All times are listed in CET (Central European Time)

Browsing Over 213 Presentations

Sort By Upcoming First
Tumour biology and pathology

Tumour biology and pathology

Room
ePoster Display
Tumour biology and pathology

1P - Real-world Frequency of Non-Small Cell Lung Cancer with Epidermal Growth Factor Receptor (EGFR) Exon 20 Insertion (Exon20ins) Mutations by Site of Insertion

Room
ePoster Display
Speakers
  • S. Viteri (Barcelona, Spain)
Authors
  • S. Viteri (Barcelona, Spain)
  • J. Bauml (Philadelphia, United States of America)
  • L. Bazhenova (La Jolla, CA, United States of America)
  • S. Ou (Orange, AL, United States of America)
  • N. Girard (Paris, France)
  • M. Schaffer (Spring House, AL, United States of America)
  • J. Rose (Spring House, PA, United States of America)
  • J. Curtin (Spring House, PA, United States of America)
  • J. Karkera (Spring House, PA, United States of America)
  • P. Mahadevia (Raritan, United States of America)
  • A. Minchom (London, Surrey, United Kingdom)

Abstract

Background

EGFR Exon20ins mutations are in-frame insertions or duplications clustered around 3 key regions of Exon 20: helical (amino acid [AA] 762–766), near loop (767–772), and far loop (773–775). The site of insertion can influence the outcome to EGFR tyrosine kinase inhibitor therapy; loop insertions are typically resistant while some helical insertions are sensitive (Yasuda Lancet Oncol 13:23; Sci Transl Med 5:216ra177). We used real-world US genomic data from patients (pts) with lung adenocarcinoma (LUAD) to determine the frequency of EGFR Exon20ins mutations by site of insertion.

Methods

Pts with EGFR Exon20ins mutations identified from US institutions by next-generation sequencing were extracted from the AACR Project Genomics Evidence Neoplasia Information Exchange (GENIE), v.8.5 (Cancer Discov 2017;7:818), FoundationInsights, and Guardant databases.

Results

From GENIE, 175 / 9,673 pts (1.8%) with LUAD had EGFR Exon20ins mutations: 72.0% were located in the near loop, 28.0% in the far loop, and none in the helical region (Table). The most frequent sites of insertion were AAs 770 (36.0%), 773 (26.9%), and 769 (21.7%).

From FoundationInsights, 627 / 36,465 pts (1.7%) with LUAD had EGFR Exon20ins mutations: 67.8% were located in the near loop, 26.6% in the far loop, and 5.4% in the helical region. The most frequent sites of insertion were AAs 770 (28.1%), 769 (23.4%), and 773 (22.8%). One Exon20ins mutation at AA 793 was also identified (Table).

From Guardant, 414 / 31,617 pts (1.3%) with LUAD had EGFR Exon20ins mutations. 65.9% were located in the near loop, 26.6% in the far loop, and 7.5% in the helical region (Table). The most frequent sites of insertion were AAs 770 (28.0%), 769 (24.2%), and 773 (22.9%).

Conclusions

Three real-world databases identified the near loop as the most frequent insertion region of Exon 20 (~70%). Exon20ins mutations were found less frequently in the far loop (~30%) and helical (0–7.5%) regions.

Table. Exon20ins Mutations by Site of Insertion

Insertion Region of Exon 20, n (%)

GENIE (N=175) FoundationInsights (N=627) Guardant (N=414)
Helical (AA 762–766) 0 34 (5.4) 31 (7.5)
763 0 33 (5.3) 31 (7.5)
764 0 1 (0.2) 0
Near Loop (AA 767–772) 126 (72.0) 425 (67.8) 273 (65.9)
767 1 (0.6) 5 (0.8) 2 (0.5)
768 0 13 (2.1) 2 (0.5)
769 38 (21.7) 147 (23.4) 100 (24.2)
770 63 (36.0) 176 (28.1) 116 (28.0)
771 17 (9.7) 66 (10.5) 35 (8.5)
772 7 (4.0) 18 (2.9) 18 (4.3)
Far Loop (AA 773–775) 49 (28.0) 167 (26.6) 110 (26.6)
773 47 (26.9) 143 (22.8) 95 (22.9)
774 1 (0.6) 17 (2.7) 15 (3.6)
775 1 (0.6) 7 (1.1) 0
Other Regions (AA 776–823) 0 1 (0.2) 0
793 0 1 (0.2) 0

Editorial acknowledgement

Medical writing support was provided by Tracy T. Cao, PhD of Janssen Global Services, LLC and funded by Janssen Global Services, LLC.

