GENES, CHROMOSOMES & CANCER 53:738–749 (2014)
Tumor Genetics and Survival of Thymic Neuroendocrine Neoplasms: A Multi-Institutional Clinicopathologic Study Philipp Str€ obel,1* Andreas Zettl,2 Konstantin Shilo,3 Wen-Yu Chuang,4 Andrew G. Nicholson,5 Yoshihiro Matsuno,6 Anthony Gal,7 Rolf Hubert Laeng,8 Peter Engel,9 Carlo Capella,10 Mirella Marino,11 John Kwok-Cheung Chan,12 Andreas Rosenwald,13 William Travis,14 Teri J. Franks,15 David Ellenberger,16 Inga-Marie Schaefer,1,17 and Alexander Marx18 1
Institute of Pathology,University Medical Center G€ ottingen,G€ ottingen,Germany Institute of PathologieViollier,Basle, Switzerland 3 Department of Pathology,Ohio State University,Columbus,OH 4 Department of Pathology,Chang Gung Memorial Hospital and Chang Gung University,Linkou,Taiwan 5 Department of Histopathology,Royal Brompton and Haref|eld NHS FoundationTrust and National Heart and Lung Institute,Imperial College,London,UK 6 Department of Surgical Pathology,Hokkaido University Hospital, Sapporo,Japan 7 Pathology & Laboratory Medicine,Emory University School of Medicine, Atlanta,GA 8 Pathologie L€anggasse,Bern, Switzerland 9 Department of Pathology,Roskilde Hospital,Roskilde,Denmark 10 Department of Human Morphology,Varese,Italy 11 Department of Pathology,Regina Elena National Cancer Institute,Rome,Italy 12 Department of Pathology,Queen Elizabeth Hospital,Hong Kong,China 13 Institute of Pathology,University of W€ urzburg,Germany 14 Department of Pathology,Memorial Sloan Kettering Cancer Center,NewYork,NY 15 Pulmonary & Mediastinal Pathology,The Joint Pathology Center, Silver Spring,MD 16 Department of Medical Statistics,University Medical Center G€ ottingen,G€ ottingen,Germany 17 Department of Pathology,Brigham and Women’s Hospital,Boston,MA 18 Institute of Pathology,University Medical Center Mannheim,University of Heidelberg,Mannheim,Germany 2
Thymic neuroendocrine tumors (TNET) are rare primary epithelial neoplasms of the thymus. This study aimed to determine clinically relevant parameters for their classification and for therapeutic decisions. We performed a comprehensive histological, clinical, and genetic study of 73 TNET cases (13 thymic typical carcinoids [TTC], 40 thymic atypical carcinoids [TAC], and 20 high-grade neuroendocrine carcinomas [HGNEC] of the thymus), contributed by multiple institutions. The mean number of chromosomal imbalances per tumor was 0.8 in TTC (31% aberrant cases) versus 1.1 in TAC (44% aberrant cases) versus 4.7 in HGNEC (75% aberrant cases). Gains of 8q24 (MYC gene locus) were the most frequent alteration and one of the overlapping features between carcinoids and HGNEC. The 5-year survival rates for TTC, TAC, and HGNEC were 100, 60, and 30%. The 10-year survival rates for TTC and TAC were 50 and 30% (P 5 0.002). Predictive mitotic cut-off values for TTC versus TAC were 2.5 per 10 high-power fields (HPF; indicating a higher death rate, P 5 0.062) and 15 per 10 HPF (indicating higher risk of recurrence, P 5 0.036) for separating HGNEC from TAC. We conclude that the current histopathologic classifications of TNET reflect tumor biology and provide important information for C 2014 Wiley Periodicals, Inc. V therapeutic management.
INTRODUCTION
Thymic neuroendocrine tumors (TNET) are rare primary epithelial neoplasms of the thymus composed of neuroendocrine cells (Moran and Suster 2000a; Suster and Moran 2001; Marx et al. 2004; Moran 2005). They comprise less than 5% of all thymic epithelial tumors. First described by Rosai and Higa in 1972, TNET are thought to arise from thymic neuroendocrine cells that constitute a minor cell population scattered in the normal human C 2014 Wiley Periodicals, Inc. V
Supported by: BMBF (A.M. and P.S.), Grant number: 01DL12027; Dr. Mildred Scheel Stiftung f€ ur Krebsforschung (I.M.S.), Grant number: 110822. Disclaimer: The views expressed in this article are those of the author and do not reflect the official policy of the Department of Defense or the United States Government. *Correspondence to: Philipp Str€ obel, MD, Institute of Pathology, University Medical Center G€ ottingen, Robert-Koch-Straße 40, D-37075 G€ ottingen, Germany. E-mail: [email protected] Received 7 January 2014; Accepted 14 April 2014 DOI 10.1002/gcc.22183 Published online 25 April 2014 in Wiley Online Library (wileyonlinelibrary.com).
739
GENETICS OF THYMIC NEUROENDOCRINE NEOPLASMS
TABLE 1. Clinical Details of 58 Patients with Available Survival Data Cases (n 5)
Gender (M/F)
Age (median)
Stage (Masaoka)
TTC
12
2/10
8–78 (56)
TAC
30
21/9
24–80 (55)
LCNEC
14
7/7
16–75 (49)
I:4 III:2 IV:5 n.a.:1 I:3 II:7 III:6 IV:9 n.a.:5 I:3 II:1 III:3 IV:7
SCC
2
1/1
62,63
IV:2
Status at last FU (alive/died)
Median survival (mths)
9/3
126
13/17
52
5/9
21
0/2
13.5
Treatment S: 3 S 1 C: 2 S 1 R: 5 S 1 C 1 R: 2 S: 10 S 1 C: 1 S 1 R: 7 S 1 C 1 R: 11 n.a.: 1 S: 5 S 1 C: 4 S 1 C 1 R: 1 S 1 R: 1 R: 1 n.a.: 2 S 1 C: 1 C: 1
FU, follow-up; TTC, thymic typical carcinoid; TAC, Thymic atypical carcinoid; SCC, small cell carcinoma of the thymus; LCNEC, large cell neuroendocrine carcinoma of the thymus; S, surgery; C, chemotherapy; R, radiotherapy; n.a., data not available.
thymus. (Rosai and Higa 1972) TNET usually manifest in adulthood and show a marked male predominance (Soga et al. 1999; Gaur et al. 2010; Crona et al. 2013). Most patients present with dyspnea or superior vena cava syndrome as a result of the mediastinal mass lesion. TNET can be accompanied by paraneoplastic syndromes such as Cushing syndrome and may be associated with multiple endocrine neoplasia syndrome type 1 (MEN1) (Rosai et al. 1972; Teh et al. 1998). Among MEN1 patients, 8% develop TNET (Gibril et al. 2003), which are histologically indistinguishable from their pulmonary or gastrointestinal counterparts. Following the nomenclature of neuroendocrine tumors (NET) of the lung, the 2004 WHO classification of TNET distinguishes two major groups: well-differentiated and poorly differentiated neuroendocrine carcinomas (Marx et al. 2004). The group of well-differentiated NET consists of the low-grade typical carcinoid (TTC) and intermediate grade atypical carcinoid (TAC). In analogy to diagnostic criteria for pulmonary NET, TTC are devoid of tumor necrosis and show 10 mitoses/10 HPF (Marx et al. 2004). In this multi-institutional study, we examined a series of 73 cases of TNET to determine the prognostic significance of the current WHO classification for risk assessment and therapeutic decision making in patients with TNET.
