630375
research-article2016
IJSXXX10.1177/1066896916630375International Journal of Surgical PathologyKiuru et al
Original Articles
Spitz Tumors: Comparison of Histological Features in Relationship to Immunohistochemical Staining for ALK and NTRK1
International Journal of Surgical Pathology 1–7 © The Author(s) 2016 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/1066896916630375 ijs.sagepub.com
Maija Kiuru, MD, PhD1,2, Achim Jungbluth, MD2, Heinz Kutzner, MD3, Thomas Wiesner, MD2, and Klaus J. Busam, MD2
Abstract Spitz tumors are a group of melanocytic neoplasms with distinct morphological features that tend to affect young individuals. Distinguishing benign from malignant Spitz tumors can be challenging, but cytogenetic and molecular tests have contributed to improvements in diagnostic accuracy. Spitz tumors harbor diverse genetic alterations, including mutations in HRAS, loss of BAP1, or kinase fusions in ROS1, NTRK1, ALK, BRAF, and RET genes. Limited data exist on the correlation between histopathological features and kinase fusions. Here, we describe the histopathological features of 105 Spitz tumors (Spitz nevi and atypical Spitz tumors), comparing lesions according to their immunoreactivity for ALK or NTRK1. Intersecting fascicular growth of fusiform melanocytes was seen in all but one ALK-positive tumor (27 of 28 or 96.4%), whereas it was infrequent in NTRK1-positive tumors (5 of 20 or 25.0%) and tumors negative for both ALK and NTRK1 (96.4% vs 25.0% vs 8.7%, P < .0027). There was a trend toward ALK-positive tumors being amelanotic compared with NTRK1-positive tumors and combined ALK-/NTRK1-negative tumors (89.3% vs 45% vs 47.4%, respectively, P = .1023) and lacking epithelioid cell morphology (0% vs 45.0% vs 41%, respectively, P = .6985). In conclusion, this study confirms that although not specific, the growth pattern of intersecting fascicles of amelanotic fusiform melanocytes is strongly associated with ALK expression. Keywords ALK, NTRK1, Spitz nevus, immunohistochemistry
Introduction Spitz tumors are a group of melanocytic tumors typically occurring in young individuals.1,2 They share a number of light microscopic findings, in particular cytological features (eg, plump spindled and large epithelioid melanocytes, multinucleated melanocytes), but also secondary epidermal or stromal changes. Most of the lesions are benign nevi, but rare melanomas with Spitz nevus–like features exist. There is a set of atypical tumors representing lesions for which a definitive distinction between benign and malignant is not always possible.3-6 Because the correct diagnosis of Spitz tumors with atypical features can be very challenging at the light microscopic level, pathologists have explored cytogenetic and molecular assays to help categorize these proliferations as benign versus malignant. Cytogenetic analysis has emerged as a useful ancillary method to improve diagnostic accuracy in difficult cases: the majority of melanomas harbor numerous and complex chromosomal aberrations, whereas
most nevi, including Spitz nevi, do not.7-10 However, it has also become apparent that a few isolated chromosomal copy number gains or losses may be present in benign Spitz nevi, and miscellaneous aberrations can be found in indolent borderline lesions.11-13 Recent studies have documented a diverse genetic background among Spitz tumors, including mutations in Harvey rat sarcoma viral oncogene homolog gene (HRAS),14 loss of BRCA1 associated protein-1 gene (BAP1),15 and kinase fusions in C-ros oncogene 1 receptor tyrosine kinase gene (ROS1), neurotrophic tyrosine kinase receptor type 1 gene (NTRK1), anaplastic lymphoma receptor tyrosine kinase gene (ALK), v-raf murine 1
University of California Davis, Sacramento, CA, USA Memorial Sloan-Kettering Cancer Center, New York, NY, USA 3 Dermatopathologie Friedrichshafen, Friedrichshafen, Germany 2
Corresponding Author: Klaus J. Busam, Department of Pathology, Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, NY 10065, USA. Email: [email protected]
