The Daily Sign-Out

Techniques to increase lymph node harvest from gastrointestinal cancer specimens: a systematic review and meta-analysis - Abbassi-Ghadi - Histopathology - Wiley Online Library

Good news/bad news. Another article supporting the use of fat clearing and methylene blue staining to increase lymph node yields in GI dissection BUT these techniques do not increase the identification of positive LNs or lead to upstaging.




From Evernote:




Techniques to increase lymph node harvest from gastrointestinal cancer specimens: a systematic review and meta-analysis - Abbassi-Ghadi - Histopathology - Wiley Online Library
Clipped from: http://onlinelibrary.wiley.com/doi/10.1111/j.1365-2559.2012.04357.x/abstract;jsessionid=41CA356777E5BB95FA5D2064D29EC704.d04t02



Abstract:

Aim: This review aims to compare different histopathological techniques for lymph node harvest from ex vivo gastrointestinal cancer specimens and to examine their influence on: (i) lymph node yield; (ii) positive lymph node detection, and; (iii) cancer staging.

Method: Systematic review of the English language literature, comparing manual nodal dissection to other techniques for lymph node harvest, up to October 10th, 2011. Methodological quality of included studies was assessed.

Results: Twenty-seven studies, examining: fat clearing, methylene blue staining, fat stretching and use of a dedicated pathology assistant, were assessed. The methodological quality of the majority of included studies was poor. Meta-analysis showed that fat clearing and methylene blue staining increased mean lymph node yield, by 13 and 15 nodes respectively when compared to manual dissection. Of the 15 studies reporting positive lymph node count, two demonstrated a significant improvement for techniques other than manual dissection. Compared to manual dissection, other techniques were not shown to influence cancer staging.

Conclusion: This review has shown that fat clearing and methylene blue staining increases the mean lymph node yield from gastrointestinal cancer specimens. There is insufficient evidence to suggest that these techniques increase positive lymph node count or lead to upstaging.

© 2012 Blackwell Publishing Ltd

Implications of clonality in multifocal non small cell lung cancer

About 5% to 8% of NSCLC patients present with 2 or more anatomically separate lung tumors at diagnosis. The distinction between synchronous primary tumors and intrapulmonary metastases has important implications for staging and, thus, treatment strategies.

Arne Warth and colleagues from University Hospital Heidelberg published a single institution study in the June 1, 2012 European Respiratory Journal (abstract) analyzing the clonal relationship between multifocal NSCLC with indistinguishable morphology in a series of 78 patients.  Their analysis included LOH analysis of 14 polymorphic microsatellite markers and EGFR and KRAS mutation screening using single-strand conformation polymorphism, followed by direct sequencing of abnormally migrating bands. 

This study included 58 adenocarcinomas and 20 squamous cell carcinomas; 59 pT3, 18 pT4, and 1 pM1a. 97.4% of all specimens could be classified as clonal, likely clonal, or non-clonal; only 2 cases were non-informative.

Some results--

1. Tumor nodules of 22 patients (28%) showed divergent LOH status--indicating non-clonal origin. This is a surprising finding--remember this study only included multifocal tumors with indistinguishable morphology.
2. Activating EGFR mutations were found in both synchronous tumors in 4/58 ADC but none of the SQC.
3. KRAS mutations were found in 20 patients (17 ADC, 3 SQC) but in 6 patients the mutation was identified in only 1 of the tumors and in 1 patient there were divergent EGFR and KRAS mutations.
4. With combined allelotyping of LOH and mutational status of EGFR and KRAS, about 36% of cases (28/78 cases) showed evidence that they were not clonally related, including:
5.  14/18 ADC and 6/10 SQC located in the same lobe.
6. Kaplan-Meier curves illustrating the impact of clonality of multifocal NSCLC on overall survival did not identify a statistically significant difference but did show a tantalizing trend that patients with 2 clonally unrelated tumors have a better OS; notably, this difference did reach statistical significance by univariate analysis.

Practical "take-home" points.

