0952.327&‘9Il/@339-03/SlO.O0
Rosta&nditts Leukotrienes and Essential Fatty Acids (1990) 39,53-57 0 Longman Group UK Ltd 1990
Prostaglandins in Squamous Cell Carcinoma of the Larynx: Tumor and Peritumor Synthesis S. Pinto*, 0. Gallo*, M. Dilaghi’, E. Gallina’, Prisco*, R. Abbate*
A. Giannini* , M. Coppo*, R. Paniccia*, D.
Clinica Medica I, Clinica Otorinolaringoiatrica II and ‘Istituto di Anatomia Patologica, * University of Florence, Viale Morgagni 85, 50134 Florence, Italy (Reprint requests to SP) ABSTRACT.
Prostaglandin (PG) E2, BketoPGFi, and Thromboxane B2 (TxB2) production by the tumor, peritumor and control tissue were investigated in specimens from patients (n=ll) with squamous cell carcinoma of the larynx, in relation to the extension and infiltration of the neoplasm and to the presence of inflammation, fibrosis and necrosis. In all specimens detectable amounts of 6ketoPGF1+ and TxB2 were found, but the predominant metabolite was PGE2. No differences in the levels of TxBz and 6ketoPGFi, were observed, but the only patient with lymphonodal involvement showed the lowest levels of 6ketoPGF,, both in tumor and peritumor tissue. Higher amounts (p< 0.05) of PGEz were synthesized by peritumor tissues in comparison to control mucosa and tumor tissue independently of the occurrence of reactive inliltration. PGs synthesis did not correlate with infhunmation, fibrosis, necrosis or staging of the neoplasm. However the two cases in stage T4 showed PGEz generation at the highest levels both in neoplastic and perineoplastic tissue. These findings indicate that in squamous cell carcinoma of the larynx an increased production of PGEz occurs, stemming not only from inflammatory cells but at least in part from neoplastic cells. This suggests that the study of arachidonic acid metabolism may contribute to characterization of the primary cancer and lead to better understanding of the mechanisms of tumor growth and diffusion.
INTRODUCTION
MATERIALS AND METHODS Subjects
The role of prostaglandins (PGs) has been widely investigated in recent years in relation to the immunologic depression of cancer patients (1, 2). PGE2 has been reported to hamper T-lymphocyte function, natural killer cell activity and antibody dependent cell-mediated cytotoxicity against tumor cells (3-6). Recently, several reports have indicated an excessive monocyte production of PGs in patients with cancer of different organs (7, 8). Moreover PGs may play a role in tumor proliferation, growth and metastatization by inducing activation or release of proteolytic enzymes (collagenase, protease, leucine amino peptidase) from lymphokine-activated macrophages (9) and fibroblasts (10). In head and neck cancer patients, an increased PGE2 production by monocytes from whole blood (11) and high amounts of prostaglandin-like material have been demonstrated in tumor tissue (12, 13). The aim of this study was to investigate the production of arachidonic acid (AA) metabolites by tumor and peritumor tissue in human epidermoid carcinoma of the larynx in order to evaluate the involvement of the different metabolites in relation to the extension and infiltration of tumor mass and the presence of inflammation, fibrosis and necrosis.
The study was carried out on 11 patients, 10 men and 1 woman, aged between 50 and 81 years (mean age 67 f. 11 years), affected by carcinoma of the larynx, who underwent surgical treatment in the II Clinic of Otorhinolaryngology of the University of Florence (Table 1). Nine of eleven patients were affected by tumor of the supraglottic region, one of the subglottic and one of the glottic. Only the patient n.2 had lateral cervical node metastasis, confirmed by histological examination.
Table 1. Clinical characteristics N”
1 2 3 4 5 6 7 8 9 10 11
53
of patients
Sex
be
Tumor
site
M M M M F M M M M M M
81 60 65 57 50 72 77 19 80 63 52
Glottic Supraglottic Supraglottic Supraglottic Subglottic Supraglottic Supraglottic Supraglottic Supraglottic Supraglottic Supraglottic
TNM T3NOMO T4N 1MO T4NOMO T3NOMO T3NOMO T3NOMO T3NOMO T3NOMO T3NOMO T2NOMO T3NOMO
54
Prostaglandins Leukotrienes and Essential Fatty Acids
All patients underwent total (n=7) or partial (n=4) laryngectomy. Patients were free from any drug interfering with AA metabolism in the two weeks preceding the surgical intervention. The histological study of the tumors showed that all cases were squamous cell carcinoma with different degrees of differentiation. The inflammatory reaction within and around the tumor, evaluated at light microscopy, was histologically classified with a semiquantitative score as: slight (+), moderate (+ +), marked (+ + +); “productive” inflammation (when lymphocytes and plasma cells were prevalent) or “exudative” (when granulocytes or/and macrophages were prevalent). Moreover the presence or absence of necrosis and stromal tumor fibrosis were noted.
