Michael H. Roh

Development and Validation of a Scalable Next-Generation Sequencing System for Assessing Relevant Somatic Variants in Solid Tumors

Next-generation sequencing (NGS) has enabled genome-wide personalized oncology efforts at centers and companies with the specialty expertise and infrastructure required to identify and prioritize actionable variants. Such approaches are not scalable, preventing widespread adoption. Likewise, most targeted NGS approaches fail to assess key relevant genomic alteration classes. To address these challenges, we predefined the catalog of relevant solid tumor somatic genome variants (gain-of-function or loss-of-function mutations, high-level copy number alterations, and gene fusions) through comprehensive bioinformatics analysis of >700,000 samples. To detect these variants, we developed the Oncomine Comprehensive Panel (OCP), an integrative NGS-based assay [compatible with < 20 ng of DNA/RNA from formalin-fixed paraffin-embedded (FFPE) tissues], coupled with an informatics pipeline to specifically identify relevant predefined variants and created a knowledge base of related potential treatments, current practice guidelines, and open clinical trials. We validated OCP using molecular standards and more than 300 FFPE tumor samples, achieving >95% accuracy for KRAS, epidermal growth factor receptor, and BRAF mutation detection as well as for ALK and TMPRSS2:ERG gene fusions. Associating positive variants with potential targeted treatments demonstrated that 6% to 42% of profiled samples (depending on cancer type) harbored alterations beyond routine molecular testing that were associated with approved or guideline-referenced therapies. As a translational research tool, OCP identified adaptive CTNNB1 amplifications/mutations in treated prostate cancers. Through predefining somatic variants in solid tumors and compiling associated potential treatment strategies, OCP represents a simplified, broadly applicable targeted NGS system with the potential to advance precision oncology efforts.

The Application of Molecular Diagnostic Studies Interrogating EGFR and KRAS Mutations to Stained Cytologic Smears of Lung Carcinoma

EGFR and KRAS mutation analyses are of increasing importance for guiding the treatment of non-small cell lung carcinomas. Insufficient cellularity of cell blocks can represent an impediment to the performance of these tests. We investigated the usefulness of cytologic direct smears as an alternative specimen source for mutation testing. Tumor cell-enriched areas from freshly prepared and archived rapid Romanowsky-stained direct smears in 33 cases of lung carcinoma were microdissected for DNA isolation and evaluated for EGFR and KRAS mutations. EGFR mutations were detected in 3 adenocarcinomas; 2 tumors had the L858R substitution and 1 an exon 19 deletion. KRAS mutations affecting codon 12, 13, or 61 were detected in 11 cases (8 adenocarcinomas and 3 non-small cell carcinomas). EGFR and KRAS mutations were mutually exclusive. Hence, archived and freshly prepared direct smears represent a robust and valuable specimen source for molecular studies, especially when cell blocks exhibit insufficient cellularity.

Ancillary techniques on direct‐smear aspirate slides

Numerous cytologic techniques aimed at effectively acquiring patient material for molecular testing have been proposed. Such techniques are becoming ever more important in an age of personalized medicine. In this commentary, the authors explored some more commonly proposed techniques to aid in the molecular testing of cytologic specimens. These techniques include the use of cell blocks, direct cytologic smears, filter paper storage, frozen samples, and enriched cellular techniques such as ThinPrep and cytospin preparations. Direct‐smeared slides demonstrate excellent preservation of DNA, are easy to prepare, and are amenable to immediate adequacy at the time of the fine‐needle aspiration (FNA) procedure as well as effective subsequent tumor purity estimation. Cell block methods cannot be assessed at the time of FNA and often demonstrate insufficiency, whereas filter paper and frozen techniques do not allow for the direct assessment of the presence and purity of tumor cells in the sample. Direct‐smeared slides are emerging as the most effective preparation and storage medium of cytologic material to be used for molecular testing. Their cost‐effectiveness, ease of use, and reliability have cemented them as the optimal solution for cytopathologists to fulfill the role of providing advanced molecular testing on patient samples. Cancer (Cancer Cytopathol) 2013.

Reply to Ancillary techniques on direct-smear aspirate slides: a significant evolution for cytopathology techniques.

