cytotechnology – Lablogatory

Three Cheers for a Three-Cell Population

Back in the olden days – wait, can I use that line yet? I graduated from my cytology program ten years ago, which feels like two decades ago or just yesterday. Depends on the day. Anyway, there are little nuggets from my training that stick with me whenever I’m screening a case. This case revolves around the concept of a two-cell population in body fluids, namely pleural, peritoneal, and pericardial. We were taught that if you see a two-cell population, the case is malignant. I remember asking in school, “But what if the diagnosis is mesothelioma?” “Okay,” my professor said, “that applies to any malignancy except mesothelioma. The two-cell population indicates a population of benign mesothelial cells and a second population of malignant cells.” Noted! I live by this rule, and every now and then, I have a case of a peritoneal fluid where the slides are covered in “wall-to-wall” adenocarcinoma, and I can’t find a benign mesothelial cell for the life of me. Obviously, we let years of experience with morphology take the reins there, but there are still those cases where the native mesothelial cells are so reactive in appearance that we start hunting for a two-cell population. This case was a perfect blend of wall-to-wall tumor and more-than-reactive mesothelials.

When a patient is being worked up for the first time or the nth time, we recognize their names and recall their clinical history and previous morphologies. Quite often we have patients with genetic predispositions to cancer who present with one type of cancer and are treated accordingly, and during surveillance imaging, come back to us with a second primary cancer. In this case, we received a right-sided pleural fluid on a 73-year-old woman with a history of ER+/PR+/HER2- breast cancer. After a history of incomplete medical follow-up, the patient presented to a local emergency room with pleuritic pain, upon which a CT Scan identified a lung mass, breast mass, and multiple liver and upper thoracic bone lesions. The assumption, given the clinical history, was metastatic breast cancer, and the clinician submitted the pleural fluid for cytology to repeat ER/PR/HER2 testing to determine if there is a better treatment strategy than her current aromatase inhibitor, which appeared to be failing.

Upon screening the cytopreparations, we observed very obvious malignant cells (Image 1). It was evident that we had a two-cell population. One of the cell populations was characterized by cohesive clusters of relatively uniform tumor cells (Image 2 & 4) and the second cell population was less cohesive with larger pleomorphic cells (Image 3 & 4). Hmm, something doesn’t make sense here. We already have a group of tumor; the other cells are supposed to be benign mesothelials cells. Is there a mixed morphology going on? Like a combined small cell carcinoma and non-small cell carcinoma that we might see in lung cancer? It’s not small cell though. These two populations are clearly two adenocarcinomas. That still can’t be. There has to be a population of mesothelials. No, no… the benign mesothelials are scattered throughout the background. It looks like a double-malignant three-cell population!

Images 1-4: Pleural Fluid, Right: 1. DQ-stained cytospin; 2-3: Pap-stained smear; 4: H&E Cell Block section (600X).

We submitted the case to the cytopathologist on service for the day. When they asked what we had, we replied, “All the adenocarcinoma,” and noted that we definitely need ER/PR/HER2 as clinically and now morphologically, it appears that the patient’s breast cancer is involving the pleural fluid. The cytopathologist agreed that there was something to the case. Metastatic breast cancer likes to appear as cannon balls in fluid (Image 2), the second cell population is too unlike the first. The cytopathologist performed immunocytochemical stains on paraffin sections of the cell block with adequate controls. The first type of cells shows positive staining for CK7, CK19, GATA3 (Image 5), and ER, and negative staining for CK20 (Image 6), CDX-2 (Image 7), BRST2, and PR. The second type of tumor cells show positive staining for CK7, CK20 (Image 6), CK19, and CDX-2 (Image 7), and negative staining for GATA3 (Image 5), ER, PR, and BRST2. The former group of cells has a morphology and immunoprofile consistent with breast origin, and the latter group of cells are morphologically and immunophenotypically suggestive of a pancreaticobiliary or gastrointestinal tract origin.

Image 5-7: Pleural Fluid, Right: 5: GATA3; 6: CK20; 7: CDX-2.

Final Diagnosis: Positive for malignancy. Adenocarcinoma, consistent with metastatic breast and metastatic pancreaticobiliary or gastrointestinal tract origin.

Interestingly, the patient had a paracentesis and liver biopsy the following week, and only the pancreaticobiliary or GI tract origin immunoprofile was positive in the peritoneal fluid and FNA. The patient’s breast cancer seemed to remain above the diaphragm.

-Taryn Waraksa-Deutsch, DHSc, SCT(ASCP)^CM, CMIAC, LSSGB, is the Cytopathology Supervisor at Fox Chase Cancer Center, in Philadelphia, Pennsylvania. She earned her master’s degree from Thomas Jefferson University in 2014 and completed her Doctorate of Health Science from Bay Path University in 2023. Her research interests include change management and continuous improvement methodologies in laboratory medicine. She is an ASCP board-certified Specialist in Cytology with an additional certification by the International Academy of Cytology (IAC). She is also a 2020 ASCP 40 Under Forty Honoree. Outside of her work, Taryn is a certified Divemaster. Scuba diving in freshwater caverns is her favorite way to rest her eyes from the microscope.

Weirdoma

A 36 year old male was referred to gastroenterology after presenting to the emergency room for hematemesis and severe fatigue. He was pale and tachycardic, and the CBC showed a hemoglobin of 4.5. An esophagastroduodenoscopy (EGD) at the time demonstrated erosive esophagitis with a visible vessel and was treated with a PPI. A repeat scope the following month no longer demonstrated a vessel but identified a 10 cm ulcerative gastric cardia mass at the GE junction. Forcep biopsies showed a gastric ulcer with granulation tissue, and the stains performed yielded results that were not consistent with carcinoma or lymphoma; however, the biopsy material was limited, and the patient was referred to our GI clinic for further workup.

The interventional gastroenterologist requested cytology be present for the patient’s endoscopic ultrasound to ensure an adequate specimen was obtained for a definitive diagnosis. During the rapid onsite evaluation (ROSE), we determined the Diff-Quik smear was adequate, and the pathologist could confidently suggest that tumor cells were present. We collected additional FNA passes in our cell block tube to run ancillary studies.

