The Daily Sign-Out

Epithelial-Mesenchymal Transition and Lung Cancer: Mechanism for Resistance to EGFR-TKIs?

Despite all of the encouraging clinical results with molecularly targeted therapies against non-small cell lung cancer, there are still important and significant questions that continue to vex.  The discovery of activating mutations in EGFR radically changed the treatment options for these patients, and, indeed, treatment of advanced stage patients with EGFR-mutated tumors with erlotinib (in the US) is now front-line therapy.  However, even patients who initially show dramatic responses invariably relapse due to the development of drug resistance.  Although such resistance can be explained in about 50% of these patients, what are the mechanisms for the other 50%?  Conversely, why do some, albeit rare, patients who are wild-type for EGFR respond to EGFR-TKIs?  Also, why don't conventional chemotherapy and molecularly targeted therapy produce a synergistic effect?  In fact, patients treated with combined therapy seem to actually do worse than if they get just one or the other--why?

Numerous research groups are trying to identify more effective biomarkers to identify predictive markers for response or resistance to EGFR-TKIs in order to optimize therapy for individual patients.  Our group at Rush University Medical Center has been focused on the role of epithelial-mesenchymal transition as a prognostic marker for early-stage NSCLC but also as a predictive marker for response to EGFR-TKIs.  So I read with great interest the article by Byers et al. recently published online in Clinical Cancer Research on December 20, 2012 (abstract here).

Some important notes:

 

• The EMT first principal component correlated better with E-cadherin protein level than did even the best CDH1 mRNA probe set.
• Mesenchymal cell lines were highly resistant to erlotinib as compared to epithelial cell lines, which agrees with previous work.  It is interesting, however, that while cell lines with EGFR-activating mutations were the most sensitive to erlotinib, in a subset of 40 cell lines WT for both EGFR and KRAS, the correlation between EMT signature and erlotinib resistance was still observed with significantly greater resistance in mesenchymal-like cell lines.  An important corollary is that the EMT signature was a better predictor of erlotinib response than mRNA probe sets for either CDH1 (E-cadherin) or VIM genes individually--a finding that agrees with the work we've done in our lab.
• EMT does not appear to be a general marker of resistance.  Indeed, mesenchymal cells were not more resistant to other targeted agents, such as sorafenib or to commonly used cytotoxic chemotherapies, including pemetrexed, docetaxel, paclitaxel, and platinum-doublets. Instead, a trend toward greater relative sensitivity was seen in mesenchymal cells as compared with epithelial for cisplatin. 
• An EMT signature may be a predictive marker in EGFR wild-type/KRAS wild-type patients with advanced stage NSCLC.  The authors analyzed the association between EMT signature and clinical outcome in patients enrolled in the BATTLE-1 trial but limited their analysis to patients whose tumors were WT for EGFR and KRAS.  Although the N is only 20, patients with disease control at 8 weeks (the primary study endpoint) showed a more epithelial-like signature as compared with those without disease control, although the difference is of borderline significance (P=0.05, by t test).  However, among the full group of 101 of 139 clinically evaluable patients (all treatment arms), expression of EMT signature genes was not prognostic of 8-week disease control or progression-free survival (PFS) in the overall group (all treatment arms).

This is a dense article but well-worth reading through in detail if you are interested in either EMT in NSCLC or mechanisms of EGFR-TKI drug resistance.  I offer a few further comments.

First, note that this is a study of NSCLC cell lines, so the findings have to be taken as hypothesis-generating for confirmation in clinical tumor specimens.  The limitations of studying cell lines ought to be well-known to you all.  But one point that has to be appreciated is that the phenomenon of EMT in vivo is context-dependent and critically depends on interaction with the tumor microenvironment--and this crucial fact is not and cannot be studied in this type of model.  Second, although it seems obvious to state this, histological subtype matters greatly with regard to EGFR mutations, EMT, and response to erlotinib.  Why do so many recently published studies still lump together non-small-cell lung cancer?  We recently focused exclusively on a group of patients with resected lung adenocarcinomas who received erlotinib upon disease progression in a study before the advent of EGFR mutation testing.  We identified two separate "epithelial" (high E-cadherin/low vimentin) and "mesenchymal" (low E-cad/high VIM) as well as intermediate/transitional tumors and found a trend toward worse DFS that did not quite reach statistical significance.  Unfortunately, our study was small (N=45) and I think that much larger numbers will be needed to show any importance difference.  In this study, out of 35 cell lines with adenocarcinoma histology, 29 (83%) had epithelial signatures, while 8 (23%) had mesenchymal signatures. Of 4 cell lines with squamous histology, there was an even distribution between epithelial and mesenchymal subsets.  Thus, the adenocarcinomas more commonly expressed an epithelial signature (P<0.0016, Chi-square test).  Finally, I have observed in our studies that EMT in vivo is a heterogenous process throughout a tumor.  Indeed, even if you focus on the invasive edge of a tumor, where one expects EMT to be occurring (as we did in a recently published abstract at AACR 2012), there is frequently significant variation in the expression and distribution of E-cadherin and vimentin.  What is the significance of this variation?  Can it be quantified and does the amount of variation in EMT have any prognostic or predictive significance?  Don't know--image analysis and quantitative immunohistochemical studies could possibly answer these questions. 

