IN THIS ISSUE:
Molecular Testing Guideline for Selection of Lung Cancer Patients for EGFR and ALK Tyrosine Kinase Inhibitors
Considering the number of publications on lung cancers bearing EGFR mutations and the rapid pace of ALK related development, there is a critical clinical need to establish molecular testing standards and recommendations to guide EGFR- and ALK-targeted therapies. Three co-chairs, a writing panel, and an advisory panel from the College of American Pathologists, International Association for the Study of Lung Cancer, and Association for Molecular Pathology were selected to critically and systematically review published data and develop evidence-based recommendations for the molecular analysis of lung cancers for these two predictive biomarkers in a clinical practice guideline (CPG). The evidence was gathered through three unbiased literature searches, which were reviewed for relevance to the recommendations, and formally graded for each recommendation by its strength in assisting users of the CPG in clinical decision making. Initial recommendations were formulated by the co-chairs and panel members at a public meeting, and the drafts were then disseminated for review sequentially to the writing panel, advisory panel, and the public prior to submission. The CPG comprises 37 guideline items that address 14 subjects, including 15 recommendations based on evidence rated with strong grades (A/B). Of these, the major recommendations are to use analysis of EGFR mutation and ALK fusion to select patients with advanced-stage adenocarcinoma for EGFR- and ALK-targeted therapy, regardless of sex, race, smoking history, or other clinical risk factors. These clinical characteristics were found to be not specific or sensitive enough to either select a patient for a targeted inhibitor, or serve as prerequisites in selecting an individual for molecular testing for aberrations in EGFR and ALK. Testing for EGFR mutations and ALK fusions should also be prioritized over other molecular analysis. The present molecular testing guideline to select lung cancer patients for EGFR and ALK TKIs was developed on limited published data, integrating expert consensus opinions, and anticipating regular updates when new testing indications or innovative technologies emerge for routine clinical implementation. (p. 823)
Four-Year Results of Low-Dose CT Screening and Nodule Management in the ITALUNG Trial
ITALUNG is a randomized clinical trial evaluating the efficacy of LDCT in lung cancer screening in Italy. This article covers the final data from the ITALUNG trial on subjects’ compliance, results of low-dose CT (LDCT) screening, and the management protocol of the screen-detected suspicious nodules in the active arm. Current or former smokers were randomized to receive either usual care or four annual LDCT screenings. Procedures included in the management protocol were CT-guided fine needle aspiration biopsy (FNAB), follow-up LDCT and 2-[18F]fluoro-2-deoxy-D glucose positron emission tomography (FDG-PET). Of 1406 subjects receiving baseline LDCT, 30.3% were positive while 15.8% of the 1263 subjects were positive in subsequent screening rounds. The rates of screen-detected lung cancers were low: 1.5% in subjects at baseline and 0.5% in subjects at subsequent screening. At least one follow-up LDCT was performed in slightly over half of the subjects. FDG-PET and FNAB both showed 84% to 90% sensitivity and specificity for malignancy. The authors concluded that high-risk individuals enrolled in LDCT screening were highly compliant throughout the screening cycle. However, the low tumor detection rate and specificity compromised the efficacy of LDCT. This confirms the need for a better defined target population and possibly the use of biomarkers in risk stratification to improve the cost effectiveness of screening. The low rate of surgery for benign lesions in this trial indicates that these surgeries could be minimized by strict adherence to a management protocol including follow-up LDCT with or without 1-month antibiotic therapy, FDG-PET, and CT-guided FNAB. (p. 866)
Improving the Quality of Lung Cancer Care in Ontario: The Lung Cancer Disease Pathway Initiative
