OUP user menu

Diagnostic Usefulness of p16/CDKN2A FISH in Distinguishing Between Sarcomatoid Mesothelioma and Fibrous Pleuritis

Di Wu PhD, Kenzo Hiroshima MDPhD, Shinji Matsumoto, Kazuki Nabeshima MDPhD, Toshikazu Yusa MDPhD, Daisuke Ozaki MDPhD, Michio Fujino MDPhD, Hisami Yamakawa MDPhD, Yukio Nakatani MDPhD, Yuji Tada MDPhD, Hideaki Shimada MDPhD, Masatoshi Tagawa MDPhD
DOI: http://dx.doi.org/10.1309/AJCPT94JVWIHBKRD 39-46 First published online: 1 January 2013


The distinction between sarcomatoid mesothelioma and fibrous pleuritis is difficult based on histology, especially when the amount of tumor tissue examined via biopsy is small and immunohistochemical examination is inconclusive. We studied the usefulness of deletion of p16 with fluorescence in situ hybridization (FISH) and p16 hypermethylation with polymerase chain reaction for the diagnosis and prognosis of malignant pleural mesothelioma (MPM). We analyzed 50 MPMs, including 22 sarcomatoid mesothelioma cases and 10 fibrous pleuritis cases. We set the cutoff value of homozygous deletion pattern as 14.4% based on FISH signaling patterns using samples of fibrous pleuritis. The percentage of homozygous deletion pattern was higher than 14.4% in 55.6% of the epithelioid mesotheliomas (10/18) and in all of the sarcomatoid mesotheliomas (22/22). Methylation of p16 was observed in 7 (20.6%) of 34 informative cases. p16 FISH analysis can be a reliable test for distinguishing between sarcomatoid mesothelioma and fibrous pleuritis and a prognostic factor for MPM.

Key Words
  • p16
  • Fluorescence in situ hybridization
  • Methylation
  • Mesothelioma
  • Prognosis
  • Pleuritis

Malignant mesothelioma (MM) is a highly aggressive neoplasm with a median survival of 8 to 14 months.1 In Japan, more than 1,000 new cases are diagnosed each year. The number of deaths between the years 2030 and 2039 is predicted to be 21 times greater than the observed number of deaths between 1990 and 1999, and the number of deaths will peak between 2030 and 2034 in Japan.2

Based on the proper clinical and radiologic context, a diagnosis of MM can be rendered when the pathologic findings are typical. However, the pathologic diagnosis of MM can be difficult because of various histopathologic patterns, such as the often deceptively bland appearance of tumor cells and the presence of pseudomesotheliomatous carcinoma. Small surgical biopsy specimens and cytologic specimens also add to the difficulty of proper diagnosis. The International Mesothelioma Panel recommends the application of at least 2 mesothelial and 2 carcinoma markers in addition to a pancytokeratin for the differential diagnosis between epitheli-oid mesothelioma and metastatic carcinoma.3

In the absence of clinical and imaging information, pathologic interpretation of benign and malignant mesothelial proliferations can be more difficult, especially when meso-thelial proliferations are observed only on a serosal surface or when spindle cells proliferate in the parietal pleura. Differential diagnosis of reactive mesothelial hyperplasia or fibrous pleuritis from malignant pleural mesothelioma (MPM) is crucial to prevent unnecessary invasive and aggressive treatments for MPM in patients with reactive mesothelial hyper-plasia or fibrous pleuritis.

Some immunohistochemical markers are useful for differentiating between epithelioid mesothelioma and reactive mesothelial hyperplasia. Epithelial membrane antigen is reported to be useful in the differential diagnosis.4,5 Strong membranous expression of epithelial membrane antigen was seen in 48 (80%) of 60 MPM cases and in 8 (20%) of 40 reactive mesothelial hyperplastic cases.5 Glut-1 is expressed in most cases of epithelioid mesothelioma but not in reactive mesothelial hyperplasia.6,7 Insulin-like growth factor II messenger RNA–binding protein 3 is also expressed in epithelioid mesothelioma but not in reactive mesothelial hyperplasia.8,9 CD146 was reported to be a sensitive and specific marker for mesotheliomas in cytologic specimens.10 The combined application of these markers may provide a potential option to differentiate epithelioid mesothelioma from reactive mesothe-lial hyperplasia.