ETS2 mediates tumor suppressor function and MET inhibition in lung adenocarcinoma

There was a very intriguing paper published online by Clinical Cancer Research on May 9 last week that should be of interest for those involved in lung cancer research.  The authors are Mohamed Kabbout, Melinda Garcia, Junya Fujimoto, et al. from M.D. Ancerson Cancer Center in Houston and the title of the article is "ETS2 mediated tumor suppressive function and MET oncogene inhibition in
human non-small cell lung cancer." 

In brief, the authors find that there is significant downregulation of the ETS2 transcription factor in lung adenocarcinomas compared to normal lung tissues.  ETS2 expression is linked to various canonical cancer-associated pathways in lung cancer cells, in particular the HGF pathway.  Moreover, knockdown of ETS2 is associated with up-regulation of the ZEB1, SNAI2, RAB3B, SERPINB5 and SPDEF genes
with concomitant with elevated cell migration and invasion.  Last, low ETS2 protein is predictive of shorter time-to-recurrence in NSCLC and lung adenocarcinoma.  These data suggest that ETS2 functions as a tumor-suppressor in lung cancer cells by, at least in part, regulation of the migration- and invasion-promoting transcriptome.

More details. . .

Gene expression profiling showed significant decreased expression of ETS2 transcription factor transcripts in lung adenocarcinoma (ADC) compared to normal controls and quantitative RT-PCR confirmed decreased ETS2 expression.  Subset analysis should significantly decreased ETS2 expression in those ADCs from smokers compared to those from never-smokers.

They analyzed ETS2 expression by IHC in a microarray set of 201 NSCLC (135 ADC, 66 squamous cell carcinoma-SQC) tumors.  Low ETS2 expression was associated with shorter time-to-recurrence in stage I and all stages of patients.  Multivariate Cox proportional hazards analysis showed that, after adjusting for stage, low ETS2 expression was an independent predictor of shorter time-to-recurrence in NSCLC and ADC.

H441 human lung cancer cell lines transfected with small interfering RNA (siRNA) specific to ETS2 showed significantly reduced ETS2 transcript compared to controls.  ETS2 knockdown significantly increased cell growth, migration, and invasion.  Conversely, ETS2 overexpression in H1299 lung cancer cells showed significantly decreased cell migration and invasion.  Of interest, the magnitude of these changes following knockdown or overexpression of ETS2 was greater in measures of migration and invasion compared to cell growth--suggesting a tumor suppressor function for ETS2 at least in vitro.

Gene expression profiling of H441 lung cancer cell lines transfected with scrambled siRNA (control) versus ETS2-specific siRNA showed significant differential expression in 1,816 transcripts.  Functional analysis by Ingenuity Pathways Analysis (IPA) showed ETS2 knockdown modulated key oncogenic pathways, including HGF, integrin, tissue factor, semaphorin, and MAPK signaling.  Notably, HGF was the top modulated canonical pathway identified following knowdown, and further, an HGF-mediated gene-interaction network was among the among the top significant gene networks following topological arrangement of the genes found to be differentially expressed by IPA.

Western blot analysis and ELISA both showed increased phosphorylation MET following ETS2 knockdown.  Importantly, co-transfection of H441 lung cancer cells with MET-specific siRNA abrogated cell migration, invasion, and, to a lesser effect, cell growth mediated by knockdown of ETS2 expression alone.  Furthermore, co-transfection of cells with MET-specific siRNA significantly attenuated the induction of SNAI2, RAB3B, SPDEF and SERPINB5 mediated by knockdown of ETS2 alone as well as reduced expression of ZEB1. These data demonstrate that ETS2 suppresses key features of the lung cell malignant phenotype, at least in part, through inhibition of the MET oncogene. 