Cancer Care Ontario (CCO) has incorporated Disease Pathway Management (DPM) to its provincial cancer control quality improvement initiatives with three goals for specific cancers: to share evidence-based best practice, to determine priorities in quality improvement and catalyze action, and to monitor performance against best practice. The authors described the process and some early successes of the establishment of the CCO lung cancer DPM since 2009. Five working groups across disciplines were formed to focus on prevention, screening, and early diagnosis; diagnosis; treatment; palliative care, end-of-life, survivorship; and patient experience. The working groups developed quality improvement projects from which a “Priorities for Action Report” was generated and disseminated across the province. Regional “roadshows” were held where regional care providers shared region-specific data on incidence, stage, treatment compliance, wait times, and other lung cancer relevant issues. CCO provided funding for possibilities of regional improvement based on the data and priorities. As a result of this initiative, the lung cancer pathways were refined. Available on CCO’s website now are the diagnostic pathway and the treatment pathways for SCLC and NSCLC, as well as educational materials for dyspnea management. There are also Lung Cancer Diagnostic Assessment Units/Programs across the province. DPM has improved access of lung cancer patients to a standardized Edmonton Symptom Assessment System to evaluate their symptoms each month. An audit to determine the challenge of uniform employment of specific evidence-based practices across the province is underway. All in all, CCO’s lung cancer DPM has raised awareness of Regional Cancer Programs (RCP) on lung cancer-specific performance and quality of care. Continuous assessment of a broad range of metrics will identify the benefits of this initiative. (p. 876)
Detection of ALK Rearrangement by Immunohistochemistry in Lung Adenocarcinoma and the Identification of a Novel EML4-ALK Variant
Lung adenocarcinomas bearing the oncogenic driver, EML4-ALK fusion gene, are responsive to the treatment of ALK inhibitors. However, the gold standard for detection of ALK rearrangement, namely fluorescence in situ hybridization (FISH) and RT-PCR, have their limitations. Immunohistochemistry (IHC) has been considered an alternative to FISH but optimal IHC protocol is needed. The authors therefore investigated the reliability of using routine IHC staining to detect ALK expression as a means to determine ALK rearrangement in lung adenocarcinoma and compared it with FISH. In this study, IHC and FISH were used to detect ALK rearrangement in 373 lung adenocarcinomas from Chinese patients. Multiplex RT-PCR was used to confirm the fusion variants in ALK positive cases by IHC or FISH. The findings indicated that all ALK negative cases determined by IHC were negative by FISH. All ALK positive cases by IHC were also positive by FISH or RT-PCR, except two cases that were FISH negative but harbored EML4-ALK variant 1 as confirmed by RT-PCR. Taken together, IHC can effectively detect ALK rearrangement in lung cancer with high sensitivity and specificity. It would be a reliable and cost effective tool in routine pathologic laboratories to identify patients suitable for ALK targeted therapy. In cases where lung cancers bearing the translocation and were ALK IHC-positive but FISH-negative, treatment with ALK inhibitors could be beneficial but further clinical trials are required. The study also led to the discovery of a novel variant of EML4-ALK (E3:ins53A20) with potent tumorigenicity potential in vivo. This variant might be important for tumorigenesis of lung cancer and further consolidates the role of ALK signaling in lung cancers. (p. 883)
RHOA-FAK Axis: Critical Vulnerability of KRAS-Driven Lung Adenocarcinomas
Dr. Scaglioni’s group set out to identify vulnerabilities in KRAS-mutant tumors, which could be exploited for cancer therapy. Cellular systems and a mouse model of high-grade lung adenocarcinoma induced by a recombinant transgene that allows activation of mutant KRAS in the respiratory epithelium, were used to elucidate the signaling pathways downstream of mutant KRAS in NSCLC. The findings showed that RHOA-focal adhesion kinase (FAK) signaling pathway is a critical requirement for the maintenance of high-grade lung tumors. Inhibition of RHOA or FAK induces selective cell death in mutant KRAS;INK4A/ARF-deficient lung cancer cells, and tumor regression specifically in mutant Kras;Cdkn2a-null mice bearing the high-grade lung cancer. This study revealed the RHOA-FAK signaling axis as a novel and critical vulnerability of mutant KRAS in NSCLCs that are INK4a/ARF- or TP53-deficient. Also, FAK inhibitors exert potent antitumor effects in NSCLCs bearing mutant KRAS in association with INK4A/ARF deficiency. This information would also provide insights for developing a biomarker to guide personalized cancer treatment using FAK inhibitors. Further studies in clinical trials using FAK inhibitors are warranted.