However, distinguishing between sarcomatoid mesothelioma and fibrous pleuritis is difficult, especially when the amount of tumor tissue in the biopsy specimen is small and immunohistochemical examination is inconclusive. Stromal or chest wall fat invasion is the most reliable criterion to diagnose sarcomatoid mesothelioma; however, it cannot be assessed in small biopsy specimens.11

The tumor suppressor genes p16 (CDKN2A) and p15 (CDKN2B) map to the 9p21 chromosomal locus and are homozygously deleted in MM and other cancers.1214 Point mutation and promoter methylation of both genes have been reported in MM.15,16 Previous studies have used fluorescence in situ hybridization (FISH) analysis to detect the loss of the p16 gene in mesothelioma15,1723 and methylation-specific polymerase chain reaction (PCR) analysis to detect p16 promoter hypermethylation.15,16 In this study, we evaluate the usefulness of 2 methods, p16 FISH and p16 promoter hyper-methylation, in discriminating sarcomatoid mesothelioma from fibrous pleuritis. We also analyze the frequency of homo-zygous deletion of the p16 gene among histologic subtypes of MPM. Our results confirmed the high prevalence of p16 deletion in MPM and suggest that detection of p16 deletions using FISH analysis could prove useful in the diagnosis of sarcomatoid mesothelioma.

Materials and Methods

Fifty MPM cases and 10 fibrous pleuritis cases seen between December 1999 and March 2012 were collected from the archives of the Department of Pathology of Yachiyo Medical Center, Chiba University, Chiba Rosai Hospital, and Chiba East Hospital Table 1. Paraffin-embedded tissue blocks were available in all cases. Informed consent was obtained from all patients. The diagnosis of MPM was based on the combination of clinical findings, imaging and gross observations at surgery, and routine H&E histology. All cases were reviewed by 2 pathologists (K.H., Y.N.). The diagnosis was confirmed on immunohistochemistry (calretinin, WT-1, D2-40, cyto-keratin [CK] 5/6, CAM5.2, CK AE1/AE3, carcinoembryonic antigen, thyroid transcription factor 1, desmin, smooth muscle actin, and S100) using the Bond MAX autoimmunostainer (Leica, Wetzlar, Germany). Representative tissue blocks were selected for FISH analysis. The fibrous pleuritis sample in this study displayed atypical spindle cells. All patients with fibrous pleuritis except 1 had a history of asbestos exposure and pleu-ral effusion. Therefore, we diagnosed these cases as benign asbestos pleural effusion.24

View this table:
Table 1


FISH was performed on formalin-fixed, paraffin-embedded, 4-μm-thick tissue sections. After paraffin sections were deparaffinized in xylene, dehydrated in ethanol, and air dried, dual-color FISH analysis was performed using a spectrum green–labeled chromosome 9 centromeric probe and a spectrum orange–labeled, locus-specific CDKN2A(p16) probe (Abbott, Abbott Park, IL). Pretreatment steps were performed using a histology FISH accessory kit (DakoCytomation, Tokyo, Japan) as follows: the section was placed in pretreat-ment solution (MES; 2-ethanesulphonic acid buffer), incubated at 121°C for 1 minute, and digested with pepsin at 37°C for 5 to 11 minutes. The probes were denatured for 5 minutes at 95°C before hybridization. The slides were then hybridized for 48 hours at 37°C and washed in 2 × SSC/0.3% Tween 20 (Sigma, St Louis, MO) at 78°C for 2 minutes, and counter-stained with DAPI/antifade (Abbott).