Downloaded from ijs.sagepub.com at UNIV CALIFORNIA SAN DIEGO on February 12, 2016
2
International Journal of Surgical Pathology
sarcoma viral oncogene homolog B gene (BRAF), and ret proto-oncogene (RET) genes.16 Some aberrations are associated with a distinct histopathological phenotype. For example, Spitz nevi with oncogenic mutations in HRAS with concomitant gain of 11p are characterized by singly disposed large epithelioid melanocytes in between thickened collagen bundles.14 Tumors with BAP1 loss are composed of large epithelioid melanocytes characterized by enlarged round to oval nuclei with open chromatin pattern and distinct nucleolus and abundant cytoplasm.15,17 Recently, we found the feature of a plexiform growth pattern of intersecting fascicles of fusiform melanocytes common in Spitz tumors with ALK rearrangement.18 Yeh at al19 subsequently published similar observations. Here, we analyze more systematically the possible association of an intersecting fascicular growth pattern with ALK expression. ALK-positive Spitz tumors were compared with NTRK1-positive lesions and Spitz tumors negative for both ALK and NTRK1.
Materials and Methods Selection of Cases Cases were selected from the files of 2 of the authors (KJB and HK). Lesions diagnosed as a Spitz nevus (n = 38) or atypical Spitz tumor (n = 67) were included in the study. Tumors diagnosed as malignant melanoma were excluded. The patients’ ages ranged from 2 to 59 years, with a mean age of 21 years. Spitzoid lesions were classified as nevi if they displayed a symmetric silhouette, bland cytology, and evidence of maturation, with no or only rare mitoses located superficially. Lesions were classified as atypical Spitz tumor if findings were present that raised concerns about the nature of the lesion, such as a mitotic rate of >2/mm2, mitoses in the midportion or deep portion of the lesion, nuclear pleomorphism, and/or incomplete maturation but were seen in context with young age and other histological features still suggesting a possible nevus (eg, relatively symmetric silhouette; sharp demarcation, predominant nested pattern), which led to an overall interpretation of the microscopic findings being insufficient for malignant melanoma. It is likely that most of the lesions classified as atypical Spitz tumor were Spitz nevi with atypical features. However, the term tumor was used to acknowledge some uncertainty about the nature of the lesion—that is, the pathologist was not entirely sure whether the lesion was completely benign. For the category of melanoma, the following parameters were used: if one of the authors thought that the lesion was an unequivocal malignant melanoma on histopathological grounds (eg, asymmetric silhouette with severe atypia and deep mitoses), if cytogenetic results met criteria for melanoma (positive test result
by fluorescence in situ hybridization (FISH), using probes for 6p25, 6q23, 9p21, 11q13, comparative genomic hybridization, or SNP array), or if clinical follow-up indicated a recurrence. Negative (for melanoma) cytogenetic test results were available in 25 lesions, including 20/28 ALK-positive tumors and 5/20 NTRK-positive tumors. Of the 28 ALKpositive tumors of this series, 12 were previously included in a prior study,18 which may have contributed to an enrichment of ALK-positive tumors that does not reflect the general distribution of NTRK1 versus ALK-positive tumors among spitzoid lesions in the general population. Although none of the lesions was known to have a recurrence, clinically follow-up was limited (ranging from a few months up to 3 years).
Microscopic Analysis The tumors were assessed for the following histopathological parameters: ulceration, tumor thickness, Clark level, tumor mitotic rate, cell type, intersecting fascicular growth pattern, inflammation, desmoplastic stroma, and the presence of Kamino bodies, melanin pigment, perineural growth, or lymphatic invasion.
