCBBcnRpY2xlIj4xNzwvcmVmLXR5cGU+PGNvbnRyaWJ1dG9ycz48YXV0aG9ycz48

YXV0aG9yPkNpYW5pLCBPLjwvYXV0aG9yPjxhdXRob3I+QnV5c2UsIE0uPC9hdXRob3I+PGF1dGhv

cj5HYXJzaWRlLCBSLjwvYXV0aG9yPjxhdXRob3I+UGF2ZXksIFQuPC9hdXRob3I+PGF1dGhvcj5T

dGVpbiwgSy48L2F1dGhvcj48YXV0aG9yPlN0ZXJuZSwgSi4gQS48L2F1dGhvcj48YXV0aG9yPlRh

eWxvciwgUi4gUy48L2F1dGhvcj48L2F1dGhvcnM+PC9jb250cmlidXRvcnM+PGF1dGgtYWRkcmVz

cz5QZW5UQUcsIEluc3RpdHV0ZSBmb3IgSGVhbHRoIFNlcnZpY2VzIFJlc2VhcmNoLCBVbml2ZXJz

aXR5IG9mIEV4ZXRlciBNZWRpY2FsIFNjaG9vbCwgVW5pdmVyc2l0eSBvZiBFeGV0ZXIsIEV4ZXRl

ciBFWDIgNFNHLCBVSy4gb3JpYW5hLmNpYW5pQHBjbWQuYWMudWs8L2F1dGgtYWRkcmVzcz48dGl0

bGVzPjx0aXRsZT5Db21wYXJpc29uIG9mIHRyZWF0bWVudCBlZmZlY3Qgc2l6ZXMgYXNzb2NpYXRl

ZCB3aXRoIHN1cnJvZ2F0ZSBhbmQgZmluYWwgcGF0aWVudCByZWxldmFudCBvdXRjb21lcyBpbiBy

YW5kb21pc2VkIGNvbnRyb2xsZWQgdHJpYWxzOiBtZXRhLWVwaWRlbWlvbG9naWNhbCBzdHVkeTwv

dGl0bGU+PHNlY29uZGFyeS10aXRsZT5CTUo8L3NlY29uZGFyeS10aXRsZT48L3RpdGxlcz48cGVy

aW9kaWNhbD48ZnVsbC10aXRsZT5CTUo8L2Z1bGwtdGl0bGU+PC9wZXJpb2RpY2FsPjxwYWdlcz5m

NDU3PC9wYWdlcz48dm9sdW1lPjM0Njwvdm9sdW1lPjxlZGl0aW9uPjIwMTMvMDEvMzE8L2VkaXRp

b24+PGtleXdvcmRzPjxrZXl3b3JkPkJpYXM8L2tleXdvcmQ+PGtleXdvcmQ+RGF0YSBDb2xsZWN0

aW9uL21ldGhvZHM8L2tleXdvcmQ+PGtleXdvcmQ+KkVwaWRlbWlvbG9naWMgUmVzZWFyY2ggRGVz

aWduPC9rZXl3b3JkPjxrZXl3b3JkPkh1bWFuczwva2V5d29yZD48a2V5d29yZD5PdXRjb21lIEFz

c2Vzc21lbnQgKEhlYWx0aCBDYXJlKS8qbWV0aG9kczwva2V5d29yZD48a2V5d29yZD5SYW5kb21p

emVkIENvbnRyb2xsZWQgVHJpYWxzIGFzIFRvcGljLyptZXRob2RzPC9rZXl3b3JkPjxrZXl3b3Jk

PlJlcHJvZHVjaWJpbGl0eSBvZiBSZXN1bHRzPC9rZXl3b3JkPjwva2V5d29yZHM+PGRhdGVzPjx5

ZWFyPjIwMTM8L3llYXI+PHB1Yi1kYXRlcz48ZGF0ZT5KYW4gMjk8L2RhdGU+PC9wdWItZGF0ZXM+

PC9kYXRlcz48aXNibj4xNzU2LTE4MzMgKEVsZWN0cm9uaWMpJiN4RDswOTU5LTgxMzggKExpbmtp

bmcpPC9pc2JuPjxhY2Nlc3Npb24tbnVtPjIzMzYwNzE5PC9hY2Nlc3Npb24tbnVtPjx1cmxzPjxy

ZWxhdGVkLXVybHM+PHVybD5odHRwczovL3d3dy5uY2JpLm5sbS5uaWguZ292L3B1Ym1lZC8yMzM2

MDcxOTwvdXJsPjwvcmVsYXRlZC11cmxzPjwvdXJscz48Y3VzdG9tMj5QTUMzNTU4NDExPC9jdXN0

b20yPjxlbGVjdHJvbmljLXJlc291cmNlLW51bT4xMC4xMTM2L2Jtai5mNDU3PC9lbGVjdHJvbmlj

LXJlc291cmNlLW51bT48L3JlY29yZD48L0NpdGU+PC9FbmROb3RlPgB=

ADDIN EN.CITE PEVuZE5vdGU+PENpdGU+PEF1dGhvcj5JbnN0aXR1dGUgZm9yIFF1YWxpdHkgYW5kIEVmZmljaWVu