The following morning, all we could make of the case was that it was a poorly-differentiated malignant neoplasm with spindle and epithelioid features. The cytoplasm was minimal and fairly wispy while the nuclei were hypochromatic and fragile with nuclear grooves and nucleoli. On the Diff-Quik smears, the cytoplasm looked blue, which pointed us in the direction of possibly lymphoma or neuroendocrine, but the clustering made me favor neuroendocrine. With the pap-stained smears, were torn between carcinoma and a neuroendocrine tumor, maybe even an epithelioid GIST, albeit an odd location. And of course, there’s always the differential of melanoma, the great mimicker. Off to IHC we go!

Images 1-4: Stomach, GE Junction, EUS-FNA. 1-2: DQ-stained smear; 3-4: Pap-stained smear.

When our immunostains were delivered later that afternoon, our pathologist came up to me and said, “I got it! I know what it is!” Ecstatic, I replied, “What is it? Lymphoma? Carcinoma? GIST? MELANOMA?” “No, it’s a WEIRDOMA! Nothing is staining positive. No epithelial markers, no definitive lymphoid markers… nothing. It’s a weird case. I have to run additional stains.”

Images 5-7: Stomach, GE Junction, EUS-FNA. 5: H&E Cell Block section (600X); 6: Vimentin-positive; 7: CD56-positive

Back to the drawing board and 20 additional recut sections later, more immunostains were ordered and a mixed profile led us down a more confusing path. The tumor cells show positive staining for vimentin, CD56, and CD10 (focal), and negative staining for AE1/AE3, Cam5.2, CK7, desmin, SMA, HHF35, CD34, CD117, DOG-1, S100, SOX-10, synaptophysin, SALL4, CD45, CD68, and CD21. Proliferative index by Ki-67 was approximately 35%. The morphology and immunoprofile of the tumor were highly unusual, suggesting a mesenchymal neoplasm, possibly a sarcoma.

The concurrent forcep biopsies demonstrated rare atypical cells that were difficult to classify due to the limited number of cells, the non-specific morphology, and the following non-specific immunophenotype: positive staining for CD99, partial positive staining for D2-40 and NSE, and focal or weak positive staining for Cam5.2, while negative for AE1/AE3, CK7, EMA, S100, CD31, desmin, PAX8, BCL2, and myogen. The biopsy tissue consisted of predominantly ulcerative tissue and a fragment of squamous mucosa with a lamina propria infiltrate of atypical cells with spindle and epithelioid morphology.

Due to the FNA cell block consisting of 80% tumor compared to the limited forcep biopsy tissue, we sent FFPE cell block sections for RNA fusion studies to help us further classify the tumor. An EWSR1::ERG (in-frame) rsa(22;21)(q12.2;q22.2) gene fusion was detected in the tissue sample, which has been reported in extraskeletal Ewing sarcoma.

Stomach, GE Junction, EUS-FNA Final Diagnosis: Ewing Sarcoma

Fortunately, the patient’s PET scan did not demonstrate any evidence of metastatic disease, and the patient along with his care team decided to pursue systemic therapy as these tumors tend to be chemosensitive. The need for radiation therapy will be reviewed depending on the tumor’s response to systemic therapy. A strange presentation, this visceral Ewing sarcoma, and a reason why immunostains and molecular profiling are so important to rendering a definitive diagnosis. In our study set files, however, it will forever be dubbed my favorite weirdoma.

-Taryn Waraksa-Deutsch, MS, SCT(ASCP)^CM, CT(IAC), has worked as a cytotechnologist at Fox Chase Cancer Center, in Philadelphia, Pennsylvania, since earning her master’s degree from Thomas Jefferson University in 2014. She is an ASCP board-certified Specialist in Cytotechnology with an additional certification by the International Academy of Cytology (IAC). She is also a 2020 ASCP 40 Under Forty Honoree.

A Diamond in the Mucin

A 35 year old male and current every day smoker transferred his care to our genitourinary (GU) clinic after undergoing a partial nephrectomy for clear cell renal cell carcinoma. A PET scan was performed, and a 5 millimeter FDG-avid right upper lobe (RUL) lung nodule was identified. While the mediastinal and hilar lymph nodes were mildly enlarged on the chest CT, they did not demonstrate FDG-avidity on the PET scan. The patient was referred to pulmonary for a workup of his subcentimeter lung nodule. Given the patient’s age and clinical history, the care team and the patient decided to pursue a lung biopsy via robotic-assisted bronchoscopy and subsequent lymph node sampling via endobronchial ultrasound (EBUS).

After the timeout, the patient was intubated, and the pulmonologist performed an airway inspection. Upon entering the right upper lobe, the pulmonologist noticed blood streaking along the bronchial wall. A right upper lobe bronchoalveolar lavage (BAL) was collected and sent for culture and cytology. Following the trachea and lung survey, the respiratory team began the robotic-assisted bronchoscopy. After confirming the correct location with both radial EBUS and fluoroscopy, multiple fine needle aspirates (FNAs) were collected from the right upper lobe lung nodule. Only benign bronchial cells were identified on rapid onsite evaluation (ROSE). A right upper lobe bronchial brushing was attempted and sent directly to cytology. Forceps were used to obtain transbronchial biopsies from the radiologically-suspicious area. The team then switched over to linear EBUS for the lymph node sampling portion of the procedure. Lymph nodes were sampled at the following stations: 11L, Level 7, and 4R. On ROSE, only lymphocytes and anthracotic pigment were identified for all three lymph nodes.

Upon e, the cytologist brought the specimens back to the laboratory for accessioning and processing. The following morning, the same cytologist screened all cytology specimens and the respective cell blocks associated with the case. The lymph nodes were signed out as negative for malignancy, with lymphocytes and anthracotic pigment present. The RUL FNA was inadequate for diagnosis as the material consisted mainly of bronchial cells and alveolar macrophages. The FNA findings were correlated with the transbronchial biopsy, which was signed out as benign pulmonary parenchyma with hemorrhage, fibrin clot, and respiratory bronchiolitis. Similarly, the bronchial brushing was negative as only benign bronchial cells were identified. In screening the BAL, there was an overabundance of hemosiderin-laden alveolar macrophages on the smear, cytospin, and SurePath liquid-based preparations (Image 1). When screening the cell block slides, abundant macrophages were also identified (Image 2).

Images 1-2. Lung, Right Upper Lobe, Bronchoalveolar Lavage 1: Pap-stained SurePath Liquid-based Prep; 2: H&E Cell Block Section (400X).

