Citations

Context.— Transbronchial cryobiopsy has recently been proposed as an alternative to surgical biopsy in the diagnosis of diffuse lung disease. Objective.— To familiarize pathologists with transbronchial cryobiopsy, including what it is, how it is performed, how it compares to other techniques of lung biopsy in diffuse lung disease, what are the technical issues relating to it, what the complications are, how cryobiopsies should be interpreted, and the clinical usefulness of cryobiopsy. Data Sources.— All the available literature on cryobiopsy in diffuse lung disease through May 2016, primarily in the last 5 years, was reviewed, and some unpublished data known to the authors were included. Conclusions.— Cryobiopsies are considerably larger than forceps biopsies and allow pattern recognition approaching that of a surgical lung biopsy in many cases. Artifacts associated with cryobiopsy are minimal. In comparison with surgical lung biopsies, the diagnosis rate with cryobiopsies is lower, in the neighborhood of 80%, versus higher than 90% for surgical lung biopsies. Cryobiopsy is proposed as an alternative to surgical lung biopsy and a technique that may appreciably decrease the number of patients who require surgical lung biopsy for diagnosis. This is important because the mortality from cryobiopsy is very small (0.1% to date) compared with surgical lung biopsy (1.7% for elective procedures and considerably higher for nonelective procedures).

Abstract Rationale The natural history of lung function in patients with bronchiolitis obliterans syndrome (BOS) after allogeneic hematopoietic cell transplant is poorly characterized. Understanding the trajectory of lung function is necessary for prompt clinical recognition and treatment and also for the rational design of prospective studies. Objectives To describe the longitudinal trajectory of lung function parameters, including FEV1, in patients with BOS after hematopoietic cell transplant. Methods Subjects with BOS defined by National Institutes of Health consensus guidelines criteria from a recent multicenter prospective trial of combination treatment with fluticasone, azithromycin and montelukast and a retrospective cohort from Fred Hutchinson Cancer Research Center were included. Longitudinal change in FEV1 for each patient was calculated on the basis of available pulmonary function tests in three periods: pre-BOS, from BOS diagnosis to 6 months, and 6–18 months after diagnosis. The effect of treatment on FEV1 trajectory was analyzed by univariate and multivariate linear regression. The Kaplan-Meier method was used to estimate survival. Measurements and Main Results: The FEV1 percent predicted value at diagnosis was 46% (interquartile range, 35–57%) for trial participants and 53% (interquartile range, 41–64%) for the retrospective cohort. There was a concomitant mild reduction in FVC, as well as a marked reduction in forced expiratory flow, midexpiratory phase, at diagnosis. While there was individual heterogeneity, the overall FEV1 trajectory was characterized by a marked decline within 6 months prior to BOS diagnosis, followed by stability of FEV1 early after diagnosis and a slow rate of decline beyond 6 months. The effect of the trial medications on FEV1 trajectory after BOS diagnosis was a mean rate of change of 0.92% predicted per month (95% confidence interval, −0.53 to 2.37) compared with the retrospective cohort, but this was not statistically significant. Two-year overall survival rates were 76% and 72% for the study participants and the retrospective cohort patients, respectively. Earlier time to diagnosis after hematopoietic cell transplant and severity of FVC at diagnosis were significantly associated with reduced survival. Conclusions The FEV1 trajectory in patients with BOS after hematopoietic cell transplant in a contemporary era of management follows a predominant pattern of rapid FEV1 decline in the 6 months prior to diagnosis, followed by FEV1 stabilization after diagnosis.

