Author of

• 29423

  +

  IBS06 - Multimodality Treatment - Realtime Data from National Registries (Ticketed Session) (ID 37)

  • Event: WCLC 2019
  • Type: Interactive Breakfast Session
  • Track: Mesothelioma
  • Presentations: 1
  • Now Available
  • Moderators:
  • Coordinates: 9/08/2019, 07:00 - 08:00, Dublin (1997)

  • 29425

    +

    IBS06.02 - Real Time Data from US (SEERS) (Now Available) (ID 3332)

    07:00 - 08:00  |  Author(s): Maaike Van Gerwen
    • Abstract
    • Presentation
    • Slides

    Abstract
    

    Real time data for Mesothelioma in US

    Maaike van Gerwen, Naomi Alpert, Andrea Wolf, Raja Flores, Emanuela Taioli

    Introduction

    The association between asbestos exposure and malignant mesothelioma has been well established. Exposure to asbestos mainly occurs through work although environmental exposure has been documented. Several countries have put in place an active epidemiologic surveillance of mesothelioma cases.

    Through a PubMed and Google Scholar search using the key words “mesothelioma” and “registry”, and by reviewing data sources of studies described in a review of environmental exposure and malignant mesothelioma, we identified existing mesothelioma registries. Countries with mesothelioma specific registries are Australia, Belgium, France, Germany, Italy, Japan, South Korea, South Africa, Turkey, and the UK. Nation-wide coverage is obtained in Italy, Australia and South Korea. Registries in Australia, France, Italy, and South Korea use interviews to obtain exposure data from the patient or a close relative, although none of these countries has a tissue bank. The UK has a mesothelioma tissue bank although it is not linked with the National Mesothelioma Audit registry. All registries have or will develop linkage with death index registries to monitor survival outcomes. (Table 1)

    The Scandinavian countries including Norway, Sweden, Finland and Denmark, have a population based cancer registry that includes mesothelioma and is linked to other databases, such as an occupational database, thus has more comprehensive information on each case.

    Mesothelioma surveillance in the US to date

    Currently, no nation-wide, mesothelioma specific registry exists in the US. Various existing databases are used to investigate mesothelioma, for instance the National Cancer Database has been used to look at prognostic factors; gender and race differences in mesothelioma survival have been studied using the Surveillance, Epidemiology, and End Results (SEER) database. All these datasets suffer of a time-lag between case occurrence, reporting, registration and eventually data availability for research purposes; these are serious limiting factors in the case of mesothelioma. Because of the rarity and lethality of the disease, a real-time capture registry is needed to thoroughly collect exposure data, complete data on treatments, quality of life before and after treatment, symptoms and pain management. All these elements are lacking in the existing databases.

    Because the US mesothelioma incidence is not decreasing as quickly as predicted and new cases still occur as well the fact that mortality rates are steady overtime, the overall health burden due to mesothelioma in the US still remains. Although the use of asbestos has been restricted or banned since 1980, several scientific questions remain open due to the long latency period between exposure and mesothelioma clinical occurrence, and to gaps in knowledge of the carcinogenesis process. Data on occupational and environmental asbestos exposure and co-exposure to other carcinogens are needed. Certain patterns, such as differences in outcomes by gender, differences in incidence rates by race, as well as geographic clusters of increased number of cases, are hard to explain with the existing data.

    Possible next steps towards a US mesothelioma registry

    “Real time” enrollment is important in order to systematically collect information on asbestos and other exposures through interviews with mesothelioma patients or a close relative. Furthermore, optimal coverage, preferably population based and nation-wide, and a simplified consent process are needed in order to capture a maximum number of cases. A centralized quality control system, standardized data collection methods, and the ability to link to relevant other existing registries are important in order to integrate the registry with clinical and prognostic information. Additionally, consistency in the design and questionnaire content with other countries would be ideal, in order to conduct comparisons and possibly pool the data. The flexibility to add or modify modules to tailor to future research questions are other preferable features for a US mesothelioma registry. (Figure 1) A discrete amount of work has been devoted to molecular markers such as mesothelin and certain germline mutations as prognostic factors. The role of these biomarkers could be validated on larger populations of patients if a comprehensive registry that includes tissue is implemented.