Slides were examined and images were obtained using a fluorescence microscope (Axio Imager 2, Carl Zeiss Microscopy, Göttingen, Germany) equipped with filter sets with single exciters for spectrum green, spectrum orange, and DAPI (UV 360 nm). Only individual and well-delineated cells were scored. Overlapping cells were excluded from the analysis. At least 100 cells were scored for each case. Homozygous deletion was defined as the absence of both 9p21 signals in the presence of at least 1 chromosome 9 centromere signal.7,18,22,25 Hemizygous deletion was defined as the presence of only one 9p21 signal in the presence of 2 chromosome 9 centromere signals or when the number of 9p21 signals was less than half the number of chromosome 9 centromere signals.26,27

DNA Extraction

DNA samples were extracted from paraffin-embedded materials. Tissue specimens were accurately removed from two 10-μm serial sections mounted on glass slides by scraping the marked area with a needle under a stereomicroscope. Tumor cells were selectively removed from the sections to minimize contamination by normal tissue specimens. The dissected sections were digested by proteinase K over 2 nights at 37°C.

Methylation-Specific PCR assay

The promoter region methylation status of the p16 gene was evaluated with methylation-specific PCR as previously reported, with slight modification28 The DNA sample was modified using bisulfite treatment with the WZ DNA Meth-ylation-Gold Kit (Zymo Research Irvine, CA). Two sets of primers were used to amplify each region of interest. One pair recognized a sequence in which CpG sites were unmethylated (bisulfite modified to UpG), and the other pair recognized a sequence in which CpG sites were methylated (unmodified by bisulfite). Reactions were hot-started at 94°C for 2 minutes. Amplification was carried out in a Takara thermal cycler (Taka-ra, Otsu, Japan) for 40 cycles (94°C/30 sec, 65°C or 60°C/30 sec, 68°C/1 min), followed by an extension at 68°C for 4 minutes. The PCR products were analyzed with electrophoresis in a 4% agarose gel and visualized with ethidium bromide staining.

Statistical Analysis

The differences among categorized groups were compared using the χ2 test. The overall survival time for the patients was defined as the time from video-assisted thoracoscopic biopsy until death or the date the patient was last known to be alive. We assessed the correlation between clinicopathologic variables (age, sex, subtype, therapy, p16 deletion, and p16 promoter methylation) and overall survival. Kaplan-Meier curves and survival estimates were calculated, and the Wilcoxon test was used to test for differences between groups. A P value less than .05 was considered statistically significant.


Clinicopathologic Characteristics

The pathologic diagnoses of the 50 patients with meso-theliomas and the 10 patients with fibrous pleuritis are summarized in Table 1. All 50 mesothelioma cases were pleural mesotheliomas. The 5 cases of desmoplastic meso-thelioma were classified as sarcomatoid mesothelioma in this study. The majority of specimens were from video-assisted thoracoscopic biopsies (78%), 8 cases were from extrapleural pneumonectomy (EPP) specimens (16%), 2 were from autopsy specimens (4%), and 1 was from resection of a metastatic lymph node (2%).

Cutoff Values for Fibrous Pleuritis

To examine whether FISH analysis of p16 can discriminate mesotheliomas from fibrous pleuritis, we evaluated the frequency of p16 deletion in MPM and in fibrous pleuritis. Cutoff levels were calculated as the mean percentage + 4 standard deviations (SDs) of cells, with deletion of nuclei in fibrous pleuritis caused by artifactual loss of signals because of nuclear sectioning Table 2 The homozygous deletion pattern (no p16/CDKN2A with at least 1 CEP-9 signal) was noted in 0 to 7.4% (mean, 3.9%, SD 2.6%) of nuclei. The hemizygous deletion pattern (p16/CDKN2A: CEP-9 ≤1:2) was observed in 1.0% to 21.0% (mean, 13.3%; SD, 6.0%) of nuclei. The homozygous or hemizygous deletion pattern was observed in 7.0% to 25.0% (mean, 17.2%; SD, 6.3%) of nuclei. Homozygous deletion was defined as more than 14.4% of nuclei showing a homozygous deletion pattern, and homozygous/hemizygous deletion was defined as more than 42.3% of nuclei showing a homozygous or hemizygous deletion pattern.

p16 Deletion in MPM

FISH analysis was successful in 48 of 50 cases. Two cases of EPP samples collected in 2002 and 2003 could not be analyzed because the signal intensity was too low. The results of p16 deletion with FISH analysis are shown in Table 3,Figure 1, and Image 1. Homozygous deletions of p16 were seen in 39 (81.3%) of 48 MPM cases. In epithelioid meso-thelioma, the percentage of homozygous deletion pattern was higher than the cutoff value in 55.6% (10/18) of the cases. In biphasic mesothelioma, the percentage of homozygous deletion pattern was higher than the cutoff value in 87.5% (7/8) of the cases. In all of the sarcomatoid mesothelioma cases, the percentage of homozygous deletion pattern was higher than the cutoff value (Figure 1A).

