The Daily Sign-Out

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.

Possible false positive interpretation of Napsin A staining in evaluating lung non-small cell carcinomas

Dr. Nelson Ordonez (MD Anderson Cancer Center, Houston) published a brief and very helpful article in the March 2012 American Journal of Surgical Pathology (abstract) that addresses a very practical issue of using IHC staining for Napsin A in evaluating primary lung NSCLC for confirmation of adenocarcinoma type.

Napsin A is normally expressed in type II alveolar pneumocytes but may also be localized immunohistochemically in intraalveolar macrophages presumably from phagocytosis.  The expression of both Napsin A and pro-surfactant protein B (ProSP-B) are regulated by thyroid transcription factor-1 (TTF-1), which is one of the common IHC stains initially performed on NSCLC not diagnostic for adenocarcinoma on routine H&E.  Napsin A is localized to lamellar bodies with ProSP-B and is thought to be involved in post-translational processing of ProSP-B.

This paper must be understood in the context of new IASLC lung adenocarcinoma classification and the algorithm recommended to histological type NSCLC.  Napsin A has emerged as a useful adjunct IHC stain, such as, in those cases in which a tumor is negative or equivocal for initial screening markers for both adenocarcinoma (ADC) and squamous cell carcinoma (SQC).  Until recently, Napsin A was found to be highly specific for lung ADC.  But, as Dr. Ordonez points out, several recent publications have found Napsin A positivity in lung squamous cell carcinomas and, thus, cast doubt over the high degree of specificity of Napsin A for lung adenocarcinoma and utility of this marker in this context.

Dr. Ordonez studied Napsin A expression by IHC in 90 primary lung SQC, including whole section and biopsy specimens, and 64 non-pulmonary SQC.  None of the primary lung SQC or the non-pulmonary SQC showed Napsin A positive staining in the malignant cells.

The key observation is the identification of entrapped type 2 pneumocytes and/or intraalveolar macrophages in desmoplastic areas adjacent to the tumor and in small clusters within the tumor--which could be a source of false-positive interpretation in SQC.  Dr. Ordonez found 5 biopsies in which these entrapped cells were difficult to distinguish from malignant cells and CK 5/6 was used in 2 of these cases to rule out the possibility that these were Napsin A-positive malignant squamous cells.

This is a critical distinction and very welcome contribution to the lung cancer pathology literature.  The reader should keep point this in mind, especially when evaluating biopsy material.  The illustrations are excellent in demonstrating these points and should also be kept close at hand.  Congratulations to Dr. Nelson Ordonez for an excellent paper.

Biomarkers in COPD

Chronic obstructive lung disease (COPD) is defined functionally as airflow limitation that is not fully reversible.  It is a MAJOR cause of morbidity and mortality in the U.S. and worldwide.  COPD has two distinct major, but often coexisting, forms that represent different manifestations of COPD: emphysema and chronic bronchitis.  Emphysema can be defined morphologically, either by radiography or pathology, while chronic bronchitis is defined clinically based upon the duration of productive cough and exclusion of other causes.  Treatment is focused on preventing progression of disease and the development of acute exacerbations, which are a major cause of hospitalizations, morbidity and mortality.  Common to both major forms is small airways inflammation and an abnormal inflammatory response either in large airways (chronic bronchitis) or in alveoli (emphysema).

COPD is also associated with low-grade systemic inflammation and biomarkers that could predict increased risk for mortality from progressive disease or acute exacerbations might be greatly beneficial in providing more intesive surveillance and treatment of this group.  A team of researchers led by Tufts University School of Medicine screened blood samples from people in the ECLIPSE (Evaluation of COPD Longitudinally to Identify Predictive Surrogate Endpoints) to see if inflammatory biomarkers could predict disease outcome (abstract).  The study included 1843 patients who were followed for 3 years.  168 of the 1843 patients (9.1%) died.  Serum or plasma levels of fibrinogen, chemokine ligand 18, surfactant protein D, C-reative protein, Clara cell secretory protein-16, interleukin-6, IL-8, and tumor necrosis factor-alpha were measured at recruitment and at subsequent visits.