There is no specific anatomical or morphological criteria that reliably indicate a clonal or nonclonal origin of multifocal NSCLC tumors. The authors reasonably recommend performing predictive molecular testing (EGFR and KRAS mutational analysis) or allelotyping separately in all cases with synchronous tumors to identify clonally related tumors. This study presents some intriguing data regarding the prognostic value of identifying nonclonalty, but the prognostic impact of clonality needs to be further analyzed in larger cohorts of NSCLC patients with multifocal synchronous tumors.

New directions for lung squamous cell carcinoma

For non-small cell lung cancer patients with localized or regional disease, treatment options vary little according to histology.  For metastatic NSCLC "targeted" therapies for patients with EGFR activating mutations or ALK translocations, pemetrexed, and the addition of bevacizumab to conventional chemotherapy have improved the treatment of metastatic NSCLC.  Unfortunately, all of these improvements apply almost exclusively to NSCLC adenocarcinoma histotype.

 For squamous cell carcinoma of the lung (SQC), the standard treatment remains cytotoxic chemotherapy, usually consisting of a platinum-containing doublet (carboplatin>US, cisplatin>Europe) with docetaxel or erlotinib used as second-line agents.

Similar to adenocarcinomas, squamous cell carcinomas are a heterogenous group of clinically, molecularly, and genetically different tumors.  Current work on SQCs is focused on identifying unique molecular patterns or driver mutations that may be therapeutic targets.  Kathryn Gold, Ignacio Wistuba and Edward Kim have published an excellent review of this recent work on SQC in Clinical Cancer Research 2012 18(11); 3002–7 that I recommend if you want a state-of-the-art update.

This review is more clinically oriented (for the pathologist) but provides connections between observed molecular abnormalities with drugs in various stages of development and current trials.  One interesting development is a phase II trial opening soon looking at the agent NCT01514864 for patients with squamous cell carcinoma with a mutation in discoidin domain receptor 2 kinase (DDR2).  I previously posted on a paper that identified DDR2 kinase mutations in lung squamous cell carcinoma about a year ago--so this is some exciting stuff! DDR2 kinase mutations occur in about 4% of SQC which doesn't seem like much until you input the denominator of the number of lung SQC diagnosed per year--this is analogous to EML-ALK translocations in lung ADC. 

Getting Ready for the ObamaCare Decision | e21 - Economic Policies for the 21st Century

Supreme Suspense: Getting Ready for the Big ObamaCare Decision | e21 - Economic Policies for the 21st Century.

This is a nicely written piece by James Capretta of the American Enterprise Institute outlining the different scenarios that might occur based on the Supreme Court decision on "ObamaCare." It is written from a conservative viewpoint but is, actually, reasonably balanced and makes some points that most everyone (right, left, and center) could agree with, at least in principle. Fundamentally, regardless of the decision, there are preexisting conditions with our health care system that must be addressed, including insurability for people with preexisting conditions, affordability of health care, providing care for the uninsured, the woeful inadequacy of the Medicaid program, the Medicare Leviathan, to name a few. 

Establishing clonality in multifocal nonsmall cell lung cancer

Up to 8% of nonsmall cell lung cancer (NSCLC) patients present with 2 or more anatomically separate lung tumors.  Distinguishing between synchronous primary tumors and intrapulmonary metastases is an important clinical question because this determination significantly changes tumor staging and subsequent therapeutic decisions.

Arne Warth et al. have published a single institution study from University Hospital Heidelberg in the current issue (1 June 2012) of European Respiratory Journal (abstract) that examines the clonal relationship in a series of patients with morphologically indistinguishable multifocal NSCLC.

The authors screened their archives for synchronous multifocal NSCLC in the same lobe (pT3), different ipsilateral lobes (pT4), and bilateral lobes (pM1a).  An important part of this study is that the authors only included those cases in which morphology between the tumors was indistinguishable using published criteria from the literature (Girard et al, Am J Surg Pathol 2009--actually a great reference).  They included 78 patients in this study.