cross-reactivity with other PGs and related substances was less than 0.3%. The detection limit of the method was 7.5 pg/ml. The CV (intraassay and interassay) were 5.5% and 9.6% respectively. TXAz assay TXA2 was assayed as its stable derivative TXB2 by RIA according to the method of Granstrom et al (16) with a commercial kit (American Biomedical Technologies, Berlin, West Germany). Cross-reactivity of TXB2 antibody was 1.8% with PGD2, less than 0.05% with PGF*,, less than 0.05% with PGEz and less than 0.05% with lZhydroxyeicosatetraenoic acid (12-HETE). The detection limit was 10 pg/ml. The intraassay and interassay CV were 8.7% and 10.4% respectively.
Experimental procedure A specimen (0.5 cm x 0.5 cm) of the tumor tissue (A), of the peritumor tissue (B) and of the healthy mucosa (C), was taken from every patient for PG study. A the specimen was taken within the tumor. B the specimen was taken at the tumor margin. C the specimen .was taken from a macroscopically unaffected area about 3 cm from the tumor. The specimens were rapidly removed and immediately preserved at -80” C. After defrosting, they were washed with tris buffer (50 mM, pH 7.5) and fibrous and fatty tissue were gently removed; then the specimens were carefully blotted, weighed, minced, and incubated with the same buffer at 37” C under continuous stirring for 30 min and subsequently centrifuged for 20 min at 6000 g at 4” C. The supernatant was stored at -80” C and subsequently assayed for PGs.
Statistical analysis Results are given as mean f SD. The statistical analysis was performed using the Wilcoxon rank sum test for paired and unpaired data and polynomial fitting.
RESULTS
PGE2 was measured by radioimmunoassay (RIA) as previously described (14) with a commercial kit (American Biomedical Technologies, Berlin, West Germany). The cross reactivity of PGE2 antiserum was 1.5% with PGA2, 1.2% with PGEl and 1% with PGFr,, while it was less than 0.14% for other substances. The detection limit was 5 pg/ml. The intraassay and interassay coefficient of variation (CV) were 9% and 10% respectively.
The study of AA metabolites showed that detectable amounts of PGE*, 6ketoPGF1, and TxBz were produced by neoplastic and perineoplastic tissue and unaffected mucosa (Table 2). PGE2 was found to be significantly higher in peritumor tissue 32.71 f 29.86 ng/lOO mg of tissue) (p < 0.05) than in normal mucosa (15.14 + 10.11 rig/100 mg of tissue) (Fig. 1). PGE2 production by tumor tissue was not different from that of normal mucosa (Table 2). No significant differences in 6ketoPGF1, or TxB2 production were observed among tumor, peritumor and healthy tissue (Fig. 2, 3). Only in the patient n. 2, affected by lateral cervical node metastasis, was bketoPGF,,j production found to be extremely low, outside the confidence limits (99%) of normal mucosa levels, both in tumor and peritumor tissue. This patient showed the highest level of PGE2 in tumor and peritumor tissue. PGE;! production did not correlate with the occurrence or pattern of inflammatory reaction or with the extent of necrosis. However, PGE2 gener-
6ketoPGF1, assay
Table 2 Eicosanoids production by tumor, peritumor and control tissue.
PGEz assay
Prostacyclin (PGi2) was assayed as its stable metabolite (6ketoPGFiJ by RIA according to Patron0 et al (15) with a commercial kit (American Biomedical Technologies, Berlin, West Germany). The maximum cross reactivity for PGs and related substances was 1.9% with 6ketoPGEz. The
Tumor tissue q/100 mg PGE? 6ketoPGF1, TxB2
17.19 k13.21 0”;; io!;“;48 .
Peritumor tissue ng/lOo mg 32.71_+29.86 2.75 + 1.45 0.714 ?I 0.489
Control tissue nghO0 mg 15.14 *lo.11 2.87 k 0.91 0.756f0.247
Prostaglandins in Squamous Cell Carcinoma of the Larynx 90.
5.5
0 0
60
m 7 .E
70
6
60
*
0 1.2.
8
al a I B
e
4
8
40 -
E" 0
30 -
P
0
cy
0
. 0
0 0 0
*
,------------
*
*
(____+__,_.
I.___________
7-
0
0.6-
0 8
0 0
0.4.
0
*
0.2.
0 I
A
C
Fig. 1 PGEz production by tumor (A), peritumor (B) and control tissue (C). 0 = T2, l = T3, * = T4.
B
C
Fig. 3 TxB2 production by tumor (A), peritumor (B) and control tissue (C). 0 = T2. 0 = T3. * = T4.
No correlation existed between the stage of the neoplasm and the amount of PGs produced; however, in the 2 patients in stage T4, PGE;! was found at very high levels both in peritumor and in tumor tissue (Fig. 1).