However, in the cell block sections, the cytologist also identified cells stranded within areas of mucin (Image 3). It may have been easy to dismiss these as macrophages, but contrary to the dusty hemosiderin-laden macrophages in this specimen, these cells had abundant vacuoles, and irregular nuclei with more prominent nucleoli (Images 4-5). The pleomorphic nuclei and larger vacuoles distinguished these cells from lipid-laden macrophages most often seen in aspiration pneumonia.

Images 3-5. Lung, Right Upper Lobe, Bronchoalveolar Lavage 3: H&E Cell Block section (100X); 4: H&E Cell Block section (400X); 5: H&E Cell Block section (600X).

The cytologist, marking the areas of interest, swiftly rendered a diagnosis of positive for malignant cells, favoring metastatic clear cell renal cell carcinoma. She enthusiastically described the case to the attending cytopathologist, who immediately ordered confirmatory immunostains on the unstained paraffin sections of the cell block.

With adequate controls, CD68 highlights the alveolar macrophages (not shown). The tumor cells show positive staining for AE1/AE3 and PAX-8 (Images 6-7), while macrophages show negative staining for AE1/AE3 and PAX-8 (Images 8-9). The tumor cells also show positive staining for vimentin (Image 10) and RCC (focal), and negative staining for TTF-1.

Images 6-10. Lung, Right Upper Lobe, Bronchoalveolar Lavage 6: AE1/AE3-positive tumor cells; 7: AE1/AE3-negative macrophages; 8: PAX-8-positive tumor cells; 9: PAX-8-negative macrophages; 10: Vimentin-positive tumor cells.

RUL BAL Final Diagnosis: Positive for malignant cells. Metastatic clear cell renal cell carcinoma.

When we attend a biopsy of an area clinically suspicious for metastatic renal cell carcinoma, we prepare ourselves for an overtly bloody sample. Even if the smears consist mainly of blood during ROSE, we rest assured that we will usually find tumor cells in the cell block sections. By obtaining a BAL of the blood-streaked RUL, the pulmonologist was able to pinpoint an area of lymphangitic spread. We often give little credit to BALs in cytology, especially when using concurrent techniques such as FNAs to sample a subcentimeter lung nodule without bulky disease. However, as seen in this case, sometimes it’s the only specimen that can help us render a diagnosis of metastatic cancer.

Thyroid Tales, Part 2

Typically, our patients present to the endocrinology clinic after their thyroid nodules are incidentally found on staging or surveillance after being diagnosed with a primary cancer in another part of the body. Based on TI-RADS criteria, the clinician either monitors the nodule or refers the patient to radiology for a thyroid FNA. When we hear “thyroid nodule,” we rarely assume anything other than thyroid tissue. Whether the imaging favors benign or suggests a high risk of malignancy, we prepare ourselves to assess the FNA smears for follicular cells (and all the levels of atypia), colloid, macrophages, Hurthle cells, lymphocytes, etc. While we must keep an open mind, we are always caught off guard when we see anything other than thyroid-related cells. So as promised in the first edition of this post, here are four thyroid FNA cases with unsuspecting findings.

Case 1

A 47-year-old male presented with throat dysphagia and odynophagia. CT scan revealed a destructive mass within the thyroid gland with compression and invasion of the thyroid cartilage and seemed contiguous with a large pharyngeal mass, spanning approximately 8 centimeters. A follow-up PET scan noted multiple hypermetabolic thyroid masses within both lobes, direct invasion of the subglottic trachea and upper esophagus, and mediastinal lymphadenopathy.

FNA passes were obtained from the right lobe of the palpable thyroid mass.

Images 1-2. Thyroid, Right Lobe, FNA 1: DQ-stained smear; 2: Pap-stained smear.

Smears (Images 1 & 2) revealed poorly differentiated neoplastic cells, follicular cells, and colloid (not visualized). No features of papillary thyroid carcinoma, medullary carcinoma, or Hurthle cell neoplasm/carcinoma were identified.

Immunohistochemical stains performed on cell block sections showed the poorly differentiated neoplastic cells to be negative for thyroglobulin, TTF-1, and calcitonin; Follicular cells, which may probably be differentiated neoplasm, were positive for thyroglobulin, TTF-1, and negative for calcitonin. Unfortunately, the scant cellularity in the cell block specimen precluded additional stains. Giant cell and spindle cell features were not identified in this specimen. Morphological features are compatible with poorly differentiated carcinoma of the thyroid gland; however, metastasis from other sites cannot be excluded.

The patient then underwent a total laryngectomy and thyroidectomy (Images 3 & 4) and level IV neck dissection, bilateral modified radical neck dissection, and tracheostomy with reconstruction performed. The patient then underwent adjuvant radiation followed by palliative re-irradiation and chemotherapy after abnormal activity was noted throughout the neck. Treatment was discontinued due to severe disease progression.

Images 3-4. Thyroid, Thyroidectomy: 3: H&E section (200X); 4: H&E section (600X).

Final diagnosis: Poorly differentiated thyroid carcinoma with squamous differentiation arising in association with differentiated follicle derived carcinoma cells.

Case 2

A 68-year-old female presented with a 3.7 centimeter left lobe-filling thyroid nodule and a history of melanoma of the left anterior tibial region that was excised a decade prior. During that time, a sentinel lymph node biopsy identified microscopic metastasis. Seven years after her initial diagnosis, the patient underwent an excision of a right upper quadrant subcutaneous nodule, demonstrating metastatic melanoma. Three months after that excision, the patient had a low anterior resection of a rectosigmoid metastasis. A breast lesion was then identified five months later, and the patient underwent a mastectomy for melanoma involving the breast. Six months after her mastectomy, the patient had a segmental resection and excision of a left posterior thigh nodule, at which point she was enrolled in a clinical trial. The next month, four additional subcutaneous nodules were excised on the left thigh, calf, and arm. After 2 years of relatively stable disease, the patient underwent a partial gastrectomy, partial small bowel resection, and left lower extremity mass for recurrent melanoma. The last PET avid area to biopsy was the left-lobed thyroid nodule. Under ultrasound guidance, multiple FNA passes of the solid and hypervascular thyroid nodule. The smears (Images 5 & 6) and cell block (Image 7) featuring single cells with eccentric nuclei and prominent nuclei are presented below.

Images 5-8. Thyroid, Left Lobe, FNA 5: DQ-stained smear; 6: Pap-stained smear; 7: H&E Cell Block section (600X); 8: A103-positive.