Collapse
Tumour biology and pathology

2P - Role of neutrophil-to-lymphocyte ratio (NLR) in the outcome of NSCLC EGFR mutated

Room
ePoster Display
Speakers
  • S. Tolu (Cagliari, Italy)
Authors
  • S. Tolu (Cagliari, Italy)
  • G. Saba (Monserrato (CA), Italy)
  • V. Impera (Monserrato, Italy)
  • D. Spanu (Cagliari, Italy)
  • N. Liscia (Cagliari, Italy)
  • G. Pinna (Monserrato, Italy)
  • A. Pireddu (Cagliari, Italy)
  • L. Demurtas (Monserrato, (CA), Italy)
  • E. Lai (Cagliari, Italy)
  • F. Manca (Monserrato (CA), Italy)
  • C. Madeddu (Monserrato, (CA), Italy)
  • E. Massa (Monserrato, (CA), Italy)
  • G. Astara (Monserrato, (CA), Italy)
  • M. Scartozzi (Monserrato, Italy)

Abstract

Background

Systemic inflammation promotes angiogenesis and cell proliferation, which play key roles during carcinogenesis. Several studies have shown a strong link between inflammation index and prognosis in non-small-cell lung cancer (NSCLC). The aim of our study is to investigate the role of NLR at diagnosis in the outcome of NSCLC EGFR mutated.

Methods

Our retrospective analysis included 103 eligible patients (pts), including 40 pts with EGFR wild type (wt) NSCLC and 63 pts with EGFR mutated NSCLC. Inclusion criteria were over 18 years of age, diagnosis of NSCLC, availability of a pre-treatment complete blood count. Exclusion criteria were active infections, autoimmune diseases, use of corticosteroids. The NLR was derived from the absolute neutrophil an the absolute lymphocyte counts. The cutoff value was determined through the use of ROC curves and through the distinction, based on overall survival (OS) at 12 months, between pts with good and bad prognosis. For each subgroup of mts, median OS was assessed. Statistical processing was performed with the aim of correlating the haematological data collected with the clinical outcome.

Results

In pts EGFR wt the NLR cut-off obtained was 3.18., in pts EGFR mutaded the cut-off was 3,5. Pts who have a value less than or equal to cut-off have a better prognosis, compared to pts who have a NLR higher than have a worse prognosis. As a pts EGFR wt the median best OS was 41.33 months, while the worst OS was a median of 10.33 months, with a difference of 31 months (p = 0.0003). In case of pts EGFR mutaded the best median OS was 21 and the worst 8,3 months, with a difference of 13 months (p = 0,013). There were no significant differences in age, ECOG PS, histology, size, therapies or treatment line.

Conclusions

The NLR remains a prognostic factor in both diseases, with or without EGFR mutation , but appear to have less impact on the outcome of EGFR mutated patients. In NSCLC EGFR mutated maybe inflammatory index could have implications on therapeutic choice, especially subsequent lines and his role deserves further study.

