Highlights in Pathology: Cytopathology (October 2019)

Dr. Hyang-Mi Ko, Toronto General Hospital / University Health Network

1. Immunocytochemistry for Predictive Biomarker Testing in Lung Cancer Cytology

Jain D, et al (Cancer Cytopathology 2019;127:325-339)

Cytology samples are an accepted source of high quality of DNA for lung cancer biomarker testing. However, immunocytochemical tests (ALK/PD-L1/ROS1) are challenging due to the variety of cytologic preparation techniques which potentially affect reliability of the results.

A review article which provides a comprehensive summary of the current status of biomarker testing using cytology samples as substrates.

The article highlights sample- and preparation type-specific challenges regarding validation and interpretation of immunocytochemistry (ICC).

It focuses on:

1. the analysis of various pre-analytical, analytical, and post-analytical factors influencing ICC results
2. discusses measures for validation of ICC protocols; and
3. provides a comprehensive summary of published data.

PD-L1 presents greater challenges than ALK and ROS1 for validation because 5 different antibody clones and assays using variable cut-offs for positivity are in use.

Concerns about sampling bias with small biopsies and cytology samples due to tumor heterogeneity are addressed.

Although most laboratories restrict ICC to cell blocks (CB), the authors clearly indicate that non-CB preparation can be useful and reliable substrates for testing after proper optimization and with rigorous quality control.

These comments are invaluable for advanced non-small cell lung carcinoma patients in which cytology samples, such as effusions and fine-needle aspirates, are often the first and only source for biomarker testing.

2. Accuracy of Diagnosing Mantle Cell Lymphoma and identifying its Variants on Fine-Needle Aspiration Biopsy

Foshat M, et al. (Cancer Cytopathol 2019;127:44-51)

Use of fine needle aspiration biopsy (FNAB) in diagnosing and subclassifying lymphomas has been debated.

In daily practice, it is important to understand in which clinical scenarios and pathologic contexts FNAB can provide a reliable diagnosis, and when  core needle or surgical biopsy  are required for patient management.

The authors demonstrate that in their practice FNA is an effective tool for the diagnosis of mantle cell lymphoma (MCL) and for subtyping (blastoid and pleomorphic variant) using a multiparametric approach.

Foshat at al. use a combination of flow cytometry, immunoperoxidase stains (incl. Ki-67), and fluorescent in situ hybridization (FISH), and approximately 93% of cases with FNAB were appropriately classified as MCL. Variant morphology (blastoid, pleomorphic, and high PI index) was accurately detected in 74% of cases.

Identification of those variants seems to be more challenging than classic MCL because of their tendency to display non-traditional immunophenotypes including complete loss of CD5, partial expression of CD23, and frequent aberrant expression of CD10, especially in the blastoid variant.

The authors used immunoperoxidase stains for cyclin D1 and SOX11 to confirm MCL in these variants, utilizing either cell block or cytospins. {Based on my experience, cytospin preparation are entirely feasible for low cellular FNA samples, work well (especially for nuclear markers) and additionally can be used for FISH!}.

The authors concluded that definitive FNAB diagnosis/classification can provide a clinically actionable diagnosis of MCL including its clinically relevant, biologically aggressive variants.

3. Cytologic Processing of Ureteral Microbiopsies is associated with Higher Sensitivity for Detection of Urothelial Carcinoma Compared to Conventional Biopsy Processing

Sheridan TB et al. (Journal of the American Society of Cytopathology 2019 epub)

Cell blocks (CB) are a common type of cytology sample preparation used for ancillary testing and sometimes they provide insights into tissue architecture as well as cytomorphology.

This research study illustrates the effectiveness of cytology processing compared to conventional surgical biopsy processing for microbiopsy specimen.

The authors use Cellient® (Hologic, Marlborough, MA) and applied this device for ureteral biopsy samples due to high rates of non-diagnostic or negative ureteral biopsies processed conventionally from patients confirmed urothelial carcinoma on ureteral washing cytology in their institution.