Objective: The Lung Injury Prediction Score identifies patients at risk for acute respiratory distress syndrome in the emergency department, but it has not been validated in non-emergency department hospitalized patients. We aimed to evaluate whether Lung Injury Prediction Score identifies non-emergency department hospitalized patients at risk of developing acute respiratory distress syndrome at the time of critical care contact. Design: Retrospective study. Setting: Five academic medical centers. Patients: Nine hundred consecutive patients (≥ 18 yr old) with at least one acute respiratory distress syndrome risk factor at the time of critical care contact. Interventions: None. Measurements and Main Results: Lung Injury Prediction Score was calculated using the worst values within the 12 hours before initial critical care contact. Patients with acute respiratory distress syndrome at the time of initial contact were excluded. Acute respiratory distress syndrome developed in 124 patients (13.7%) a median of 2 days (interquartile range, 2–3) after critical care contact. Hospital mortality was 22% and was significantly higher in acute respiratory distress syndrome than non-acute respiratory distress syndrome patients (48% vs 18%; p < 0.001). Increasing Lung Injury Prediction Score was significantly associated with development of acute respiratory distress syndrome (odds ratio, 1.31; 95% CI, 1.21–1.42) and the composite outcome of acute respiratory distress syndrome or death (odds ratio, 1.26; 95% CI, 1.18–1.34). A Lung Injury Prediction Score greater than or equal to 4 was associated with the development of acute respiratory distress syndrome (odds ratio, 4.17; 95% CI, 2.26–7.72), composite outcome of acute respiratory distress syndrome or death (odds ratio, 2.43; 95% CI, 1.68–3.49), and acute respiratory distress syndrome after accounting for the competing risk of death (hazard ratio, 3.71; 95% CI, 2.05–6.72). For acute respiratory distress syndrome development, the Lung Injury Prediction Score has an area under the receiver operating characteristic curve of 0.70 and a Lung Injury Prediction Score greater than or equal to 4 has 90% sensitivity (misses only 10% of acute respiratory distress syndrome cases), 31% specificity, 17% positive predictive value, and 95% negative predictive value. Conclusions: In a cohort of non-emergency department hospitalized patients, the Lung Injury Prediction Score and Lung Injury Prediction Score greater than or equal to 4 can identify patients at increased risk of acute respiratory distress syndrome and/or death at the time of critical care contact but it does not perform as well as in the original emergency department cohort.

Abstract Background: We recently have reported that co-mutations within KRAS-mutated lung adenocarcinoma (LUAD) define a distinct biological phenotype and may offer insights into therapeutic targeting using this classification system. The analysis revealed striking differences in checkpoint ligand expression and T cell infiltration between TP53 and STK11 (LKB1) co-mutated KRAS LUAD. We, therefore, hypothesized that loss of LKB1promotes an immune-suppressed phenotype in KRAS LUAD. To test this, we analyzed 100 immune-related genes from The Cancer Genome Atlas datasets, as well as three other sets of resected tumors from NSLC patients from MD Anderson Cancer Center. Results: Our results revealed that tumors lacking LKB1 (LKB1-loss), as defined by mutation or gene-expression signature, resulted in a down regulation of: tumor associated immunosuppressive ligands, PD-L1, PD-L2, B7-H3, B7-H4, HVEM; T cell associated immunosuppressive ligands PD-1, CTLA-4, BTLA; co-stimulation ligands such as CD80, -86, -40L, and -40. We further examined if this loss of immunosuppression and co-stimulation resulted in changes in markers of T cell phenotypes. Indeed, tumors lacking LKB1-loss tumors demonstrated a distinct lack absence of several T cell markers including CD3, -4, & -8, but also RORC, T-bet, GATA-3, and FoxP3. To determine if myeloid cell populations, known mediators of immune suppression, were also diminished in LKB1-loss tumors, we analyzed expression of CD11c, -103, -11b, -68, and CD33. Consistent with T cell makers and suppressive ligand markers, myeloid markers were down regulated in LKB1-loss tumors. Because common immune suppression markers were lost, we then examined the expression of antigen presentation (HLA-I, HLA-II) and amino acid transporters as possible alternative methods for immune suppression. In LKB1-loss tumors, both HLA-I and HLA-II markers were down regulated, compared to tumors with intact LKB1 expression. Furthermore, genes associated with the antigen presentation complex, LMP2, TAP1&2, and B2M were also down regulated. In contrast to the immune markers described above, cationic transporters (SLC7A2), anionic transporters (SLC7A11), and neutral transporters (SLC3A2) were universally upregulated in LKB1-loss tumors. Conclusions: LKB1 loss is associated with an inert immune phenotype. Our data suggests that reduced antigen presentation and amino acid depletion, but not checkpoint factors or MDSCs, are likely mediator of this phenotype. Citation Format: Warren L. Denning, Ferdinandos Skoulidis, Li Shen, Vassiliki Papadimitrakopoulou, Lixia Diao, Yanyan Lou, Lauren A. Byers, Jing Wang, Jaime R. Canales, Ignacio I. Wistuba, John Weinstein, Don L. Gibbons, John V. Heymach. Loss of LKB1 mediates an immune inert phenotype in human lung adenocarcinoma. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 4137.