Immunostains were performed on cell block sections, and the neoplastic cells are positive for A103 (Image 8), HMB45 (scattered cells), and SOX-10, while negative for CD45, TTF-1, thyroglobulin, and calcitonin.

The patient began treatment with temozolomide and completed 33 cycles of pembrolizumab. Her most recent metastasis demonstrated extensive tumor cell necrosis, and disease progression has slowed tremendously.

Final diagnosis: Melanoma.

Case 3

After developing sudden shortness of breath and chest tightness, a 40-year-old female patient presented to the emergency department. A large mediastinal mass compressing the heart and central structures in the chest was identified on CT scan. Two thyroid nodules were also noted during that time. The patient underwent a mediastinal biopsy, which demonstrated small cell lung cancer, and the patient underwent thoracic radiation and six cycles of chemotherapy, as well as whole brain radiation. Two years later, the patient established care with endocrinology for her 1.6 centimeter solid left lobe thyroid nodule and a 1.2 cm complex thyroid nodule in the right lobe. While the right nodule was consistent with a hyperplastic nodule, the smears and cell block of the left thyroid nodule are presented below (Images 9-11).

Images 9-11. Thyroid, Left Lobe, FNA 9: DQ-stained smear; 10: Pap-stained smear; 7: H&E Cell Block section (400X).

Immunohistochemical stains performed on cell block sections demonstrate the neoplastic cells were positive for TTF-1, AE1/AE3, synaptophysin, and CD56.

The patient then completed four subsequent cycles of chemotherapy with concurrent chemoradiation and is currently on active surveillance showing no evidence of disease for over 12 months.

Final diagnosis: Metastatic small cell carcinoma.

Case 4

A 59-year-old female presented to her primary care physician for gross hematuria and fatigue. Her thyroid workup demonstrated hypothyroidism on her thyroid function panel and a 2.3 centimeter solid and hypervascular thyroid nodule in the right lobe. Her urology workup revealed a 6.7 centimeter exophytic left kidney mass, and the follow-up CT scan identified a lytic lesion in the right iliac bone. The thyroid biopsy was performed in the endocrinology clinic while she was also establishing care with the urologic oncology team the same day. The smears and cell block specimen from multiple FNA passes are presented below (Images 12-14).

Images 12-15. Thyroid, Right Lobe, FNA 12: DQ-stained smear; 13: Pap-stained smear; 14: H&E Cell Block section (400X); 15: Vimentin-positive.

Immunocytochemical stains were performed on paraffin sections of the cell block. Tumor cells show positive staining for vimentin (Image 15), focal staining for e-cadherin, and negative staining for CK7, TTF-1, thyroglobulin, CD10, and RCC.

The patient was referred to radiology for a CT-guided biopsy of the lytic bone lesion, which demonstrated similar cells. The patient had a radical left nephrectomy, followed by sunitinib. The thyroid nodule was not responding to treatment, so they patient underwent a total thyroidectomy, which showed metastatic high-grade clear cell carcinoma with sarcomatoid progression, consistent with renal primary. In some areas, the thyroid follicles were proliferating and appear atypical, probably reactive to the metastatic carcinoma. A checkpoint inhibitor was added to the patient’s therapy, but the disease continued to progress, and the patient elected for palliative care.

Final diagnosis: Poorly differentiated carcinoma, consistent with metastatic renal cell carcinoma.

That’s a wrap! Stay tuned for the next series of cytology case studies!

A Series of Infectious Events

Working in a cancer center, our cytologists are well-versed in cancer morphology being able to diagnosis primary malignancies, distant metastases, and even combined metastatic disease in the same lymph node. What we don’t see as often as community hospitals are infectious diseases. However, we do have many immunocompromised patients at our institution, so the rare opportunistic infection does occur. And boy, do we get excited to pass the case around! Please find a series of infectious events embedded within this post. And unfortunately, we do not live in an area where coccidiomycosis is endemic, so beyond school, we haven’t had the pleasure of identifying those in our daily work.

Case 1. Lung, Bilateral, BAL (Bronchoalveolar Lavage)

A 73-year-old male patient was admitted to the ICU with pneumonia. The pulmonologist performed a bilateral bronchoalveolar lavage (BAL) to rule out pneumocystis pneumonia. We prepared a pap-stained smear, two cytospins, a SurePath liquid based prep, and a cell block. Two additional cytospins were sent to histology for GMS staining. While no malignant cells were identified, fragments of squamous epithelium with acute inflammation and necrosis were present. Multiple viral inclusions were identified, appearing as ground glass within the nuclei. (Image 1). These cells present with classic 3 M features: molding, multinucleation, & margination of chromatin. The cell block also highlights viral inclusions, but demonstrates pseudohyphae and spores associated with surrounding squamous cells as well (Image 2).

Images 1-2. Lung, Bilateral, BAL. 1: SurePath LBP; 2: H&E Cell Block section (400X).

Diagnosis: Herpes Simplex Virus (HSV) and Candida.

Case 2. Lung, Left Lower Lobe, CT-guided FNA

A 72-year old male with stage IIA squamous cell carcinoma underwent a VATS right upper lobectomy and mediastinal lymph node dissection. He completed adjuvant carboplatin/gemcitabine therapy. On a surveillance CT scan, the treated area demonstrated progression as well as multiple bilateral lung nodules. To determine whether the new left lower lobe superior segment lung nodule was a metastasis or new primary, a CT-guided biopsy was performed. The smears and cell block sections were negative for malignancy but demonstrated inflammatory cells and necrotic debris, consistent with a necrotizing inflammatory process (Images 3-5). A separate pass was sent for microbiological cultures to correlate our findings. The following day, Kinyoun and GMS stains were performed on paraffin-embedded sections of the cell block. No fungal organisms were identified on GMS, but acid-fast bacilli were noted by the cytologist on the Kinyoun-stained section (Image 6).

Images 3-6. Lung, Left Lower Lobe, CT-guided FNA. 3: DQ-stained smear; 4: Pap-stained smear; 5: H&E section (100x); 6: Kinyoun stain (600x).

Diagnosis: Acid-fast bacilli (AFB), consistent with Mycobacterium Avium Complex. Isolated and confirmed by microbiology.