MATERIAL AND METHODS Patients and Tumor Specimens
Seventy-three cases of TNET (71 primary and one metastasis) were contributed by 13 international cooperating centers. The study had been approved by the local ethical committees of the contributing centers (study no. 117/04, ethical committee, medical faculty, University of W€ urzburg, Germany). A small proportion of the cases had previously been published focusing on different aspects of clinical and biological features of TNET (Goto et al. 2001; Gal et al. 2004). Only cases that were successfully analyzed by comparative genomic hybridization (CGH) were included. All cases were reevaluated by three expert pathologists (A.M., T.F., and K.S.) with regards to tumor subtype, stage, and mitotic count. There were no cases where all three experts disagreed and discrepancies were related only to mitotic counts. In the few cases with discrepancies, the case was reevaluated Genes, Chromosomes & Cancer DOI 10.1002/gcc
Diagnosis
TTC TTC TTC TTC TTC TTC TTC TTC TTC TTC TTC TTC TTC TAC TAC TAC TAC TAC* TAC* TAC TAC TAC TAC TAC TAC TAC TAC TAC TAC TAC* TAC TAC TAC TAC TAC TAC TAC
TAC
No.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37
38
Genes, Chromosomes & Cancer DOI 10.1002/gcc
6
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 3 3 3 3 3 4 4 4 5 5 5 6
Mitotic count/10 HPF
56
8 78 43 53 49 47 50 45 53 46 64 67 45 40 46 52 68 56 35 61 48 24 52 51 43 68 60 44 49 47 54 54 63 36 57 64 60
Age
M
F F M M M M M M M M M M M F M M F M M M M M M M M M M F F M M M M M F M M
Sex
9
3 12 20 16.5 2.9 5 15 8 11 11.5 n.a. 12 n.a. 7 1 6 n.a. 6 18 14 7 6 4 8 13 7 16 n.a. 7 25 4 8 6 8 4 7 8
Size (cm)
3
1 3 n.a 4b 4 1 1 3 1 4b 4 4 n.a. 1 4 4 n.a. 1 4b 2 2 3 3 4 4a 4b 3 n.a. 4b n.a. n.a. 1 2 4 4b n.a. 2
Stage
D
A D A n.a. A D D A A A A A n.a. A D D n.a. D A A A n.a. A D D D D n.a. D n.a. n.a. D A D n.a. A D
Status
YES
NO YES YES NO YES YES YES NO NO NO YES NO n.a. NO YES YES n.a. YES YES NO NO n.a. NO YES YES YES YES n.a. YES n.a. n.a. YES NO YES NO NO YES
Recurrence rev ish enh(2q24, 9q13q22) rev ish dim(3p11p24) rev ish enh(2q, 22), dim(1p, 3) rev ish enh(1,7), dim(13q14q22) rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish enh(10) rev ish no imbalances rev ish dim(1q41q42) rev ish enh(12q23qter) rev ish enh(8), dim(2q22q24) rev ish enh(15q) rev ish dim(3q12q21, 6q) rev ish enh(5q32q33), dim(10q) rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish dim(3q, 4q, 13q) rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish dim(6, 13q) rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish enh(8q24qter), dim(5q21q23, 11q21qter, 13q12q21) rev ish enh(1q), dim(7, 16q)
CGH karyotype
4
2 1 5 3 0 0 0 0 0 0 0 0 0 2 0 1 1 3 1 2 2 0 0 0 0 3 0 0 0 0 3 0 0 0 0 0 4
No of CGH net changes
1
2 0 2 2 0 0 0 0 0 0 0 0 0 2 0 0 1 2 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1
No of gains
3
0 1 3 1 0 0 0 0 0 0 0 0 0 0 0 1 0 1 0 2 1 0 0 0 0 3 0 0 0 0 3 0 0 0 0 0 3
No of losses
TABLE 2. Clinicopathological and Cytogenetic Details of 73 Thymic Neuroendocrine Tumors Including Four Cases* with Associated MEN-1
0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
No of amplifications
740 € STROBEL ET AL.
LCNEC LCNEC
LCNEC
67 68
69
TAC TAC TAC TAC TAC TAC TAC TAC LCNEC LCNEC LCNEC LCNEC LCNEC LCNEC LCNEC LCNEC LCNEC
46 47 48 49 50 51 52 53 54 55 56 57 58 59 60Me 61 62
LCNEC LCNEC LCNEC LCNEC
TAC TAC TAC TAC TAC TAC TAC*
39 40 41 42 43 44 45
63 64 65 66
Diagnosis
No.
94
41 67
20 22 26 26
8 8 8 8 9 9 10 10 11 12 12 13 13 18 17 18 19
6 7 7 7 7 8 8
Mitotic count/10 HPF
52
67 50
53 79 58 70
53 41 59 45 56 58 60 49 35 75 51 16 29 36 36 70 46
80 74 42 68 67 35 58
Age
M
F F
F M F M
M M M M F M F M M M M M M F F M M
F M M M M F M
Sex
13.5
15 n.a.
6 12 n.a. 2.5
n.a. 11 5 n.a. 13 10 n.a. n.a. 20 17 6 9 n.a. 7 7 9 n.a.
11 18.5 12 n.a. 7 7 10
Size (cm)
n.a.
n.a. 4
1 1 4 2
n.a. n.a. 2 n.a. 3 1 n.a. n.a. 4 3 3 1 n.a. 3 3 4 n.a.
2 n.a. 3 n.a. 4 1 2
Stage
A
D D
D D A D
n.a. A A n.a. A A n.a. n.a. D A A n.a. n.a. D D A D
D D D n.a. D D n.a.
Status
YES
YES YES
YES YES NO YES
n.a. YES NO n.a. NO NO n.a. n.a. YES NO NO n.a. n.a. YES YES YES YES
NO YES YES n.a. YES YES NO
Recurrence
CGH karyotype rev ish enh(8q24), dim(11q23qter) rev ish enh(12q23qter) rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish enh(12q24qter), dim(13q21q22) rev ish enh(7, 14q), dim(2p23pter, 5q14q34, 10, 18q21qter) rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish enh(1q), dim(6q25qter) rev ish no imbalances rev ish enh(7, 16p), dim(17p12ter) rev ish no imbalances rev ish enh(2p21pter, 2q) rev ish enh(8) rev ish no imbalances rev ish enh(8) rev ish no imbalances rev ish enh(1q, 7, 9) rev ish enh(1q, 7, 9p23pter, 9q) rev ish no imbalances rev ish enh(7q11q11, 8q21qter, 12q12q14, 17), dim(4q21qter, 5q12q23, 12q24qter, 13q14q31) rev ish dim(13q) rev ish enh(8), dim(13q) rev ish enh(2p23), dim(3p, 5q) rev ish enh(1q, 8q24), dim(4q, 5q15q22, 9p21pter) rev ish no imbalances rev ish enh(1p, 1q12q31, 4p, amp8q22qter), dim(2p, 3, 8p, 8q11q21, 9p, 9q13q32, 10q26qter, 12p) rev ish enh(2p21pter), dim(6q23qter, 18p)
TABLE 2. (Continued)
3
0 14
1 3 3 5
0 0 0 0 2 0 4 0 2 2 0 2 0 5 5 0 9
2 1 0 0 0 2 8
No of CGH net changes
1
0 4
0 2 1 2
0 0 0 0 1 0 3 0 2 2 0 2 0 5 5 0 5
1 1 0 0 0 1 3
No of gains
2
0 9
1 1 2 3
0 0 0 0 1 0 1 0 0 0 0 0 0 5 0 0 4
1 0 0 0 0 1 5
No of losses
0
0 1
0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0
No of amplifications
GENETICS OF THYMIC NEUROENDOCRINE NEOPLASMS
741
Genes, Chromosomes & Cancer DOI 10.1002/gcc
0 5 10
5
0 13 21
8
0 4 14
10
1 9 6 16
YES 4b SCC 73
114
62
F
n.a.