*
6-
5-
*
o
0
4-
DISCUSSION
0 0 0
0 . ______ ._____
.@
a
._________*.
l*
0.
0
l
l
*
*
A
o_,$__x_ 0
0..
:
B
C
Fig. 2 6ketoPGF, production by tumor (A), peritumor (B) and control tissue (C). 0 = T2, l = T3, * = T4.
tended to be higher in the tumor (21.95 + 12.58 ngl’100 mg of tissue, n=4) and peritumor (54.8 + 31.68 ng/lOO mg of tissue, n=4) specimens in which necrosis was completely absent in comparison to tumor (14.47 + 13.73 ng/lOO mg of n=7) tissue, and peritumor (21.88 * 23.25 ng/lOO mg of tissue, n=7) samples with necrosis of varying degree. PG production was not related to tumour size or degree of differentiation. In all specimens there was little exudative inflammation. ation
00
:
0
l
B
l-
0
0.0
;
2-
0.6
% :!
.a
3-
*
0 i__-_________
*
0
0
0
8
2
0 P
LO-
Our results indicate that detectable amounts of PGE2 6ketoPGF1, and TxB2 were produced in all specimens of tumor, peritumor tissue and normal mucosa from patients with squamous cell carcinoma of the larynx. PGE2 was the predominant AA metabolite in the three areas and its production was found increased in peritumor tissue in comparison to healthy mucosa. This finding could be attributed to the stromal inflammatory reaction most manifest at the tumor margin, In fact inflammatory cells are able to produce prostaglandins, especially PGE2 (17), and therefore the inflainmatory infiltrate might be the source of enhanced amounts of PGs. However, the absence of a clear relationship between the production of PGs and the entity of reactive infiltrate seems to suggest that the enhanced PC synthesis does not only stem from inflammatory cells but also from neoplastic and/or normal host cells. In fact, in squamous cell carcinoma, PG synthesis takes place, as has been specifically demonstrated by immunofluorescence studies (13). In our study the enhanced PG production at the margin of the neoplasm could, at least in part, be due to tumor
56
Prostaglandins Leukotrienes and Essential Fattv Acids
cells, which are more activated in this area so facilitating a larger neoplasm diffusion (18). Even if the exact role of PGs in neoplastic disease is not yet well defined, PGE2 is known to represent a suppressor signal for many immune functions and to promote growth of the tumoral mass (19). In our study no relationship was found between the stage of the neoplasm and the amounts of PGs produced, but most of patients (8/11) had a tumor classified as T3. It seems noteworthy that the two cases in stage T4 showed a PGE2 generation at the highest levels both in peritumor and tumor tissue. No significant differences in TxB2 and 6ketoPGF1, production were observed among the specimens studied. The importance of the balance of TxB2/PG12 for metastatization has been recently stressed (20); an alteration of this ratio in the direction of TxB2 has been claimed to determine a higher diffusion of the neoplasm, by inducing platelet aggregation which has been demonstrated to promote tumor growth and metastasis (21). Moreover, pretreatment with PGI,? or with a PGIZ production stimulator is reported to inhibit metastatic cells proliferation (22). The absence of alteration of TxB2 and PG12 is in keeping with the absence of metastasis, which was the patient selection criterion for operability. Only in one case had regional lymphonode involvement occurred, and it seems remarkable that in this case the lowest levels of PGI;! were found both in tumor and peritumor tissue, while in the healthy mucosa PG12 production was within the normal range. However, more investigations are needed to confirm the role of the balance between TxB2 and PGI? production. This could have important implications for a therapeutic approach which aims to affect separately the individual AA metabolites. The administration of cyclooxygenase inhibitors in head, neck and mammary cancer in order to counteract the effects of PGE* has led to contradictory results such as a regression of primary neoplasm on one side and an increased incidence of metastasis on the other (2325). In this study no relationship was observed between PG synthesis, tumor necrosis and stromal fibrosis, the latter being scarcely represented in all specimens. As far as necrosis is concerned, the absence of significant differences in PGE2 synthesis between the specimens with and without necrosis could be attributed to the smallness of the two groups. our findings indicate that in In conclusion, squamous cell carcinoma of the larynx an increased
PGE2 production occurs in peritumor tissue, suggesting that the study of the eicosanoid profile could with the other morphological and contribute, biological parameters, to characterization of the primary neoplasm and understanding of the
