ORIGINAL ARTICLE
Lymphoepithelioma-like Hepatocellular Carcinoma An Uncommon Variant of Hepatocellular Carcinoma With Favorable Outcome Anthony W.H. Chan, FRCPA,* Joanna H.M. Tong, PhD,*wz Yi Pan, MMed,*z Stephen L. Chan, FRCP,wy Grace L.H. Wong, MD,w8 Vincent W.S. Wong, MD,w8 Paul B.S. Lai, FRCS,z and Ka-Fai To, FRCPA*wz
Abstract: Lymphoepithelioma-like hepatocellular carcinoma (LEL-HCC) is an uncommon variant of HCC with only 22 cases reported in the literature. To better determine the incidence, clinicopathologic features, prognostic significance, and molecular pathogenesis of LEL-HCC, we presented the largest series of LEL-HCC from a 9-year retrospective cohort of patients with HCC undergoing surgical resection. LEL-HCC was identified in 20 patients (4.9%). Compared with patients having HCC without significant tumor-infiltrating lymphocyte (TIL), patients with LEL-HCC had a relatively lower frequency of male sex (P = 0.022), tended to present at early-stage disease (80.0% vs. 56.3% as AJCC stage I, P = 0.037; 100% vs. 77.3% as BCLC stage 0/A, P = 0.010), and all harbored a solitary tumor only (P = 0.006). There was no significant difference in the age at presentation, underlying chronic liver disease, cirrhotic background, serum a-fetoprotein level, tumor size, histologic grade, and frequencies of vascular invasion. Most of the TILs in LELHCC were cytotoxic T lymphocytes. None of the LEL-HCCs From the Departments of *Anatomical and Cellular Pathology; yClinical Oncology, State Key Laboratory in Oncology in South China, Prince of Wales Hospital; wInstitute of Digestive Disease, Partner State Key Laboratory of Digestive Disease; zLi Ka Shing Institute of Health Science, Sir Y.K. Pao Cancer Center; 8Department of Medicine and Therapeutics; and zDepartment of Surgery, Division of Hepatobiliary and Pancreatic Surgery, The Chinese University of Hong Kong, Hong Kong. Drafting of the manuscript: A.W.H.C. Critical revision of the manuscript: S.L.C., G.L.H.W., V.W.S.W., P.B.S.L. Pathologic examinations: A.W.H.C., J.H.M.T., K.F.T. Molecular studies: J.H.M.T., Y.P. Clinical managements of the patients: S.L.C., G.L.H.W., V.W.S.W., P.B.S.L. Conflicts of Interest and Source of Funding: G.L.H.W. has served as an advisory committee member for Otsuka and Gilead. She has also served as a speaker for Bristol-Myers Squibb, Echosens, Furui, and Otsuka. V.W.S.W. has served as an advisory committee member for Roche, Novartis, Gilead, and Otsuka. He has also served as a speaker for Bristol-Myers Squibb, Roche, Novartis, Abbott Diagnostics, and Echosens. Correspondence: Ka-Fai To, FRCPA, Department of Anatomical and Cellular Pathology, Prince of Wales Hospital, The Chinese University of Hong Kong, Ngan Shing Street, Shatin, NT, Hong Kong (e-mail: [email protected]). Supplemental Digital Content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Website, www.ajsp.com. Copyright r 2014 Wolters Kluwer Health, Inc. All rights reserved.
304 | www.ajsp.com
was associated with Epstein-Barr virus. LEL-HCC was associated with better overall (5-y survival: 94.1% vs. 63.9%; P = 0.007) and progression-free (5-y survival: 87.8% vs. 46.6%; P = 0.002) survivals compared with HCC without significant TIL. The multivariate analysis revealed that LEL-HCC was an independent prognostic factor for overall and progression-free survivals. The adjusted hazard ratio of cancer death and tumor progression for LEL-HCC was 0.12 (P = 0.037) and 0.14 (P = 0.002), respectively. LEL-HCC did not differ in frequencies of microsatellite instability, BRAF mutation, and DNA hypermethylation. In brief, LEL-HCC is a distinct uncommon variant of HCC characterized by dense cytotoxic T-cell infiltration and favorable prognosis. Key Words: hepatocellular carcinoma, lymphoepithelioma, lymphoid stroma, tumor-infiltrating lymphocyte, microsatellite instability, BRAF, hypermethylation (Am J Surg Pathol 2015;39:304–312)
L
ymphoepithelioma is characterized by syncytial sheets of undifferentiated carcinoma cells intermingled with dense tumor-infiltrating lymphocytes (TILs) and was originally described in the nasopharynx. The tumors, which develop outside of the nasopharynx and exhibit the morphology of lymphoepithelioma, are designated as lymphoepithelioma-like carcinomas (LELCs). LELCs arising from the head and neck region, thymus, lung, and stomach have variable associations with Epstein-Barr virus (EBV), whereas those occurring in the skin and urogenital tract have no known association with EBV. Primary hepatic LELC is rare and composed of 2 distinct entities: lymphoepithelioma-like cholangiocarcinoma (LEL-CC) and lymphoepithelioma-like hepatocellular carcinoma (LEL-HCC). LEL-CC is frequently associated with EBV and characterized by marked female predominance, favorable overall survival, and distinctively frequent DNA hypermethylation.1 In contrast, LEL-HCC has not been well characterized yet. Only 22 cases have been reported in the literature until now.2–9 There are no universally accepted diagnostic criteria for LEL-HCC. Only 18.2% of LEL-HCCs reported in the literature exhibit typical histologic features of LELC, whereas remaining cases are conventional HCCs with significant Am J Surg Pathol
Volume 39, Number 3, March 2015
Am J Surg Pathol
Volume 39, Number 3, March 2015
TILs.4–7 Wada et al3 reported the largest series of 11 cases of LEL-HCC and were the first to use quantitative criteria (>100 TILs in ten 400 high-power fields) to report definite significant TILs in LEL-HCC. Previous diagnostic terms—HCC with lymphocytic infiltration and HCC with lymphoid stroma—are probably more appropriate than LEL-HCC, but the latter is the current terminology used in the latest World Health Organization classification.10 Because of the rarity of LEL-HCC, information regarding clinicopathologic features and clinical outcome is limited. The association between LELHCC and EBV is also controversial. In the digestive tract, some carcinomas contain dense TILs. One of these carcinomas is colorectal medullary carcinoma, which is a histologic variant of colorectal carcinoma, characterized by poor differentiation, prominent TILs, and favorable prognosis.11 Medullary carcinoma is associated with microsatellite instability (MSI), BRAF mutation, and CpG island hypermethylation.12 The association between these molecular alterations and LEL-HCC is unknown. In this study, we report 20 patients with LEL-HCC from a retrospective cohort of 409 patients undergoing surgical resection for primary HCC. Our aim was to determine the incidence, clinicopathologic features, prognostic significance, and molecular changes (MSI, BRAF mutation, and CpG island hypermethylation) of LEL-HCC.
MATERIALS AND METHODS Patients The cohort has recruited patients who received curative surgery for primary HCC from January 2002 to December 2010 at Prince of Wales Hospital, Hong Kong. The study has been approved by the institutional review board. Baseline clinical and laboratory parameters (namely, complete blood count, prothrombin time and international normalized ratio, liver and renal biochemistries, hepatitis B/C [HBV/HCV] serology, a-fetoprotein [AFP], etc.) were retrieved and reviewed from the hospital database. The blood tests, excluding viral serology, were taken within 1 month before surgery. The Child-Pugh score and the Model for End-stage Liver Disease score were calculated on the basis of the baseline clinical and biochemical data as described. Tumor staging was determined by the seventh edition of American Joint Committee on Cancer (AJCC) Tumor-Node-Metastasis (TNM) staging system and the Barcelona Clinic Liver Cancer (BCLC) staging system. The neutrophil to lymphocyte ratio was calculated by dividing the absolute neutrophil count by the absolute lymphocyte count. The platelet to lymphocyte ratio was estimated as the absolute platelet count divided by the absolute lymphocyte count. The prognostic nutritional index was the sum of serum albumin level (g/L) and 5 the absolute lymphocyte count (109/L). All of the patients underwent surveillance in the clinic with regular ultrasonography and measurement of AFP according to the local practice. The duration of Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
Lymphoepithelioma-like HCC
follow-up was defined from the date of operation to the last follow-up before we analyzed the data or to the date of death. Overall survival was defined from the time of surgery to the time of HCC-related death. Progression-free survival was defined from the time of surgery to the time of radiologic evidence of tumor relapse or metastasis.