D
YES D 4 12 SCC 72
105
63
M
4 LCNEC 71
100
37
M
15
D
YES
rev ish enh(2p, 6p21pter, 7q32qter, amp8q24, 11q12q13, 11q23qter, 12q24qter, 14q22qter), dim(3p11p13, 3q, 4, 5q14q23, 8p, 9p21pter, 10q21, 13q) rev ish enh(1q, 3p, 12p, 14), dim(3q, 4q32qter, 6q, 9p, 10, 13, 17q22qter, 18) rev ish enh(1q, 5p15, 5q34qter, 6p21, 8, 9q34, 12q24qter, 14q22q24, 15q22qter, 16p, 17), dim(2, 3p11p12, 3q12q13, 4, 13q14qter, 16q12q21) rev ish enh(1p33pter, 8q, 9q, 12q24qter, 14), dim(3p11p14, 3p22p24, 8p21p22, 11q14q22, 13q14q31, Xp21) n.a. 4b LCNEC 70
94
66
F
11
n.a.
No of amplifications No of losses No of gains Stage Diagnosis No.
Mitotic count/10 HPF
Age
Sex
Size (cm)
Status
Recurrence
CGH karyotype
No of CGH net changes TABLE 2. (Continued)
Genes, Chromosomes & Cancer DOI 10.1002/gcc
MEN-1, multiple neuroendocrine neoplasia type 1; TTC, thymic typical carcinoid; TAC, thymic atypical carcinoid; LCNEC, large cell neuroendocrine carcinoma; SCC, small cell carcinoma; Me, metastasis; A, alive; D, died; n.a., not available.
€ STROBEL ET AL.
742
and accepted if at least two reviewers agreed. Diagnosis was based on the clinical setting and imaging (tumor mass in the mediastinum, no evidence of pulmonary primary), morphology, and immunohistochemistry (protocols available upon request). All cases showed strong expression of at least one of the following neuroendocrine markers: chromogranin, synaptophysin, neuron-specific enolase, and CD56 in at least 50% of tumor cells. Cases were classified based on hematoxylin-eosin (H&E)stained slides using the 2004 WHO classification criteria (Marx et al. 2004). Tumor stage was determined following the modified Masaoka–Koga classification (Koga et al. 1994): I, encapsulated tumor; II, infiltration of mediastinal fat; III, infiltration of neighboring organs; IVa, pleural or pericardial dissemination; and IVb, lymphatic or hematogenous metastases. Mitotic counts were determined in 10 HPF on H&E stained sections, using a Zeiss Axioplan microscope (403 objective, field-of-view diameter of 0.65 mm, resulting in 1 HPF 5 0.332 mm2). For maximal reproducibility, only mitoses with clearly visible chromosomal spindles in areas of viable tumor were counted (Travis et al. 1998). Clinical Characteristics and Statistical Analysis
Clinical information on history, type, and duration of symptoms at initial presentation, and type of treatment (surgery/chemotherapy/radiation) was available for 58 patients. Follow-up information was obtained by reviewing the clinical records, contacting the attending physicians, or reviewing autopsy files. In these archival retrospective cases collected over long time periods, detailed data on treatment (such as regimen or dosage of chemotherapy or radiation therapy etc.) were often incomplete or missing and, therefore, omitted. CGH Analysis
In all cases, DNA was extracted from formalinfixed, paraffin-embedded tissue blocks using a phenol-chloroform-based extraction method as described previously (Zettl et al. 2000). The length of the extracted DNA was analyzed on a 0.7% agarose gel with suitable DNA length standards. CGH was performed according to published protocols (Zettl et al. 2000). Signals were visualized with a Zeiss Axiophot fluorescence microscope and analyzed with ISIS digital image analysis system (MetaSystems, Altlussheim, Germany). Ratio values of 1.25 and 0.8 were used as upper and lower thresholds for identification of chromosomal gains
743
GENETICS OF THYMIC NEUROENDOCRINE NEOPLASMS
Figure 1. Correlation between (A) histology of thymic typical carcinoid (TTC), thymic atypical carcinoid (TAC), large cell (LCNEC) and small cell carcinoma (SCC), and (B) Masaoka tumor stage and survival.
TABLE 3. Overlapping Genetic Alterations Between Carcinoids (Typical and Atypical) and High -Grade Neuroendocrine Carcinomas (HGNEC) Frequency Chr.
All tumors (n 5 73)
Carcinoids (n 5 53)
HGNEC (n 5 20)
1p 1qa 2p 2q 2q 3p 3q 4q 5q 5q 6q 7p 7q 8p 8q 9q 10p 10q 11q 12q 13q 14q 15q 16p 16q 18q
3 (4%) 9 (12%) 3 (4%) 3 (4%) 2 (3%) 7 (10%) 7 (10%) 6 (8%) 2 (3%) 6 (8%) 5 (7%) 4 (5%) 6 (8%) 5 (7%) 12 (16%) 5 (7%) 2 (3%) 5 (7%) 3 (4%) 6 (8%) 12 (16%) 5 (7%) 2 (3%) 2 (3%) 2 (3%) 2 (3%)
1 (2%) 3 (6%) 1 (2%) 2 (4%) 1 (2%) 2 (4%) 3 (6%) 1 (2%) 1 (2%) 2 (4%) 3 (6%) 2 (4%) 2 (4%) 1 (2%) 3 (6%) 1 (2%) 1 (2%) 2 (4%) 2 (4%) 3 (6%) 5 (9%) 1 (2%) 1 (2%) 1 (2%) 1 (2%) 1 (2%)
2 (10%) 6 (30%) 2 (10%) 1 (5%) 1 (5%) 5 (25%) 4 (20%) 5 (25%) 1 (5%) 4 (20%) 2 (10%) 2 (10%) 4 (20%) 4 (20%) 9 (45%) 4 (20%) 1 (5%) 3 (15%) 1 (5%) 3 (15%) 7 (35%) 4 (20%) 1 (5%) 1 (5%) 1 (5%) 1 (5%)
Alteration type Gain Gain Loss Gain Loss Loss Loss Loss Gain Loss Loss Gain Gain Gain Gain Gain Loss Loss Loss Gain Loss Gain Gain Gain Loss Loss
a
Alterations identified in at least 10 percent of all cases are marked in bold.
or losses, respectively. High-level amplification was defined as an overrepresentation of genetic material with the fluorescence ratio values exceeding 2.0 or based on the observation of strong focal signals in the fluorescein isothiocyanate fluorescence.