Collapse
Tumour biology and pathology

3P - AZ12756122, a novel Fatty Acid Synthase (FASN) inhibitor, reduces resistance properties in Gefitinib- and Osimertinib resistant EGFR-mutated non-small cell lung cancer models

Room
ePoster Display
Speakers
  • E. Polonio (Girona, Spain)
Authors
  • E. Polonio (Girona, Spain)
  • S. Palomeras (Girona, Spain)
  • R. Porta-Balanya (Girona, Spain)
  • J. Bosch-Barrera (Girona, Spain)
  • C. Vásquez (Girona, Spain)
  • J. Ciurana (Girona, Spain)
  • S. Ruiz-Martínez (Girona, Spain)
  • T. Puig (Girona, Spain)

Abstract

Background

Non-small cell lung cancer (NSCLC) is the most common subtype corresponding to roughly 85% of lung cancer patients. The discovery of activating mutations in the epidermal growth factor receptor (EGFRm) gene has led to an era of targeted therapy in lung cancer with EGFR tyrosine kinase inhibitors (TKIs). Although great progress has been accomplished, therapeutic improvement is necessary due to the constant appearance of resistance. Cancer stem cells (CSCs) are a small subpopulation responsible for tumour initiation and progression, recurrence, metastasis, and resistance to anti-cancer therapies. Furthermore, the overexpression and hyperactivation of fatty acid synthase (FASN) has been related to tumour aggressiveness and therapy resistance. Previously, we have shown that FASN inhibition causes cytotoxic effects in EGFRm lung adenocarcinoma cell models sensitive and resistant to EGFR-TKIs. Additionally, the combination of AZ12756122, a novel FASN inhibitor, with osimertinib exhibited a synergistic interaction. Therefore, we studied the effect of AZ12756122 in combination with osimertinib on EGFR, HER2, MAPK, STAT3 and AKT proteins in EGFRm NSCLC models sensitive and resistant to EGFR-TKIs. Moreover, the ability of AZ12756122 to target the CSC population was evaluated.

Methods

EGFRm lung adenocarcinoma models sensitive to EGFR-TKIs (PC9) and T790M- resistant to both gefitinib and osimertinib (PC9-GR3) were used. Gene and protein expression were assessed through qRT-PCR and Western blot, respectively. CSC population was evaluated by means of sphere-formation and clonogenic assays.

Results

The combination of AZ12756122 with osimertinib reduced FASN protein expression and the EGFR, HER2, AKT, MAPK, and STAT3 protein activation in the PC9-GR3 model. AZ12756122 treatment decreased the CSC properties tested of PC9 and PC9-GR3 models.

Conclusions

In conclusion, FASN inhibition caused by AZ12756122 arises as an encouraging therapeutic alternative to overcome resistance to EGFR-TKIs due to its synergistic interaction with osimertinib and its capacity to reduce CSC properties in EGFRm NSCLC cell models.

Collapse
Tumour biology and pathology

4P - Prevalence of MET Exon 14-mutations or MET amplification in non-small cell lung cancer in Swiss patients

Room
ePoster Display
Speakers
  • S. Savic Prince (Basel, Switzerland)
Authors
  • S. Savic Prince (Basel, Switzerland)
  • M. Bihl (Basel, Switzerland)
  • S. Eppenberger-Castori (Basel, Switzerland)
  • M. Matter (Basel, Switzerland)
  • N. Zellweger (Basel, Switzerland)
  • S. Rothschild (Basel, Switzerland)
  • L. Bubendorf (Basel, Switzerland)

Abstract

Background

MET-targeted treatment has improved by the introduction of highly selective MET inhibitors for advanced stage MET Exon 14-mutated or MET-amplified non-small cell lung cancer (NSCLC). MET exon 14 skipping mutations occur in 3% and MET amplification in 1 to 6% of unselected NSCLC. The aim of our study was to investigate the prevalence of these MET alterations in treatment-naïve pre-selected NSCLC from our routine clinical practice, were MET analyses were generally performed sequentially in NSCLC with high MET expression and wild type status for EGFR, KRAS, ALK and ROS1.

Methods

580 consecutive treatment-naïve NSCLC who underwent routine predictive testing were retrospectively evaluated. High MET expression, as determined by immunohistochemistry (IHC), was defined as complete membranous MET staining with at least moderate intensity in ≥50% of tumor cells. MET exon 14 analysis was performed by Sanger sequencing. MET gene copy numbers were evaluated by fluorescence in situ hybridization (FISH) and MET amplification was defined as a MET/CEN7 ratio of ≥2.0. MET IHC was performed on histology specimens and cellblocks, but not on conventional cytology preparations. Therefore, prevalence data of MET alterations were analyzed separately for IHC pre-selected and non-preselected NSCLC wild type for other oncogenic driver alterations.