Histology, Immunohistochemistry, and EBVencoded RNA In Situ Hybridization The histologic diagnosis of HCC was reviewed and confirmed by 2 pathologists (A.W.H.C. and K.F.T.). LEL-HCC diagnosis was made by the criteria previously described by Wada et al3 (>100 in ten 400 high-power fields). Hepatitic activity in the background was evaluated by Ishak modification for hepatic activity index for scoring of necroinflammatory activity.13 Fatty liver was defined as the presence of steatosis in Z5% of hepatocytes. The degree of steatosis (grade 0, 20 mg/L)
All (n = 409) (n [%])
HCC Without TIL (n = 389) (n [%])
LEL-HCC (n = 20) (n [%])
P
346 (84.6) 55.7 ± 11.0 197 (48.2) 150 (36.7) 343 (83.9) 16 (3.9) 166 (40.6) 244 (59.7)
333 (85.6) 55.6 ± 10.9 186 (47.8) 145 (37.3) 326 (83.8) 16 (4.1) 155 (39.8) 236 (60.7)
13 (65.0) 57.5 ± 12.3 11 (55.0) 5 (25.0) 17 (85.0) 0 (0.0) 11 (55.0) 8 (40.0)
0.022 0.444 0.552 0.263 1.000 1.000 0.202 0.066
52 ± 35 351 (85.8) 94 (23.0) 36 (8.8) 108 (26.4)
53 ± 35 334 (85.9) 94 (24.2) 34 (8.7) 105 (27.0)
38 ± 21 17 (85.0) 0 (0.0) 2 (10.0) 3 (15.0)
0.060 1.000 0.006 0.692 0.235 0.056
235 73 95 6
219 72 93 5
16 1 2 1
(57.5) (17.8) (23.2) (1.5)
(56.3) (18.5) (23.9) (1.3)
(80.0) (5.3) (10.5) (5.3) 0.037
235 (57.5) 174 (42.5)
219 (56.3) 170 (43.7)
16 (80.0) 4 (20.0)
320 (78.2) 89 (21.8) 68.0 [5.0-705.0] 240 (58.7)
300 (77.1) 89 (22.9) 64.5 [5.0-729.0] 227 (58.4)
20 (100.0) 0 (0.0) 203.0 [22.2-466.5] 13 (65.0)
0.010 0.528 0.257
IQR indicates interquartile range.
TABLE 2. Liver Biochemistry, Hepatitic Activity, Systemic Inflammatory Indices, and Molecular Studies of HCC With or Without Significant TIL Serum albumin (g/L) Serum total bilirubin (mmol/L) Serum alanine transaminase (IU/L) Serum alkaline phosphatase (IU/L) Model for end-stage liver disease (MELD) Ishak activity score Neutrophil count ( 109/L) Lymphocyte count ( 109/L) Neutrophil to lymphocyte ratio Platelet to lymphocyte ratio Prognostic nutritional index
HCC Without TIL (n = 389)
LEL-HCC (n = 20)
P
39.2 ± 5.2 12.8 ± 10.4 59.1 ± 44.0 98.4 ± 44.8 8.7 ± 2.5 2.7 ± 1.7 4.0 ± 1.7 1.5 ± 0.6 3.0 ± 1.9 128.3 ± 72.9 46.6 ± 6.6
41.5 ± 2.8 9.2 ± 2.9 45.4 ± 22.1 84.9 ± 38.0 7.9 ± 1.3 2.4 ± 1.6 3.7 ± 1.5 1.6 ± 0.4 2.6 ± 2.1 117.1 ± 42.4 48.9 ± 4.0
0.064 0.130 0.180 0.199 0.182 0.440 0.509 0.757 0.490 0.516 0.205
HCC Without TIL (n = 38)
LEL-HCC (n = 19)
65.8 21.1 13.2 0.0
57.8 21.1 21.1 0.0
1.000
32.2 41.2 43.8 22.2 60.5 44.7 15.8
57.9 37.5 33.3 21.1 68.4 47.4 26.3
0.070 0.680 0.470 0.920 0.560 0.851 0.342
MSI status MSS MSI-L MSI-H BRAF mutation Hypermethylation p15 p16 RASSF1A GSTP1 1 or more of the above genes 2 or more of the above genes 3 or more of the above genes
0.732
MSI-H indicates high-frequency MSI; MSI-L, low-frequency MSI; MSS, microsatellite stable.
Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
www.ajsp.com |
307
Chan et al
and sixty-two patients (64.1%) were alive at the end of the follow-up period. The 2- and 5-year overall survival rates were 80.8% and 65.3%, respectively. One hundred and ninety-eight patients (48.4%) did not suffer from tumor relapse during the follow-up period. The 2- and 5-year
Am J Surg Pathol
Volume 39, Number 3, March 2015
progression-free survival rates were 59.5% and 48.5%, respectively. LEL-HCC was associated with better overall (2-y survival rate of 100% vs. 79.9%; 5-y survival rate of 94.1% vs. 63.9%; P = 0.007) and progression-free (2-y survival rate
FIGURE 1. Pathology of LEL-HCC. A and B, The tumor was composed of trabeculae and sheets of tumor cells with dense lymphocytic infiltrate (hematoxylin and eosin). The TILs are composed of mainly cytotoxic T cells (CD3+/CD8+). C, CD20. D, CD3. E, CD8. F, Foxp3.
308 | www.ajsp.com
Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
Am J Surg Pathol
Volume 39, Number 3, March 2015
of 100% vs. 57.5%; 5-y survival rate of 87.8% vs. 46.6%; P = 0.002) survivals compared with HCC without significant TIL (Fig. 2). On univariate analysis, the following parameters were also associated with unfavorable overall and progression-free survivals: older age, larger tumor size, moderate/poor histologic differentiation, tumor multiplicity, the presence of vascular invasion, advanced AJCC and BCLC stages, and marked elevated serum AFP of >1000 mg/L (Tables 3, 4). The multivariate analysis on these parameters revealed that LEL-HCC was one of the independent prognostic factors for overall and progressionfree survivals (Tables 3, 4). The adjusted hazard ratio of cancer death and tumor progression for LEL-HCC was 0.12 (95% confidence interval: 0.02-0.88, P = 0.037) and 0.14 (95% confidence interval: 0.04-0.48, P = 0.002), respectively. The presence of significant TIL in HCC could reduce the risk of cancer death and progression by over 85%.
MSI, BRAF Mutation, and CpG Island Hypermethylation of LEL-HCC Nineteen of 20 LEL-HCCs (1 case was not included because of insufficient material) and 38 age/sex/stagematched HCCs without significant TIL were subjected for molecular studies. Eleven patients with LEL-HCC (57.8%) were classified as microsatellite stable, 4 patients (21.1%) as low-frequency MSI, and 4 patients (21.1%) as high-frequency MSI. The frequencies of MSI status were not different from those of matched HCC without significant TIL (P = 0.732). None of the 19 LEL-HCCs nor 38 HCCs without significant TIL showed BRAF mutation (P = 1.000). Four genes (p15, p16, RASSF1A, and GSTP1), commonly silenced by hypermethylation in HCC, were examined to compare methylation status between LEL-HCC and HCC without significant TIL.15 The frequencies of hypermethylation of these 4 genes were not different between both groups. Results of molecular studies are summarized in Table 2.