    Summary and conclusions

    In conclusion, with the remaining health burden due to mesothelioma, the changing landscape of asbestos exposure, and the many unanswered scientific questions, a nation-wide, real-time US mesothelioma registry is urgently needed. Methods for data sharing, linkage to existing tissue banks, and data access should be implemented and tested on a small scale before being implemented nationwide. One of the most practical outputs of these efforts would be the ability to conduct pragmatic trials that could be built out of a “real time” case capture system.

    

    

    Only Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login, select "Add to Cart" and proceed to checkout. If you would like to become a member of IASLC, please click here.

    Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.

• 29924

  +

  MA02 - Miscellaneous Topics in the Management of Early Stage Lung Cancer (ID 116)

  • Event: WCLC 2019
  • Type: Mini Oral Session
  • Track: Treatment of Early Stage/Localized Disease
  • Presentations: 1
  • Now Available
  • Moderators:Mireia Serra-Mitjans, Thomas A. D'Amico
  • Coordinates: 9/08/2019, 10:30 - 12:00, Interlaken (1988)

  • 29931

    +

    MA02.09 - Prognostic Impact of Immune Cell Biomarkers in Surgically Resectable Non-Small Cell Lung Cancer (Now Available) (ID 2646)

    10:30 - 12:00  |  Author(s): Maaike Van Gerwen
    • Abstract
    • Presentation
    • Slides

    Background
    

    Immune cells within the tumor microenvironment (TME) play an important role in the development, progression and eventual outcomes of non-small cell lung cancer (NSCLC). The relative balance of immune effector and regulatory cell subpopulations may tilt the TME to be either detrimental or supportive of tumorogenesis and will have a profound impact on the tumor’s eventual destiny. In early-stage lung cancer, the role of individual immune cell subtypes in the TME on survival outcomes following surgical resection is unknown.

    Method
    

    This project made use of The Cancer Genome Atlas (TCGA) Program data. We computed sample-specific scores for different immune cells using xCell, a new model for estimating different immune cell types from RNAseq data, for all stage I-IIIA NSCLCs. Then, we assessed the association between each cell type and survival with Cox Regression, while adjusting for important clinical variables (i.e., stage, age, gender, smoking status). We stratified the analysis according to histological subtype.

    Result
    

    There were 910 surgically resected early-stage NSCLC analyzed, of which 438 were adenocarcinomas (LUADs) and 472 were squamous cell (LUSC) samples. Higher levels of natural killer cells, neutrophils, and mast cells within tumors were associated with significantly improved survival in LUAD patients, whereas no immune cell type was associated with survival for LUSC patients or the combined analysis.

    Figure 1: Adjusted Survival According to Estimated Immune Cell Infiltration

    

    Hazard ratios are adjusted for stage, gender, age and smoking status

    Conclusion
    

    Innate and adaptive immune cells within the TME may have prognostic value in early-stage NSCLC patients undergoing surgical resection. However, the role of individual immune cells may vary according to histological subtype. Prospective research should continue to assess the association of the immune cell composition of the TME with clinical outcomes.

    Only Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login, select "Add to Cart" and proceed to checkout. If you would like to become a member of IASLC, please click here.

    Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.

• 30016

  +

  MA07 - Clinical Questions and Potential Blood Markers for Immunotherapy (ID 125)

  • Event: WCLC 2019
  • Type: Mini Oral Session
  • Track: Immuno-oncology
  • Presentations: 1
  • Now Available
  • Moderators:David R Spigel, Roberto Ferrara
  • Coordinates: 9/08/2019, 13:30 - 15:00, Vancouver (2003)

  • 30024

    +

    MA07.10 - The Influence of Sex on Immunotherapy Efficacy in Non-Small Cell Lung Cancer (Now Available) (ID 712)

    13:30 - 15:00  |  Author(s): Maaike Van Gerwen
    • Abstract
    • Presentation
    • Slides

    Background
    

    Patient’s sex impacts clinical outcomes for multiple cancers, including non-small cell lung cancer (NSCLC). A recent meta-analysis demonstrated sex may also impact response to novel immunotherapeutic agents, where men appear to derive greater benefit than women. However, the role of important clinical confounders of immunotherapy response that differ according to sex was not accounted for. The aim of this project was to investigate the effect of sex on immunotherapy benefit for NSCLC patients using a large, nationally representative database while adjusting for important clinical confounders.

