[Anaplastic lymphoma kinase-positive adenocarcinoma of lung: a cytopathologic analysis].

OBJECTIVE: To study the cytomorphologic features of anaplastic lymphoma kinase (ALK)-rearranged pulmonary adenocarcinoma. METHODS: The morphologic features in 153 pulmonary adenocarcinoma cytology specimens encountered during the period from September, 2011 to April, 2015 in Shanghai Cancer Hospital were retrospectively reviewed. Fluorescence in-situ hybridization (FISH) and/or immunohistochemistry (Ventana D5F3) for ALK gene rearrangement were carried out. The samples studied included 34 pleural effusion specimens, 40 endobronchial ultrasound-guided transbronchial needle aspirates (EBUS-TBNA) and 79 fine needle aspirates of palpable masses on body surface. RESULTS: Thirty-nine cases (25.5%) of ALK-rearranged samples were identified by FISH and/or immunohistochemistry, including 3 cases diagnosed by FISH and 36 cases by both technologies. The median age of the ALK-positive group was 50 years, significantly younger than that of the ALK-negative group (60 years old, P = 0.002). Only 4 of the ALK-positive patients were smokers, which was significantly less than that of the ALK-negative group (P < 0.01). In ALK-positive group, 3 cases showed cribriform pattern with prominent nucleoli, 3 cases showed cribriform pattern with mucin-rich cells and 8 cases showed extracellular mucus with mucin-rich cells. The above cytomorphologic patterns were significantly less common in ALK-negative tumors (P < 0.01). CONCLUSIONS: ALK-rearranged lung adenocarcinoma is associated with certain distinctive morphologic patterns, including cribriform architecture, presence of prominent nucleoli, mucin-rich cells and extracellular mucus, which can be observed in cytology specimens (including conventional smears and cell block sections). These findings, when combined with clinical features, may give clues to detection of ALK-positive cases.

High sensitivity of touch imprint cytology in young patients with invasive lobular carcinoma of the breast.

Invasive lobular carcinoma (ILC) of the breast is a rare type of breast cancer. Metastatic ILC is difficult to identify in sentinel lymph nodes (SLNs) because of its low-grade cytomorphology and its tendency to resemble lymphocytes. Touch imprint cytology (TIC) is a rapid method for evaluating SLNs intraoperatively. We sought to evaluate this technique in the identification of SLN metastases in ILC. A total of 230 SLNs from 88 patients diagnosed with ILC were examined with TIC, and the results were compared with their histological diagnosis. We first confirmed some typical cytological features of SLN metastases in ILC such as a thick eosinophilic background and hyperplasia of small cell lacking adhesion. Further analysis showed that sensitivity of TIC was 60.5 % on per-node basis and 58.3 % on per-patient basis; specificity of TIC was 100 % on per-node basis and 100 % on per-patient basis. Interestingly, when we divided patients according to their ages, the sensitivity of TIC in patients younger than 50 years old greatly increased to 90.9 % on per-node basis and 85.7 % on per-patient basis, whereas the specificity remained 100 % on both per-node basis and per-patient basis. However, different tumor sizes did not clearly change the sensitivity of TIC compared to the overall sensitivity. Patients with tumor size bigger than 1 cm revealed a TIC sensitivity of 59.4 % on per-node basis and 55.0 % on per-patient basis. These results suggest that TIC can be used as a reliable method of detecting SLN metastasis only in young patients with ILC.

[Value of cytopathology in endobronchial ultrasound-guided transbronchial needle aspiration].

OBJECTIVE: To evaluate the role of cytopathology in endobronchial ultrasound-guided transbronchial needle aspiration (EBUS-TBNA) for lung tumor diagnosis and staging. METHODS: Two-hundred consecutive cases of lung tumor with EBUS-TBNA performed during the period from April, 2009 to September, 2010 in Shanghai Cancer Hospital were retrospectively reviewed. The cytologic diagnoses were categorized as non-diagnostic, negative, suspicious and malignant. When available, cell block preparation and immunohistochemistry were performed. On the 22 positive cases diagnosed by on-site evaluation, epidermal growth factor receptor (EGFR) mutation study was carried out. RESULTS: In the 200 cases of cytology specimens, 122 cases (69.3%) were diagnosed as malignant, 42 cases (23.9%) as benign and 12 cases (6.8%) as suspicious for malignancy. The non-diagnostic rate was 12.0% (24/200). Amongst the 200 cases studied, 140 cases (70.0%) had histologic correlation available (via core biopsy, mediastinoscopic biopsy or surgical excision). The sensitivity and specificity of EBUS-TBNA cytologic diagnoses were 94.4% and 100%, when using histopathologic findings and clinical follow-up data as gold standard. The cell block preparation and immunohistochemistry were useful in subtyping and diagnosis of extrathoracic malignancy. EGFR mutations were detected in 8 cytology samples (36.4%). CONCLUSIONS: EBUS-TBNA is a sensitive and specific tool for diagnosis and staging of lung cancer. The cytology samples can be used for further ancillary investigations including cell block preparation, immunohistochemistry and molecular studies.

