Background

Adequate and timely testing for molecular alterations in NSCLC is necessary to enable treatment with tyrosine kinase inhibitors (TKI) when a certain mutation or rearrangement is present. On a nationwide basis, we aimed to assess the performance of molecular testing for EGFR and/or KRAS mutation, and ALK and ROS rearrangement in a cohort of metastatic NSCLC.

Methods

All stage IV non-squamous NSCLC from 2013 and 2015 were identified from the Netherlands Cancer Registry and matched to the Dutch Pathology Registry (PALGA). Using information extracted from pathology reports, proportions of tumors tested for EGFR and/or KRAS and ALK, and in 2015 also for ROS, within 3 months after diagnosis, were determined, and variation between 48 laboratories was assessed. In a best practice session with 4 laboratories with highest testing proportions, we tried to identify a process for the best possible flow and highest possible testing proportions.

Results

In total, 6,619 tumors were included (2013: N = 3,195; 2015: N = 3,424). In 2013, EGFR and/or KRAS testing was performed in 73.1% (variation between laboratories 30.6% to 91.7%) and was significantly higher in 2015: 78.9% (40.0% to 91.0%). Of the EGFR/KRAS wildtype (wt) tumors, 49.5% underwent ALK testing in 2013 (6.3% to 100%) and 77.4% in 2015 (32.5% to 100%), which was significantly higher. ROS testing was performed for 50.9% (0% to 100%) of the EGFR/KRAS wt tumors from 2015. In 2015, 6, 7 and 13 laboratories tested significantly less often for EGFR/KRAS, ALK and ROS, respectively, than the national proportion. Insufficient tissue was the most stated reason for not testing. The best practice session showed that, among other, dedicated specialized personnel, good communication with short lines, and a work culture of critical openness and honesty are essential for adequate molecular testing.

Conclusions

Although molecular testing proportions were significantly higher in 2015, improvement remains possible in some laboratories/hospitals, considering that some patients were possibly unjustly not eligible for TKI. Feedback on molecular testing performance was sent to individual laboratories, so they can review, and, if needed, improve their practice.

Legal entity responsible for the study

University Medical Centre Utrecht.

Funding

AstraZeneca, Pfizer, Roche.

Disclosure

C. Kuijpers, S. Willems: Funding: AstraZeneca, Pfizer, Roche (these companies had no role in study design, analyses and reporting). All other authors have declared no conflicts of interest.

Background

FRα is present on the cell surface, mediates its intracellular transport via receptor-mediated endocytosis, and is also involved in cell division. FRα is expressed at high levels on various cancer cells, and it is considered that FRα could be a potential therapeutic target in FRα-expressing cancers.

Methods

We investigated the correlations between FRα expression in tissue microarray (TMA) constructed from surgical specimens of lung adenocarcinoma (LADC) and clinicopathological features including EGFR mutation retrospectively. Two TMA cores per a resected LADC specimen were created using TMA Master (3DHISTECH, Budapest, Hungary) at National Cancer Center Hospital between January 2007 and November 2016. Based on previous studies, if the percentage of stained tumor cells was greater than or equal 5%, FRα was considered as positive, and if H-score was more than or equal 60 (range, 0-300), FRα was considered as high expression.

Results

Overall, 466 TMA cores created from 233 patients who underwent surgery for LADC were evaluated; FRα positive expression (FRα-pos)/negative (FRα-neg), 222/11; FRα high expression (FRα-HE)/low (FRα-LE), 190/43, respectively. 124 patients (53.2%) were with EGFR mutation. The median H-scores of FRα expression, rates of FRα-pos, and rates of FRα-HE for EGFR mutation and wild type were 159/104 (p = 0.0002), 97.6/92.7% (p = 0.077), and 88.7/73.4% (p = 0.0026). The median disease free survival (months) were 18.7/23.3 (p = 0.30) for FRα-pos/neg and 20.2/12.8 (p = 0.42) for FRα-HE/LE, respectively. The median overall survival were 92.9/not reached (p = 0.42) for FRα-pos/neg and 92.9/96.4 (p = 0.42) for FRα-HE/LE, respectively. FRα expression was not considered as a prognostic factor in LADC. 82 patients with EGFR mutation received EGFR-TKI after recurrence (FRα-pos/neg, 79/3; FRα-HE/LE, 71/11). Overall response rates and median progression free survival (months) of EGFR-TKI were 59.2/66.6% (p = 0.80) and 16.1/11.5 (p = 0.016) for FRα-pos/neg; 61.8/45.5% (p = 0.31) and 16.2/13.2 (p = 0.95) for FRα-HE/LE, respectively.

Conclusions

FRα expression was significantly higher in LADC with EGFR mutation than that with wild-type. The efficacy of EGFR-TKI was better in FRα-pos than in FRα-neg.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

Y. Fujiwara: Research funding: AbbVie, AstraZeneca, BMS, Chugai, Daiichi Sankyo, Eisai, Eli Lilly, Incyte, Merck Serono, MSD, Novartis; Advisory role: AstraZeneca, BMS, MSD, Novartis, ONO; Speakers bureau: AstraZeneca, BMS, MSD, Novartis, ONO, Sysmex, Taiho. S. Kitano: Personal fee, grant: Eisai; Honoraria: ONO, Bristol-Myers Squibb, AstraZeneca, Chugai, Pfizer, Sanofi, Nippon Kayaku, Boehringer Ingelheim, Meiji, Taiho, Novartis, Daiichi Sankyo, MSD, Kirin, Celgene, Sumitomo Dainippon; Grant: Regeneron, Astellas, Gilead, AMED, JSPS. T. Shimizu: Advisory board: Takeda, The Consortium on Harmonization of Institutional Requirements for Clinical Research; Corporate sponsored research: Bristol-Myers Squibb, Daiichi Sankyo, Lilly, Novartis, Eisai, Takeda, PharmaMar, FivePrime, 3D-Medicine, Symbio, Chordia. N. Yamamoto: Grants: Quintiles, Astellas, Chugai, Eisai, Taiho, BMS, Pfizer, Novartis, Daiichi Sankyo, Bayer, Boehringer Ingelheim, Kirin, Takeda, ONO, Janssen, Lilly, MSD, Merck; Advisory: Eisai, Takeda, Otsuka, Boehringer Ingelheim, Cimic; Speakers: BMS, Pfizer, AstraZeneca, Lilly, ONO, Chugai. N. Motoi: Personal fee (lecture, Advisory board): Bristol-Myers Squibb, Miraca Life Sciences, AstraZeneca, Chugai Pharma, MSD, Agilent, Novartis, Roche Diagnostics; Institutional research funding: Roche Diagnostics. All other authors have declared no conflicts of interest.

Background

Circular RNAs (circRNA), a subclass of noncoding RNA characterized by covalently closed continuous loops, play emerging roles in tumorigenesis and aggressiveness. Somatic copy number variations (CNV) may drive cancer progression through both coding and noncoding transcripts. In lung adenocarcinomas, tumor stage is positively correlated with the proportion of CNV. CNV-associated circRNAs in lung cancer have rarely been reported.

Methods

To identify critical circRNAs that contribute to lung adenocarcinoma progression, ribosomal RNA-depleted RNA sequencing was performed in 5 paired lung adenocarcinoma tissues and adjacent non-tumor tissues. Maltose-binding protein (MBP)-affinity purification and RNA immunoprecipitation (RIP) were used to identify RNA-binding proteins (RBPs) that were associated with circXPO1. Patient-derived xenograft (PDX) models were performed to assess the clinical benefits of targeting circXPO1.

Results

By integrating bioinformatics analyses of alerted circRNAs and focal CNV in lung adenocarcinoma, we identify a proto-oncogenic circular RNA (circXPO1), produced from the XPO1 gene at 2p15 amplicon. CircXPO1 was overexpressed in lung adenocarcinoma tissues and promoted proliferation and tumorigenesis of lung adenocarcinoma. Our mechanistic studies indicated that circXPO1 functions, at least in part, through its interaction with IGF2BP1. CircXPO1 epigenetically upregulated β-Catenin expression in lung adenocarcinoma cancer cells by interacting with IGF2BP1 to promote the stabilization of β-Catenin. Intratumor injection of cholesterolconjugated siRNA specifically targeting circXPO1 inhibited tumor growth in a patient-derived lung adenocarcinoma xenograft model.

Conclusions

In summary, our study identified a CNV-associated circular RNA, circXPO1, which is overexpressed in lung adenocarcinoma tissues, due to the amplification of 2q15 locus, promotes proliferation and tumorigenesis of lung adenocarcinoma via interacting with IGF2BP1 and subsequently promote the stabilization of β-Catenin. Meanwhile, understanding the precise role of circXPO1 provides a novel therapeutic strategy for lung adenocarcinoma in the sight of circular RNAs.

Legal entity responsible for the study

Qi Huang, Mantang Qiu, Fan Yang, Jun Wang.

Funding

National Natural Science Foundation of China.

Disclosure

All authors have declared no conflicts of interest.

Background

Somatic mutations of the gene encoding epidermal growth factor receptor (EGFR) are detected in approximately 30%–50% of patients with non-small cell lung cancers (NSCLC), so detection of EGFR mutation is the pivotal step of treatment in patients with advanced NSCLC. However, difficulty in obtaining sufficient tissue and bias from the heterogeneity of the tumor samples are the major obstacles. Although analyzing EGFR with circulating tumor DNA (ctDNA) in plasma is a breakthrough, accuracy is the problem in variable methods. Peptide nucleic acid (PNA) clamping-assisted fluorescence melting curve analysis (PANAMutyper®) is a novel and highly sensitive method of detecting EGFR mutation in tumor tissues. This study was designed to evaluate PANAMutyper® for detecting EGFR mutation with ctDNA of patients with lung cancer.

Methods

EGFR mutation status detected by PNA clamp with tissue samples and by PANAMutyper® with ctDNA was compared. Tissue biopsy was done in 158 patients with lung tumor, in which 23 cases were excluded and 135 cases were enrolled. All the plasma samples of the cases with mutant EGFR in tissue samples were verified by an already known highly sensitive method of droplet digital polymerase chain reaction (ddPCR).

Results

EGFR mutation rate was 23.0% (31/135) in overall patients. The concordance rate of tissue and plasma samples was 91.9% (124/135). The sensitivity, specificity, negative predictive value, and positive predictive value were 64.5%, 100.0%, 90.4%, and 100.0%, respectively, according to the tissue samples as a standard.

Conclusions

PANAMutyper® method was not inferior to ddPCR for the detection of EGFR mutation including T790M with ctDNA. These results suggest that the detection of EGFR mutation status using ctDNA in plasma by PANAMutyper® is a feasible test prior to tissue biopsy.

Legal entity responsible for the study

Choonhee Son.

Funding

Has not received any funding.

Disclosure

The author has declared no conflicts of interest.

Background

Tumor Treating Fields (TTFields) are a clinically approved anti-mitotic treatment modality delivered via noninvasive application of low intensity, intermediate frequency, alternating electric fields. In this study, we evaluated whether TTFields-induced cell death can be perceived as immunogenic.

Methods

Human and murine lung carcinoma cell lines were treated with TTFields using the inovitro system. Immunogenic cell death was evaluated by changes in the levels of calreticulin (CRT) on the surface of treated cells, phosphorylation of the translation initiation factor eIF2α, and secretion of ATP and High mobility group box 1 (HMGB1). Activation of immune cells was evaluated using co-culture of bone marrow derived dendritic cells (DCs) with TTFields treated cells. For in-vivo studies, mice orthotopically implanted with lung tumors were treated with TTFields, the immune checkpoint inhibitor anti-PD-1 or a combination of the two modalities. Tumor volume was monitored and flow cytometry analysis was performed for phenotypic characterization of infiltrating immune cells.

Results

We demonstrate that cancer cells that died under TTFields application exhibited release of HMGB1, ATP secretion from cells, and ER stress leading to CRT translocation to the cell surface, all of which are cardinal signs of immunogenic cell death. TTFields treated cells promoted in vitro phagocytosis by DCs and DC maturation as well as initiation of inflammation in vivo. The combined treatment of lung tumor-bearing mice with TTFields plus the immune checkpoint inhibitor anti-PD-1 led to a significant decrease in tumor volume compared to anti-PD-1 alone or to the control group. Significant increases in CD45+ tumor infiltrating cells were observed in the TTFields plus anti-PD-1 group. These infiltrating cells demonstrated upregulation of surface PD-L1 expression. Correspondingly, cytotoxic T-cells isolated from these tumors showed higher levels of IFN-γ production relative to untreated mice.

Conclusions

Our results demonstrate the potential of TTFields therapy to induce immunogenic cell death and increase the efficacy of anti PD-1 therapy by further enhancing antitumor immunity.

Legal entity responsible for the study

Novocure Israel.

Funding

Novocure Israel.

Disclosure

U. Weinberg, T. Voloshin Sela, Y. Porat, M. Munster, R.S. Schneiderman, C. Tempel Brami, S. Cahal, M. Giladi, A. Kinzel, Y. Palti: Full time employee: Novocure Israel; Stock options, stocks: Novocure. N. Kaynan, S. Davidi, A. Shteingauz, K. Gotlib, E. Zeevi: Full time employee: Novocure Israel; Stock options: Novocure. E. Kirson: Full time employee: Novocure Israe; Stock options, stocks: Novocure; Senior leadership position: Novocure.

Background

Complex network of chemokines is a part of immune reaction targeted against tumor cells. Chemokines influence cancer growth. It is not clear whether the concentrations of chemokines at the time of diagnosis of NSCLC reflect extent of the disease. Aim: To compare chemokine concentrations (CCL2, CCL8, CXCL12) in plasma of patients with resectable NSCLC to those without cancer. To find out whether the chemokine concentrations differ according to stage of the disease.

Methods

We enrolled 69 patients undergoing surgery for proven/suspected NSCLC (37 males- mean age 66 years, 32 females- mean age 64 years). They underwent standard diagnostic and staging procedure to determine resectability (CT/PET-CT, bronchoscopy, ev. abdominal ultrasound), then surgery was made. We diagnosed 42 patients with NSCLC, 27 patients in the control group had benign lung pathologies (e.g. chondrohamatoma, fibrous tissue) and remained as the control group. We took plasma samples before surgery in treatment naive patients. The chemokine concentrations were assesed by ELISA LSBio Kits. The results were stated in pg/ml (CCL2, CCL8), ng/ml (CXCL12). Parametric statistics was used for results analysis.

Results

Table: 8P

Conclusions

Chemokine values did not differ in NSCLC versus control group. Patients with NSCLC stages II and IIIA had significantly higher CXCL12 concentrations than control group. This could be because CXCL12 promotes tumor growth and metastasis.

Legal entity responsible for the study

The authors.

Funding

Internal grant of Thomayer Hospital.

Disclosure

All authors have declared no conflicts of interest.

Background

Patients with non-small cell lung cancer (NSCLC) often harbors driver mutations in multiple oncogenes, including EGFR, KRAS, ALK, ROS1, BRAF, HER2, RET, etc. Driver genetic alterations are used as predictive biomarkers for molecular targeted therapies. Therefore, identifying mutations in oncogenes and tailoring therapy accordingly become a standard in clinical cancer management. A genetic testing assay based on next-generation sequencing (NGS) technology, named AmoyDx Essential NGS Panel, has been developed for multiplexed and targeted deep sequencing of variants in 10 driver genes. Clinical validation was conducted in the present study.

Methods

A total of 372 formalin-fixed and paraffin-embedded (FFPE) tissue samples collected from Chinese patients with NSCLC were detected by AmoyDx Essential NGS Panel, which enables the simultaneous detection of single-nucleotide variants (SNVs), insertions/deletions and fusions in 10 driver genes (EGFR, KRAS, NRAS, BRAF, PIK3CA, ALK, ROS1, HER2, RET, and MET) in DNA samples. DNA sequencing was performed on the Illumina platform. Golden standard Sanger sequencing as reference method was used as a reference method to test the same cohort. The concordance of variants determined with the AmoyDx Essential NGS Panel was assessed compared to the reference method.

Results

In total of 372 samples, 98.39% (366/372) of patients were successfully detected by both AmoyDx Essential NGS Panel and Sanger sequencing. 73.50% (269/366) were identified to shown variants by NGS as listed in the table . The overall concordance rate of mutations determined with AmoyDx Essential NGS Panel compared with reference was 98.80%.

Conclusions

The NGS analysis with AmoyDx Essential NGS Panel represents an accurate and efficient approach to detect genomic alterations in 10 driver genes with high concordance rate of 98.80% compared with Sanger sequencing.

Legal entity responsible for the study

Amoy Diagnostics Co., Ltd.

Funding

Has not received any funding.

Disclosure

H. Ge, L. Ruan, Q. Gao, X. Yang, W. Shi, X. Li, G. Zhu: Employee: Amoy Diagnostics Co., Ltd. All other authors have declared no conflicts of interest.

