New algorithm to estimate risk of natalizumab-associated progressive multifocal leukoencephalopathy (PML) in anti-JCV antibody positive patients: analyses of clinical trial data to provide further temporal precision and inform clinical practice
Author(s):
H. Koendgen
,
H. Koendgen
Affiliations:
I. Chang
,
I. Chang
Affiliations:
B. Sperling
,
B. Sperling
Affiliations:
G. Bloomgren
,
G. Bloomgren
Affiliations:
B. Haddock
,
B. Haddock
Affiliations:
S. Richman
,
S. Richman
Affiliations:
P.-R. Ho
,
P.-R. Ho
Affiliations:
N. Campbell
N. Campbell
Affiliations:
ECTRIMS Online Library. Koendgen H. 09/16/16; 145932; P1249
Harold Koendgen
Harold Koendgen
Contributions
Abstract

Abstract: P1249

Type: Poster

Abstract Category: Therapy - disease modifying - Risk management for disease modifying treatments

Background: Previous PML risk estimates in natalizumab-treated patients using postmarketing data were stratified by 3 established risk factors: presence of anti-JCV antibodies, prior immunosuppressant (IS) use, and treatment duration. As individual risk factor information was not available for all patients, population-based assumptions were needed. While essential since 2012 in guiding patient care, these estimates can now be made more precise using patient-level clinical trial data to assess risk associated with anti-JCV antibody levels ("index").

Objective: To derive annual index-stratified PML risk estimates from clinical trials stratified by the 3 risk factors and based on individual patient data.

Methods: Data were pooled from the STRATIFY-2 (N=24,402), TOP (N=5691), TYGRIS (N=6489), and STRATA (N=1094) trials. The life-table method provided forward-looking PML risk estimates for anti-JCV antibody positive patients in yearly epochs. Further index-stratified risk estimates for patients without prior IS were derived by combining these risk estimates with the probability distribution of index values in patients with/without PML using Bayes rule.

Results: The pooled cohort comprised 37,249 natalizumab-treated patients, including 156 patients with confirmed PML. Patients received a median (range) of 42 (1-129) natalizumab doses, 58% were anti-JCV antibody positive, and 14% had prior IS use. For patients with index ≤0.9, estimated PML risks (per 1000 patients) in yearly epochs 1, 2, 3, 4, 5, and 6 of natalizumab exposure were 0.1, 0.1, 0.2, 0.4, 0.5, 0.6, respectively; for those with index >0.9 to ≤1.5, respective estimated risks were 0.1, 0.3, 0.8, 2, 2, 3; for those with index >1.5, respective estimated risks were 0.2, 0.9, 3, 7, 8, 10; for those with prior IS use, respective estimated risks were 0.3, 0.4, 4, 8, 8, 6. Cumulative PML risk over time will be presented.

Conclusions: Although consistent with previous risk estimates from postmarketing data, these updated risk estimates achieve greater temporal precision in yearly treatment epochs using individual patient data and do not require population-based assumptions. By incorporating index in the algorithm, PML risk can be further stratified for anti-JCV antibody positive patients without prior IS use. In the prior algorithm, the largest numeric increase in PML risk occurred after >2 years of treatment; with increased precision using clinical trial data, the largest increase is observed after >3 years.

Disclosure:

Supported by Biogen



Harold Koendgen: employee of and may hold stock and/or stock options in Biogen.

Ih Chang: employee of and may hold stock and/or stock options in Biogen.

Bjoern Sperling: employee of and may hold stock and/or stock options in Biogen.

Gary Bloomgren: employee of and may hold stock and/or stock options in Biogen.

Bill Haddock: employee of and may hold stock and/or stock options in Biogen.

Sandra Richman: employee of and may hold stock and/or stock options in Biogen.

Pei-Ran Ho: employee of and may hold stock and/or stock options in Biogen.

Nolan Campbell: employee of and may hold stock and/or stock options in Biogen.

Abstract: P1249

Type: Poster

Abstract Category: Therapy - disease modifying - Risk management for disease modifying treatments

Background: Previous PML risk estimates in natalizumab-treated patients using postmarketing data were stratified by 3 established risk factors: presence of anti-JCV antibodies, prior immunosuppressant (IS) use, and treatment duration. As individual risk factor information was not available for all patients, population-based assumptions were needed. While essential since 2012 in guiding patient care, these estimates can now be made more precise using patient-level clinical trial data to assess risk associated with anti-JCV antibody levels ("index").

Objective: To derive annual index-stratified PML risk estimates from clinical trials stratified by the 3 risk factors and based on individual patient data.

Methods: Data were pooled from the STRATIFY-2 (N=24,402), TOP (N=5691), TYGRIS (N=6489), and STRATA (N=1094) trials. The life-table method provided forward-looking PML risk estimates for anti-JCV antibody positive patients in yearly epochs. Further index-stratified risk estimates for patients without prior IS were derived by combining these risk estimates with the probability distribution of index values in patients with/without PML using Bayes rule.

Results: The pooled cohort comprised 37,249 natalizumab-treated patients, including 156 patients with confirmed PML. Patients received a median (range) of 42 (1-129) natalizumab doses, 58% were anti-JCV antibody positive, and 14% had prior IS use. For patients with index ≤0.9, estimated PML risks (per 1000 patients) in yearly epochs 1, 2, 3, 4, 5, and 6 of natalizumab exposure were 0.1, 0.1, 0.2, 0.4, 0.5, 0.6, respectively; for those with index >0.9 to ≤1.5, respective estimated risks were 0.1, 0.3, 0.8, 2, 2, 3; for those with index >1.5, respective estimated risks were 0.2, 0.9, 3, 7, 8, 10; for those with prior IS use, respective estimated risks were 0.3, 0.4, 4, 8, 8, 6. Cumulative PML risk over time will be presented.

Conclusions: Although consistent with previous risk estimates from postmarketing data, these updated risk estimates achieve greater temporal precision in yearly treatment epochs using individual patient data and do not require population-based assumptions. By incorporating index in the algorithm, PML risk can be further stratified for anti-JCV antibody positive patients without prior IS use. In the prior algorithm, the largest numeric increase in PML risk occurred after >2 years of treatment; with increased precision using clinical trial data, the largest increase is observed after >3 years.

Disclosure:

Supported by Biogen



Harold Koendgen: employee of and may hold stock and/or stock options in Biogen.

Ih Chang: employee of and may hold stock and/or stock options in Biogen.

Bjoern Sperling: employee of and may hold stock and/or stock options in Biogen.

Gary Bloomgren: employee of and may hold stock and/or stock options in Biogen.

Bill Haddock: employee of and may hold stock and/or stock options in Biogen.

Sandra Richman: employee of and may hold stock and/or stock options in Biogen.

Pei-Ran Ho: employee of and may hold stock and/or stock options in Biogen.

Nolan Campbell: employee of and may hold stock and/or stock options in Biogen.

By clicking “Accept Terms & all Cookies” or by continuing to browse, you agree to the storing of third-party cookies on your device to enhance your user experience and agree to the user terms and conditions of this learning management system (LMS).

Cookie Settings
Accept Terms & all Cookies