    Method
    

    Advanced metastatic NSCLC patients diagnosed between 2013-2015 were identified in the National Cancer Database (NCDB). A Cox Proportional Hazards model was used to assess the interaction between sex and immunotherapy treatment for overall survival. This model was also adjusted for histology, stage, age, race, tumor size, comorbidities and other treatment (i.e. chemotherapy, radiation).

    Result
    

    Of 103,525 advanced NSCLC patients, 69,120 (67%) had adequate follow-up information for survival analysis. Of these, 37,423 (54.1%) were males and 31,697 (45.9%) females; 4,012 patients received immunotherapy as first-course treatment. In the adjusted model, both males (Hazard Ratio [HR]_adj: 0.77, 95% Confidence Interval [CI] 0.73-0.81) and females (HR_adj: 0.80, 95% CI 0.76-0.85) receiving immunotherapy had improved survival compared to those not receiving immunotherapy. The interaction between sex and immunotherapy was not significant (p=0.2539) after adjusting for clinical variables. Among the covariates, younger age, adenocarcinoma histology, Black race, smaller tumor size, lower comorbidity score and additional cancer treatment (either chemotherapy or radiation) were independently associated with better survival (p<0.0001 for all comparisons).

    Conclusion
    

    Patient sex does not appear to affect the benefit of immunotherapy in advanced NSCLC patients after adjusting for potential clinical confounders. Other clinical factors may play a role in immunotherapy response and should be explored in future research.

    Only Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login, select "Add to Cart" and proceed to checkout. If you would like to become a member of IASLC, please click here.

    Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.

• 31431

  +

  P1.13 - Staging (ID 181)

  • Event: WCLC 2019
  • Type: Poster Viewing in the Exhibit Hall
  • Track: Staging
  • Presentations: 1
  • Moderators:
  • Coordinates: 9/08/2019, 09:45 - 18:00, Exhibit Hall

  • 31432

    +

    P1.13-01 - The Importance of Staging of Lung Cancers, 30 mm or Less, Separately for Subsolid and Solid Nodules (ID 1413)

    09:45 - 18:00  |  Author(s): Maaike Van Gerwen
    • Abstract

    Background
    

    To determine pathologic results on non-small-cell lung cancers (NSCLCs), 30 mm or less in maximum diameter, separately by tumor consistency (solid, subsolid) on CT scans as we had shown that long-term survival was significantly different by tumor consistency and by type of parenchymal invasion.

    Method
    

    We reviewed all patients enrolled in the Initiative for Early Lung Cancer Research on Treatment (IELCART), a prospective cohort study of patients with first primary T1a-T1c NSCLC between 2016 and 2018 who had surgical resection. Short-axis diameter of N1-N3 lymph node on CT and SUVmax uptake on FDG-PET, if performed, were documented with values ≥ 2.5 defined as PET positive. Pathology reports were reviewed for N1-N3 lymph nodes (LNs) metastases and parenchymal invasion.

    Result
    

    Among 347 patients, 280 (80.7%) and 67 (19.3%) had solid and subsolid NSCLCs, respectively; all subsolid NSCLCs were adenocarcinoma. There was FDG-PET uptake in 253 (93.3%) with solid NSCLCs and in 55 (91.7%) with subsolid NSCLCs.

    None of the 67 subsolid NSCLCs had N1 or N2 LN metastases (Table 1). Among the 280 solid NSCLCs, none of the 42 NSCLCs≤ 10 mm had N1 or N2 metastases, while 5 of the 238 solid NSCLCs greater than 10 mm had N2 and 14 had N1 LN metastases. None of the N2 LNs were positive on FDG-PET and only 4(28.6%) of the 14 N1 LNs were positive on FDG-PET.

    Angiolymphatic invasion was most frequently, followed by pleural and major vascular invasion (Table 1). For solid NSCLCs, invasion increased with increasing tumor diameter.

    Conclusion
    

    No N1-N3 LN metastases were identified in solid NSCLCs ≤ 10 mm; none in subsolid NSCLCs ≤ 30 mm. None with N2 LN metastases were positive on FDG-PET. This suggests that for NSCLCs, 30 mm or less, clinical staging be based on solely tumor size. For pathologic staging, we recommend differentiating staging classification by tumor consistency in line with the latest recommendations for pathologic assessment.