Y3kgaW4gSGVhbHRoIENhcmU8L0F1dGhvcj48WWVhcj4yMDExPC9ZZWFyPjxSZWNOdW0+NzU8L1Jl

Y051bT48RGlzcGxheVRleHQ+KEluc3RpdHV0ZSBmb3IgUXVhbGl0eSBhbmQgRWZmaWNpZW5jeSBp

biBIZWFsdGggQ2FyZSAyMDExLCBDaWFuaSwgQnV5c2UgZXQgYWwuIDIwMTMpPC9EaXNwbGF5VGV4

dD48cmVjb3JkPjxyZWMtbnVtYmVyPjc1PC9yZWMtbnVtYmVyPjxmb3JlaWduLWtleXM+PGtleSBh

cHA9IkVOIiBkYi1pZD0iMnh4OWQ1d3cwMnphd3NlZng5bDU5NTkyZHh4YXNzMGFmMDVkIiB0aW1l

c3RhbXA9IjE1MzUwODUzNTEiPjc1PC9rZXk+PC9mb3JlaWduLWtleXM+PHJlZi10eXBlIG5hbWU9

IkpvdXJuYWwgQXJ0aWNsZSI+MTc8L3JlZi10eXBlPjxjb250cmlidXRvcnM+PGF1dGhvcnM+PGF1

dGhvcj5JbnN0aXR1dGUgZm9yIFF1YWxpdHkgYW5kIEVmZmljaWVuY3kgaW4gSGVhbHRoIENhcmUs

PC9hdXRob3I+PC9hdXRob3JzPjwvY29udHJpYnV0b3JzPjx0aXRsZXM+PHRpdGxlPlZhbGlkaXR5

IG9mIHN1cnJvZ2F0ZSBlbmRwb2ludHMgaW4gb25jb2xvZ3k6IEV4ZWN1dGl2ZSBzdW1tYXJ5IG9m

IHJhcGlkIHJlcG9ydCBBMTAtMDUsIFZlcnNpb24gMS4xLjwvdGl0bGU+PHNlY29uZGFyeS10aXRs

ZT5odHRwczovL3d3dy5uY2JpLm5sbS5uaWguZ292L2Jvb2tzL05CSzE5ODc5OS8gKGFjY2Vzc2Vk

IDI0IEF1Z3VzdCAyMDE4KTwvc2Vjb25kYXJ5LXRpdGxlPjwvdGl0bGVzPjxwZXJpb2RpY2FsPjxm

dWxsLXRpdGxlPmh0dHBzOi8vd3d3Lm5jYmkubmxtLm5paC5nb3YvYm9va3MvTkJLMTk4Nzk5LyAo

YWNjZXNzZWQgMjQgQXVndXN0IDIwMTgpPC9mdWxsLXRpdGxlPjwvcGVyaW9kaWNhbD48ZGF0ZXM+

PHllYXI+MjAxMTwveWVhcj48L2RhdGVzPjx1cmxzPjwvdXJscz48L3JlY29yZD48L0NpdGU+PENp

dGU+PEF1dGhvcj5DaWFuaTwvQXV0aG9yPjxZZWFyPjIwMTM8L1llYXI+PFJlY051bT43NjwvUmVj

TnVtPjxyZWNvcmQ+PHJlYy1udW1iZXI+NzY8L3JlYy1udW1iZXI+PGZvcmVpZ24ta2V5cz48a2V5

IGFwcD0iRU4iIGRiLWlkPSIyeHg5ZDV3dzAyemF3c2VmeDlsNTk1OTJkeHhhc3MwYWYwNWQiIHRp

bWVzdGFtcD0iMTUzNTA4NTQxMCI+NzY8L2tleT48L2ZvcmVpZ24ta2V5cz48cmVmLXR5cGUgbmFt

ZT0iSm91cm5hbCBBcnRpY2xlIj4xNzwvcmVmLXR5cGU+PGNvbnRyaWJ1dG9ycz48YXV0aG9ycz48

YXV0aG9yPkNpYW5pLCBPLjwvYXV0aG9yPjxhdXRob3I+QnV5c2UsIE0uPC9hdXRob3I+PGF1dGhv

cj5HYXJzaWRlLCBSLjwvYXV0aG9yPjxhdXRob3I+UGF2ZXksIFQuPC9hdXRob3I+PGF1dGhvcj5T

dGVpbiwgSy48L2F1dGhvcj48YXV0aG9yPlN0ZXJuZSwgSi4gQS48L2F1dGhvcj48YXV0aG9yPlRh

eWxvciwgUi4gUy48L2F1dGhvcj48L2F1dGhvcnM+PC9jb250cmlidXRvcnM+PGF1dGgtYWRkcmVz

cz5QZW5UQUcsIEluc3RpdHV0ZSBmb3IgSGVhbHRoIFNlcnZpY2VzIFJlc2VhcmNoLCBVbml2ZXJz

aXR5IG9mIEV4ZXRlciBNZWRpY2FsIFNjaG9vbCwgVW5pdmVyc2l0eSBvZiBFeGV0ZXIsIEV4ZXRl

ciBFWDIgNFNHLCBVSy4gb3JpYW5hLmNpYW5pQHBjbWQuYWMudWs8L2F1dGgtYWRkcmVzcz48dGl0

bGVzPjx0aXRsZT5Db21wYXJpc29uIG9mIHRyZWF0bWVudCBlZmZlY3Qgc2l6ZXMgYXNzb2NpYXRl

ZCB3aXRoIHN1cnJvZ2F0ZSBhbmQgZmluYWwgcGF0aWVudCByZWxldmFudCBvdXRjb21lcyBpbiBy

YW5kb21pc2VkIGNvbnRyb2xsZWQgdHJpYWxzOiBtZXRhLWVwaWRlbWlvbG9naWNhbCBzdHVkeTwv

dGl0bGU+PHNlY29uZGFyeS10aXRsZT5CTUo8L3NlY29uZGFyeS10aXRsZT48L3RpdGxlcz48cGVy

aW9kaWNhbD48ZnVsbC10aXRsZT5CTUo8L2Z1bGwtdGl0bGU+PC9wZXJpb2RpY2FsPjxwYWdlcz5m

NDU3PC9wYWdlcz48dm9sdW1lPjM0Njwvdm9sdW1lPjxlZGl0aW9uPjIwMTMvMDEvMzE8L2VkaXRp

b24+PGtleXdvcmRzPjxrZXl3b3JkPkJpYXM8L2tleXdvcmQ+PGtleXdvcmQ+RGF0YSBDb2xsZWN0

aW9uL21ldGhvZHM8L2tleXdvcmQ+PGtleXdvcmQ+KkVwaWRlbWlvbG9naWMgUmVzZWFyY2ggRGVz

aWduPC9rZXl3b3JkPjxrZXl3b3JkPkh1bWFuczwva2V5d29yZD48a2V5d29yZD5PdXRjb21lIEFz

c2Vzc21lbnQgKEhlYWx0aCBDYXJlKS8qbWV0aG9kczwva2V5d29yZD48a2V5d29yZD5SYW5kb21p

emVkIENvbnRyb2xsZWQgVHJpYWxzIGFzIFRvcGljLyptZXRob2RzPC9rZXl3b3JkPjxrZXl3b3Jk