The authors also find that ETS2 suppresses EGF-mediated signaling in lung cancer cells.  HGF treatment increased ETS2 protein expression and phosphorylation of MET and ERK1/2 MAPK in H441 cells.  Conversely, expression of ETS2, phospho-MET, and phospho-ERK1/2 was attenuated by co-treatment of H441 cells with the MET TKI PHA-665752.  HGF treatment also significantly increased the expression of SNAI1, RAB3B, SPDEF, ZEB1, and SERPINB5 genes.  Moreover, the expression of these genes was significantly higher in HGF-treated cells transfected with ETS2-specific siRNA compared to similarly treated control cells.  Conversely, overexpression of ETS2 in H1299 cells decreased HGF-induced phosphorylated MET levels and decreased HGF-induced cell migration in the wound healing assay. 

Finally, NSCLC (p=0.01) or ADC (p=0.02) patients with relatively low ETS2 protein expression and higher phospho-MET membrane immunoreactivity showed shortest tiime-to-recurrence compared to other patient groups, specifically those with higher ETS2 and higher phospho-MET
expression.

NKX2-1/TTF-1 drives pulmonary-specific differentiation in lung adenocarcinoma

Snyder et al. recently published an intriguing report in Molecular Cell (21 March 2013) (featured free article) in which they convincingly demonstrate how tumors in which Nkx2-1 was deleted show striking different morphology from those in which Nkx2-1 is expressed.

Nkx2-1/TTF-1 is a highly conserved homeodomain-containing transcription factor that is expressed at the onset of lung and thyroid development and has been shown from murine knockout experiments to be required for branching morphogenesis in very early lung development. Nkx2-1 is expressed in about 80% of human lung adenocarcinomas (ADCs). Recent studies have shown that Nkx2-1-negative ADCs have a worse prognosis and have a higher histologic grade (more poorly differentiated) compared to Nkx2-1-positive tumors. These data suggest that Nkx2-1 may coordinate a lineage-specific differentiation program on lung adenocarcinomas that restrains their malignant potential. Further data have shown that the Nkx2-1 gene is amplified in 10%–15% of human lung ADCs, suggesting that it can also act as a lineage-survival oncogene in a subset of tumors. The authors' previous work using a mouse model for lung ADC showed that Nkx2-1 restrained the ability of Kras-driven lung tumors to evolve to a poorly differentiated, Hmga2-positive state. 

The authors cleverly designed a mouse model in which Cre recombinase can activate a conditional allele of Kras ^G12D and delete both alleles of Nkx2-1 from the lungs of mice.  Simultaneous activation of Kras ^G12D and deletion of Nkx2-1 yielded peripheral lung ADCs that showed a distinct glandular growth pattern of malignant cells with abundant MUC5A-positive mucin resembling human lung mucinous ADC. In contrast, control Nkx2-1-positive mice developed tumors showing a predominantly papillary non-mucinous pattern. They also showed in a different model that in established tumors, Nkx2-1 deletion induced a glandular pattern and mucin production similar to de novo tumors and promoted long-term tumor growth. In addition, this study demonstrated that Nkx2-1 deletion induced diffuse alveolar epithelial hyperplasia in the alveolar space.  The hyperplastic cells lacked expression of canonical Nkx2-1 targets and exhibited increasing mucinous accumulation and morphology over time.

The authors compared Kras-mutant/Nkx2-1-deleted tumors and Nkx2-1-positive tumors by mRNA expression profiling and found 1,828 differentially expressed genes, including 1,137 upregulated and 691 downregulated genes. MetaCore analysis for disease-specific gene collections and networks identified genes associated with gastric diseases that were enriched in  ‘‘stomach neoplasms’’ and ‘‘stomach diseases’’ as categories that were enriched in Kras-mutant/Nkx2-1-deleted tumors.  Finally, they determined whether human mucinous lung adenocarcinomas also exhibit evidence of gastric differentiationby evaluating the expression of two stomach-restricted proteins (GKN1 and CTSE) in 37 human lung adenocarcinomas. In 11 NKX2-1-negative mucinous human lung adenocarcinomas (n = 11), they found that GKN1 was expressed in 6 out of 11 cases and that CTSE was strongly and diffusely expressed in all 11 tumors. In contrast, NKX2-1-positive lung adenocarcinomas (n = 26) were entirely negative for GKN1. Most NKX2-1-positive lung tumors were either negative (n = 14) or only focally positive (n = 10) for CTSE. 

