Title:An Innovative Telomere-associated Prognosis Model in AML: Predicting Immune Infiltration and Treatment Responsiveness
Volume: 33
Issue: 1
Author(s): Binyang Song, Jinzhan Lou, Lijun Mu, Xiao Lu, Jian Sun*Bo Tang*
Affiliation:
- Department of Hematology, The Second Affiliated Hospital of Dalian Medical University, Dalian, 116023, People's Republic of China
- Department of Hematology, The Second Affiliated Hospital of Dalian Medical University, Dalian, 116023, People's Republic of China
Keywords:
Acute myeloid leukemia, telomere-related genes, prognostic model, immune infiltration, drug sensitivity, OBFC1, SMN1.
Abstract:
Aims: To build an innovative telomere-associated scoring model to predict
prognosis and treatment responsiveness in acute myeloid leukemia (AML).
Background: AML is a highly heterogeneous malignant hematologic disorder with a
poor prognosis. While telomere maintenance is frequently observed in tumors, investigations
into telomere-related genes (TRGs) in AML remain limited.
Objectives: This study aimed to identify prognostic TRGs using the least absolute shrinkage
and selection operator (LASSO) Cox regression and multivariate Cox regression,
evaluate their predictive value, explore the association between TRG scores and immune
cell infiltration, and assess the sensitivity of high-scoring AML patients to chemotherapeutic
agents.
Methods: Univariate Cox regression analysis was conducted on the TCGA cohort to
identify prognostic TRGs and to develop the TRG scoring model using LASSO-Cox
and multivariate Cox regression. Validation was performed on the GSE37642 cohort. Immune
cell infiltration patterns were assessed through computational analysis, and the sensitivity
to chemotherapeutic agents was evaluated.
Results: Thirteen prognostic TRGs were identified, and a seven-TRG scoring model (including
NOP10, OBFC1, PINX1, RPA2, SMG5, MAPKAPK5, and SMN1) was developed.
Higher TRG scores were associated with a poorer prognosis, as confirmed in the
GSE37642 cohort, and remained an independent prognostic factor even after adjusting
for other clinical characteristics. The high-score group was characterized by elevated infiltration
of B cells, T helper cells, natural killer cells, tumor-infiltrating lymphocytes,
regulatory T (Treg) cells, M2 macrophages, neutrophils, and monocytes, along with reduced
infiltration of gamma delta T cells, CD4- T cells, and resting mast cells. Moreover,
high infiltration of M2 macrophages and Tregs was associated with poor overall
survival compared to low infiltration. Notably, high-risk AML patients were resistant to
Erlotinib, Parthenolide, and Nutlin-3a, but sensitive to AC220, Midostaurin, and Tipifarnib.
Additionally, using RT-qPCR, we observed significantly higher expression of
two model genes, OBFC1 and SMN1, in AML tissues compared to control tissues.
Conclusion: This innovative TRG scoring model demonstrates considerable predictive
value for AML patient prognosis, offering valuable insights for optimizing treatment
strategies and personalized medicine approaches. The identified TRGs and associated
scoring models could aid in risk stratification and guide tailored therapeutic interventions
in AML patients.