Table: 9P Variants detected by NGS

Background

Although roles of JAK2 mutations in myeloproliferative neoplasms (MPN) are well established, roles of JAK2 mutations in the pathogenesis of non-small cell lung cancer (NSCLC) remain unclear. The aim of this study is to investigate mutations and prognosis of NSCLC harboring JAK2 mutations.

Methods

A total of 766 patients with non-small-cell lung cancer were recruited between July 2012 and December 2016. The status of JAK2 mutations and other genes were detected by next generation sequencing.

Results

JAK2 gene mutation rate was 2.35% (18/766) in non-small cell lung cancer, including V617F (3 patients), L43I (2 patients), D768H (1 patient), G1066A (1 patient), W1020C (1 patient), S465R (1 patient), Q21E (1 patient), T842A (1 patient), C452* (1 patient), I404M (1 patient), P708Q (1 patient), S488* (1 patient), S1035L (1 patient), V387L(1 patient) and E575* (1 patient), and median overall survival (OS) for these patients was 19.0 months. Among them, all patients were JAK2 gene with co-occurring mutations. Briefly, patients with (n = 4) or without (n = 14) co-occurring EGFR mutations had a median OS of 23.0 months and 13.0 months respectively (P = 0.05); patients with (n = 11) or without (n = 7) co-occurring TP53 mutations had a median OS of 20.0 months and 7.0 months respectively (P = 0.01); patients with (n = 3) or without (n = 15) co-occurring ARID1A mutations had a median OS of 20.0 months and 13.0 months respectively (P = 0.32); patients with (n = 6) or without (n = 12) co-occurring KRAS mutations had a median OS of 11.5 months and 20.0 months respectively (P = 0.25).

Conclusions

TP53 accompanied mutations might play a good prognosis in JAK2 gene mutation NSCLC. Next generation sequencing provides a simplified strategy and reasonably high detection rate for JAK2 mutation, which suggested application of the strategies into clinical molecular diagnostics.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

The efficacy and safety of multikinase inhibitor anlotinib has been confirmed in the treatment of advanced non-small cell lung cancer (NSCLC). However, the direct functional mechanisms of tumor lethality mediated by anlotinib were not fully elucidated, and the underlying mechanisms related to resistance remains largely elusive.

Methods

Cell viability, colony formation, apoptosis and tumor growth assays were performed to examine the effect of anlotinib on lung cancer cells in vitro and in vivo. The punctate patterns of LC3-I/II were detected by confocal microscopy. HUVECs migration was detected using Transwell and scratch wound-healing assay. To visualize the microvessels, tublar formation assay was performed. The expression of LC3-I/II and beclin-1 and the changes of JAK2/STAT3/VEGFA pathway were detected by western blotting. The VEGFA levels in tumor supernatant were measured by ELISA.

Results

Anlotinib treatment decreased cell viability and induced apoptosis in Calu-1 and A549 cells. Moreover, anlotinib induced human lung cancer cell autophagy in a dose- and time-dependent manner. Blocking autophagy enhanced the cytotoxicity and anti-angiogenic ability of anlotinib as evidenced by HUVECs migration, invasion, and tublar formation assay. Co-administration of anlotinib and chloroquine (CQ) further reduced VEGFA level in the tumor supernatant, compared with that of anlotinib or CQ treatment alone. When autophagy was induced by rapamycin, the JAK2/STAT3 pathway was activated and VEGFA was elevated, which was attenuated after deactivating STAT3 by S3I-201. Further in vivo studies showed that anlotinib inhibited tumor growth, induced autophagy and suppressed JAK2/STAT3/VEGFA pathway, and CQ enhanced this effect.

Conclusions

Anlotinib induced apoptosis and protective autophagy in human lung cancer cell lines. Autophagy inhibition further enhanced the cytotoxic effects of anlotinib, and potentiated the anti-angiogenic property of anlotinib through JAK2/STAT3/VEGFA signaling.

Legal entity responsible for the study

The authors.

Funding

National Natural Science Foundation of China grants 81472792 and Postgraduate Research & Practice Innovation Program of Jiangsu Province grants SJCX18_0707.

Disclosure

All authors have declared no conflicts of interest.

Background

Variation at TP63 has recently been shown to be associated with non-small cell lung cancer patients (NSCLC) in the Chinese population. There is some clinical evidence for the use of TP63 mutations as prognostic and predictive biomarker. The aim of this study is to investigate mutations and prognosis of NSCLC harboring TP63 mutations.

Methods

A total of 1236 patients with non-small-cell lung cancer were recruited between July 2012 and December 2016. The status of TP63 mutations and other genes were detected by next generation sequencing.

Results

TP63 gene mutation rate was 2.02% (25/1236) in non-small cell lung cancer, including R643* (1 patient), H247N (1 patient), A139V (1 patient), V626F (1 patient), Q70* (1 patient), Q274E (1 patient), H615D (1 patient), R350T (1 patient), Y202Kfs*29 (1 patient), P229H (1 patient), M40V (1 patient), E409Q (1 patient), V179M (1 patient), W658* (1 patient), S365* (1 patient), T193M (1 patient), L50F (1 patient), A554E (1 patient), R226H (1 patient), Q99* (1 patient), S310N (1 patient), T169N (1 patient), R266Q (1 patient), D372H (1 patient), and P492T (1 patient), and median overall survival (OS) for these patients was 15.0 months. Among them, all patients were TP63 gene with co-occurring mutations. Briefly, patients with (n = 5) or without (n = 20) co-occurring EGFR mutations had a median OS of 22.5 months and 14.0 months respectively (P = 0.23); patients with (n = 21) or without (n = 4) co-occurring TP53 mutations had a median OS of 15.0 months and 13.0 months respectively (P = 0.33); patients with (n = 5) or without (n = 20) co-occurring BRAF mutations had a median OS of 14.0 months and 15.0 months respectively (P = 0.72); patients with (n = 5) or without (n = 20) co-occurring KRAS mutations had a median OS of 6.0 months and not up to now respectively (P < 0.01).

Conclusions

TP63 is structurally and functionally similar to TP53 and their activity as transcription factors is regulated by a wide repertoire of shared and unique post-translational modifications and interactions with regulatory cofactors. EGFR, TP53 and BRAF gene accompanied may have less correlation with TP63 mutation in NSCLC patients. KRAS accompanied mutations might play a worse prognosis in TP63 gene mutation NSCLC.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Mast cells (MCs) are key effectors in allergic reactions, but are also involved in tissue remodeling, wound healing and protection against pathogens. MCs infiltrate tumors and their number within the tumor microenvironment in certain cancer types, such as lung cancer, have been correlated with poor prognosis. The nature of crosstalk between lung cancer and MCs remain poorly resolved. In this study, we investigated the activation patterns within the MCs following cell-to-cell contact with lung cancer cells showing CD73 involvement.

Methods

Human MCs (HMC-1 and LAD-2) were exposed to Human lung cancer cells (H1299), derived membranes to recapitulate cell contact-mediated activation. Lysates of MCs were tested for protein expression and phosphorylation by targeted western blotting. We unraveled the intracellular signaling molecules that are necessary for this signaling pathway by a pharmacological approach using several inhibitors. Each condition was repeated at least three times.

Results

H1299 membrane exposure activated the ERK 1/2 MAP kinases in HMC-1 and in LAD-2 cells. AKT signaling was also activated in LAD-2 cells as a result of this contact. CD73 dephosphorylates AMP to adenosine within the MCs. Interestingly enough, this ERK 1/2 activation was inhibited by CD73 inhibitor and A3 receptor antagonists in HMC-1 cells. ERK 1/2 activation was inhibited by A3 receptor antagonists and PI3K in LAD-2 cells. Furthermore, we discovered that protein kinase C (PKC) inhibitor augments the activation of ERK 1/2 in LAD-2 cells. In contrast, PKC inhibitor inhibits the activation of ERK 1/2 in HMC-1 cells.

Conclusions

Our results suggest that H1299 membranes activate ERK 1/2 in HMC-1 cells by a mechanism that involves autocrine formation of adenosine and is mediated by CD 73 and A3 receptor. In addition, we discovered that there is an important difference between the ERK 1/2 MAP kinase signal transduction in HMC-1 and LAD-2 cells, PKC is an inhibitor of the H1299 activation of ERK 1/2 in LAD-2 cells. In contrast, the H1299 membrane activation of ERK 1/2 kinase in HMC-1 cells is mediated by PKC. This could be explained by the fact that LAD-2 MCs express wild type c-kit receptor, and HMC-1 MCs express an oncogenic constitutively active c-kit mutant.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

MET exon 14 skipping is a potential driver alteration in lung cancer targetable. Treatment with crizotinib can cause dramatic responses in patients whose cancers have MET exon 14 skipping. The current study was aiming to determine the clinical and pathological characteristics in non-small cell lung cancers (NSCLC).

Methods

A total of 2926 patients with NSCLC were recruited between July 2012 and 2015. The status of MET exon 14 skipping and other genes were detected by next generation sequencing.

Results

MET exon 14 skipping rate was 1.06% (31/2926) in NSCLC, including X1009_spilce (10 patients), X963_spilce (6 patients), D1010H (5 patients), D1010N (3 patients), X1008_spilce (1 patient), X1006_spilce (1 patient), X1007_spilce (1 patient), D1010Y (1 patient), Y1003S (1 patient), D1002G (1 patient), P1008A (1 patient). Among them, EGFR mutations+ MET skipping [7 patients (2 patients with 19 del+ 5 patients with L858R)], ALK fusion+ MET skipping (2 patients) and ROS1 fusion+ MET skipping (1 patient).

Conclusions

MET exon 14 skipping defined a new molecular subset of NSCLC with identifiable clinical characteristics. The therapeutic crizotinib might be an alternative treatment for patients with MET exon 14 skipping NSCLC.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Patients with thyroid transcription factor 1 (TTF1) negative lung adenocarcinoma (ADC) have been reported to have a worse prognosis and to lack epidermal growth factor receptor (EGFR) mutations. The aim of this study was to analyse the prognostic significance of TTFI status in a series of tumor samples from patients with clinically confirmed lung adenocarcinoma.

Methods

A real-world data study of TTF1-negative ADC was performed in all lung cancer patients diagnosed in the University Hospital of Lleida from January 2011 to December 2016, using TTF1 clone 8G7G3/1(DAKO^®). Each patient’s clinical history, pathology specimens and molecular results were noted. The control group consisted of 231 patients with TTF1 positive lung ADC .

Results

Three hundred and thirty four (334) patients with ADC were identified; TTF1 results were available in 258 (77,2%) cases. Of these, 27 (10,4%) patients were identified with TTF1-negative ADC (74% males). The median age of these patients was 66 years and their smoking history was as follows: 55,5% former smoker, 29,6% ex smokers and 14,8% never smokers. The clinical stages were as follows: stage I or II (n = 3 [11%]), stage III (n = 4 [14,8%]) and stage IV (n = 20 [74%]). Patients’ mean survival was 7,6 vs 24,4 months in ADC TTF1-negative patients versus TTF1-postive patients (P = 0,00001). When compared with the control group, TTF1-negative patients had shorter overall survival (P = 0,00001), regardless of whether patients were treated with radical or palliative intent: 34,4 vs 10, months P = 0,00001 (radical treatment) and 14,6 vs 6,7 months P = 0,007 (palliative intent). EGFR mutations were less frequent (P = 0,046) in TTF-negative tumors compared to the control group (5,5% vs 25,3%).

Conclusions

Patients with TTF1-negative NSCLC ADC have worse overall survival, regardless of treatment intention, and a lower frequency of EGFR mutations.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Oxidative and electrophilic changes in cellular redox balance are mainly coordinated by the Keap1/Nrf2 axis and is strongly related to tumor progression, chemo- and radio-therapy resistance of cancer cells. In lung tumors this system is constitutively activated mainly by the loss of Keap1 or gain of Nrf2 functions due to point mutations in the key interacting domains of these two proteins. Beside of genetic lesions, Keap1/Nrf2 epigenetic abnormalities, as aberrant Keap1 promoter methylation and regulation by microRNAs were reported as emerging mechanisms of deregulation. Here we investigated the contribution of miRNA machinery on the modulations of Keap1/Nrf2 activity in lung tumors by analyzing panels of NSCLC and SCLC cell lines.

Methods

We searched for candidate miRNAs interacting with Nrf2 and Keap1 by using a combination of published data and in silico analyses performed by multiple bioinformatics tools (miRTarBase for known miRNAmRNA interactions; TargetScan, MiRanda, microRNA.org, miRBase also for predicted interactions). After this preliminary analysis we selected a list of 11 miRNAs that are experimentally validated in other tumors and/or predicted to be associated to NRF2 or KEAP1 and profiled their expression levels by real-time PCR in lung cancer cell lines and tissues.

Results

miR-27 family (miR-27a and miR-27b) was found to be significantly downregulated in NSCLC and SCLC cell lines. Conversely, miR-200 family (miR-200a and miR-141) was found to be significantly upregulated. Afterward, the expression data for miR-27 family and miR-200a was validated on an available training set of 29 tumor/normal paired tissues from NSCLC patients. As expected, miR- 200a was significantly up-regulated (p < 0,01, t-test), whereas miR27a and miR-27b significantly downregulated (p < 0,001, t-test) in tumors compared to normal tissues.

Conclusions

Since the obtained results refer to currently un-investigated miRNAs related to Keap1/Nrf2 axis in lung cancer, we plan to extend our analysis and confirm their role and impact on KEAP1/NRF2 modulation by in vitro studies. Our preliminary results suggest that redoxi-miRNA in lung cancer should add a new order of complexity to the regulation of the Nrf2/ARE pathway and will need to be more deeply explored in the future.

Legal entity responsible for the study

The authors.

Funding

AIRC.

Disclosure

All authors have declared no conflicts of interest.

Background

The immunohistochemical expression of PD-L1 protein has been evaluated as a predictor of prognosis and response to immunotherapy, with controversial results. We herein aimed to further study its clinical relevance in patients with non-small cell lung cancer (NSCLC).

Methods

The clinical records, pathology and molecular reports of 152 patients with histologically or cytologically confirmed NSCLC, diagnosed with early-stage or metastatic disease, were retrospectively studied. The respective archival formalin-fixed, paraffin-embedded blocks were also retrieved for immunohistochemistry (IHC) testing. PD-L1 IHC was performed using the Ventana PD-L1 (SP263) assay on an automated staining system (Ventana BenchMark ULTRA), and the results were correlated with demographic, clinicopathological and molecular features of patients.

Results

The most common subtype of NSCLC in our cohort was adenocarcinoma (82.2%). Overall, positive PD-L1 expression was observed in 47.4% of patients’ samples. EGFR and KRAS mutations and ALK rearrangement were observed in 13.0%, 7.1% and 5.4% of patients, respectively. The percentage of PD-L1 positive expression was significantly higher in cases with malignant pleural effusion (p = 0.026), more advanced disease stage (p = 0.045) and positive lymph node status (p = 0.049). PD-L1 staining showed no association with molecular data. Correlation of PD-L1 expression with the presence of pleural effusion (OR: 4.57; 95% CI: 1.20 – 17.39; p = 0.026) and positive lymph node status (OR: 1.95; 95% CI: 1.01 – 3.81; p = 0.048) was further confirmed in multivariate analysis.

Conclusions

Our study results support the independent correlation of positive expression of PD-L1 protein with adverse prognostic parameters and aggressive features, such as the presence of malignant pleural effusion and infiltrated lymph nodes.

Legal entity responsible for the study

National and Kapodistrian University of Athens.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Increasing evidence has shown that FOXP3 is not only expressed in immune cells but also in tumor cells. However, the results of tumor FOXP3 is inconsistent and even the opposite. Unlike mice, humans express multiple isoforms of FOXP3. However, the expression and role of FOXP3 isoforms in cancers are still largely unknown. The role of FOXP3 and FOXP3Δ3 in NSCLC remains unclear.

Methods

The expression of FOXP3 and four FOXP3 isoforms was analyzed using the RSEM method for 32 cancer types from TCGA databases. The correlation between the expression of FOXP3 and prognosis of NSCLC patients was analyzed (http://kmplot.com/). Western blots assay, colony formation assay, MTT assay, wound-healing assay, and cell invasion assay were performed to determine the cell functions and to measure the expression of related molecules after the ectopic expression of FOXP3 and FOXP3Δ2 in NSCLC cells.

Results

The expression of FOXP3 was upregulated in NSCLC tissues compared to adjacent non-tumor lung tissues. The FOXP3Δ3 was the most frequent FOXP3 isoform detected in NSCLC as well as in almost all types of cancers. The high expression of FOXP3 was correlated with poor prognosis in NSCLC. Tumor cells with FOXP3Δ3 were much more oncogenic than those without, and the ectopic expression of either FOXP3FL or FOXP3Δ3 could promote the proliferation, migration, and invasion of A549 and H23 cells. The ectopic expression of FOXP3 and FOXP3Δ3 downregulated E-cadherin and upregulated N-cadherin, snail, slug, MMP2, and MMP7.