As expected, results showed that cytologic processing of bladder microbiopsies provides higher sensitivity than surgical processing, which appears to be related to a higher recovery of denuded cells with ThinPrep and minimization of tissue loss during processing.

ThinPrep (high grade nuclear morphology) and cell block (structure) provide complementary diagnostic information and, in some cases, permits confirmation of invasion in cell block. FISH testing was performed to exclude the possibility of false positive cytologic diagnosis.

One caveat of this study is the use of a commercial automated cell block machine for which one of authors is listed as inventor.

The paper suggests that, in order to maximize cell recovery from small tissue fragments, an approach of admittedly more labor-intensive cytologic processing can be pursued which may be more effective than conventional biopsy processing.   

4. A Combination of MTAP and BAP1 Immunohistochemistry in Pleural Effusion Cytology for the Diagnosis of Mesothelioma

Kinoshita Y, et al.(Cancer Cytopathol. 2018 Jan;126(1):54-63)

Malignant mesothelioma is diagnostically challenging in effusion cytology. In order to provide a more definitive diagnosis I am always looking for a diagnostic marker to differentiate between benign and malignant mesothelial cells.

This paper explored a novel IHC marker, MTAP, one of the 9p21 gene clusters, and proposes it as a potential alternative to FISH. MTAP nuclear/cytoplasmic IHC alone shows high specificity (100%) like FISH, but its sensitivity is somewhat lower (42.2% vs. 62.2%).

In combination with BAP1 nuclear immunostaining the combined sensitivity for mesothelioma is increased up to 77.8% which is close to the sensitivity of a combination BAP1 IHC and FISH (84.4%).

The authors set the cutoff value at 50% for MTAP and BAP1 IHC and 10% for 9p21 FISH.

Testing of cell blocks and matched tissue demonstrated complete correlation.

Few limitations are worth to mention: 

1. The sarcomatoid subtype is not included
2. Assessment of MTAP IHC (as with BAP1) requires adequate cellularity and has challenges in some cases with weak and/or heterogeneous staining
3. A relatively small number of cases were analyzed to confirm the usefulness of MTAP.

Despite these limitations, the BAP1+MTAP panel seems to be a highly reliable marker combination especially where 9p21 FISH is not readily available.

5. The International Academy of Cytology Yokohama System for Reporting Breast Fine Needle Aspiration Biopsy Cytopathology

Field AS, et al. (Acta Cytologica 2019;63:257-273)

When I was working in a middle-income Asian country, breast FNAB was a popular procedure to investigate breast lumps and rapid, accurate, and cost-effective diagnostic test. I have observed a diminished role of FNAB in the developed world with the increasing role of core needle biopsy (CNB).

Since many breast problems are first discovered by women themselves, often by chance, I still believe in the diagnostic role of FNAB for patients with a palpable lesion, particularly in reducing patient anxiety immediately when a benign diagnosis can be confirmed.

A new IAC System for reporting Breast Fine Needle Aspiration Biopsy (FNAB) will be published as a textbook soon. The paper provides the rational for and a summary of this new system.

The goals of the System are to:

• stimulate the appropriate use of breast FNAB
• improve the reporting of breast FNAB
• facilitate the communication between the cytopathologist and breast clinicians
• to facilitate optimal breast patient care.

This IAC Yokohama system defines five categories with a risk of malignancy (ROM) and a suggested management algorithm;

• Insufficient/Inadequate (ROM: 2.6-4.8%)
• benign (1.4-2.3%)
• atypical (13-15.7%)
• suspicious of malignancy (84.6-97.1%)
• malignant (99.0-100%).

Management suggestions for each of the categories are stratified for Best Practice management vs. Low/Middle Income Countries management dependent on the availability of local medical resources and local practices.

If rapid on-site evaluation (ROSE) is available during the procedure, CNB is recommended except for benign and “triple test” malignant cases.

The System emphasizes high quality performance and preparation and recommend direct smears rather than liquid-based cytology.

The provided key diagnostic cytologic features and definitions for each of the categories which are useful for anyone involved in the evaluation of breast fine needle specimens.