The importance of baseline viral load when assessing relative efficacy in treatment-naïve HBeAg-positive chronic hepatitis B: A systematic review and network meta-analysis

Stuart Mealing, Isabella Ghement, Neil Hawkins, David A. Scott, Benedicte Lescrauwaet, Maureen Watt, Mark Thursz, Pietro Lampertico, Lorenzo Mantovani, Edith Morais, Bruno Bregman, Michel Cucherat

Research output: Contribution to journalArticle

Abstract

Background: To date no network meta-analysis (NMA) has accounted for baseline variations in viral load when assessing the relative efficacy of interventions for chronic hepatitis B (CHB). We undertook baseline-adjusted and unadjusted analyses using the same data to explore the impact of baseline viral load (BVL) on CHB treatment response. Methods: We searched Embase, Medline, Medline in Process and the Cochrane CENTRAL databases for randomised clinical trials (RCTs) of monotherapy interventions at licensed doses for use in CHB. Search strategies comprised CHB disease and drug terms (a combination of controlled vocabulary and free text terms) and also a bespoke RCT filter.The NMA was undertaken in WinBUGs using fixed and random effects methods, using data obtained from a systematic review. Individual patient data (IPD) from an entecavir clinical trial were used to quantify the impact of different baseline characteristics (in particular undetectable viral load (UVL) at 1 year) on relative treatment effect. Study level mean baseline values from all identified studies were used. Results were generated for UVL and presented as relative risks (RRs) and 95% credible intervals (CrIs) using entecavir as reference treatment.Results: Overall, for all eight relevant interventions we identified 3,000 abstracts. Following full text review a total of 35 (including the contents of six clinical study reports) met the inclusion critera; 19 were in hepatitis B e antigen (HBeAg)-positive patients and 14 of the 19 contained outcome information of relevance to the NMA.Entecavir and tenofovir studies had heterogeneous patient populations in terms of BVL (mean values 9.29 and 8.65 log10 copies/ml respectively). After adjusting UVL for BVL using an informative prior based on the IPD analysis, the difference between entecavir and tenofovir was not statistically significant (RR 1.27, 95% CrI 0.96 to 1.47 - fixed effects). A similar conclusion was found in all sensitivity analyses. Adjusted tenofovir results were more consistent with observed clinical trial response rates. Conclusions: This study demonstrates the importance of adjusting for BVL when assessing the relative efficacy of CHB interventions in achieving UVL. This has implications for both clinical and economic decision making.

Original languageEnglish
Article number21
JournalSystematic Reviews
Volume3
Issue number1
DOIs
Publication statusPublished - Mar 7 2014

Fingerprint

Hepatitis B e Antigens
Chronic Hepatitis B
Viral Load
Tenofovir
Randomized Controlled Trials
Clinical Trials
Controlled Vocabulary
Network Meta-Analysis
Therapeutics
Economics
Databases
entecavir

Keywords

  • Entecavir
  • Network meta-analysis
  • Relative efficacy
  • Systematic review
  • Virologic response

ASJC Scopus subject areas

  • Medicine (miscellaneous)

Cite this

The importance of baseline viral load when assessing relative efficacy in treatment-naïve HBeAg-positive chronic hepatitis B : A systematic review and network meta-analysis. / Mealing, Stuart; Ghement, Isabella; Hawkins, Neil; Scott, David A.; Lescrauwaet, Benedicte; Watt, Maureen; Thursz, Mark; Lampertico, Pietro; Mantovani, Lorenzo; Morais, Edith; Bregman, Bruno; Cucherat, Michel.

In: Systematic Reviews, Vol. 3, No. 1, 21, 07.03.2014.