Results

66% (302/457) of NSCLC had high MET expression. There was no difference between the prevalence of MET exon 14 mutations in IHC pre-selected and non-preselected NSCLC wild type for other oncogenic driver alterations (8.3% (9/108) and 8.1% (7/86), respectively, p=1.0). In contrast, the prevalence of MET amplification was higher in MET IHC pre-selected NSCLC (11.5% (11/96) and 3.2% (1/31), respectively, p=0.29).

Conclusions

MET exon14 skipping mutations are enriched in NSCLC wild type for other oncogenic driver alterations with the prevalence of >8% being higher compared to the published literature in unselected NSCLC (3%). High MET expression does not enrich for MET exon 14 mutations. However, MET IHC seems to be a cost-effective approach for pre-screening NSCLC for further MET FISH analysis.

Collapse
Tumour biology and pathology

5P - False positives errors in RNA based next generation sequencing of exon 14 skipping mutations in NSCLC

Room
ePoster Display
Speakers
  • M. Suryavanshi (New Delhi, India)
Authors
  • M. Suryavanshi (New Delhi, India)
  • S. Mattoo (Delhi, India)
  • U. Batra (delhi, India)
  • S. Sharma (New Delhi, India)
  • D. Kumar (Delhi, India)
  • A. Mehta (Delhi, India)

Abstract

Background

Exon 14 skipping mutations are found in approximately 3% of patients with NSCLC.Robust approaches for detection of MET exon 14 skipping events are crucial for treatment. About one-third of the mutations occur between exons 13 and 14 at acceptor site of exon 14, and two-thirds occur between exons 14 and 15 at donor site. In addition, MET exon 14 skipping can result from large deletions that can span not only all of exon 14 but large portions of the intronic sequence. This mutation can be detected by sequencing of DNA or RNA or both. DNA based approaches alone are not able to detect more than 60% of these mutations. RNA based sequencing detects the product of exon 14 skipping mutations which is “fusion” of exon 13 to 15 regardless of the underlying genomic event. Most studies have favoured a RNA based approach.

Methods

During the period from August 2017 to January 2021, a total of 231 samples of NSCL were assessed by routine clinical application of the Thermofischer Ion Torrent™ Oncomine™ Focus 52 gene Assay . Sequencing data were processed with the Ion Torrent Suite software . This assay detects MET mutations at 3′ end splice donor site by DNA sequencing and RNA sequencing for exon 13 and exon 15 fusion for exon 14 skipping mutation. All positive cases on RNA sequencing were reanalysed by PCR and Sanger sequencing for confirmation.

Results

Exon 13 and exon 15 fusion by RNA was detected in 20 cases. Read counts ranged from 143 to 7980. Two cases had additional MET amplification,one case had EGFR deletion 19,one case had CTNNB1 p.Ser37Phe and KRAS p.Gly12Asp, one case had RET KIF5B Fusion(read count 458), one case had EGFR amplification and in remaining cases exon 13 and exon 15 fusion was the sole abnormality. Only 6 out of 20 cases detected by NGS were confirmed by Sanger sequencing. All cases above the read count of 1607 were detected by sanger sequencing. All true positive cases had exon 3 and exon 15 fusion as the sole abnormality. Cases with MET amplification were also negative on sanger sequencing.

Conclusions

RNA based exon 13 and exon 15 fusion for detection of exon 14 skipping mutations can have false positive calls by Ion torrent based sequencing and should be confirmed by alternate methods.

Collapse
Tumour biology and pathology

6P - Clinical and molecular characteristics in non-small cell lung cancer patients with alteration in PIK3 pathway.