Case 3. Lung, Right Upper Lobe, CT-guided FNA

A 58-year-old male presented with multiple lung nodules and a brain mass. We reviewed the brain mass excision from an outside institution and agreed with the original diagnosis of anaplastic oligodendroglioma, WHO grade III with a Ki-67 proliferation index that approached 20%. EGFR was not amplified (ratio 1.2), but 1p/q19 co-deletions were noted in greater than 75% of tumor cells. To rule out primary versus metastatic disease, the patient had a CT scan-guided biopsy of right upper lobe lung mass. No malignant cells were identified in the sample; however, necrotic debris and abundant fungal hyphae were noted (Images 7-9). A portion of the sample was sent to Microbiology for culture. The following day, a GMS and PAS stains were performed on paraffin-embedded sections of the cell block which demonstrated the same fungal hyphae seen in the smears and cell block preparations (Images 10 & 11).

Images 7-11. Lung, Right Upper Lobe, CT-guided FNA. 7: DQ-stained smear; 8: Pap-stained smear; 9: H&E section (400x); 10: GMS stain (400x); 11: PAS stain (400x).

Diagnosis: Abundant fungal hyphae, consistent with Aspergillus

Case 4. Left Hilum, EBUS-FNA

A 20-year-old female patient presented with patches, pain, and inflammation on her legs, and she was diagnosed with erythema nodosum. When her swelling and pain worsened, a chest X-ray demonstrated a left hilar mass, and a subsequent CT demonstrated the mass to be encircling the left superior pulmonary artery and obstructing the pulmonary vein along with multiple peribronchial ground-glass opacities and hilar lymphadenopathy. The concern from the referring physician was thymoma versus lymphoma given her age and clinical presentation. The patient underwent an endobronchial ultrasound to assess the hilar mass and lymphadenopathy. The lymph node aspirates appeared benign, with flow cytometry supporting the cytologic diagnosis. On the left hilum FNA, there were aggregates of lymphocytes, plasma cells, and epithelioid histiocytes with caseating necrosis and fibrosis (Image 12-14). Kinyoun, PAS, and GMS stains were performed on paraffin-embedded sections of the cell block. No acid-fast bacilli were identified. Fungal organisms in the form of budding yeast were noted on GMS (Image 15) and PAS stain. The patient was prescribed a 12-week course of antifungal medication.

Images 12-15. Lung, Left Hilum, EBUS-FNA. 12: DQ-stained smear; 13: Pap-stained smear; 14: H&E section (400x); 15: GMS stain (400x)

Diagnosis: Necrotizing inflammation with fungal organisms, suggestive of Histoplasmosis.

Case 5. Lung, Right Middle Lobe, BAL (Bronchoalveolar Lavage)

A 44-year-old male patient with uncontrolled Type II diabetes and hypertension presented to pulmonary after imaging demonstrated diffuse mediastinal and hilar lymphadenopathy. The differential diagnosis was sarcoidosis versus a lymphoproliferative process. An endobronchial ultrasound was performed to evaluate the lymph nodes, all of which came back as reactive. A BAL was performed and sent for cell count, cytology, flow cytometry, and microbiology. Flow cytometry analysis demonstrated a reversed CD4:CD8 ratio, and upon further testing, the patient was determined to have HIV. Eosinophilic froth or casts were identified on the cytopreparations of the BAL (Images 16). GMS and PAS stains were performed with adequate controls, and the PAS was negative for other fungal organisms while the GMS demonstrate positive staining for what we in cytology refer to as cups or crushed ping pong balls (Image 17). He was treated with Bactrim.

Images 16-17. Lung, Right Middle Lobe, BAL. 16: Pap-stained cytospin; 17: GMS stain (600x)

Diagnosis: No malignant cells identified. Positive for Pneumocystis jirovecii.

Case 6. Lung, Right Lower Lobe, CT-guided FNA

A 68-year-old male patient with a history of a renal transplant presented with an endobronchial mass in the left lower lobe that was biopsied and diagnosed as adenocarcinoma at an outside institution. We reviewed the slides in-house and determined the original tumor to be a mucoepidermoid carcinoma. After an unsuccessful staging procedure, a mediastinoscopy was performed, and the mediastinal lymph nodes showed hyalinizing non-necrotizing granulomata, suggesting underlying sarcoidosis. No microorganisms were identified with AFB, GMS, or PAS stains. The patient did not receive adjuvant therapy following the resection of his endobronchial tumor. Seven years later, he presented to the ER for syncope and 30 lbs. weight loss in 5 months. A CT scan was performed demonstrating a thick-walled cavitary lung mass in the right lower lobe. The patient was referred to radiology for a CT-guided FNA of the RLL mass. Fibrous tissue and abundant microorganisms with a polysaccharide capsule were identified on both FNA and core biopsy (Images 18-20). The PAS, GMS, Mucicarmine (Image 21), and Fontana Masson special stains were performed on cell block sections, with proper controls, highlighting abundant microorganisms. The patient was prescribed an antifungal for his cryptococcoma (cryptococcal lung abscess).

Images 18-21. Lung, Right Lower Lobe, CT-guided FNA. 18: DQ-stained smear; 19: Pap-stained smear; 20: H&E section (600x); 21: Mucicarmine stain (400x)

Diagnosis: No malignant cells identified. Abundant microorganisms, morphologically consistent with Cryptococcus species.

If you enjoyed this special series, look out for more in the future! And feel free to recommend or request interesting cases!

Histio Makes History

An 81 year old female presented to the head and neck clinic after being diagnosed with cutaneous T cell lymphoma of the posterior mid-parietal scalp at an outside institution. She was initially treated with Brentuximab every three weeks but developed significant toxicities. The patient’s previous “T cell lymphoma” material was reviewed at our institution and the immunophenotypic report described the neoplastic cells as being positive for CD45, CD2, CD4, BCL6+, CD3 (subset), and CD123 (scattered), while negative for CD7, CD8, CD20, CD30, CD56, EBER ISH, PAX5, and lysozyme. Immunohistochemical slides were not provided for review. Flow cytometric analysis determined that there was no immunophenotypic evidence of a clonal T cell population in the patient’s peripheral blood.

A second scalp biopsy was performed at another outside institution, and the findings were similar to the parietal scalp; however, there were atypical pleomorphic cells which displayed irregular contours, hyperchromasia, and multiple nucleoli. The atypical cells were predominantly positive for CD4 and diffuse positivity for CD1a. These same pleomorphic cells were negative for CD3, CD8, CD20, CD30, ALK1, BCL6, CD56, EBER, AE1/AE3, SOX10, Desmin, PAX5, MUM1, CD5, and Cam 5.2.

