Search

e18003 Background: Head and neck squamous cell carcinoma (HNSCC) is often diagnosed at an advanced stage and has a dismal prognosis. Anti-PD1/PD-L1 checkpoint inhibitors have shown survival benefit for the treatment on HNSCC, and are better tolerated than chemotherapy. Unfortunately, the response rate of immunotherapy is low for advanced HNSCC. Therefore, the identification of predictive biomarkers of response to anti-PD1/PD-L1 is a key point for better selecting patients that would benefit from immunotherapy. Methods: The present study included 44 patients with advanced HNSCC who received anti-PD-1 immunotherapy. Follow-up was available for them. Primary tumor tissues and matched blood samples were collected. Genome profiles were analyzed using a designed 543-gene panel based on NGS, and were available on 39 of 44 patients. Profiles of infiltrating immune cells were available for 44 patients. The infiltrating intensity of PD-1, PD-L1, CD8, CD68 and CD57 positive cells were assessed by multiplex immunohistochemistry. Results: The most frequently mutated genes were TP53 (79%), CDKN2A (33%), NOTCH1 (21%), CASP8 (13%) and PIK3CA (10%). Patients with mutations in the PI3K pathway showed a trend of longer OS (P = 0.06). Moreover, the TMB-high group showed significant superior OS than the TMB-low group (p = 0.0018). PD-L1 expression was positively correlated with CD8 expression within stroma, and PD-1 expression correlated significantly with PD-L1 and CD8 expression within both tumor and stroma areas. In addition, tumor/stromal PD-L1 and PD-1 expression were both associated with better OS (P < 0.05). Prolonged OS was also observed in patients with more tumor and stromal infiltration of CD8+ cells, CD8+PD-L1+ cells, CD68+ macrophages, or nature killer (NK, CD57+) cells. Conclusions: Higher TMB, PD-L1 expression, as well as the infiltration of immune cells within both tumor and stromal area might serve as favorable prognostic markers in advanced HNSCC patients treated with anti-PD-1 immunotherapy.

e20580 Background: Small cell lung cancer (SCLC) is an aggressive neuroendocrine subtype of lung cancer with high mortality. Recently, Immune checkpoint inhibitors (ICIs) have been shown to have the potential to improve the prognosis of SCLC, but little is known about immunotherapeutic biomarkers. Homologous recombination deficiency (HRD) is demonstrated to be a response predictor to immunotherapies in gynecologic cancers, while limited studies were reported in small cell lung cancer. Herein, we analyze the mutational pattern of HRR related genes in a Chinese SCLC cohort and further analyze the relationship between HRR-gene mutations and tumor mutational burden. Methods: Target gene sequencing (543 genes) was performed in 133 Genecast cohort with small cell lung cancer. PD-L1 expression were evaluated for 90 among 133 patients using the SP142 PD-L1 immunohistochemistry assay. Results: Among 133 patients, 47 (35.3%) had HRR-gene mutations. ATM (8.3%), NBN (4.5%) and BRCA2 (4.5%) were the top 3 mutated HRR-gene in the cohort，followed by ATR (3.8%), BARD1 (3.8%), BRCA1 (3.8%), PALB2 (3.8%), RAD50 (3.8%), CHEK2 (3.0%), BLM (3.0%), BRIP1(2.3%), CHEK1(1.5%), RAD52(1.5%), and MRE11A (0.8%). Pathogenic somatic and germline mutations of HRR genes were identified in 11 (11/47, 23.4%) and 3 (3/47, 6.4%) patients, respectively. 1 (1/47, 2.1%) patient carried both germline and somatic variants. Genomic landscape revealed that TP53 and RB1 were commonly mutated genes in SCLC cohort. Mutations in KMT2D, AR and RTK-RAS pathway occurred more frequently in the HRR-Mut group, compared with the wildtype ones. Furthermore, we found that mutations in HRR-gene were associated with high TMB (Wilcoxon, p = 0.048), and patients with high TMB (≥median) showed a higher proportion of positive PD-L1 expression in 90 SCLC patients. Conclusions: Our data indicated that genomic alterations associated with HRR-genes have a positive correlation with high TMB, and detection of HRR-gene mutation status probably could help identify patients who might benefit from immune checkpoint blockade therapy. Keywords: Small cell lung cancer, Homologous recombination deficiency, Immunotherapy.

