scholarly journals Predicting the Prognosis of Non-Small Cell Lung Cancer by Integrating Microarray and Clinical Data with Deep Learning

2019 ◽  
Author(s):  
Yu-Heng Lai ◽  
Wei-Ning Chen ◽  
Te-Cheng Hsu ◽  
Che Lin ◽  
Yu Tsao ◽  
...  

AbstractNon-small cell lung cancer (NSCLC) is one of the most common lung cancers worldwide. Accurate prognostic stratification of NSCLC can become an important clinical reference when designing therapeutic strategies for cancer patients. With this clinical application in mind, we developed a deep neural network (DNN) combining heterogeneous data sources of gene expression and clinical data to accurately predict the prognosis of NSCLC patients. Based on microarray data from a cohort set (614 patients), seven well-known NSCLC markers were used to group patients into marker- and marker+ subgroups. Using a systems biology approach, prognosis relevance values (PRV) were then calculated to select eight additional novel prognostic gene markers. Gene markers along with clinical data were then used to develop an integrative DNN via bimodal learning to predict the 5-year survival rate of NSCLC patients with tremendously high accuracy (AUC: 0.8163, accuracy: 75.44%), which is superior to all other existing methods based on AUC. Using the capability of deep learning, we believe that our predicted cancer prognosis can be a promising index helping oncologists and physicians develop personalized therapy and build the foundation of precision medicine in the future.

2022 ◽  
Vol 11 ◽  
Author(s):  
Jiexia Zhang ◽  
Shuangfeng Tang ◽  
Chunning Zhang ◽  
Mingyao Li ◽  
Yating Zheng ◽  
...  

BackgroundPALB2, a gene in the homologous recombination repair (HRR) pathway of the DNA damage response (DDR), is associated with the efficacy of platinum-based chemotherapy, immunotherapy, and PARP inhibitor therapy in several tumors. However, the PALB2 characteristics, its correlation with immunotherapy biomarker, and the prognostic effect of immunotherapy in non-small cell lung cancer (NSCLC) were unknown.MethodsTumor tissue samples from advanced Chinese NSCLC patients were analyzed by next-generation sequencing (NGS) (panel on 381/733-gene). Tumor mutation burden (TMB) is defined as the total number of somatic non-synonymous mutations in the coding region. Microsatellite instability (MSI) was evaluated by NGS of 500 known MSI loci. Programmed Cell Death-Ligand 1 (PD-L1) expression was evaluated using immunohistochemistry (Dako 22C3 or SP263). One independent cohort (Rizvi2018.NSCLC.240.NGS cohort) containing genomic and clinical data from 240 patients with advanced NSCLC and two cohorts (the OAK and POPLAR study cohort) containing genomic and clinical data from 429 patients with advanced NSCLC were used to analyze the prognostic effect of PALB2 on immunotherapy.ResultsGenetic mutation of 5,227 NSCLC patients were analyzed using NGS, of which 162 (3.1%) harbored germline PALB2 mutation (PALB2gmut) and 87 (1.66%) harbored somatic PALB2 mutation (PALB2smut). In NSCLC patients with PALB2gmut and PALB2smut, the most frequently mutated gene was TP53 (65%, 64%). PALB2smut (14.52 Muts/Mb) was associated with higher TMB (p < 0.001) than PALB wild-type (PALB2wt) (6.15 Muts/Mb). However, there was no significant difference in TMB between PALB2gmut (6.45 Muts/Mb) and PALB2wt (6.15 Muts/Mb) (p = 0.64). There was no difference in PD-L1 expression among PALB2gmut, PALB2smut, and PALB2wt. In the Rizvi2018.NSCLC.240.NGS cohort, there was no difference in progression-free survival (PFS) (HR =1.06, p = 0.93) between PALB2 mutation (3.15 months) and PALB2wt (3.17 months). The OAK and POPLAR study cohort of NSCLC patients showed that there was no difference in overall survival (OS) (HR =1.1, p = 0.75) between PALB2 mutation (10.38 months) and PALB2wt (11.07 months).ConclusionsThese findings suggest that PALB2 may not be used as a biomarker for determining prognosis on immunotherapy in NSCLC.


Cancers ◽  
2021 ◽  
Vol 13 (15) ◽  
pp. 3828
Author(s):  
Anello Marcello Poma ◽  
Rossella Bruno ◽  
Iacopo Pietrini ◽  
Greta Alì ◽  
Giulia Pasquini ◽  
...  

Pembrolizumab has been approved as first-line treatment for advanced Non-small cell lung cancer (NSCLC) patients with tumors expressing PD-L1 and in the absence of other targetable alterations. However, not all patients that meet these criteria have a durable benefit. In this monocentric study, we aimed at refining the selection of patients based on the expression of immune genes. Forty-six consecutive advanced NSCLC patients treated with pembrolizumab in first-line setting were enrolled. The expression levels of 770 genes involved in the regulation of the immune system was analysed by the nanoString system. PD-L1 expression was evaluated by immunohistochemistry. Patients with durable clinical benefit had a greater infiltration of cytotoxic cells, exhausted CD8, B-cells, CD45, T-cells, CD8 T-cells and NK cells. Immune cell scores such as CD8 T-cell and NK cell were good predictors of durable response with an AUC of 0.82. Among the immune cell markers, XCL1/2 showed the better performance in predicting durable benefit to pembrolizumab, with an AUC of 0.85. Additionally, CD8A, CD8B and EOMES showed a high specificity (>0.86) in identifying patients with a good response to treatment. In the same series, PD-L1 expression levels had an AUC of 0.61. The characterization of tumor microenvironment, even with the use of single markers, can improve patients’ selection for pembrolizumab treatment.


Cancers ◽  
2021 ◽  
Vol 13 (8) ◽  
pp. 1794
Author(s):  
Alice Indini ◽  
Erika Rijavec ◽  
Francesco Grossi

Immune checkpoint inhibitors (ICIs) targeting the programmed cell death (PD)-1 protein and its ligand, PD-L1, and cytotoxic T-lymphocyte-associated antigen (CTLA)-4, have revolutionized the management of patients with advanced non-small cell lung cancer (NSCLC). Unfortunately, only a small portion of NSCLC patients respond to these agents. Furthermore, although immunotherapy is usually well tolerated, some patients experience severe immune-related adverse events (irAEs). Liquid biopsy is a non-invasive diagnostic procedure involving the isolation of circulating biomarkers, such as circulating tumor cells (CTC), cell-free DNA (cfDNA), and microRNAs (miRNAs). Thanks to recent advances in technologies, such as next-generation sequencing (NGS) and digital polymerase chain reaction (dPCR), liquid biopsy has become a useful tool to provide baseline information on the tumor, and to monitor response to treatments. This review highlights the potential role of liquid biomarkers in the selection of NSCLC patients who could respond to immunotherapy, and in the identification of patients who are most likely to experience irAEs, in order to guide improvements in care.


2020 ◽  
Vol 31 ◽  
pp. S851
Author(s):  
C. Dellepiane ◽  
S. Coco ◽  
M.G. Dal Bello ◽  
G. Rossi ◽  
E. Rijavec ◽  
...  

Sign in / Sign up

Export Citation Format

Share Document