-
Categories
-
Pharmaceutical Intermediates
-
Active Pharmaceutical Ingredients
-
Food Additives
- Industrial Coatings
- Agrochemicals
- Dyes and Pigments
- Surfactant
- Flavors and Fragrances
- Chemical Reagents
- Catalyst and Auxiliary
- Natural Products
- Inorganic Chemistry
-
Organic Chemistry
-
Biochemical Engineering
- Analytical Chemistry
-
Cosmetic Ingredient
- Water Treatment Chemical
-
Pharmaceutical Intermediates
Promotion
ECHEMI Mall
Wholesale
Weekly Price
Exhibition
News
-
Trade Service
Prof.
Hui Zhou, Director of the Department of Lymphoma and Hematology, Hunan Cancer Hospital, Dr.
Selin Merdan, Department of Industrial and Systems Engineering, Georgia Institute of Technology, USA.
Risk prediction and immune infiltration profile of patients with diffuse large B-cell lymphoma based on gene expression profiles.
Research background Diffuse large B-cell lymphoma (DLBCL) relies on the International Prognostic Index (IPI) to identify high-risk patients in clinical risk stratification
.
The latest research shows that the immune microenvironment plays an important role in the prediction of DLBCL treatment response and survival
.
Research methods This study established a risk prediction model and evaluated the correlation between the biological significance of the model and immune infiltration
.
Research results We conducted gene expression profiling on 718 patients with DLBCL, in which RNA sequencing data and clinical trial covariates were derived from the study of Reddy (2017) et al
.
Using unsupervised and supervised machine learning methods to identify survival-related genetic features, a multivariate survival model was constructed
.
Use CIBERSORT deconvolution tool to analyze the composition of tumor infiltrating immune cells
.
A score based on four genetic characteristics was developed to divide patients into high-risk groups and low-risk groups
.
The combination of scoring based on gene expression and IPI improves the recognition rate of high-risk patients with DLBCL
.
These genetic features were successfully verified by the deconvolution tool
.
Correlating the deconvolution results with genetic markers and risk scores found that CD8+ T cells and naïve CD4+ T cells are associated with a good prognosis
.
Figure: Research conclusions on gene expression levels of survival prediction models.
By using systematic methods to analyze gene expression data, a risk prediction model that is superior to existing risk assessment methods has been established and verified
.
Expert opinion Professor Zhou Hui The author of this article collects RNA-seq data from DLBCL patients and analyzes their gene expression profiles, using unsupervised and supervised machine learning methods to identify survival-related gene features and construct multivariate survival The model developed a gene feature score based on the expression of four genes and combined with the IPI score to evaluate the patient's risk classification and prognosis, and calculated the composition of tumor immune infiltrating cells through CIBERSORT deconvolution analysis, and combined the results of CIBERSORT deconvolution with gene features and risks The scores are combined to obtain a robust survival prediction model, which is helpful for the prognosis assessment and risk stratification of DLBCL patients
.
Dialogue between Chinese and Foreign Scholars Professor Zhou Hui This article defines whether there are B cells as TME and tumor.
Can this method be listed in the DLBCL to support the definition method? The definition of using B cells is based on an article published by Reddy et al.
in Cell in 2017
.
1001 patients received a rituximab-based treatment regimen
.
The target of rituximab is CD20, which is an antigen expressed on B cells
.
We know that the DLBCL subtypes of all these patients are dominated by B cells, so CD20 is the first choice for treatment
.
Based on this, tumors can be divided into tumors (with B cells) and TME (without B cells)
.
All DLBCL samples used can be classified as GCB or ABC, both of which are based on B cells
.
Once again proved that B cells can be used to distinguish tumors from TME
.
Author Dr.
Selin Merdan, Professor Hui Zhou, in Figure 2, the first four columns have 22 immune cell data, while the last two columns only have 20 immune cell data, and the text does not explain the meaning of the data in the heat map.
Is it a ratio or a P value? Thank you for letting us notice this
.
We recently received a notice about an error in Figure 2, and we are contacting the publisher to update it
.
The heat map on Figure 2 shows the P value of the t-test calculated based on the log2 change
.
This is explained further in the "Evaluation of tumor-infiltrating immune cells" section of the article "Methods"
.
The original text is as follows: "The change in the proportion of immune cell subtypes between groups was evaluated by log2 changes, and poorer prognostic factors were used as a reference
.
Except for these clinical risk patients, patients with different survival outcomes (survival (Or death) The difference multiple of immune infiltration in the risk group was evaluated, where each cell subtype was calculated using log2 (basic average survival/basic average death)
.
" The author, Dr.
Selin Merdan, Professor Zhou Hui, uses WALD in the third part of the results.
The tested P<0.
1 is the cutoff value to obtain 1989 differential genes.
What is the basis for setting the P value? The article uses a similar method to that used by SSDave et al.
(Dave SS, et al.