mechanisms of tumor growth, metastasis development and host immune response. Moreover, a better knowledge of the interaction between PGs and neoplasm may be useful in making a therapeutic choice among drugs interfering with AA metabolism. References 1. Plescia 0 J. Smith H, Grinwich K. Subversion of immune system by tumor cells and role of prostaglandins. Proceedings of the National Academy of Sciences of the United States of America 72: 1848-1851, 1975. 2. Goodwin J S. Prostaglandins and host defense in cancer. Medical Clinics of North America 65: 829-844,198l. 3. Goodwin J S, Webb D R. Regulation of the immune response by prostaglandins. Clinical Immunology and Immunopathology 115: 106-122, 1980. 4. Brunda M J, Heberman R B, Holden T H. lnhibition of murine natural killer cell activity by prostaglandins. Journal of Immunology 124: 2682-2686,198O. 5. Leung K H, Koren H S. Protective effect of interferon on NK cells from suppression by PGE2. Journal of Immunology 129: 1742-1747, 1983. 6. Droller M J. Schneider M N, Perlmann P. A possible role of prostaglandins in the inhibition of natural and antibody-dependent cell-mediated cytotoxicity against tumor cells. Cellular Immunology 39: 165-177, 1978. 7. Balch C H, Dougherty P A, Tilden A B. Excessive PGE? production by suppressor monocytes in head and neck cancer patients. Annals of Surgery 196: 645-648, 1982. 8. Goodwin J S, Messner R P, Bankhurst A D. Prostaglandin-producing suppressor cells in Hodgkin’s disease. New England Journal of Medicine 297: 963-68, 1977. 9. Wah! L M, Wahl S M, Meryenhage S E, Martin G R. Collagenase production by lymphokine-activated macrophages. Science 187: 261-263.1975. 10. Huang C C, Blitzer A, Abramson M. Collagenase in human head and neck tumors and rat tumors and fibroblasts in monolayer cultures. Annals of Otology, Rhinology and Laryngology 95: 158-161. 1986. 11. Berlinger N T. Deficient immunity in head and neck cancer due to excessive monocyte production of prostaglandins. Laryngoscope 94: 1407-1410, 1984. 12. Bennet A, Cater R L, Stamford I F, Tanner N S B. Prostaglandins like material extracted from squamous carcinomas of the head and neck. British
Journal of Cancer 41: 204-208, 1980. 13. Jung T T K, Berlinger N T, Juhn S K. Prostaglandins in squamous cell carcinoma of the head and neck. A preliminary study. Laryngoscope 95: 307-312.1985. 14 Neri Serneri G G. Gensini G F, Abbate R, Prisco D.. Roeasi P G. Castellani S, Casolo G C, Matucci ’ M. Fanyini F, Di Donato M, Dabizzi R P. Spontaneous and cold pressure test-induced prostaglandin biosynthesis by human heart. American Heart Journal 110: 50-55, 1985. 15. Patron0 C. Pueliese F. Ciabattoni G. Patrienani P. Maseri A. Chi&chia S, Peskar B A, Cinotc G A, Simonetti B M, Pierucci A. Evidence for a direct stimulatory effect of prostacyclin on renin release in man. Journal of Clinical Investigation 69: 231-239. 1982.
Prostaglandins 16. Granstrom E, Kindahl H, Samuelsson B. A radioimmunoassay for thromboxane B2. Analytical Letters 9: 611-627, 1976. 17. Lewis G P. Immunoregulatory activity of metabolites of arachidonic acid and their role in inflammation. British Medical Bulletin 39: 243-248, 1983. 18. Heppner G H. Tumor heterogeneity. Cancer Research 44: 2259-2265, 1984. 19. Bennet A, Houghton J, Leaper D J. Stamford I F. Cancer growth, response to treatment, and survival time in mice: beneficial effect of the prostaglandin-synthesis inhibitor flurbiprofen Prostaelandins 17: 179-191. 1979. 20. Honn K V, Dunn J R, Meyer J. Thromboxanes and prostacyclin: positive and negative modulators of tumor cell prolifer$ion. p 375 in Prostaglandins and Cancer. First International Conference. New York. (Powles T J, Bockman R S, Honn K V, Ramwell P eds) Alan R Liss, Inc. New York, 1982. 21. Honn K V, Bockman R S, Marnett L J. Prostaglandins and cancer: a review of tumor
22.
23.
24
25.
in Squamous Cell Carcinoma of the Larynx
initiation through tumor metastasis. Prostaglandins 21: 833-864,198l. Honn K V. Prostacycliflromboxane ratios in tumor growth and metastasis. p 733 in Prostaglandins and Cancer. First International Conference. New York. (Powles T J, Bockman R S. Honn K V. Ramwell P eds) Alan R Liss. Inc, New York, 1982. Panje W R. Regression of head and neck carcinoma with a prostaglandin-synthesis inhibitor. Archives of Otolaryngology 107: 658-663, 1981. Rolland P H, Martin P M, Jacquemier J, Rolland M A, Tooga M. Prostaglandin in human breast cancer: evidence suggesting that an elevated Pg production is a marker of high metastatic potential for neoplastic cells. Journal of the National Cancer Institute 64: 1061-1070. 1980. Blitzer A. Huang C C. The effect of indomethacin on the growth of epidermoid carcinoma of the palate in rats. Archives of Otolaryngology 109: 719-723. 1983.