View this table:
Table 2
View this table:
Table 3
Image 1

Fluorescence in situ hybridization showing homozygous (loss of both red signals; A) and hemizygous (loss of one red signal; B) loss of p16 in 2 mesothelioma cases, respectively. Arrow indicates normal cells showing 2 green and 2 red signals or 1 green and 1 red signal.

Figure 1

Percentage of nuclei with a deletion pattern in cases with each histologic subtype of malignant pleural mesothelioma and fibrous pleuritis. A, Homozygous deletion pattern. Cutoff value was set as 14.4% based on mean percentage + 4 standard deviations (SDs) of nuclei showing a homozygous deletion pattern on fluorescence in situ hybridization (FISH) in fibrous pleuritis. B, Homozygous/hemizygous deletion pattern. Cutoff value was set as 42.3% based on mean percentage + 4 SDs of nuclei showing a homozygous or hemizygous deletion pattern on FISH in fibrous pleuritis. B, biphasic subtype; E, epithelioid subtype; P, pleuritis; S, sarcomatoid subtype.

We analyzed the homozygous deletion pattern and the hemizygous deletion pattern together. The mean percentage + 4 SDs for fibrous pleuritis was 42.3%. When we applied this cutoff value, 66.7% (12/18) of epithelioid mesothelioma could be diagnosed with p16 FISH analysis (Figure 1B).

Frequency of Methylation in MPM and Fibrous Pleuritis

We performed methylation-specific PCR for the p16 promoter in 35 mesothelioma and 7 fibrous pleuritis cases Figure 2. Methylation-specific PCR was not successful in 1 mesothelioma case because the amount of extracted DNA was low. Of the remaining mesothelioma cases, the frequency of promoter methylation was 20.6% (7/34). The frequency of promoter methylation in fibrous pleuritis cases was 42.9% (3/7). No significant difference was observed in the frequency of promoter methylation between MPM and fibrous pleuritis. We investigated the relationship between the homozygous deletion and promoter methylation of the p16 gene in MPM. Twenty-two (81.5%) of the 27 mesothelioma cases with homozygous deletion were without methylation; however, 5 (18.5%) of the 27 mesothelioma cases with homozygous deletion did show methylation. Only 2 of 6 mesotheliomas without homozygous deletion that were analyzed showed promoter methylation. Therefore, no association was found between p16 homozygous deletion and p16 promoter methylation.

Figure 2

Methylation status of the p16 gene in 7 fibrous pleuritis and 7 malignant pleural mesothelioma (MPM) cases. Numbers indicate individual patients. bp, base pair; DW, distilled water as negative control; M, polymerase chain reaction (PCR) product with methylated-specific primers; U, PCR product with unmethylated-specific primers.

Prognostic Analysis of p16 Deleted vs Nondeleted MPM

In this study, we analyzed the prognosis of 43 patients with MPM who had been treated during the period between December 1999 and March 2012. Table 4 summarizes the predictors of prognosis in MPM. Patients with sarcomatoid MPM showed the worst survival, whereas patients with epi-thelioid MPM showed better survival (Table 4). The median survival times for patients with respect to histologic subtypes of MPM are as follows: 22 months for epithelioid mesothe-lioma, 12 months for biphasic mesothelioma, and 5 months for sarcomatoid mesothelioma (P = .0015). p16 homozygous deletion is associated with shorter survival (Table 4). For patients with p16 homozygous deletion, the median survival time was 7 months. In contrast, patients without p16 homo-zygous deletion showed a longer survival time of 22 months (P = .0363). Patients without p16 methylation had longer survival times than patients with p16 methylation (P = .0442).


