Maria Cecilia Mengoli

Large cell carcinoma of the lung: clinically oriented classification integrating immunohistochemistry and molecular biology

This study aimed at challenging pulmonary large cell carcinoma (LLC) as tumor entity and defining different subgroups according to immunohistochemical and molecular features. Expression of markers specific for glandular (TTF-1, napsin A, cytokeratin 7), squamous cell (p40, p63, cytokeratins 5/6, desmocollin-3), and neuroendocrine (chromogranin, synaptophysin, CD56) differentiation was studied in 121 LCC across their entire histological spectrum also using direct sequencing for epidermal growth factor receptor (EGFR) and v-Ki-ras2 Kirsten rat sarcoma viral oncogene homolog (KRAS) mutations and FISH analysis for ALK gene translocation. Survival was not investigated. All 47 large cell neuroendocrine carcinomas demonstrated a true neuroendocrine cell lineage, whereas all 24 basaloid and both 2 lymphoepithelioma-like carcinomas showed squamous cell markers. Eighteen out of 22 clear cell carcinomas had glandular differentiation, with KRAS mutations being present in 39 % of cases, whereas squamous cell differentiation was present in four cases. Eighteen out of 20 large cell carcinomas, not otherwise specified, had glandular differentiation upon immunohistochemistry, with an exon 21 L858R EGFR mutation in one (5 %) tumor, an exon 2 KRAS mutation in eight (40 %) tumors, and an ALK translocation in one (5 %) tumor, whereas two tumors positive for CK7 and CK5/6 and negative for all other markers were considered adenocarcinoma. All six LCC of rhabdoid type expressed TTF-1 and/or CK7, three of which also harbored KRAS mutations. When positive and negative immunohistochemical staining for these markers was combined, three subsets of LCC emerged exhibiting glandular, squamous, and neuroendocrine differentiation. Molecular alterations were restricted to tumors classified as adenocarcinoma. Stratifying LCC into specific categories using immunohistochemistry and molecular analysis may significantly impact on the choice of therapy.

Activating c-KIT mutations in a subset of thymic carcinoma and response to different c-KIT inhibitors

BACKGROUND To analyze a multi-institutional series of type C thymic carcinomas (TCs) (including neuroendocrine tumors), focusing on the expression and mutations of c-KIT. MATERIALS AND METHODS Immunohistochemical expression of c-KIT/CD117, p63, CD5 and neuroendocrine markers, as well as mutational analysis of c-KIT exons 9, 11, 13, 14, 17 by direct sequencing of 48 cases of TCs. Immunohistochemical and molecular data were statistically crossed with clinicopathological features. RESULTS Overall, 29 tumors (60%) expressed CD117, 69% were positive for CD5 and 85% (41 cases) for p63. Neuroendocrine markers stained all six atypical carcinoids and five poorly-differentiated thymic squamous cell carcinomas. Overall, six CD117-positive cases (12.5%) showed c-KIT mutation. No mutation was detected in CD117-negative tumors and carcinoids. All the mutations were found in poorly-differentiated thymic squamous cell carcinomas expressing CD117, CD5, p63 and lacking neuroendocrine markers (6 of 12 cases with these features). Mutations involved exon 11 (four cases: V559A, L576P, Y553N, W557R), exon 9 (E490K) and exon 17 (D820E). CONCLUSIONS All TCs need an immunohistochemical screening with CD117, while c-KIT mutation analysis is mandatory only in CD117-positive cases, particularly when coexpressing CD5 and p63, lacking neuroendocrine differentiation. The finding of c-KIT mutation can predict efficacy with different c-KIT inhibitors.BACKGROUND To analyze a multi-institutional series of type C thymic carcinomas (TCs) (including neuroendocrine tumors), focusing on the expression and mutations of c-KIT. MATERIALS AND METHODS Immunohistochemical expression of c-KIT/CD117, p63, CD5 and neuroendocrine markers, as well as mutational analysis of c-KIT exons 9, 11, 13, 14, 17 by direct sequencing of 48 cases of TCs. Immunohistochemical and molecular data were statistically crossed with clinicopathological features. RESULTS Overall, 29 tumors (60%) expressed CD117, 69% were positive for CD5 and 85% (41 cases) for p63. Neuroendocrine markers stained all six atypical carcinoids and five poorly-differentiated thymic squamous cell carcinomas. Overall, six CD117-positive cases (12.5%) showed c-KIT mutation. No mutation was detected in CD117-negative tumors and carcinoids. All the mutations were found in poorly-differentiated thymic squamous cell carcinomas expressing CD117, CD5, p63 and lacking neuroendocrine markers (6 of 12 cases with these features). Mutations involved exon 11 (four cases: V559A, L576P, Y553N, W557R), exon 9 (E490K) and exon 17 (D820E). CONCLUSIONS All TCs need an immunohistochemical screening with CD117, while c-KIT mutation analysis is mandatory only in CD117-positive cases, particularly when coexpressing CD5 and p63, lacking neuroendocrine differentiation. The finding of c-KIT mutation can predict efficacy with different c-KIT inhibitors.