Conclusions

FOXP3Δ3 is the major isoform of FOXP3 in NSCLC and it may be a predictive factor for NSCLC patients.

Legal entity responsible for the study

The Chinese University of Hong Kong.

Funding

Research Grants Council of the Hong Kong SAR (No: CUHK462613) and the National Natural Science Foundation of China (No: 81472742).

Disclosure

All authors have declared no conflicts of interest.

Background

Air pollution is still a major environmental problem affecting the health of the respiratory system. Studies have confirmed that air pollution is closely related to the occurrence of lung cancer. Air pollution (mainly PM2.5) has been classified as a carcinogen by the International Agency for Research on Cancer. Previous evidence of the relationship between PM2.5 and lung cancer mostly comes from countries with low PM2.5 exposure level. Exposure to fine particulate matter (PM2.5) may increase the risk of lung cancer, but the underlying mechanism is still unclear.There is no direct evidence that PM2.5 can directly induce lung cancer .The aim of our study was to investigate the atypical hyperplasia of bronchial epithelial cells induced by chronic exposure to PM2.5 in mice and the possible mechanism of PM2.5 in the pathogenesis of severe atypical hyperplasia.

Methods

30 BALB/c mice were treated with noninvasive tracheal instillation of PM2.5 suspension at different doses (2.5mg/kg, 5mg/kg, 10mg/kg) for 90 days (one time every 3 days), and the blank group and saline group were set as a control group. Tissue samples were taken at the end of the endotracheal infusion. Histopathological changes of lung tissue in mice and the expression of TTF-1, CK7 and Ki67 were detected by Hematoxylin-Eosin staining and immunohistochemical staining respectively.

Results

Histopathological changes showed atypical hyperplasia of bronchiolo epithelia, part of glands papillary hyperplasia, some luminal glands crowded and disordered, glandular cavity back-to-back phenomenon, pulmonary septum widened, alveolar macrophages with engulfed particles and lymphocyte aggregation in bronchiole and alveolar, especially at high dose exposure group. Electron microscopic observation showed that normal structure of alveolar and bronchial epithelial cells disappeared and a large number of autophagic bodies were formed in the alveolar and bronchial epithelial cells of mice. The expression of TTF-1, CK7 and Ki67 in the lung tissue with 10mg/kg PM2.5 were positive.

Conclusions

Chronic exposure to PM2.5 induced severe atypical hyperplasia of the bronchial epithelial cells in mice, it suggested that the model of severe atypical hyperplasia of lung in mice was established.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Checkpoint inhibitors targeting programmed death receptor (PD)-1 or PD-ligand 1 (PD-L1) became the cornerstone in the treatment of advanced NSCLC. Several phase III trials showed a better overall survival by treating with combination chemotherapy and checkpoint inhibition, suggesting that addition of chemotherapy increased the response to checkpoint inhibitors. Recently, peripheral blood biomarkers such as Ki67+PD-1+CD8 cells were found to be predictive for clinical outcome with PD-1 treatment. Knowing more about immune modulatory capacities of chemotherapy can help us to design better treatment strategies. We investigated the immune-modulatory effects of paclitaxel/carboplatin/bevacizumab (PCB), focusing on known immune populations associated with response to checkpoint inhibitors in peripheral blood.

Methods

In the NVALT 12 study, 223 patients with advanced NSCLC were enrolled to receive PCB, with or without nitroglycerin patch. At baseline and after the first and second treatment cycle, peripheral blood was drawn. By flow cytometry, the proportions of T cells and several subsets and co-inhibitory receptors of these, B cells and monocytes were determined.

Results

6 weeks after starting treatment with PCB, the proportions of T cells were significantly increased compared to baseline values. Within the T cells subsets, proliferation of CD4 T cells remained stable whereas proliferation of CD8 T cells (Ki67+) were significantly increased. The proliferating Ki67+ CD8 T cells expresses more PD-1 compared to non-proliferating CD8 T cells. However, patients with >2 fold increased proliferation of T cells did not show a better outcome.

Conclusions

Paclitaxel/ carboplatin/ bevacizumab induces proliferation of CD8 T cells which expresses more co-inhibitory checkpoint molecules. Progression free and overall survival was unchanged by this increase on its own, showing the rationale to combine PCB with checkpoint inhibition in lung cancer, as used in Impower 150.

Clinical trial identification

NCT01171170.

Legal entity responsible for the study

The authors.

Funding

NVALT.

Disclosure

D. Dumoulin: Speakers fee: BMS, Roche, Pfizer, Novartis. A-M.C. Dingemans: Advisory boards, lectures: Roche, Eli Lilly, Boehringer Ingelheim, AstraZeneca, Pfizer, BMS, Amgen, Novartis, MSD, Takeda (honoraria to institution). J.G. Aerts: Advisory boards: BMS, Boehringer Ingelheim, MSD, AstraZeneca, Eli Lilly, Takeda, Amphera; Stock owner: Amphera B.V. All other authors have declared no conflicts of interest.

Background

iSenescence is a remodeling of immune functions with a multifactorial etiology (i.e. aging, chronic inflammation, cancer). Although the absence of CD28 and the expression of CD57 and KLRG1 on circulating T-lymphocytes are hallmarks of iSenescence, the characterization of such phenotype in aNSCLC pts and the correlation with clinical characteristics and benefit from IO or PCT are currently unknown.

Methods

A senescent immune phenotype (SIP) defined as % of circulating CD8⁺CD28^-CD57⁺KLRG1⁺ T-lymphocytes was assessed by flow cytometry (FC) on fresh blood from aNSCLC pts treated with IO or PCT in a single institution. A log-rank maximization method was used to identify a SIP cut-off level and dichotomize pts accordingly. The objective was to correlate SIP with clinical characteristics and RECIST response by univariate logistic regression analysis.

Results

37 aNSCLC pts were evaluable for SIP before IO: 32% ≥ 65 years, 91% non-squamous, 43% KRAS mutated, 51% with PD-L1 expression ≥1%, 8% chemotherapy naïve. 43% had PD, 41% stability (SD), 16% partial response (PR). Median PFS and OS were 2.7 (95% CI 1.8; 7.3) and 13 (95% CI 4.8-NR) months, respectively, median follow-up was 9.3 (95% CI 6.2-14.9) months. SIP (% CD28^-CD57⁺KLRG1⁺) median value on circulating CD8⁺ lymphocytes was 12.2% (min 1.7%, max 56.1%). 32% of pts had >20.47% CD8⁺ lymphocytes with a CD28^-CD57⁺KLRG1⁺ phenotype, being classified SIP⁺. SIP status did not significantly correlate with age, pts’ characteristics or CT exposure. 2 (17%) of 12 SIP⁺ had PR/SD (DCR), vs 19 (76%) of 25 SIP^- pts (p = 0.001); median PFS was significantly lower in SIP⁺ (1.5 months 95% CI 1;2.2) vs SIP^- pts (7.4 months 95% CI 5.5, 9.3) (p = 0.001). Among 61 aNSCLC pts treated with 1^st line PCT, 18% had PD, 43% SD, 39% PR. SIP median value on circulating CD8⁺ lymphocytes was 17.9% (min 0.89%, max 66.1%), 43% of pts were SIP⁺. SIP did not significantly correlate with DCR (OR: 0.82, 95% CI 0.22-3.13, p = 0.82) upon PCT.

Conclusions

iSenescence, monitored by FC measurement of 3 surface molecules on circulating CD8 ⁺ lymphocytes, is observed in 32% and 43% of aNSCLC pts before IO or PCT, respectively. SIP correlated with lower DCR upon IO and not PCT.

Legal entity responsible for the study

Institut Gustave Roussy.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

The tumor mutation burden (TMB) is emerging as a prognostic and predictive marker for the response to immune checkpoint blockade (ICB) drugs. We aimed to develop a new method for the definition of gene panels that can precisely estimate the TMB with a considerably lower amount of genome.

Methods

We developed a bioinformatic pipeline which allows the design of gene panels suited for TMB estimation. The method is particularly efficient in optimizing the balance between the precision of the TMB estimate and the length of the gene panel created. We tested in silico the efficiency of different panels obtained with our method in an independent cohort of patients with lung adenocarcinoma (LUAD). We also compared in the same cohort of patients the performance of our panels with the performance of existing gene panels.

Results

We designed a 0.080 Megabases (Mb) long gene panel which estimated TMB in an independent LUAD cohort with an acceptable precision (adjusted R²=0.745; Spearman ρ = 0.827, Pearson ρ = 0.864). The panel showed 0.89 accuracy in the identification of TMB-high patients (25/28 patients, CI: 0.73 – 0.96). Every unitary increase of our TMB estimate was associated with lower risk of disease progression (univariate analysis: HR = 0.78; CI: 0.65-0.93; p = 0.006; multivariate analysis: HR = 0.8; CI: 0.63-1.01; p = 0.0621). Different existing panels of less than 1 Mb long showed a lower adjusted R² when compared to our gene panel (Table). Two other commercial panels of 1.9 Mb and 1.1 Mb showed a similar adjusted R² to panels of the same lengths built with our method; nevertheless, they showed a lower accuracy in TMB-high patients definition (ROC curves AUC of 0.862, 0.870, 0.946 and 0.967 were observed, respectively, for the commercial 1.9 Mb panel, the commercial 1.1 Mb panel, our 0.080 Mb panel and our 2.0 Mb panel)Table: 24P

Conclusions

Our method allows the design of gene panels particularly suitable for the TMB estimation in patients with LUAD and outperforms existing gene panels less than 1 Mb long.

Legal entity responsible for the study

Paolo Manca.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Treatment resistance and metastasis are linked to cancer stem cells (CSCs). This population represents a promising target, but remains unexplored in lung cancer. The main objective of this study was to characterize lung CSCs and discover new therapeutic strategies.

Methods

The study was performed on NSCLC cells from 8 resected patients and 12 cell lines. Suspension cultures (tumorspheres) were established for CSCs enrichment and differentiated tumor cells were cultured as monolayers (2D). The CSCs properties of tumorspheres were assessed in vitro and in vivo. The expression of 60 CSC-related genes was analyzed by RTqPCR and the expression of 12 proteins was evaluated by immunoblot (IB) and immunofluorescence (IF). High-throughput screening was performed using Prestwick and Myria libraries. Selected drugs were administered intraperitoneally to NOD/SCID mice with tumors induced by NSCLC patient and H1650 tumorspheres.

Results

Lung tumorspheres showed unlimited exponential growth (>30 passages), great tumor initiation potential, differentiation capacity, and high resistance to chemotherapy agents, but not to salinomycin. Tumorspheres had significantly higher expression of CSC-related genes (ALDH1A1, KLF4, NANOG, CD44, CD90, CDKN1A, JUNB, MDM2), invasion promoters (MMP9, SNAI1, ITGA6), and Notch (NOTCH1, NOTCH3, DLL4, JAG1) and Wnt (CTNNB1, GSK3B) components than their paired adherent-cultured cells. IB confirmed the overexpression of proteins encoded by CD44, NANOG, CDKN1A, SNAI1, ITGA6, and NOTCH3, and IF showed different localization patterns on lung adenocarcinoma tumorspheres compared with the 2D cultures. Three novel drugs [Disulfiram (DSF), Compound 1 (COMP1) and Compound 2 (COMP2)] with greater cytotoxic potential against lung tumorspheres than monolayer cells were identified. These results were validated in vivo, demonstrating the capacity of these drugs to reduce tumor growth in mice.

Conclusions

Tumorspheres are a useful culture platform for CSCs characterization in a simple and cost-effective way. We found three drugs which are able to diminish the formation and viability of tumorspheres, constituting promising therapies against lung CSCs. Supported by CB16/12/00350, PI12-02838, and PI15-00753 from ISCIII.

Legal entity responsible for the study

Fundación de Investigación Hospital General Universitario de Valencia.

Funding

Centro de Investigación Biomédica en Red de Oncología (CIBERONC) and Instituto de Salud Carlos III (ISCIII).

Disclosure

All authors have declared no conflicts of interest.

Background

The introduction of immune-checkpoint inhibitors (ICIs) in the management of aNSCLC has led to great outcome improvement, but reliable predictive biomarkers are still a need. Liquid biopsy has the potential to monitor biological effects of treatment. Aim of the study is to explore the potential predictive value of its dynamic analysis in aNSCLC treated with ICIs.

Methods

aNSCLC patients consecutively treated with ICIs at Istituto Oncologico Veneto were prospectively enrolled and genotyped in tissue. Plasma samples were collected at baseline (T1), after three or four weeks according to the administration schedule (T2) and at the moment of the first radiological evaluation (T3). Patients carrying KRAS mutation in tissue were analyzed in plasma with droplet digital PCR (ddPCR). Semiquantitative index of fractional abundancy of mutated allele (MAFA) was used.

Results

aNSCLC patients (N: 54) were prospectively enrolled and tissue genotyped, 24 of them carried KRAS mutation in tissue and 11 (46%) were positive in plasma at baseline. Positivity was not associated with tumor burden or other clinical features. We evaluated the impact of the presence and the quantitative variation of MAFA during treatment on outcome in terms of progression free-survival (PFS). After a median follow-up of 10.9 months, the presence of sentinel mutation at T1 did not affect PFS, whereas the MAFA increase from baseline to T2 and to T3 were associated with shorter PFS (HR: 5.9, 95%CI: 1.2-27.7, p:0.02 and HR: 12.1, 95%CI: 2.3-23.8, p:0.003, respectively). Median PFS was 5.2 months in the presence of MAFA increase T1-2, while it was not reached in case of decreased/stable MAFA.

Conclusions

Increase in MAFA from baseline to three or four weeks after the start of ICI is associated with shorter PFS. Predictive value of dynamic analysis of sentinel mutations in plasma during ICIs treatment warrants further validation in aNSCLC.

Legal entity responsible for the study

Istituto Oncologico Veneto.

Funding

Istituto Oncologico Veneto.

Disclosure

All authors have declared no conflicts of interest.

Background

The introduction of cancer immunotherapies has improved survival rates of certain groups of patients. According to first studies, that used whole exome sequencing, the accumulation of tumor neoantigens and therefore tumor mutational burden (TMB) could be used as molecular marker for prediction. In this study, large gene panel assays of different suppliers were tested on 28 DNA samples derived from formalin fixed, paraffin-embedded (FFPE) tumor tissue with regard to implementation in routine diagnostics.

Methods

After DNA isolation the following kits of three different suppliers have been used for library preparation: A custom SureSelect^XTHS Target Assay (Agilent Technologies, Santa Clara, CA, USA), the NEOplus v2 RUO assay (New Oncology, Cologne, Germany) and the TruSight Oncology 500 kit (Ilumina, San Diego, CA, USA, TSO500) following manufacturers’ protocol. All assays were sequenced on the NextSeq 500 system (Illumina). For quality control the TapeStation 4200 System (Agilent Technologies) was used.

Results

All kits tested showed different final concentration after library preparation. Fragment sizes were as expected and libraries were suitable for sequencing. In terms of time needed for library preparation the NEOplus v2 RUO assay was the most laborious kit. The resulting TMB-scores showed similar picture for all kits and samples tested. For all panels access to the results of the sequenced genes and detected variants is possible. Data access differs between the developers in terms of availability and manageability. Since not all kits use unique molecular identifiers (UMI) the filtering capabilities are different.

Conclusions

In this study, three different gene panel assays for TMB-determination were tested. All panels provided a similar TMB-score for the analysed samples. In terms of usage and in respect of implementation in routine diagnostics, time and efficiency are the most important parameters. Here, the TSO500 kit and the SureSelect^XTHS Target Assay showed the best potential for the use in routine diagnostics.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

C. Heydt: Honoraria: AstraZeneca, BMS, Illumina. S. Merkelbach-Bruse: Honoraria: AstraZeneca; Advisory role: AstraZeneca, Roche. All other authors have declared no conflicts of interest.

Background

Identification of activating mutations in the EGFR enabled to change the therapeutic approach for NSCLC patients as therapeutic targets. Liquid biopsy is a new option for analysis of biomarkers with valuable prognostic and predictive information, allowing the detection of EGFR mutations with a reproducible and minimally invasive approach. The aim of this study was to correlate EGFR mutational status in ctDNA at diagnosis and follow-up of NSCLC patients as a complementary/alternative to tissue-based molecular profiling.

Methods

Study included 20 patients with EGFR-mutated advanced NSCLC. Blood samples were collected at diagnosis, during follow-up and at progression. CtDNA was obtained from plasma and EGFR mutations were assessed by BEAMing dPCR (Sysmex®). Concordance between tissue and plasma EGFR mutation status was calculated as the number of positive plasma samples out of the total number of tissue samples. Cases at which T790M were first detected in blood were compared to date of progression as determined by radiological imaging in standard practice.