The addtion of a panel of biomarkers significantly improved predictive power and was able to better predict outcomes compared with current clinical variables, including age, BODE index, and hospitalization history. The addition of IL-6 provided the most improvement of discriminatory power, but this was improved further with the addition of all biomarkers measured.

Aberrant Wnt1 and beta-catenin expression in NSCLC

This month's April 2011 Journal of Thoracic Oncology has an interesting paper by Xu and colleagues (abstract) that examines Wnt1 expression in non-small cell lung cancer (NSCLC) in relation to downstream Wnt signaling molecules, including beta-catenin, and correlates different marker expression with traditional clinicopathological parameters.

This is an immunohistochemical study of a tissue microarray composed of 262 NSCLC resected specimens.    The authors define aberrant beta-catenin expression as: 1) decreased membranous pattern in less than 70% of tumor cells, 2) cytoplasmic pattern of expression, or 3) nuclear pattern of expression.  As expected, the majority of patients are stage 1 or 2 but nearly 36% of the study population are never-smokers--over 50% in the adenocarcinoma histology group (!).  The authors found that Wnt1 overexpression was significantly associated with never-smoker status and ADC histology; aberrant beta-catenin expression was associated with SQC histology.  While each Wnt1 overexpression and aberrant beta-catenin expression were independently associated with decreased OS regardless of TNM stage, patients that demonstrated both had worse OS than other combinations of Wnt1/beta-catenin expression.  Further, Wnt1/beta-catenin expression was also an independent factor adjusted by stage, pleural invasion, lymphatic invasion, ECOG PS or stage, age, gender, histology, and p53 expression.

A few comments about this paper.  One significant caveat to the findings is the over-representation of never smokers (35%) in this Asian study population, especially nearly 54% of never smokers in the ADC group, compared to a more typical American population in a community hospital setting (about 10%).  Overexpression of Wnt1 was significantly associated with never-smoking and as well as ADC histology, but, somewhat paradoxically, Wnt1 overexpression was associated with significantly shorter overall survival.  Unfortunately, the authors do not break this further down into Wnt1 overexpression and OS with respect to histological type--perhaps some SQCs showing Wnt1 overexpression and dismal OS are skewing the data here since one would expect never-smokers with ADC to have a better prognosis in general.  Another is the usual issues with using TMAs to represent complex heterogenous tumors.  The authors do not sufficiently describe from where they target obtaining cores for constructing the TMA--are they targeting the invasive margin, the central tumor, wherever?  what about mixed histology tumors?  These are critical methodological issues--especially when extrapolating the data concerning marker expression.  One admirable feature of this paper is the clear definition, supported by appropriate photomicrographs, of "aberrant" beta-catenin expression.  As I am currently working on projects that are similarly assessing beta-catenin expression in NSCLC, I found this very helpful in assessing this paper's findings.

Lymphomatoid granulomatosis: new insights

A great deal has been learned over the past few years about lymphomatous granulomatosis (LG), a rare pulmonary lymphoproliferative disease with a high mortality rate.  The eminent pulmonary pathologist, Anna-Louise Katzenstein, and co-authors, Erika Doxtader and Sonia Narendra, have published an excellent (free!) review e-article in American Journal of Surgical Pathology that I recommend highly.

The pathologic features of LG and post-transplant lymphoproliferative disorder (PTLD) are essentially identical, so the diagnosis of LG should not be made in organ transplant recipients.  Similarly, the pathologic features of LG are so similar to iatrogenic immunodeficiency-associated lymphoproliferative disorder that the latter diagnosis should be given consideration if the patient has a history of receiving methotrexate or other immunosuppresive drugs.

Fetal Lung Interstitial Tumor--report of proposed new entity

So many great articles at the end of the year, it's going to take me 'til Easter to catch up!

The December 2010 issue of American Journal of Surgical Pathology has a fascinating article from Dishop and (multiple) colleagues reporting 10 morphologically distinct tumors from multiple American institutions.  The patients' ages ranged from 0 days to 3 months and imaging in each case revealed a well-circumscribed lobar-based mass.  Lobectomy or wedge resection was performed in each case.