Immunohistochemistry Immunohistochemistry was performed using 5-µm-thick unstained sections of formalin-fixed, paraffin-embedded archival material. For the detection of ALK, monoclonal antibody (mAb) clone D5F3 (Cell Signaling, Danvers, MA) was used. Immunostaining was performed on a Benchmark Ultra instrument (Ventana Medical Systems, Tucson, AZ) using an avidin-biotin–based secondary detection system (iView, Ventana). NTRK1 was detected with mAb clone EP1058Y (Abcam, Cambridge, MA) on a Leica Bond-RX platform (Leica, Buffalo Grove, IL); A polymer-based kit (Refine, Leica) was used as a secondary. Appropriate negative and positive control slides were included in all assays.
Fluorescence In Situ Hybridization FISH was performed in selected cases as previously described18 to confirm ALK or NTRK1 rearrangement. Commercially available break-apart probes were used according to the manufacturer’s protocol (Abbott Molecular, Des Plaines, IL). The probes were hybridized on 5-mm-thick tissue sections. The number and localization of the hybridization signals was assessed in a minimum of 100 interphase nuclei. At least 50% of tumor cells had to show a split signal to report a rearrangement of a kinase. FISH using commercially (Abbott Molecular, Des Plaines, IL) available probes for chromosomes 6p25,
Downloaded from ijs.sagepub.com at UNIV CALIFORNIA SAN DIEGO on February 12, 2016
3
Kiuru et al 6q23, centromere 6 (Cep6), 11q13, and 9p21 was also used for case selection. Cases were excluded, if the FISH result was positive. A positive FISH result required at least 1 of the following chromosomal copy number changes: >29% of cells with >2 copies of 6p25; >55% of cells with more than 2 copies of 6p25 than Cep6; >38% of cells with >2 copies of 11q13; or >42% of cells with fewer copies of 6q23 compared with Cep6. For the 9p21 probe, more than 29% of cells needed to display homozygous deletion of 9p21 for a positive test result.
Statistical Analysis The ANOVA was used to compare continuous data, and χ2 and Fisher’s exact test were used for comparisons of categorical data. A P value
research-article2016
IJSXXX10.1177/1066896916630375International Journal of Surgical PathologyKiuru et al
Original Articles
Spitz Tumors: Comparison of Histological Features in Relationship to Immunohistochemical Staining for ALK and NTRK1
International Journal of Surgical Pathology 1–7 © The Author(s) 2016 Reprints and permissions: sagepub.com/journalsPermissions.nav DOI: 10.1177/1066896916630375 ijs.sagepub.com
Maija Kiuru, MD, PhD1,2, Achim Jungbluth, MD2, Heinz Kutzner, MD3, Thomas Wiesner, MD2, and Klaus J. Busam, MD2
Abstract Spitz tumors are a group of melanocytic neoplasms with distinct morphological features that tend to affect young individuals. Distinguishing benign from malignant Spitz tumors can be challenging, but cytogenetic and molecular tests have contributed to improvements in diagnostic accuracy. Spitz tumors harbor diverse genetic alterations, including mutations in HRAS, loss of BAP1, or kinase fusions in ROS1, NTRK1, ALK, BRAF, and RET genes. Limited data exist on the correlation between histopathological features and kinase fusions. Here, we describe the histopathological features of 105 Spitz tumors (Spitz nevi and atypical Spitz tumors), comparing lesions according to their immunoreactivity for ALK or NTRK1. Intersecting fascicular growth of fusiform melanocytes was seen in all but one ALK-positive tumor (27 of 28 or 96.4%), whereas it was infrequent in NTRK1-positive tumors (5 of 20 or 25.0%) and tumors negative for both ALK and NTRK1 (96.4% vs 25.0% vs 8.7%, P < .0027). There was a trend toward ALK-positive tumors being amelanotic compared with NTRK1-positive tumors and combined ALK-/NTRK1-negative tumors (89.3% vs 45% vs 47.4%, respectively, P = .1023) and lacking epithelioid cell morphology (0% vs 45.0% vs 41%, respectively, P = .6985). In conclusion, this study confirms that although not specific, the growth pattern of intersecting fascicles of amelanotic fusiform melanocytes is strongly associated with ALK expression. Keywords ALK, NTRK1, Spitz nevus, immunohistochemistry
