Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2021 Jul;27(4):e441-e449.
doi: 10.1111/hae.14325. Epub 2021 May 14.

Performance of a clinical risk prediction model for inhibitor formation in severe haemophilia A

Shermarke Hassan  1   2 , Roberta Palla  2 , Carla Valsecchi  3 , Isabella Garagiola  3 , Amal El-Beshlawy  4 , Mohsen Elalfy  5 , Vijay Ramanan  6 , Peyman Eshghi  7 , Mehran Karimi  8 , Samantha Claudia Gouw  1   9 , Pier Mannuccio Mannucci  2   3 , Frits Richard Rosendaal  1 , Flora Peyvandi  2   3 ; SIPPET study group
Affiliations

Performance of a clinical risk prediction model for inhibitor formation in severe haemophilia A

Shermarke Hassan et al. Haemophilia. 2021 Jul.

Abstract

Background: There is a need to identify patients with haemophilia who have a very low or high risk of developing inhibitors. These patients could be candidates for personalized treatment strategies.

Aims: The aim of this study was to externally validate a previously published prediction model for inhibitor development and to develop a new prediction model that incorporates novel predictors.

Methods: The population consisted of 251 previously untreated or minimally treated patients with severe haemophilia A enrolled in the SIPPET study. The outcome was inhibitor formation. Model discrimination was measured using the C-statistic, and model calibration was assessed with a calibration plot. The new model was internally validated using bootstrap resampling.

Results: Firstly, the previously published prediction model was validated. It consisted of three variables: family history of inhibitor development, F8 gene mutation and intensity of first treatment with factor VIII (FVIII). The C-statistic was 0.53 (95% CI: 0.46-0.60), and calibration was limited. Furthermore, a new prediction model was developed that consisted of four predictors: F8 gene mutation, intensity of first treatment with FVIII, the presence of factor VIII non-neutralizing antibodies before treatment initiation and lastly FVIII product type (recombinant vs. plasma-derived). The C-statistic was 0.66 (95 CI: 0.57-0.75), and calibration was moderate. Using a model cut-off point of 10%, positive- and negative predictive values were 0.22 and 0.95, respectively.

Conclusion: Performance of all prediction models was limited. However, the new model with all predictors may be useful for identifying a small number of patients with a low risk of inhibitor formation.

Keywords: factor VIII; haemophilia A; immunogenicity; inhibitors; prediction.

PubMed Disclaimer

Conflict of interest statement

SH, CV, IG, AE, ME, VR, PE, MK and FRR have no interests to disclose. RP has received travel support from Pfizer and Kedrion and honoraria for participating as a speaker at educational symposia organized by Novo Nordisk. SG has received an unrestricted research grant from Sobi. PMM is a member of the scientific board for the Bayer Awards. He has also received from Bayer, Kedrion and Novo Nordisk honoraria for lectures at educational symposia. FP has received honoraria for participating as a speaker at satellite symposia organized by Bioverativ, Grifols, Roche, Sanofi, Sobi, Spark and Takeda. F.P. reports participation at advisory board of Sanofi and Sobi.

Figures

FIGURE 1
FIGURE 1
Calibration plot of 2015 model and new model. The figure shows the calibration plot of the 2015 model (A) and of the new model (B). On the X‐axis, the predicted probability of inhibitor formation according to the model is plotted against the observed risk on the Y‐axis. (0 = no inhibitor, 1 = developed an inhibitor) The blue dots represent the proportion of patients experiencing an event, stratified by quintiles of increasing predicted risk. Quintiles with a higher predicted inhibitor risk should have a higher proportion of patients who develop the outcome (ie, a higher observed risk). Alternatively, a LOWESS (Locally Weighted Scatter plot Smoothing) line was estimated to examine calibration across the whole range (shown here as a red dotted line). The grey line represents perfect prediction, meaning that the predicted risk is exactly the same as the observed risk across the whole range. Ideally, both the quintiles and the LOWESS line should lie exactly on top of the grey line
See this image and copyright information in PMC

References

    1. Peyvandi F, Garagiola I, Young G. The past and future of haemophilia: diagnosis, treatments, and its complications. Lancet. 2016;388(10040):187‐197. - PubMed
    1. Mannucci PM, Mancuso ME, Santagostino E, Franchini M. Innovative pharmacological therapies for the hemophilias not based on deficient factor replacement. Semin Thromb Hemost. 2016;42(5):526‐532. - PubMed
    1. ter Avest PC, Fischer K, Mancuso ME, et al. Risk stratification for inhibitor development at first treatment for severe hemophilia A: a tool for clinical practice. J Thromb Haemost. 2008;6(12):2048‐2054. - PubMed
    1. Hashemi SM, Fischer K, Moons KG, van den Berg HM. Improved prediction of inhibitor development in previously untreated patients with severe haemophilia A. Haemophilia. 2015;21(2):227‐233. - PubMed
    1. Peyvandi F, Mannucci PM, Garagiola I, et al. A randomized trial of factor VIII and neutralizing antibodies in hemophilia A. N Engl J Med. 2016;374(21):2054‐2064. - PubMed

LinkOut - more resources