The authors performed EGFR and KRAS mutation screening by using single-strand conformation polymorphism sensitive gel elctrophoresis and then subjected aberrently migrating bands to direct DNA Sanger sequencing.  The authors analyzed 14 polymorphic microsatellite markers for loss of heterozygosity. A clever part of the analysis of clonality in this study was the utilization of individual clonality scores for each tumor pair by multiplying the likelihood of the observed findings for each informative variable.  Thus, the clonality score defines the statistical error probability of a clonal or non-clonal relationship.

Results

1. The study included 58 adenocarcinomas (ADC) and 20 squamous cell carcinomas (SQC).  59 were pT3, 18 pT4, and 1 pM1a.  The mean follow up for all patients was 25.6 months. 97.4% of all specimens could be classified as clonal, likely clonal or nonclonal.
2. Tumors from 22 patients (28%) showed divergent LOH status indicative of non-clonal origin (14/58 ADC, 8/20 SQC).
3. Activating EGFR mutations were found in both synchronous tumors in 4/58 ADC and none of the SQC showed EGFR mutation.
4. KRAS mutations were found in 20 cases (17 ADC, 3 SQC), including in both synchronous tumors in 13 cases but in only 1 tumor in 6 cases. 1 patient had divergent EGFR and KRAS mutations in separate tumors.
5. Combined allelotyping of LOH and EGFR and KRAS mutational status showed that 36% of synchronous primary tumors were not clonally related.  Remember these were tumors that were determined to be morphologically indistinguishable.
6. Impact of clonality of multifocal NSCLC on overall survival: Kaplan-Meier curves show a trend that patients with 2 clonally unrelated tumors have a better outcome but this did not reach statistical significance by univariate analysis.

Conculsions

There is no objective, specific anatomical or morphological criteria that reliably indicate a clonal or non-clonal origin of multiple NSCLC tumors.  Therefore, the authors recommend performing predictive molecular testing (EGFR and KRAS) separately in all cases with synchronous tumor nodules or allelotyping to identifiy clonally related tumors.

This paper will influence the way I approach these situations.  First, I will specifically incorporate the Girard criteria to identify morphologically indistinguishable cases.  Second, I will send out separate tumor specimens for EGFR and KRAS mutation analysis before assigning my pathologic staging to the case.  A couple practical questions are not addressed by this study.  Given the remarkable morphological heterogeneity of lung ADC--and the high percentage of "mixed" subtype tumors, isn't it possible that morphological disparate tumors could still be clonally related?  The authors don't specifically address the issue of mixed subtype in relation to "morphologically indistinguishable" tumors--but this would be a major practical issue.  Should we simply assume that morphologically different-ish tumors are non-clonal or do the testing anyway?  A great complimentary study would test all synchronous tumors regardless of histology since this study showed that morphological criteria could not predict clonal origin even in a narrow series of synchronous tumors that looked alike in pairs.  An intriguing point of this study is the possible relationship between clonal relationship and OS.  The authors rightly conclude that this should be tested in larger cohorts of NSCLC patients to identify any statistically significance to the trend that they identified.  I am perplexed as to why clonally unrelated tumors should have better outcomes.  If any one has any thoughts, I'd love to hear them--please post a comment with your thoughts. 

Flagship Biosciences Acquires IHCtech

Flagship Biosciences released news today of their acquisition of IHCtech, a company founded by Patsy Ruegg in 2002 at the University of Colorado Health Sciences Center Pathology Department, that serves research investigators all over the U.S. with IHC services.  This is an interesting development for Flagship Biosciences and would seem to create some vertical dimension to their already formidable service line for quantitative IHC analysis.  Here's an excerpt from the press release (provided to me courtesy of Mr. Steven Potts, CEO and Chairman of Flagship):

“We have partnered with Flagship Biosciences on a number of new techniques and approaches for quantitation in immunohistochemistry,” said Patsy Ruegg, owner and founder of IHCtech. “Their approaches to whole slide analysis and commitment to quantitative pathology makes a perfect partner with IHCtech’s expertise in high quality histology and immunohistochemistry. We enthusiastically look to further innovation by evaluating all aspects of the tissue chain -- tissue procurement, fixation and processing, with the ability to measure with whole slide analysis how each of these steps contribute to variability in the overall process.”