Introduction Spitz tumors are a group of melanocytic tumors typically occurring in young individuals.1,2 They share a number of light microscopic findings, in particular cytological features (eg, plump spindled and large epithelioid melanocytes, multinucleated melanocytes), but also secondary epidermal or stromal changes. Most of the lesions are benign nevi, but rare melanomas with Spitz nevus–like features exist. There is a set of atypical tumors representing lesions for which a definitive distinction between benign and malignant is not always possible.3-6 Because the correct diagnosis of Spitz tumors with atypical features can be very challenging at the light microscopic level, pathologists have explored cytogenetic and molecular assays to help categorize these proliferations as benign versus malignant. Cytogenetic analysis has emerged as a useful ancillary method to improve diagnostic accuracy in difficult cases: the majority of melanomas harbor numerous and complex chromosomal aberrations, whereas
most nevi, including Spitz nevi, do not.7-10 However, it has also become apparent that a few isolated chromosomal copy number gains or losses may be present in benign Spitz nevi, and miscellaneous aberrations can be found in indolent borderline lesions.11-13 Recent studies have documented a diverse genetic background among Spitz tumors, including mutations in Harvey rat sarcoma viral oncogene homolog gene (HRAS),14 loss of BRCA1 associated protein-1 gene (BAP1),15 and kinase fusions in C-ros oncogene 1 receptor tyrosine kinase gene (ROS1), neurotrophic tyrosine kinase receptor type 1 gene (NTRK1), anaplastic lymphoma receptor tyrosine kinase gene (ALK), v-raf murine 1
University of California Davis, Sacramento, CA, USA Memorial Sloan-Kettering Cancer Center, New York, NY, USA 3 Dermatopathologie Friedrichshafen, Friedrichshafen, Germany 2
Corresponding Author: Klaus J. Busam, Department of Pathology, Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, NY 10065, USA. Email: [email protected]
Downloaded from ijs.sagepub.com at UNIV CALIFORNIA SAN DIEGO on February 12, 2016
2
International Journal of Surgical Pathology
sarcoma viral oncogene homolog B gene (BRAF), and ret proto-oncogene (RET) genes.16 Some aberrations are associated with a distinct histopathological phenotype. For example, Spitz nevi with oncogenic mutations in HRAS with concomitant gain of 11p are characterized by singly disposed large epithelioid melanocytes in between thickened collagen bundles.14 Tumors with BAP1 loss are composed of large epithelioid melanocytes characterized by enlarged round to oval nuclei with open chromatin pattern and distinct nucleolus and abundant cytoplasm.15,17 Recently, we found the feature of a plexiform growth pattern of intersecting fascicles of fusiform melanocytes common in Spitz tumors with ALK rearrangement.18 Yeh at al19 subsequently published similar observations. Here, we analyze more systematically the possible association of an intersecting fascicular growth pattern with ALK expression. ALK-positive Spitz tumors were compared with NTRK1-positive lesions and Spitz tumors negative for both ALK and NTRK1.