Abstract BACKGROUND Genomic intra- and intertumor heterogeneity is one main reason for relapse and resistance to therapy. There is a shortage of studies characterizing the level of genomic intertumoral heterogeneity of known cancer genes in multiple longitudinal biopsies of individual patients. Additionally, a workflow to deconvulate the intermixture of tumor and stromal components is missing. To overcome these limitations and to decipher the genomic heterogeneity and clonal evolution, we developed a method and applied it on matched (primary-recurrence/metastasis) non-squamous, non-small cell lung cancers (NSCLC). METHODS Multiparameter Ploidy Profiling (MPP) comprises the isolation of nuclei from frozen or formalin and paraffin embedded (FFPE) tissues, followed by multiparamter flow sorting. Sorted populations were subjected to genomic profiling by high resolution array comparative genomic hybridization (aCGH) and massively parallel sequencing (MPS). DAPI allowed to seperate populations by ploidy and anti-TTF1 antibodies was used to control for tumor nuclei. Array-CGH, combined with ploidy, was used to retrieve genome wide copy numbers. The Comprehensive Cancer Panel that covers all-exons of 409 cancer genes was applied on all sorted tumor and stromal populations to detect somatic mutations and their variant allelic frequency (VAF). RESULTS MPP was successfully applied to 44 frozen or FFPE tissue specimens from 19 patients. Array-CGH and MPS of TTF1-negative, normal cells were concordant to germline controls. Sequencing revealed that 50% of mutations are shared between primary tumors and metastases. Except for one patient, mutations with VAF&gt;0.3 are shared between primary and metastasis. The variant allelic frequency was significantly higher in shared mutations than mutations that were private to one tumor. Besides common activating mutations in EGFR and KRAS we found biallelic inactivation in tumor suppressor genes like KEAP1, NF1, STK11 and TP53. Two clonal evolutionary patterns were found: 1) early and 2) late divergence. Matched tumors without any shared mutations were classified as unrelated primary tumors. CONCLUSION The power of MPP is to increase the precision of downstream analysis, due to the sorting of pure populations of tumor cells. It permitted to infer the clonal evolution of tumor populations with unprecedented confidence. The low level of genomic heterogeneity of mutations with VAF&gt;0.3 in this study is in line with recent data from Zhang et al., who showed that the poor precision of low depth sequencing (&lt;277x) contributes to an overestimation of genomic heterogeneity (Zhang et al., Science 2014). Integrational analysis of ploidy, chromosomal aberrations and mutations in 409 cancer genes allowed to draw the evolutionary picture of each patient's disease. MPP appears as a promising tool for identification of genomic vulnerabilities that could be exploited for tailored treatment. Citation Format: Thomas Lorber, Tanja Dietsche, Valeria Perrina, Darius Juskevicius, Arthur Krause, David Mueller, Michael Barrett, Christian Ruiz, Lukas Bubendorf. Clonal evolution and genomic tumor heterogeneity in non-small cell lung cancer deciphered by multiparameter ploidy profiling. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 158.