DISCUSSION LEL-HCC is a distinct uncommon variant of HCC. Since it was firstly described by Shirabe and colleagues in 1995, a total of 22 cases have been reported in the literature.2–9 Herein, we present 20 patients with LEL-HCC from a 9-year retrospective cohort. The frequency of LEL-HCC in 2 independent large series of HCC patients in Japan (n = 163) and France (n = 162) was 6.7% and 3.6%, respectively,3,4 and comparable to that of our cohort (4.9%, n = 409). In all 42 patients with LEL-HCC, the mean age of presentation was 58 years (range, 39 to 81 y), with a male to female ratio of 2.8:1. Half of these patients had cirrhosis (50.0%). HBV and HCV accounted for 54.8% and 38.1% of the underlying etiology of these patients. The mean size of the tumor was 32 mm (range, 15 to 95 mm). The majority of the tumors were solitary (87.8%) and free of vascular invasion (84.7%). About three quarter (76.9%) of patients presented at AJCC stage I disease. Only a single case was reported to be associated with EBV.5 The clinical outcome was good with 82.9% of patients alive without disease. Patients who suffered from Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
Lymphoepithelioma-like HCC
disease recurrence and cancer death account for 9.8% and 9.8%, respectively. Our cohort further consolidated the favorable prognosis of LEL-HCC by showing the presence of significant TIL in HCC as an independent prognostic factor for overall and progression-free survivals in multivariate analyses. In many but not all malignant tumors, those containing prominent infiltration of cytotoxic T-cell TILs tend to have a better prognosis.16,17 The antitumor effect of cytotoxic T-cell TILs contributes to less aggressive behavior of LEL-HCC. The pathogenesis of LEL-HCC remains unknown. Firstly, EBV has been shown to be associated with many LELCs. Almost all nasopharyngeal carcinomas harbor EBV.18 Pulmonary and gastric LELCs in Asian populations show strong associations with EBV.19,20 One type of primary hepatic LELC, LEL-CC, is also frequently associated with EBV (69.6%).1 Is there any association between LEL-HCC and EBV? Only 1 case of LEL-HCC has been reported.5 Even in our region endemic with EBV, none of 20 LEL-HCCs in our cohort shows the presence of EBV. Although several groups have demonstrated the presence of EBV-DNA in HBV/HCVrelated HCC tumors, no direct evidence supporting the causative role of EBV in HCC development is available yet.21 The role of EBV in carcinogenesis of HCC, particularly LEL-HCC, remains disputable. Secondly, malignant transformation of hepatocellular adenoma (HCA) is a rare but well-documented phenomenon with a reported frequency of 4.2%.22 One histologic subtype of HCA is inflammatory HCA (IHCA), which is featured by the presence of focal or dense intratumoral inflammatory infiltrate, sinusoidal dilatation and peliosis, and frequent ductular reaction. Immunoreactivity to C-reactive protein and SAA is characteristic. IHCA is frequently associated with obesity, glycogenosis, and fatty liver disease.23 Could LEL-HCC represent a malignant counterpart of IHCA? In our cohort of LEL-HCC, none of the tumor shows significant peliosis, ductular reaction, or immunoreactivity to SAA. Although the frequency of fatty liver in the nontumorous livers of LEL-HCC is higher than those of HCC without significant TIL, the difference is statistically insignificant. Moreover, none of the patients with LEL-HCC has underlying glycogenosis. Apart from intratumoral inflammatory infiltrate linking up these 2 primary hepatocellular neoplasms, LEL-HCC and IHCA appear not to have any association in clinical manifestations, pathologic features, and tumorigenesis. Thirdly, prominent lymphocytic infiltration in colorectal carcinoma is associated with MSI, BRAF mutation, and DNA hypermethylation.12 Could these molecular alterations explain significant TILs in LEL-HCC? We did not demonstrate any association of LEL-HCC with MSI, BRAF mutation, and DNA hypermethylation. The exact carcinogenesis of LEL-HCC requires further studies. The diagnosis of LEL-HCC is usually straightforward by pathologic examination. However, a few differential diagnoses, including LEL-CC, metastatic LELC from other sites, and lymphoma, may need to be considered. Firstly, LEL-CC is a rare variant of cholangiocarcinoma, with only www.ajsp.com |
309
Chan et al
Am J Surg Pathol
Volume 39, Number 3, March 2015
FIGURE 2. A, Kaplan-Meier overall survival curves and (B) Kaplan-Meier progression-free survival curves for LEL-HCC and HCC without significant TIL.
23 cases reported in the literature. It is primary hepatic adenocarcinoma with varied glandular differentiation, dense lymphoplasmacytic infiltrate, and variable expression of biliary-type cytokeratins (CK7 and CK19) and stemness markers (CD133 and EpCAM).1 Although a small subset of HCC may express CK19, CD133, and/or EpCAM, LEL-CC does not demonstrate any hepatocellular differentiation
evident by typical histology and negative immunoreactivity to hepatic markers (HepPar-1 and Arginase-1).24 Secondly, metastatic LELC of other sites can be differentiated from LEL-HCC by clinicoradiologic correlations. Moreover, metastatic LELC does not express hepatic markers by immunohistochemistry. Thirdly, lymphoma, either primary or metastatic, may mimic LEL-HCC. The presence of
TABLE 3. Univariate and Multivariate Analyses on the Overall Survival of HCC Patients Univariate Patient factors Male sex Age >60 y HBV HCV Fatty liver Cirrhosis Tumor factors Tumor size (mm) Moderate/poor differentiation Multiple tumors Major vessel invasion Microvascular invasion AJCC stage II vs. I III vs. I IV vs. I BCLC stage B vs. 0/A C/D vs. 0/A Serum AFP (>1000 mg/L) LEL-HCC
Multivariate* P
Hazard Ratio
95% CI
1.602 1.487 0.720 1.271 0.783 1.345
0.953-2.693 1.058-2.089 0.479-1.082 0.594-2.721 0.556-1.103 0.959-1.887
0.076 0.022 0.114 0.537 0.162 0.086
1.009 2.186 3.024 2.587 4.353
1.006-1.014 1.236-3.867 2.150-4.253 1.625-4.119 3.125-6.064
4.450 5.664 22.268 2.245 3.334 1.812 0.108
P
Hazard Ratio
95% CI
1.308
0.916-1.868
0.140
< 0.001 0.007 < 0.001 < 0.001 < 0.001
1.002 1.144
0.997-1.007 0.628-2.084
0.398 0.661
2.884-6.886 3.798-8.447 9.328-53.156
< 0.001 < 0.001 < 0.001
4.375 4.814 27.196
2.766-6.921 3.012-7.695 10.987-67.321
1.256-4.015 2.289-4.857 1.263-2.599 0.015-0.770
0.006 < 0.001 0.001 0.026
1.150 0.120
0.785-1.684 0.017-0.876
< 0.001 < 0.001 < 0.001
0.473 0.037
*Tumor multiplicity, vascular invasion, and BCLC stage were excluded in the multivariate analysis to minimize multicollinearity. CI indicates confidence interval.