The smears contained large, highly pleomorphic cells with irregular, elongated, and multilobated nuclei, frequent nuclear grooves and folds, fine chromatin, prominent nucleoli, and variable amounts of pale, eosinophilic cytoplasm, alt.

The outside tissue block on the original scalp biopsy was requested, and our pathology department performed additional immunostains. The neoplastic cells of interest were positive for CD1a, S100, CD68 (a small subset), and negative for lysozyme, CD21, CD30, and CD3. Ki67 proliferation index was interpreted at approximately 60%. An unstained FFPE tissue section was sent to a reference laboratory, and the neoplastic cells were strongly positive for Langerin.

While the Brentuximab treatment initially appeared to have a positive impact on the overall disease burden, the PET CT following 3 cycles showed a mixed response, including resolution of cervical lymphadenopathy and identification of multiple new lung nodules and bulky mediastinal lymphadenopathy. Between that and numerous reported toxicities, the treatment protocol was discontinued. The patient was then referred to radiology for a CT-scan guided right lower lobe lung biopsy measuring 2.2 x 1.3 centimeters with an SUV or 29.6.

In the CT Scan suite, we received multiple FNA passes from the interventional radiologist and made air-dried and alcohol-fixed smears, rinsing the residual needle material into a tube of balanced salt solution for a cell block preparation. We determined our specimen was adequate for scant tumor cells, as depicted on the Diff-Quik smears below.

Images 1-2. Lung, right lower lob, CT-guided FNA. Diff-Quik stained smears.

In comparison to the material from the second scalp biopsy, the cells from the lung biopsy appeared identical. Our Pap-stained smears and H&E cell block sections also demonstrated the highly pleomorphic cells described above.

Images 3-6. Lung, Right Lower Lobe, CT-guided FNA. 3-4: Pap-stained smears, 5-6: H&E sections (5: 100x, 6: 400x).

Immunostains performed on the cell block slides with adequate controls show that the tumor cells are positive for CD1a, CD4, partially positive for CD45 and S100, negative for AE1/3, TTF-1, and p40.

Images 7-8. Lung, Right Lower Lobe, CT-guided FNA. Cell block section immunohistochemistry. 7: CD1a-positive; 8: partially S-100-positive.

Our pathologists felt the cells from the second scalp biopsy and the lung biopsy were representative of a Langerhans cell sarcoma, a form of malignant histiocytosis, rather than a T-cell lymphoma. It is possible that the first scalp biopsy’s diagnosis of T-cell lymphoma was due to sampling error and the pleomorphic cells of interest were missed. The Ki-67 proliferative index of 60% helped to distinguish between Langerhans cell histiocytosis and Langerhans cell sarcoma.

Molecular testing performed on the core biopsy was negative for a BRAF mutation and positive for an NF1 inactivating mutation. The tumor may then be sensitive to mTOR inhibitors and MAPK pathway inhibitors, such as MEK inhibitors. Appeals for a MEK inhibitor were denied by insurance, but fortunately, the tumor also demonstrated high PD-L1 expression at 90%, making this specific patient a candidate for pembrolizumab, which was fully covered by insurance.

____________________________________________________________________________________________

I can’t help but think about the disparities associated with cancer and the inaccessibility of potentially lifesaving or life-prolonging treatments. Sure, there may be viable alternatives, such as this case, but what if we had equal access to cutting edge, personalized therapies? What if the only therapy available was too costly to bear? Just because a cancer might be rare, such as Langerhans cell sarcoma, it doesn’t mean access to a proven effective therapy should also be rare. Even with drug assistance programs, so many patients face the harsh reality of tapping into their life savings to just to save their own life. When we became medical laboratory professionals, we promised to provide timely and accurate for all of our patients. Now, it’s time that pharmaceutical companies and our healthcare system as a whole work together to provide high quality, low-cost, readily accessible and personalized treatment options to every patient. They deserve that chance to overcome or at least manage their cancer.

Tumor on the Brain

Back in my Master’s program at Jefferson, I fondly remember the week we covered central nervous system (CNS) tumors. I was fascinated by the mnemonic tools we would use to identify different CNS tumors, such as “fried eggs” for oligodendrogliomas, perivascular pseudorosettes in ependymomas, and the whorling associated with meningiomas. Fortunately, for our patients, and unfortunately, for our diagnostic curiosity, we rarely see CNS tumors at my institution. Brain lesions resulting from metastatic carcinomas are typically well-identified via imaging and treated appropriately by the surgical, medical, and radiation oncology teams, but cytologists are available to screen cerebrospinal fluids (CSFs) for CNS involvement. For primary CNS tumors, however, we’re left recollecting the core memory of the second semester of our didactic phase. When a metastatic CNS tumor made its way into our lab, our cytology team swooned with excitement. (Yes, I know, but please introduce me to a lab professional who doesn’t embrace their quirks.) A 27-year-old male patient presented to radiation oncology three years after surgical debulking of a brain tumor at an outside institution. The patient, who was referred to radiation oncology at to treat the residual tumor at the original institution, did not follow up and developed an 8 centimeter recurrence a year after the initial resection. At this point, the patient experienced complete vision loss and underwent a biparietal-occipital craniectomy. A repeat brain MRI was performed a year later, and once again, a large enhancing extra-axial mass was identified along with multiple smaller masses also increasing in size. The patient received radiation after worsening difficulty with ambulation. After almost completing the planned fractions of radiation, the patient elected to stop their radiation therapy due to worsening seizures. A left neck mass was identified six months prior, and while the mass had not grown or caused pain, the patient was referred to head and neck surgical oncology for evaluation. Surveillance imaging demonstrated an enlarged left level 5A lymph node, suggestive of metastatic disease. Multiple ultrasound-guided fine needle aspiration biopsies were obtained from the lymph node, and ROSE was performed. The Diff-Quik-stained and concurrent Pap-stained smears demonstrated lesional tissue, although everything from epithelioid histiocytes to spindle cell melanoma to a renal primary were considered as a differential. Based on the location, a salivary gland primary was also a possibility for this case. The streaked cytoplasm and pseudoinclusions in both smears were concerning for a metastasis of the patient’s primary CNS tumor, but we were still hesitating to make the call.

