Histopathology 1986, 10, 1153-1162

Staining for factor VIII related antigen and Ulex europaeus agglutinin I (UEA-I) in 230 tumours. An assessment of their specificity for angiosarcoma and Kaposi’s sarcoma

M.LEADER, M.COLLINS, J.PATEL & K.HENRY Department of Histopathology, Westminster Site, Charing Cross & Westminster Medical School, London, U K

Accepted for publication 28 April 1986

LEADER M., COLLINS M., PATEL J. & HENRY K. (1986) Histopathology 10, 1153-1162

Staining for factor VIII related antigen and Ulex europueus agglutinin 1 (UEA-I) in 230 tumours. An assessment of their specificity for angiosarcoma and Kaposi’s sarcoma

In this study we examined the staining reactivity of commercially available antisera to factor VIII related antigen (F VIII RAg) and Ulex europaeus agglutinin I (UEA-I) on sections from 230 formalin fixed paraffin embedded tumours. These included 196 sarcomas, 20 carcinomas and 14 angiomas. All angiomas showed positive staining for F VIII RAg; all carcinomas showed negative staining; the vasoformative areas of all angiosarcomas stained positively but only four of six angiosarcomas showed positive staining of their solid areas; of seven Kaposi’s sarcomas, all showed positive staining of vessels and six showed positive staining of the spindle cell component. In the remaining 181 non-vascular sarcomas there was a false positive result in four tumours (2.2%), three of which had a history of irradiation. Pre-radiotherapy biopsies of these three tumours stained negatively with anti-F VIII RAg. UEA-I was demonstrated in all the angiomas studied, in all angiosarcomas (including the solid components) and in well-formed vessels of all Kaposi‘s sarcomas, but only in the spindle cell component of 316. However, there was an unacceptably high rate of false positive staining amongst the carcinomas and non-vascular sarcomas. In conclusion, F VIII RAg is a specific but not a sensitive marker of angiosarcomas; UEA-I is a sensitive but not a specific marker of angiosarcomas.

Keywords: Ulex europaeus agglutinin I, factor VIII related antigen, angioma, angiosar- coma, Kaposi’s sarcoma, sarcoma, carcinoma

Introduction

The histopathological diagnosis of soft tissue tumours can be exceedingly difficult and it is not always possible, even for experts, to distinguish them from pseudosar- comatous carcinomas by using conventional staining methods. However, precise

Address for correspondence: Dr M.Leader, Department of Histopathology, Westminster Medical School, Horseferry Road, London SWlP 2AR, UK.

1153

1154 M . Leader et al.

diagnosis is essential for optimal patient management and is also vital for the purpose of controlled clinical trials comparing various treatment regimes. The recently developed immunocytochemical markers may well provide useful diagnostic tools. Tissue-specific antigens can be recognized and the antibodies raised against them can then be used as tissue-specific markers. As it is presumed that tissues can retain their specific antigens even during neoplastic transformation (Ramaekers et al. 1983), these markers can now be applied to tumour diagnosis.

In this report sections from a series of 230 tumours have been reacted with antisera to factor VIII related antigen (F VIII RAg) and Ulex europaeus agglutinin I (UEA-I). The results with these markers are compared.

Materials and methods

Details of the 196 sarcomas, 14 angiomas and 20 carcinomas included in this study are summarized in Table 1. All were fixed in neutral buffered formalin and all were paraffin embedded. The sarcomas were obtained from the files of the Westminster Hospital. The age of the blocks ranged from 1 to 28 years. All sarcomas were examined by Professor D.H.MacKenzie (MacKenzie 1970) and diagnosed on the basis of H&E and other non-immunocytochemical staining. The carcinomas were chosen retrospectively from the files of the Westminster Hospital and the blocks ranged in age from 1 to 4 years. All 230 tumours were stained with F VIII RAg antiserum and 228 tumours with anti-UEA-I. In addition the angiosarcomas were stained with CAM 5.2, a cytokeratin antibody.