Room
ePoster Display
Speakers
  • Y. Lage (Madrid, Spain)
Authors
  • Y. Lage (Madrid, Spain)
  • P. Álvarez Ballesteros (Madrid, Spain)
  • M. Olmedo García (Madrid, Spain)
  • A. Gómez Rueda (Ciudad Real, Spain)
  • E. Corral de la Fuente (Madrid, Spain)
  • J. Torres Jiménez (Madrid, Spain)
  • E. Vida Navas (Madrid, Spain)
  • J. Soto Castillo (Madrid, Spain)
  • M. Lario (Madrid, Spain)
  • A. Benito Berlinches (Madrid, Spain)
  • A. Santón (Madrid, Spain)
  • P. Garrido Lopez (Madrid, Spain)

Abstract

Background

Currently non-small cell lung cancer (NSCLC) is a tumor with a broad spectrum of targeted therapies already available or in clinical trials. Molecular characterization of the tumor using next generation sequencing (NGS) technology has become a key tool for the molecular profiling of NSCLC, facilitating treatment decisions.

The phosphoinositide-3-kinase (PI3K) signaling pathway has a critical role in both tumorigenesis and the progression of disease and it represents an attractive target for novel anticancer therapies in NSCLC.

Methods

We performed a retrospective study of patients with NSCLC treated in Hospital Universitario Ramón y Cajal, with alterations in PIK3 pathway identified by NGS. Clinical characteristics and concurrent mutations were described.

Results

1745 patients with lung carcinoma were treated in our institution between 2011 to 2020. We analyzed 479 patients with NSCLC who underwent NGS and 61 patients (12.7%) were identified with alteration in PIK3 pathway (tissue NGS in 43 and blood-based NGS in 19 patients).

57.3% of patients were diagnosed at IV stage. Most of them were men (67%) with smoking history (13% non-smokers). The most common histological types were adenocarcinoma (42%) and squamous cell carcinoma (36%). 27% of tumors were PD-L1 negative and 25 patients had determination of tumor mutational burden (> 10 mut/Mb in 52%).

PI3-kinase catalytic subunit alpha (PIK3CA) was the more common molecular finding (77% of patients): mutations exon 9 and 20 (82.5%), amplification (13.7%) and both (3.4%).

We also found the presence of mutations of PTEN (tumor suppressor that negatively regulates the PIK3/AKT/mTOR pathway) in 6.5% and TP 53 in 59 % of patients.

Other mutations were detected in 27 patients (44.2%): EGFR (8%), KRAS (8,1%), ALK rearrangement (3.2%), BRAF (1.6%), MET(3.2%), FGFR (6.5%) , CDKs (22.9%) .

Conclusions

Our cohort shows that alteration in PI3K pathway is more frequent in men with smoking history NSCLC patients and is not mutually exclusive to other mutations, that remark the relationship between pathways. Clinical trials are needed to predict the potential clinical benefit from the use of PI3K pathway inhibitors.

Collapse
Tumour biology and pathology

7P - STK11 and Galectin-3 Tissue Expression Entails a Prognostic Signature in Immunotherapy Treated NSCLC patients

Room
ePoster Display
Speakers
  • G. Mazzaschi (Parma, Italy)
Authors
  • G. Mazzaschi (Parma, Italy)
  • R. Minari (Parma, Italy)
  • L. Gnetti (Parma, Italy)
  • N. Campanini (Parma, Italy)
  • T. Zielli (Parma, Italy)
  • M. Baucina (Parma, Italy)
  • F. Perrone (Parma, Italy)
  • A. Leonetti (Parma, Italy)
  • F. Quaini (Parma, Italy)
  • M. Tiseo (Parma, Italy)

Abstract

Background

The profound and long-lasting response to immune checkpoint inhibitors (ICIs) in a subset of non-small cell lung cancer (NSCLC) patients makes it urgent the need of predictive biomarkers. As critical modulators of the tumor microenvironment, STK11 and Galectin-3 (Gal-3) demonstrated significant impact on patient outcome. Thus, we aimed to dissect the prognostic and predictive role of STK11 and Gal-3 tissue expression in ICI-treated advanced NSCLC.