Fluorescence in Situ Hybridization Analysis
Fluorescence in situ hybridization (FISH) was performed in three LCNEC cases as previously described (Manner et al. 2010). Low-level amplification was defined by gene signal to centromere Genes, Chromosomes & Cancer DOI 10.1002/gcc
744
€ STROBEL ET AL.
curves were made using the log-rank test as well as multivariate Cox regression to adjust for possible confounders. For binary outcomes, groups were compared by logistic regression and using the v2-test. Diagnostic tests to define cut-offs were performed using the Youden index. A P-value of 25% of tumor cells. Presence of >9 gene signals or tight clusters of signals was considered high-level amplification.
Statistical Analysis
Statistical analyses were performed using SAS version 9.3 (SAS Institute Inc. Cary, NC), Statistica 10 (StatSoft, Hamburg, Germany) and R Stat 2.15 (R Foundation, Vienna, Austria). Prognostic factors for survival were analyzed according to Kaplan–Meier, and comparisons between Genes, Chromosomes & Cancer DOI 10.1002/gcc
The clinical data are summarized in Table 1. Of the 73 cases of TNET (72 primaries and one metastasis), 13 (18%) were classified as TTC, 40 (55%) as TAC, and 20 (27%) as HGNEC. The patients’ median age was 52 years (range, 8–80). Fifty four patients were male and 18 female. The mean tumor size at diagnosis was 8 cm (range, 1–25). Among the 52 cases with information on histological stage, 10 cases were in Masaoka–Koga stage I, eight cases stage II, 11 cases stage III, and 23 stage IV. Six patients (two with TTC, three with TAC, and one with HGNEC) presented with paraneoplastic Cushing syndrome, and one patient with TAC had paraneoplastic hypercalcemia. Four patients (all TAC) had evidence of MEN1 syndrome. Among the 58 patients with available survival data, 31 (53%) had died (overall survival: 1–228 months after initial diagnosis), 28 patients were still alive (follow-up period: 6–180 months) (data summarized in Tables 1 and 2). Among 12 patients with TTC, nine (75%) were still alive with a median follow-up of 54 months and three patients (25%) had died (median survival 126 months). Of the 30 patients with TAC, 13 (43%) patients were alive with a median follow-up of 58 months and 17 (57%) had died (median survival 52 months). Of the 16 patients with HGNEC, five (31%) were alive with a median follow-up of 50 months and 11 (69%) had died with a median survival of 10 months. On multivariate analysis, only stage (P < 0.02), recurrence (P < 0.02), and number of mitoses/histological tumor type (P < 0.03) correlated significantly with survival (Fig. 1), whereas age, sex, and tumor size did not. Among Masaoka–Koga tumor stage groups, statistically significant difference was observed only between stage I and IV (P 5 0.03). The 5- and 10-year survival rates were 100 and 60% for TTC, and 50 and 30% for
GENETICS OF THYMIC NEUROENDOCRINE NEOPLASMS
745
Figure 3. Grafic representation of overlapping genetic features among thymic typical carcinoids (TTC), thymic atypical carcinoids (TAC) and high-grade neuroendocrine carcinomas (HGNEC). Lines represent individual cases; only cases with overlapping alterations found in at least two groups are shown. Green bars: chromosomal gains, red bars: chromosomal losses.
TAC. The 5-year survival rate for LCNEC was 30 and 0% for SCC. Cytogenetic Profiles of TNET Subgroups
The genetic alterations found in TNET are detailed in Tables 2 and 3 and in Figures 2 and 3. There was an incremental increase of genetic alterations from TCC to TAC, LNCEC, and SCC (Fig. 2). The mean number of aberrations per tumor was 0.8 (0.5 gains, 0.4 losses) in TTC, 1.1 (0.5 gains, 0.7 losses) in TAC, 4.7 (2.3 gains, 2.6 losses, 0.1 amplifications) in LCNEC, and 15.5 (nine gains, 6.5 losses) in SCC. Chromosomal gains in carcinoids were mainly located on 1q (four cases), 2q (two cases), 6q (three cases), 7 (three cases), 8 (three cases), and 12q (three cases). Chromosomal losses in carcinoids were mainly located on 2q (two cases), 3q (three cases), 5q (two cases), 11q (two cases), and 13q (five cases). The average number of chromosomal alterations in HGNEC was significantly higher than in carcinoids (2.3 6 2.0 vs. 1.0 6 1.2 (P < 0.05)). Frequently overlapping genetic alterations found both in carcinoids and HGNEC included gains on 1q and 8q and losses on 3 and 13q (Table 3 and Fig. 3). In addition, HGNEC showed recurrent gains at 2p, 9p, and 17q and losses at 4p, 8p, 9p, and 18p that were not observed in carcinoids. FISH analysis performed in three HGNEC cases suggested amplification of MYC as a potential target of the recurrent gains on 8q24 (Fig. 4). Gains of 8q24 occurred significantly more frequently in HGNEC than in carcinoids (P < 0.05). Cut-off Values for Clinical Risk Assessment
Using the number of mitoses/10 HPF and the total number of chromosomal imbalances per tumor versus tumor recurrence and patient death to define clinical risk groups, the following cut-off values were identified: a mitotic count of 15/10 HPF correlated with a higher rate of recurrence
Figure 4. FISH showing high-level gene amplification (green signal) of MYC relative to the centromeric probe (red signal) in a case of a large cell neuroendocrine carcinoma that had shown gain of 8q24 on comparative genomic hybridization.
(P 5 0.036), and a mitotic count 2.5 correlated with a higher death rate (P 50.062). A cut-off of 2.5 chromosomal imbalances (mean value per tumor) correlated with a higher rate of recurrence (P 5 0.111) and a higher death rate (P 5 0.063) (Youden index, v2 test) (Table 4 and Fig. 5). DISCUSSION
In this retrospective and multi-institutional, multinational collaborative study, we tried to establish criteria for clinical decision making in patients with TNET. Our results show clearcut histomorphologic, genetic, and clinical differences between the four tumor subtypes recognized in the current WHO classification. TNET showed an incremental loss of differentiation, accompanied by an increase in mitotic activity and chromosomal aberrations ranging from the low-grade TTC to intermediate grade TAC to the high-grade LCNEC to SCC. Genes, Chromosomes & Cancer DOI 10.1002/gcc
€ STROBEL ET AL.