PlJlcHJvZHVjaWJpbGl0eSBvZiBSZXN1bHRzPC9rZXl3b3JkPjwva2V5d29yZHM+PGRhdGVzPjx5

ZWFyPjIwMTM8L3llYXI+PHB1Yi1kYXRlcz48ZGF0ZT5KYW4gMjk8L2RhdGU+PC9wdWItZGF0ZXM+

PC9kYXRlcz48aXNibj4xNzU2LTE4MzMgKEVsZWN0cm9uaWMpJiN4RDswOTU5LTgxMzggKExpbmtp

bmcpPC9pc2JuPjxhY2Nlc3Npb24tbnVtPjIzMzYwNzE5PC9hY2Nlc3Npb24tbnVtPjx1cmxzPjxy

ZWxhdGVkLXVybHM+PHVybD5odHRwczovL3d3dy5uY2JpLm5sbS5uaWguZ292L3B1Ym1lZC8yMzM2

MDcxOTwvdXJsPjwvcmVsYXRlZC11cmxzPjwvdXJscz48Y3VzdG9tMj5QTUMzNTU4NDExPC9jdXN0

b20yPjxlbGVjdHJvbmljLXJlc291cmNlLW51bT4xMC4xMTM2L2Jtai5mNDU3PC9lbGVjdHJvbmlj

LXJlc291cmNlLW51bT48L3JlY29yZD48L0NpdGU+PC9FbmROb3RlPgB=

ADDIN EN.CITE.DATA (Institute for Quality and Efficiency in Health Care 2011, Ciani, Buyse et al. 2013). There are no universally accepted threshold values for interpretation of high association for neither of these measures. Correlation coefficients range from -1.00 to +1.00. Values close to -1.00 or +1.00 indicate high negative or positive correlation, while values close to 0 indicate lack of correlation between two variables. Institute for Quality and Efficiency in Health Care suggests that the correlation can be considered high when lower limit of the 95% confidence interval (CI) of the correlation coefficient is ≥ 0.85, low when the upper limit of the 95% CI is ≤ 0.70, and moderate otherwise ADDIN EN.CITE <EndNote><Cite><Author>Institute for Quality and Efficiency in Health Care</Author><Year>2011</Year><RecNum>75</RecNum><DisplayText>(Institute for Quality and Efficiency in Health Care 2011)</DisplayText><record><rec-number>75</rec-number><foreign-keys><key app="EN" db-id="2xx9d5ww02zawsefx9l59592dxxass0af05d" timestamp="1535085351">75</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Institute for Quality and Efficiency in Health Care,</author></authors></contributors><titles><title>Validity of surrogate endpoints in oncology: Executive summary of rapid report A10-05, Version 1.1.</title><secondary-title> (accessed 24 August 2018)</secondary-title></titles><periodical><full-title> (accessed 24 August 2018)</full-title></periodical><dates><year>2011</year></dates><urls></urls></record></Cite></EndNote>(Institute for Quality and Efficiency in Health Care 2011). There is also no guidance on the interpretation of relative odds ratio or relative risk ratio values. Values close to 1.00 indicate that the treatment effects of surrogate outcomes and clinical outcomes are similar. Values below 1.00 indicate overestimation of treatment effects and values above 1.00 indicate underestimation of treatment effects by the surrogate outcome for a ‘bad outcome’ (an outcome where lower proportions or values are preferable, for example proportion dead or symptom score). The reverse is true for a ‘good outcome’ (an outcome where higher proportions or values are preferable, for example, cure or health-related quality of life). One can find whether the differences in the effect estimates between the surrogate outcome and clinical outcome are statistically significant