57
Rosta&nditts Leukotrienes and Essential Fatty Acids (1990) 39,53-57 0 Longman Group UK Ltd 1990
Prostaglandins in Squamous Cell Carcinoma of the Larynx: Tumor and Peritumor Synthesis S. Pinto*, 0. Gallo*, M. Dilaghi’, E. Gallina’, Prisco*, R. Abbate*
A. Giannini* , M. Coppo*, R. Paniccia*, D.
Clinica Medica I, Clinica Otorinolaringoiatrica II and ‘Istituto di Anatomia Patologica, * University of Florence, Viale Morgagni 85, 50134 Florence, Italy (Reprint requests to SP) ABSTRACT.
Prostaglandin (PG) E2, BketoPGFi, and Thromboxane B2 (TxB2) production by the tumor, peritumor and control tissue were investigated in specimens from patients (n=ll) with squamous cell carcinoma of the larynx, in relation to the extension and infiltration of the neoplasm and to the presence of inflammation, fibrosis and necrosis. In all specimens detectable amounts of 6ketoPGF1+ and TxB2 were found, but the predominant metabolite was PGE2. No differences in the levels of TxBz and 6ketoPGFi, were observed, but the only patient with lymphonodal involvement showed the lowest levels of 6ketoPGF,, both in tumor and peritumor tissue. Higher amounts (p< 0.05) of PGEz were synthesized by peritumor tissues in comparison to control mucosa and tumor tissue independently of the occurrence of reactive inliltration. PGs synthesis did not correlate with infhunmation, fibrosis, necrosis or staging of the neoplasm. However the two cases in stage T4 showed PGEz generation at the highest levels both in neoplastic and perineoplastic tissue. These findings indicate that in squamous cell carcinoma of the larynx an increased production of PGEz occurs, stemming not only from inflammatory cells but at least in part from neoplastic cells. This suggests that the study of arachidonic acid metabolism may contribute to characterization of the primary cancer and lead to better understanding of the mechanisms of tumor growth and diffusion.
INTRODUCTION
MATERIALS AND METHODS Subjects
The role of prostaglandins (PGs) has been widely investigated in recent years in relation to the immunologic depression of cancer patients (1, 2). PGE2 has been reported to hamper T-lymphocyte function, natural killer cell activity and antibody dependent cell-mediated cytotoxicity against tumor cells (3-6). Recently, several reports have indicated an excessive monocyte production of PGs in patients with cancer of different organs (7, 8). Moreover PGs may play a role in tumor proliferation, growth and metastatization by inducing activation or release of proteolytic enzymes (collagenase, protease, leucine amino peptidase) from lymphokine-activated macrophages (9) and fibroblasts (10). In head and neck cancer patients, an increased PGE2 production by monocytes from whole blood (11) and high amounts of prostaglandin-like material have been demonstrated in tumor tissue (12, 13). The aim of this study was to investigate the production of arachidonic acid (AA) metabolites by tumor and peritumor tissue in human epidermoid carcinoma of the larynx in order to evaluate the involvement of the different metabolites in relation to the extension and infiltration of tumor mass and the presence of inflammation, fibrosis and necrosis.
The study was carried out on 11 patients, 10 men and 1 woman, aged between 50 and 81 years (mean age 67 f. 11 years), affected by carcinoma of the larynx, who underwent surgical treatment in the II Clinic of Otorhinolaryngology of the University of Florence (Table 1). Nine of eleven patients were affected by tumor of the supraglottic region, one of the subglottic and one of the glottic. Only the patient n.2 had lateral cervical node metastasis, confirmed by histological examination.
Table 1. Clinical characteristics N”
1 2 3 4 5 6 7 8 9 10 11
53
of patients
Sex
be
Tumor
site
M M M M F M M M M M M
81 60 65 57 50 72 77 19 80 63 52
Glottic Supraglottic Supraglottic Supraglottic Subglottic Supraglottic Supraglottic Supraglottic Supraglottic Supraglottic Supraglottic
TNM T3NOMO T4N 1MO T4NOMO T3NOMO T3NOMO T3NOMO T3NOMO T3NOMO T3NOMO T2NOMO T3NOMO
54
Prostaglandins Leukotrienes and Essential Fatty Acids
All patients underwent total (n=7) or partial (n=4) laryngectomy. Patients were free from any drug interfering with AA metabolism in the two weeks preceding the surgical intervention. The histological study of the tumors showed that all cases were squamous cell carcinoma with different degrees of differentiation. The inflammatory reaction within and around the tumor, evaluated at light microscopy, was histologically classified with a semiquantitative score as: slight (+), moderate (+ +), marked (+ + +); “productive” inflammation (when lymphocytes and plasma cells were prevalent) or “exudative” (when granulocytes or/and macrophages were prevalent). Moreover the presence or absence of necrosis and stromal tumor fibrosis were noted.
cross-reactivity with other PGs and related substances was less than 0.3%. The detection limit of the method was 7.5 pg/ml. The CV (intraassay and interassay) were 5.5% and 9.6% respectively. TXAz assay TXA2 was assayed as its stable derivative TXB2 by RIA according to the method of Granstrom et al (16) with a commercial kit (American Biomedical Technologies, Berlin, West Germany). Cross-reactivity of TXB2 antibody was 1.8% with PGD2, less than 0.05% with PGF*,, less than 0.05% with PGEz and less than 0.05% with lZhydroxyeicosatetraenoic acid (12-HETE). The detection limit was 10 pg/ml. The intraassay and interassay CV were 8.7% and 10.4% respectively.