746
TABLE 4. Pathological and Genetic Risk Groups of Thymic Neuroendocrine Tumors Mitotic count per 10 HPF (mean value)
Morphology (defining features) CARCINOIDS
TTC TAC HGNEC
LCNEC
SCC
Polygonal cells arranged in ribbons, festoons, solid nests and rosette-like glands, 6 necrosis, low to moderate mitotic activity Low -mitotic activity, no necrosis Low to moderate mitotic activity, 6 necrosis High -grade tumors with some neuroendocrine morphology, necrosis often extensive, mitotic activity usually high Large cells with cytologic and histologic neuroendocrine feaures (e.g., trabecular growth, palisading), frequent nucleoli, chromatin vesicular or finely granular, high -mitotic activity, usually extensive necrosis, Small cells with ill-defined cell borders and finely granular nuclear chromatin, scant cytoplasm, nucleoli are inconspicuoua or absent, high mitotic activity, often extensive necrosis
0–10 (5.1)
Tumor Genetics and Survival of Thymic Neuroendocrine Neoplasms: A Multi-Institutional Clinicopathologic Study Philipp Str€ obel,1* Andreas Zettl,2 Konstantin Shilo,3 Wen-Yu Chuang,4 Andrew G. Nicholson,5 Yoshihiro Matsuno,6 Anthony Gal,7 Rolf Hubert Laeng,8 Peter Engel,9 Carlo Capella,10 Mirella Marino,11 John Kwok-Cheung Chan,12 Andreas Rosenwald,13 William Travis,14 Teri J. Franks,15 David Ellenberger,16 Inga-Marie Schaefer,1,17 and Alexander Marx18 1
Institute of Pathology,University Medical Center G€ ottingen,G€ ottingen,Germany Institute of PathologieViollier,Basle, Switzerland 3 Department of Pathology,Ohio State University,Columbus,OH 4 Department of Pathology,Chang Gung Memorial Hospital and Chang Gung University,Linkou,Taiwan 5 Department of Histopathology,Royal Brompton and Haref|eld NHS FoundationTrust and National Heart and Lung Institute,Imperial College,London,UK 6 Department of Surgical Pathology,Hokkaido University Hospital, Sapporo,Japan 7 Pathology & Laboratory Medicine,Emory University School of Medicine, Atlanta,GA 8 Pathologie L€anggasse,Bern, Switzerland 9 Department of Pathology,Roskilde Hospital,Roskilde,Denmark 10 Department of Human Morphology,Varese,Italy 11 Department of Pathology,Regina Elena National Cancer Institute,Rome,Italy 12 Department of Pathology,Queen Elizabeth Hospital,Hong Kong,China 13 Institute of Pathology,University of W€ urzburg,Germany 14 Department of Pathology,Memorial Sloan Kettering Cancer Center,NewYork,NY 15 Pulmonary & Mediastinal Pathology,The Joint Pathology Center, Silver Spring,MD 16 Department of Medical Statistics,University Medical Center G€ ottingen,G€ ottingen,Germany 17 Department of Pathology,Brigham and Women’s Hospital,Boston,MA 18 Institute of Pathology,University Medical Center Mannheim,University of Heidelberg,Mannheim,Germany 2
Thymic neuroendocrine tumors (TNET) are rare primary epithelial neoplasms of the thymus. This study aimed to determine clinically relevant parameters for their classification and for therapeutic decisions. We performed a comprehensive histological, clinical, and genetic study of 73 TNET cases (13 thymic typical carcinoids [TTC], 40 thymic atypical carcinoids [TAC], and 20 high-grade neuroendocrine carcinomas [HGNEC] of the thymus), contributed by multiple institutions. The mean number of chromosomal imbalances per tumor was 0.8 in TTC (31% aberrant cases) versus 1.1 in TAC (44% aberrant cases) versus 4.7 in HGNEC (75% aberrant cases). Gains of 8q24 (MYC gene locus) were the most frequent alteration and one of the overlapping features between carcinoids and HGNEC. The 5-year survival rates for TTC, TAC, and HGNEC were 100, 60, and 30%. The 10-year survival rates for TTC and TAC were 50 and 30% (P 5 0.002). Predictive mitotic cut-off values for TTC versus TAC were 2.5 per 10 high-power fields (HPF; indicating a higher death rate, P 5 0.062) and 15 per 10 HPF (indicating higher risk of recurrence, P 5 0.036) for separating HGNEC from TAC. We conclude that the current histopathologic classifications of TNET reflect tumor biology and provide important information for C 2014 Wiley Periodicals, Inc. V therapeutic management.
INTRODUCTION
Thymic neuroendocrine tumors (TNET) are rare primary epithelial neoplasms of the thymus composed of neuroendocrine cells (Moran and Suster 2000a; Suster and Moran 2001; Marx et al. 2004; Moran 2005). They comprise less than 5% of all thymic epithelial tumors. First described by Rosai and Higa in 1972, TNET are thought to arise from thymic neuroendocrine cells that constitute a minor cell population scattered in the normal human C 2014 Wiley Periodicals, Inc. V
Supported by: BMBF (A.M. and P.S.), Grant number: 01DL12027; Dr. Mildred Scheel Stiftung f€ ur Krebsforschung (I.M.S.), Grant number: 110822. Disclaimer: The views expressed in this article are those of the author and do not reflect the official policy of the Department of Defense or the United States Government. *Correspondence to: Philipp Str€ obel, MD, Institute of Pathology, University Medical Center G€ ottingen, Robert-Koch-Straße 40, D-37075 G€ ottingen, Germany. E-mail: [email protected] Received 7 January 2014; Accepted 14 April 2014 DOI 10.1002/gcc.22183 Published online 25 April 2014 in Wiley Online Library (wileyonlinelibrary.com).
739
GENETICS OF THYMIC NEUROENDOCRINE NEOPLASMS
TABLE 1. Clinical Details of 58 Patients with Available Survival Data Cases (n 5)
Gender (M/F)
Age (median)
Stage (Masaoka)
TTC
12
2/10
8–78 (56)
TAC
30
21/9
24–80 (55)
LCNEC
14
7/7
16–75 (49)
I:4 III:2 IV:5 n.a.:1 I:3 II:7 III:6 IV:9 n.a.:5 I:3 II:1 III:3 IV:7
SCC
2
1/1
62,63
IV:2
Status at last FU (alive/died)
Median survival (mths)
9/3
126
13/17
52
5/9
21
0/2
13.5
Treatment S: 3 S 1 C: 2 S 1 R: 5 S 1 C 1 R: 2 S: 10 S 1 C: 1 S 1 R: 7 S 1 C 1 R: 11 n.a.: 1 S: 5 S 1 C: 4 S 1 C 1 R: 1 S 1 R: 1 R: 1 n.a.: 2 S 1 C: 1 C: 1
FU, follow-up; TTC, thymic typical carcinoid; TAC, Thymic atypical carcinoid; SCC, small cell carcinoma of the thymus; LCNEC, large cell neuroendocrine carcinoma of the thymus; S, surgery; C, chemotherapy; R, radiotherapy; n.a., data not available.
thymus. (Rosai and Higa 1972) TNET usually manifest in adulthood and show a marked male predominance (Soga et al. 1999; Gaur et al. 2010; Crona et al. 2013). Most patients present with dyspnea or superior vena cava syndrome as a result of the mediastinal mass lesion. TNET can be accompanied by paraneoplastic syndromes such as Cushing syndrome and may be associated with multiple endocrine neoplasia syndrome type 1 (MEN1) (Rosai et al. 1972; Teh et al. 1998). Among MEN1 patients, 8% develop TNET (Gibril et al. 2003), which are histologically indistinguishable from their pulmonary or gastrointestinal counterparts. Following the nomenclature of neuroendocrine tumors (NET) of the lung, the 2004 WHO classification of TNET distinguishes two major groups: well-differentiated and poorly differentiated neuroendocrine carcinomas (Marx et al. 2004). The group of well-differentiated NET consists of the low-grade typical carcinoid (TTC) and intermediate grade atypical carcinoid (TAC). In analogy to diagnostic criteria for pulmonary NET, TTC are devoid of tumor necrosis and show 10 mitoses/10 HPF (Marx et al. 2004). In this multi-institutional study, we examined a series of 73 cases of TNET to determine the prognostic significance of the current WHO classification for risk assessment and therapeutic decision making in patients with TNET.