[Evaluation of intraoperative touch imprint cytology of sentinel lymph node for breast cancer].

OBJECTIVE: To evaluate the intraoperative touch imprint cytology as an diagnostic method of sentinel lymph node for breast cancer patient. METHODS: Sentinel lymph node biopsy was performed in 105 selected early breast cancer patients, and sentinel lymph node was identified in 101 (96.19%) of these patients. Axillary lymph node dissection was also performed in almost all the patients. All the sentinel lymph nodes were cut into 2-3 mm pieces along the long axis. Touch imprint was made of each piece of the sentinel lymph node, then air-dried, and finally stained with H&E. Intraoperative touch imprint cytology results were compared with the final paraffin H&E pathology. All sentinel nodes were cut into 4 microm sections every 100-microm interval, and the series sections were stained with H&E. RESULTS: 202 sentinel lymph nodes were identified in 101 breast cancer patients. The sensitivity, specificity, accuracy, positive predictive value, negative predictive value of intraoperative imprint cytology for 202 sentinel nodes was 92.1%, 98.8%, 97.5%, 94.6% and 98.2%, respectively; which was 89.3%, 98.6%, 96.0%, 96.2% and 96.0%, respectively in the 101 patients with identified sentinel node. Compared with the series sections, the sensitivity, specificity, accuracy, positive predictive value, negative predictive value of intraoperative imprint cytology for sentinel nodes was 83.3%, 98.8%, 95.5%, 94.6% and 95.8%, respectively; and it was 81.3%, 100.0%, 94.1%, 100.0% and 92.0%, respectively in 101 patients with identified sentinel node. CONCLUSION: Touch imprint cytology is a simple, effective and rapid method for intraoperative pathological evaluation of sentinel lymph node for breast cancer patient, which has a high concordance with the paraffin results, and can provide accurate and rapid diagnosis information for the surgeon during operation.

[Evaluation of the results of fine-needle aspiration liver biopsies and the complications in 2528 cases].

OBJECTIVE: To study the value of fine-needle aspiration biopsy (FNAB) in diagnosing primary liver cancer (PLC) and its major complications. METHODS: From June 1, 1985 to May 31, 2005, 2528 patients who were presented with suspected PLC underwent ultrasound-guided FNAB in the Cancer Hospital of Fudan University. The results were retrospectively reviewed. RESULTS: Of those 2528 cases, there was malignancy in 2061 patients (81.53%), of which 1704 were diagnosed as primary liver neoplasms; 41 were diagnosed as metastatic carcinoma, and 316 were not further classified as primary or metastatic. No malignancy was found in 431 cases (17.05%). In 36 cases (1.42%), suspicious malignancy or anaplasia was suggested. Follow-up results showed that all the 2061 positive cases were verified to be malignant and there were no false positive cases. 163 of the 431 negative cases were verified to be malignant in the follow-ups, of which 136 cases were PLC; 28 of the 36 suspicious malignancy or anaplasia were proven to be malignant (all were PLC). The sensitivity, specificity, positive predictive value, negative predictive value and overall accuracy of our FNAB for diagnosing liver malignancy were 91.52%, 100.00%, 100.00%, 59.10% and 92.44%, respectively, and 81.01% cases were diagnosed by FNAB in all the 2096 cases with PLC. Cytological examinations of the smears obtained by FNAB correctly distinguished primary and secondary malignancy in 77.49% of the patients. After FNAB, 11 patients (0.44%) had intraperitoneal hemorrhages and 5 cases (0.20%) had needle tract implantation metastases. CONCLUSIONS: FNAB is important and effective for determining the malignancy potential of liver tumors, especially for PLC. Complications related to FNA were rather rare, therefore, this technique may be easily applied to clinical practice.