Konstantinidou G, Ramadori G, Torti F, et al. RHOA-FAK is a required signaling axis for the maintenance of KRAS-driven lung adenocarcinomas. Cancer Discov 2013;3:444–457.
Esophageal Adenocarcinoma (EAC): Dissecting the Mutational Complexity by Exome and Whole-Genome Sequencing
The increased incidence and poor prognosis of EAC reflect an urgent need to better understand the underlying cause for EAC and to help with the development of diagnostics or targeted therapies. Dulak and colleagues elucidated the genomic alterations of 149 EAC cases untreated by chemotherapy or radiation before surgery, using whole-exome sequencing of paired tumor and normal tissues. Fifteen of these were also subjected to whole-genome sequencing. Data from whole-genome sequencing revealed a mutation profile defined by high frequencies of A>C transversions at AA dinucleotide sites (34% of the total mutations), which appeared to be unique to EAC. A greater fraction of the transversions was found within tubular esophagus, which suggests stomach acid in the lower esophagus to be a potential mutagenic trigger. Data from whole-exome sequencing identified 26 significantly mutated genes, including TP53, CDKN2A, SMAD4, ARID1A and PIK3CA which have previously been reported in EAC. New mutations found include chromatin-modifying factors and candidate contributors SPG20, TLR4, ELMO1, and DOCK2. ELMO1 and DOCK2 are upstream regulators of the RAC1 GTPase. Functional analyses of EAC-derived mutations in ELMO1 demonstrated augmented cellular invasion. Therefore, RAC1 pathway could potentially play a role in EAC tumorigenesis. The current data also suggest that mutation could be a potential biomarker, in addition to overexpression and genomic amplification of ERBB2, to guide the use of ERBB2-targeting agents in the treatment of EAC. More functional studies of these genes are needed to elucidate their roles in EAC tumorigenesis and to determine the underlying cause of the unique transversions at AA dinucleotides in EAC.
Dulak AM, Stojanov P, Peng S, et al. Exome and whole-genome sequencing of esophageal adenocarcinoma identifies recurrent driver events and mutational complexity. Nat Genet 2013;45:478–486.
Estrogen Receptor 1 (ESR1) Expression as Prognostic and Predictive Marker in Non-Small Cell Lung Cancer (NSCLC)
Adjuvant cisplatin-based chemotherapy is the accepted standard of care for the management of stage II to IIIA completely resected NSCLC. Some patients with resected NSCLC are at a high risk of relapse after adjuvant chemotherapy (ACT) while some remain relapse free without ACT. Hence, prognostic and predictive markers for clinical outcome are required. Studies showed a correlation between tumors with high expression of estrogen receptors and improved response to anti-estrogens or aromatase inhibitors in breast cancers. Estrogen receptors are also expressed in lung tumors. In this study, the authors evaluated the association of estrogen receptor 1 (ESR1) expression with survival in patients with completely resected NSCLC. Publicly available gene microarray, clinical, and follow-up data from studies in early stage NSCLC were used. Expression values were determined against clinical and survival data in a training set (n = 138) and a test set (comprising subpopulation from the adjuvant JBR.10 study). High ESR1 expression was of significant positive prognostic value for early stage NSCLC patients in the training set population and in the observational arm of the test set population. Tumors with low ESR1 levels showed a benefit from ACT with a 5-year survival of 77.8% compared with 33.3% in the observational arm. Patients with high ESR 1 levels did not benefit from ACT. It is the first report that indicates that ESR1 expression in NSCLC is of prognostic value and could also be predictive in this setting. Further studies on ESR1 levels in larger cohorts of patients with NSCLC are warranted. This finding reflects the possibility to clinically stratify NSCLC patients by ESR1 expression in decision making on adjuvant therapy. It also prompts further studies to evaluate anti-estrogenic agents for treating NSCLC patients expressing high level of ESR1.
Brueckl WM, Al-Batran SE, Ficker JH, et al. Prognostic and predictive value of estrogen receptor 1 expression in completely resected non-small cell lung cancer. Int J Cancer April 12, 2013 [Epub ahead of print].