Prediction of survival in follicular lymphoma based on molecular features of tumor-infiltrating immune cells.
[J].
The New England Journal of Medicine,2004,351( 21):2159.
)
.
As explained in that article, we deliberately chose a relatively free cut-off point of statistical significance, realizing that individual genes are potential noise in biological processes
.
The author Dr.
Selin Merdan, Professor Hui Zhou, why didn’t other GEO datasets be used as the validation set for the article? The article sample is large and multi-centered, so there is no need to compare with other GEO data sets
.
Author Dr.
Selin Merdan, Professor Hui Zhou.
The number of missing IPI scores in the data of this article is slightly higher.
Is there a slight lack of data integrity? The article has complete survival information of 718 patients, of which 593 patients have available IPI scores.
Although 125 patients have missing IPI scores, the sample size of the remaining patients is sufficient and more reliable statistical results can be obtained
.
The author Dr.
Selin Merdan, Professor Hui Zhou, the model includes a large amount of genes, is it possible to further search for core genes
?
Further research and analysis of the core genes involved are beyond the scope of our work
.
However, we think this issue is worthy of further discussion
.
The author Dr.
Selin Merdan, Professor Hui Zhou, this model is difficult to apply clinically, and would like to ask about the prospects of clinical application
.
Applying this integrated approach to clinical practice is challenging
.
If there is enough time, we plan to develop the survival prediction model into an online application, which requires clinicians to upload consistent data
.
Online algorithms will automatically summarize and quantify the data, and process them to obtain gene expression measurements
.
These gene expression measurements will be incorporated into the survival prediction formula developed
.
Then, the patients are classified into low or high risk categories through online algorithms, as shown in Figure 3 in the original text
.
Author Selin Merdan, Ph.
D.
Editing Perspective, Dr.
Martin Chopra, Editor-in-Chief of Oncology and Hematology Therapy of Springer Nature's Targeted Oncology and Adis series of journals Key Points: The authors of this study developed a transcription signature-based risk stratification model, and applying this risk score across different IPI risk groups demonstrated improved patient stratification.
The authors also derived information on the immune cell-composition of the tumor microenvironment from the gene expression data.
The gene expression-based score was independent of the IPI score in predicting overall survival, and the combination of the gene expression-based score and the IPI score improved discrimination on the entire set over the gene expression-based predictor score alone.
In settings where transcriptome sequencing is part of the standard assessment of patients with Diffuse Large B-cell Lymphoma,These results could therefore help improve patient stratification, and might contribute to a more risk-based treatment approach.
Article source: Merdan, S.
, Subramanian, K.
, Ayer, T.
et al.
Gene expression profiling-based risk prediction and profiles of immune infiltration in diffuse large B-cell lymphoma.
Blood Cancer J.
11, 2 (2021).
https://doi.
org/10.
1038/s41408-020-00404-0
Hui Zhou, Director of the Department of Lymphoma and Hematology, Hunan Cancer Hospital, Dr.
Selin Merdan, Department of Industrial and Systems Engineering, Georgia Institute of Technology, USA.
Risk prediction and immune infiltration profile of patients with diffuse large B-cell lymphoma based on gene expression profiles.
Research background Diffuse large B-cell lymphoma (DLBCL) relies on the International Prognostic Index (IPI) to identify high-risk patients in clinical risk stratification
.
The latest research shows that the immune microenvironment plays an important role in the prediction of DLBCL treatment response and survival
.
Research methods This study established a risk prediction model and evaluated the correlation between the biological significance of the model and immune infiltration
.
Research results We conducted gene expression profiling on 718 patients with DLBCL, in which RNA sequencing data and clinical trial covariates were derived from the study of Reddy (2017) et al
.
Using unsupervised and supervised machine learning methods to identify survival-related genetic features, a multivariate survival model was constructed
.
Use CIBERSORT deconvolution tool to analyze the composition of tumor infiltrating immune cells
.
A score based on four genetic characteristics was developed to divide patients into high-risk groups and low-risk groups
.
The combination of scoring based on gene expression and IPI improves the recognition rate of high-risk patients with DLBCL
.
These genetic features were successfully verified by the deconvolution tool
.
Correlating the deconvolution results with genetic markers and risk scores found that CD8+ T cells and naïve CD4+ T cells are associated with a good prognosis
.
Figure: Research conclusions on gene expression levels of survival prediction models.
By using systematic methods to analyze gene expression data, a risk prediction model that is superior to existing risk assessment methods has been established and verified
.
Expert opinion Professor Zhou Hui The author of this article collects RNA-seq data from DLBCL patients and analyzes their gene expression profiles, using unsupervised and supervised machine learning methods to identify survival-related gene features and construct multivariate survival The model developed a gene feature score based on the expression of four genes and combined with the IPI score to evaluate the patient's risk classification and prognosis, and calculated the composition of tumor immune infiltrating cells through CIBERSORT deconvolution analysis, and combined the results of CIBERSORT deconvolution with gene features and risks The scores are combined to obtain a robust survival prediction model, which is helpful for the prognosis assessment and risk stratification of DLBCL patients
.