Table 1. Anti-F VIll RAg and anti-UEA-I staining of 230 and 228 tumours respectively

Anti-F VlII RAg Anti-UEA-I

Tumours No. Positive Negative No. Positive Negative

Angiosarcomas Kaposi's sarcomas Leiom yosarcomas Rhabdomyosarcomas Synovial sarcomas Clear cell sarcomas Neurofibrosarcomas Haemangiopericytomas Malignant fibrous histiocytomas Liposarcomas Fibrosarcomas Sarcomas unclassified Angiomas Squamous cell carcinomas Adenocarcinomas

7 8

24 22 20 8

18 19 24 20 20 6

14 10 10

4* 6 t 0 2 0 1 0 0 0 0 0 1

14 0 0

2* *i

24 20 20 7

18 19 24 20 20 5 0

10 10

I 6

24 22 20

8 18 19 24 20 20

6 14 10 10

6* 0" 3 t 3 t 2 22 0 22 2 18 2 6 2 16 1 18 4 20 2 18 1 19 0 6

14 0 I 3 5 5

~

*These results relate only to six angiosarcomas with a solid component. tThese results relate only to the spindle cell components of seven Kaposi's sarcomas, i.e. seven tumours. t=Equivocal staining in one case.

F Vlll RAg and UEA-I staining in 230 tumours 1155

The anti-F VIII RAg, a polyclonal rabbit anti-human antiserum, was obtained from Dakopatts (Copenhagen). The reagent was tested at differing dilutions, at varying temperatures and using various trypsinization times. The best and most reliable results were obtained using a dilution of anti-F VIII RAg of 1:300 on sections trypsinized for 20 min at a temperature of 37°C. The peroxidase-anti-peroxidase technique was used.

The method used for detecting UEA-I receptors was as follows: after blocking endogenous peroxidase activity, sections were incubated in UEA-I (Sigma) for 2 h at room temperature followed by washing and by incubation in rabbit anti-UEA-I (Dakopatts) at a dilution of 1 : l O O for 30 min, again followed by washing and by incubation in swine anti-rabbit serum, peroxidase labelled (Dakopatts), at a dilution of 1:150.

Sections reacted with CAM 5.2 were stained using a dilution of 1:lO of the primary antibody and the indirect immunoperoxidase technique.

Each coded section was examined using a x4, x10 and x40 objective on each section. A positive or negative result was reported on the basis o f immunocytochemi- cal staining and with no prior knowledge of the histological diagnosis. The case numbers were then decoded and the result of the staining tabulated for each tumour.

Results

The results are summarized in Table 1. For a more detailed analysis they are divided into the following categories: angiomas, angiosarcomas, Kaposi’s sarcomas, ‘non- vascular’ sarcomas and carcinomas.

A N G I O M A S

All angiomas showed positive granular cytoplasmic staining with both markers.

Table 2. Angiosarcomas. Factor VIII RAg and UEA-I staining

Vasoformative areas Solid areas

F VIIl F VIII Site Sex Age Grade RAg UEA-I RAg UEA-I

Mucosa of cheek M ? 3 + + Negative + Mucosaof cheek M 66 2 + + Occasional cells + Breast F 29 2 + + + + Thyroid F 61 1 + + None present None present

Back M 16 3 + + Occasional cells +

strongly positive

Thigh M 46 3 + + Negative +

Back M 72 3 + + Most cells strongly + weakly positive

positive

16

1156 M . Leader et al.

A N G I O S A R C O M A S

Seven angiosarcomas were examined (Table 2). For the purpose of recording results the tumours were divided into vasoformative and solid areas. Vasoformative areas were seen in all the tumours and all showed positive granular cytoplasmic staining with both markers (Figure 1). Solid areas were absent in one tumour. F VIII RAg staining was observed in the solid components of four of six cases (Figure 2).

Interpretation of the staining results with anti-UEA-I was more problematical. This was due to the difficulty of distinguishing between the positively staining red blood cells scattered throughout the tumour and the tumour cells. In many cases the use of a X 100 objective lens was required. However, positively staining tumour cells were demonstrated in all six angiosarcomas. In some cases only occasional cells stained. The positively staining cells were generally seen lining clefts within the tumour. Instead of granular cytoplasmic staining, a thin brown line of staining at the edge of the positive cells was accepted as a positive reaction. There was no correlation between the degree of differentiation of the tumour and the number of cells that

Figure 1. Grade 1 angiosarcoma; the vascular spaces are lined by endothelial cdls, many protruding into the vascular lumen. These have stained strongly with F VIII RAg antiserum. ~ 4 0 0 .