by using methods described by Altman et al. to compare the effect estimates ADDIN EN.CITE <EndNote><Cite><Author>Altman</Author><Year>2003</Year><RecNum>77</RecNum><DisplayText>(Altman and Bland 2003)</DisplayText><record><rec-number>77</rec-number><foreign-keys><key app="EN" db-id="2xx9d5ww02zawsefx9l59592dxxass0af05d" timestamp="1535086572">77</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Altman, Douglas G</author><author>Bland, J Martin</author></authors></contributors><titles><title>Interaction revisited: the difference between two estimates</title><secondary-title>BMJ</secondary-title></titles><periodical><full-title>BMJ</full-title></periodical><pages>219</pages><volume>326</volume><number>7382</number><dates><year>2003</year></dates><urls><related-urls><url> %J BMJ</electronic-resource-num></record></Cite></EndNote>(Altman and Bland 2003), although some adjustments have to be made to the calculations of the pooled standard error because the estimates have been obtained from the same samples ADDIN EN.CITE <EndNote><Cite><Author>Borenstein</Author><Year>2009</Year><RecNum>78</RecNum><DisplayText>(Borenstein, Hedges et al. 2009)</DisplayText><record><rec-number>78</rec-number><foreign-keys><key app="EN" db-id="2xx9d5ww02zawsefx9l59592dxxass0af05d" timestamp="1535087018">78</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Borenstein, M</author><author>Hedges, L</author><author>Higgins, J</author><author>Rothstein, HR</author></authors></contributors><titles><title>Chapter 4: Effect sizes based on means</title><secondary-title>Introduction to meta-analysis</secondary-title></titles><section>21-32</section><dates><year>2009</year></dates><publisher>John Wiley &amp; Sons Inc</publisher><urls></urls></record></Cite></EndNote>(Borenstein, Hedges et al. 2009).Appendix 3 Examples of surrogate outcomes that failed to be a good substitute for clinical outcomesThere are several examples of treatments that cause an improvement in the surrogate outcome (which is correlated with the clinical outcome) but actually result in harm to patients. Ventricular ectopic beats are associated with adverse prognosis in patients with myocardial infarction and class 1 antiarrhythmic agents effectively suppress ventricular arrhythmias in animals and humans; however, these drugs increased human mortality when evaluated in randomised controlled trials ADDIN EN.CITE <EndNote><Cite><Author>Bucher</Author><Year>1999</Year><RecNum>82</RecNum><DisplayText>(Bucher, Guyatt et al. 1999)</DisplayText><record><rec-number>82</rec-number><foreign-keys><key app="EN" db-id="2xx9d5ww02zawsefx9l59592dxxass0af05d" timestamp="1535090162">82</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Bucher, H. C.</author><author>Guyatt, G. H.</author><author>Cook, D. J.</author><author>Holbrook, A.</author><author>McAlister, F. A.