• 31728

  +

  P1.17 - Treatment of Early Stage/Localized Disease (ID 188)

  • Event: WCLC 2019
  • Type: Poster Viewing in the Exhibit Hall
  • Track: Treatment of Early Stage/Localized Disease
  • Presentations: 2
  • Now Available
  • Moderators:
  • Coordinates: 9/08/2019, 09:45 - 18:00, Exhibit Hall

  • 31779

    +

    P1.17-14 - Prognostic Value of Immune Cell Biomarkers in Surgically Resectable Non-Small Cell Lung Cancer: A Meta-Analysis (Now Available) (ID 2658)

    09:45 - 18:00  |  Author(s): Maaike Van Gerwen
    • Abstract
    • Slides

    Background
    

    Immune cells within the tumor microenvironment (TME) play an important role in the development, progression and outcomes of non-small cell lung cancer (NSCLC). We conducted the first meta-analysis of studies assessing the role of individual immune cells in surgically resected stage I-III NSCLC to evaluate the prognostic value of immune cell biomarkers.

    Method
    

    PubMed was searched to identify eligible studies assessing clinical outcomes of surgically resected stage I-III NSCLC patients according to immune cell subsets: CD3+ T cells, CD4+ T Helper cells, CD8+ T cytotoxic cells, CD20+ B cells, FoxP3+T Regulatory cells, Natural Killer cells (CD56/CD57+), macrophages (CD68+), and Mast Cells. Meta-analysis was performed using a linear mixed-effects model to determine overall survival (OS) and disease free survival (DFS), and was reported according to whether or not the study adjusted for clinical covariates. I² was used to assess heterogeneity across studies.

    Result
    

    42 articles (7,906 patients) were included in this analysis. Higher levels of CD20+ B cells were associated with better OS, while increased FoxP3+ T regulatory cells and Mast Cells were associated with worse OS in unadjusted studies. However, in studies adjusting for clinical variables, CD8+ T cytotoxic cells and Natural Killer cells were also found to be associated with improved OS, while mast cells no longer were significantly detrimental. CD20+ B cells were associated with better DFS in unadjusted studies; in adjusted studies, CD8+ T cytotoxic cells were associated with better survival and FoxP3+ T Regulatory cells were associated with worse survival, with CD20+ B cells no longer significantly associated survival. I² did not show substantial heterogeneity between studies.

    Table 1: Meta-Analysis Survival Estimates

    	OS				DFS
    Biomarker	Unadjusted HR (95% CI)	N, Heterogeneity (%)	Adjusted HR (95% CI)	N, Heterogeneity (%)	Unadjusted HR (95% CI)	N, Heterogeneity (%)	Adjusted HR (95% CI)	N, Heterogeneity (%)
    CD3+ T cells	0.98 (0.87-1.12)	6 articles I2= 0.00	0.71 (0.37-1.37)	3 articles I2= 0.00	0.98 (0.86-1.12)	5 articles I2= 0.00%	-------	-------
    CD4+ T Helper cells	0.64 (0.37-1.10)	4 articles I2= 13.65	0.80 (0.50-1.28)	4 articles I2= 0.00	0.72 (0.39-1.32)	3 articles I2= 0.00%	0.59 (0.11-3.12)	1 article
    CD8+ T Cytotoxic cells	0.99 (0.95-1.04)	14 articles I2= 0.00	0.71 (0.52-0.96)	12 articles I2= 20.95	0.94 (0.77-1.16)	8 articles I2= 16.65%	0.60 (0.41,-0.87)	9 articles I2= 20.43%
    CD20+ B cells	0.45 (0.22-0.93)	5 articles I2= 52.29	0.16 (0.04-0.64)	1 article	0.57 (0.33,-1.00)	4 articles I2= 0.00%	0.51 (0.20-1.32)	1 article
    FoxP3+ T Regulatory cells	1.78 (1.20-2.64)	9 articles I2= 14.93	2.38 (1.56-3.65)	6 articles I2= 0.00	1.45 (0.81-2.59)	3 articles I2= 0.00%	2.07 (1.10-3.90)	3 articles I2= 0.00%
    Natural Killer cells	0.66 (0.35-1.25)	3 articles I2= 13.63	0.50 (0.26-0.95)	4 articles I2= 0.00	1.35 (0.39-4.66)	1 article	0.59 (0.27-1.28)	2 articles I2= 0.00%
    Macrophages	1.11 (0.65-1.90)	5 articles I2= 18.11	1.04 (0.69-1.55)	6 articles I2= 0.00	-------	-------	1.88 (0.87-4.08)	4 articles I2= 43.97%
    Mast Cells	1.81 (1.01-3.15)	3 articles I2=0.00	2.01 (0.88-3.92)	3 articles I2=43.99	2.30 (1.20-4.70)	1 article	1.50 (0.60-3.60)	1article