 “While Flagship Biosciences has established strong partnerships with a number of highly respected IHC laboratories, both within the United States and internationally, this acquisition gives us the opportunity to internally evaluate tissue staining and implement new processes that better equip immunohistochemistry operations for use in quantitative pathology with whole slide imaging analysis,” said Dr. David Young, President of Flagship Biosciences. “Patsy Ruegg’s leadership in the area of high-quality research immunohistochemistry has been clear for many years. She has been a consistent expert in new IHC techniques like dual staining as well as a longtime leader in training of histology personnel.”

Usual disclaimer that I have no financial relationship etc with either company--I'm just an innocent bystander passing along information.  But I encourage you to visit Flagship's website.

 

Call It Cancer

A couple of recent editorials in The Oncologist have reacted to the recent NIH consensus panel on prostate cancer and this panel's statement that it may be "a mistake" to call low-grade, low-volume prostate cancer a cancer on the basis that it may cause unnecessary anxiety and provoke unnecessary action when the majority of men with this diagnosis are elderly and never experience morbidity or mortality with this disease.

Bruce Chabner and Matthem Smith (editorial, free registration required) opine that:

Although we agree that the connotation of lethality for some patients is unfortunate, the diagnosis of cancer depends on pathological findings (my emphasis), specifically, pathological appearance and invasive behavior. Indolent cancers occur as part of the spectrum of breast cancers, lymphoma, and other forms of malignancy, and those will continue to be called cancer.

Marshall Lichtman in another editorial draws comparison with the tortured terminology of myelodysplasic syndromes ("refractory anemia" "excess blasts") and the detrimental effects this mislabeling had on the understanding of this disease:

A profound error was made of the sort that Drs. Chabner and Smith are trying to help us avoid when the clonal cytopenias and oligoblastic myelogenous leukemia were designated myelodysplasia. By using such anachronisms (and euphemisms) as refractory anemia, and illogical terms as “excess blasts”—as if one can have an excess of malignant cells akin to excess fat cells—a neoplasia is thereby mislabeled a dysplasia, thus setting back thinking about the disorders by 50 years.

The confusion of students trying to understand why a disease with leukemic blast cells in blood and marrow is not called leukemia and of basic scientists thinking they were studying a preneoplastic condition was entirely unnecessary, if the correct language had been chosen.

This got me thinking about a similar and frequent situation I find myself with the term "carcinoid tumor"--a term that, at a minimum obscures and at worst is just downright wrong, in terms of clearly indicating the malignant potential of tumors showing typical characteristics.  So what does atypical carcinoid tumor mean?  As awkward as it is to explain metastatic carcinoid tumor in tumor board conferences, I can only imagine how it would be explaining to a patient, especially if that patient has done some research on the Internet prior to our visit.

What we call things does indeed matter and we, as pathologists, should make every effort to use terminology that reflects the behavior of what we are trying to describe.  It is some comfort that some of our oncology colleagues recognize our contribution to the diagnostic process but we must also be accountable for being complicit in using and perpetuating confusing and abstruse language.

Abstracts and Posters Accepted Now for Pathology Visions 2012

As a courtesy to the Digital Pathology Association, I am passing along this information regarding the 2012 Pathology Visions conference in Baltimore, Maryland from October 28-31, 2012.  Please consider submitting an abstract for oral presentation and/or poster presentation.  This is a great conference to attend!

  

ABSTRACT SUBMISSION DEADLINE FOR PATHOLOGY VISIONS 2012 APPROACHING

What’s the best way to stay in the loop on the most pressing and intriguing topics in the field of digital pathology? By attending and speaking at the Digital Pathology Association’s (DPA) annual meeting, Pathology Visions.

Pathology Visions will be held October 28 – 31, 2012 at the Hilton Baltimore. The Pathology Visions Conference will give you the opportunity to present and learn about real world, practical applications in the ever evolving field of digital pathology. In addition, Pathology Visions will give you direct access to industry leaders, the opportunity to earn CME credits, your choice of cutting-edge workshops with distinguished trade visionaries, the chance to see the latest product solutions, and much more!