Methods

Tissue and peripheral blood (PB) samples were collected at baseline from 25 consecutive NSCLC undergoing ICI. STK11 and Gal-3 expression was detected by Immunohistochemistry (IHC). STK11 mutation was determined by Next Generation Sequencing (NGS) in a subset of cases (n: 8, so far - other cases are ongoing). Soluble PD-L1 (sPD-L1) (immunoassay), PB CD8+PD1+ and NK cells (FACS) together with inflammatory features including LDH were measured. Tissue parameters were correlated with clinical outcome and circulating immune profile.

Results

Absence of STK11 expression by IHC (STK11neg) was detected in 18 (72%) NSCLC among which 74% were KRAS-mutant. Lack of STK11 expression at tissue and genomic levels was concordant in 62%. STK11neg cases displayed a trend towards a worse PFS (median PFS [mPFS], 4.3 vs 13.5 months, P=0.09) and OS (median OS, 12.8 months vs not reached, P=0.190). Although PFS was shorter in Gal-3high (2+/3+ IHC) cases, Gal-3 status did not impact on OS nor DCR. Significantly higher sPD-L1 levels (P=0.037) and lower circulating NK (P=0.042) were observed in STK11neg cases, while Gal-3high were associated to higher LDH values and lower CD3+ and NK. Patients carrying STK11neg/Gal-3high tumors had significantly reduced PFS (mPFS, 1.7 vs 13.5 months, P=0.001) and lower benefit from ICI (DCR, 30% vs 80%, P=0.03) compared to their counterpart. Moreover, STK11neg/Gal-3high signature had a relevant impact on CD4+, CD8+ and NK number and LDH levels.

Conclusions

STK11neg and Gal-3high may represent a poor prognostic trait in ICI-treated NSCLC patients. The correlation of STK11 and Galectin with circulating factors underlies a close interplay between tissue and peripheral blood compartments, ultimately featuring the tumor-host interaction and the proneness to respond to ICIs.

Collapse
Tumour biology and pathology

8P - Distinct tumor bacterial microbiome in lung adenocarcinomas manifested as radiological subsolid nodules

Room
ePoster Display
Speakers
  • Y. Ma (Beijing, China)
Authors
  • Y. Ma (Beijing, China)
  • M. Qiu (Beijing, China)

Abstract

Background

Early stage lung adenocarcinoma (LUAD) manifested as subsolid nodules (SSN) in CT have been considered as a special clinical subtype and are less aggressive than pure solid LUAD. Here, we firstly report the microbiome diversity between subsolid and solid LUAD.

Methods

We performed 16S rRNA sequencing of 35 pairs of LUAD tumor tissues and adjacent normal tissues, including 10 SSN and 25 SN. 29 patients were at TNM stage I, 2 were at TNM stage II, and 4 at were TNM stage IIIA. Age, gender, BMI and TNM stage were matched between SSN and SN. Machine learning was used to identify microbial signatures and construct predictive models.

Results

SSN and SN had 102 and 232 unique OTUs, respectively; while tumor and normal lung tissues had almost identical OTUs. Interestingly, at phylum level, both SSN and SN were mainly composed of Actinobacteria, Proteobacteria, Firmicutes, and Bacteroidetes. SSN had significantly higher abundance of Chloroflexi and Gemmatimonadetes. At genus level, Rhodococcus, Ochrobactrum are main compositions in these two groups. For alpha-diversity, SSN group has greater bacterial richness (number of OTUs, p=0.017) and diversity (shannon index, p=0.17). Both the unweighted unifrac and PCoA analyses confirmed SSN and SN have statistically different microbiome composition (Anosim, R = 0.213, p = 0.016). Moreover, LEfSe analysis revealed 54 features that may discriminate SSN and SN (LDA > 2.5, p < 0.05). At genus level, increased bacteria such as Cloacibacterium, Subdoligranulum, and Mycobacterium and decreased bacteria like Lachnoclostridium were strong discriminative features for SSN. Probiotics with anti-cancer potential, like Lactobacillus, showed elevated levels in normal tissues. Based on a five genera-signature, SSN could be accurately discriminated from SN using random forest algorithm with a sensitivity of 1, specificity of 0.8, and AUC of 0.90 (95% CI, 0.77–1.00). To verify the accuracy, we performed 10-fold cross-validation, resulting in a mean AUC of 0.933.