Each tumor was a well-circumscibed intraparenchymal mass with a solid-to-spongelike surface that appeared to be at least partially surrounded by a fibrous capsule.  The microscopic appearance was characterized by an abrupt transition from normal saccular-stage lung to more immature-appearing, variably-sized and -shaped spaces separated by cellular septa with variable width.  The lining epithelium is composed of regularly distributed non-ciliated flattened to cuboidal cells.  The interstitium is composed of a monotonous population of uniformly distributed round-to-ovoid mesenchymal cells.  None of the cases showed any of the malignant features associated with pleuropulmonary blastoma (PPB).

The interstitial cells showed diffuse positivity for vimentin and variable staining for smooth muscle actin and desmin; the epithelial cells were positive for cytokeratin (clone not specified) and EMA but TTF-1 positivity was reported in only 1 case.

Fetal lung interstitial tumor (FLIT) is the term proposed by the authors to describe a lobar-based tumor well-demarcated from surrounding normal lung that presents either prenatally or within the first 3 months of age, is composed of variably sized cysts composed of histologically benign epithelium and interstitial cells, and shows apparently benign clinical behavior.

The authors discuss the possible relationship between "FLIT," type I (cystic) PPB, and congenital peribronchial myofibroblastic tumor (CPMT) and posit a growth disorder in pulmonary interstitial cells as a common histogenesis for these rare, curious tumors.

I realize that this is a highly subspecialized report that is probably of little general interest.  BUT check this: tumors such as these point to the complex interactions between endodermally-derived epithelium and mesodermally-derived mesenchyme during lung development--an interaction which plays itself out again in more common lung diseases, like ILDs and lung cancer!

Uncommon Lung Adenocarcinoma Variants: a mini-review

There are five common types of adenocarcinoma (actually I would say four since "mixed" doesn't seem to me a specific 'type") recognized in the WHO classification of lung cancer.  But this classification also includes five rare but distinctive variants: fetal adenocarcinoma, mucin-producing adencarcinomas (mucinous "colloid" adenocarcinoma, mucinous cystadenocarcinoma, and signet ring adenocarcinoma), and clear cell adenocarcinoma.  

While rare to exceedingly rare as pure tumors, the variant patterns are important to be aware of because they occur much more commonly as a component of mixed type adenocarcinomas, are associated with particular clinicopathologic features, and often require the exclusion of a different primary site.  Another challenging variant is the nonmucinous lung adenocarcinoma which resembles colorectal adenocarcinoma but I'll review this variant in another post.  Hope you all find this of use.

Fetal adenocarcinoma

• Unencapsulated but well-circumscribed composed of closely packed branching tubules with little intervening stroma resembling fetal lung seen between 10 and 16 weeks' development (pseudoglandular stage).
• Cribriform arrangements can be prominent and cells arranged in cords and ribbons with peripheral palisading also described.
• The tubules are lined by "endometrioid" nonciliated pseudostratified columnar cells with frequent subnuclear or supranuclear vacuoles.
• Structures resembling squamous morules as in endometrioid adenocarcinoma are often present and the nuclei in these foci often appear optically clear (nuclear pseudoinclusions) due to biotin accumulation.
• Frequent mitoses.  Variable necrosis--focal to extensive.
• Low-grade and high-grade variants are recognized which differ significantly in clinical and histopathologic parameters:
• • Low-grade: mean age 34 years, about equal M:F, usually present at early stage, associated with a good prognosis; orderly arrangement of glands, focal necrosis (if present at all), small fairly uniform nuclei with inconspicuous nucleoli, morules common; slight to moderate loose fibromyxoid stroma, neuroendocrine differentiation almost always found, occasional AFP-positive cells, p53 protein overexpression uncommon, nuclear and cytoplasmic β-catenin expression
  • High-grade: mean age 64 years, male predominance, usually advanced stage at presentation, associated with poor prognosis (42% DOD@24 mos.); disorderly arrangement of glands, extensive areas of necrosis common, large pleomorphic nuclei with prominent nucleoli, morules absent, abundant desmoplastic stroma, always AFP-positive cells, p53 overexpression common, membranous β-catenin expression

Mucin-producing adenocarcinomas

Morphologically similar to non-pulmonary mucinous carcinomas from GI tract, pancreas, ovary, and breast.

