The Daily Sign-Out

Is Notch signaling involved in lung carcinogenesis?

This is the question asked by Joyce Osanyingbemi-Obidi and colleagues from Johns Hopkins in an interesting recent paper published in Molecular Cancer Research.  The cancer stem cell hypothesis proposes that there is a small proportion of tumor cells with stem cell-like characteristics, including self-renewal capability, as well as a much larger proportion of "differentiated" cells with limited proliferative capability.  While stem cell-like cells have been characterized in a variety of malignancies, such a population for NSCLC has not been reported.  I previously blogged on a murine model for evidence of lung stem/progenitor cells. The authors cite the emerging body of data (perhaps not surprisingly) suggesting that the same key signaling pathways in embryonic and fetal development (e.g. Notch, Wnt, and Hedgehog) are also involved in the regulation of cancer stem cells.  The intracellular domain of Notch receptors (NICD) is released upon ligand binding and translocates to the nucleus where it converts the CSL complex from a transciptional repressor into an activator.  However, the NICD-CSL complex requires mastermind-like (MAML) proteins to recruit and bind coactivators that drive expression of downstream targets that promote cell growth and proliferation.  Aberrant Notch signaling has been described in a variety of malignancies (albeit with both activating and inactivating Notch mutations) and, indeed, elevated expression of Notch family members and a key downstream target, "hairy and enhancer of split 1" (Hes1), have been reported in NSCLC.

Using NSCLC cell lines showing overexpression of Notch3 and Hes1, the authors demonstrated that pharmacologic inhibition of Notch was associated with suppression of Hes1 expression and inhibition of clonogenic growth.  The authors used a clever LSL-KRAS^G12D mouse model in which an oncogenic KRAS allele is silenced by a transcription stop site flanked by lox P sites; inhalation of an adenovirus vector containing the Cre recombinase promotes rearrangment of the lox sites and deletion of the transcriptional stop and selective overexpression of mutant KRAS in bronchial and alveolar cells.  The authors showed a transient induction of Hes1 expression by immunohistochemistry in airway epithelial cells 1 week postinfection with adenovirus and gradually diminshing Hes1-positive cells in tumors over time.  They then used an novel inducible double-negative MAML knock-in murine model to show that inhibition of Notch signaling in vivo was not sufficient to suppress carcinogenesis or tumor progression in the context of mutant KRAS-driven carcinogenesis.

So--these data suggest that Notch signaling may have a role in early lung adenocarcinogenesis through the transformation of a tumorigenic stem cell-like cell but is not likely required for adenocarcinogenesis.  The authors provide several notable caveats and qualifications of this conclusion but, I think the key consideration is that Notch signaling is considerably more complex that we are given to understand.  Notch signaling in development can maintain stem cell characteristics in certain cell populations or drive cell differentiation in other.  Likewise, Notch pathway activation in neoplasia appears to have tumor-promoting and tumor-suppressing effects.  Also, not all Notch family members appear to equally contribute to lung neoplasia.  Finally, individual Notch isoforms may have opposing effects on carcinogenesis.

This paper may not be of general interest, especially for the busy practicing pathologist, but is worthwhile to reflect upon and ruminate over.  Intriguing stuff.

Second opinions in pathology and medicine

Laura Landro, in her Wall Street Journal column, "The Informed Patient," has taken on the controversial subject of second opinions in medicine.  

Inevitably, the "attention-grabber" of such an article is an anecdotal patient initially "misdiagnosed" at one institution who goes to another institution, gets the right diagnosis (and the right treatment), and is healthy now.  I certainly don't want to minimize the issues here--not at all.  But pathologists really need to get out in front of this issue.  Hopefully, the College of American Pathologists will officially provide a response.

But more importantly, we must implement and communicate the quality assurance activities in place designed to minimize misdiagnosis in surgical pathology and cytopathology.  The author does quote Hardeep Singh, chief of the health policy and quality program at Houston's DeBakey VA Medical Center as stating that "a growing number of centers are requiring an internal second review of pathology reports to prevent misdiagnosis. If the second opinion differs markedly, a third opinion may be necessary to get a consensus on what course of treatment is best."  Who isn't already doing this?  A more nuanced and balanced approach to this article might have been to interview a (at least one!) practicing pathologist to get some context (check out the sidebar "Changing the Diagnosis"--where did this come from?) as well as specifics on the challenges of making diagnoses and what we do to confirm diagnoses--especially when cancer is being considered.  There has been a great deal of pathology literature in the last ten years identifying and addressing the problems and concerns raised in this article.  It sure would have been nice if this had been noted.

Despite what the public is led to (or wants to) believe, there is no infallible, 100% guaranteed diagnostic test out there for cancer--and (sorry!) never will be (with all due respect to the Human Genome Project). Sure, a wand would be nice.  But, in the meantime, there will be an irreducible level of uncertainty because of variation at every step of the diagnostic process from the patient herself through molecular testing on the tumor to individual response to treatment.  We cannot be nihilistic about this and must try to minimize variation at the diagnostic/interpretive level.  But we can also clearly communicate to ourselves, our colleagues, and the public the steps we are taking to minimize misdiagnoses.

Please share any innovative strategies you or your colleagues are taking for quality assurance in surgical pathology.

Lab uses Groupon to market testing

This is a really intriguing use of social media by a lab to market their services!  Affordable Labs in Vernon Hills, Illinois (a northern suburb of Chicago) is using Groupon to market a lab panel including cholesterol and liver function tests.  Affordable Labs' website states that they are not contracted with insurance companies and are pricing their services directly to consumers, including those with high-deductible plans or no insurance at all.  Disclaimer: I have no financial interest in any of this.  But this is an innovative way to market lab services that deserves wider consideration.

Researchers Map Mutations Lung Cancer in Never-Smokers

Researchers Map Potential Genetic Origins, Pathways of Lung Cancer in Never-Smokers.

Interesting preliminary data presented at AACR/IASLC Joint Conference on Molecular Origins of Lung Cancer currently underway in San Diego.  The team from the Translational Genomics Research Institute performed whole-genome sequencing (WGS) and whole-transcriptome sequencing (WTS) on tumors from 2 women who were never-smokers (one with early-stage disease and one with advanced stage disease) and compared these with a woman who was a smoker with early-stage disease.  One curious and unexpected finding is the lack of EGFR and KRAS mutations and ALK/EML4 translocation in the never-smokers' tumors.  They also found very few mutations in the tumor from the never smoker with early stage adenocarcinoma (so why did they develop a tumor?), but did find mutations in the usual "classic" tumor suppressor genes in the tumor from the never smoker with late stage disease.  They hypothesize that mutations in the classic tumor-suppressor genes might be a factor in late stage lung cancer, at least in never-smokers.  While this is a very limited "study" of only two never-smoker patients, the investigators are currently examining a larger group of about 30 never-smokers with adenocarcinoma and 60 clinical-matched smokers with adenocarcinoma.  This is a very interesting approach that is likely to yield some unexpected and provocative results.