    HR= Hazard Ratio, CI = Confidence Interval

    Conclusion
    

    Immune cell subsets are able to provide prognostic information in early-stage NSCLC undergoing surgical resection. When evaluating these immune biomarkers, adjustment for clinical covariates can have a profound impact on survival estimates.

    Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.

  • 31763

    +

    P1.17-28 - Refusal of Surgery for Early Stage Lung Cancer: Risk Factors and Survival (ID 1395)

    09:45 - 18:00  |  Author(s): Maaike Van Gerwen
    • Abstract
    • Slides

    Background
    

    The standard of care for treatment of early stage non-small cell lung cancer (NSCLC) is surgical resection. However, suitable patients refuse surgery for various reasons. We characterize the clinical and sociodemographic predictors of those who refuse recommended surgery for early stage NSCLC cancers and explore the influence of surgical compliance on survival.

    Method
    

    Primary Stage I and II NSCLC cases diagnosed between 2007 and 2014 were selected from the Surveillance, Epidemiology and End Results (SEER) database (n = 36,926). All cases were recommended surgery for treatment. Predictors of surgery refusal were examined using multivariate logistic regression. The likelihood of mortality after refusing surgery and any chemotherapy/radiation and refusing surgery but receiving alternative forms of treatment was performed using a cox proportional hazards model. Propensity-matched survival analysis was then performed between those who received surgery and those who refused surgery. SAS v9.4 was used for statistical analyses.

    Result
    

    The majority of the sample was non-Hispanic White (79%), female (52%), married (57%) and Stage I (83%). Most were between 50-64 years old (31%) or 65-79 years old (55%) and had tumor sizes of 11-20 mm (32%), 21-40 mm (42%), or > 40 mm (20%); 909 cases (2.5%) refused surgical intervention. Of these, 634 (69.7%) cases underwent radiation therapy and 87 (9.6%) received chemotherapeutic treatment. At multivariable analysis, non-Hispanic blacks (OR_adj 2.14, 95% CI: 1.76-2.61), increasing age (65-79 years old: OR_adj 4.29, 95% CI: 2.02-9.11), increasing tumor size (21-40 mm: OR_adj 2.79, 95% CI: 1.73-4.50), and single marital status (OR_adj 2.14, 95% CI: 1.85-2.48) were associated with increased odds of surgical refusal. Stage II lung cancer was inversely associated (OR_adj 0.75, 95% CI: 0.62-0.92) with surgical refusal. Refusing surgery and having no chemotherapy/radiation (HR_adj 4.16, 95% CI: 3.48-4.96) as well as refusing surgery but undergoing alternative forms of treatment (HR_adj 2.90, 95% CI: 2.55-3.29) were associated with increased mortality compared to those who received recommended surgery. After propensity matching (n = 1790), refusing surgery was associated with increased likelihood of mortality (HR_adj 2.77, 95% CI: 2.22-3.46).

    Conclusion
    

    Identifiable risk factors exist for refusing recommended surgery for Stage I/II NSCLC, and refusal is associated with an increased likelihood of mortality. Recognizing that certain subgroups are more likely to refuse surgery is vital when providing treatment choices and reducing disparities in survival for early stage lung cancer.

    Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.

• 31105

  +

  P2.06 - Mesothelioma (ID 170)

  • Event: WCLC 2019
  • Type: Poster Viewing in the Exhibit Hall
  • Track: Mesothelioma
  • Presentations: 1
  • Now Available
  • Moderators:
  • Coordinates: 9/09/2019, 10:15 - 18:15, Exhibit Hall

  • 31124

    +

    P2.06-19 - A Meta-Analysis of Mesothelin, Osteopontin, and Fibulin-3 as Biomarkers of Malignant Pleural Mesothelioma (Now Available) (ID 1729)

    10:15 - 18:15  |  Author(s): Maaike Van Gerwen
    • Abstract
    • Slides

    Background
    

    Malignant Pleural Mesothelioma (MPM) is a rare and aggressive cancer; the use of mesothelin, fibulin-3, and osteopontin as biomarkers has been proposed as a screening tool. The following meta-analysis examines average levels of these biomarkers in blood and pleura in comparison to several control groups.