Abstract Background: Genomic tumor heterogeneity is one reason for recurrence and resistance to cancer therapy. It is still not known what leads to the relapse or the emergence of metastasis in individual patients. To contribute to answer this question, we investigated the evolutionary patterns in squamous cell carcinoma of the lung (SCC) in 10 patients with multiple matched tumors (primary vs recurrence/metastasis). Methods: To identify the genomic profile, frozen tissues from SCC patients were subjected to multiparameter ploidy profiling (MPP). Tumor populations were sorted based on ploidy (DNA-content using DAPI) and p40 was used to control for diploid tumor populations. Subsequent genomic profiling by array comparative genomic hybridization (aCGH) and targeted massive parallel sequencing (Ion Torrent Comprehensive Cancer Panel with an all-exon coverage of 409 cancer genes) was applied. Results: The usage of MPP enabled the determination of the clonal composition of the SCC and define genomic characteristics. Due to flow sorting the tumor populations were effectively isolated and existence of distinct clonal populations was indicated. Aneuploid and diploid tumors were successfully revealed using the additional marker p40. Therefore, the absence of non-malignant cells increased the precision in the genomic analyses Conclusion: The study uncovered the composition of human SCC in the sense of different clonal populations, which harbor private genomic aberrations and mutations confirmed by aCGH and NGS. Even if our data is still preliminary we provide evidence that applying MPP increases the precision of aCGH and sequencing analysis. Genomic analyses of sorted pure tumor population allows to reconstruct the clonal evolution and facilitates the understanding of tumor heterogeneity and their evolutionary pattern, which could have potentially an impact on therapeutic response or metastasis development. Citation Format: Arthur Krause, Thomas Lorber, Valeria Perrina, Tanja Dietsche, Michael Barrett, Christian Ruiz, Lukas Bubendorf. The analysis of the genomic evolution of squamous cell carcinoma of the lung. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 155.

Abstract Several papers have implicated mutations in the AT-rich interactive domain containing protein 1B (ARID1B) in carcinoma formation in multiple tissue types, including the lung. The ARID1B protein functions as part of SWI/SNF, a multi-subunit chromatin remodeling complex that repositions nucleosomes and may act as a tumor suppressor in cancer. ARID1B is located in a region in 6q that was previously shown to have strong evidence of linkage with lung cancer risk. As it is one of the more promising candidates within the target region, this study presents the viability of ARID1B as a candidate gene for familial lung cancer. Here, we chose individuals from families in our previous linkage study that showed the strongest linkage signal on 6q. 75 individuals from 9 families were sequenced for 37Mb of chromosome 6 from 130Mb to 167Mb using Illumina technology and a custom Agilent kit. Genotypes were imported into GoldenHelix SVS 7 for quality control and dropped if read depth &lt; 10, quality score &lt; 10 or a quality score to read depth ratio &lt; 0.5. Variants with more than 1 Mendelian error were dropped. Variants with a single Mendelian inconsistency were dropped only for that family. Two-point parametric linkage analysis at theta = 0 was performed using R and an implementation of the Elston-Stewart algorithm in the paramlink package using a penetrance model of 0.01/0.1/0.1 (dd/Dd/DD) and a disease allele (D) frequency of 0.01. Family 102 showed strong evidence for linkage near ARID1B. The highest LOD score (1.34) out of all sequenced variants was located near this gene in a known conserved intergenic region between ARID1B and NOX3. Four other LOD scores &gt; 0.7 were found in the intergenic region between these two genes. Two are transcription factor binding sites; two are conserved genomic elements. We also observed two linked intronic variants of ARID1B in this family; an enhancer (LOD = 0.89) and a transcribed region (LOD = 0.73). Two other families (47 and 59) displayed evidence of linkage in ARID1B. Here, the interest is not the value of the LOD scores (top scores were 0.64 and 0.54), but the centering of the linked haplotype around the gene. Family 47 has 60 variants with LODs between 0.64 and 0.61; 45 located within ARID1B or known regulatory regions. Family 59 has 61 variants with LODs between 0.54 and 0.51; 29 located within ARID1B or known regulatory regions The variants appear to be spread evenly throughout ARID1B, suggesting the presence of linked haplotypes. Analyses are currently underway to reconstruct possible haplotypes in these families. The results of this study show that variants in the ARID1B gene at least partially account for the linkage signal on 6q in three of the nine most strongly linked families in our cohort and thus we believe that ARID1B is a plausible candidate gene for further studies involving familial lung cancer. Replication studies using 27 new extended families are currently underway to confirm this initial result. Citation Format: Anthony M. Musolf, Claire L. Simpson, Mariza de Andrade, Diptasri Mandal, Colette Gaba, Ping Yang, Ming You, Elena Y. Kupert, Marshall W. Anderson, Ann G. Schwartz, Susan M. Pinney, Christopher I. Amos, Joan E. Bailey-Wilson. Chromatin remodeling gene ARID1B is linked to familial lung cancer. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 2542.