Experimental procedure A specimen (0.5 cm x 0.5 cm) of the tumor tissue (A), of the peritumor tissue (B) and of the healthy mucosa (C), was taken from every patient for PG study. A the specimen was taken within the tumor. B the specimen was taken at the tumor margin. C the specimen .was taken from a macroscopically unaffected area about 3 cm from the tumor. The specimens were rapidly removed and immediately preserved at -80” C. After defrosting, they were washed with tris buffer (50 mM, pH 7.5) and fibrous and fatty tissue were gently removed; then the specimens were carefully blotted, weighed, minced, and incubated with the same buffer at 37” C under continuous stirring for 30 min and subsequently centrifuged for 20 min at 6000 g at 4” C. The supernatant was stored at -80” C and subsequently assayed for PGs.
Statistical analysis Results are given as mean f SD. The statistical analysis was performed using the Wilcoxon rank sum test for paired and unpaired data and polynomial fitting.
RESULTS
PGE2 was measured by radioimmunoassay (RIA) as previously described (14) with a commercial kit (American Biomedical Technologies, Berlin, West Germany). The cross reactivity of PGE2 antiserum was 1.5% with PGA2, 1.2% with PGEl and 1% with PGFr,, while it was less than 0.14% for other substances. The detection limit was 5 pg/ml. The intraassay and interassay coefficient of variation (CV) were 9% and 10% respectively.
The study of AA metabolites showed that detectable amounts of PGE*, 6ketoPGF1, and TxBz were produced by neoplastic and perineoplastic tissue and unaffected mucosa (Table 2). PGE2 was found to be significantly higher in peritumor tissue 32.71 f 29.86 ng/lOO mg of tissue) (p < 0.05) than in normal mucosa (15.14 + 10.11 rig/100 mg of tissue) (Fig. 1). PGE2 production by tumor tissue was not different from that of normal mucosa (Table 2). No significant differences in 6ketoPGF1, or TxB2 production were observed among tumor, peritumor and healthy tissue (Fig. 2, 3). Only in the patient n. 2, affected by lateral cervical node metastasis, was bketoPGF,,j production found to be extremely low, outside the confidence limits (99%) of normal mucosa levels, both in tumor and peritumor tissue. This patient showed the highest level of PGE2 in tumor and peritumor tissue. PGE;! production did not correlate with the occurrence or pattern of inflammatory reaction or with the extent of necrosis. However, PGE2 gener-
6ketoPGF1, assay
Table 2 Eicosanoids production by tumor, peritumor and control tissue.
PGEz assay
Prostacyclin (PGi2) was assayed as its stable metabolite (6ketoPGFiJ by RIA according to Patron0 et al (15) with a commercial kit (American Biomedical Technologies, Berlin, West Germany). The maximum cross reactivity for PGs and related substances was 1.9% with 6ketoPGEz. The
Tumor tissue q/100 mg PGE? 6ketoPGF1, TxB2
17.19 k13.21 0”;; io!;“;48 .
Peritumor tissue ng/lOo mg 32.71_+29.86 2.75 + 1.45 0.714 ?I 0.489
Control tissue nghO0 mg 15.14 *lo.11 2.87 k 0.91 0.756f0.247
Prostaglandins in Squamous Cell Carcinoma of the Larynx 90.
5.5
0 0
60
m 7 .E
70
6
60
*
0 1.2.
8
al a I B
e
4
8
40 -
E" 0
30 -
P
0
cy
0
. 0
0 0 0
*
,------------
*
*
(____+__,_.
I.___________
7-
0
0.6-
0 8
0 0
0.4.
0
*
0.2.
0 I
A
C
Fig. 1 PGEz production by tumor (A), peritumor (B) and control tissue (C). 0 = T2, l = T3, * = T4.
B
C
Fig. 3 TxB2 production by tumor (A), peritumor (B) and control tissue (C). 0 = T2. 0 = T3. * = T4.
No correlation existed between the stage of the neoplasm and the amount of PGs produced; however, in the 2 patients in stage T4, PGE;! was found at very high levels both in peritumor and in tumor tissue (Fig. 1).