Results

A total of 262 plasma samples from 20 patients were analyzed, 75% percent of patients were females; 60% had never smoked; with a median of 18 months of follow- up (range: 5.13-61.77). The Positive Percent Agreement (PPA) at baseline for EGFR del19 and L858R mutation status between plasma and tissue was 75%; in 12 cases the clearance of the primary mutation happens during the first four to eight weeks after initiation of EGFR TKI. Furthermore, eight patients with radiological progression presented the resistance mutation T790M (66%), remarking that the majority of these tumors presented again the sensitizing mutation. In 6 of these patients plasma T790M-positivity was detected an average of 16 weeks (range: 4-24) prior to radiological progression. These results suggest that periodic monitoring of EGFR status in ctDNA could provide information regarding diagnoses, response or resistance to TKI-treatment.

Conclusions

Analyses of EGFR mutations in cfDNA have important clinical applicability and can be a useful alternate biomarker of diagnoses and early detection of TKIs resistance mechanisms.

Legal entity responsible for the study

Fundación de Investigación del Hospital General Universitario de Valencia.

Funding

Work supported in part by AstraZeneca (ISSIRES 0110); CIBERONC (CB16/12/00350); López-Trigo Grant.

Disclosure

All authors have declared no conflicts of interest.

Background

T790M mutation detection previos to 1^st/2^nd generation EGFR TKIs treatmentis rare by conventional methods (1-8%), although PCR ultrasensitive methods increase their prevalence to 34-80%. The aim of the study is to evaluate the presence of T790M mutations by 2 different ultrasensitive assays and their impact efficacy.

Methods

127 patients (pts) with NSCLC and EGFR mutations were analyzed. 73 of them had sufficient pretreatment tissue for T790M testing. The T790M mutation was screened by PNA Clamp TaqMan Assay (Applied Biosystems) and by ddPCR (BioRad).

Results

The limit of detection (LOD) of variant allele frequency was 0.081% and 0.136% for PNA Clamp and ddPCR, respectively. Prevalence of T790M mutations pretreatment was 23% (17/73) and 32% (23/71), respectively. A total of 71 samples gave a certain result in both methodologies (71/73; 95.89%). Agreement between the results of both techniques was 91.55%. Finally, 31 pts who met the eligibility criteria: diagnoses of advanced NSCLC, treated with 1^st/2on generation TKIs, have pre-treatment tissue for the analysis and the result of the T790M ultrasensitive analysis was conclusive; were selected to assess the relationship between the results of T790M determined by ultrasensitive methods and the prognosis. Nine (29%) of them had pretreatment T790M+ mutation by both ultrasensitive methods. Median PFS and OS were superior for patients with T790M detected by ultrasensitive PCR assay pre-treatment with TKIs: 24,9 vs. 19,6 m (p = 0.107) and 47,3 vs. 19,9 m (p = 0.76), respectively. To evaluate the association between T790M pretreatment and in the moment of progression we selected 16 pts with T790M mutation analyzed at progression (8 T790M+ and 8 T790M-). T790M at the time of progression was detected in 7 (63,6%) of 11 T790M- diagnosis patients, whilst only 1 (20%) of 5 T790M+ diagnosis patients were positive at the time of progression.

Conclusions

Both methods were highly effective for detection of EGFR T790M mutation. The detection of T790M mutation pretreatment of 1^st/2^ndgeneration EGFR TKIs by ultrasensitive PCR assay was associated with better PFS and OS.

Legal entity responsible for the study

Precision Medicine and Biomarkers Unit. IISLAFE.

Funding

AstraZeneca.

Disclosure

O.J. Juan Vidal: Honoraria, advisory role: Boehringer Ingelheim, Bristol-Myers Squibb, Merck Sharp & Dohme, Roche/Genetech, AstraZeneca, Pfizer, Eli Lilly, AbbVie; Institutional research funding: Bristol-Myers Squibb, AstraZeneca. All other authors have declared no conflicts of interest.

Background

Insulin-like growth factor 1(IGF-1) signaling pathway has been suggested as an important oncogenic mediator in various malignancies, including lung cancer. In this study, we aimed to evaluate serum levels of IGF-1, IGF-2 and IGF-Binding Protein 3 (IGF-BP3) before and after chemotherapy or chemoradiotherapy treatment in patients with locally advanced/metastatic non-small cell lung cancer (NSCLC) and their potential correlation with response to therapy and patient survival.

Methods

Seventy-three patients with stage III/IV NSCLC were included in the study. Expression of IGF1, IGF2, and IGFBP3 was evaluated in peripheral blood samples in two separate time points, at baseline and 3 months post treatment. Analysis and quantification of circulating levels of IGF-1, IGF-2, IGF-BP3 were performed by total ELISA technique. IGF-1/IGF-BP3 ratio is considered an indicator of IGF-1 bioavailability.

Results

In univariate analysis, the median value of IGF-1 after treatment (130.80 vs 98.00 ng/ml, p = 0.087), and the median ratio of IGF-1/IGF-BP3 (0.01044 vs 0.00678, p = 0.056) were higher in responders. Further analysis of the variation of each biomarker before and after treatment showed that the median value of IGF-1 was decreased after treatment in the total population (125.82 from 133.4 ng/ml, p = 0.087), albeit no difference was found between subgroups. In addition, the median value of IGF-1/IGF-BP3 ratio was found to be lower after treatment in the total population (0.01006 from 0.01252, p = 0.011). Importantly, the post-treatment value of the ratio was significantly reduced in responders (0.01044 from 0.01255, p = 0.02). No correlation was found between variation in biomarker values, patient OS and PFS.

Conclusions

In a cohort of patients with stage III/IV NSCLC treated with chemotherapy or chemoradiotherapy, reduction of IGF-1/IGF-BP3 ratio, which is an indicator of IGF-1 bioavailability, was statistically significant in patients that responded to treatment. If validated in larger cohorts, our results support the use of IGF-1/IGFBP3 as a predictive tool for response to chemotherapy in NSCLC.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Bronchoalveolar lavage fluid (BALF) examination was found to be useful in the evaluation of local immune status in the site of lung cancer development. This procedure may be performed during lung cancer diagnosis and repeated during therapy. It especially concerns the majority of lung cancer cases, which are beyond the possibility of resection at recognition. However, the procedure of BAL in the lung affected by cancer is often complicated. The aim of this study was an analysis of differences in the character of immune response between the lung affected by cancer versus the opposite, ’healthy‘ lung. May BALF from the lung free of tumor serve for the characterization of local immune status?

Methods

BALF of 36 patients was harvested from the lung affected by cancer and opposite site: the lung without tumor. BALF cells were analyzed by means of the following sets of antibodies anti: the first probe CD4, CD127, CD25, Foxp3, CTLA4, the second probe: CD3, CD16/CD56, CD45, CD19, CD8. For macrophage phenotyping the immunofluorescent method with antibodies anti: CCR7 and CD163 was used. The concentrations of 27 cytokines were analyzed in the duplicate using Bio-Plex Pro Human Cytokine Assay. The next part of the study was performed in other 32 patients by the same method with a panel of antibodies anti: CD3, CD4, CD8, CD127, CD45RA, CD69, CTLA-4, PD-1.

Results

By the analysis of the BALF profile we found that only the cells and cytokines involved in regulatory pathways showed significant differences between both the ’healthy‘ and the cancerous lung, being elevated in the latter. The proportion of Tregs, CTLA4+Tregs, activated and memory PD-1+ lymphocytes, CCR7_lowCD163^high macrophages were significantly higher in the lung with cancer than in the healthy lung. On the other hand, the proportion of effectory cells and inflammatory cytokines concentration did not differ between both lungs, which may indicate that they form an integrated immune system and individual homeostasis.

Conclusions

Regulatory mechanisms which are up-regulated in the tumourbearing lung affect BALF profile in the lung with cancer, while the examination of BALF from the “healthy” lung may serve for the evaluation of individual basic profile of the anti-cancer response.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

J. Domagala-Kulawik: Travel grant, consultant: BMS, MSD. All other authors have declared no conflicts of interest.

Background

Exosomes are membranous vesicles around 40-130 nm secreted by cells, carrying key information to distant tissues. Exosomes have been detected in different types of clinical samples and may play a key role in NSCLC. They intervene in several processes such as angiogenesis, metastatic niche formation, immunosuppression; being relevant in stem cell differentiation. The main objective of this study was to analyze exosomal cargo from NSCLC cell lines and primary cultures under two conditions: suspension cultures with cancer stem cells features (3D tumorspheres) and adherent cultures (2D).

Methods

Cell cultures were established from NSCLC resected patients and cell lines. Exosomes isolation was performed by ultracentrifugation. Characterization was carried out by NTA, electron microscopy, immunoblot and flow cytometry. Exosomal DNA was extracted to determine the mutational status of EGFR and RAS genes by BEAMing dPCR (Sysmex®). Transcriptomic analysis has been carried out from exosomal RNA with Clariom D Human microarrays (Affymetrix®), (p ≤ 0.01).

Results

Regarding exosomes characterization, NTA and electron microscopy showed an exosome size from 108-125 nm. Exosomes surface markers were detected by immunoblot and flow cytometry. Mutational analysis of EGFR and RAS genes shown the same pattern displayed by the origin cells. Transcriptomic analysis showed an expression of mRNAs, miRNAs and precursors significantly different between 3D and 2D exosomes. Afterwards, their targets were identified and a pathway enrichment analysis was performed to know in which biological processes (cancer-related) are involved. Significant differential expressions were also found between ADC vs SCC-derived exosomes. Interestingly, 7 exosomal miRNAs (miR-200c, 29a, 339, 224, 31, 21, 33a) had already been identified as overexpressed in NSCLC tissue by our group. Moreover, miR-339 and miR-21 were related to prognosis (p < 0.05) in ADC.

Conclusions

Differences in exosomal cargo have been observed between: i) 3D vs 2D cultures and ii) ADC vs SCC. In addition, the same mutational pattern was detected in exosomes as compared with parental cultures. Therefore, exosomes can be a useful source for biomarkers analysis in NSCLC.

Legal entity responsible for the study

Fundación de Investigación del Hospital General Universitario de Valencia (FIHGUV).

Funding

Supported by grant GV/2018/026 & Asociación Española contra el Cáncer (AECC Valencia).

Disclosure

All authors have declared no conflicts of interest.

Background

According to results of PACIFIC trial, immune checkpoint inhibitors (CPIs) are an integral part of the multimodal treatment approach for inoperable locally-advanced non-small cell lung cancer (LA-NSCLC). The purpose of this single-center study was to investigate a prognostic value of CD8-positive tumor stroma-infiltrating lymphocytes (TILs) and PD-L1 positive tumor cells at initial biopsy in patients treated with definitive chemoradiotherapy (CRT).

Methods

We collected initial tumor tissue samples and reviewed the clinical characteristics of 36 LA-NSCLC patients treated at our center with definitive CRT between 2000 and 2004. The analyzed tumor tissue was taken before start of multimodal treatment. An experienced pathologist performed the immunohistochemistry analysis and interpretation. Tumor cells and CD8-positive tumor stroma-infiltrating lymphocytes were analyzed separately. Based on PD-L1 expression on tumor cells, patients were divided into three subgroups (0%, 1-5%, >5%). Two patient subgroups were defined according to CD8 expression in the tumor-surrounding stroma (low and high density: 0-40% vs. 41-100%).

Results

Thirty-six patients with LA-NSCLC were treated with definitive CRT (stage II, III and IV: 2, 30 and 4 patients). 18 (50%) patients had squamosa cell carcinoma, 14 patients (39%) adenocarcinoma and 4 patients (11%) NOS. Patients with high density of CD8 positive tumor stroma-infiltrating lymphocytes (TILS) showed a significantly reduced overall survival than patients with low density (13 vs. 19 months, one-year survival 56.5 vs. 69.2% p = 0.047); Patients without (0%) and low expression (1-5%) of PD-L1 on tumor cells showed a significantly improved overall survival compared to patients with a PD-L1 expression on tumor cells over 5% (median survival: 13.8 vs. 6.6 months, 1-year survival rate: 67.7 vs. 33.3%; p = 0.039).

Conclusions

High density of CD8-positive tumor stroma-infiltrating lymphocytes and PDL1 expression on tumor cells over 5% at initial biopsy correlates significantly with reduced overall survival in patients with LA-NSCLC treated with CRT.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Approximately 80% of lung cancer is non-small cell lung cancer (NSCLC) with a median survival time of 8 to 10 months. It is critical and urgent to find the effective interventions for patients suffering from NSCLC. Histone deacetylase 2 (HDAC2) is increasingly associated with development of several solid tumors. However, its role in lung cancer remained mostly lacking.

Methods

Immunohistochemistry was used to analyze the distribution patterns of HDAC2 protein in lung cancer and the results were further statistically correlated with clinicopathological characteristics and survival rate. XTT, invasion assay, flowcytometry, immunoblotting, and immunoprecipitation were used to explore the bio-impacts and underlying mechanisms of HDAC2 and its interplay with nicotine in lung cancer cells.

Results

The Kaplan–Meier plotter shows that 1926 lung cancer patients with high HDAC2 mRNA levels will have a poor overall survival rate (p = 0.004). Increased HDAC2 mRNA expression predicts an improved overall survival of non-smoker patient group (p = 0.018) but not that of smoker patient group. Further immunohistochemistry showed similar results (p < 0.001). Intriguingly, this association seen in smoker and non-smoker groups was demonstrated in only squamous cell lung cancer, providing the specific interplay between HDAC2 and carcinogenic content of cigarette. Treatment with nicotine induced the NO production through rapidly increasing NOS2 expression and HDAC2 was subsequently nitrosylated. Knockdown of HDAC2 reduced nicotine-induced NOS2 expression and invasive activity. Ectopic HDAC2 overexpression enhanced nicotine-mediated NOS2 induction. Interestingly, ectopic overexpression of Cys262/274Ala HDAC2 mutant, un-nitrosylated, inhibited the invasive activity. HDAC2 specific siRNA and inhibitor enhanced the cytotoxicity of cisplatin and abolished the cisplatin-mediated cyclin B1 induction.

Conclusions

Our results implicated that HDAC2 might be required for cigarette carcinogenic content-mediated advanced development or treatment failure in lung cancer. NOS2 provides the link between nicotine and HDAC2 in lung cancer. Inhibition of HDAC2 activity may increase therapeutic effects of cisplatin-based treatments in NSCLC.

Legal entity responsible for the study

Yao-Tsung Yeh.

Funding

MOST, Taiwan (R.O.C).

Disclosure

All authors have declared no conflicts of interest.

Background

Primary lymphoepithelioma-like carcinoma (LELC) is a rare form of lung cancer, since genetic alternations participate in the carcinogenesis in lung cancer, whether critical driver genes such as ROS1 fusion, anaplastic lymphoma kinase (ALK) rearrangement and epidermal growth factor receptor (EGFR) mutation play roles in LELC remains unclear.

Methods

We collected a total of 30 LELC samples for genetic and prognosis analysis retrospectively in ShangHai Chest Hospital, EGFR gene mutation, ALK rearrangement and ROS1 fusion status were extracted from digital database. Clinicopathological characteristics with genetic profiling and survival were analyzed.

Results

All samples appeared negative for genes (EGFR, ALK and ROS1) alternations detection. Female gender acted as an independently prognostic factor in poorer disease-free survival (DFS) (P = 0.034), and tumors locate in the left lobe associate with worse DFS in univariate analysis (significant trend, P = 0.051), moreover, serum positive NSE level also indicate a shorter DFS after adjustment (significant trend, P = 0.057). Most of LELC are diagnosed at early stage, with 93.3% (27/30) patients obtained the opportunities for surgery.

Conclusions

Since no classical genetic alternations appeared in present study, more investigations of other genes distributions should be explored in the future for better understanding of this rare subtype of lung cancer. Serum positive NSE level seemed to be a prognostic biomarker for DFS in LELC patients.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Immunotherapy brings evident benefits to lung cancer patients. However, the answer to this treatment is highly individualized and seems to be better in ‘hot’ tumors than in ’cold’ ones. The vast majority of lung cancers are in their advanced stages at recognition, therefore the access to large fragments of tumor tissue and tumor environment is limited. The results of our previous studies showed the usefulness of bronchoalveolar lavage fluid (BALF) examination in the evaluation of immune status in a lung tumor site. The aim of this study was to assess if BALF may reflect the signs of ‘hot’ vs. ‘cold’ tumors.

Methods

36 lung cancer patients were investigated. The number of lymphocytes and the proportion of following lymphocyte subtypes were analysed by flow cytometry: T, B, T helper, T cytotoxic, NK cells, CD25⁺Th cells, T regulatory cells, CTLA4⁺Tregs. The concentration of IFNγ, TNFα, VEGF, TGFβ, IL-2, IL-10 was measured by Bio-Plex Pro Human Cytokine Assay. We defined ’hot‘ tumors by the presence of cell proportion or cytokine concentration above the median value and ’cold‘ tumors where these markers were lower than the median value.