Genome-Wide Analysis Links Gene Expression, Cancer Cell Invasion and Anti-Cancer Compound Sensitivity
The authors conducted a genome-wide analysis to understand the molecular profile of invasion heterogeneity in cancer cells and its relationship to anti-cancer drug sensitivity. Using the NCI-60 panel, they identified 633 invasion-associated (IA) genes by invasion assay and Affymetrix gene expression data. The association of IA genes to drug sensitivity was evaluated from the analysis of chemosensitivty data from 99 anti-cancer compounds with known mechanism of action. Eight genes (EGFR, ITGA3, MYLK, RAI14, AHNAK, GLS, IL32, and NNMT) were identified with significant correlation with paclitaxel, docetaxel, erlotinib, everolimus, and dasatinib. This eight-gene signature was validated for its prediction for chemosensitivity in 107 independent experiments in 78 tumor cell lines (mostly outside the NCI-60 panel). Published clinical and molecular data from lung and breast cancer cohorts receiving adjuvant chemotherapy were analyzed using the eight-gene signature and it predicted relapse-free survival for the lung and breast cancer patients. Taken together, this study enhanced the NCI-60 model with important in vitro characteristics of cancer and revealed a set of genes that are associated to cancer invasion and chemosensitivity, which could be used to predict survival in patients with lung or breast cancer.
Hsu YC, Chen HY, Yuan S, et al. Genome-wide analysis of three-way interplay among gene expression, cancer cell invasion and anti-cancer compound sensitivity. BMC Medicine 2013;11:106.
Anti-HER2 Agents for Lung Cancer with a HER2 Mutation
Patients with NSCLC that harbor HER-2 overexpression (approximately 20%) do not respond to anti-HER2 agent, whereas those bearing a HER2 mutation (approximately 2–3%) respond to these drugs, according to this study. The largest retrospective study to date in Europe was conducted by Julien Mazières and colleagues to examine the effect of anti-HER2 agents in patients with HER2 mutation. They identified 65 (1.7%) NSCLC patients with a HER2 in-frame insertion in exon 20. All 65 tumors were adenocarcinomas and 50% were stage IV, mostly women (69%), and 52.3% were never-smokers. Sixteen patients, with stage IV lung cancer and prior conventional therapy, were treated with one or more anti-HER2 agents: afatinib, trastuzumab, lapatinib, and masatinib. Progressive disease was observed in four patients, disease stabilizations in seven patients, and partial responses in eleven patients. Progression-free survival for patients with anti-HER2 therapies was 5.1 months. Median survival was 89.6 and 22.9 months for early-stage and stage IV patients, respectively. Of note, a disease control rate (DCR) of 93% was observed in trastuzumab-based treatment and a DCR of 100% with afatinib. Prospective clinical trials will be required to confirm the benefit of the anti-HER2 drugs in patients with HER2 mutations. The findings illustrate the potential of anti-HER2 agents in treating patients of this population and underscore the importance of screening for HER2 mutations in these patients.
Mazieres J, Peters S, Lepage B, et al. Lung cancer that harbors a HER2 mutation: epidemiologic characteristics and therapeutic perspectives. J Clin Oncol April 22, 2013 [Epub ahead of print].
Dissecting the Genomic Profiles on Recurrence in Patients with Non–Small-Cell Lung Cancer (NSCLC) by Next-Generation Sequencing
Genomic changes in archived primary NSCLC tumors were compared with those in matched metastatic tumors for the genomic profiling of recurrence. Matched primary and metastatic tumors from 15 patients were evaluated by next-generation sequencing. About 20% of the genomic changes identified were known recurrent alterations, of which TP53 mutations were the most frequent recurrent alterations (12 patients). Ten patients have tumors bearing two or more recurrent alterations. A concordance rate of 94% was determined by comparative analysis of recurrent alterations between primary tumor and matched metastasis while non-recurrent alterations show a rate of 63%. The findings suggest the application of archived primary tumors in delineating genomic profiles on recurrence by detecting key recurrent alterations in matched metastatic tumors, which could be valuable to guide treatment on recurrence.