Images 1-4. Lymph Node, Neck, Left, Level 5A, US-guided FNA. 1-2: Diff-Quik-stained smears, 3-4: Pap-stained smears.

The following morning, the H&E-stained FFPE cell block sections demonstrated the characteristic whorls expected for the patient’s primary, although the idea of metastasis was uncanny.

Images 5-6. Lymph Node, Neck, Left, Level 5A, US-guided FNA. H&E sections (6: 100x, 7: 400x).

We then used immunohistochemical studies to confirm our morphologic diagnosis. Immunostains performed on the cell block slides with adequate controls show that the tumor cells are positive for vimentin and PR (focal), while negative for AE1/AE3, EMA, CK7, CK20, TTF-1, Napsin A, p40, Pax8, synaptophysin, and S-100. The Ki-67 proliferation index fell at 18%, which is consistent with intermediate aggressive disease in a WHO Grade 2 atypical meningioma.

Images 7-8. Lymph Node, Neck, Left, Level 5A, US-guided FNA. Cell block section immunohistochemistry. 7: Vimentin-positive; 8: focally PR-positive.

The patient had next gen sequencing performed on his tissue, which demonstrated an NF-2 mutation, indicating he may benefit from MTOR inhibitors, but he elected not to pursue systemic therapy.

Where meningiomas account for 36% of primary brain tumors, atypical meningiomas comprise only 5-15% of all meningiomas (Cai et al., 202. Extracranial metastasis of atypical meningioma is a rare event, with only a few cases documented in the literature. While meningioma metastases are uncommon, a thorough collaboration between clinical impression and pathologic interpretation is necessary to ensure the possibility is not entirely excluded.

References

Cai C., Kresak J.L., Yachnis A.T. (2021) Atypical meningioma. Pathology Outlines. Retrieved October 11th, 2022, from https://www.pathologyoutlines.com/topic/cnstumoratypicalmeningioma.html.

P.S. I’d like to take this opportunity for a shameless plug. My Doctor of Health Science (DHSc) research survey is live now through November 23^rd, 2022. If you’re a medical laboratory professional or pathologist, please consider contributing to our field of laboratory medicine! Click the following link to read the consent form and take the one-time anonymous survey. Thank you for your time!

https://www.surveymonkey.com/r/leadinglabs

Triaging Times

As a clinical instructor and lead cytologist at my institution, I like to remind our newer cytologists and cytology students of the importance of being prepared for FNA biopsies so they develop good habits or best practices as they become more experienced. This level of preparation helps to create a culture of ongoing learning and improvement, which is necessary for the laboratory. In my experience, I’ve met some cytologists who prefer to go into a case blind, with the mindset that knowing the patient’s clinical history in advance muddies their knowledge, skills, and abilities, limiting their mindset by excluding the possibility of other diagnoses. While diving into the unknown might seem exciting, it is also a hindrance and could result in errors, especially when the clinical history helps us triage the patient’s sample. For example, knowing that the patient has a history of lymphoma or that the presentation state includes bulky lymphadenopathy prompts us to collect additional needle passes to send for flow cytometry analysis. Another concern is not knowing whether the patient has a history of breast, gastric, or esophageal cancer, and consequently processing the specimen routinely, which may result in an extended cold ischemic time. This delay in fixation along with insufficient formalin fixation can yield false negatives on ER/PR IHC in breast cancers and HER2 FISH in breast, gastric, and esophageal cancers, which could restrict the use of hormone therapies, such as tamoxifen and aromatase inhibitors for hormone receptor-positive (HR+) cancers, or trastuzumab for HER2+ cancers. I cannot overemphasize the importance of familiarizing yourself with clinical history and communicating case specifics while you act as a mediator between clinician and pathologist.

Whether the clinical history impacts the pre-analytical phase, such as specimen collection (limiting cold ischemic time or collecting additional needle passes for ancillary studies) or the analytical phase, as such processing (formalin fixation) and diagnosis (selecting an appropriate immunoprofile), we must remain vigilant and proactive in laboratory medicine. In this case, knowing the patient’s clinical history was of the utmost significance as it helped to reduce the number of immunostains and ancillary studies necessary to make the diagnosis. Using morphologic criteria in tandem with the patient’s clinical history narrowed the differential diagnoses to just two possible types of cancer, presented below.

A 59 year old male patient presented to the emergency room after an automobile accident. On imaging, the X-ray and CT scan identified a left humerus mass and fracture, and bloodwork was performed. His medical record was sparse and uneventful with no recent visits or encounters. To build a more comprehensive wellness profile and prepare for surgery, he was also offered a one-time screening for Hepatitis C, as an adult who was born between 1945 and 1965.

The left humerus mass was biopsied via CT-scan guidance and two passes were obtained. The Diff-Quik stained smears demonstrate large polygonal cells, some with abundant, granular cytoplasm and some isolated cells with naked nuclei. Vessels also appear to traverse some of the cell groups.

Images 1-2: Bone, Humerus, Left, CT-guided FNA. Diff-Quik-stained smears.

The Pap-stained smears also demonstrate polygonal cells with granular cytoplasm, nuclei with coarse chromatin, and prominent nucleoli. An interesting feature frequently identified in this case is the intranuclear inclusions, and in hindsight, a focus on these may have further reduced the number of immunostains performed.

Images 3-5: Bone, Humerus, Left, CT-guided FNA. Pap-stained smears.

The H&E-stained cell block sections show trabeculae with endothelial wrapping around the cell cords. While renal cell carcinoma was listed as a differential diagnosis due to its telltale oncocytic cytoplasm and vascularity, hepatocellular carcinoma was favored.

Images 6-7: Bone, Humerus, Left, CT-guided FNA. H&E sections (6: 100x, 7: 400x).

Immunostains were performed using proper positive and negative controls on the cell block sections, and the tumor cells show positive staining for Arginase, cam5.2, and Hepar1, while negative staining for CK7 and PAX8 (not shown).

Images 8-10: Bone, Humerus, Left, CT-guided FNA. Cell block section immunohistochemistry. 8: Arginase-positive; 9: cam5.2-positive; 10: Hepar1-positive.

Fortunately, before ordering immunostains, both our cytologist and pathologist working on the case peered into the patient’s medical record and noticed that he had recent bloodwork which demonstrated a positive Hepatitis C screening. This diagnosis was as recent as the identification of his humerus mass. Had it not been for his car accident, I can’t imagine how long he would have gone undiagnosed for both hepatitis and metastatic hepatocellular carcinoma. Incidental findings save lives, folks.