e15552 Background: As the fifth most common malignancies and the third leading cause of cancer death in the world, gastric cancer (GC) is often diagnosed in locally advanced or metastatic stage and, therefore, has a poor prognosis. Copy number instability (CNI) score, interpreted as the general measure of genomic instability, was reported as a prognosis predictor of some types of cancer, but its role in prognosis of GC patients remains unknown. The aim of the present study was to investigate its prognostic value in patients with GC after surgical resection and adjuvant chemotherapy. Methods: The present study included 120 patients who had received gastrectomy and adjuvant chemotherapy with stage II-IV GC. Follow-up was available for them. DNA was extracted from primary formalin-fixed paraffin-embedded (FFPE) tumor specimens and matched blood samples. Genomic profiles were analyzed by using a designed panel (1408 genes) based on next generation sequencing (NGS). Results: The most frequently mutated genes were TP53 (49%), PIK3CA (8%), RNF43 (6%) ERBB3 (6%) and APC (6%), and the most frequently amplified genes were TRPS1 (41%), COL1A2 (31%), CSMD3 (29%), ZFHX4 (29%), NAV3 (23%). CNI score was negatively correlated with OS, the CNI-high group had markedly shorter OS than CNI-low group ( p= 0.0093). In addition, there were statistically significant differences in OS between different clinical staging ( p= 0.0468). Moreover, the Cox proportional hazard model showed that CNI score was an independent prognosis factor in GC. Conclusions: The current study indicates that CNI score in primary tumor tissue is an independent predictive prognostic biomarker for GC.

e15555 Background: Gastric cancer (GC) is associated with high morbidity and mortality rates in the world with poor prognosis and limited treatment options. The level of copy number alteration (CNA), termed CNA burden, is reported as a pan-cancer prognostic factor associated with recurrence and death. The current study aims to investigate association between CNA burden in primary tumor tissue and overall survival (OS) of Chinese patients with GC after surgical resection. Methods: The present study included 78 patients who had received surgical resection and adjuvant chemotherapy with poorly differentiated GC. The primary outcome was OS. Tumor specimens were obtained from surgery and submitted for next generation sequencing (NGS) with matched normal tissue samples. A 1408-gene panel was used to identify genome profiles. Data were analyzed using Cox proportional hazards models and Kaplan-Meier survival analysis. Results: The most frequently altered genes were TP53 (47%), PIK3CA (10%), PTEN (9%), NOTCH1 (8%) and RNF43 (6%), and copy numbers of TRPS1 (65%), COL1A2 (50%), CSMD3 (45%), ZFHX4 (45%), NAV3 (36%) varied most frequency in current cohort. Greater tumor CNA burden correlated with an increase in death from disease compared to a lower tumor CNA burden ( p= 0.0066). In addition, there were statistically significant differences in OS between different clinical staging ( p= 0.0011). Moreover, the Cox proportional hazard model showed that CNA burden was an independent prognosis factor in GC. Finally, we performed an independent signature that includes CNA burden and clinical staging to predict survival of GC. Conclusions: This study indicates that tumor CNA burden is an independent predictive survival biomarker for Chinese gastric cancers. CNA burden combined with clinical staging is a better predictor for postoperative survival prediction of gastric cancer.

e15730 Background: Pancreatic cancer is one of the most aggressive human malignancies and has a poor prognosis. Large-scale studies have reported on its initiation, progression, diagnosis and prognosis. However, over the past several decades, the overall 5-year survival rate of pancreatic cancer has remained at less than 5%. Thus, the current major challenge in the postoperative management of pancreatic cancer is to identify high-risk recurrent patients. We retrospectively investigated the clinical, pathological and outcome data for 48 pancreatic cancer patients to clarify the associations of molecular mechanisms and prognosis. Methods: Eligible pancreatic cancer patients were included, and formalin-fixed paraffin-embedded (FFPE) tumor specimens and matched blood samples were collected at Ruijin Hospital, Shanghai Jiao tong University School of Medicine. Genome profiles were analyzed by using a designed 1408-gene panel based on next-generation sequencing (NGS). The copy number instability (CNI) score in primary tumor tissue was calculated to investigate the relationship between molecular features of the primary tumor and prognosis. Results: The CNI score in primary tumor tissue was positively correlated with lymph node metastasis, TP53 mutation, and early recurrence. Moreover, preoperative Carbohydrate antigen 19-9 (CA19-9) levels and CNI scores in primary tumor tissue were significant independent predictors associated with PFS in pancreatic ductal adenocarcinoma (PDAC). Finally, we performed an independent signature that includes CNI score and Carbohydrate antigen 19-9 (CA19-9) level to predict prognosis of pancreatic cancer. Conclusions: These results suggest that CNI score in primary tumor tissue is an independent predictive prognostic biomarker for PDAC. CNI combined with CA19-9 is a better predictor for postoperative prognostic prediction of pancreatic cancer.