We thank da Cunha Santos et al for their interest in our commentary, and agree that the use of FTA cards is most likely a sound choice for long-term DNA storage. In our experience, the use of Diff-Quik–stained cytologic smears, stored for years, is also satisfactory for the isolation of high-quality DNA. Several points deserve clarification regarding these methods. As the authors point out, the breakage of slides during transport is a potential issue; nonetheless, careful packing of slides can prevent this and we do not believe that this is a serious drawback to the use of smears. With regard to the possible problem with the stability of unstained smears, we acknowledge heat and humidity as potential problematic issues. This can be circumvented by storing the slides in climate-controlled rooms. In our practice, any remaining air-dried, unstained slides are stained with Diff-Quik, coverslipped, and stored long term after ancillary tests have been completed. If additional DNA is needed in the future, one of the extra Diff-Quik–stained slides can be decoverslipped for tumor cell microdissection and DNA purification. The authors fail to provide convincing data that filter papers are any less susceptible than unstained smears to heat and humidity. Because it is possible that the filter papers can attract moisture, a side-by-side comparison would be interesting. More clarification is needed regarding the authors’ insistence that analysis of a corresponding stained cytospin slide obtained from the same needle rinse as the FTA card helps ensure the identity of the material in the FTA card. When assessing tumor purity in a specimen, or whether tumor cells are even present, there is no better substitute than the ability to examine the actual cells to be extracted with the microscope. This is possible with stained slides. Slides are not homogenous; they often display contaminating, benign cells as well as areas of relatively pure tumor cells. Microdissection of tumorenriched areas is possible with smears whereas this is not possible with FTA cards. The use of cytospin preparations to infer tumor cellularity and purity on FTA cards is still an extrapolative exercise; extrapolation is not necessary with smears because “what you see is what you get.” More data are needed to conclude whether FTA cards or stained slides provide higher quality nucleic acids on extraction. Molecular testing methodology is constantly evolving, and the requirements of nucleic acid quality will require constant evaluation. During this very exciting era, cytopathologists are starting to investigate various cytopreparatory platforms (smears, FTA cards, ThinPrep slides, etc.) and slide staining conditions for the isolation of nucleic acids. We believe that no method is perfect and each can be continually improved. A major advantage of cytologic sample processing lies in the versatility of the sample slide preparation methods available, and we believe these different platforms are complementary. Collaborative efforts to continuously optimize nucleic acid purification from various specimen preparations is essential for cytopathologists worldwide to better serve their patients in this constantly evolving era of precision medicine.

The use of stained cytologic direct smears for ALK gene rearrangement analysis of lung adenocarcinoma

Rearrangements involving the anaplastic lymphoma kinase (ALK) gene are present in approximately 5% of lung adenocarcinomas. Crizotinib is approved for the treatment of lung adenocarcinomas harboring ALK rearrangements. Patients with advanced stage lung cancer are not candidates for surgical resection of their primary tumors. For these patients, cytologic specimens often represent the only diagnostic tissue available. Cell blocks (CBs) are routinely used for molecular studies; however, insufficient CB cellularity can impede the performance of these assays.

Application of immunocytochemistry and BRAF mutational analysis to direct smears of metastatic melanoma

The cytodiagnosis of melanoma in fine‐needle aspiration (FNA) specimens can be challenging, often requiring the use of immunocytochemistry. As constitutively activating mutations in the BRAF oncogene are present in at least 40% of melanomas, the use of FNA material to interrogate the BRAF mutational status is likely to increase. Because cell blocks, traditionally used for these studies, can occasionally exhibit insufficient tumor cellularity, the authors investigated the utility of direct smears for immunocytochemistry and BRAF mutational analysis.

Group consensus review minimizes the diagnosis of "follicular lesion of undetermined significance" and improves cytohistologic concordance

We conducted a group consensus review of thyroid aspirates that were previously interpreted as “atypia of undetermined significance/follicular lesion of undetermined significance” (AUS/FLUS) and followed by surgical interventions. The study aimed to investigate if consensus review would minimize the diagnosis of AUS/FLUS with an optimal interobserver agreement and also promote a better cytohistologic concordance. A group of reviewers who were blinded to the corresponding histologic findings simultaneously evaluated a total of 50 aspirates at a multiheaded light microscope. Using the Bethesda System for Reporting Thyroid Cytopathology as a guideline, a consensus interpretation was reached upon review of each aspirate. Interobserver agreement was calculated and recorded. The cytohistologic correlation was then performed between the consensus interpretation and the corresponding histologic diagnosis. The consensus review reclassified 26 (52%) aspirates as non‐neoplasia/benign, 10 (20%) as follicular neoplasm/suspicious for a follicular neoplasm, 1 (2%) as papillary thyroid carcinoma, and 2 (4%) as nondiagnostic. Eleven (22%) aspirates remained AUS/FLUS. The interobserver agreement across the five diagnostic categories ranged from 71.6% to 100% with an average level of 88.8%. Cytohistologic concordance was achieved in 24 of 26 (92.3%) and 9 of 11 (81.8%) aspirates that were reclassified as non‐neoplasia/benign and neoplasia/malignancy, respectively. A diagnostic accuracy of 89.2% (33/37) was obtained in reclassified cases. In conclusion, the group consensus review minimized AUS/FLUS, offered an optimal level of interobserver agreement, and most importantly, promoted excellent cytohistologic concordance in reclassified cases and, therefore, could play a substantial role in the future in reducing reaspiration and/or unnecessary surgeries. Diagn. Cytopathol. 2012.