Granted, in settings of unknown primaries with widespread metastatic disease, such as carcinomatosis, an extensive workup is almost always inevitable. Narrowing down possible etiology based on information such as gender, age, and environmental or occupational exposure can help, but that doesn’t always yield a definitive answer as time- or cost-effectively as possible. In this case, that one clue of untreated Hepatitis C was all the cytopathology team needed. A rarity, sure, but as we are asked to do more personalized tests with less material, think of the patient’s specimen as a puzzle and keep your eye out for a clue both under the microscope and behind the computer. You never know what you might find that reduces errors and unnecessary testing while efficiently leading to a definitive diagnosis.

Cytology Case Study: Strike a Chord

Every FNA ROSE attended where the patient is conscious and attentive can be tricky to navigate. You have to remain cognizant of your word choice, your demeanor, and the delivery of your adequacy statement to the clinician. The patient is already in a heightened state of awareness because he or she is about to be probed with a needle (or six!) for a test that is likely to rule out a benign or malignant process. I prefer to go into my biopsies with some sort of clinical picture and as many details as I can retain – is there a previous history of cancer? Where is the lesion located? Is it a single mass or are there multiple lesions? What does the radiologic imaging suggest? Are there any elevated serum tumor markers? I need to be able to walk the walk and talk the talk. However, there are rare instances when cytotechs are asked to rush down to an unscheduled add-on biopsy where we have yet to research the impression documented in the patient’s medical record. In those situations, I ask the clinician (typically an interventional radiologist) all the questions I can think of while still emulating some form of confidence to the patient.

I entered the procedure room and greeted the radiologist, radiology fellow, tech, nurse, and the patient, a 57-year-old male who was prone and alert on the table. I jotted down notes during the timeout and pulled the radiologist aside to ask, “does the patient have a history of cancer?” In this case, the answer was “they have a soft tissue tumor in the left gluteus, which is what we’re biopsying.” “Alright, let’s get those differentials rolling – sarcoma; after my hibernoma experience – a lipomatous tumor; or could it be a carcinoma (because yes, I’ve seen a lung adenocarcinoma metastasize to the gluteal muscle before)? Hmm… what else? What other mesenchymal tumors could originate here… or metastasize here?” My brainstorming balloon was popped by the radiologist asking if I was ready for the first needle pass. I replied, “Yes, of course!” I glance over at the patient and smile, trying to assure him AND myself that I’ll be able to give him a definitive answer to his puzzle.

Here’s what I visualized under the microscope after I stained the first air-dried smear in our Diff-Quik solutions.

Images 1-2. Left gluteal FNA, DQ-stained smears.

My inner monologue became: “Well, it’s not a sarcoma or a carcinoma. It doesn’t look malignant. Not quite a hibernoma. What is with that myxoid matrix? It’s not mucinous or serous, so… what is it…? It’s granular! Wait. Those nuclei. They’re so… what’s the word? It’s definitely representative of the lesion. Regardless, it’s adequate!” I turned away from my microscope to face the team – “The sample is adequate. May I have a few more passes for my cell block, and will you collect core biopsies, too? “Yes and yes,” the radiologist replied. I smiled again at the patient, and he mouthed, “thank you.” “Phew, mission accomplished,” I thought. “Now what the heck are those hallmark cells called mixed in with a majority of epithelioid cells arranged in chords?”

I climb the stairs up to the lab and do a quick Google search. “DUH! Physaliphorous cells!” These cells have a distinct feature where the nucleus is centrally located but is also scalloped by cytoplasmic vacuoles. There weren’t as many physaliphorous (physaliferous) cells as I had hoped to appreciate. Some of the cells looked lipoblastic in nature with larger vacuoles displacing the nuclei to the periphery, almost signet ring in nature, many were cuboidal. But that was it… those cells! Now, imagine the scene in Finding Nemo where Nemo struggles to tell his classmates he lives in an anemone. That was my garbled attempt at pronouncing “physaliphorous” to the attending pathologist when sharing my interpretation. She looked at me like I was saying anything other than the word I was trying to reproduce. I cannot blame her; I still turn beet-red at the memory. But I was convinced that a chordoma was the tumor I presented to her.

After I processed my FNA, I examined the patient’s electronic health record to see if he had any previously biopsied neoplasms on file, and much to my surprise, the patient was diagnosed with a primary chordoma of the sacrum and treated with en bloc resection and radiation in 2013. Mutation analysis was performed on the resection of this chordoma, which exhibited a homozygous loss of CDK2NA (p16). The patient had one recurrence at an outside facility in 2015 and transferred his care to our institution for follow-up. Now, the patient presented with this gluteal metastasis and soon thereafter, a paraspinal metastasis. As the patient’s chordoma did not completely respond to radiation, the clinical care team turned to the tyrosine kinase inhibitor, Gleevec, which was discontinued due to disease progression. The patient’s regimen then went on to include sunitinib, which was also discontinued due to disease progression, palbociclib, then nivolumab, followed by radiation to the thoracic spine, sorafenib, and now is on a clinical trial for patients with advanced refractory cancers.

When I turned in my Diff-Quik & Pap-stained slides and the cell block H&E sections with a diagnosis of chordoma the next day, the attending cytopathologist paged through one of our cytology texts to a tidbit on chordomas before signing out the case. She reviewed the section with me. Other than the unique physaliphorous cells, it turns out a diagnosis of chordoma is fairly rare, as it is the only malignancy derived from the notochord, typically occurring at either end of the axial skeleton.¹ Metastasis of these tumors is also rare, so this case of widespread metastatic disease was even more intriguing to me.

Images 3-8. Left gluteal FNA . Images 3-5, Pap-stained smears; 6-8, H&E cell block sections.

References

Cibas, E. S., & Ducatman, B. S. (2009). Cytology: Diagnostic Principles and Clinical Correlates, Expert Consult – Online and Print (3rd ed.). Saunders.

-Taryn Waraksa, MS, SCT(ASCP)^CM, CT(IAC), has worked as a cytotechnologist at Fox Chase Cancer Center, in Philadelphia, Pennsylvania, since earning her master’s degree from Thomas Jefferson University in 2014. She is an ASCP board-certified Specialist in Cytotechnology with an additional certification by the International Academy of Cytology (IAC). She is also a 2020 ASCP 40 Under Forty Honoree.