Figure 2. Grade 3 angiosarcoma; large pleomorphic cells are seen staining positively for F VIII RAg. x400.

F VII I R A g and U E A - I staining in 230 tumours 1157

stained positively, or the intensity of staining with either marker (Table 2). The seven angiosarcomas were also stained with an anti-epithelial monoclonal antibody CAM 5.2 (Makin, Bobrow & Bodmer 1984) to exclude the possibility of their being pseudo- sarcomatous carcinomas. All showed negative staining (Leader et al. 1986).

K A P O S I ’ S S A R C O M A S

Eight Kaposi’s sarcomas were stained with anti-F VIII RAg. The tumours were categorized for the purpose of recording results into areas with well-formed vessels, areas where tumour cells lined clefts containing red blood cells and spindle cell areas. Positive staining was seen in the vessels and in cells lining clefts in all eight. Seven of the eight tumours showed well-formed spindle cell areas. Positive staining was seen in six (Figure 3), but was equivocal in the seventh.

Six tumours were stained with anti-UEA-I. Positive staining of vessels was seen in all cases but positively staining spindle cells were seen only in three cases. Again a x 100 objective was required to distinguish between the positively staining extrava- sated red blood cells and tumour cells.

Figure 3. Kaposi’s sarcoma; spindle cells show positive staining for F VIII RAg. A gradation in intensity of staining is seen with some cells staining weakly (arrow) and some intensely (arrowhead). X400.

Figure 4. Prostatic carcinoma. Intense staining of tumour cells is seen. Blood vessels show endothelial staining. Anti-UEA-I. X400.

1158 M . Leader et al.

“ON-VASCULAR’ S A R C O M A S

One hundred and eighty-one tumours were stained for F VIII RAg (Table 1). Four showed positive staining. These comprised two rhabdomyosarcomas, one clear cell sarcoma and one undifferentiated sarcoma of uncertain histogenesis. Three of these four tumours had been irradiated prior to the current surgery. One of the rhab- domyosarcomas showed negative staining of the vast majority of the tumour cells. However, focal positive granular staining was seen in a small number of rhab- domyoblasts. These were seen adjacent to an area of necrosis and many of the positive cells showed pyknotic nuclei. In the second rhabdomyosarcoma a considerable num- ber of the rhabdomyoblasts stained positively, but on this occasion many of the cells were viable and showed no degenerative features. It was interesting to note that irradiated viable skeletal muscle included in the block also stained positively.

The clear cell sarcoma was composed of cells with round and ovoid vesicular nuclei and large eosinophilic nucleoli. Many of the cells contained clear cytoplasm whilst others showed eosinophilic cytoplasm. The positive staining was quite focal and some of the nuclei also stained positively. The positively reacting undifferentiated sarcoma was composed of sheets of large ovoid and spindle cells with large vesicular nuclei many of which had prominent nucleoli. The mitotic index was high. A diagnosis of malignant fibrous histiocytoma was suspected but could not be confirmed. Only small numbers of tumour cells showed positive staining, restricted to the cytoplasm.

These four tumours were reacted with pre-adsorbed F VIII RAg antiserum and the positive staining was removed in all of them. In addition pre-radiotherapy biopsies were available from three cases and all stained negatively for F VIII RAg.

When stained with anti-UEA-I, 16 of the 181 non-vascular sarcomas showed positive reactions (Table 1). None of these 16 tumours had stained with anti-F VIII RAg.

C A R C I N O M A S

Twenty carcinomas were stained with F VIII RAg antiserum. All showed negative staining. Twelve of the 20 carcinomas stained with anti-UEA-I showed positive staining. Especially strong staining was seen in well differentiated squamous cell carcinomas, in a prostatic carcinoma (Figure 4), in a large bowel adenocarcinoma and in an endometrial carcinoma.

Discussion

F VIII RAg and UEA-I are recognized markers of endothelial cells (Holthofer et al. 1981, Miettinen et al. 1983). This study set out to assess the specificity of each marker for endothelial derived neoplasms and to evaluate the extent of false positive and false negative reactions associated with each marker.