Concomitant EGFR mutation and ALK rearrangement in lung adenocarcinoma is more frequent than expected: report of a case and review of the literature with demonstration of genes alteration into the same tumor cells.

Oncogenic drivers in lung non-small-cell lung cancer (NSCLC) are considered mutually exclusive, but a review of the literature reveals that concomitant EGFR mutations and ALK rearrangement may occur in a subset of NSCLC. We report here a case of pulmonary adenocarcinoma with concomitant EGFR mutation in exon 21 (L858R) and ALK rearrangement in naive and relapsed tumors. Tumor cells seem to harbor both gene alterations and the patient had a long-lasting response both to EGFR inhibitor in second line and ALK inhibitor once tumor progressed. A speculative discussion on molecular mechanisms underlying this uncommon phenomenon and practical points about epidemiologic, clinicopathologic features and therapeutic options in this intriguing subset of double-positive tumor are reported.

K-RAS mutations indicating primary resistance to crizotinib in ALK-rearranged adenocarcinomas of the lung: Report of two cases and review of the literature

The paradigm of mutually exclusive alterations among oncogenic drivers in non-small-cell lung cancer (NSCLC) is challenged by the increasing evidence of detection of two or more driver alterations in the same tumor using highly-sensitive molecular assays. We report here two cases of ALK-rearranged adenocarcinomas harboring concomitant exon 2 K-RAS mutations (G13D and Q61H). The patients, a 49-year-old smoker man and a 59-year-old non-smoking woman, experienced a rapid disease progression and primary resistance to crizotinib. Search for similar cases in the literature reveals that concomitant K-RAS mutations and ALK rearrangement occur in a subset of NSCLC and seems to lead to resistance to crizotinib. Among 8 similar cases receiving crizotinib previously reported (4 in first line and 4 in second line), 1 had a partial response, 1 stable disease and 6 disease progression. One patient still had progression disease when switching to ceritinib. At the end, K-RAS mutations seem to represent a negative predictive marker in ALK-rearranged adenocarcinomas treated with ALK inhibitor.

Does Immunohistochemistry Represent a Robust Alternative Technique in Determining Drugable Predictive Gene Alterations in Non-Small Cell Lung Cancer?

Immunohistochemistry (IHC) is a widely-tested, low-cost and rapid ancillary technique available in all laboratories of pathology. This method is generally used for diagnostic purposes, but several studies have investigated the sensitivity and specificity of different immunohistochemical antibodies as a surrogate test in the determination of predictive biomarkers in non-small cell lung cancer (NSCLC), particularly for Epidermal Growth Factor Receptor (EGFR) gene mutations, Anaplastic Lymphoma Kinase (ALK) gene and ROS1 rearrangements. In this review, a critical examination of the works comparing the consistency of IHC expression and conventional molecular techniques to identify genetic alterations with predictive value in NSCLC is discussed. Summarizing, data on sensitivity and specificity of antibodies against ALK and ROS1 are very consistent and time has come to trust in IHC at least as a cost-effective screening tool to identify patients with rearranged tumors in clinical practice. On the other hand, mutant-specific antibodies against EGFR demonstrate a good specificity but a lowto- fair sensitivity, raising some cautions on their employment as robust predictive biomarkers. A brief comment on preliminary experiences with antibodies against BRAF, RET, HER2 and c-MET is also included.

MUC5AC, cytokeratin 20 and HER2 expression and K‐RAS mutations within mucinogenic growth in congenital pulmonary airway malformations

Rossi G, Gasser B, Sartori G, Migaldi M, Costantini M, Mengoli M C, Piccioli S, Cavazza A & Rivasi F 
(2012) Histopathology 60, 1133–1143

Localized pleuropulmonary crystal-storing histiocytosis: 5 cases of a rare histiocytic disorder with variable clinicoradiologic features

Crystal-storing histiocytosis (CSH) localized to the thoracic region is a rare occurrence, often secondary to lymphoproliferative or plasma cell diseases. About 10 case reports have been previously published, and 3 of these have no relationship with clonal hematologic disorders. We collected here the first series of 5 consecutive cases of CSH involving lungs (4 cases) and pleura (1 case). There were 3 women and 2 men with a mean age at diagnosis of 65 years. All cases had an underlying hematologic disorder (2 B-cell marginal-zone lymphomas, 2 monoclonal gammopathy of undetermined significance and 1 pulmonary plasmacytoma). Despite a common morphology characterized by a dense and irregular growth of large eosinophilic histiocytes with intracytoplasmic refractile crystals, 2 cases presented with cystic changes at gross and imaging examinations, calcified amyloid was found in 2 cases, and 1 case showed an interstitial lung disease with nonspecific interstitial pneumonia pattern. Histiocytes were immunoreactive for CD68 (clones PGM-1 and KP-1) but were not for CD1a and S100; the associated lymphoplasmacellular disorder had a clonal profile on molecular analysis with &kgr; light-chain restriction. Two cases were originally misdiagnosed as cystic fibrohistiocytic tumor and carcinoid tumor, thus confirming that CSH localized to this site may result in a diagnostic challenge with a broad spectrum of differential diagnoses. The presence of intracytoplasmic crystals and a plasma cell infiltrate around a histiocytic proliferation should alert the pathologist to consider CSH and to carefully investigate the presence of clonal hematologic disease.