The DL

There is much more to unpack from this article than I can concisely summarize in this post. The fascinating point for me is that lung alveolar epithelial cells have a latent gastric differentiation program that becomes activated by default when Nkx2-1/TTF-1 is deleted and this is most dramatically evinced in pulmonary mucinous adenocarcinomas. These particular tumors are nettlesome diagnostically because of the question raised of whether the tumor is primary or metastatic.

Wonderful paper and lots to think about!

 

CAP releases guidelines for validating whole slide imaging

The College of American Pathologists has released guidelines for validating whole slide imaging. While stating the lack of FDA approval for using WSI for primary diagnosis, the College is acknowledging the very likely adoption of WSI for primary diagnosis in the near future. The CAP is offering numerous resources related to this topic and I highly urge you to start familiarizing yourself with the various aspects of slide imaging and image analysis.

The CAP will be hosting a webcast on Friday, May 17 to provide an overview of the guideline.

BRAF mutations in colorectal cancer: mini-review

There has been a recent flurry of papers and reports from ASCO-GI 2013 pushing more molecular testing in colorectal cancer (CRC) and its clinical significance with regard to prognostic and predictive factors, especially with MMR/MSI.

One area that caught my interest was BRAF mutations in colorectal cancer. Given the justifiable excitement around BRAF-mutated malignant melanoma responses with vemurafenib, I was curious about BRAF mutation in CRC. Here's my notes--

• 5% to 10% of CRCs have the BRAF mutation and the predominant mutation is similar to that seen in a large minority of melanomas, viz, the V600E mutation.
• Activating mutations in BRAF lead to constitutive activation of downstream signaling through the MAPK/MEK/ERK pathway.
• Recognition of this led to reflexive testing for BRAF if a CRC tumor was found to be KRAS-wildtype, since BRAF-mutant tumors would also be resistant to therapy with cetuximab similar to KRAS-mutant tumors.
• BRAF mutation appears to occur early in CRC progression and is associated with CpG island methylation.
• It leads to serrated adenoma-type polyps that have defective mismatch repair leading to a tumor that displays microsatellite instability.
• Important clinicopathologic associations: female gender, older age, right-sided location, high histologic grade, MSI-high--but not HNPCC.
• Advanced stage BRAF-mutated tumors have a poor prognosis: in a study of 524 patients with metastatic CRC (mCRC) by Tran et al. in Cancer (2011), OS for patients with BRAF-mutant tumors was 10.4 months versus 34.7 months in patients with BRAF-wild type tumors.
• Unlike the 81% response rate with vemurafinib in BRAF-mutated melanomas, the response rate in patients with mCRC is only about 5%.
•         ?more heterogenous distribution of BRAF mutation in CRC
•         similar mutations do not equal similar tumor biology (and thus) response to same targeted therapy
• In CRC (and in general?), mutations need to be viewed in context of other signaling pathways--
• Data from CRC cell lines and clinical tissue specimens show that BRAF mutation alone is insufficient for vemurafinib sensitivity: CRC can escape through activation of the PTEN/PI3K/AKT pathway and feedback activation of EGFR or other RTKs. This suggests multiple pathway inhibition as a possible therapeutic strategy in BRAF-mutant CRC.

Still lots of questions with this distinct subset of CRC--stay tuned.

 

Atomic force microscopy! EMT and idiopathic pulmonary fibrosis

Here is a very intesting article I recently found with some very cool technology:

Abstract:

Epithelial-mesenchymal transition (EMT) is closely implicated in the pathogenesis of idiopathic pulmonary fibrosis. Associated with this phenotypic transition is the acquisition of an elongated cell morphology and establishment of stress fibers. The extent to which these EMT-associated changes influence cellular mechanics is unclear. We assessed the biomechanical properties of alveolar epithelial cells (A549) following exposure to TGF-β1. Using atomic force microscopy, changes in cell stiffness and surface membrane features were determined. Stimulation with TGF-β1 gave rise to a significant increase in stiffness, which was augmented by a collagen I matrix. Additionally, TGF-β1-treated cells exhibited a rougher surface profile with notable protrusions. Simultaneous quantitative examination of the morphological attributes of stimulated cells using an image-based high-content analysis system revealed dramatic alterations in cell shape, F-actin content and distribution. Together, these investigations point to a strong correlation between the cytoskeletal-associated cellular architecture and the mechanical dynamics of alveolar epithelial cells undergoing EMT.