UEA-I appears to be a sensitive marker of endothelial derived neoplasms as positive staining was seen in all angiosarcomas. Unfortunately this marker was associ- ated with a high false positive rate, especially amongst carcinomas (12/20), but also amongst non-vascular sarcomas. As the differential diagnosis of a grade I11 angiosar-

F VIII RAg and UEA-I staining in 230 tumours 1159

coma commonly includes a poorly differentiated carcinoma, this marker would have little discriminatory value. Other workers have also noted cross-reactivity with carcin- omas (Ordonez & Batsakis 1984, Walker 1984, Lee et al. 1985, Prime et al. 1985). Anti-UEA-I has a further disadvantage; interpretation of its staining reactions with angiosarcomas is much more difficult than interpretation of anti-F VIII RAg staining. This is due to the positive staining of red blood cells by anti-UEA-I.

All workers confirm the finding of F VIII RAg in the endothelial cells of benign haemangiomas (Burgdorf, Mukai & Rosai 1981). Smaller vessels appear more consis- tently positive than larger vessels (Burgdorf et al. 1981). Anti-F VIII RAg staining of lymphatic endothelial cells and of glomerular endothelial cells is more controversial. Some workers report positive staining and some negative staining (Mukai, Rosai & Burgdorf 1980, Sehested & Hou-Jensen 1981, Burgdorf et al. 1981). Sinusoidal lining cells of the liver and spleen stain positively for F VIII RAg (Burgdorf etal. 1981) as do the stromal cells of haemangioblastomas (Jurco et al. 1982) and cardiac myxoma cells (McComb 1984). Anti-F VIII RAg is also useful in highlighting vascular invasion by tumour cells (Bettelheim, Mitchell & Gusterson 1984). It is not detected in adeno- matoid tumours (Bell & Flotte 1982, Said, Nash & Lee 1982).

F VIII RAg can only be regarded as a useful marker in soft tissue tumour sar- comas if it specifically stains vascular tumours. It must also be demonstrable within the solid areas of these tumours. In this study solid foci in only four of six angiosarcomas stained positively. Therefore a negative result does not exclude a diagnosis of angio- sarcoma. Review of the literature confirms the limitations of anti-F VIII RAg staining of angiosarcomas. In various studies 23/24 tumours (Capo et al. 1985), 15/36 tumours (Pfaltz et al. 1983) and 13/20 tumours (Ruchti, Gerber & Schaffner 1984) stained positively. However, these authors do not comment as to whether the positive cells were in vasoformative or solid areas. Guarda et al. (1982), in a study of 27 angiosar- comas, found positivity within solid areas of 5/22 tumours. There is no clear-cut correlation between the degree of positive staining and tumour differentiation (Ruchti et al. 1984). The negative staining of angiosarcomas in our study is not due to fixation or technical problems as non-tumour vessels in all sections stained positively. Nega- tive staining may suggest that the more undifferentiated cells lose their ability to synthesize F VIII RAg. Some workers have shown that I: VIII RAg may exist in a polymeric form (Eckert & Chavin 1977). Perhaps this could explain some negative staining.

F VIII RAg is useful in making a diagnosis of Kaposi’s sarcoma. Six of seven of these tumours showed positively staining spindle cells. Staining was equivocal in the seventh. Taking a combined review of eight papers documenting the F VIII RAg staining patterns of Kaposi’s sarcomas, three groups of workers found that the spindle cells of the tumours stained negatively (Burgdorf et al. 1981, Sehested & Hou-Jensen 1981, Millard & Heryet 1985). The five other groups found the spindle cells in many of the tumours studied stained positively (Nadji et al. 1981, Guarda et al. 1981, Flotte, Hatcher & Friedman-Kien 1984, Ahktar et al. 1984, Ordonez & Batsakis 1984). Millard & Heryet (1985), in their review, conclude that the positive staining found by some workers is due to an over-zealous interpretation of a positive result. They define a positive reaction as a granular brown appearance in the cell cytoplasm. Our

1160 M . Leader et al.

positively staining Kaposi’s sarcomas fulfilled this definition. We therefore consider that these tumours do stain positively for F VIII RAg. Other workers have also demonstrated UEA-I receptors in Kaposi’s sarcomas (Ordonez & Batsakis 1984).