Vignot S, Frampton GM, Soria JC, et al. Next-generation sequencing reveals high concordance of recurrent somatic alterations between primary tumor and metastases from patients with non-small-cell lungcancer. J Clin Oncol April 29, 2013 [Epub ahead of print].
Pooled Analysis of KRAS Mutation and Subtype in Prognosis and Predicting Outcome of Early-Stage Resected Non–Small-Cell Lung Cancer
This study analyzed four trials of adjuvant chemotherapy (ACT) versus observation (OBS) to determine the prognostic role of KRAS mutation and predictive effects of KRAS codons on survival benefit in non-small-cell lung cancer (NSCLC). They explored the relationship between KRAS mutation, mutation subtype (codons 12, 13 and 14), and survival outcome using a multivariate Cox model in a population of 1543 patients. The results showed no significant prognostic value for KRAS mutations in NSCLC. Twenty four patients with codon 13 mutations appeared to have worse outcome from ACT but this needs to be validated despite being statistically significant. At this stage, KRAS mutation cannot be recommended as a marker in NSCLC patient selection for ACT. This was the first report of an association of KRAS mutation status with an almost 3-fold risk of developing second primary tumors in the OBS cohort. This implicates the potential of KRAS status in identifying high risk patients developing second primary tumors without ACT, whom might benefit from more intense follow-up.
Shepherd FA, Domerg C, Hainaut P, et al. Pooled analysis of the prognostic and predictive effects of KRAS mutation status and KRAS mutation subtype in early-stage resected non-small-cell lung cancer in four trials of adjuvant chemotherapy. J Clin Oncol April 29, 2013 [Epub ahead of print].
NEWS IN BRIEF:
ESTRO 2013: Mathematical Models Versus Doctors in Prediction of Outcomes and Responses to Treatment
Oberije and her colleagues from the MAASTRO Clinic, Maastricht University Medical Center, Maastricht, The Netherlands, compared mathematical prediction models with radiation oncologists at predicting the outcomes and responses of lung cancer patients to treatment. They used published mathematical models and incorporated previous patients’ information to generate a statistical formula for the prediction for future patients. The findings of the study, presented at the 2nd Forum of the European Society for Radiotherapy and Oncology (ESTRO), demonstrated that mathematical prediction models out-perform doctors in the prediction with a better predictive value than that of the doctors: 0.71 vs. 0.56 for 2-year survival, 0.76 vs. 0.59 for dyspnea and 0.72 vs. 0.52 for dysphagia, respectively. The negative predictive value was comparable between the models and the doctors. Further studies are needed to integrate prediction models into standard clinical care as well as to improve models with the latest discoveries in genetics and imaging, for instance.
Saving $63 Billion in Health Care Costs with Federal Cigarette Tax Increase
American Cancer Society Cancer Action Network (ACS CAN) estimated that the President’s proposal to increase the federal cigarette tax by 94 cents would save the country more than $63 billion in long-term health care spending, which includes more than $821 million from fewer smoking-related lung cancer cases over 10 years. The total federal cigarette tax would be $1.95 per pack under the proposed raise. The increase of the federal tax on other tobacco products would also take place under the budget proposal, which would improve public health and health care savings.
A Mathematical Model Tells the Future Pathways and Timescales of Metastatic Lung Cancer
A Markov chain/Monte Carlo (MCMC) mathematical model was developed by Newton and colleagues for cancer progression based on a large set of autopsy data to determine the multidirectional pathways and timescales associated with metastasis in lung cancer. The major findings, published in Cancer Research, show that metastatic lung cancer is a systemic and multidirectional stochastic process. The model demonstrated the main aspects that determine the future pathways and timescales of progression are combined characteristics of the primary site and the first metastatic site as a spreader or sponge. Lung is a significant self-seeder, in which circulating tumor cells from the primary tumor re-enter the primary site. Other significant mechanisms of multidirectionality in lung cancer progression are metastasis reseeding of the regional and distant lymph nodes and adrenal gland, primary lung reseeding through regional lymph nodes, and metastasis reseeding of the liver. The model could potentially affect surgical decisions in metastatic disease as different outcomes for patients could be predicted when the disease site is a spreader or sponge.