310 | www.ajsp.com
Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
Am J Surg Pathol
Volume 39, Number 3, March 2015
Lymphoepithelioma-like HCC
TABLE 4. Univariate and Multivariate Analyses on the Progression-free Survival of HCC Patients Univariate Patient factors Male sex Age >60 y HBV HCV Fatty liver Cirrhosis Tumor factors Tumor size (mm) Moderate/poor differentiation Multiple tumors Major vessel invasion Microvascular invasion AJCC stage II vs. I III vs. I IV vs. I BCLC stage B vs. 0/A C/D vs. 0/A Serum AFP (>1000 mg/L) LEL-HCC
Multivariate* P
Hazard Ratio
95% CI
1.318 1.396 0.781 1.502 0.788 1.253
0.885-1.963 1.048-1.860 0.547-1.114 0.817-2.763 0.593-1.046 0.947-1.660
0.174 0.023 0.173 0.191 0.099 0.115
1.007 2.041 2.617 2.230 3.156
1.004-1.011 1.299-3.206 1.948-3.516 1.475-3.371 2.376-4.192
3.172 4.211 14.985 2.138 2.769 1.998 0.199
P
Hazard Ratio
95% CI
1.247
0.919-1.664
0.160
< 0.001 0.002 < 0.001 < 0.001 < 0.001
1.000 1.283
0.996-1.004 0.802-2.053
0.815 0.298
2.225-4.522 3.052-5.810 6.490-34.602
< 0.001 < 0.001 < 0.001
2.846 3.728 37.194
1.973-4.106 2.586-5.414 14.459-95.681
1.292-3.538 1.984-3.864 1.473-2.709 0.064-0.623
0.003 < 0.001 < 0.001 0.006
1.510 0.141
1.097-2.078 0.042-0.476
< 0.001 < 0.001 < 0.001
0.011 0.002
*Tumor multiplicity, vascular invasion, and BCLC stage were excluded in the multivariate analysis to minimize multicollinearity. CI indicates confidence interval.
architectural/cytologic atypia in lesional hepatocytes and the absence of an aberrant immunophenotype of the lymphoid component support the diagnosis of LEL-HCC. Our study is limited by its retrospective nature and relatively small sample size. Completeness of clinical and pathologic data and long duration of follow-up (up to 13.1 y) can partly compensate for this limitation. Moreover, LEL-HCC is a rare variant of HCC. Our cohort has already accounted for 42.5% of all LEL-HCCs in the literature. Another limitation is that the mechanism of recruitment of TILs in LEL-HCC remains unanswered. Our study provides circumstantial evidence that the presence of TIL in LEL-HCC is unlikely correlated with necroinflammatory activity of the underlying chronic hepatitis and systemic inflammatory responses. We also could not establish any association of LEL-HCC with EBV, IHCA, MSI, BRAF mutation, and DNA hypermethylation. A multicentric prospective study recruiting more LEL-HCC patients is needed to further investigate the underlying mechanism of recruitment of TILs in LEL-HCC. In conclusion, LEL-HCC is a distinct uncommon variant of HCC characterized by dense cytotoxic T-cell infiltration and favorable prognosis. The association with EBV is only very weak. The mechanism of recruitment of dense TILs requires further studies.
3. 4. 5. 6. 7. 8. 9. 10.
11.
12.
REFERENCES 1. Chan AW, Tong JH, Sung MY, et al. Epstein-Barr virus-associated lymphoepithelioma-like cholangiocarcinoma: a rare variant of intrahepatic cholangiocarcinoma with favourable outcome. Histopathology. 2014;65:674–683. 2. Shirabe K, Matsumata T, Maeda T, et al. A long-term surviving patient with hepatocellular carcinoma including lymphocytes
Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
13. 14.
infiltration—a clinicopathological study. Hepatogastroenterology. 1995;42:996–1001. Wada Y, Nakashima O, Kutami R, et al. Clinicopathological study on hepatocellular carcinoma with lymphocytic infiltration. Hepatology. 1998;27:407–414. Emile JF, Adam R, Sebagh M, et al. Hepatocellular carcinoma with lymphoid stroma: a tumour with good prognosis after liver transplantation. Histopathology. 2000;37:523–529. Si MW, Thorson JA, Lauwers GY, et al. Hepatocellular lymphoepithelioma-like carcinoma associated with epstein barr virus: a hitherto unrecognized entity. Diagn Mol Pathol. 2004;13:183–189. Chen CJ, Jeng LB, Huang SF. Lymphoepithelioma-like hepatocellular carcinoma. Chang Gung Med J. 2007;30:172–177. Nemolato S, Fanni D, Naccarato AG, et al. Lymphoepitelioma-like hepatocellular carcinoma: a case report and a review of the literature. World J Gastroenterol. 2008;14:4694–4696. Park HS, Jang KY, Kim YK, et al. Hepatocellular carcinoma with massive lymphoid infiltration: a regressing phenomenon? Pathol Res Pract. 2009;205:648–652. Shinoda M, Kadota Y, Tsujikawa H, et al. Lymphoepithelioma-like hepatocellular carcinoma: a case report and a review of the literature. World J Surg Oncol. 2013;11:97. Theise ND, Curado MP, Franceshi S, et al. Hepatocellular carcinoma. In: Bosman FT, Carneiro F, Hruban RH, et al, eds. WHO Classification of Tumours of the Digestive System. Lyon: International Agency for Research on Cancer; 2010:205–216. Hamiton SR, Bosman FT, Boffetta P, et al. Carcinoma of the colon and rectum. In: Bosman FT, Carneiro F, Hruban RH, et al, eds. WHO Classification of Tumours of the Digestive System. Lyon: International Agency for Research on Cancer; 2010:131–182. Ogino S, Nosho K, Irahara N, et al. Lymphocytic reaction to colorectal cancer is associated with longer survival, independent of lymph node count, microsatellite instability, and CpG island methylator phenotype. Clin Cancer Res. 2009;15:6412–6420. Ishak K, Baptista A, Bianchi L, et al. Histological grading and staging of chronic hepatitis. J Hepatol. 1995;22:696–699. Boland CR, Thibodeau SN, Hamilton SR, et al. A National Cancer Institute Workshop on Microsatellite Instability for cancer detection and familial predisposition: development of international criteria for
www.ajsp.com |
311
Chan et al
15. 16. 17. 18. 19.
the determination of microsatellite instability in colorectal cancer. Cancer Res. 1998;58:5248–5257. Selaru FM, David S, Meltzer SJ, et al. Epigenetic events in gastrointestinal cancer. Am J Gastroenterol. 2009;104:1910–1912. Zhang L, Conejo-Garcia JR, Katsaros D, et al. Intratumoral T cells, recurrence, and survival in epithelial ovarian cancer. N Engl J Med. 2003;348:203–213. Galon J, Costes A, Sanchez-Cabo F, et al. Type, density, and location of immune cells within human colorectal tumors predict clinical outcome. Science. 2006;313:1960–1964. Lo KW, Chung GT, To KF. Deciphering the molecular genetic basis of NPC through molecular, cytogenetic, and epigenetic approaches. Semin Cancer Biol. 2012;22:79–86. Song HJ, Srivastava A, Lee J, et al. Host inflammatory response predicts survival of patients with Epstein-Barr virus-associated gastric carcinoma. Gastroenterology. 2010;139:84–92, e82.
312 | www.ajsp.com
Am J Surg Pathol
Volume 39, Number 3, March 2015
20. Liang Y, Wang L, Zhu Y, et al. Primary pulmonary lymphoepithelioma-like carcinoma: fifty-two patients with long-term follow-up. Cancer. 2012;118:4748–4758. 21. Petrova M, Kamburov V. Epstein-Barr virus: silent companion or causative agent of chronic liver disease? World J Gastroenterol. 2010;16:4130–4134. 22. Stoot JH, Coelen RJ, De Jong MC, et al. Malignant transformation of hepatocellular adenomas into hepatocellular carcinomas: a systematic review including more than 1600 adenoma cases. HPB (Oxford). 2010;12:509–522. 23. Bioulac-Sage P, Sempoux C, Possenti L, et al. Pathological diagnosis of hepatocellular cellular adenoma according to the clinical context. Int J Hepatol. 2013;2013:253–261. 24. Chan AW, Tong JH, Chan SL, et al. Expression of stemness markers (CD133 and EpCAM) in prognostication of hepatocellular carcinoma. Histopathology. 2014;64:935–950.
Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
Lymphoepithelioma-like Hepatocellular Carcinoma An Uncommon Variant of Hepatocellular Carcinoma With Favorable Outcome Anthony W.H. Chan, FRCPA,* Joanna H.M. Tong, PhD,*wz Yi Pan, MMed,*z Stephen L. Chan, FRCP,wy Grace L.H. Wong, MD,w8 Vincent W.S. Wong, MD,w8 Paul B.S. Lai, FRCS,z and Ka-Fai To, FRCPA*wz
Abstract: Lymphoepithelioma-like hepatocellular carcinoma (LEL-HCC) is an uncommon variant of HCC with only 22 cases reported in the literature. To better determine the incidence, clinicopathologic features, prognostic significance, and molecular pathogenesis of LEL-HCC, we presented the largest series of LEL-HCC from a 9-year retrospective cohort of patients with HCC undergoing surgical resection. LEL-HCC was identified in 20 patients (4.9%). Compared with patients having HCC without significant tumor-infiltrating lymphocyte (TIL), patients with LEL-HCC had a relatively lower frequency of male sex (P = 0.022), tended to present at early-stage disease (80.0% vs. 56.3% as AJCC stage I, P = 0.037; 100% vs. 77.3% as BCLC stage 0/A, P = 0.010), and all harbored a solitary tumor only (P = 0.006). There was no significant difference in the age at presentation, underlying chronic liver disease, cirrhotic background, serum a-fetoprotein level, tumor size, histologic grade, and frequencies of vascular invasion. Most of the TILs in LELHCC were cytotoxic T lymphocytes. None of the LEL-HCCs From the Departments of *Anatomical and Cellular Pathology; yClinical Oncology, State Key Laboratory in Oncology in South China, Prince of Wales Hospital; wInstitute of Digestive Disease, Partner State Key Laboratory of Digestive Disease; zLi Ka Shing Institute of Health Science, Sir Y.K. Pao Cancer Center; 8Department of Medicine and Therapeutics; and zDepartment of Surgery, Division of Hepatobiliary and Pancreatic Surgery, The Chinese University of Hong Kong, Hong Kong. Drafting of the manuscript: A.W.H.C. Critical revision of the manuscript: S.L.C., G.L.H.W., V.W.S.W., P.B.S.L. Pathologic examinations: A.W.H.C., J.H.M.T., K.F.T. Molecular studies: J.H.M.T., Y.P. Clinical managements of the patients: S.L.C., G.L.H.W., V.W.S.W., P.B.S.L. Conflicts of Interest and Source of Funding: G.L.H.W. has served as an advisory committee member for Otsuka and Gilead. She has also served as a speaker for Bristol-Myers Squibb, Echosens, Furui, and Otsuka. V.W.S.W. has served as an advisory committee member for Roche, Novartis, Gilead, and Otsuka. He has also served as a speaker for Bristol-Myers Squibb, Roche, Novartis, Abbott Diagnostics, and Echosens. Correspondence: Ka-Fai To, FRCPA, Department of Anatomical and Cellular Pathology, Prince of Wales Hospital, The Chinese University of Hong Kong, Ngan Shing Street, Shatin, NT, Hong Kong (e-mail: [email protected]). Supplemental Digital Content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal’s Website, www.ajsp.com. Copyright r 2014 Wolters Kluwer Health, Inc. All rights reserved.
304 | www.ajsp.com
was associated with Epstein-Barr virus. LEL-HCC was associated with better overall (5-y survival: 94.1% vs. 63.9%; P = 0.007) and progression-free (5-y survival: 87.8% vs. 46.6%; P = 0.002) survivals compared with HCC without significant TIL. The multivariate analysis revealed that LEL-HCC was an independent prognostic factor for overall and progression-free survivals. The adjusted hazard ratio of cancer death and tumor progression for LEL-HCC was 0.12 (P = 0.037) and 0.14 (P = 0.002), respectively. LEL-HCC did not differ in frequencies of microsatellite instability, BRAF mutation, and DNA hypermethylation. In brief, LEL-HCC is a distinct uncommon variant of HCC characterized by dense cytotoxic T-cell infiltration and favorable prognosis. Key Words: hepatocellular carcinoma, lymphoepithelioma, lymphoid stroma, tumor-infiltrating lymphocyte, microsatellite instability, BRAF, hypermethylation (Am J Surg Pathol 2015;39:304–312)
L
ymphoepithelioma is characterized by syncytial sheets of undifferentiated carcinoma cells intermingled with dense tumor-infiltrating lymphocytes (TILs) and was originally described in the nasopharynx. The tumors, which develop outside of the nasopharynx and exhibit the morphology of lymphoepithelioma, are designated as lymphoepithelioma-like carcinomas (LELCs). LELCs arising from the head and neck region, thymus, lung, and stomach have variable associations with Epstein-Barr virus (EBV), whereas those occurring in the skin and urogenital tract have no known association with EBV. Primary hepatic LELC is rare and composed of 2 distinct entities: lymphoepithelioma-like cholangiocarcinoma (LEL-CC) and lymphoepithelioma-like hepatocellular carcinoma (LEL-HCC). LEL-CC is frequently associated with EBV and characterized by marked female predominance, favorable overall survival, and distinctively frequent DNA hypermethylation.1 In contrast, LEL-HCC has not been well characterized yet. Only 22 cases have been reported in the literature until now.2–9 There are no universally accepted diagnostic criteria for LEL-HCC. Only 18.2% of LEL-HCCs reported in the literature exhibit typical histologic features of LELC, whereas remaining cases are conventional HCCs with significant Am J Surg Pathol
Volume 39, Number 3, March 2015
Am J Surg Pathol
Volume 39, Number 3, March 2015
TILs.4–7 Wada et al3 reported the largest series of 11 cases of LEL-HCC and were the first to use quantitative criteria (>100 TILs in ten 400 high-power fields) to report definite significant TILs in LEL-HCC. Previous diagnostic terms—HCC with lymphocytic infiltration and HCC with lymphoid stroma—are probably more appropriate than LEL-HCC, but the latter is the current terminology used in the latest World Health Organization classification.10 Because of the rarity of LEL-HCC, information regarding clinicopathologic features and clinical outcome is limited. The association between LELHCC and EBV is also controversial. In the digestive tract, some carcinomas contain dense TILs. One of these carcinomas is colorectal medullary carcinoma, which is a histologic variant of colorectal carcinoma, characterized by poor differentiation, prominent TILs, and favorable prognosis.11 Medullary carcinoma is associated with microsatellite instability (MSI), BRAF mutation, and CpG island hypermethylation.12 The association between these molecular alterations and LEL-HCC is unknown. In this study, we report 20 patients with LEL-HCC from a retrospective cohort of 409 patients undergoing surgical resection for primary HCC. Our aim was to determine the incidence, clinicopathologic features, prognostic significance, and molecular changes (MSI, BRAF mutation, and CpG island hypermethylation) of LEL-HCC.
MATERIALS AND METHODS Patients The cohort has recruited patients who received curative surgery for primary HCC from January 2002 to December 2010 at Prince of Wales Hospital, Hong Kong. The study has been approved by the institutional review board. Baseline clinical and laboratory parameters (namely, complete blood count, prothrombin time and international normalized ratio, liver and renal biochemistries, hepatitis B/C [HBV/HCV] serology, a-fetoprotein [AFP], etc.) were retrieved and reviewed from the hospital database. The blood tests, excluding viral serology, were taken within 1 month before surgery. The Child-Pugh score and the Model for End-stage Liver Disease score were calculated on the basis of the baseline clinical and biochemical data as described. Tumor staging was determined by the seventh edition of American Joint Committee on Cancer (AJCC) Tumor-Node-Metastasis (TNM) staging system and the Barcelona Clinic Liver Cancer (BCLC) staging system. The neutrophil to lymphocyte ratio was calculated by dividing the absolute neutrophil count by the absolute lymphocyte count. The platelet to lymphocyte ratio was estimated as the absolute platelet count divided by the absolute lymphocyte count. The prognostic nutritional index was the sum of serum albumin level (g/L) and 5 the absolute lymphocyte count (109/L). All of the patients underwent surveillance in the clinic with regular ultrasonography and measurement of AFP according to the local practice. The duration of Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
Lymphoepithelioma-like HCC
follow-up was defined from the date of operation to the last follow-up before we analyzed the data or to the date of death. Overall survival was defined from the time of surgery to the time of HCC-related death. Progression-free survival was defined from the time of surgery to the time of radiologic evidence of tumor relapse or metastasis.