Research output: Contribution to journalArticle

Mealing, S, Ghement, I, Hawkins, N, Scott, DA, Lescrauwaet, B, Watt, M, Thursz, M, Lampertico, P, Mantovani, L, Morais, E, Bregman, B & Cucherat, M 2014, 'The importance of baseline viral load when assessing relative efficacy in treatment-naïve HBeAg-positive chronic hepatitis B: A systematic review and network meta-analysis', Systematic Reviews, vol. 3, no. 1, 21. https://doi.org/10.1186/2046-4053-3-21
Mealing, Stuart ; Ghement, Isabella ; Hawkins, Neil ; Scott, David A. ; Lescrauwaet, Benedicte ; Watt, Maureen ; Thursz, Mark ; Lampertico, Pietro ; Mantovani, Lorenzo ; Morais, Edith ; Bregman, Bruno ; Cucherat, Michel. / The importance of baseline viral load when assessing relative efficacy in treatment-naïve HBeAg-positive chronic hepatitis B : A systematic review and network meta-analysis. In: Systematic Reviews. 2014 ; Vol. 3, No. 1.
@article{448267a7afb3409b82e59667cc68769a,
title = "The importance of baseline viral load when assessing relative efficacy in treatment-na{\"i}ve HBeAg-positive chronic hepatitis B: A systematic review and network meta-analysis",
abstract = "Background: To date no network meta-analysis (NMA) has accounted for baseline variations in viral load when assessing the relative efficacy of interventions for chronic hepatitis B (CHB). We undertook baseline-adjusted and unadjusted analyses using the same data to explore the impact of baseline viral load (BVL) on CHB treatment response. Methods: We searched Embase, Medline, Medline in Process and the Cochrane CENTRAL databases for randomised clinical trials (RCTs) of monotherapy interventions at licensed doses for use in CHB. Search strategies comprised CHB disease and drug terms (a combination of controlled vocabulary and free text terms) and also a bespoke RCT filter.The NMA was undertaken in WinBUGs using fixed and random effects methods, using data obtained from a systematic review. Individual patient data (IPD) from an entecavir clinical trial were used to quantify the impact of different baseline characteristics (in particular undetectable viral load (UVL) at 1 year) on relative treatment effect. Study level mean baseline values from all identified studies were used. Results were generated for UVL and presented as relative risks (RRs) and 95{\%} credible intervals (CrIs) using entecavir as reference treatment.Results: Overall, for all eight relevant interventions we identified 3,000 abstracts. Following full text review a total of 35 (including the contents of six clinical study reports) met the inclusion critera; 19 were in hepatitis B e antigen (HBeAg)-positive patients and 14 of the 19 contained outcome information of relevance to the NMA.Entecavir and tenofovir studies had heterogeneous patient populations in terms of BVL (mean values 9.29 and 8.65 log10 copies/ml respectively). After adjusting UVL for BVL using an informative prior based on the IPD analysis, the difference between entecavir and tenofovir was not statistically significant (RR 1.27, 95{\%} CrI 0.96 to 1.47 - fixed effects). A similar conclusion was found in all sensitivity analyses. Adjusted tenofovir results were more consistent with observed clinical trial response rates. Conclusions: This study demonstrates the importance of adjusting for BVL when assessing the relative efficacy of CHB interventions in achieving UVL. This has implications for both clinical and economic decision making.",
keywords = "Entecavir, Network meta-analysis, Relative efficacy, Systematic review, Virologic response",
author = "Stuart Mealing and Isabella Ghement and Neil Hawkins and Scott, {David A.} and Benedicte Lescrauwaet and Maureen Watt and Mark Thursz and Pietro Lampertico and Lorenzo Mantovani and Edith Morais and Bruno Bregman and Michel Cucherat",
year = "2014",
month = "3",
day = "7",
doi = "10.1186/2046-4053-3-21",
language = "English",
volume = "3",
journal = "Systematic Reviews",
issn = "2046-4053",
publisher = "BioMed Central",
number = "1",

}

TY - JOUR

T1 - The importance of baseline viral load when assessing relative efficacy in treatment-naïve HBeAg-positive chronic hepatitis B

T2 - A systematic review and network meta-analysis

AU - Mealing, Stuart

AU - Ghement, Isabella

AU - Hawkins, Neil

AU - Scott, David A.