</author><author>for the Evidence-Based Medicine Working, Group</author></authors></contributors><titles><title>Users&apos; guides to the medical literature: XIX. applying clinical trial results a. how to use an article measuring the effect of an intervention on surrogate end points</title><secondary-title>JAMA</secondary-title></titles><periodical><full-title>JAMA</full-title></periodical><pages>771-778</pages><volume>282</volume><number>8</number><dates><year>1999</year></dates><isbn>0098-7484</isbn><urls><related-urls><url>;(Bucher, Guyatt et al. 1999). This probably resulted in tens of thousands of deaths in people with non-lethal arrhythmias because of the reliance on the surrogate outcomes ADDIN EN.CITE <EndNote><Cite><Author>Bucher</Author><Year>1999</Year><RecNum>82</RecNum><DisplayText>(Bucher, Guyatt et al. 1999)</DisplayText><record><rec-number>82</rec-number><foreign-keys><key app="EN" db-id="2xx9d5ww02zawsefx9l59592dxxass0af05d" timestamp="1535090162">82</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Bucher, H. C.</author><author>Guyatt, G. H.</author><author>Cook, D. J.</author><author>Holbrook, A.</author><author>McAlister, F. A.</author><author>for the Evidence-Based Medicine Working, Group</author></authors></contributors><titles><title>Users&apos; guides to the medical literature: XIX. applying clinical trial results a. how to use an article measuring the effect of an intervention on surrogate end points</title><secondary-title>JAMA</secondary-title></titles><periodical><full-title>JAMA</full-title></periodical><pages>771-778</pages><volume>282</volume><number>8</number><dates><year>1999</year></dates><isbn>0098-7484</isbn><urls><related-urls><url>;(Bucher, Guyatt et al. 1999). In people with heart failure, angiotensin-converting enzyme inhibitors improved exercise capacity and decreased mortality as demonstrated by randomised controlled trials, but other classes of drugs such as milrinone and epoprostenol caused increased cardiovascular mortality despite improving the exercise tolerance ADDIN EN.CITE <EndNote><Cite><Author>Bucher</Author><Year>1999</Year><RecNum>82</RecNum><DisplayText>(Bucher, Guyatt et al. 1999)</DisplayText><record><rec-number>82</rec-number><foreign-keys><key app="EN" db-id="2xx9d5ww02zawsefx9l59592dxxass0af05d" timestamp="1535090162">82</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Bucher, H. C.</author><author>Guyatt, G. H.</author><author>Cook, D. J.</author><author>Holbrook, A.</author><author>McAlister, F. A.</author><author>for the Evidence-Based Medicine Working, Group</author></authors></contributors><titles><title>Users&apos; guides to the medical literature: XIX. applying clinical trial results a. how to use an article measuring the effect of an intervention on surrogate end points</title><secondary-title>JAMA</secondary-title></titles><periodical><full-title>JAMA</full-title></periodical><pages>771-778</pages><volume>282</volume><number>8</number><dates><year>1999</year></dates><isbn>0098-7484</isbn><urls><related-urls><url>;(Bucher, Guyatt et al. 1999). In diabetes, glycosylated haemoglobin (HbA1c) is used as a surrogate marker for clinical outcomes. However, some oral