MATERIAL AND METHODS Patients and Tumor Specimens
Seventy-three cases of TNET (71 primary and one metastasis) were contributed by 13 international cooperating centers. The study had been approved by the local ethical committees of the contributing centers (study no. 117/04, ethical committee, medical faculty, University of W€ urzburg, Germany). A small proportion of the cases had previously been published focusing on different aspects of clinical and biological features of TNET (Goto et al. 2001; Gal et al. 2004). Only cases that were successfully analyzed by comparative genomic hybridization (CGH) were included. All cases were reevaluated by three expert pathologists (A.M., T.F., and K.S.) with regards to tumor subtype, stage, and mitotic count. There were no cases where all three experts disagreed and discrepancies were related only to mitotic counts. In the few cases with discrepancies, the case was reevaluated Genes, Chromosomes & Cancer DOI 10.1002/gcc
Diagnosis
TTC TTC TTC TTC TTC TTC TTC TTC TTC TTC TTC TTC TTC TAC TAC TAC TAC TAC* TAC* TAC TAC TAC TAC TAC TAC TAC TAC TAC TAC TAC* TAC TAC TAC TAC TAC TAC TAC
TAC
No.
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37
38
Genes, Chromosomes & Cancer DOI 10.1002/gcc
6
1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 2 2 2 2 2 2 2 2 2 2 3 3 3 3 3 4 4 4 5 5 5 6
Mitotic count/10 HPF
56
8 78 43 53 49 47 50 45 53 46 64 67 45 40 46 52 68 56 35 61 48 24 52 51 43 68 60 44 49 47 54 54 63 36 57 64 60
Age
M
F F M M M M M M M M M M M F M M F M M M M M M M M M M F F M M M M M F M M
Sex
9
3 12 20 16.5 2.9 5 15 8 11 11.5 n.a. 12 n.a. 7 1 6 n.a. 6 18 14 7 6 4 8 13 7 16 n.a. 7 25 4 8 6 8 4 7 8
Size (cm)
3
1 3 n.a 4b 4 1 1 3 1 4b 4 4 n.a. 1 4 4 n.a. 1 4b 2 2 3 3 4 4a 4b 3 n.a. 4b n.a. n.a. 1 2 4 4b n.a. 2
Stage
D
A D A n.a. A D D A A A A A n.a. A D D n.a. D A A A n.a. A D D D D n.a. D n.a. n.a. D A D n.a. A D
Status
YES
NO YES YES NO YES YES YES NO NO NO YES NO n.a. NO YES YES n.a. YES YES NO NO n.a. NO YES YES YES YES n.a. YES n.a. n.a. YES NO YES NO NO YES
Recurrence rev ish enh(2q24, 9q13q22) rev ish dim(3p11p24) rev ish enh(2q, 22), dim(1p, 3) rev ish enh(1,7), dim(13q14q22) rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish enh(10) rev ish no imbalances rev ish dim(1q41q42) rev ish enh(12q23qter) rev ish enh(8), dim(2q22q24) rev ish enh(15q) rev ish dim(3q12q21, 6q) rev ish enh(5q32q33), dim(10q) rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish dim(3q, 4q, 13q) rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish dim(6, 13q) rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish enh(8q24qter), dim(5q21q23, 11q21qter, 13q12q21) rev ish enh(1q), dim(7, 16q)
CGH karyotype
4
2 1 5 3 0 0 0 0 0 0 0 0 0 2 0 1 1 3 1 2 2 0 0 0 0 3 0 0 0 0 3 0 0 0 0 0 4
No of CGH net changes
1
2 0 2 2 0 0 0 0 0 0 0 0 0 2 0 0 1 2 1 0 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1
No of gains
3
0 1 3 1 0 0 0 0 0 0 0 0 0 0 0 1 0 1 0 2 1 0 0 0 0 3 0 0 0 0 3 0 0 0 0 0 3
No of losses
TABLE 2. Clinicopathological and Cytogenetic Details of 73 Thymic Neuroendocrine Tumors Including Four Cases* with Associated MEN-1
0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
No of amplifications
740 € STROBEL ET AL.
LCNEC LCNEC
LCNEC
67 68
69
TAC TAC TAC TAC TAC TAC TAC TAC LCNEC LCNEC LCNEC LCNEC LCNEC LCNEC LCNEC LCNEC LCNEC
46 47 48 49 50 51 52 53 54 55 56 57 58 59 60Me 61 62
LCNEC LCNEC LCNEC LCNEC
TAC TAC TAC TAC TAC TAC TAC*
39 40 41 42 43 44 45
63 64 65 66
Diagnosis
No.
94
41 67
20 22 26 26
8 8 8 8 9 9 10 10 11 12 12 13 13 18 17 18 19
6 7 7 7 7 8 8
Mitotic count/10 HPF
52
67 50
53 79 58 70
53 41 59 45 56 58 60 49 35 75 51 16 29 36 36 70 46
80 74 42 68 67 35 58
Age
M
F F
F M F M
M M M M F M F M M M M M M F F M M
F M M M M F M
Sex
13.5
15 n.a.
6 12 n.a. 2.5
n.a. 11 5 n.a. 13 10 n.a. n.a. 20 17 6 9 n.a. 7 7 9 n.a.
11 18.5 12 n.a. 7 7 10
Size (cm)
n.a.
n.a. 4
1 1 4 2
n.a. n.a. 2 n.a. 3 1 n.a. n.a. 4 3 3 1 n.a. 3 3 4 n.a.
2 n.a. 3 n.a. 4 1 2
Stage
A
D D
D D A D
n.a. A A n.a. A A n.a. n.a. D A A n.a. n.a. D D A D
D D D n.a. D D n.a.
Status
YES
YES YES
YES YES NO YES
n.a. YES NO n.a. NO NO n.a. n.a. YES NO NO n.a. n.a. YES YES YES YES
NO YES YES n.a. YES YES NO
Recurrence
CGH karyotype rev ish enh(8q24), dim(11q23qter) rev ish enh(12q23qter) rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish enh(12q24qter), dim(13q21q22) rev ish enh(7, 14q), dim(2p23pter, 5q14q34, 10, 18q21qter) rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish no imbalances rev ish enh(1q), dim(6q25qter) rev ish no imbalances rev ish enh(7, 16p), dim(17p12ter) rev ish no imbalances rev ish enh(2p21pter, 2q) rev ish enh(8) rev ish no imbalances rev ish enh(8) rev ish no imbalances rev ish enh(1q, 7, 9) rev ish enh(1q, 7, 9p23pter, 9q) rev ish no imbalances rev ish enh(7q11q11, 8q21qter, 12q12q14, 17), dim(4q21qter, 5q12q23, 12q24qter, 13q14q31) rev ish dim(13q) rev ish enh(8), dim(13q) rev ish enh(2p23), dim(3p, 5q) rev ish enh(1q, 8q24), dim(4q, 5q15q22, 9p21pter) rev ish no imbalances rev ish enh(1p, 1q12q31, 4p, amp8q22qter), dim(2p, 3, 8p, 8q11q21, 9p, 9q13q32, 10q26qter, 12p) rev ish enh(2p21pter), dim(6q23qter, 18p)
TABLE 2. (Continued)
3
0 14
1 3 3 5
0 0 0 0 2 0 4 0 2 2 0 2 0 5 5 0 9
2 1 0 0 0 2 8
No of CGH net changes
1
0 4
0 2 1 2
0 0 0 0 1 0 3 0 2 2 0 2 0 5 5 0 5
1 1 0 0 0 1 3
No of gains
2
0 9
1 1 2 3
0 0 0 0 1 0 1 0 0 0 0 0 0 5 0 0 4
1 0 0 0 0 1 5
No of losses
0
0 1
0 0 0 0
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0
No of amplifications
GENETICS OF THYMIC NEUROENDOCRINE NEOPLASMS
741
Genes, Chromosomes & Cancer DOI 10.1002/gcc
0 5 10
5
0 13 21
8
0 4 14
10
1 9 6 16
YES 4b SCC 73
114
62
F
n.a.
