Abstract 25P
Background
The Cancer Genome Atlas (TCGA) Project defined four prognostic subgroups of Endometrial Cancer (EC): POLE (favorable prognosis), MSI and Copy Number Low (CNL) the intermediate prognosis; and Copy Number High (CNH) unfavorable prognosis. The aim of the study is to perform a characterization at immune level of the prognostic EC-TCGA groups to identify those cases that could be benefited of immunotherapy.
Methods
A total of 48 FFPE EC were retrospectively selected (Ref): POLE (n=6); MSI (n=9); CNL (n=16) and CNH (n=17). Transcriptomic profiling was performed with HTG EdgeSeq Precision Immuno-Oncology Panel (PIO), which interrogates 1392 genes involved in tumor/immune interaction. The estimation of the relative abundance of immune and stromal cellular content and cell types was assigned to the 23 HTG EdgeSeq Reveal software immunophenotyping signatures. DESeq2 R package was used for differential expression analysis. Statistical analysis and data visualization was performed in R version 4.1.2. Clinicopathological information is collected in the table.
Results
Unsupervised analysis showed 2 clusters with different prognosis in terms of OS (p=0.0033) and DFS (p=0.00055). Cluster 1 was constituted by 8 MSI, 6 POLE, 14 CNL and 7 CNH while cluster 2 was mainly formed by CNH (10), followed by 1 MSI and 2 CNL. Differential expression analysis resulted in 897 genes between clusters, 400 overexpressed and 497 under expressed in Cluster 1. The 6 most significant genes were TGFB1, BEX1, CDK4, TUBB, RFC4 and TRIP13. In addition, cluster 1 was related with higher PD-1 (p=4.4·10-4), PD-L1(p=9.9·10-6) and CTL4 (p=4·10-4) expression and immune related scores (immune; p=0.00019 and stroma; p=0.00014). Table: 25P
| Cluster 1 | Cluster 2 | Total | ||
| No of patients (%) | 35 (73) | 13 (27) | 48 (100) | |
| Histological type (%) | Endometrial | 31 (65) | 9 (19) | 40 (84) |
| Serous | 4 (8) | 4 (8) | 8 (16) | |
| Grade (%) | 1 | 17 (35) | 6 (13) | 23 (48) |
| 2 | 11 (23) | 3 (6) | 14 (29) | |
| 3 | 7 (15) | 4 (8) | 11 (23) | |
| Stage (%) | I | 25 (52) | 10 (21) | 35 (73) |
| II | 1 (2) | 0 (0) | 1 (2) | |
| III | 9 (19) | 3 (6) | 12 (25) | |
| TCGA Group (%) | POLE | 6 (12.5) | 0 (0) | 6 (12.5) |
| MSI | 8 (16.7) | 1 (2.1) | 9 (18.8) | |
| CNL | 14 (29.2) | 2 (4.2) | 16 (33.3) | |
| CNH | 7 (14.6) | 7 (14.6) | 14 (29.2) | |
| Median follow-up [range] months | 89,9 [5-152] | |||
| Median OS | 101,1 [26-152] | 85,33 [6-143] | 92,02 [6-152] | |
| Median DFS | 101,1 [10-152] | 81,17 [5-105] | 89,87 [5-152] | |
| Relapse (%) | 4/35 (11,4) | 7/13 (53,8) | 11/48 (22,9) | |
| Éxitus (%) | 1/35 (2,9) | 4/13 (30,8) | 5/48 (10,4) |
Conclusions
Cluster stratification suggests implication of immune-related features in classification of EC-patients beyond TCGA subgroups. These findings could be useful in the clinical management of the disease, constituting an open window in the selection of EC patients for immunotherapy.
Legal entity responsible for the study
The authors.
Funding
Has not received any funding.
Disclosure
All authors have declared no conflicts of interest.