http://dx.doi.org/10.1016/j.nano.2011.06.021

I know this is a little out there for the pedestrian pathologist (ahem, like me) but I thought this was so cool--definitely worth at least scanning for the images of atomic force microscopy alone! But this article also provides data at the cellular level to explain some of the mechanical forces behind the changes that occur in IPF.

Imaging tumors in the future

Baohong Yuan and Joshua Rychak have written an interesting review in the February 2013 American Journal of Pathology that outlines emerging techniques based on ultrasound to obtain functional and molecular imaging of tumors.  Apart from being fascinating reading, this article underscores how the disciplines of radiology and pathology may merge at certain points sometime in the future.  Abstract:

Tumor functional and molecular imaging has significantly contributed to cancer preclinical research and clinical applications. Among typical imaging modalities, ultrasonic and optical techniques are two commonly used methods; both share several common features such as cost efficiency, absence of ionizing radiation, relatively inexpensive contrast agents, and comparable maximum-imaging depth. Ultrasonic and optical techniques are also complementary in imaging resolution, molecular sensitivity, and imaging space (vascular and extravascular). The marriage between ultrasonic and optical techniques takes advantages of both techniques. This review introduces tumor functional and molecular imaging using microbubble-based ultrasound and ultrasound-mediated optical imaging techniques. (Am J Pathol 2013, 182: 305e311; http://dx.doi.org/10.1016/j.ajpath.2012.07.036)

 

If you have a chance, I highly recommend checking this one out!

Pathologists to Blame for Hindering Acquisition of Cancer Tissue for Genomic Testing?

Medscape recently posted an article that discusses the need for standards for acquiring, preserving and storing cancer tissue, citing a recent Viewpoint article in JAMA by Drs. Laura Goetz, Kelly Bethel, and Eric Topol. If you have the chance, I highly recommend you pathologists read the full article as there is a valuable discussion of the need for appropriate specimens for next-generation cancer genomic testing.  Basically, we need to move beyond the idea that "diagnosis" just means having enough tissue or cells to make a morphological diagnosis (including immunohistochemical stains) to one where the "diagnosis" is extended to include molecular testing, even whole-genome sequencing.

I do admire Dr. Topol's principled stand to get patients to drive demand for obtaining appropriate tissue for genomic testing (a refreshing free market approach!).  However, I must take exception to comments made by Dr. Topol in the Medscape article where he states (bold is my emphasis):

Dr. Topol notes that there is much resistance from pathologists, in that it would be expensive to have fresh frozen specimens because it requires increased space, monitoring, and electricity for storage in freezers, there would be problems with reimbursement, and so on. "Basically, every excuse under the sun," he said. "But if it was one of them having a new diagnosis of cancer, they would probably like to have as much information as possible."

After having read the JAMA article, I have no idea why Dr. Topol takes such a gratuitous--and unnecessary swipe at pathologists.  His remarks and imputations are as ignorant as they are offensive.  While Dr. Topol correctly assesses the factors that are contributing to the slow development of processes, standards, and facilities to obtain, preserve, and maintain tumor specimens appropriate and optimal for molecular and genomic testing, resistance from pathologists is absolutely not one of the factors. In fact, the College of American Pathologists, our leading voice, has taken a lead role in developing such standards and educating pathologists in molecular and genomic oncology. See the CAP's resource page for molecular oncology.

Dr. Topol's assessment of the factors that are hindering the development of more widespread tissue procurement for genetic testing is solid--viz., trend of less tissue obtained for diagnosis, logistical issues with fresh frozen specimens, reimbursement--but his conclusion blaming pathologists is absolutely wrong.  The resistance that pathologists are putting forth has nothing to do with the concept of acquiring fresh frozen tissue (which we agree wholeheartedly with!) but with the facts that 1) we are responsible for establishing and maintaining an accredited facility for providing this service and 2) we are largely responsible for paying for this service which is neither reimbursed professionally or technically or adds to the "bottom line" of the institutions that we serve.  Since about 75% of cancer care is delivered at the community hospital level, perhaps Dr. Topol has some solutions for how these hospitals specifically are going to manage and pay for this--regardless if one of the pathologists (or even one of the administrators) has a new diagnosis of cancer.