D
YES D 4 12 SCC 72
105
63
M
4 LCNEC 71
100
37
M
15
D
YES
rev ish enh(2p, 6p21pter, 7q32qter, amp8q24, 11q12q13, 11q23qter, 12q24qter, 14q22qter), dim(3p11p13, 3q, 4, 5q14q23, 8p, 9p21pter, 10q21, 13q) rev ish enh(1q, 3p, 12p, 14), dim(3q, 4q32qter, 6q, 9p, 10, 13, 17q22qter, 18) rev ish enh(1q, 5p15, 5q34qter, 6p21, 8, 9q34, 12q24qter, 14q22q24, 15q22qter, 16p, 17), dim(2, 3p11p12, 3q12q13, 4, 13q14qter, 16q12q21) rev ish enh(1p33pter, 8q, 9q, 12q24qter, 14), dim(3p11p14, 3p22p24, 8p21p22, 11q14q22, 13q14q31, Xp21) n.a. 4b LCNEC 70
94
66
F
11
n.a.
No of amplifications No of losses No of gains Stage Diagnosis No.
Mitotic count/10 HPF
Age
Sex
Size (cm)
Status
Recurrence
CGH karyotype
No of CGH net changes TABLE 2. (Continued)
Genes, Chromosomes & Cancer DOI 10.1002/gcc
MEN-1, multiple neuroendocrine neoplasia type 1; TTC, thymic typical carcinoid; TAC, thymic atypical carcinoid; LCNEC, large cell neuroendocrine carcinoma; SCC, small cell carcinoma; Me, metastasis; A, alive; D, died; n.a., not available.
€ STROBEL ET AL.
742
and accepted if at least two reviewers agreed. Diagnosis was based on the clinical setting and imaging (tumor mass in the mediastinum, no evidence of pulmonary primary), morphology, and immunohistochemistry (protocols available upon request). All cases showed strong expression of at least one of the following neuroendocrine markers: chromogranin, synaptophysin, neuron-specific enolase, and CD56 in at least 50% of tumor cells. Cases were classified based on hematoxylin-eosin (H&E)stained slides using the 2004 WHO classification criteria (Marx et al. 2004). Tumor stage was determined following the modified Masaoka–Koga classification (Koga et al. 1994): I, encapsulated tumor; II, infiltration of mediastinal fat; III, infiltration of neighboring organs; IVa, pleural or pericardial dissemination; and IVb, lymphatic or hematogenous metastases. Mitotic counts were determined in 10 HPF on H&E stained sections, using a Zeiss Axioplan microscope (403 objective, field-of-view diameter of 0.65 mm, resulting in 1 HPF 5 0.332 mm2). For maximal reproducibility, only mitoses with clearly visible chromosomal spindles in areas of viable tumor were counted (Travis et al. 1998). Clinical Characteristics and Statistical Analysis
Clinical information on history, type, and duration of symptoms at initial presentation, and type of treatment (surgery/chemotherapy/radiation) was available for 58 patients. Follow-up information was obtained by reviewing the clinical records, contacting the attending physicians, or reviewing autopsy files. In these archival retrospective cases collected over long time periods, detailed data on treatment (such as regimen or dosage of chemotherapy or radiation therapy etc.) were often incomplete or missing and, therefore, omitted. CGH Analysis
In all cases, DNA was extracted from formalinfixed, paraffin-embedded tissue blocks using a phenol-chloroform-based extraction method as described previously (Zettl et al. 2000). The length of the extracted DNA was analyzed on a 0.7% agarose gel with suitable DNA length standards. CGH was performed according to published protocols (Zettl et al. 2000). Signals were visualized with a Zeiss Axiophot fluorescence microscope and analyzed with ISIS digital image analysis system (MetaSystems, Altlussheim, Germany). Ratio values of 1.25 and 0.8 were used as upper and lower thresholds for identification of chromosomal gains
743
GENETICS OF THYMIC NEUROENDOCRINE NEOPLASMS
Figure 1. Correlation between (A) histology of thymic typical carcinoid (TTC), thymic atypical carcinoid (TAC), large cell (LCNEC) and small cell carcinoma (SCC), and (B) Masaoka tumor stage and survival.
TABLE 3. Overlapping Genetic Alterations Between Carcinoids (Typical and Atypical) and High -Grade Neuroendocrine Carcinomas (HGNEC) Frequency Chr.
All tumors (n 5 73)
Carcinoids (n 5 53)
HGNEC (n 5 20)
1p 1qa 2p 2q 2q 3p 3q 4q 5q 5q 6q 7p 7q 8p 8q 9q 10p 10q 11q 12q 13q 14q 15q 16p 16q 18q
3 (4%) 9 (12%) 3 (4%) 3 (4%) 2 (3%) 7 (10%) 7 (10%) 6 (8%) 2 (3%) 6 (8%) 5 (7%) 4 (5%) 6 (8%) 5 (7%) 12 (16%) 5 (7%) 2 (3%) 5 (7%) 3 (4%) 6 (8%) 12 (16%) 5 (7%) 2 (3%) 2 (3%) 2 (3%) 2 (3%)
1 (2%) 3 (6%) 1 (2%) 2 (4%) 1 (2%) 2 (4%) 3 (6%) 1 (2%) 1 (2%) 2 (4%) 3 (6%) 2 (4%) 2 (4%) 1 (2%) 3 (6%) 1 (2%) 1 (2%) 2 (4%) 2 (4%) 3 (6%) 5 (9%) 1 (2%) 1 (2%) 1 (2%) 1 (2%) 1 (2%)
2 (10%) 6 (30%) 2 (10%) 1 (5%) 1 (5%) 5 (25%) 4 (20%) 5 (25%) 1 (5%) 4 (20%) 2 (10%) 2 (10%) 4 (20%) 4 (20%) 9 (45%) 4 (20%) 1 (5%) 3 (15%) 1 (5%) 3 (15%) 7 (35%) 4 (20%) 1 (5%) 1 (5%) 1 (5%) 1 (5%)
Alteration type Gain Gain Loss Gain Loss Loss Loss Loss Gain Loss Loss Gain Gain Gain Gain Gain Loss Loss Loss Gain Loss Gain Gain Gain Loss Loss
a
Alterations identified in at least 10 percent of all cases are marked in bold.
or losses, respectively. High-level amplification was defined as an overrepresentation of genetic material with the fluorescence ratio values exceeding 2.0 or based on the observation of strong focal signals in the fluorescein isothiocyanate fluorescence.