Results

There were 36% of BALF samples with more than 2 parameters below the median values, which may reflect ’cold‘ tumors. The indications of immune response (more than 7, e.g. 50%) were elevated in 53% samples. The majority of ’hot‘ tumors were in the type of adenocarcinoma or NOS, but without any relation to the EGFR mutations nor ALK alterations. No relation between the intensity of immune response in the BALF and the clinical stage of cancer was observed.

Conclusions

BALF examination may help in distinguishing ’hot‘ and ’cold‘ lung cancer and may be useful in the characterization of local immune status before treatment. By extrapolating the results of this study we could recommend BALF as a source of useful biomarkers before immunotherapy.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

J. Domagala-Kulawik: Travel grants, consultations: BMS, MSD. All other authors have declared no conflicts of interest.

Background

During the last years there is an expansion of our knowledge in lung cancer immunology. In addition, there is a growing number of studies on the role of lymphotoxin beta receptor (LTβR), a member of the tumor necrosis factor (TNF) family, which plays an important role in lymphoid system formation and homeostasis as well as in immune system regulation mainly through NF-κB signaling. The aim of the current study was to investigate the clinical relevance of LTβR single nucleotide polymorphism (SNP) rs10849448 (A/G) with the susceptibility to NSCLC, the clinicopathological parameters, the relapse and the survival rates of NSCLC patients, as well as with the protein expression of LTβR.

Methods

LTβR SNP was genotyped in 268 randomly selected NSCLC patients and 279 age- and gender-matched healthy donors. Immunohistochemical analysis for LTβR was performed on 127 NSCLC tumors. The studied cohort was under observation during a five-year period.

Results

Genotype frequencies of rs10849448 (AA, AG, and GG) varied between healthy controls and patients, but the difference did not reach statistical significance (P = 0.054). AA homozygotes were found to have lower risk for NSCLC compared to G allele carriers in univariate as well as in multivariate analysis (both P = 0.016). Moreover, rs10849448 was associated with development of metastases with A allele carriers developing less often metastatic disease in adrenals (P = 0.013). Interestingly, rs10849448 was related to membranous LTβR protein expression (P = 0.035) in malignant cells, with AA homozygotes being associated with higher protein levels.

Conclusions

The present findings suggest that the investigated SNP rs10849448 may be associated with NSCLC initiation as well as with the development of metastatic disease. More association and functional studies are needed in order to further clarify it’s role in NSCLC.

Legal entity responsible for the study

H. Kalofonos.

Funding

This research was co-funded by the Hellenic Society of Medical Oncology (HeSMO) through a research funding program.

Disclosure

All authors have declared no conflicts of interest.

Background

Lung cancer is the leading cause of death among all other malignancies. Despite the wide use and the high efficacy of immunotherapy, most patients with advanced or metastatic disease need chemotherapeutic treatment. Primary and acquired resistance to chemotherapeutic agents are the main cause of reduced overall survival. The Hedgehog (HH) pathway has been shown to be crucial in cell development and survival. Activated in several types of cancer it might be a potent bypass mechanism mediating chemo resistance. Thus, this study aims at investigating the role of HH signalling in chemo resistant NSCLC cells.

Methods

HCC827 NSCLC cells were treated with increasing sub-lethal doses of pemetrexed and vinorelbine to produce pemetrexed and vinorelbine resistant cells. RNA was isolated from both chemo resistant and naive cells. RT-qPCR was performed to evaluated mRNA gene expression. To measure cell growth capacity we used a MTT assay to evaluate the impact of the HH-inhibitor vismodegib in naïve and chemo resistant cell lines.

Results

Pemetrexed resistant cells showed clearly increased expression levels of most of HH signalling genes. The HH transcription factors GLI1, GLI2 and GLI3 were increased 3.7, 4.3 and 3.5 fold compared to non-resistant cells. The trans-membrane HH receptor PTCH1 showed to be elevated 2.4 fold and the ligand of HH signalling SSH revealed to be expressed 5.9 fold compared to naïve cells. However, vinorelbine resistant cells did not show a significant activation of HH signalling. Supporting these results pemetrexed resistant cells treated with the HH inhibitor vismodegib showed reduced proliferation compared to naïve cells treated with vismodegib.

Conclusions

These are the first results of our study investigating the impact of HH signalling on chemo resistant NSCLC cells. In the case of pemetrexed resistance activation of HH pathway could play an important role to mediate this resistance. However, further investigation is needed to clarify the role of the HH pathway in NSCLC treatment.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

This study aimed to investigate the potential systemic anti-tumor effects of stereotactic ablative radiotherapy (SABR) and apatinib (a novel vascular endothelial growth factor receptor 2 tyrosine kinase inhibitor) via reversing the immunosuppressive tumor microenvironment for lung carcinoma.

Methods

Lewis lung cancer cells were injected into C57BL/6 mice in the right hindlimb (primary tumor; irradiated) and in the left flank (secondary tumor; nonirradiated). When both tumors grew to the touchable size, mice were randomly divided into eight treatment groups. These groups received normal saline or three distinct doses of apatinib (50 mg/kg, 150 mg/kg, 200 mg/kg) daily for 7 days, in combination with a single dose of 15 Gy radiotherapy or not to the primary tumor. The further tumor growth/regression of mice were followed and observed.

Results

Only primary tumor growth could be delayed with single SABR, while both primary and secondary tumor growth could be regressed with apatinib administration in a dose-dependent propensity. When combining SABR and apatinib, significant retardation of primary tumor growth could be observed. In addition, when combining SABR and apatinib of 200 mg/kg, growth of the secondary tumors could also be significantly inhibited (P < 0.01). This result indicated the namely “abscopal effect”. Mechanism analysis suggested that increased programmed death ligand-1 expression in tumor tissue was counteract by additional apatinib therapy. Furthermore, when apatinib was combined with SABR, the composition of immune cells could be changed and tumor specific T cell activation could be induced. More importantly, this two-pronged approach evoked tumor antigen-specific immune responses and the mice were resistant to another tumor rechallenge, finally, long-term survival was improved.

Conclusions

Our results suggested that the tumor microenvironment could be managed with high-dose of apatinib, which was effective in eliciting an abscopal effect induced by SABR.

Legal entity responsible for the study

The authors.

Funding

This work was supported by National Natural Science Foundation of China grants 81472792 and 81702813, Postgraduate Research & Practice Innovation Program of Jiangsu Province grants SJCX18_0707, and the Technology Office Foundation of Lianyungang City (NO.SH1613).

Disclosure

All authors have declared no conflicts of interest.

Background

In 2011, metabolic reprogramming was defined as one of the cancer hallmarks, and several drugs targeting cancer metabolism have been developed as anti-cancer drugs. To improve medication efficiency of these anti-metabolic drugs, we tried to sub-classify non-small cell lung cancer (NSCLC) cells by their glycolytic dependency and treat them by targeting different metabolism-related proteins.

Methods

We stratified NSCLC cells according to their Oxygen Consumption Rate (OCR) and Extracellular Acidification Rate (ECAR) measured by the Seahorse XF analyzer. Metformin, an inhibitor of mitochondrial respiratory complex I, was used as an example to measure the sensitivities of NSCLC cells in different metabolic subtypes to drugs targeting oxidative phosphorylation (OXPHOS) pathways.

Results

According to the OCR and ECAR, measured by the Seahorse XF analyser, NSCLC cells could be further separated into two metabolic subtypes, OXPHOS-dependent and glycolysis-dependent subtypes. We found that the OXPHOS-dependent subtype was sensitive to metformin, while the glycolysis-dependent subtype was resistant. To target the glycolysis-dependent subtype of NSCLC cells, monocarboxylate transporter 4 (MCT4), a proton-linked lactate transporter, was identified. MCT4 was highly expressed in the glycolysis-dependent subtype of NSCLC cells and its’ expression was important for these cells. When we knocked down the expression or blocked the function of MCT4, the proliferation of these cells was significantly inhibited.

Conclusions

In this report, we established a method to sub-classify NSCLC cells and predict the sensitivity of these cells to metformin. Besides, we identified MCT4 transporter as a druggable target to inhibit the proliferation of glycolysis-dependent subtype of NSCLC.

Legal entity responsible for the study

The authors.

Funding

Ministry of Science and Technology in Taiwan.

Disclosure

All authors have declared no conflicts of interest.

Background

Epidermal growth factor receptor (EGFR) mutation testing in plasma cell-free DNA (cfDNA) from advanced lung cancer patients is an emerging clinical tool. This non-invasive approach may offer genotyping information but prevent patients from the shortcomings of invasive and repeated biopsies. This meta-analysis was designed to determine the diagnostic accuracy of two common PCR systems, ddPCR and ARMS-PCR, for detecting EGFR mutation in cfDNA.

Methods

A systematic search was carried out based on electronic databases. Data from eligible studies were extracted and pooled to calculate the sensitivity, specificity, diagnostic odds ratio (DOR), and area under the summary receiver-operating characteristic curve (AUROC), using tissue biopsy results as the standard method.

Results

Twenty-three studies involving 4,725 cases were identified and included. The plasma testing sensitivity, specificity, DOR and AUROC, compared with the matched tumor tissues, were 71.7% (95% CI, 67.6%-75.5%), 96.4% (95% CI, 94.2%-97.9%), 39.3 (95% CI, 21.2-73.1), 0.87 (95% CI, 0.80-0.94) for ddPCR, and 65.3% (95% CI, 62.9%-67.6%), 98.2% (95% CI, 97.6%-98.7%), 52.8 (95% CI, 26.3-106.1), 0.71 (95% CI, 0.52-0.91) for ARMS-PCR, respectively. In studies simultaneously comparing the two platforms, we observed no significant difference in specificity (P = 0.473) and sensitivity (P = 0.960) between ddPCR and ARMS-PCR, regardless of EGFR mutation type, tumor stage and EGFR-TKI treatmentTable: 41P

Conclusions

This study demonstrates that both ddPCR and ARMS-PCR have a high specificity with a practical sensitivity for detecting EGFR mutation in cfDNA of advanced lung cancer patients, which supports their application as a supplement or an alternative to tissue biopsy in clinical practice for genotyping.

Legal entity responsible for the study

The First Affiliated Hospital of Guangzhou Medical University.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

As part of I-O Optimise, a multinational research platform providing real-world insights into the management of lung cancers, temporal trends for initial Tx and OS are reported for pts diagnosed with NSCLC at Leeds Teaching Hospitals Trust (LTHT), hosting one of the largest integrated cancer centres in the UK.

Methods

A retrospective cohort study using longitudinal data collected from electronic medical records at LTHT, including all adult pts diagnosed with NSCLC from Jan 2007 to Aug 2017 (follow-up to Mar 2018). Initial Tx was the 1st Tx received after diagnosis (categorised as surgery, radiotherapy or systemic anti-cancer therapy [SACT]). In this analysis, 2 diagnostic periods were compared: 2007–2012 and 2013–2017. Kaplan–Meier methods were used for OS.

Results

Among 3739 pts with incident NSCLC, TNM stage distribution was: I, 19.2% (this increased noticeably from 14.0% in 2007–2012 to 24.8% in 2013–2017); II, 11.6%; IIIA, 12.5%; IIIB, 9.0%; and IV, 47.7%. Overall, 29.7% had non-squamous cell carcinoma (NSQ), 21.9% had squamous cell carcinoma (SQ), 11.7% had NSCLC NOS (this decreased noticeably from 18.8% in 2007 to 8.4% in 2016) and 2.4% had other histology; 34.3% of pts were diagnosed without pathology. Tx patterns for NSQ and SQ pts are shown by stage and year of diagnosis (Table). Between 2007–2012 and 2013–2017, median OS (in months) increased noticeably in stage I–IIIA NSQ (28.7 to 50.0) and SQ pts (17.4 to 32.1), but not in stage IIIB–IV NSQ (4.1 to 5.0) or SQ pts (5.3 to 4.8)Table: 44P

Conclusions

OS improved in stage I–IIIA pts over time, most likely due to an increased proportion diagnosed at stage I and the evolution of initial Tx through the study period. In contrast, among stage IIIB–IV pts, SACT use remains relatively low and OS outcomes remain unchanged. Future analyses will help assess the impact of new therapies, including immunotherapy and 2nd-generation tyrosine kinase inhibitors, on Tx patterns and OS in the UK.

Editorial acknowledgement

Editorial assistance was provided by Richard Daniel, PhD, of Parexel and funded by Bristol-Myers Squibb.

Legal entity responsible for the study

Bristol-Myers Squibb.

Funding

Bristol-Myers Squibb through an EU wide RWD initiative called IO-Optimize sponsored by BMS.

Disclosure

M. Snee: Grants, personal fees: IQVIA, during the conduct of the study. L. Lacoin: Consultant epidemiologist contracted by Bristol-Myers Squibb. W. Sopwith: Personal fees: IQVIA, during the conduct of the study and outside the submitted work. C. Chaib: Employee: Bristol-Myers Squibb; Bristol-Myers Squibb (BMS) is funding this work through an EU wide RWD initiative called IO-Optimize sponsored by BMS. J.R. Penrod: Employment, stock ownership: Bristol-Myers-Squibb. J.C. O’Donnell: Employee, shareholder: BMS. All other authors have declared no conflicts of interest.

Background

Understanding Tx patterns and related outcomes in the rapidly changing NSCLC landscape informs clinical decision-making. As part of I-O Optimise, a multinational research platform providing real-world insights into the management of lung cancers, Tx patterns and OS are reported for pts with NSCLC prior to immunotherapy reimbursement in Portugal.

Methods

IPO-Porto, Portugal’s largest oncology hospital, has a research database linked to the RORENO cancer registry, covering northern Portugal. This database has collected data on 1524 adult pts with NSCLC since 2012; the current analysis includes pts diagnosed at stage IIIB–IV from Jan 2015 (when systemic anticancer therapy [SACT] data became available) to Dec 2016 (follow-up to Jun 2017). Tx patterns are shown for the first 3 Tx lines. Kaplan–Meier methods were used for OS and time to next Tx or death (TTNT) from SACT initiation.

Results

Of 595 pts diagnosed at stages I–IV in 2015–2016, 54 (9.1%) had stage IIIB and 338 (56.8%) had stage IV NSCLC (median age, 65 yrs [range: 27–90; 9.2% ≥80]; male, 77.3%; brain metastases, 16.3%; non-squamous [NSQ], 73.7%; squamous [SQ], 21.2%). Of the 372 NSQ/SQ pts, 284 (76.3%) had 1st-line SACT; 101 (27.2%) had 2nd-line, and 20 (5.4%) had 3rd-line (Table). Median OS from SACT initiation was 12.6 (NSQ) or 10.3 (SQ) months. Median OS was longer in stage IV pts without brain metastases (11.6 vs 7.6 months), and in treated NSQ pts with EGFR/ALK alterations vs wild type (14.4 vs 9.3 months). Most NSQ pts with EGFR/ALK alterations (73.0%) had a 1st-line tyrosine kinase inhibitor (TKI). Median TTNT was 11.0 months in NSQ pts on 1st-line TKI and 6.3 (NSQ) or 7.3 (SQ) months in pts on 1st-line platinum chemotherapyTable: 45P

Conclusions

Pts with advanced NSCLC have a high burden of disease, with most diagnosed at stage IV and a short OS from SACT initiation. Future analyses will assess the post-reimbursement impact of immunotherapies (and new TKIs) on Tx and OS in Portugal.

Editorial acknowledgement

Professional medical writing assistance was provided by Richard Daniel, PhD, of Parexel and funded by Bristol-Myers Squibb.

Legal entity responsible for the study

Bristol-Myers-Squibb.

Funding

Bristol-Myers-Squibb through an EU wide RWD initiative called IO-Optimize sponsored by BMS.

Disclosure

M.A.S. Soares: Advisory boards, talks: Roche; Lilly, BMS, MSD; Boehringer Ingelheim, Pfizer, AstraZeneca, Novartis, outsider the submitted work. L. Antunes, P. Redondo, M.J. Bento: Grants: IQVIA, during the conduct of the study. M. Borges: Grants: IQVIA to IPO Porto, during the conduct of the study. R. Hermans, F. Grimson, R. Munro: Employee: IQVIA Real-World Insight Solutions (vendor paid by BMS). D. Patel: Personal fees: BMS, during the conduct of the study. C. Chaib: Employee: Bristol-Myers Squibb. L. Lacoin: Consultant epidemiologist contracted by Bristol-Myers-Squibb. All other authors have declared no conflicts of interest.

Background

As part of I-O Optimise, a multinational research platform providing real-world insights into the management of lung cancers, the SCAN-LEAF study aims to describe the epidemiology, clinical care, and outcomes for pts with NSCLC in Scandinavia. Here, we report temporal OS trends among pts diagnosed with incident NSCLC from 2005 to 2015 in Denmark and Sweden.