*
6-
5-
*
o
0
4-
DISCUSSION
0 0 0
0 . ______ ._____
.@
a
._________*.
l*
0.
0
l
l
*
*
A
o_,$__x_ 0
0..
:
B
C
Fig. 2 6ketoPGF, production by tumor (A), peritumor (B) and control tissue (C). 0 = T2, l = T3, * = T4.
tended to be higher in the tumor (21.95 + 12.58 ngl’100 mg of tissue, n=4) and peritumor (54.8 + 31.68 ng/lOO mg of tissue, n=4) specimens in which necrosis was completely absent in comparison to tumor (14.47 + 13.73 ng/lOO mg of n=7) tissue, and peritumor (21.88 * 23.25 ng/lOO mg of tissue, n=7) samples with necrosis of varying degree. PG production was not related to tumour size or degree of differentiation. In all specimens there was little exudative inflammation. ation
00
:
0
l
B
l-
0
0.0
;
2-
0.6
% :!
.a
3-
*
0 i__-_________
*
0
0
0
8
2
0 P
LO-
Our results indicate that detectable amounts of PGE2 6ketoPGF1, and TxB2 were produced in all specimens of tumor, peritumor tissue and normal mucosa from patients with squamous cell carcinoma of the larynx. PGE2 was the predominant AA metabolite in the three areas and its production was found increased in peritumor tissue in comparison to healthy mucosa. This finding could be attributed to the stromal inflammatory reaction most manifest at the tumor margin, In fact inflammatory cells are able to produce prostaglandins, especially PGE2 (17), and therefore the inflainmatory infiltrate might be the source of enhanced amounts of PGs. However, the absence of a clear relationship between the production of PGs and the entity of reactive infiltrate seems to suggest that the enhanced PC synthesis does not only stem from inflammatory cells but also from neoplastic and/or normal host cells. In fact, in squamous cell carcinoma, PG synthesis takes place, as has been specifically demonstrated by immunofluorescence studies (13). In our study the enhanced PG production at the margin of the neoplasm could, at least in part, be due to tumor
56
Prostaglandins Leukotrienes and Essential Fattv Acids
cells, which are more activated in this area so facilitating a larger neoplasm diffusion (18). Even if the exact role of PGs in neoplastic disease is not yet well defined, PGE2 is known to represent a suppressor signal for many immune functions and to promote growth of the tumoral mass (19). In our study no relationship was found between the stage of the neoplasm and the amounts of PGs produced, but most of patients (8/11) had a tumor classified as T3. It seems noteworthy that the two cases in stage T4 showed a PGE2 generation at the highest levels both in peritumor and tumor tissue. No significant differences in TxB2 and 6ketoPGF1, production were observed among the specimens studied. The importance of the balance of TxB2/PG12 for metastatization has been recently stressed (20); an alteration of this ratio in the direction of TxB2 has been claimed to determine a higher diffusion of the neoplasm, by inducing platelet aggregation which has been demonstrated to promote tumor growth and metastasis (21). Moreover, pretreatment with PGI,? or with a PGIZ production stimulator is reported to inhibit metastatic cells proliferation (22). The absence of alteration of TxB2 and PG12 is in keeping with the absence of metastasis, which was the patient selection criterion for operability. Only in one case had regional lymphonode involvement occurred, and it seems remarkable that in this case the lowest levels of PGI;! were found both in tumor and peritumor tissue, while in the healthy mucosa PG12 production was within the normal range. However, more investigations are needed to confirm the role of the balance between TxB2 and PGI? production. This could have important implications for a therapeutic approach which aims to affect separately the individual AA metabolites. The administration of cyclooxygenase inhibitors in head, neck and mammary cancer in order to counteract the effects of PGE* has led to contradictory results such as a regression of primary neoplasm on one side and an increased incidence of metastasis on the other (2325). In this study no relationship was observed between PG synthesis, tumor necrosis and stromal fibrosis, the latter being scarcely represented in all specimens. As far as necrosis is concerned, the absence of significant differences in PGE2 synthesis between the specimens with and without necrosis could be attributed to the smallness of the two groups. our findings indicate that in In conclusion, squamous cell carcinoma of the larynx an increased
PGE2 production occurs in peritumor tissue, suggesting that the study of the eicosanoid profile could with the other morphological and contribute, biological parameters, to characterization of the primary neoplasm and understanding of the