Histology, Immunohistochemistry, and EBVencoded RNA In Situ Hybridization The histologic diagnosis of HCC was reviewed and confirmed by 2 pathologists (A.W.H.C. and K.F.T.). LEL-HCC diagnosis was made by the criteria previously described by Wada et al3 (>100 in ten 400 high-power fields). Hepatitic activity in the background was evaluated by Ishak modification for hepatic activity index for scoring of necroinflammatory activity.13 Fatty liver was defined as the presence of steatosis in Z5% of hepatocytes. The degree of steatosis (grade 0, 20 mg/L)
All (n = 409) (n [%])
HCC Without TIL (n = 389) (n [%])
LEL-HCC (n = 20) (n [%])
P
346 (84.6) 55.7 ± 11.0 197 (48.2) 150 (36.7) 343 (83.9) 16 (3.9) 166 (40.6) 244 (59.7)
333 (85.6) 55.6 ± 10.9 186 (47.8) 145 (37.3) 326 (83.8) 16 (4.1) 155 (39.8) 236 (60.7)
13 (65.0) 57.5 ± 12.3 11 (55.0) 5 (25.0) 17 (85.0) 0 (0.0) 11 (55.0) 8 (40.0)
0.022 0.444 0.552 0.263 1.000 1.000 0.202 0.066
52 ± 35 351 (85.8) 94 (23.0) 36 (8.8) 108 (26.4)
53 ± 35 334 (85.9) 94 (24.2) 34 (8.7) 105 (27.0)
38 ± 21 17 (85.0) 0 (0.0) 2 (10.0) 3 (15.0)
0.060 1.000 0.006 0.692 0.235 0.056
235 73 95 6
219 72 93 5
16 1 2 1
(57.5) (17.8) (23.2) (1.5)
(56.3) (18.5) (23.9) (1.3)
(80.0) (5.3) (10.5) (5.3) 0.037
235 (57.5) 174 (42.5)
219 (56.3) 170 (43.7)
16 (80.0) 4 (20.0)
320 (78.2) 89 (21.8) 68.0 [5.0-705.0] 240 (58.7)
300 (77.1) 89 (22.9) 64.5 [5.0-729.0] 227 (58.4)
20 (100.0) 0 (0.0) 203.0 [22.2-466.5] 13 (65.0)
0.010 0.528 0.257
IQR indicates interquartile range.
TABLE 2. Liver Biochemistry, Hepatitic Activity, Systemic Inflammatory Indices, and Molecular Studies of HCC With or Without Significant TIL Serum albumin (g/L) Serum total bilirubin (mmol/L) Serum alanine transaminase (IU/L) Serum alkaline phosphatase (IU/L) Model for end-stage liver disease (MELD) Ishak activity score Neutrophil count ( 109/L) Lymphocyte count ( 109/L) Neutrophil to lymphocyte ratio Platelet to lymphocyte ratio Prognostic nutritional index
HCC Without TIL (n = 389)
LEL-HCC (n = 20)
P
39.2 ± 5.2 12.8 ± 10.4 59.1 ± 44.0 98.4 ± 44.8 8.7 ± 2.5 2.7 ± 1.7 4.0 ± 1.7 1.5 ± 0.6 3.0 ± 1.9 128.3 ± 72.9 46.6 ± 6.6
41.5 ± 2.8 9.2 ± 2.9 45.4 ± 22.1 84.9 ± 38.0 7.9 ± 1.3 2.4 ± 1.6 3.7 ± 1.5 1.6 ± 0.4 2.6 ± 2.1 117.1 ± 42.4 48.9 ± 4.0
0.064 0.130 0.180 0.199 0.182 0.440 0.509 0.757 0.490 0.516 0.205
HCC Without TIL (n = 38)
LEL-HCC (n = 19)
65.8 21.1 13.2 0.0
57.8 21.1 21.1 0.0
1.000
32.2 41.2 43.8 22.2 60.5 44.7 15.8
57.9 37.5 33.3 21.1 68.4 47.4 26.3
0.070 0.680 0.470 0.920 0.560 0.851 0.342
MSI status MSS MSI-L MSI-H BRAF mutation Hypermethylation p15 p16 RASSF1A GSTP1 1 or more of the above genes 2 or more of the above genes 3 or more of the above genes
0.732
MSI-H indicates high-frequency MSI; MSI-L, low-frequency MSI; MSS, microsatellite stable.
Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
www.ajsp.com |
307
Chan et al
and sixty-two patients (64.1%) were alive at the end of the follow-up period. The 2- and 5-year overall survival rates were 80.8% and 65.3%, respectively. One hundred and ninety-eight patients (48.4%) did not suffer from tumor relapse during the follow-up period. The 2- and 5-year
Am J Surg Pathol
Volume 39, Number 3, March 2015
progression-free survival rates were 59.5% and 48.5%, respectively. LEL-HCC was associated with better overall (2-y survival rate of 100% vs. 79.9%; 5-y survival rate of 94.1% vs. 63.9%; P = 0.007) and progression-free (2-y survival rate
FIGURE 1. Pathology of LEL-HCC. A and B, The tumor was composed of trabeculae and sheets of tumor cells with dense lymphocytic infiltrate (hematoxylin and eosin). The TILs are composed of mainly cytotoxic T cells (CD3+/CD8+). C, CD20. D, CD3. E, CD8. F, Foxp3.
308 | www.ajsp.com
Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
Am J Surg Pathol
Volume 39, Number 3, March 2015
of 100% vs. 57.5%; 5-y survival rate of 87.8% vs. 46.6%; P = 0.002) survivals compared with HCC without significant TIL (Fig. 2). On univariate analysis, the following parameters were also associated with unfavorable overall and progression-free survivals: older age, larger tumor size, moderate/poor histologic differentiation, tumor multiplicity, the presence of vascular invasion, advanced AJCC and BCLC stages, and marked elevated serum AFP of >1000 mg/L (Tables 3, 4). The multivariate analysis on these parameters revealed that LEL-HCC was one of the independent prognostic factors for overall and progressionfree survivals (Tables 3, 4). The adjusted hazard ratio of cancer death and tumor progression for LEL-HCC was 0.12 (95% confidence interval: 0.02-0.88, P = 0.037) and 0.14 (95% confidence interval: 0.04-0.48, P = 0.002), respectively. The presence of significant TIL in HCC could reduce the risk of cancer death and progression by over 85%.
MSI, BRAF Mutation, and CpG Island Hypermethylation of LEL-HCC Nineteen of 20 LEL-HCCs (1 case was not included because of insufficient material) and 38 age/sex/stagematched HCCs without significant TIL were subjected for molecular studies. Eleven patients with LEL-HCC (57.8%) were classified as microsatellite stable, 4 patients (21.1%) as low-frequency MSI, and 4 patients (21.1%) as high-frequency MSI. The frequencies of MSI status were not different from those of matched HCC without significant TIL (P = 0.732). None of the 19 LEL-HCCs nor 38 HCCs without significant TIL showed BRAF mutation (P = 1.000). Four genes (p15, p16, RASSF1A, and GSTP1), commonly silenced by hypermethylation in HCC, were examined to compare methylation status between LEL-HCC and HCC without significant TIL.15 The frequencies of hypermethylation of these 4 genes were not different between both groups. Results of molecular studies are summarized in Table 2.