hypoglycaemic drugs that reduce HbA1c increase the risk of cardiovascular events ADDIN EN.CITE <EndNote><Cite><Author>Yudkin</Author><Year>2011</Year><RecNum>83</RecNum><DisplayText>(Yudkin, Lipska et al. 2011)</DisplayText><record><rec-number>83</rec-number><foreign-keys><key app="EN" db-id="2xx9d5ww02zawsefx9l59592dxxass0af05d" timestamp="1535091303">83</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Yudkin, John S</author><author>Lipska, Kasia J</author><author>Montori, Victor M</author></authors></contributors><titles><title>The idolatry of the surrogate</title><secondary-title>BMJ</secondary-title></titles><periodical><full-title>BMJ</full-title></periodical><volume>343</volume><dates><year>2011</year></dates><urls><related-urls><url> %J BMJ</electronic-resource-num></record></Cite></EndNote>(Yudkin, Lipska et al. 2011). In cancer, Kim et al. found that only about 15% of the 36 cancer drugs approved by the US FDA between 2008 and 2012 on the basis of reduction in tumour size or volume, progression-free survival, or disease-free survival improved overall survival ADDIN EN.CITE <EndNote><Cite><Author>Kim</Author><Year>2015</Year><RecNum>84</RecNum><DisplayText>(Kim and Prasad 2015)</DisplayText><record><rec-number>84</rec-number><foreign-keys><key app="EN" db-id="2xx9d5ww02zawsefx9l59592dxxass0af05d" timestamp="1535091871">84</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Kim, C.</author><author>Prasad, V.</author></authors></contributors><titles><title>Cancer drugs approved on the basis of a surrogate end point and subsequent overall survival: An analysis of 5 years of us food and drug administration approvals</title><secondary-title>JAMA Internal Medicine</secondary-title></titles><periodical><full-title>JAMA Internal Medicine</full-title></periodical><pages>1992-1994</pages><volume>175</volume><number>12</number><dates><year>2015</year></dates><isbn>2168-6106</isbn><urls><related-urls><url>;(Kim and Prasad 2015). In 50% of the cancer drugs, it was clearly demonstrated that there was no improvement in overall survival with uncertainty remaining in the remaining 35% of the drugs ADDIN EN.CITE <EndNote><Cite><Author>Kim</Author><Year>2015</Year><RecNum>84</RecNum><DisplayText>(Kim and Prasad 2015)</DisplayText><record><rec-number>84</rec-number><foreign-keys><key app="EN" db-id="2xx9d5ww02zawsefx9l59592dxxass0af05d" timestamp="1535091871">84</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Kim, C.</author><author>Prasad, V.</author></authors></contributors><titles><title>Cancer drugs approved on the basis of a surrogate end point and subsequent overall survival: An analysis of 5 years of us food and drug administration approvals</title><secondary-title>JAMA Internal Medicine</secondary-title></titles><periodical><full-title>JAMA Internal Medicine</full-title></periodical><pages>1992-1994</pages><volume>175</volume><number>12</number><dates><year>2015</year></dates><isbn>2168-6106</isbn><urls><related-urls><url>;(Kim and Prasad 2015). Rupp et al. evaluated