Mucinous (colloid) adenocarcinoma

• Rare as a subtype but mixed tumors frequently show mucinous morphology
• Appears same as in other organs in which it is more frequently encountered: dissecting mucin pools containing islands and strips of mucinous epithelium with malignant cytologic features
• In the lung, mucinous columnar cells can focally line alveoli in a lepidic pattern.
• The neoplastic goblet cells express enteric markers (CDX2, MUC2 and CK20) with inconsistent, focal and weak CK7 and TTF-1 expression--thus, clinical correlation is needed to diagnose this as a pure lung primary.
• • Primary lung mucinous BAC does not show mucin pools invading stroma, is primarily composed of neoplastic mucinous cells lining alveolar walls, and does not show expression of CDX2 or MUC2.
• This neoplasm should be considered a low-grade carcinoma with more favorable survival rates compared to mucinous BAC or conventional mixed adenocarcinoma.

Mucinous cystadenocarcinoma

• Like the ovarian counterparts, this groups of tumors shows a histologic spectrum encompassing benign (mucinous cystadenoma), borderline (mucinous cystic tumor of borderline malignancy) and malignant (mucinous cystadenocarcinoma).
• The cytological and architectural criteria for differentiating these tumors are also similar to those in the ovary, except that invasion of surrounding lung is diagnosed as a mixed subtype of adenocarcinoma (for reasons I'm not quite clear but there it is).
• The mucinous epithelium in all of these cystic lesions differ from the mucinous adenocarcinoma type in that these express CK7 but inconsistently express CK20 and TTF-1.  These are quite rare and have been studied for MUC2 or CDX2.

Signet ring adenocarcinomas

• Rare as a pure subtype and uncommon (around 2%) as a component of mixed adenocarcinomas.
• Younger average age when predominant as a component compared to average age of lung cancer patients in general
• Unlike the other subtypes, this variant is defined by cytology--the classic signet-ring morphology.
• Intracytoplasmic material stains with PAS diastase, mucicarmine, and Alcian blue @ pH 2.7--indicating both neutral and acid mucins
• Over 95% of tumors express CK7 and over 80% express TTF-1 but are negative for CK20; also positive for CEA, MUC1 and MUC5AC--a staining profile that helps distinguish this as a primary lung tumor versus metastasis from breast, gastric and colon primaries.
• This is the most aggressive mucinous-type lung adenocarcinoma.
• Tumors with at least 50% signet-ring component have a poorer 5-year survival rate compared to the typical adenocarcinoma prognosis.

Clear cell adenocarcinoma

• Extremely rare as a pure subtype--but over a third of mixed adenocarcinomas show at least some focal clear cell change.
• Tumor cells form sheets with at least focal tubal formation (the latter required for diagnosis) and may form papillae.
• Intracytoplasmic material is PAS-positive, diastase-sensitive--indicative of glycogen.
• Pure tumors have not been well-studied by IHC--one reported tumor was CK7+/CK20-/TTF-1+.
• Distinguishing primary lung clear cell adenocarcinoma from a renal metastasis requires
• • clinical and radiologic correlation
  • identifying a mixed growth pattern and intracytoplasmic mucin
  • finding IHC markers specific for RCC
     

Mucinous bronchioloalveolar carcinoma of lung and ALK mutation

Last month I received a generous note from a patient who responded to a previous post on molecular markers in lung cancer from the USCAP 2010 meeting.  I made the statement that "every adenocarcinoma should be tested for EGFR mutation (with the exception of mucinous ACa/BAC which are always KRAS mutants)."  This patient's experience should cause us to not be so categorical about such statements.  Moreover, as we develop testing algorithms, we should consider an approach which will allow us to detect "exceptions to the rule."

The patient was diagnosed with mucinous BAC.  But not only was the tumor wild-type for KRAS mutation (i.e., no mutation was detected) but instead an ALK mutation identified in the tumor.  The patient subsequently received the drug crizotinib, an inhibitor of the fusion protein EML4/ALK, and has had a sustained response to treatment.  

This is the drug that generated so much attention at the recent ASCO 2010 meeting in Chicago.  In an international phase I study of advanced, heavily pretreated NSCLC patients selected for ALK protein expression, crizotinib produced a 57% response rate with a 6-month PFS predicted to be 72%.  This data was presented at the Plenary Session.  Moreover, it was noted that the majority of patients showed some kind of response (63% including objective and unconfirmed partial responses), the response often occurred dramatically within the first 2 weeks of treatment, and the responses were often durable.  In this study, 96% of responders had adenocarcinoma histology--primarily signet ring morphology.