Loss of the p16 gene is one of the most frequent genetic alterations in MPM.15 Illei et al18 found homozygous deletion of p16 in 70 (74%) of 95 MPM cases, including 49 (69%) of 71 epithelioid, 16 (84%) of 19 biphasic, and 5 (100%) of 5 sarcomatoid mesotheliomas. In a recent study of MPM, Chiosea et al22 detected homozygous 9p21 deletion with FISH analysis in 35 (67%) of 52 cases of MPM. Our study showed similar results, with homozygous deletion of p16 seen in 81.3% (39/48) of Japanese MPM cases. Furthermore, homozygous deletion was observed in 55.6% (10/18) of epithelioid mesotheliomas, 100% (22/22) of sarcomatoid mesotheliomas, and 87.5% (7/8) of biphasic mesotheliomas. These results indicate that p16 FISH analysis in MPM is a useful test that may guide diagnosis and treatment decisions for patients.

View this table:
Table 4

Previous studies on p16 FISH analysis used a cutoff value of 20%.7,18,20,22,25,29 It is reported that no reactive pleu-ral mesothelial proliferations showed p16 deletion.7,2123,26 We believe that spindle cells in fibrous pleuritis do not harbor deletion of the p16 gene; therefore, we established an alternative cutoff value for p16 deletion based on FISH signaling patterns obtained from spindle cells observed in fibrous pleuritis. In a standard data distribution, 99.994% of the data values are within 4 SDs. Therefore, in this study we established the cutoff value for homozygous deletion as 14.4%. Chung et al23 proposed a cutoff value for the diagnosis of mesothelioma as 10% for homozygous deletion and 44% for hemizygous deletion based on FISH signaling patterns obtained from benign controls. Because of the variability in FISH analysis and scoring methods from one institute to another, each institute should develop its own applicable cutoff value.

Illei et al17 demonstrated that homozygous p16 deletion detected with FISH is a very powerful technique for confirming the diagnosis of MPM over reactive mesothelial cells with cytology. Differentiation of sarcomatoid mesothelioma and fibrous pleuritis via histochemical and immunohistochemi-cal staining is difficult, especially in small biopsy specimens. However, few sarcomatoid mesothelioma cases have been analyzed with FISH to date. To our knowledge, the number of sarcomatoid mesothelioma cases in each report did not exceed 5 and the results are contradictory.1821,23 We investigated 22 cases of sarcomatoid mesothelioma with FISH analysis and found that all sarcomatoid mesotheliomas showed a homozy-gous deletion pattern in more than 14.4% of tumor cells, and all of the fibrous pleuritis showed a homozygous deletion pattern in less than 14.4% of spindle cells. Both sensitivity and specificity of p16 FISH analysis using a cutoff value of 14.4% for the diagnosis of mesothelioma are 100%.

Most studies of FISH analysis reported that homozygous deletion was observed frequently in MPM17,18,21,22,26,27; more than 90% of analyzed MPM cells showed a homozygous deletion pattern.22,25 Hemizygous deletion was not identified in any of the analyzed cases of MPM in some reports.22,25 However, Chung et al23 reported that the mean percentage of nuclei with a homozygous deletion pattern was 61% (range, 1%-87%), with 3 cases showing less than 30%; hemizygous deletion occurred in 10 (18%) of 54 MPM cases. They reported that hemizygous deletion is sufficient for the diagnosis of MPM.23 Frequency of p16 gene homozygous deletion in epithelioid mesothelioma was 55.6% (10/18) in our study. When we combined the homozygous and hemizygous patterns, the cutoff value for mesothelioma was 42.3%. Two of the 6 epithelioid mesothelioma cases without homozygous deletion had a higher value than our established cutoff value. Sensitivity of p16 deletion for the diagnosis of epithelioid mesothelioma increased to 66.7% (12/18) when we combined both homo-zygous and hemizygous patterns. The p16 gene is a known tumor suppressor, and loss of both alleles is required for malignant transformation.12 Therefore, homozygous deletion fully explains the association between the p16 gene deletion pattern and a diagnosis of MPM observed in our study. Alternatively, hemizygous deletion would be the first hit of the Knudson 2-hit hypothesis that would require a second hit, such as promoter hypermethylation, to eliminate the second gene copy.30