whether the 50% of the cancer drugs with no improvement in overall survival improved the health-related quality of life (HRQoL). They found information on health-related quality of life for only 7 of the 18 drugs: only one of the drugs improved HRQoL; two drugs worsened the HRQoL; there was no difference in HRQoL or inconsistent results about HRQoL for the remaining drugs ADDIN EN.CITE <EndNote><Cite><Author>Rupp</Author><Year>2017</Year><RecNum>85</RecNum><DisplayText>(Rupp and Zuckerman 2017)</DisplayText><record><rec-number>85</rec-number><foreign-keys><key app="EN" db-id="2xx9d5ww02zawsefx9l59592dxxass0af05d" timestamp="1535092289">85</key></foreign-keys><ref-type name="Journal Article">17</ref-type><contributors><authors><author>Rupp, T.</author><author>Zuckerman, D.</author></authors></contributors><titles><title>Quality of life, overall survival, and costs of cancer drugs approved based on surrogate endpoints</title><secondary-title>JAMA Internal Medicine</secondary-title></titles><periodical><full-title>JAMA Internal Medicine</full-title></periodical><pages>276-277</pages><volume>177</volume><number>2</number><dates><year>2017</year></dates><isbn>2168-6106</isbn><urls><related-urls><url>;(Rupp and Zuckerman 2017).References ADDIN EN.REFLIST Altman, D. G. and J. M. Bland (2003). "Interaction revisited: the difference between two estimates." BMJ 326(7382): 219.Borenstein, M., L. Hedges, J. Higgins and H. Rothstein (2009). "Chapter 4: Effect sizes based on means." Introduction to meta-analysis.Bucher, H. C., G. H. Guyatt, D. J. Cook, A. Holbrook, F. A. McAlister and G. for the Evidence-Based Medicine Working (1999). "Users' guides to the medical literature: XIX. applying clinical trial results a. how to use an article measuring the effect of an intervention on surrogate end points." JAMA 282(8): 771-778.Ciani, O., M. Buyse, R. Garside, T. Pavey, K. Stein, J. A. Sterne and R. S. Taylor (2013). "Comparison of treatment effect sizes associated with surrogate and final patient relevant outcomes in randomised controlled trials: meta-epidemiological study." BMJ 346: f457.Institute for Quality and Efficiency in Health Care (2011). "Validity of surrogate endpoints in oncology: Executive summary of rapid report A10-05, Version 1.1." (accessed 24 August 2018).Kim, C. and V. Prasad (2015). "Cancer drugs approved on the basis of a surrogate end point and subsequent overall survival: An analysis of 5 years of us food and drug administration approvals." JAMA Internal Medicine 175(12): 1992-1994.Lee Rodgers, J. and W. A. Nicewander (1988). "Thirteen Ways to Look at the Correlation Coefficient." The American Statistician 42(1): 59-66.Rupp, T. and D. Zuckerman (2017). "Quality of life, overall survival, and costs of cancer drugs approved based on surrogate endpoints." JAMA Internal Medicine 177(2): 276-277.Yudkin, J. S., K. J. Lipska and V. M. Montori (2011). "The idolatry of the surrogate." BMJ 343. ................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download