Diagnostic utility of PAX8 and PAX2 immunohistochemistry in the identification of metastatic Müllerian carcinoma in effusions

Morphologic distinction of Müllerian carcinomas from non‐Müllerian carcinomas in effusion specimens by cytomorphology alone can be diagnostically challenging. Therefore, immunohistochemical adjuncts can be useful in differentiating Müllerian from non‐Müllerian metastases. In this study, we evaluated the expression of PAX8 and PAX2 in malignant effusions collected from patients with known Müllerian and non‐Müllerian carcinomas. Sections from cell blocks prepared from 152 effusion specimens (54 and 98 cases representing metastases from Müllerian and non‐Müllerian primaries, respectively) were immunostained with rabbit polyclonal antibodies against PAX8 and PAX2. Immunopositivity was defined as the presence of strong nuclear staining in at least 25% of the tumor cells. Fifty‐two (96%) and 13 (24%) of the 54 Müllerian carcinomas were positive for PAX8 and PAX2, respectively. PAX8 positivity was seen in only four (4%) of 98 non‐Müllerian carcinomas; these represented metastasis from a large cell neuroendocrine lung carcinoma, papillary thyroid carcinoma, renal cell carcinoma, and acinic cell carcinoma of the parotid gland. PAX2 positivity was not seen in any of the non‐Müllerian carcinomas. The results demonstrate that both PAX8 and PAX2 are highly specific markers for metastatic Müllerian carcinomas in cell block preparations from effusion specimens (96% and 100%, respectively). PAX8, however, is more sensitive than PAX2 in identifying Müllerian carcinomas in fluids (96% versus 24%). Overall, immunohistochemistry for PAX8 and PAX2 represent diagnostically useful adjuncts in identifying a Müllerian carcinoma as a source of a malignant effusion. Diagn. Cytopathol. 2010.

The application of cytogenetics and fluorescence in situ hybridization to fine‐needle aspiration in the diagnosis and subclassification of renal neoplasms

Percutaneous fine‐needle aspiration (FNA) cytology is an important diagnostic test for the evaluation and management of selected renal masses. Cytogenetic analysis of cytology specimens can serve as an adjunct for precise classification because certain tumors are associated with specific chromosomal aberrations. This study summarizes our experience with the application of conventional cytogenetics and fluorescence in situ hybridization (FISH) to renal FNA specimens.

Utility of PAX8 and PAX2 immunohistochemistry in the identification of renal cell carcinoma in diagnostic cytology

The diagnosis of metastatic renal cell carcinoma (RCC) in cytology specimens may be difficult to confirm on the basis of cytomorphology alone. Often, immunohistochemistry serves as an important adjunct in confirming this diagnosis. Recently, PAX2 was shown to be useful in this regard. In this study, we sought to compare the utility of PAX8 to that of PAX2 immunohistochemistry in the diagnosis of RCC in cytology specimens. First, we verified the performance of PAX8 immunohistochemistry on a tissue microarray (TMA) composed of 54 cases of RCC; PAX8 immunoreactivity was seen in at least 10% of the tumor cells in all cases. Next, we applied PAX8 immunohistochemistry to cell block sections prepared from 24 cases of RCC, obtained from fine‐needle aspirates and effusion specimens. PAX2 immunohistochemistry was performed for comparison. Immunopositivity was defined as the presence of nuclear staining in at least 10% of tumor cell nuclei. Immunoreactivity for PAX8 and PAX2 was seen in 21 (88%) and 20 (83%) of the 24 cases, respectively. The presence of either PAX8 or PAX2 immunostaining was present in 22 of 24 cases, thus showing a total sensitivity of 92%. Overall, the results indicate that PAX8 and PAX2 are diagnostically useful adjuncts in confirming the diagnosis of RCC in cytology specimens. Diagn. Cytopathol. 2012.