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
. 2022 Mar 22:12:862526.
doi: 10.3389/fcimb.2022.862526. eCollection 2022.

A Predicted Model for Refractory/Recurrent Cytomegalovirus Infection in Acute Leukemia Patients After Haploidentical Hematopoietic Stem Cell Transplantation

Meng-Zhu Shen  1 , Shen-Da Hong  2 , Jie Wang  1   3 , Xiao-Hui Zhang  1 , Lan-Ping Xu  1 , Yu Wang  1 , Chen-Hua Yan  1 , Huan Chen  1 , Yu-Hong Chen  1 , Wei Han  1 , Feng-Rong Wang  1 , Jing-Zhi Wang  1 , Kai-Yan Liu  1 , Xiao-Jun Huang  1   4   5 , Xiao-Dong Mo  1   5
Affiliations

A Predicted Model for Refractory/Recurrent Cytomegalovirus Infection in Acute Leukemia Patients After Haploidentical Hematopoietic Stem Cell Transplantation

Meng-Zhu Shen et al. Front Cell Infect Microbiol. .

Abstract

Objective: We aimed to establish a model that can predict refractory/recurrent cytomegalovirus (CMV) infection after haploidentical donor (HID) hematopoietic stem cell transplantation (HSCT).

Methods: Consecutive acute leukemia patients receiving HID HSCT were enrolled (n = 289). We randomly selected 60% of the entire population (n = 170) as the training cohort, and the remaining 40% comprised the validation cohort (n = 119). Patients were treated according to the protocol registered at https://clinicaltrials.gov (NCT03756675).

Results: The model was as follows: Y = 0.0322 × (age) - 0.0696 × (gender) + 0.5492 × (underlying disease) + 0.0963 × (the cumulative dose of prednisone during pre-engraftment phase) - 0.0771 × (CD34+ cell counts in graft) - 1.2926. The threshold of probability was 0.5243, which helped to separate patients into high- and low-risk groups. In the low- and high-risk groups, the 100-day cumulative incidence of refractory/recurrent CMV was 42.0% [95% confidence interval (CI), 34.7%-49.4%] vs. 63.7% (95% CI, 54.8%-72.6%) (P < 0.001) for total patients and was 50.5% (95% confidence interval (CI), 40.9%-60.1%) vs. 71.0% (95% CI, 59.5%-82.4%) (P = 0.024) for those with acute graft-versus-host disease. It could also predict posttransplant mortality and survival.

Conclusion: We established a comprehensive model that could predict the refractory/recurrent CMV infection after HID HSCT.

Clinical trial registration: https://clinicaltrials.gov, identifier NCT03756675.

Keywords: cytomegalovirus; haploidentical donor; hematopoietic stem cell transplant; predicted model; refractory.

PubMed Disclaimer

Conflict of interest statement

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

Figures

Figure 1
Figure 1
Flow diagram of building the machine learning model.
Figure 2
Figure 2
Receiver operating characteristic (ROC) curve and confusion matrix for refractory/recurrent cytomegalovirus (CMV) infection model in the training (A) and validation cohorts (B).
Figure 3
Figure 3
The 100-day cumulative incidence of refractory/recurrent (A) and total (B) cytomegalovirus (CMV) infection in the low- and high-risk groups.
Figure 4
Figure 4
The 100-day cumulative incidence of refractory/recurrent cytomegalovirus (CMV) infection in patients without acute graft-versus-host disease (aGVHD) (A) and with aGVHD (B).
Figure 5
Figure 5
The 1-year cumulative incidence of relapse (A), non-relapse mortality (NRM; B), leukemia-free survival (LFS; C), and overall survival (OS; D) in the low- and high-risk groups.
See this image and copyright information in PMC

References

    1. Abu-Mostafa Y. S. (1989). The Vapnik-Chervonenkis Dimension: Information Versus Complexity in Learning. Neural Comput. 1 (3), 312–317. doi: 10.1162/neco.1989.1.3.312 - DOI
    1. Arcuri L. J., Schirmer M., Colares M., Maradei S., Tavares R., Moreira M. C. R., et al. . (2020). Impact of Anti-CMV IgG Titers and CD34 Count Prior to Hematopoietic Stem Cell Transplantation From Alternative Donors on CMV Reactivation. Biol. Blood Marrow Transplant. 26 (11), e275–e279. doi: 10.1016/j.bbmt.2020.07.034 - DOI - PubMed
    1. Red Blood Cell Disease (Anemia) Group, Chinese Society of Hematology, Chinese Medical Association. (2017). Chinese Expert Consensus on the Diagnosis and Treatment of Aplastic Anemia (2017). Zhonghua Xue Ye Xue Za Zhi 38 (1), 1–5. doi: 10.3760/cma.j.issn.0253-2727.2017.01.001 - DOI - PMC - PubMed
    1. Bittencourt H., Rocha V., Chevret S., Socié G., Espérou H., Devergie A., et al. . (2002). Association of CD34 Cell Dose With Hematopoietic Recovery, Infections, and Other Outcomes After HLA-Identical Sibling Bone Marrow Transplantation. Blood 99 (8), 2726–2733. doi: 10.1182/blood.V99.8.2726 - DOI - PubMed
    1. Blumer A., Ehrenfeucht A., Haussler D., Warmuth M. K. (1989). Learnability and the Vapnik-Chervonenkis Dimension. J. ACM. 36 (4), 929–865. doi: 10.1145/76359.76371 - DOI

Publication types

MeSH terms

Associated data