mechanisms of tumor growth, metastasis development and host immune response. Moreover, a better knowledge of the interaction between PGs and neoplasm may be useful in making a therapeutic choice among drugs interfering with AA metabolism. References 1. Plescia 0 J. Smith H, Grinwich K. Subversion of immune system by tumor cells and role of prostaglandins. Proceedings of the National Academy of Sciences of the United States of America 72: 1848-1851, 1975. 2. Goodwin J S. Prostaglandins and host defense in cancer. Medical Clinics of North America 65: 829-844,198l. 3. Goodwin J S, Webb D R. Regulation of the immune response by prostaglandins. Clinical Immunology and Immunopathology 115: 106-122, 1980. 4. Brunda M J, Heberman R B, Holden T H. lnhibition of murine natural killer cell activity by prostaglandins. Journal of Immunology 124: 2682-2686,198O. 5. Leung K H, Koren H S. Protective effect of interferon on NK cells from suppression by PGE2. Journal of Immunology 129: 1742-1747, 1983. 6. Droller M J. Schneider M N, Perlmann P. A possible role of prostaglandins in the inhibition of natural and antibody-dependent cell-mediated cytotoxicity against tumor cells. Cellular Immunology 39: 165-177, 1978. 7. Balch C H, Dougherty P A, Tilden A B. Excessive PGE? production by suppressor monocytes in head and neck cancer patients. Annals of Surgery 196: 645-648, 1982. 8. Goodwin J S, Messner R P, Bankhurst A D. Prostaglandin-producing suppressor cells in Hodgkin’s disease. New England Journal of Medicine 297: 963-68, 1977. 9. Wah! L M, Wahl S M, Meryenhage S E, Martin G R. Collagenase production by lymphokine-activated macrophages. Science 187: 261-263.1975. 10. Huang C C, Blitzer A, Abramson M. Collagenase in human head and neck tumors and rat tumors and fibroblasts in monolayer cultures. Annals of Otology, Rhinology and Laryngology 95: 158-161. 1986. 11. Berlinger N T. Deficient immunity in head and neck cancer due to excessive monocyte production of prostaglandins. Laryngoscope 94: 1407-1410, 1984. 12. Bennet A, Cater R L, Stamford I F, Tanner N S B. Prostaglandins like material extracted from squamous carcinomas of the head and neck. British
Journal of Cancer 41: 204-208, 1980. 13. Jung T T K, Berlinger N T, Juhn S K. Prostaglandins in squamous cell carcinoma of the head and neck. A preliminary study. Laryngoscope 95: 307-312.1985. 14 Neri Serneri G G. Gensini G F, Abbate R, Prisco D.. Roeasi P G. Castellani S, Casolo G C, Matucci ’ M. Fanyini F, Di Donato M, Dabizzi R P. Spontaneous and cold pressure test-induced prostaglandin biosynthesis by human heart. American Heart Journal 110: 50-55, 1985. 15. Patron0 C. Pueliese F. Ciabattoni G. Patrienani P. Maseri A. Chi&chia S, Peskar B A, Cinotc G A, Simonetti B M, Pierucci A. Evidence for a direct stimulatory effect of prostacyclin on renin release in man. Journal of Clinical Investigation 69: 231-239. 1982.
Prostaglandins 16. Granstrom E, Kindahl H, Samuelsson B. A radioimmunoassay for thromboxane B2. Analytical Letters 9: 611-627, 1976. 17. Lewis G P. Immunoregulatory activity of metabolites of arachidonic acid and their role in inflammation. British Medical Bulletin 39: 243-248, 1983. 18. Heppner G H. Tumor heterogeneity. Cancer Research 44: 2259-2265, 1984. 19. Bennet A, Houghton J, Leaper D J. Stamford I F. Cancer growth, response to treatment, and survival time in mice: beneficial effect of the prostaglandin-synthesis inhibitor flurbiprofen Prostaelandins 17: 179-191. 1979. 20. Honn K V, Dunn J R, Meyer J. Thromboxanes and prostacyclin: positive and negative modulators of tumor cell prolifer$ion. p 375 in Prostaglandins and Cancer. First International Conference. New York. (Powles T J, Bockman R S, Honn K V, Ramwell P eds) Alan R Liss, Inc. New York, 1982. 21. Honn K V, Bockman R S, Marnett L J. Prostaglandins and cancer: a review of tumor
22.
23.
24
25.
in Squamous Cell Carcinoma of the Larynx
initiation through tumor metastasis. Prostaglandins 21: 833-864,198l. Honn K V. Prostacycliflromboxane ratios in tumor growth and metastasis. p 733 in Prostaglandins and Cancer. First International Conference. New York. (Powles T J, Bockman R S. Honn K V. Ramwell P eds) Alan R Liss. Inc, New York, 1982. Panje W R. Regression of head and neck carcinoma with a prostaglandin-synthesis inhibitor. Archives of Otolaryngology 107: 658-663, 1981. Rolland P H, Martin P M, Jacquemier J, Rolland M A, Tooga M. Prostaglandin in human breast cancer: evidence suggesting that an elevated Pg production is a marker of high metastatic potential for neoplastic cells. Journal of the National Cancer Institute 64: 1061-1070. 1980. Blitzer A. Huang C C. The effect of indomethacin on the growth of epidermoid carcinoma of the palate in rats. Archives of Otolaryngology 109: 719-723. 1983.
57