Materials and Methods Selection of Cases Cases were selected from the files of 2 of the authors (KJB and HK). Lesions diagnosed as a Spitz nevus (n = 38) or atypical Spitz tumor (n = 67) were included in the study. Tumors diagnosed as malignant melanoma were excluded. The patients’ ages ranged from 2 to 59 years, with a mean age of 21 years. Spitzoid lesions were classified as nevi if they displayed a symmetric silhouette, bland cytology, and evidence of maturation, with no or only rare mitoses located superficially. Lesions were classified as atypical Spitz tumor if findings were present that raised concerns about the nature of the lesion, such as a mitotic rate of >2/mm2, mitoses in the midportion or deep portion of the lesion, nuclear pleomorphism, and/or incomplete maturation but were seen in context with young age and other histological features still suggesting a possible nevus (eg, relatively symmetric silhouette; sharp demarcation, predominant nested pattern), which led to an overall interpretation of the microscopic findings being insufficient for malignant melanoma. It is likely that most of the lesions classified as atypical Spitz tumor were Spitz nevi with atypical features. However, the term tumor was used to acknowledge some uncertainty about the nature of the lesion—that is, the pathologist was not entirely sure whether the lesion was completely benign. For the category of melanoma, the following parameters were used: if one of the authors thought that the lesion was an unequivocal malignant melanoma on histopathological grounds (eg, asymmetric silhouette with severe atypia and deep mitoses), if cytogenetic results met criteria for melanoma (positive test result
by fluorescence in situ hybridization (FISH), using probes for 6p25, 6q23, 9p21, 11q13, comparative genomic hybridization, or SNP array), or if clinical follow-up indicated a recurrence. Negative (for melanoma) cytogenetic test results were available in 25 lesions, including 20/28 ALK-positive tumors and 5/20 NTRK-positive tumors. Of the 28 ALKpositive tumors of this series, 12 were previously included in a prior study,18 which may have contributed to an enrichment of ALK-positive tumors that does not reflect the general distribution of NTRK1 versus ALK-positive tumors among spitzoid lesions in the general population. Although none of the lesions was known to have a recurrence, clinically follow-up was limited (ranging from a few months up to 3 years).
Microscopic Analysis The tumors were assessed for the following histopathological parameters: ulceration, tumor thickness, Clark level, tumor mitotic rate, cell type, intersecting fascicular growth pattern, inflammation, desmoplastic stroma, and the presence of Kamino bodies, melanin pigment, perineural growth, or lymphatic invasion.
Immunohistochemistry Immunohistochemistry was performed using 5-µm-thick unstained sections of formalin-fixed, paraffin-embedded archival material. For the detection of ALK, monoclonal antibody (mAb) clone D5F3 (Cell Signaling, Danvers, MA) was used. Immunostaining was performed on a Benchmark Ultra instrument (Ventana Medical Systems, Tucson, AZ) using an avidin-biotin–based secondary detection system (iView, Ventana). NTRK1 was detected with mAb clone EP1058Y (Abcam, Cambridge, MA) on a Leica Bond-RX platform (Leica, Buffalo Grove, IL); A polymer-based kit (Refine, Leica) was used as a secondary. Appropriate negative and positive control slides were included in all assays.
Fluorescence In Situ Hybridization FISH was performed in selected cases as previously described18 to confirm ALK or NTRK1 rearrangement. Commercially available break-apart probes were used according to the manufacturer’s protocol (Abbott Molecular, Des Plaines, IL). The probes were hybridized on 5-mm-thick tissue sections. The number and localization of the hybridization signals was assessed in a minimum of 100 interphase nuclei. At least 50% of tumor cells had to show a split signal to report a rearrangement of a kinase. FISH using commercially (Abbott Molecular, Des Plaines, IL) available probes for chromosomes 6p25,
Downloaded from ijs.sagepub.com at UNIV CALIFORNIA SAN DIEGO on February 12, 2016
3
Kiuru et al 6q23, centromere 6 (Cep6), 11q13, and 9p21 was also used for case selection. Cases were excluded, if the FISH result was positive. A positive FISH result required at least 1 of the following chromosomal copy number changes: >29% of cells with >2 copies of 6p25; >55% of cells with more than 2 copies of 6p25 than Cep6; >38% of cells with >2 copies of 11q13; or >42% of cells with fewer copies of 6q23 compared with Cep6. For the 9p21 probe, more than 29% of cells needed to display homozygous deletion of 9p21 for a positive test result.
Statistical Analysis The ANOVA was used to compare continuous data, and χ2 and Fisher’s exact test were used for comparisons of categorical data. A P value