Our study showed no correlation between the deletion of p16 as determined with FISH and methylation status of the p16 promoter. Four of 6 mesotheliomas without p16 homo-zygous deletion showed no promoter methylation, whereas 5 of 27 mesotheliomas with p16 homozygous deletion also showed promoter methylation. If p16 is deleted in tumor cells, the corresponding promoter region would also be deleted and methylation of p16 would not be observed. In some cases of mesothelioma that showed a homozygous deletion pattern of p16, the tumor cells also showed a normal p16 pattern on FISH analysis, and therefore displayed promoter hypermeth-ylation because the promoter was not deleted in these tumor cells. Alternatively, these tumor cell samples may have been contaminated by non–tumor cells that, when analyzed with methylation-specific PCR, showed promoter methylation. Our results showed that the hypermethylation of the p16 gene was observed in 3 (42.9%) of 7 fibrous pleuritis cases. Studies have reported aberrantly methylated p15 alleles in the lymphocytes of healthy donors,31p16 hypermethylation in normal breast tissue,32 and p15 hypermethylation and p16 hypermeth-ylation in normal colon mucosa.33 However, further study is needed to evaluate the relationship between methylation status of the p16 gene and tumorigenesis.

It is reported that the most important predictors of poor prognosis in MPM are poor performance status, gender, low hemoglobin concentration, high platelet count, and high WBC count.3436 Our study showed that younger patients and patients treated with EPP have a better survival. Biphasic or sarcomatous subtypes are reported to be associated with poor survival.3438 Our study results showed that patients with epithelioid mesothelioma had longer survival than those with nonepithelioid mesothelioma. Homozygous deletion of p16 is reported to be associated with a worse prognosis.23,29 Dacic et al29 demonstrated that loss of p16, as detected with FISH in 21 (60%) of 35 MPM cases with epithelioid subtype, was more frequently associated with a short survival. The estimated median survival time was 12 months for subjects positive for p16 deletion. Loss of p16 immunoreactivity in mesothe-liomas is also reported to be a negative prognostic factor in pleural or peritoneal mesothelioma.29,39 However, these studies analyzed mainly patients with epithelioid mesothelioma. Our study included not only epithelioid mesotheliomas but also biphasic and sarcomatoid mesotheliomas. Using Wil-coxon analysis, we confirmed that histology is a strong factor influencing overall survival, and p16 deletion is a prognostic predictor for patients with MPM.

In summary, our data demonstrated that FISH analysis using a commercially available dual-color FISH probe can be reliably performed on archival paraffin-embedded tissue that has been kept at room temperature for a long duration. However, signal intensity did decrease 8 years after biopsy. We also demonstrated that p16 deletion detected with FISH analysis is useful for differentiating sarcomatoid mesothelioma from fibrous pleuritis. Analysis of methylation status of the p16 gene may not be useful for the diagnosis of mesothelioma. Homozygous deletion of p16 detected with FISH analysis and p16 promoter methylation evaluated with methylation-specific PCR may predict the prognosis of patients with MPM.


Upon completion of this activity you will be able to:

  • describe histologic features of fibrous pleuritis and of sarcomatoid mesothelioma.

  • recognize borderline cases in which histology is indeterminate to allow definitive diagnosis of fibrous pleuritis vs sarcomatoid mesothelioma.

  • outline a menu of immunohistochemical stains that may assist in differentiating fibrous pleuritis from sarcomatoid mesothelioma, including p16/CDKN2A fluorescence in situ hybridization, with expected results and staining thresholds for each pathologic entity.

The ASCP is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians. The ASCP designates this journal-based CME activity for a maximum of 1 AMA PRA Category 1 Credit ™ per article. Physicians should claim only the credit commensurate with the extent of their participation in the activity. This activity qualifies as an American Board of Pathology Maintenance of Certification Part II Self-Assessment Module.

The authors of this article and the planning committee members and staff have no relevant financial relationships with commercial interests to disclose.

Questions appear on p 124. Exam is located atwww.ascp.org/ajcpcme.


  • This study was partially supported by Grants-in-Aid from the Ministry of the Environment of Japan and a Grant-in-Aid from the Nichias Corporation, Tokyo, Japan.


View Abstract