DISCUSSION LEL-HCC is a distinct uncommon variant of HCC. Since it was firstly described by Shirabe and colleagues in 1995, a total of 22 cases have been reported in the literature.2–9 Herein, we present 20 patients with LEL-HCC from a 9-year retrospective cohort. The frequency of LEL-HCC in 2 independent large series of HCC patients in Japan (n = 163) and France (n = 162) was 6.7% and 3.6%, respectively,3,4 and comparable to that of our cohort (4.9%, n = 409). In all 42 patients with LEL-HCC, the mean age of presentation was 58 years (range, 39 to 81 y), with a male to female ratio of 2.8:1. Half of these patients had cirrhosis (50.0%). HBV and HCV accounted for 54.8% and 38.1% of the underlying etiology of these patients. The mean size of the tumor was 32 mm (range, 15 to 95 mm). The majority of the tumors were solitary (87.8%) and free of vascular invasion (84.7%). About three quarter (76.9%) of patients presented at AJCC stage I disease. Only a single case was reported to be associated with EBV.5 The clinical outcome was good with 82.9% of patients alive without disease. Patients who suffered from Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
Lymphoepithelioma-like HCC
disease recurrence and cancer death account for 9.8% and 9.8%, respectively. Our cohort further consolidated the favorable prognosis of LEL-HCC by showing the presence of significant TIL in HCC as an independent prognostic factor for overall and progression-free survivals in multivariate analyses. In many but not all malignant tumors, those containing prominent infiltration of cytotoxic T-cell TILs tend to have a better prognosis.16,17 The antitumor effect of cytotoxic T-cell TILs contributes to less aggressive behavior of LEL-HCC. The pathogenesis of LEL-HCC remains unknown. Firstly, EBV has been shown to be associated with many LELCs. Almost all nasopharyngeal carcinomas harbor EBV.18 Pulmonary and gastric LELCs in Asian populations show strong associations with EBV.19,20 One type of primary hepatic LELC, LEL-CC, is also frequently associated with EBV (69.6%).1 Is there any association between LEL-HCC and EBV? Only 1 case of LEL-HCC has been reported.5 Even in our region endemic with EBV, none of 20 LEL-HCCs in our cohort shows the presence of EBV. Although several groups have demonstrated the presence of EBV-DNA in HBV/HCVrelated HCC tumors, no direct evidence supporting the causative role of EBV in HCC development is available yet.21 The role of EBV in carcinogenesis of HCC, particularly LEL-HCC, remains disputable. Secondly, malignant transformation of hepatocellular adenoma (HCA) is a rare but well-documented phenomenon with a reported frequency of 4.2%.22 One histologic subtype of HCA is inflammatory HCA (IHCA), which is featured by the presence of focal or dense intratumoral inflammatory infiltrate, sinusoidal dilatation and peliosis, and frequent ductular reaction. Immunoreactivity to C-reactive protein and SAA is characteristic. IHCA is frequently associated with obesity, glycogenosis, and fatty liver disease.23 Could LEL-HCC represent a malignant counterpart of IHCA? In our cohort of LEL-HCC, none of the tumor shows significant peliosis, ductular reaction, or immunoreactivity to SAA. Although the frequency of fatty liver in the nontumorous livers of LEL-HCC is higher than those of HCC without significant TIL, the difference is statistically insignificant. Moreover, none of the patients with LEL-HCC has underlying glycogenosis. Apart from intratumoral inflammatory infiltrate linking up these 2 primary hepatocellular neoplasms, LEL-HCC and IHCA appear not to have any association in clinical manifestations, pathologic features, and tumorigenesis. Thirdly, prominent lymphocytic infiltration in colorectal carcinoma is associated with MSI, BRAF mutation, and DNA hypermethylation.12 Could these molecular alterations explain significant TILs in LEL-HCC? We did not demonstrate any association of LEL-HCC with MSI, BRAF mutation, and DNA hypermethylation. The exact carcinogenesis of LEL-HCC requires further studies. The diagnosis of LEL-HCC is usually straightforward by pathologic examination. However, a few differential diagnoses, including LEL-CC, metastatic LELC from other sites, and lymphoma, may need to be considered. Firstly, LEL-CC is a rare variant of cholangiocarcinoma, with only www.ajsp.com |
309
Chan et al
Am J Surg Pathol
Volume 39, Number 3, March 2015
FIGURE 2. A, Kaplan-Meier overall survival curves and (B) Kaplan-Meier progression-free survival curves for LEL-HCC and HCC without significant TIL.
23 cases reported in the literature. It is primary hepatic adenocarcinoma with varied glandular differentiation, dense lymphoplasmacytic infiltrate, and variable expression of biliary-type cytokeratins (CK7 and CK19) and stemness markers (CD133 and EpCAM).1 Although a small subset of HCC may express CK19, CD133, and/or EpCAM, LEL-CC does not demonstrate any hepatocellular differentiation
evident by typical histology and negative immunoreactivity to hepatic markers (HepPar-1 and Arginase-1).24 Secondly, metastatic LELC of other sites can be differentiated from LEL-HCC by clinicoradiologic correlations. Moreover, metastatic LELC does not express hepatic markers by immunohistochemistry. Thirdly, lymphoma, either primary or metastatic, may mimic LEL-HCC. The presence of
TABLE 3. Univariate and Multivariate Analyses on the Overall Survival of HCC Patients Univariate Patient factors Male sex Age >60 y HBV HCV Fatty liver Cirrhosis Tumor factors Tumor size (mm) Moderate/poor differentiation Multiple tumors Major vessel invasion Microvascular invasion AJCC stage II vs. I III vs. I IV vs. I BCLC stage B vs. 0/A C/D vs. 0/A Serum AFP (>1000 mg/L) LEL-HCC
Multivariate* P
Hazard Ratio
95% CI
1.602 1.487 0.720 1.271 0.783 1.345
0.953-2.693 1.058-2.089 0.479-1.082 0.594-2.721 0.556-1.103 0.959-1.887
0.076 0.022 0.114 0.537 0.162 0.086
1.009 2.186 3.024 2.587 4.353
1.006-1.014 1.236-3.867 2.150-4.253 1.625-4.119 3.125-6.064
4.450 5.664 22.268 2.245 3.334 1.812 0.108
P
Hazard Ratio
95% CI
1.308
0.916-1.868
0.140
< 0.001 0.007 < 0.001 < 0.001 < 0.001
1.002 1.144
0.997-1.007 0.628-2.084
0.398 0.661
2.884-6.886 3.798-8.447 9.328-53.156
< 0.001 < 0.001 < 0.001
4.375 4.814 27.196
2.766-6.921 3.012-7.695 10.987-67.321
1.256-4.015 2.289-4.857 1.263-2.599 0.015-0.770
0.006 < 0.001 0.001 0.026
1.150 0.120
0.785-1.684 0.017-0.876
< 0.001 < 0.001 < 0.001
0.473 0.037
*Tumor multiplicity, vascular invasion, and BCLC stage were excluded in the multivariate analysis to minimize multicollinearity. CI indicates confidence interval.