Bronchial fibroepithelial polyp: a clinico-radiologic, bronchoscopic, histopathological and in-situ hybridisation study of 15 cases of a poorly recognised lesion.

Bronchial fibroepithelial polyp is an uncommon, poorly recognised lesion, lacking clear diagnostic criteria at histology, but possibly mimicking neoplastic growth on clinico‐radiologic and histopathological grounds. The aim of this study was to define the clinico‐pathological features, bronchoscopic appearance and treatment of bronchial fibroepithelial polyp.

Napsin A expression in pulmonary sclerosing haemangioma

codon 12) have also been associated with these lesions. However, a number of other cases describe the development of the secondary PNET in the original tumour bed, and our case illustrates an evolution of one tumour type into another with admixture of cell types in the third specimen. This argues towards a theory of radiation-induced transformation of original tumour into PNET. To the best of our knowledge, no consistent molecular alteration has been described in this latter group of cases. Regardless of pathogenesis, the outcome for these tumours is uniformly dismal, with reported mean survival ranging from 6 to 12 months. This may be due at least in part to the fact that many of the patients had already received their full dose of irradiation to the tumour area.

Pulmonary Large Cell Neuroendocrine Carcinoma: A True High-Grade Neuroendocrine Tumor Needing Prospective Therapeutic Data

To the Editor: We read with interest the article by Varlotto et al.1 on large cell neuroendocrine carcinoma (LCNEC) of the lung. Based on Surveillance Epidemiology and End Results (SEER) database, the authors collected 1211 LCNEC cases (324 without radiotherapy) and compared the clinicopathologic and survival features of LCNEC with those of 8295 (1120 without radiotherapy) large cell carcinoma (LCC) cases and 35,304 (355 without radiotherapy) small cell lung cancer (SCLC) cases. Briefly, the conclusion of this article is that patients with LCNEC should be classified and treated as having a LCC rather than a SCLC. We would here take the position that the results of this work do not entirely justify the final statements and may raise some confusion among oncologists. First, diagnosis of LCNEC is difficult requiring a careful histopathologic and immunohistochemical examination on surgical specimens. Even in the hands of expert pathologists, a major agreement is achieved only in 40 to 50% of cases.2 In a previous multi-institutional experience with LCNEC, we originally selected 122 cases.3 Fifteen cases were subsequently excluded because LCNEC was combined with other tumor types, and 24 cases were finally excluded because they lacked an unanimous diagnostic agreement among expert pathologists.3 For this reason, in our opinion, a centralized pathological review is necessary to confirm the diagnosis, a requisite that is lacking in the article by Varlotto et al.1 Second, molecular and biomarker analyses strongly support the fact that LCNEC is very similar to SCLC, from gene expression profiling to immunohistochemical expression of neuroendocrine (NE) markers, p53, Rb, bcl-2, Ki67/MIB-1, TTF-1, and several other biomarkers along with apoptosis and cell proliferation pathways.4,5 While LCNEC more often have a clinical and radiologic presentation resembling a non-small cell lung cancer (NSCLC)/LCC (i.e., as a peripheral mass), pooled analysis of therapy and survival data seem to confirm that LCNEC is an aggressive carcinoma with a poor prognosis in stage I (about 40% at 5 years) and receiving a significant benefit from pre-/postoperative chemotherapy with cisplatin-based regimens.6,7 Response rate to platinum-based chemotherapy may reach about 80%, a figure more similar to SCLC than LCC/ NSCLC.6 Curiously, all these data have not been considered in the discussion section in the article by Varlotto et al.1 The diagnosis of LCNEC requires immunohistochemical or electron microscopy confirmation of neuroendocrine differentiation. Data reported by Varlotto et al.1 on the increase from 8 to 21% of LCNEC diagnoses among LCC likely reflect a wide and possibly not appropriate use of immunostains for NE markers in poorly differentiated NSCLC (i.e., have pathologists used nonspecific antibodies such as NSE or PGP9.5 to identify NE differentiation?). In conclusion, to our eyes, there is a consistent body of evidence justifying the classification of LCNEC as a true high-grade NE carcinoma of the lung. Further supporting this fact, identical carcinomas have been recognized and recently accepted in the new World Health Organization classification of the digestive tract among NE tumors.8 On the other hand, prospective clinical data validating retrospective analyses concerning the most effective therapeutic approach for LCNEC are needed. By contrast, it is difficult to accept per se the conclusions of the work by Varlotto et al.1 based on the absence of expert pathologic review of the SEER cases. Finally, from a pathologic viewpoint, more stringent and objective criteria on quoting the NE differentiation (i.e., does a positive staining in more than 10% of tumor cells for only one NE marker represents a good cutoff in defining a LCNEC?) are clearly required.