    Method
    

    A PubMed search was conducted for studies that measured the levels of mesothelin, osteopontin, and fibulin-3 in patients with MPM compared to several control groups (other malignancies, benign lung disease, or healthy participants).

    Result
    

    Thirty two studies with 28 distinct datasets were identified as reporting mesothelin levels. Statistically significant mean differences (MD) were observed between mesothelin in the blood of MPM patients and each of the control groups. The overall MD (95% confidence interval) between MPM and non-MPM controls was 3.88 (2.73-5.05) nM/L (nMPM=455). Statistically significant MDs in pleural levels of mesothelin were also seen between MPM patients and non-MPM cancers and benign lung diseases. The overall MD (95%CI) was 31.45 (23.63-39.27) nM/L (nMPM=571). Twelve articles with 10 distinct datasets reported the mean level of osteopontin in MPM cases and controls. Statistically significant MDs in blood levels of osteopontin were seen between MPM patients and benign lung diseases or healthy controls (overall MD (95% CI) 103.75 (54.77-152.72) ng/mL (nMPM=504)), but not with other cancers. Studies on pleural levels of osteopontin were too scarce to conduct a meta-analysis. Nine articles measured fibulin-3 in blood or pleura of MPM patients compared to controls. There was a statistically significant MD in blood levels of fibulin-3 between MPM and each of the control groups. The overall MD (95%CI) was 46.52 (34.04-59.00) ng/mL (nMPM=330). There was also a statistically significant MD in the pleural levels of fibulin-3 between MPM and the cancer and benign lung disease groups. The overall MD (95% CI) was 385.35 (287.98-482.73)ng/mL (nMPM=137).

    Conclusion
    

    This meta-analysis supports the concept that mesothelin and fibulin-3 in control groups are significantly lower than in MPM, and thus they might be useful as screening biomarkers. Further studies are necessary to better assess the value of these promising biomarkers.

    Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.

• 31680

  +

  P2.16 - Treatment in the Real World - Support, Survivorship, Systems Research (ID 187)

  • Event: WCLC 2019
  • Type: Poster Viewing in the Exhibit Hall
  • Track: Treatment in the Real World - Support, Survivorship, Systems Research
  • Presentations: 1
  • Moderators:
  • Coordinates: 9/09/2019, 10:15 - 18:15, Exhibit Hall

  • 31681

    +

    P2.16-01 - Risk Factors for Short-Term Post-Operative Events Following Lung Cancer Resection (ID 1381)

    10:15 - 18:15  |  Author(s): Maaike Van Gerwen
    • Abstract
    • Slides

    Background
    

    Lung cancer represents 13% of all newly diagnosed US cancers and is the leading cause of all cancer deaths (25%). Reducing postoperative complications and improving quality of life following surgery remains central when providing treatment. We examined risk factors for short-term post-opeartive complications following lung cancer resection in patients treated at Mount Sinai, New York City.

    Method
    

    Data was selected from the General Thoracic Surgery Database (2012-2018). Patients with a primary diagnosis of lung cancer who underwent a Lobectomy (n = 603) or a Wedge resection (n = 659) were included. Preoperative assessments of health, including the Zubrod Performance Score, % predicted FEV1, heart (congestive heart failure, coronary artery disease, peripheral vascular disease) lung (pulmonary hypertension, chronic obstructive pulmonary disease, and interstitial fibrosis) comorbidities, diabetes, and smoking status were measured for each patient. The primary outcome was the occurence of at least one post-operative event (pulmonary/ cardiovascular/gastrointestinal/neurological complications, infections, and unexpected readmissions to the operating room or intensive care unit) during their hospital stay or within 30 days of their surgery; SAS v9.4 was used for statistical analyses.