AU - Lescrauwaet, Benedicte

AU - Watt, Maureen

AU - Thursz, Mark

AU - Lampertico, Pietro

AU - Mantovani, Lorenzo

AU - Morais, Edith

AU - Bregman, Bruno

AU - Cucherat, Michel

PY - 2014/3/7

Y1 - 2014/3/7

N2 - Background: To date no network meta-analysis (NMA) has accounted for baseline variations in viral load when assessing the relative efficacy of interventions for chronic hepatitis B (CHB). We undertook baseline-adjusted and unadjusted analyses using the same data to explore the impact of baseline viral load (BVL) on CHB treatment response. Methods: We searched Embase, Medline, Medline in Process and the Cochrane CENTRAL databases for randomised clinical trials (RCTs) of monotherapy interventions at licensed doses for use in CHB. Search strategies comprised CHB disease and drug terms (a combination of controlled vocabulary and free text terms) and also a bespoke RCT filter.The NMA was undertaken in WinBUGs using fixed and random effects methods, using data obtained from a systematic review. Individual patient data (IPD) from an entecavir clinical trial were used to quantify the impact of different baseline characteristics (in particular undetectable viral load (UVL) at 1 year) on relative treatment effect. Study level mean baseline values from all identified studies were used. Results were generated for UVL and presented as relative risks (RRs) and 95% credible intervals (CrIs) using entecavir as reference treatment.Results: Overall, for all eight relevant interventions we identified 3,000 abstracts. Following full text review a total of 35 (including the contents of six clinical study reports) met the inclusion critera; 19 were in hepatitis B e antigen (HBeAg)-positive patients and 14 of the 19 contained outcome information of relevance to the NMA.Entecavir and tenofovir studies had heterogeneous patient populations in terms of BVL (mean values 9.29 and 8.65 log10 copies/ml respectively). After adjusting UVL for BVL using an informative prior based on the IPD analysis, the difference between entecavir and tenofovir was not statistically significant (RR 1.27, 95% CrI 0.96 to 1.47 - fixed effects). A similar conclusion was found in all sensitivity analyses. Adjusted tenofovir results were more consistent with observed clinical trial response rates. Conclusions: This study demonstrates the importance of adjusting for BVL when assessing the relative efficacy of CHB interventions in achieving UVL. This has implications for both clinical and economic decision making.

AB - Background: To date no network meta-analysis (NMA) has accounted for baseline variations in viral load when assessing the relative efficacy of interventions for chronic hepatitis B (CHB). We undertook baseline-adjusted and unadjusted analyses using the same data to explore the impact of baseline viral load (BVL) on CHB treatment response. Methods: We searched Embase, Medline, Medline in Process and the Cochrane CENTRAL databases for randomised clinical trials (RCTs) of monotherapy interventions at licensed doses for use in CHB. Search strategies comprised CHB disease and drug terms (a combination of controlled vocabulary and free text terms) and also a bespoke RCT filter.The NMA was undertaken in WinBUGs using fixed and random effects methods, using data obtained from a systematic review. Individual patient data (IPD) from an entecavir clinical trial were used to quantify the impact of different baseline characteristics (in particular undetectable viral load (UVL) at 1 year) on relative treatment effect. Study level mean baseline values from all identified studies were used. Results were generated for UVL and presented as relative risks (RRs) and 95% credible intervals (CrIs) using entecavir as reference treatment.Results: Overall, for all eight relevant interventions we identified 3,000 abstracts. Following full text review a total of 35 (including the contents of six clinical study reports) met the inclusion critera; 19 were in hepatitis B e antigen (HBeAg)-positive patients and 14 of the 19 contained outcome information of relevance to the NMA.Entecavir and tenofovir studies had heterogeneous patient populations in terms of BVL (mean values 9.29 and 8.65 log10 copies/ml respectively). After adjusting UVL for BVL using an informative prior based on the IPD analysis, the difference between entecavir and tenofovir was not statistically significant (RR 1.27, 95% CrI 0.96 to 1.47 - fixed effects). A similar conclusion was found in all sensitivity analyses. Adjusted tenofovir results were more consistent with observed clinical trial response rates. Conclusions: This study demonstrates the importance of adjusting for BVL when assessing the relative efficacy of CHB interventions in achieving UVL. This has implications for both clinical and economic decision making.

KW - Entecavir

KW - Network meta-analysis

KW - Relative efficacy

KW - Systematic review

KW - Virologic response

UR - http://www.scopus.com/inward/record.url?scp=84907275210&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84907275210&partnerID=8YFLogxK

U2 - 10.1186/2046-4053-3-21

DO - 10.1186/2046-4053-3-21

M3 - Article

C2 - 24602249

AN - SCOPUS:84907275210

VL - 3

JO - Systematic Reviews

JF - Systematic Reviews

SN - 2046-4053

IS - 1

M1 - 21

ER -