310 | www.ajsp.com
Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
Am J Surg Pathol
Volume 39, Number 3, March 2015
Lymphoepithelioma-like HCC
TABLE 4. Univariate and Multivariate Analyses on the Progression-free Survival of HCC Patients Univariate Patient factors Male sex Age >60 y HBV HCV Fatty liver Cirrhosis Tumor factors Tumor size (mm) Moderate/poor differentiation Multiple tumors Major vessel invasion Microvascular invasion AJCC stage II vs. I III vs. I IV vs. I BCLC stage B vs. 0/A C/D vs. 0/A Serum AFP (>1000 mg/L) LEL-HCC
Multivariate* P
Hazard Ratio
95% CI
1.318 1.396 0.781 1.502 0.788 1.253
0.885-1.963 1.048-1.860 0.547-1.114 0.817-2.763 0.593-1.046 0.947-1.660
0.174 0.023 0.173 0.191 0.099 0.115
1.007 2.041 2.617 2.230 3.156
1.004-1.011 1.299-3.206 1.948-3.516 1.475-3.371 2.376-4.192
3.172 4.211 14.985 2.138 2.769 1.998 0.199
P
Hazard Ratio
95% CI
1.247
0.919-1.664
0.160
< 0.001 0.002 < 0.001 < 0.001 < 0.001
1.000 1.283
0.996-1.004 0.802-2.053
0.815 0.298
2.225-4.522 3.052-5.810 6.490-34.602
< 0.001 < 0.001 < 0.001
2.846 3.728 37.194
1.973-4.106 2.586-5.414 14.459-95.681
1.292-3.538 1.984-3.864 1.473-2.709 0.064-0.623
0.003 < 0.001 < 0.001 0.006
1.510 0.141
1.097-2.078 0.042-0.476
< 0.001 < 0.001 < 0.001
0.011 0.002
*Tumor multiplicity, vascular invasion, and BCLC stage were excluded in the multivariate analysis to minimize multicollinearity. CI indicates confidence interval.
architectural/cytologic atypia in lesional hepatocytes and the absence of an aberrant immunophenotype of the lymphoid component support the diagnosis of LEL-HCC. Our study is limited by its retrospective nature and relatively small sample size. Completeness of clinical and pathologic data and long duration of follow-up (up to 13.1 y) can partly compensate for this limitation. Moreover, LEL-HCC is a rare variant of HCC. Our cohort has already accounted for 42.5% of all LEL-HCCs in the literature. Another limitation is that the mechanism of recruitment of TILs in LEL-HCC remains unanswered. Our study provides circumstantial evidence that the presence of TIL in LEL-HCC is unlikely correlated with necroinflammatory activity of the underlying chronic hepatitis and systemic inflammatory responses. We also could not establish any association of LEL-HCC with EBV, IHCA, MSI, BRAF mutation, and DNA hypermethylation. A multicentric prospective study recruiting more LEL-HCC patients is needed to further investigate the underlying mechanism of recruitment of TILs in LEL-HCC. In conclusion, LEL-HCC is a distinct uncommon variant of HCC characterized by dense cytotoxic T-cell infiltration and favorable prognosis. The association with EBV is only very weak. The mechanism of recruitment of dense TILs requires further studies.
3. 4. 5. 6. 7. 8. 9. 10.
11.
12.
REFERENCES 1. Chan AW, Tong JH, Sung MY, et al. Epstein-Barr virus-associated lymphoepithelioma-like cholangiocarcinoma: a rare variant of intrahepatic cholangiocarcinoma with favourable outcome. Histopathology. 2014;65:674–683. 2. Shirabe K, Matsumata T, Maeda T, et al. A long-term surviving patient with hepatocellular carcinoma including lymphocytes
Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
13. 14.
infiltration—a clinicopathological study. Hepatogastroenterology. 1995;42:996–1001. Wada Y, Nakashima O, Kutami R, et al. Clinicopathological study on hepatocellular carcinoma with lymphocytic infiltration. Hepatology. 1998;27:407–414. Emile JF, Adam R, Sebagh M, et al. Hepatocellular carcinoma with lymphoid stroma: a tumour with good prognosis after liver transplantation. Histopathology. 2000;37:523–529. Si MW, Thorson JA, Lauwers GY, et al. Hepatocellular lymphoepithelioma-like carcinoma associated with epstein barr virus: a hitherto unrecognized entity. Diagn Mol Pathol. 2004;13:183–189. Chen CJ, Jeng LB, Huang SF. Lymphoepithelioma-like hepatocellular carcinoma. Chang Gung Med J. 2007;30:172–177. Nemolato S, Fanni D, Naccarato AG, et al. Lymphoepitelioma-like hepatocellular carcinoma: a case report and a review of the literature. World J Gastroenterol. 2008;14:4694–4696. Park HS, Jang KY, Kim YK, et al. Hepatocellular carcinoma with massive lymphoid infiltration: a regressing phenomenon? Pathol Res Pract. 2009;205:648–652. Shinoda M, Kadota Y, Tsujikawa H, et al. Lymphoepithelioma-like hepatocellular carcinoma: a case report and a review of the literature. World J Surg Oncol. 2013;11:97. Theise ND, Curado MP, Franceshi S, et al. Hepatocellular carcinoma. In: Bosman FT, Carneiro F, Hruban RH, et al, eds. WHO Classification of Tumours of the Digestive System. Lyon: International Agency for Research on Cancer; 2010:205–216. Hamiton SR, Bosman FT, Boffetta P, et al. Carcinoma of the colon and rectum. In: Bosman FT, Carneiro F, Hruban RH, et al, eds. WHO Classification of Tumours of the Digestive System. Lyon: International Agency for Research on Cancer; 2010:131–182. Ogino S, Nosho K, Irahara N, et al. Lymphocytic reaction to colorectal cancer is associated with longer survival, independent of lymph node count, microsatellite instability, and CpG island methylator phenotype. Clin Cancer Res. 2009;15:6412–6420. Ishak K, Baptista A, Bianchi L, et al. Histological grading and staging of chronic hepatitis. J Hepatol. 1995;22:696–699. Boland CR, Thibodeau SN, Hamilton SR, et al. A National Cancer Institute Workshop on Microsatellite Instability for cancer detection and familial predisposition: development of international criteria for
www.ajsp.com |
311
Chan et al
15. 16. 17. 18. 19.
the determination of microsatellite instability in colorectal cancer. Cancer Res. 1998;58:5248–5257. Selaru FM, David S, Meltzer SJ, et al. Epigenetic events in gastrointestinal cancer. Am J Gastroenterol. 2009;104:1910–1912. Zhang L, Conejo-Garcia JR, Katsaros D, et al. Intratumoral T cells, recurrence, and survival in epithelial ovarian cancer. N Engl J Med. 2003;348:203–213. Galon J, Costes A, Sanchez-Cabo F, et al. Type, density, and location of immune cells within human colorectal tumors predict clinical outcome. Science. 2006;313:1960–1964. Lo KW, Chung GT, To KF. Deciphering the molecular genetic basis of NPC through molecular, cytogenetic, and epigenetic approaches. Semin Cancer Biol. 2012;22:79–86. Song HJ, Srivastava A, Lee J, et al. Host inflammatory response predicts survival of patients with Epstein-Barr virus-associated gastric carcinoma. Gastroenterology. 2010;139:84–92, e82.
312 | www.ajsp.com
Am J Surg Pathol
Volume 39, Number 3, March 2015
20. Liang Y, Wang L, Zhu Y, et al. Primary pulmonary lymphoepithelioma-like carcinoma: fifty-two patients with long-term follow-up. Cancer. 2012;118:4748–4758. 21. Petrova M, Kamburov V. Epstein-Barr virus: silent companion or causative agent of chronic liver disease? World J Gastroenterol. 2010;16:4130–4134. 22. Stoot JH, Coelen RJ, De Jong MC, et al. Malignant transformation of hepatocellular adenomas into hepatocellular carcinomas: a systematic review including more than 1600 adenoma cases. HPB (Oxford). 2010;12:509–522. 23. Bioulac-Sage P, Sempoux C, Possenti L, et al. Pathological diagnosis of hepatocellular cellular adenoma according to the clinical context. Int J Hepatol. 2013;2013:253–261. 24. Chan AW, Tong JH, Chan SL, et al. Expression of stemness markers (CD133 and EpCAM) in prognostication of hepatocellular carcinoma. Histopathology. 2014;64:935–950.
Copyright
r
2014 Wolters Kluwer Health, Inc. All rights reserved.