Fluorescence in Situ Hybridization Analysis
Fluorescence in situ hybridization (FISH) was performed in three LCNEC cases as previously described (Manner et al. 2010). Low-level amplification was defined by gene signal to centromere Genes, Chromosomes & Cancer DOI 10.1002/gcc
744
€ STROBEL ET AL.
curves were made using the log-rank test as well as multivariate Cox regression to adjust for possible confounders. For binary outcomes, groups were compared by logistic regression and using the v2-test. Diagnostic tests to define cut-offs were performed using the Youden index. A P-value of 25% of tumor cells. Presence of >9 gene signals or tight clusters of signals was considered high-level amplification.
Statistical Analysis
Statistical analyses were performed using SAS version 9.3 (SAS Institute Inc. Cary, NC), Statistica 10 (StatSoft, Hamburg, Germany) and R Stat 2.15 (R Foundation, Vienna, Austria). Prognostic factors for survival were analyzed according to Kaplan–Meier, and comparisons between Genes, Chromosomes & Cancer DOI 10.1002/gcc
The clinical data are summarized in Table 1. Of the 73 cases of TNET (72 primaries and one metastasis), 13 (18%) were classified as TTC, 40 (55%) as TAC, and 20 (27%) as HGNEC. The patients’ median age was 52 years (range, 8–80). Fifty four patients were male and 18 female. The mean tumor size at diagnosis was 8 cm (range, 1–25). Among the 52 cases with information on histological stage, 10 cases were in Masaoka–Koga stage I, eight cases stage II, 11 cases stage III, and 23 stage IV. Six patients (two with TTC, three with TAC, and one with HGNEC) presented with paraneoplastic Cushing syndrome, and one patient with TAC had paraneoplastic hypercalcemia. Four patients (all TAC) had evidence of MEN1 syndrome. Among the 58 patients with available survival data, 31 (53%) had died (overall survival: 1–228 months after initial diagnosis), 28 patients were still alive (follow-up period: 6–180 months) (data summarized in Tables 1 and 2). Among 12 patients with TTC, nine (75%) were still alive with a median follow-up of 54 months and three patients (25%) had died (median survival 126 months). Of the 30 patients with TAC, 13 (43%) patients were alive with a median follow-up of 58 months and 17 (57%) had died (median survival 52 months). Of the 16 patients with HGNEC, five (31%) were alive with a median follow-up of 50 months and 11 (69%) had died with a median survival of 10 months. On multivariate analysis, only stage (P < 0.02), recurrence (P < 0.02), and number of mitoses/histological tumor type (P < 0.03) correlated significantly with survival (Fig. 1), whereas age, sex, and tumor size did not. Among Masaoka–Koga tumor stage groups, statistically significant difference was observed only between stage I and IV (P 5 0.03). The 5- and 10-year survival rates were 100 and 60% for TTC, and 50 and 30% for
GENETICS OF THYMIC NEUROENDOCRINE NEOPLASMS
745
Figure 3. Grafic representation of overlapping genetic features among thymic typical carcinoids (TTC), thymic atypical carcinoids (TAC) and high-grade neuroendocrine carcinomas (HGNEC). Lines represent individual cases; only cases with overlapping alterations found in at least two groups are shown. Green bars: chromosomal gains, red bars: chromosomal losses.
TAC. The 5-year survival rate for LCNEC was 30 and 0% for SCC. Cytogenetic Profiles of TNET Subgroups
The genetic alterations found in TNET are detailed in Tables 2 and 3 and in Figures 2 and 3. There was an incremental increase of genetic alterations from TCC to TAC, LNCEC, and SCC (Fig. 2). The mean number of aberrations per tumor was 0.8 (0.5 gains, 0.4 losses) in TTC, 1.1 (0.5 gains, 0.7 losses) in TAC, 4.7 (2.3 gains, 2.6 losses, 0.1 amplifications) in LCNEC, and 15.5 (nine gains, 6.5 losses) in SCC. Chromosomal gains in carcinoids were mainly located on 1q (four cases), 2q (two cases), 6q (three cases), 7 (three cases), 8 (three cases), and 12q (three cases). Chromosomal losses in carcinoids were mainly located on 2q (two cases), 3q (three cases), 5q (two cases), 11q (two cases), and 13q (five cases). The average number of chromosomal alterations in HGNEC was significantly higher than in carcinoids (2.3 6 2.0 vs. 1.0 6 1.2 (P < 0.05)). Frequently overlapping genetic alterations found both in carcinoids and HGNEC included gains on 1q and 8q and losses on 3 and 13q (Table 3 and Fig. 3). In addition, HGNEC showed recurrent gains at 2p, 9p, and 17q and losses at 4p, 8p, 9p, and 18p that were not observed in carcinoids. FISH analysis performed in three HGNEC cases suggested amplification of MYC as a potential target of the recurrent gains on 8q24 (Fig. 4). Gains of 8q24 occurred significantly more frequently in HGNEC than in carcinoids (P < 0.05). Cut-off Values for Clinical Risk Assessment
Using the number of mitoses/10 HPF and the total number of chromosomal imbalances per tumor versus tumor recurrence and patient death to define clinical risk groups, the following cut-off values were identified: a mitotic count of 15/10 HPF correlated with a higher rate of recurrence
Figure 4. FISH showing high-level gene amplification (green signal) of MYC relative to the centromeric probe (red signal) in a case of a large cell neuroendocrine carcinoma that had shown gain of 8q24 on comparative genomic hybridization.
(P 5 0.036), and a mitotic count 2.5 correlated with a higher death rate (P 50.062). A cut-off of 2.5 chromosomal imbalances (mean value per tumor) correlated with a higher rate of recurrence (P 5 0.111) and a higher death rate (P 5 0.063) (Youden index, v2 test) (Table 4 and Fig. 5). DISCUSSION
In this retrospective and multi-institutional, multinational collaborative study, we tried to establish criteria for clinical decision making in patients with TNET. Our results show clearcut histomorphologic, genetic, and clinical differences between the four tumor subtypes recognized in the current WHO classification. TNET showed an incremental loss of differentiation, accompanied by an increase in mitotic activity and chromosomal aberrations ranging from the low-grade TTC to intermediate grade TAC to the high-grade LCNEC to SCC. Genes, Chromosomes & Cancer DOI 10.1002/gcc
€ STROBEL ET AL.
746
TABLE 4. Pathological and Genetic Risk Groups of Thymic Neuroendocrine Tumors Mitotic count per 10 HPF (mean value)
Morphology (defining features) CARCINOIDS
TTC TAC HGNEC
LCNEC
SCC
Polygonal cells arranged in ribbons, festoons, solid nests and rosette-like glands, 6 necrosis, low to moderate mitotic activity Low -mitotic activity, no necrosis Low to moderate mitotic activity, 6 necrosis High -grade tumors with some neuroendocrine morphology, necrosis often extensive, mitotic activity usually high Large cells with cytologic and histologic neuroendocrine feaures (e.g., trabecular growth, palisading), frequent nucleoli, chromatin vesicular or finely granular, high -mitotic activity, usually extensive necrosis, Small cells with ill-defined cell borders and finely granular nuclear chromatin, scant cytoplasm, nucleoli are inconspicuoua or absent, high mitotic activity, often extensive necrosis
0–10 (5.1)