The DPA is currently accepting oral presentation abstracts and poster presentation abstracts. The oral abstracts, submissions which provide researchers the opportunity to present for 30 minutes on a topic related to digital pathology, will be due on June 1, 2012. The poster presentation abstracts, allowing attendees the opportunity to display a poster for two days during Pathology Visions, will be due August 1, 2012.

Submitted abstracts will be reviewed by the Pathology Visions Program Committee and chosen presenters will be provided a conference registration pass (valued at $699), economy class round trip airfare, and lodging expenses for a maximum of 3 nights (October 28 - 31) at the host hotel.

Accepted presenters will be notified by June 15, 2012. To submit your abstract please now please click here. For more information regarding Pathology Visions please click here.

 

About the DPA

The Digital Pathology Association, located in Indianapolis, IN, was founded in 2009. Its mission is to facilitate education and awareness of digital pathology applications in healthcare and life sciences. Members will be encouraged to share best practices and promote the use of technology among colleagues in order to demonstrate efficiencies, awareness, and its ultimate benefits to patient care.

 

New proposal for histological grading in colorectal carcinoma

I find it surprising that histological grading for two of the most common adult malignancies--colon and lung--lacks any kind of consensus criteria for grading elements let alone clinical validation as a prognostic marker.  I have previously blogged on proposals for grading for lung adenocarcinoma but the February 2012 issue of American Journal of Surgical Pathology has an intriguing proposal for criteria for grading of colorectal cancer by Hideki Ueno et al. (abstract).

I am most familiar with Dr. Ueno's previous work on tumor budding in colon cancer (you should be too!) and this work builds on and expands on this work.  In contrast to tumor budding, defined as invasive single cells and/or clusters with <5 cells, this work describes and defines the significance of "poorly differentiated clusters" for grading.  Poorly differentiated clusters are defined by these authors as "cancer clusters in the stroma composed of >5 cancer cells and lacking a gland-like structure."  These clusters are identified at low power and then counted per 20X (1-mm diameter) field.  The color illustrations included in the text are particularly well-selected and beautifully reproduced.  Grading was also assessed independently by assessing quantitatively the area of the tumor involved by poorly differentiated clusters and by tumor budding.

Although this is a single institution study, it included 500 consecutive patients with stage 2 or stage 3 CRC who underwent curative resection and had follow-up on 97% (!) of these patients (483 patients) for a median of 68 months for the 378 surviving patients.

Three things stand out from the results.  First, there was excellent separation in 5-Year DFS using quantity of poorly differentiated clusters as the criterion for grading: grade 1--96%, grade 2--85%, grade 3--59%.  Second, tumor grading based on poorly differentiated clusters, T and N stages, and venous invasion were independent prognostic parameters related to survival on mutlivariate analysis.  FInally, the authors measured intra- and inter-observer variability in tumorg grading based on poorly differentiated clusters.

This is an interesting and status quo-challenging article that should provoke some good discussion amongst GI specialist and general pathologists.

The new IASLC/ATS/ERS Lung Adencarcinoma Classification is a Stage-Independent Predictor of Survival

 I blogged on this new classification over a year ago when it was first published, but now papers are rolling out showing its utility.  My take on this recent paper below is that it confirms the idea of histologically grading lung ADC according to the predominant different subtypes (lepidic predominant grade 1 "well-differentiated" versus solid predominant being grade 3 "poorly-differentiated".  Moreover, it asserts that it is sufficient to regard only the predominant histologic type with regard to survival--which is a relief from going through the tedious task of estimating percentages of different minor subtypes after complete histologic examination.  Get this article--good for discussion with surgeons, oncologists, tumor conference, etc. 

The Novel Histologic International Association for the Study of Lung Cancer/American Thoracic Society/European Respiratory Society Classification System of Lung Adenocarcinoma Is a Stage-Independent Predictor of Survival.