Dialogue between Chinese and Foreign Scholars Professor Zhou Hui This article defines whether there are B cells as TME and tumor.
Can this method be listed in the DLBCL to support the definition method? The definition of using B cells is based on an article published by Reddy et al.
in Cell in 2017
.
1001 patients received a rituximab-based treatment regimen
.
The target of rituximab is CD20, which is an antigen expressed on B cells
.
We know that the DLBCL subtypes of all these patients are dominated by B cells, so CD20 is the first choice for treatment
.
Based on this, tumors can be divided into tumors (with B cells) and TME (without B cells)
.
All DLBCL samples used can be classified as GCB or ABC, both of which are based on B cells
.
Once again proved that B cells can be used to distinguish tumors from TME
.
Author Dr.
Selin Merdan, Professor Hui Zhou, in Figure 2, the first four columns have 22 immune cell data, while the last two columns only have 20 immune cell data, and the text does not explain the meaning of the data in the heat map.
Is it a ratio or a P value? Thank you for letting us notice this
.
We recently received a notice about an error in Figure 2, and we are contacting the publisher to update it
.
The heat map on Figure 2 shows the P value of the t-test calculated based on the log2 change
.
This is explained further in the "Evaluation of tumor-infiltrating immune cells" section of the article "Methods"
.
The original text is as follows: "The change in the proportion of immune cell subtypes between groups was evaluated by log2 changes, and poorer prognostic factors were used as a reference
.
Except for these clinical risk patients, patients with different survival outcomes (survival (Or death) The difference multiple of immune infiltration in the risk group was evaluated, where each cell subtype was calculated using log2 (basic average survival/basic average death)
.
" The author, Dr.
Selin Merdan, Professor Zhou Hui, uses WALD in the third part of the results.
The tested P<0.
1 is the cutoff value to obtain 1989 differential genes.
What is the basis for setting the P value? The article uses a similar method to that used by SSDave et al.
(Dave SS, et al.
Prediction of survival in follicular lymphoma based on molecular features of tumor-infiltrating immune cells.
[J].
The New England Journal of Medicine,2004,351( 21):2159.
)
.
As explained in that article, we deliberately chose a relatively free cut-off point of statistical significance, realizing that individual genes are potential noise in biological processes
.
The author Dr.
Selin Merdan, Professor Hui Zhou, why didn’t other GEO datasets be used as the validation set for the article? The article sample is large and multi-centered, so there is no need to compare with other GEO data sets
.
Author Dr.
Selin Merdan, Professor Hui Zhou.
The number of missing IPI scores in the data of this article is slightly higher.
Is there a slight lack of data integrity? The article has complete survival information of 718 patients, of which 593 patients have available IPI scores.
Although 125 patients have missing IPI scores, the sample size of the remaining patients is sufficient and more reliable statistical results can be obtained
.
The author Dr.
Selin Merdan, Professor Hui Zhou, the model includes a large amount of genes, is it possible to further search for core genes
?
Further research and analysis of the core genes involved are beyond the scope of our work
.
However, we think this issue is worthy of further discussion
.
The author Dr.
Selin Merdan, Professor Hui Zhou, this model is difficult to apply clinically, and would like to ask about the prospects of clinical application
.
Applying this integrated approach to clinical practice is challenging
.
If there is enough time, we plan to develop the survival prediction model into an online application, which requires clinicians to upload consistent data
.
Online algorithms will automatically summarize and quantify the data, and process them to obtain gene expression measurements
.
These gene expression measurements will be incorporated into the survival prediction formula developed
.
Then, the patients are classified into low or high risk categories through online algorithms, as shown in Figure 3 in the original text
.
Author Selin Merdan, Ph.
D.
Editing Perspective, Dr.
Martin Chopra, Editor-in-Chief of Oncology and Hematology Therapy of Springer Nature's Targeted Oncology and Adis series of journals Key Points: The authors of this study developed a transcription signature-based risk stratification model, and applying this risk score across different IPI risk groups demonstrated improved patient stratification.
The authors also derived information on the immune cell-composition of the tumor microenvironment from the gene expression data.
The gene expression-based score was independent of the IPI score in predicting overall survival, and the combination of the gene expression-based score and the IPI score improved discrimination on the entire set over the gene expression-based predictor score alone.
In settings where transcriptome sequencing is part of the standard assessment of patients with Diffuse Large B-cell Lymphoma,These results could therefore help improve patient stratification, and might contribute to a more risk-based treatment approach.
Article source: Merdan, S.
, Subramanian, K.
, Ayer, T.
et al.
Gene expression profiling-based risk prediction and profiles of immune infiltration in diffuse large B-cell lymphoma.
Blood Cancer J.
11, 2 (2021).
https://doi.
org/10.
1038/s41408-020-00404-0