New Device to Check Margins in Breast Cancer Surgery

Just a follow-up to yesterday's post regarding surgical margins in breast cancer. Will this device obviate inking the specimen or the debate about how much margin is "enough"?

Quoted from Medscape article by Zosia Chustecka:

The new device, MarginProbe (Dune Medical Devices Inc), consists of a single-use probe that analyzes breast tissue removed during a lumpectomy. Using radiofrequency spectroscopy, the probe analyzes "electromagnetic signatures" at the margins of the surgically removed breast tissue and compares them to an internal library of signatures taken from both healthy and cancerous tissue. Within the space of 5 minutes, the probe provides an assessment of the excised tissue, indicating whether or not it is cancerous.

 

Controversy: Surgical Margins in Breast Cancer

What is a clear margin in breast-conserving cancer surgery?

ASCO Connection (link) hosts a point-counterpoint discussion by two contributors regarding surgical margins in breast cancer that would be of general interest to surgical pathologists.  It is fairly brief (you can read in about 10 minutes) but also is well-referenced and would be a good resource to have.

The issues around surgical margin assessment and reporting is one that I find and have found vexing and exasperating. One of my professors while I was a resident refused to report margins on lumpectomies--much to the irritation of the oncologists, radiation oncologists, and surgeons--on the principle that the structure of fat and the extravasation of ink rendered any judgment of the margin status based upon the microscopic absence of (non-cauterized) recognizable tumor painted with ink meaningless. Although I couldn't understand his perspective at the time, I certainly do understand where he was coming from after years of experience and numerous clinical conversations about "close," "negative," and "positive" margins.  Yet, the "standard-of-care" dictates that we report margins, so we do.

Take a look at our colleagues' take on this subject. I found Dr. Silverstein's presentation of the split reduction excision particularly interesting and wonder if there are any readers who have had experience with this type of specimen and would be willing to "guest post" on this.

Video: Dr. Jack West discusses changing views on molecular testing in NSCLC

The Lungevity blog features a video of Dr. Jack West from Swedish Cancer Institute in Seattle discussing his evolving views regarding molecular testing for actionable driver mutations in non-small-cell lung cancer (NSCLC). Dr. West is a thought-leader in lung oncology and I found this a tidy and brief summary of current issues and reasonable argument for expansion of such testing.

This is important for pathologists as we are often tasked with triaging testing as the time of diagnosis, managing results reporting, and driving testing. My own views as a pathologist with a concentration on lung cancer have changed over the last two years based on clinical practice, research meetings, recent publications, and listening to clinicians' views. Initially, it seemed to me that reflex testing of all advanced stage (stage 2B and up) lung adenocarcinomas and excluding tumors with squamous histology; since we generally are unaware of the smoking history, I ignored smoking history as a criterion for testing. I okayed this with the oncologists and off we went. Dr. West succinctly contrasts the two views concerning molecular testing in NSCLC: essentially testing all patients with advanced stage NSCLC versus testing focused on "enriched" populations, e.g. adenocarcinoma histology, never/light smokers.

My own views began to change as a result of one memorable lung cancer research meeting at Rush we had when one of the thoracic surgeons related how a patient grilled him on why he received an expensive bill for all this molecular testing without any discussion with the surgeon or oncologist and, moreover, when he had already expressed his preference to not receive any further treatment following biopsy and mediastinoscopy. These aren't cheap tests and someone (not the pathologist) must explain these results to the patient. I also reviewed our experience with molecular testing and found several instances where testing was sent on early-stage patients and squamous histology tumors, which I didn't think was appropriate. Based on another round of discussions with the oncologists, I decided that we should consult with the oncologist before ordering molecular testing.

But as Dr. West says, "Some patient's cancer don't read the books." He relates several recent experiences that have caused him to revise his views: squamous tumors that show driver mutations, including one tumor with both EGFR activating mutation and ALK rearrangement. He advocates broader, more expansive molecular testing as well as consideration of testing recurrences and re-biopsy of clinically stable lesions.

The bottom line is that this is still an evolving area--we must be aware of new data coming out and be prepared to modify our views and practices.