Methods

The SCAN-LEAF Danish and Swedish cohorts were established by linking respective national registries and include all adult pts diagnosed with incident NSCLC from Jan 2005 to Dec 2015 (follow-up to Dec 2016). The Kaplan–Meier method was used to estimate OS at 1, 3, and 5 yrs by histology (non-squamous cell [NSQ] or squamous cell [SQ]), TNM stage, and yr of diagnosis; changes in OS over time were assessed using the Cochrane–Armitage test.

Results

31,939 pts in Denmark and 30,067 pts in Sweden were diagnosed with NSCLC from 2005 to 2015. Most were diagnosed at stage IV (51.6% and 48.4%, respectively) and had NSQ histology (54.4% and 60.4%). Statistically significant trends (P < 0.05) for improved OS accompanied by an absolute OS rate increase of > 5% over the analysis period were seen for NSQ pts at 1 yr for all stages in both countries (Table); at 3 yrs for stages I–IIIB in Denmark (P ≤ 0.027), and stages I–II (P ≤ 0.0013) in Sweden; and at 5 yrs for stages I–II (P ≤ 0.026) in both countries. For SQ pts, this was seen only at 1 yr for stage IIIA in Denmark and stage I in Sweden (Table), and at 5 yrs for stage IIIA in Denmark (P = 0.02)Table: 46P

Conclusions

Despite some improvements between 2005 and 2015, mainly in the short-term survival of pts with early-stage NSCLC, long-term OS rates for pts with late-stage disease did not change significantly and remained low. Even in pts with early-stage disease, OS outcomes were suboptimal, with a particular unmet need in the SQ population. Future analyses including data after 2015 will evaluate the potential impact on OS of increased use of new TKIs and immune checkpoint inhibitors.

Editorial acknowledgement

Professional medical writing assistance was provided by Richard Daniel, PhD, of Parexel funded by Bristol-Myers Squibb.

Legal entity responsible for the study

Bristol-Myers Squibb.

Funding

Bristol-Myers Squibb.

Disclosure

S. Ekman: Grants: BMS, during the conduct of the study. P. Horvat: Employee: IQVIA. D. Patel: Personal fees: BMS, during the conduct of the study. M. Roselund, A. Mette-Kejs: Fees for service to institution (IQVIA) during the conduct of this study: BMS; Employee: IQVIA. A. Juarez-Garcia: Employee: BMS. L. Lacoin: Consultant epidemiologist contracted by Bristol-Myers-Squibb for the SCAN-LEAF Project. All other authors have declared no conflicts of interest.

Background

As part of I-O Optimise, a multinational research platform providing real-world insights into the management of lung cancers, the SCAN-LEAF study aims to describe the epidemiology, clinical care and outcomes for pts with NSCLC in Scandinavia. We report initial Tx and OS for pts with NSCLC prior to the availability of immunotherapies in Sweden.

Methods

The analysis includes all adult pts diagnosed with NSCLC at Uppsala and Karolinska (Stockholm) University Hospitals from 2012 to 2015 (follow-up to Dec 2016). Electronic medical record data were extracted using Pygargus CXP software and linked with national registries. Bespoke rule-based algorithms were applied to describe Tx patterns; Kaplan–Meier methods were used to estimate OS.

Results

2779 pts were diagnosed with incident NSCLC (median age, 70 yrs [range: 22–96; 14.2% ≥80]; male, 48.5%; histology: non-squamous (NSQ), 70.9%, squamous (SQ), 17.7%, other, 11.4%; stage distribution: I, 19.3%; II, 7.7%; IIIA, 12.3%; IIIB, 7.2%; IV, 51.2%). Initial Tx (≤6 months from diagnosis) by stage and yr of diagnosis is shown in the table. Median OS (months) for NSQ and SQ pts: not reached and 52.8 in stage I, 43.2 and 23.6 in stage II, 26.7 and 20.4 in stage IIIA, 12.5 and 12.9 in stage IIIB, and 7.6 and 6.1 in stage IV, respectively. Among stage IIIB–IV pts, 60.7% (NSQ) and 53.5% (SQ) had ≥1 line of systemic anti-cancer therapy (SACT); median OS was 12.2 (NSQ) and 10.4 (SQ) months in pts on SACT, and 3.1 (NSQ) and 3.7 (SQ) months in pts not on SACT. Ongoing analyses will assess factors associated with SACT receipt in stage IIIB–IV ptsTable: 47P

Conclusions

Swedish pts with NSCLC had a high burden of disease, with most diagnosed at stage IV and a median OS of ∼1 yr in late-stage pts receiving SACT. There is also scope for improved prognosis in pts diagnosed at early stages, particularly in SQ pts. Future analyses will assess the potential impact of recent improvements in diagnostics and therapeutics on Tx patterns and OS in Swedish NSCLC pts.

Editorial acknowledgement

Professional medical writing assistance was provided by Richard Daniel, PhD, of Parexel funded by Bristol-Myers Squibb.

Legal entity responsible for the study

Bristol-Myers Squibb.

Funding

Bristol-Myers Squibb.

Disclosure

S. Ekman: Grants: BMS, during the conduct of the study. P. Horvat, A. Mette Kejs: Employee: IQVIA. D. Patel: Personal fees: BMS, during the conduct of the study. M. Rosenlund: Employed: BMS, during the conduct of the study. A. Juarez-Garcia: Employed: BMS, outside the submitted work. L. Lacoin: Consultant epidemiologist contracted by Bristol-Myers-Squibb for the SCAN-LEAF Project. All other authors have declared no conflicts of interest.

Background

Education has been shown to be inversely associated with the incidence of lung cancer at several conventional observational studies. However, this association may be biased owing to the methodological limitations of traditional observational study-confounding, reverse causation, and measurement error. Therefore, we aimed to investigate whether more years spent in education is causally associated with risk of lung cancer through a two sample mendelian randomisation study.

Methods

The main analysis used publicly available genetic summary data from two large consortiums (International Lung Cancer Consortium (ILCCO) and Social Science Genetic Association Consortium (SSGAC)). Genetic variants used as instrumental variables for lung cancer and years of education were derived from two large genome wide association studies: ILCCO and SSGAC, respectively. Finally, genetic data from three additional consortia (TAG, GLGC, GIANT) were analyzed to investigate whether longer education can causally alter the common lung cancer risk factors. The exposure was the genetic predisposition to higher levels of education, measured by 73 SNPs from SSGAC. The primary outcome was the risk of lung cancer (11348 events in ILCCO). Secondary outcomes based on different histologic subtypes were also examined.

Results

Genetic predisposition towards 3.6 years of additional education was associated with a 52% lower risk of lung (odds ratio 0.48, 95% confidence interval 0.34 to 0.66; p = 1.02 × 10−5). Sensitivity analyses were consistent with a causal interpretation in which major bias from genetic pleiotropy was unlikely. The Mendelian randomisation assumptions did not seem to be violated. Genetic predisposition towards longer education was additionally associated with less smoking, lower body mass index, and a favorable blood lipid profile. Table: 48P

Mendelian randomisation estimates of the associations between education attainment and risk of lung cancer overall and histologic types

Conclusions

Our present mendelian randomisation study provided strong evidence to support that higher education attainment plays a causal role in lowering the risk of lung cancer. Furthermore, more work is needed to elucidate the potential mechanisms which mediate the association between education and lung cancer.

Legal entity responsible for the study

The authors.

Funding

National Key R&D Program of China (Grant No. 2016YFC0905500, 2016YFC0905503).

Disclosure

All authors have declared no conflicts of interest.

Background

Crohn’s disease is associated with increased lung cancer risk in observational studies, but the causality of this association is uncertain. Here we conducted a two-sample Mendelian randomisation (MR) study to examine whether Crohn’s disease is causally related to lung cancer.

Methods

We used the summary data of the International Inflammatory Bowel Disease Genetics Consortium (IIBDGC, 5,956 cases and 14,927 controls; European ancestry) and International Lung Cancer Consortium (ILCCO, 11348 lung cancer cases and 15861 controls; European ancestry). 52 single nucleotide polymorphisms (SNPs) associated with Crohn’s disease were used as instrumental variables in the MR analysis. We utilized various MR methods (inverse-variance weighting (IVW), MR Egger, weighted median regression) to explore the causality. To investigate whether the effect of Crohn’s disease is associated with the histology of lung cancer, we also performed similar analyses on two different histologic subtypes of lung cancer (adenocarcinoma (LUAD) and squamous cell carcinoma (LUSC).

Results

Crohn’s disease was not significantly associated with risk of lung cancer (odds ratio [OR], 1.01; 95% confidence interval [CI], 0.97-1.04; p = 0.76, IVW method). Weighted median (OR, 1.03; 95% CI, 0.98-1.07; p = 0.26) and MR Egger analysis (OR, 1.03; 95% CI, 0.95-1.11; p = 0.49) showed similar effect estimates of Crohn’s disease on lung cancer. The MR-Egger regression analysis showed that there was no evidence for the presence of horizontal pleiotropy (intercept β=-0.005, p = 0.532). Our leave-one-out analysis revealed that no single SNP strongly drove the overall effect of Crohn’s disease on lung cancer. The similar causal trend was observed in both LUAD and LUSC (LUAD, OR 0.99, 95% CI 0.94-1.04, p = 0.67, IVW method; LUSC, OR 0.98, 95% CI 0.94-1.03, p = 0.47, IVW method).

Conclusions

Our findings indicated that Crohn’s disease might not be causally associated with lung cancer, although patients with Crohn’s disease were at increased lung cancer risk in observational studies. Our study suggested there might be other risk factors associated with lung cancer in patients Crohn’s disease, which needs to be confirmed in the further study.

Legal entity responsible for the study

The authors.

Funding

National Key R&D Program of China (2016YFC0905500).

Disclosure

All authors have declared no conflicts of interest.

Background

To evaluate the capability of pulmonary MR imaging with zero echo time (ZTE) in pulmonary nodules assessments, using low-dose computed tomography (LDCT) as the reference standard.

Methods

Institutional review board approval and informed consent were obtained. Eight-nine consecutive patients (51 males: mean age, 69 years and 38 females: mean age, 71 years) with various pulmonary nodules were examined with chest LDCT and pulmonary MR imaging with ZTE. The interval between two examinations was within two weeks for each patient. Lung images were visually scored by two experienced radiologists who assessed the probability of nodule presence in comparison with LDCT. To compare nodule detection capability of the two methods, consensus for performances was evaluated by means of kappa statistics and χ² test, and receiver operating characteristic analyses were used to compare diagnostic performance of both methods.

Results

There was no significant difference (F = 0.75, P = 0.56) in figure of merit between methods (LDCT, 0.84 Vs. MR imaging with ZTE, 0.86). Intermethod agreements between ZTE MR imaging and LDCT were significant (0.70 ≤ κ ≤ 0.98; P < 0.001). Areas under the curve for pulmonary nodules on LDCT were significantly larger than those on MR imaging with ZTE (P = 0.02).

Conclusions

Pulmonary thin-section MR imaging with ZTE was useful in nodule detection, and it is considered at least as effective as low-dose thin-section CT.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Rapid Access Lung Lesions Clinic (RALLC) is designed to expedite the diagnostic pathway of suspected lung cancer, to allow treatment in a timely fashion. Invasive investigations (i.e. bronchoscopy, EBUS biopsy, CT-guided biopsy) are the usual tools to obtain tissue for histological diagnosis and each carries a small procedure-related complications risk, which could potentially delay definitive treatment initiation until the complication resolves.

Methods

Data was prospectively collected of all patients referred to a RALLC clinic in a tertiary cancer centre in regional Australia between July 2017 and December 2018. The need and choice of invasive investigation was determined based on tumour location, clinical suspicion of malignancy and multidisciplinary recommendation. Time to treatment was calculated from the date of diagnostic investigation to the first date of treatment (i.e. chemo/radio/immunotherapy, surgery). All data were analysed using descriptive statistics.

Results

Over 18-months 202 patients were referred to the RALLC clinic for evaluation. 105 patients underwent invasive investigation [CT-guided lung biopsy 46 (44%), Bronchoscopy 32 (30%), EBUS biopsy 22 (21%) & Surgical biopsy 5 (5%)]. Malignancy was confirmed in 89 patients (85% diagnostic yield), 80 being lung cancer. The only investigation-related complication was pneumothorax, which occurred in 15 of the CT-guided lung biopsy (33%) with 11 patients needing hospital admission. The median time to first treatment for the overall cohort of lung biopsy patients was 28 days (mean 28.3, range 0-72). In the non-pneumothorax group, the median was 27 days (mean 27.7, range 0-65) while the pneumothorax group was 29 days (mean 29.8, range 5-72).

Conclusions

Although CT-guided lung lesion biopsy resulted in a significant number of pneumothorces in our cohort of patients, ultimately this did not significantly delay definitive treatment compared to the overall cohort. However, the high pneumothorax rate increases health care burden and additional invasive management procedures for the patient. Further investigation is warranted to explore factors associated with this.

Legal entity responsible for the study

Ballarat Health Services, Ballarat, Victoria, Australia.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Treatment options for lung cancer patients are progressing rapidly. A clinical decision support system (CDSS) which affords first-line treatment recommendations for cancer patients was implemented at Jupiter Medical Center, on March 1, 2017. This pilot analysis studied system performance and its impact on patients and providers.

Methods

Lung cancer cases managed in the CDSS between October 2017 and November 2018 were incorporated. CDSS results were viewed by providers, discussed in tumor board, and shared with the patients. Navigators directed a survey about CDS impact and user satisfaction to clinician and patients at conclusion of visits. Summary responses were calculated.

Results

Thirty cases were processed by the CDSS; 23 surveys were completed. The treatment team agreed completely with CDSS recommendations for 18 cases (78.3%) and partly for 4 (17.4%). Clinicians reported that the CDSS impacted the treatment decision 9 (39.1%) cases and saved time in treatment design for 17 (73.9%). The CDSS confirmed the treatment plan for 16 cases (69.6%), recommended an option that was designated for 5 (21.7%), and offered a useful option not chosen for 1 (4.3%). For most cases, clinicians conveyed they were pleased with the CDSS results (n = 22; 95.7%), realized the system easy to use (n = 21; 91.3%), and would use it again (n = 22; 95.7%). For most cases, clinicians responded that they strongly agreed or agreed that the CDSS provided information that was relevant (n = 22, 95.7%), easy to understand (n = 21, 95.6%), and actionable (n = 21, 95.6%). Most patients answered that they were pleased with the CDSS analysis of their treatment plan (n = 19; 82.6%) and would want their provider to use the system in the future (n = 19; 82.6%).

Conclusions

This study provides initial evidence that a CDSS can positively support and influence the care of lung cancer patients. Treatment teams agreed with CDSS recommendations in most cases, and the CDSS impacted treatment conclusions. Providers and patients were pleased with the CDSS and reported being willing to use it again.

Legal entity responsible for the study

Jupiter Medical Center.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Bangladesh is ranked 6th among the Tuberculosis high burden countries. Sometimes, pulmonary tuberculosis (PTB) mimics the early symptoms of lung cancer. Developing countries like Bangladesh, with limited diagnostic facilities, often misdiagnose early lung cancer as smear negative PTB. According to national guidelines for PTB, smear negative PTB is defined as the cases diagnosed on the basis of X-ray abnormalities or suggestive histology and extra pulmonary cases without laboratory confirmation. This observational study has been done to find out the ratio of misdiagnosis of lung cancer as Smear Negative PTB.

Methods

This observational study was done between June, 2017 to January, 2018 at OPD of NIDCH (National Institute of Diseases the Chest & Hospital), Dhaka, Bangladesh. Inclusion criteria for the confirmation of PTB were: symptoms not improving after 3 weeks of anti-TB treatment; non-improving chest X-ray; new symptoms; no prior anti-TB treatment; smear negative PTB. For confirmation of AFB (Acid fast bacillus) smear test, CT scan of chest, GeneXpert of sputum, Mycobacterium culture and histological confirmation were performed. During this period, 172 cases were included in the study. After confirmation of the diagnosis, every case was treated according to national TB guidelines and refered to other tertiary specialized hospitals. After changing of the treatment protocol, 3-weekly follow up was done for 2 months with chest X-ray and other blood parameters.

Results

The investigations mentioned above were performed in all cases included in the study: 35 cases (20.34%) were diagnosed as lung cancer by histological confirmation. 81 cases were (47.09%) diagnosed as multidrug-resistant TB and 2 cases (1.16%) were diagnosed as extensively drug-resistant TB. Other cases (31.39%) were confirmed as PTB.