There was a ‘false positive rate’ of 2.2% with anti-F VIII RAg amongst the non- vascular sarcomas. On reviewing the histopathology of these four cases, we confirmed the earlier diagnoses. Rhabdomyoblasts were visible in the two rhabdomyosarcomas. The clear cell sarcoma stained positively with Perk’ stain, anti-S100 protein and neurone specific enolase antiserum. The fourth case showed no vasoformative areas or haemosiderin deposition and was thought to present a malignant fibrous histo- cytoma. Staining of the four cases with pre-adsorbed antiserum was negative and this suggests that the positive results with anti-F VIII RAg were genuine. N o single cause adequately explains these false positive results. However, it may be significant that three cases had been previously irradiated. The fourth case was an amputation specimen, but we were unable to discover whether prior radiotherapy had been given or not. It is theoretically possible that irradiation may have damaged the tumour cell epitopes and accounted for the false positive staining. In support of this is the fact that irradiated skeletal muscle included in one of the sections also stained positively and that pre-radiotherapy biopsies in the three cases stained negatively with anti-F VIII RAg. Therefore, a positive result with F VIII RAg antiserum in a tumour which has been irradiated should be interpreted with caution.

Other immunological techniques described for the identification of endothelial cells include anti-actomyosin and anti-tissue factor (Burgdorf et al. 1981) but both are limited in value by the fact that they require the use of fresh tissue. i n conclusion, it would seem that the ideal marker of endothelial cells is not yet available. F VIII RAg antiserum will stain the vasoformative areas of angiosarcomas in all cases but is associated with a 33% ‘false negative’ rate in staining the solid components of these tumours. Despite this high false negative rate it is quite a specific marker being associated with only a 2.2% ‘false positive’ rate. UEA-I lectin by contrast stains a higher percentage of angiosarcomas but is associated with a high false positive rate in the staining of carcinomas. In addition to their role in the diagnosis of angiosarcomas, both markers seem to be of use in the diagnosis of Kaposi’s sarcomas.

Acknowledgements

We are most grateful to Professor D.H.MacKenzie who very kindly allowed us full access to his collection of soft tissue sarcomas. Without his diagnostic expertise this unique collection of cases would not have been available to us and this study could not have taken place. We are indebted to DAKO Pharmaceuticals who provided us with antisera. Dr Leader is in receipt of a grant from the North West Thames Area Health Authority and the Westminster Hospital Research Trust.

References

AHKTAR M., BUNUAN H., ALI M.A. & GODWIN J.T. (1984) Kaposi’s sarcoma in renal transplant reci- pients: Ultrastructural and irnrnunoperoxidase study of four cases. Cancer 53, 258-266

F VZZZ RAg and UEA-I staining in 230 turnours 1161

BELL D.A. & FLOTE T.J. (1982) Factor VIII related antigen in adenomatoid tumours. Cancer SO, 932- 938

BETTELHEIM R., MITCHELL D. & GUSTERSON B.A. (1984) lmmunocytochemistry in the identification of vascular invasion in breast cancer. Journal of Clinical Pathology 37, 364-366

BURCDORF W.H., MUKAI K. & ROSA[ J . (1981) Immunohistochemical identification of Factor VIII related antigen in endothelial cells of alleged vascular nature. America/ Journalof Clinical Pathology 75,167- 171

CAPO U., OZZELLO L., FENOCLID C.M., LOMBARDI L. & RILKE F. (198.5) Angiosarcomas arising in oedematous extremities: Imrnunostaining for Factor VIIl related antigen and ultrastructural fea- tures. Human Pathology 16, 144-150

ECKERT H. & CHAVIN S.I . (1977) Electrophoreticstudies of human Factor VIII related antigen. Biochemi- cal Society Transactions 5, 1439-141 1

FLOTTET.J., HATCHERV.A. & FRIEDMAN-KIEN A.E. (1984) Factor VlII RAgin Kaposi’ssarcomain young homosexual men. Archives of Dermatology 120, 180-182

GUARDA L.G., SILVA E.G., ORDONEZ N.G. & SMITH J.L. (1982) Factor VIII in Kaposi’s sarcoma. American Journal of Clinical Pathology 76, 197-200

HOLTHOFER H., VIRTANEN I., PETERSSON E., TORNROTH T., ALFTHAN O., LINDER E. & MIETTINEN A. (1981) Lectins as fluorescence markers for saccharides in human kidney. Laboratory Investigation 45,

JURCOS.,NADJI M., HARVEY D.G., P A R K E R J . ~ . , FoNTR.L. &MORALESA.R. (1982) Hemangioblastomas: Histogenesis of the stromal cell studied by immunocytochemistry. Human Pathology 13, 13-18