    Result
    

    There were 1262 patients (age 18-95 years); 60% were female, 64% were White, and 55% were Stage 1a / 1b. More patients underwent surgery with a Video-Assisted Thoracoscopic Surgery (VATS) approach (58%) rather than an Open approach (42%). The majority of patients was classified as past smokers (58%), while the remainder were never (27%) and current (15%) smokers. 17% of patients presented with at least one heart comorbidity, while 21% with at least one lung comorbidity. At multivariable analysis, female gender (OR_adj 0.64, 95% CI: 0.49-0.96), Wedge resection (OR_adj 0.64, 95% CI: 0.44-0.92 compared to Lobectomy), VATS (OR_adj 0.54, 95% CI: 0.38-0.77 compared to an OPEN approach), and an increasing % predicted FEV1 were inversely associated with the occurence of a post-operative event. Increasing age (OR_adj 1.03, 95% CI: 1.01-1.04), Zubrod performance score (OR_adj 1.53, 95% CI: 1.21-1.93), duration of surgery in hours (OR_adj 1.10, 95% CI: 1.01-1.21), and currently smoking (OR_adj 1.97, 95% CI: 1.14-3.40) were associated with increased odds of a post-operative event.

    Conclusion
    

    Significant risk factors exist for the occurence of a short-term post-operative event following lung cancer resection, which should be recognized to improve post-surgery quality of life. A VATS wedge resection should be utilized whenever appropriate to reduce the likelihood of post-operative complications.

    Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.

• 31780

  +

  P2.17 - Treatment of Early Stage/Localized Disease (ID 189)

  • Event: WCLC 2019
  • Type: Poster Viewing in the Exhibit Hall
  • Track: Treatment of Early Stage/Localized Disease
  • Presentations: 1
  • Moderators:
  • Coordinates: 9/09/2019, 10:15 - 18:15, Exhibit Hall

  • 31793

    +

    P2.17-11 - Evaluating PD-L1 as a Prognostic Biomarker in Surgically Resectable Non-Small Cell Lung Cancer (ID 1959)

    10:15 - 18:15  |  Author(s): Maaike Van Gerwen
    • Abstract

    Background
    

    Identifying biomarkers that can predict which early stage non-small cell lung cancers (NSCLCs) are likely to recur following surgical resection is critical to improving survival outcomes. Programmed Cell Death Ligand-1 (PD-L1) is an immune regulatory protein expressed on tumor cells that inhibits T effector response against tumors. Therefore, high PDL-1 expression on tumor cells may enable evasion of the anti-tumor response and be associated with worst outcomes. We conducted a meta-analysis to determine if PDL-1 expression is associated with survival in surgically resected early stage NSCLCs.

    Method
    

    PubMed was searched to identify eligible studies comparing survival of surgically resected stage I-III NSCLC patients according to PD-L1 tumor expression. Included studies were grouped according to measurement criteria of PDL-1 expression: 1%, 5%, 50% cutoffs or as a continuous variable expressed as H-score. The latter is calculated from the percentage of cells expressing PDL1 multiplied by the intensity of staining. Meta-analysis was performed using a linear mixed-effects model to determine overall survival (OS). I² was used as a measure of heterogeneity across studies.

    Result
    

    Of the 519 articles queried, 31 articles met eligibility criteria, accounting for 6,713 patients. Nine studies (n=2,407) used H score, where higher PD-L1 expression was associated with worse OS (hazard ratio [HR]_meta: 1.68, 95% confidence interval [CI]: 1.17-2.41, I²= 15.45%). Five studies used a 1% cutoff (n=1,073), 14 studies reported using a 5% value (n=2,310) and 6 studies (n=1,572) used a 50% cutoff for evaluating high vs. low PD-L1 expression. PD-L1 expression was not statistically significantly associated with survival according to these cutoffs and there was high inter-study heterogeneity (HR_meta: 1.47, 95% CI 0.89-2.41; I²= 17.57%; HR_meta: 1.09, 95% CI 0.79-1.52; I²= 35.49%; HR_meta: 1.21, 95% CI 0.61-2.40; I²= 58.26% for 1%, 5%, 50%, respectively).

    Conclusion
    

    Higher PD-L1 expression in early stage NSCLCs as measured by H scores was associated with worse post-surgical survival. The measurement of PD-L1 expression as a continuous variable (i.e. H score) may provide more accuracy for predicting survival outcomes over percentage cutoffs. Future research is needed to validate PD-L1 expression as a predictive biomarker of survival for early-stage NSCLCs undergoing surgical resection.