Conclusions

Early stage LUAD manifested as radiological SSN has increased microbiome richness and diversity compared with SN, and the microbiome composition of SSN is distinct from that of SN. Subsolid lung adenocarcinoma has a special microbiome subtype .

Collapse
Tumour biology and pathology

9P - Lung cancer mutational state assessed by NGS in a Portuguese Center

Room
ePoster Display
Speakers
  • D. Magalhães (Senhora da Hora, Portugal)
Authors
  • D. Magalhães (Senhora da Hora, Portugal)
  • M. Vilaça (Senhora da Hora, Portugal)
  • É. Cipriano (Senhora da Hora, Portugal)
  • A. Tavares (Senhora da Hora, Portugal)
  • D. Silva (Senhora da Hora, Portugal)
  • L. Cirnes (Porto, Portugal)
  • H. Magalhães (Senhora da Hora, Portugal)
  • F. Estevinho (Senhora da Hora, Portugal)

Abstract

Background

In advanced non-squamous non-small cell lung cancer (NSCLC), and other selected NSCLC patients molecular testing is crucial to detect oncogenic driver mutations and to direct treatment regarding the molecular profile. Next-generation sequencing (NGS) allows accurate and efficient multiple genes and mutation analyses. This technique offers an extensive molecular tumor characterization in a cost-effective and timely manner.

Our aim was to evaluate the NGS mutational profile and to correlate the results with clinicopathological characteristics and PD-L1 expression in patients with NSCLC.

Methods

Retrospective study of the NGS analysis, carried out in patients (pts) with NSCLC between January 2017 and December 2020 in our oncological center in Portugal. Clinicopathological characteristics were accessed by review of medical records. A descriptive analysis was performed, and Qui square was used to compare categorical variables. P-value reported was two-sided, and tests were conducted at the 0.05 significance level, SPSS®20.

Results

A total of 299 pts with NSCLC were analyzed, 293 had a valid test. The median age was 69 years . 71% (n=211) were males, 21% (n=62) were non-smokers and 33% (n=99) were smokers. 65% (n=190) of the pts had at least one mutation. 73 pts (38%) had a mutation with a therapeutic target approved by EMA.

KRAS was the most found mutation (31%), followed by EGFR (14%) and ALK (5%). EGFR, ALK, ROS1 and RET were more prevalent in females (p<0.05), KRAS was more common in males. EGFR and RET were more common in pts younger than 50 years old (p<0.05). EGFR was most found in non-smokers, KRAS was the most prevalent in smokers and former smokers (p<0.05).

Regarding tumor PD-L1 expression, BRAF and MET variants were more common in PD-L1 ≥50% (p<0.05), while EGFR was more common in PD-L1 <50% (p<0.05).

Conclusions

KRAS and EGFR prevalence were in accordance with the previous reported in Caucasian population. Some mutations correlate with clinic-pathological characteristics as sex, smoking status, and PD-L1 expression. Our study provides results from broad molecular profiling of Portuguese NSCLC pts treated in one hospital. In the future, multicentre data analysis will be presented.

Collapse
Tumour biology and pathology

10P - Impact of the number of metastatic lymph nodes on survival in pathological stage II-N1 non-small cell lung cancer

Room
ePoster Display
Speakers
  • J. De Ruiter (Amsterdam, Netherlands)
Authors
  • J. De Ruiter (Amsterdam, Netherlands)
  • A. De Langen (Amsterdam, Netherlands)
  • K. Monkhorst (Amsterdam, Netherlands)
  • A. Veenhof (Amsterdam, Netherlands)
  • H. Klomp (Amsterdam, Netherlands)
  • R. Damhuis (Utrecht, Netherlands)
  • K. Hartemink (Amsterdam, Netherlands)

Abstract

Background

Several single-centre studies investigated whether the number of metastatic lymph nodes would be a relevant supplementary prognostic factor in pathological N1 (pN1) non-small cell lung cancer (NSCLC), however, with conflicting results. The aim of this study was to determine the prognostic impact of the number of metastatic N1 lymph nodes in different histological subtypes for patients with stage II-N1 NSCLC.