LEADER M., PATEL I., COLLINS M. & HENRY K. (1986) An analysis of the sensitivity and specificity of the cytokeratin marker CAM 5.2 for epithelial tumours. Results of a study of 202 sarcomas, 50 carcin- omas and 28 malignant melanomas. In press

LEE A.K., DELELLIS R.A., ROSEN P.P., SAICO P.E., GANCI M.D., BAGIN R., GROSHEN S. & WOLFE H.J . (1985) ABH blood group isoantigen expression in breast carcinomas-an immunohistochernical evaluation using monoclonal antibodies. American Journal of Clinical Pathology 83, 308-319

MACKENZ~E D.H. (1970) The Differentiat Diagnosis of Fibroblastic Disorders. Blackwell Scientific Publications, Oxford

MAKIN C.A., BOBROW L.G. & BODMER W.F. (1984) Monoclonal antibody to cytokeratin for use in routine histopathology. Journal of Clinical Pathology 37, 975-983

MCCOMB R.D. (1984) Heterogeneous expression of Factor VIlIiVon Willebrand factor by cardiac myx- oma cells. American Journal of Surgical Pathology 8, 539-544

MlETrlNEN M., HOLTHOFERH., LEHTO V.P., MIETTINEN A . & VIRTANEN I. (1983) Ulexeuropaeus I lectin as a marker for tumours derived from endothelial cells. American Journal of Clinical Pathology 79,

MILLARD P.R. & HERYET A.R. (1985) An immunohistological study of Factor VIlI related antigen and Kaposi’s sarcoma using polyclonal and monoclonal antibodies. Journal of Pathology 146, 31-38

MUKAI K. , ROSA] J . & BURCDORF H.C. (1980) Localization of Factor VIII related antigen in vascular endothelial cells using an irnmunoperoxidase method. American Journal of Surgical Pathology 4,

NADJI M., MORALES A.R., ZIECLES-WEISSMAN J. & PENNEYS N.S. (1981) Kaposi’s sarcoma. Immuno- histological evidence for an endothelial origin. Archives of Pathology and Laboratory Medicine 105, 274-275

ORDONEZ N.G. & BATSAKIS J.G. (1984) Comparison of UIex europaeus 1 lectin and Factor VIlI related antigen in vascular lesions. Archives of Pathology and Laboratory Medicine 108, 129-132

PFALTZ M., HEDINCER C. , SAREMASLANI P. & ECLOFF B. (1983) Malignant haemangioendothelioma of the thyroid and Factor VIII related antigen. Virchows Archiv. A . Pafhological Anafomy and Hirtopatho- logy 401, 177-184

PRIME S.S., ROSSER T.J., MALAMOS D., SHEPHERD J.P. & SCULLY C. (1985) The use of t h e lectin Ll1e.r europaew to study epithelial cell differentiation in neoplastic and non-neoplastic oral white lesions. Journal of Pathology 147, 173-179

RAMAEKERS F.C.S., PUTTS J.J.G., MOESKER O., KANT A., HUYSMANS A, , HAAC D., JAP P.H.K, HERMAN

391-399

32-36

273-216

1162 M . Leader et al.

C.J. & VooiJs G.P. (1983) Antibodies to intermediate filament proteins in the immunohistochemical identification of human tumours: an overview. Hisiochemical Journal 15,691-713

RUCHTI G., GERBER H.A. & SCHAFFNER T. (1984) Factor VIIl related antigen in malignant haemangioen- dothelioma of the thyroid. Additional evidence for the endothelial origin of this tumour. American Journal of Clinical Pathology 82, 474-480

SAID J.W., NASH G. & LEE M. (1982) Immunoperoxidase localization of keratin proteins, car- cinoernbryonic antigen and Factor VIII in adenomatoid tumours. Human Pathology 13, 1106-1180

SEHESTED M. & HOU-JENSEN K. (1981) Factor VIII related antigen as an endothelial cell marker in benign and malignant diseases. Virchows Archiv. A . Pathological Anatomy and Histopathology 391,218-255

WALKER R.A. (1984) The binding of peroxidase-labelled lectins to human breast epithelium. 111. Altered fucose-binding patterns of breast carcinomas and their significance. Journal of Pathology 144, 101- 117