Abstract Lung cancer accounts for about 27% of the cancer related deaths in the USA annually. Pathologically, it is a very complex disease broadly classified into small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC). NSCLCs are further classified into squamous cell carcinoma, large cell carcinoma and adenocarcinoma. Achaete-scute homolog 1 (ASCL1), an important transcription factor essential in the development of neuroendocrine cells (NE) in lungs is shown to be specifically expressed in lung NE cancers and 10-20% of adenocarcinomas (AD) with NE differentiation (NED), thus suggesting a role in the pathogenesis of these tumors. Our previous study showed that ASCL1 is a regulator of the RET oncogene in AD with NED. RET is a receptor tyrosine kinase with two isoforms in humans: RET9 (short) and RET51 (long). We performed survival analysis to study implications of RET isoforms in ASCL1+ tumors and found that elevated expression of the long RET mRNA was associated with poor survival. Subsequent in vitro experiments demonstrated that treatment with EGF robustly induced phosphorylation of RET in HCC1833 and H1755 cell lines which have high endogenous levels of ASCL1 and RET but not in VMRC-LCD cell line which has high level of ASCL1 but low level of RET. EGF induced phosphorylation of RET was diminished by gefitinib and by EGFR siRNA. Immunoprecipitation results indicated direct binding between EGFR and RET in presence of EGF. Furthermore, a high throughput drug screening found 8 EGFR inhibitors that were 10 - 250 fold more cytotoxic in ASCL1+ compared with ASCL1- AD cells. These results implicate EGFR as a key regulator of RET activation in ASCL1+ AD and suggest that EGFR inhibitors may be therapeutic for this population of patients. Citation Format: Kaustubh N. Bhinge, Yang Lin, Hamed Rahi, Marie Christine Aubry, Aaron Mansfield, Irina Kovtun, Stephen Murphy, Ping Yang, Dennis Wigle, Joanne (Eunhee) Yi, Aqsa Nasir, Simone Terra, Julian Molina, George Vasmatzis, Farhad Kosari. EGFR-mediated activation of RET in ASCL1+ lung adenocarcinoma. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 1129.

An effort over 3 years to conduct a data pooling project regarding the association between smoking and lung cancer treatment outcomes has shown that, of 32 completed studies of lung cancer treatments in 1 National Cancer Institute Cooperative Group, only a minority have usable data concerning smoking‐related variables. Clinical trials research is needed to inform patients with cancer and their physicians about the impact of smoking, and how smoking cessation may affect the course of their disease.

Ciliated muconodular papillary tumors (CMPTs) are rare peripheral lung lesions, characterized by papillary architecture and ciliated columnar cells admixed with mucinous cells and basal cells. They often have prominent surrounding intra-alveolar mucin, which can lead to diagnostic confusion with mucinous adenocarcinoma. Recognition of the ciliated component is the key to diagnosis of CMPT. The literature contains few reported cases to date, all occurring in East-Asian patients. Although follow-up data are limited, CMPT seems to be an indolent tumor with very good prognosis, leading some to question whether it is a reactive or hamartomatous lesion. However, a very recent molecular study has identified BRAF (40%) and EGFR (30%) alterations in CMPT, supporting a truly neoplastic process. Here for the first time, we report 4 cases of morphologically typical CMPT in western patients, occurring in 1 man (60 y) and 3 women (71 to 83 y). Interestingly, 1 case occurred in background of pronounced small airway disease with necrotizing bronchiolitis and multiple carcinoid tumorlets. We further analyzed 1 tumor using a 50 gene next-generation sequencing oncology panel that identified 2 pathogenic mutations (BRAF V600E and AKT1 E17K). Our study is the first to describe that CMPT can occur in western (non-Asian) patients. Our data confirm BRAF V600E mutation as a probable driver in a subset of these tumors, along with AKT1 mutation, which further supports that CMPT are indolent pulmonary neoplasms.