Methods

We performed a retrospective cohort study using data from the Netherlands Cancer Registry, including patients treated with a surgical resection for stage II-N1 NSCLC in 2010-2016. Overall survival (OS) was compared between patients with single (pN1a) versus multiple (pN1b) metastatic pN1 nodes. With a multivariable analysis we evaluated the impact of the number of metastatic pN1 nodes in different histological subtypes.

Results

The series involved 871 (67%) patients with pN1a and 438 (33%) with pN1b NSCLC. 5-year OS was 53% for pN1a versus 51% for pN1b. 54.1% of patients with pN1a and 54.8% of patients with pN1b received adjuvant chemotherapy. In multivariable analysis, the difference between pN1b and pN1a was statistically significant, but small (HR 1.19, 95% CI 1.01-1.40). Furthermore, histological subtype, age, pathological T stage, and adjuvant chemotherapy were independent prognostic factors. When stratifying for histological subtype, the prognostic impact of pN1a/b was only observed in adenocarcinoma patients (HR 1.44, 95% CI 1.15-1.81). Beneficial effects of adjuvant chemotherapy were not different for pN1a versus pN1b.

Conclusions

The number of metastatic pN1 nodes was associated with a minor difference in survival and its impact was limited to adenocarcinoma histology. Biological parameters such as histological subtype may be more appropriate for inclusion in future TNM revisions and may guide the use of new (neo)adjuvant systemic therapies.

Collapse
Tumour biology and pathology

11P - LINC00926 is a B cell-specific long noncoding RNA in lung adenocarcinoma and is associated with the prognosis of patients with this disease

Room
ePoster Display
Speakers
  • J. Li (Beijing, China)
Authors
  • J. Li (Beijing, China)
  • H. Guo (Beijing, China)
  • Y. Ma (Beijing, China)
  • H. Chen (Beijing, China)
  • M. Qiu (Beijing, China)

Abstract

Background

A previous report has shown that LINC00926 might be involved in the suppression of breast cancer metastasis. However, studies regarding the function of LINC00926 in lung adenocarcinoma are rare.

Methods

10X single cell RNA sequencing (scRNA-seq) data of 8 healthy lung tissues and 5 lung adenocarcinoma tissues were obtained from our previous study. The sequence of LINC00926 was obtained from Noncode (http://www.noncode.org/) and the cellular location of the lncRNAs was predicted by lncLocator (http://www.csbio.sjtu.edu.cn/bioinf/lncLocator/). The target miRNA of the lncRNAs were predicted by Starbase (http://starbase.sysu.edu.cn/). The target genes of miRNA were predicted by miRDB (http://mirdb.org/). Functional annotation of the target genes was made by DAVID (https://david.ncifcrf.gov/). The overall survival of patients in TCGA-LUAD with different expression level of LINC00926 was compared by GEPIA (http://gepia.cancer-pku.cn/).

Results

LINC00926 was specifically expressed in B cells in lung adenocarcinoma. However, LINC00926 was not expressed in B cells in normal lungs. LINC00926 was mainly located in the cytoplasm, suggesting that LINC00926 might function through RNA-sponge. LINC00926 could bind the microRNA hsa-miR-3194-5p. The target genes of hsa-miR-3194-5p were enriched in Gene Ontology (GO) terms such as cell junction, protein kinase binding and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways such as Ras signaling pathway. High expression of LINC00926 was associated with better overall survival of patients with lung adenocarcinoma.

Conclusions

LINC00926 might be a B-cell specific marker and might play a protective role in patients with lung adenocarcinoma. Further research is needed to explore how LINC00926 functions in patients with lung adenocarcinoma.

Collapse