Conclusions

In countries with a high tuberculosis burden and lack of facilities for guided biopsy or fine needle aspiration, except in tertiary hospitals, plays an important role in delay of early diagnosis of lung cancer. In addition, flexible national guidelines and DOT (direct observed therapy) of anti-TB drugs are given at no cost to patients are also contributory factors. In this short observational study, it was found that a number of patients received anti-TB drug based on clinical diagnosis has been diagnosed as Lung cancer that were previously diagnosed as Smear negative PTB on clinical ground and Chest X-ray. This finding suggests that national policy makers should consider that patients with smear negative PTB (clinical diagnosed PTB) need close follow up and proper referral to specialized & tertiary hospitals where there are facilities for confirming the diagnosis. To make and change in national guideline this observational study should be run in multi center in the country.

Legal entity responsible for the study

NIDCH.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Radiomics is a promising tool for identification of new prognostic biomarkers. Radiomic models are often based on single-institution data however multi-centric data, highly heterogeneous due to different scanning protocols, reflect better the clinical reality. Robustness studies are crucial to find features robust to e.g. scanner settings. We study if a CT radiomics overall survival (OS) model trained on multi-centric data with robust feature pre-selection can achieve a similar performance as a model on standardized data.

Methods

Pre-treatment CT data from 121 IIIA/N2 NSCLC patients from a prospective multi-centric randomized trial (SAKK 16/00, neoadj. chemo- or radiochemotherapy prior to surgery) were used to calculate 1404 radiomic features on the primary tumor. Two OS radiomic models were trained on (1) a sub-cohort with standardized imaging protocol (native CT, standard kernel, n = 84) and on (2) the entire heterogeneous cohort but with robust radiomic feature pre-selection. Robust features were extracted from four robustness studies (contrast, convolution kernel, motion, delineation) using the intra-class correlation coefficient (> 0.9 considered stable). Seperately for each model, principal component (PC) analysis was performed and PCs describing in total 95% data variance were selected. The feature with highest correlation to the PCs were used for the multivariate Cox model with backward selection. Model performances were quantified using Concordance Index (CI), verified with 10-fold cross-validation and compared using bootstrap with resampling.

Results

Robustness studies revealed 113 stable features. The convolution kernel was the largest influence on the feature stability. Final OS model on the entire heterogeneous data consisted of four and on standardized data of six features (all identified as unstable). The model on standardized data showed significant better prognostic performance compared to the model with robust feature pre-selection based on the entire heterogeneous data (CI = 0.64 and 0.61, p < 0.05, resp.).

Conclusions

For our prognostic NSCLC radiomic models image protocol standardization appears superior to using larger but heterogeneous imaging data combined with robust feature pre-selection.

Legal entity responsible for the study

The authors.

Funding

Swiss National Science Foundation (SNSF) Swiss Group for Clinical Cancer Research SAKK.

Disclosure

M. Guckenberger: Board member: European Society for Therapeutic Radiation Oncology (ESTRO). All other authors have declared no conflicts of interest.

Background

Brain metastases in NSCLC patients is the major cause of the poor survival and quality of life. Symptomatic brain metastasis at baseline are present in 5-10% of NSCLC and over the time course about 40% patients develop brain metastasis. Screening for asymptomatic brain metastasis in advanced NSCLC is controversial. Here we present our initial experience of mandatory brain imaging of all NSCLC patients irrespective of CNS symptoms and stage.

Methods

It is a prospective observation study of NSCLC patients with mandatory baseline brain imaging. Data was analysed for patients enrolled in the study from Jan 2018- Nov 2018. All patients with histology proven NSCLC were incuded irrespective of stage. Brain imaging was performed by CECT. A comprehensive CNS examination was performed at diagnosis. All statistical analysis was done by STATA version 13.Univariate analysis was done by Cox regression model.

Results

There were 238 patients enrolled. Median age was 58yrs (range 26-83yrs) with male predominance (n = 183 , 76.9%). Smokers were 65.1% ( n = 153) and majority had ECOG PS 1&2 (82%). Most common histology was adenocarcinoma - 64.6% (n = 151) . EGFR mutation could be tested in 98 of adenocarcinoma patients and was positive in 32 (32.6%). ALK (IHC-D5F3) was tested in 78 patients and was positive in 11 (14.1%) . Majority patients had stage 4 disease, 74.8% (n = 176). There was upstaging to stage 4 after brain imaging in 8% (n = 20) patients. Brain metastases was present in 22.7% (n = 54) of which 52% (n = 28) were asymptomatic. Majority had solitary lesion - 48% (n = 26), predominatly located supratentorially - 66% (n = 36). Majority of these patients received chemotherapy - 46.2% (n = 25), TKI -27.7% (n = 15), WBRT -14.8%(n = 8), BSC -22.2% (n = 12). Univariate analyses was done for age, sex, histological subtype, EGFR and ALK status, out of which younger age (<40 yrs) and adenocarcinoma histology were found to be significant predictors for presence brain metastases.

Conclusions

Brain metastasis with mandatory baseline brain imaging (atleast CECT) were found in 22.7% patients in our study and resulted in upstaging of disease in 8%. Frequency of brain metastasis were significantly higher in adenocarcinoma histology and younger age (<40 yrs).

Legal entity responsible for the study

All India Institute of Medical Sciences.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Patients with lung cancer and co-existing interstitial lung disease (ILD) are at increased risk of treatment-related toxicity after both surgery and radiotherapy. A care path was implemented at our institution for patients presenting to the lung tumor board with a possible ILD, and we report on our experience using this structured approach.

Methods

Since 2015, patients with possible lung cancer and ILD were referred to the general ILD clinic for assessment. In 2017, a dedicated ILD lung tumor board was established in order to facilitate quick assessment of treatment-related risks. An ethics-approved institutional database containing details of all these patients was accessed.

Results

24 patients with lung tumors and a co-existing ILD were identified (Table). The mean interval between referral to, and consultation at our ILD-board was 2 weeks. A prior diagnosis of ILD was available in 9 of 17 (53%) patients, but review led to a re-classification of the ILD subtype in 8 of the former. Treatments for lung cancer included radiotherapy alone (n = 14), surgery (n = 6), sequential chemoradiation (n = 3), and concurrent CRT followed by salvage surgery (n = 1). 6 patients developed progression of ILD after radiation; of these, 2 had received nintedanib during treatment. One patient died because of progressive ILD and in another 3 patients ILD-related deaths could not be excluded.

Conclusions

A dedicated care path for ILD patients resulted in a fast evaluation of lung cancer patients. A previous ILD-diagnosis was revised in a majority of patients, a process which can allow for a better understanding of treatment-related risks in different subgroups of ILD patients, and also assess the role of ILD-directed therapies.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

S. Senan: Grants, during the conduct of the study: ViewRay Inc.; Personal fees, outside the submitted work: Varian Medical Systems. All other authors have declared no conflicts of interest.

Background

Precise tumour volume delineation is a major challenge in lung cancer radiotherapy (RT) because of respiration induced motion. Four-dimensional computed tomography (4DCT) using maximum intensity projection (MIP) dataset is commonly used method to generate internal target volume (ITV). This study compared ITV generation by tumour delineation on MIP and on all phases of the respiratory cycle.

Methods

Thirty consecutive patients who underwent 4DCT from Jan 2014 to Mar 2017 were included. After 4DCT image acquisition, ITV was generated by tumour delineation on MIP (ITV_MIP) and on all ten phases of respiratory cycle(ITV₁₀). The Correlation between ITV_MIP and ITV₁₀ was analysed using Matching index (M.I=overlapping/encompassing volume) and 3D centroid shift. M.I >0.8 was considered as good agreement. Time required to generate ITV_MIP and ITV₁₀ was also documented_. Mann Whitney test was used for analysis and p < 0.05 was considered significant.

Results

Mean volume of ITV₁₀ was 134 cc (range 13 -627) and ITV_MIP was 113 cc (range 11 - 569)(p < 0.001). Median volume of ITV_MIP not getting covered by ITV₁₀ was 4 cc (5.2%) and vice versa was 19.5cc (19%). The mean M.I between ITV_MIP and ITV₁₀ was 0.76 (SD = 0.08, range 0.57-0.88). For 15 patients with tumour close to high density structures like mediastinum, mean MI was 0.73 and for peripheral tumours was 0.78 (p = 0.003). Mean M.I was 0.75 for both tumour ≤5 and >5 cm. For peripheral and ≤ 5 cm tumours, M.I was 0.8. The average time for ITV_MIP and ITV₁₀ delineation was 9 and 96 minutes respectively. The centre of mass of ITV₁₀ and ITV_MIP was same 0.15 cm in sup/inf axis, 0.11 cm in medio/lateral axis and -0.06 and -0.04 cm in ant/post axis respectively when compared with free breathing CT. The 3D Centroid shift between ITV_MIP and ITV₁₀ was 0.01 cm.

Conclusions

ITV delineation using MIP is significantly smaller and may be imprecise in tumours adjacent to mediastinum, chest wall and diaphragm. This difference is likely due to difficulty in delineating tumour edges caused by blurring of images. However, delineation using MIP dataset is significantly less time consuming, hence, we recommend its continued use in peripheral tumour and advises caution in central tumours.

Legal entity responsible for the study

The authors.

Funding

TMC Intramural.

Disclosure

All authors have declared no conflicts of interest.

Background

Small-cell lung cancer (SCLC), accounting for about 15% of all lung cancers, is a subtype of lung cancer with poor prognosis. It has a 5-year survival rate of 7% and kills an estimated 250,000 people worldwide annually. Although many studies have estimated the prognosis of SCLC, most of them were conducted without considering competing risks. This study aimed to evaluate the probability of cause-specific death for patients with stage I small cell lung cancer (SCLC) with a competing risk analysis.

Methods

We identified patients with stage I SCLC between 2004 and 2010 in the Surveillance Epidemiology, and End Results (SEER) database. We calculated the cumulative incidence function (CIF) for all the SCLC patients, and the differences in CIF between subgroups were estimated by Gray’s test. Proportional subdistribution hazard model was constructed to predict cancer-specific death for patients with stage I SCLC. We also built a competing risk nomogram based on Fine and Gray’s model to predict the 3-year, the 5-year prognosis of SCLC patients. We evaluated the model performance by the c-index and calibration plot using a bootstrap cross-validation method with 200 resamples. All statistical analyses and visualization were performed on R statistical software version 3.4.4 (Institute for Statistics and Mathematics, Vienna, Austria). Statistical significance was set as a 2-sided p < 0.05.

Results

We identified 864 stage I SCLC patients. The 5-year cumulative incidence of cause-specific death for stage I SCLC was 56.2% and 13.7% for other causes of death. Predictive factors for the prognosis of stage I SCLC included age, surgery, chemotherapy, and radiotherapy (Table). Fine and Gray competing risk regression model indicated that age at diagnosis, surgery treatment, and radiotherapy could be independent predictive factors of SCLC cause-specific death. Those who were diagnosed with SCLC at an older age were more like to die of lung cancer, with a subdistribution hazard ratios (sdHR) of 1.02 (95% CI, 1.012-1.03). Patients without treatment were at an elevated risk of SCLC cause-specific death except for chemotherapy, with an sdHR of 2.85 (95% CI, 2.29-3.54) and 1.89 (95% CI, 1.53-2.33) for patients without surgery and radiotherapy, respectively. No statistical significance was detected between chemotherapy and SCLC cause-specific death. The competing risk nomogram based on the Fine and Gray’s model was established to predict the 3-year and 5-year cause-specific death. The c-index for SCLC cause-specific mortality model was 0.66, and the calibration curves suggested that the nomogram was well-calibrated.

Conclusions

In the study, we performed a competing risk analysis in patients with stage I SCLC based on the SEER database. We discovered independent predictive factors of death due to SCLC and built a nomogram to calculate the 3- and 5-year cause-specific mortality. The competing risk nomogram might be a convenient tool to evaluate crude mortalities of stage I SCLC, and help clinicians to choose appropriate treatment strategies.

Legal entity responsible for the study

The authors.

Funding

National Key R&D Program of China (2016YFC0905500).

Disclosure

All authors have declared no conflicts of interest.

Table: 60P

Five-year cumulative incidences of death among patients with stage I SCLC

Background

Small cell lung carcinoma (SCLC) rarely underwent surgical treatment. The aim of the study was to investigate the value of postoperative radiotherapy for limited stage small cell lung cancer based on a prognostic scoring model.

Methods

We searched the Surveillance, Epidemiology, and End Results (SEER) database for all cases of limited stage SCLC treated by surgical resection and chemotherapy between 2004-2014. A prognostic scoring model was built to evaluate the prognostic value of postoperative radiotherapy on the basis of a Cox proportional hazards model. Overall survival (OS) and cancer-specific survival (CSS) was compared in the different risk groups.

Results

Table: 61P

Multivariable analysis of the association between clinical variables and overall survival (N = 767)

Conclusions

On the basis of the prognostic scoring model, postoperative-radiotherapy in addition to surgery provides better outcomes than surgery alone for limited stage SCLC in high risk group. Prospective studies are needed to validate these results.

Legal entity responsible for the study

Jing Yu.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

In 2016, a survival score for limited stage small cell lung cancer (LS-SCLC) patients was developed to characterize prognostic sub-groups who underwent multimodal treatment. Herein, we validate the score in an independent external patient cohort.

Methods

We reviewed the medical charts of 78 LS-SCLC patients treated with CRT at our institution. The survival score was calculated by independent prognostic factors: gender, Karnofsky performance status (50-70% vs. 80-100%), Tumor sub-stage (very limited vs. limited disease) and hemoglobin level before the start of radiation (<12 g/dl vs. ≥12 g/dl). Scoring points were derived from 2-year survival rates divided by 10 and the individual values for each prognostic factor were tallied. Three risk subgroups were defined (high, intermediate vs. low-risk: 9-13, 14-18 vs.19-26 points).

Results

Median overall survival in the validation cohort was 17 months (range: 1-123months). The 2-year survival rates were 0% in the high, 35% in the intermediate and 43% in the low-risk subgroup, respectively (p = 0.018). The difference in 2-year survival between the high and intermediate risk subgroups was significant (p = 0.007), whereas the 2yr-OS between the intermediate and low risk was not (p = 0.602). After stratification for the concurrent treatment mode, 2-year survival rates were 0% in the high, 60% in the intermediate and 58% in the low-risk subgroups, respectively (p = 0.004).

Conclusions

The survival score was reproducible to estimate 2-year survival rates of patients with LS-SCLC, especially in the high and intermediate-risk subgroups. In order to better characterize the prognostic difference between intermediate and low-risk patients, the scoring system needs further optimization.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Around 15% of newly diagnosed lung cancer patients are characterized as SCLC mostly with extensive stage disease. While SCLC may be responsive to 1^st or 2^nd line chemotherapy, there is no approved drug for the 3^rd line with guidelines recommending best supportive care. Rovalpituzumab Teserine (Rova-T) is a novel first–in-class antibody-drug conjugate and targets the delta-like protein 3 (DLL3), a newly discovered target highly expressed in SCLC and other high-grade neuroendocrine carcinomas. ROVA-T showed an encouraging single-agent anti-tumor activity and manageable safety profile in Phase 1 and 2 studies. The present study is a retrospective analysis to evaluate DLL3 testing and real-life experience with ROVA-T for patients with an unmet medical need after failure of at least 2 systemic treatments.

Methods

DLL3 Immunohistochemistry (VENTANA DLL3 (SP347) Assay) was performed for 68 Patients with high grade neuroendocrine carcinoma (61 SCLC and 7 LCNEC). The percentages of stained tumor cells was determined - TC-counts of < 25% were interpreted as negative, counts of ≥ 25 to < 75% as positive and ≥75% as highly positive. 16 patients were eligible for treatment with Rova-T and received at least 1 of 2 planned cycles with a dose of 0.3 mg/kg.

Results

DLL3 staining was seen in most of the patients with high grade neuroendocrine carcinoma, 49 specimens (72.1%) were interpreted as highly positive, 10 specimens (14.7%) were considered positive. 9 (13.2 %) were considered negative, 4 (5,9%) of them showed no staining for IHC. All eligible patients had an ED with an ECOG of 0-1. 2 of the treated patients were DLL3 negative, 4 positive and 10 were highly positive. Of the 2 planned cycles 7 patients received both and 9 received only one due to disease progression or adverse events. Overall, 4 patients (25%) showed a PR, 4 (25%) had SD and 8 developed disease progression (50%). Side effects were manageable.

Conclusions

DLL3 staining is frequent in SCLC patients but further studies are needed to proof it to be a useful predictive biomarker. Rova-T seems to be an option for ED SCLC patients in later lines with a high unmet need for treatment and showed a clinical benefit in selected patients with a manageable safety profile.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

SCLC is an aggressive cancer subtype, with early metastatic spread and poor prognosis. We review Irish patient outcomes, outside of a clinical trial setting, to determine if patients who commenced radiotherapy (RT) within 30days of chemotherapy (CT) had a longer PFS.