This is surely a very exceptional case since mucinous ACa/BAC is a very distinctive histological subtype and has been also characterized at the molecular level as being associated with KRAS mutation.  The point is that histology is not a perfect predictor of potentially druggable molecular pathologies.  If targeted therapy is being contemplated in NSCLC, we should adopt a systematic approach to all NSCLC regardless of histology.  At this point, while we are identifying patients who may potentially benefit from targeted therapy, we are building a database that will allow a more full appreciation of the relationship between histological type and molecular pathology.

Thanks to the patient who took the time to respond to my blog and for helping to edify us all!

More on granulomatous lung disease: Pneumocystis jiroveci

Speaking of granulomatous lung disease (previous post), Hartel et al. have written a clinicopathologic review of the largest published series (20) of cases of pulmonary granulomatous inflammation associated with Pneumocystits jiroveci (formerly known as Prince).

I recall first seeing a case of Pneumocystis pulmonary granulomatous inflammation in an Illinois Registry of Anatomic Pathology case probably about 17-18 years ago--apparently, it made quite an impression!  Of course, I've only seen one similar case since them "in real life."

Several important findings from the paper:

1. The granulomas were well-formed in 80% of cases, necrotizing in 80% of cases, and multiple in 90% of cases.
2. Radiographic appearance is atypical for Pneumocystis pneumonia: Nodular infiltrates were present in 50% (8/16) of cases with studies and a solitary nodule was present in 19% of cases (3/16).
3. Open biopsy was required to make the diagnosis in 65% of cases and the diagnosis was made at autopsy in 25%.  Bronchoalveolar lavage--the main diagnostic tool in conventional Pneumocystis pneumonia--is of low yield in the diagnosis of granulomatous Pneumocystis pneumonia--as are bronchial washings, FNA, and transbronchial biopsy.  The fact that these tests are usually negative in this setting may explain why the diagnosis was made post-mortem a quarter of the time.
4. The authors discuss the role of host factors as most likely contributing to the atypical pathological manifestations of Pneumocystis since studies have repeatedly concluded that identical organisms are present in the various histologic reactions identified.
5. Good discussion of distinguishing between P. jiroveci and H. capsulatum on GMS stain: the lack of budding forms, shape and size (Pneumocystis--spherical, Histoplasma--ovoid)of yeast, and size of capsular dots (larger in Pneumocystis versus Histoplasma).

Review of Granulomatous Lung Disease: a few highlights

Drs. Mukhopadhyay and Gal have written an excellent review of granulomatous lung disease in the May 2010 issue of Archives of Pathology and Laboratory Medicine, which has a "Special Section" featuring non-neoplastic lung disease.

One of their points, that cannot be emphasized enough IMHO, is the crucial importance of trying to identify an organism since infection is the most common cause of pulmonary granulomas.  This is tedious work and requires a certain grim determination.  It often feels like trying to "find a needle in a haystack."  But I can also say that it is very rewarding to identify an organism so that proper therapy can be administered to the patient.  Re-emphasizing a point made by Ulbright and Katzenstein (1980), examination of additional blocks or sections showing necrosis with special stains is indicated when the histologic appearance suggests infection.  The authors also provide some helpful guidance in reporting situations when the stains are still negative.

Another very helpful item is in Table 2. which lists key diagnostic features of major non-infectious causes of granulomatous inflammation in the lung.  Keep this in your vade mecum.

Keep in mind the distribution of the granulomas.  I find this to be very helpful in generating a differential diagnosis, especially when special stains for organisms are negative.  The authors nicely reiterate this point.

Another important point is to be aware of PCR and other immunohistochemical and molecular methods for identifying organisms in FFPE tissue, especially for suspected mycobacterial infection.  It is worth considering ordering these additional tests when tissue was not obtained for culture or when mycobacterial infection is suspected clinically but routine ZN stains are negative.

The descriptions of the various major non-infectious entities associated with pulmonary granulomatous disease are as good as found in the major specialty textbooks and are a great reference.  I mean, 196 references, are you serious?!