Methods

SCLC cases diagnosed between 2006-2016 were identified from a prospectively maintained lung cancer registry at a tertiary referral centre. Electronic patient records, paper charts, and pathology reports were reviewed to extract relevant data. In order to detect a statistical difference between the two groups with 95% confidence interval and a 5% margin of error, a sample size of 125 patients with SCLC was needed, and a sample size of 56 with LSSCLC. A two-sided t-test was used where data was normally distributed. A p value of < 0.05 was accepted. Kaplan-Meier analysis was used to assess progression free survival and a Mann-Whitney test to assess differences between groups.

Results

265 patients identified, 70 with LSSCLC, 44 had concurrent treatment. Median time to starting CT was 20 days (95%CI; 15.8-24.2days). The median time between diagnosis and CT was 28days, from CT to RT was 28.5 days. 21 patients started RT within 30 days of commencing CT; median time 19days. 19 patients started RT more than 30 days post CT; median time 67days. Median time from diagnosis to CT was 12days in this group. Median PFS where the time from day1 CT to day1 of RT was less than 30days was 12.1months. For patients where the treatment interval was greater than 30 days, median PFS was 10.6months. Using a non-parametric Mann-Whitney test to evaluate the median PFS produced a U value of 127, and a P-value of 0.41.

Conclusions

This study reports real-world outcomes in patients with LSSCLC over a 10-year period. Although not reaching statistical significance there is a numerical difference in PFS favouring shorter time to combination CRT. We have also identified areas where there is room to improve our practice. Further attention will focus on strategies to reduce both the interval between histological diagnosis and commencing CT, and the interval between CT and RT in this setting. A study expansion is planned.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

We conducted this randomized trial at our center to assess benefits of prophylactic cranial irradiation (PCI) in cases of extensive stage small cell lung cancer (eSCLC), who showed complete or partial response to 4 cycles of cisplatin and etoposide chemotherapy-based regime.

Methods

For the study, 71 patients of histology proven SCLC with extensive stage (with no symptomatic and radiological brain metastasis) on clinico-radiological examination were randomized into two arms (Arm A and Arm B). Patients in Arm A were treated with PCI whilw those in Arm B were kept on observation, patient in both arms were followed for 1 year. Primary endpoint of the study was to assess the time of onset of symptomatic brain metastasis in both arms. Computer Tomography (CT scan) or Magnetic Resonance Imaging (MRI) was performed in patients who had any symptom suggestive of brain metastasis.

Results

Patients in both arms (36 in Arm A and 35 in Arm B) were comparable for baseline characteristics. Patients in Arm A had significantly low risk of symptomatic brain metastasis then Arm B (p value = <0.005). The risk of brain metastasis in Arm A was 19.4% and in Arm B it was 51.4%. Mean duration for freedom from symptomatic brain metastasis in Arm A was 8.36 months while it was 4.49 months for Arm B. Overall 1 year median survival rate in Arm A was significantly higher in Arm A then Arm B (8 months vs 4 months). Patients in Arm A had higher toxicities, but they were mostly manageable and were not associated with poor quality of life.

Conclusions

PCI is associated with significantly low risk of symptomatic brain metastasis and higher overall survival without affecting quality of life in eSCLC.

Legal entity responsible for the study

Aacharya Tulsi Regional Cancer Treatment and Research Center, Bikaner.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

For pts with ES-SCLC, guidelines recommend carbo or cis + etop as first-line (1L) tx. However, real-world data (RWD) on tx patterns and outcomes are limited. Here, we describe pt characteristics, tx duration and clinical outcomes associated with these 2 regimens.

Methods

Pts with ES-SCLC diagnosis (or limited-stage [LS] SCLC who initiated second-line [2L] tx) between 1 Jan 2013 and 31 Aug 2017 (follow-up to 31 Aug 2018) were identified from the US-based Flatiron Health electronic health record–derived database. Pts receiving tx with either carbo + etop or cis + etop were included in the analysis.

Results

RWD on 2161 pts from 156 tx centres were included; 84% of pts received carbo + etop. See table below for pt characteristics. The median tx duration was 3.4 mo (95% CI: 3.4, 3.4) with carbo + etop and 3.0 mo (95% CI: 2.8, 3.4) with cis + etop. The distribution of tx cycles administered was similar between carbo and cis, with 20% and 28% of pts completing 4 or 6 cycles, respectively. Median overall survival (OS) was 8.3 mo (95% CI: 8.1, 8.7) with carbo + etop and 9.7 mo (95% CI: 9.3, 11.0) with cis + etop. The 1-yr OS rates were 30% (95% CI: 28, 33) and 41% (95% CI: 36, 47), respectively. In pts with Eastern Cooperative Oncology Group performance status (ECOG PS) 0-1 and ≥ 2, median OS was 9.3 mo (95% CI: 8.6, 9.9) and 7.1 mo (95% CI: 6.3, 8.3), respectively. Pts with unknown ECOG PS had a median OS of 8.4 mo (95% CI: 8.0, 8.9).

Conclusions

Tx duration was similar between the 2 regimens. Pts who received cis + etop had numerically increased OS vs pts who received carbo + etop, as did pts with ECOG PS 0-1. However, these findings may be due to pts receiving cis + etop being fitter (younger and lower ECOG PS) at baseline.

Editorial acknowledgement

Medical writing assistance for this abstract was provided by Steffen Biechele, PhD, of Health Interactions and funded by F. Hoffmann-La Roche, Ltd.

Legal entity responsible for the study

F. Hoffmann-La Roche, Ltd.

Funding

F. Hoffmann-La Roche, Ltd.

Disclosure

M. Sebastian: Honoraria, consulting: AZ, BI, BMS, Eli Lilly, Merck Sharp & Dohme, Novartis, Pfizer, Roche; Honoraria: Pierre Fabre; Consulting: Celgene. F. Barlesi: Personal fees/clinical trials (inst.): AZ, BMS, BI, Lilly, Roche, Novartis, Merck, MSD, Pierre Fabre, Pfizer, Takeda; Clinical trials (inst.): AbbVie, ACEA, Amgen, Bayer, Eisai, Genentech, Ipsen, Ignyta, Innate Pharma, Loxo, Medimmune, Sanofi-Aventis. R. Califano: Honoraria/consult: BI; Stock: Christie Private Care; Grants/nonremunerated activities: Clovis, AbbVie; Leadership (nonrem): ESMO; Nonrem membership: EORTC; Honoraria/consult/grants/nonrem activities: AZ, Roche, Pfizer, Lilly, MSD, Takeda, Novartis, BMS. A.S. Mansfield: Research funding (inst.): Novartis, Verily; Honoraria/Ad board (inst.): Genentech, AbbVie, BMS Mesothelioma Applied Research Foundation; Board member (non-rem.) ASCO; Lung Cancer Education Committee member (non-rem). F.H. Blackhall: Grant/research: Roche, BI, AZ, Cellmedica, AbbVie, Pfizer; Advisory board: AZ, Cellmedica, AbbVie, Ipsen, Takeda, Roche; Consulting/speakers: Takeda; Honoraria: Takeda, Roche, AbbVie, Ipsen; Study, editorial support: Roche. E.M. Flahavan: Employee: Roche; Stock: Lilly, Roche; support of parent study and funding of editorial support: Roche. J. Davies: Employee: Roche. P. Arnold: Employee: Roche; Stock: Novartis Pharma AG. S. Morris: Employee, Stock: Roche. M. Reck: Support of parent study, funding of editorial support: Roche; Honoraria for lectures and consulting: Amgen, AbbVie, BI, BMS, Celgene, Merck-Serono, MSD, Lilly, Novartis, Pfizer, Roche.

Table: 66P

Characteristics

Background

Limited stage small-cell lung cancer (LS-SCLC), which represents ∼30% of newly diagnosed SCLC, remains an area of high unmet medical need. Standard of care, which has not changed for several decades, consists of curative intent platinum-based chemotherapy concurrent with radiotherapy (cCRT) followed by prophylactic brain irradiation (PCI) and observation. Despite good response to cCRT, outcomes remain poor, with median progression-free survival (PFS) ∼15 months and overall survival (OS) ∼25 months. Durvalumab (D) is a selective, high-affinity, human IgG1 monoclonal antibody (mAb) that blocks programmed cell death ligand-1 binding to programmed cell death-1 and CD80. Tremelimumab (T) is a selective human IgG2 mAb against CTLA-4. D demonstrated a PFS and OS advantage over placebo in locally advanced NSCLC following cCRT. D and D + T demonstrated a tolerable safety profile and antitumour activity in pretreated extensive stage SCLC. The ADRIATIC trial (NCT03703297) will assess if treatment with D ± T is beneficial vs placebo in patients (pts) with LS-SCLC who have not progressed following cCRT.

Trial design

ADRIATIC is a Phase 3, randomised, double-blind, multicentre, placebo-controlled international trial. Pts (N∼600) will be randomised 1:1:1 to receive D + placebo T, D + T, or dual placebo, stratified by Stage (I/II vs III) and receipt of PCI at the investigator’s discretion (yes vs no). Eligible pts must have confirmed inoperable Stage I–III LS-SCLC; WHO/ECOG PS 0/1; and completed 4 cycles of cCRT with a response of stable disease or better within 1–42 days prior to randomisation. Pts will receive the assigned treatment until clinical, RECIST v1.1-defined progressive disease, intolerable toxicity or for a maximum of 24 months, whichever comes first. Primary objectives are PFS and OS for D ± T vs placebo. Key secondary endpoints include health-related quality of life, and safety and tolerability. Recruitment is ongoingTable: 67TiP

Clinical trial identification

NCT03703297.

Editorial acknowledgement

Medical writing support, which was in accordance with Good Publication Practice (GPP3) guidelines, was provided by Craig Turner, MSc, of Cirrus Communications, an Ashfield company (Macclesfield, UK), and was funded by AstraZeneca.

Legal entity responsible for the study

AstraZeneca PLC.

Funding

AstraZeneca.

Disclosure

S. Senan: Departmental research grant: AstraZeneca, Varian Medical Systems, ViewRay Inc.; Advisory boards: AstraZeneca, MSD, Eli Lilly, Celgene. N. Shire: Employment, stock: AstraZeneca, outside the submitted work. G. Mak W. Yao, H. Jiang: Employment, stock: AstraZeneca, outside the conduct of the study.

Background

Stereotactic body radiation therapy (SBRT) has become increasingly accepted as a treatment for patients with inoperable early stage lung cancer. We studied the patterns of treatment utilization for stage I non-small cell lung cancer (NSCLC) in three European countries in patients diagnosed in 2015-2016.

Methods

Information from population-based registries in England, Norway and the Netherlands were retrieved, and treatment patterns and two-year survival for patients with clinical stage I were analysed.

Results

Resection rates were higher in Norway (55%) and England (53%) than in the Netherlands (47%). SBRT use was highest in the Netherlands (41%), followed by Norway (29%) and England (12%). Failure to perform curative intent treatment was more common in England (26%) than either Norway (13%) or the Netherlands (9%). In patients treated with SBRT, pathology confirmation of diagnosis was more common in Norway (70-73%) than in England (44-51%) or the Netherlands (47-50%). Two-year survival rates were better for surgery than for SBRT or conventional radiotherapy and treatment-specific survival was higher in Norway than in the other two countries.

Conclusions

For clinical stage I NSCLC, SBRT has largely replaced conventional radiotherapy in both Norway and the Netherlands, while just 67% of radiotherapy patients in England receive SBRT. The proportion of patients not receiving any curative treatment was higher in England, especially in the elderly.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

S. Senan: Research grants: Varian Medical Systems, ViewRayInc; Membership of advisory boards: AstraZeneca, MSD. All other authors have declared no conflicts of interest.

Background

Stereotactic body radiotherapy (SBRT) is a standard treatment for early stage non-small cell lung cancer (esNSCLC) patients who are not eligible for surgery. We aimed to evaluate the prognostic significance of the level of change in SUV_max (ΔSUV_max) between pre and post-treatment PET/CT in esNSCLC patients treated with SBRT.

Methods

Between November 2009-February 2018, pathologically proven esNSCLC patients (T1-2N0M0) treated with CyberKnife as primary treatment alone and had pre and post-treatment PET/CT (in 3 weeks before and 12-16 weeks after SBRT) were retrospectively identified. The ΔSUV_max was calculated using formula ΔSUV_max. The area under the curve (AUC) was used to verify the accuracy; the product of maximum sensitivity and specificity was chosen as the cutoff value. Then, we stratified the study cohort above and below AUC and the survival data were estimated by Kaplan Meier method. Univariate and multivariate analyses were carried out by use of a Cox proportional hazards model.

Results

All patients’ clinicopathological and treatment characteristics are presented in the table. Median dose was 45 Gy/3 fr (range: 45-60 Gy/3-5 fr). According to EORTC metabolic response criteria, 8 (16.7%) patients achieved complete, and 35 (72.9%) patients achieved partial response. AUC was calculated as 0.62 for cutoff ΔSUV_max (sensitivity 79%, specificity 45%). ΔSUV_max was ≥0.62 in 7 of 8 patients with complete response and <0.62 in progressive 2 patients. At a median follow-up of 23 (range: 6-92) months local, regional and distant relapse had developed in 16 (33.3%), 11 (22.9%), and 16 (33.3%) patients, respectively. 29 (60.4%) patients were still alive at the time of analysis. Median PFS was 15 vs 59 months (p=.012) and median OS was 36 vs 70 months (p=.045). In univariate analysis, we could not find any significant effect of sex, age, KPS, T stage, and BED₁₀ on PFS, and OS. In both univariate and multivariate analysis, the lower ΔSUV_max (as both dichotomous and continuous variable) was determined as a negative prognostic factor on PFS (p=.02, .003 for univatiate and .013 and .003 for multivariate) and it has been showed that the lower ΔSUV_max (only as a dichotomous variable) is a negative prognostic factor on OS in multivariate analysis (p=.009)Table: 74P

Clinicopathological and treatment characteristics

Conclusions

ΔSUV_max is a prognostic factor in esNSCLC patients treated with SBRT and patients with higher ΔSUV_max (≥0.62) have better PFS and OS.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.

Background

Localization of ground-glass opacifications (GGOs) with more than 1cm distance to the pleura are still a challenge to thoracic surgeons during video-assisted thoracoscopic surgery (VATS). Efficiency, damage avoidance, and safety are uppermost factors to be considered when choosing the location technique. Our center tested the thoracoscopic ultrasound (US) as a complementary method during VATS GGOs surgery.

Methods

We evaluated 40 patients with GGOs divided into group A (VATS-US) and group B (VATS). We used positive pressure in the chest cavity during surgery in group A to decrease the time needed to collapse the lung. With the same excision standard between the two groups, we recorded and compared the time needed, ratio of converting to thoracotomy, blooding volume, and any complications observed.

Results

We found that the use of ultrasound during VATS surgery could improve the success rate of localization (p < 0.05), especially in patients where it was difficult to conduct preoperative CT guided localization (p < 0.01). There were no statistically significant differences between the two groups in operation time, bleeding during the operation, or postoperative complications (p > 0.05).

Conclusions

We concluded that thoracoscopic ultrasound could be an excellent supplement to preoperative CT localizing technique for localization of GGOs, if not a replacement at present.

Legal entity responsible for the study

Hongbin Zhang.

Funding

Peking University International Hospital.

Disclosure

The author has declared no conflicts of interest.

Background

Development of endoscopic therapeutic methods for early central lung cancer.

Methods

During the period from 1984 to 2019 in P.A.Hertsen Moscow Research Oncologic Institute has accumulated clinical experience in endoscopic diagnostics and treatment of early central lung cancer in 129 patients (176 tumors). X-Ray negative ECLC was observed in 86% of cases. As a treatment approach - electrocoagulation was in used from 1984 to 1993 in 73 patients (91 tumors) with ECLC, PDT and argon plasma coagulation (APC) were used from 1992 to 2019 in 56 patients (85 tumors). The following agents for used for PDT: radachlorine, photohem, photosense photolon. Endoscopic methods in 90% of cases was selected due to unfeasibility of the conventional surgery.

Results

In the group of patients with ECLC treated by electrocoagulation the complete tumor regression was observed in 89% of cases. The tumor recurrence observed only in 17% of cases. In the patients who received PDT and APC the results were determined by the tumor size. When the tumor size was not greater than 1 cm, complete regression was achieved in 100% of cases. When the tumor size ranged from 1 to 1.5 cm complete resorption was achieved in 75% of cases. and with the tumor size up to 2.5 cm complete regression was achieved in 29% of cases. In 9% of cases the recurrences were diagnosed. Follow-up period lasted up to 15 years. Median survival was 5.05 years. Five year survival rate was 50±12 %.

Conclusions

Intraluminal endoscopic treatment methods - photodynamic therapy and argon-plasma coagulation in early central lung cancer patients, declined of conventional surgery, provides complete tumor regression in most of cases. 5-year disease-free survival achieved in half of patients group.

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.