Combination assay for tumor markers in oral squamous cell carcinoma

Combination assay for tumor markers in oral squamous cell carcinoma

J Oral Maxillofac Surg 55964-966, 1997 Combination Assay for Tumor Markers in Oral Squamous Cell Carcinoma HIDE0 KUROKAWA, SHINGO TOKUDOME, DDS, PHD...

326KB Sizes 0 Downloads 19 Views

J Oral Maxillofac Surg 55964-966, 1997

Combination Assay for Tumor Markers in Oral Squamous Cell Carcinoma HIDE0 KUROKAWA, SHINGO TOKUDOME,

DDS, PHD,” YOSHIHIRO YAMASHITA, DDS,t DDS,$ AND MINORU KAJIYAMA, DDS, MD5

Purpose: The study evaluated use of a combination assay of tumor markers in the diagnosis of oral squamous cell carcinoma. Patients and Methods: Serum levels of four tumor markers (carcinoembryonit antigen [CEA], squamous cell carcinoma antigen [SCCA], immunosuppressive acidic protein [IAP], and cytokeratin 19 fragment [Cyfra]) were simultaneously measured in 42 patients with oral squamous cell carcinoma (0-SCC) and in 12 patients with oral benign diseases. Results: The positive rates were 31.0% for CEA, 38.1% for SCCA, 52.4% for IAP, and 38.1% for Cyfra in patients with 0-SCC. These rates were significantly different (P < .Ol) from those of control patients with oral benign diseases. The sensitivity (81 .O%) and accuracy (77.8%) of the combination assay uses higher than that obtained with individual markers. Conclusion: A combination assay with CEA, SCCA, IAP, and Cyfra may be useful for the screening of patients with suspected oral squamous cell carcinoma.

Tumor markers are tumor-mediated substances produced by tumor tissue.’ The use of markers has been accepted as a valuable tool for diagnosis, determining such prognosis, and for treatment monitoring.2-6 The tumor markers that have been used in patients with oral squamous cell carcinoma (0-SCC) include carcinoembryonic antigen (CEA),7,8 squamous cell carcinoma antigen (SCCA),7%9 immunosuppressive acidic protein (IAP),7 and cytokeratin 19 fragment (Cyfra).” However, all of these markers have low sensitivity in 0-SCC and therefore are not useful for clinical purposes. Most authors have measured single tumor markers,2-6 and there are very few reports concerning the use of combination assays in which two or more markers were measured simultaneously.“~‘2 In this study

various tumor markers were analyzed singly and in combination in patients suffering from 0-SCC. Patients

Forty-two consecutive patients with 0-SCC treated in our department between 1990 and 1994 were entered in the study. Blood specimens were collected after informed consent was received. Four tumor markers were simultaneously measured: CEA, SCCA, IAP, and Cyfra. The patients’ ages ranged from 38 to 89 years; 21 were men and 21 were women. The anatomic sites of the tumors and their stage, according to the UICC classification,‘3 are summarized in Table 1 and 2. Sera from patients with benign tumors or cystic lesions who were treated in our department during the study period were used as a control. Serum levels of CEA were determined by radioimmunoassay (CEA Roche Z gel kit, Roche, Tokyo, Japan), originally developed by Thomson et a1.14 The upper normal limit was set at 2.5 ng/mL on the basis of previous reports.’ The sensitivity of the test was 0.5 ng/mL. For the measurement of serum SCCA, a double-antibody radioimmunoassay kit was used (SCC RIA BEAD, Dainabot Co, Tokyo, Japan). An upper

Received from the Second Department of Oral and Maxillofacial Surgery, Kyushu Dental College, Kitakyushu, Japan. * Associated Professor. t Assistant. $ Assistant. $ Professor. Address correspondence and reprint requests to Dr Kurokawa: Second Department of Oral and Maxillofacial Surgery, Kyushu Dental College, 2-6-l Manazuru Kokurakita-ku, Kitakyushu City, 803 Japan. 0 1997 American

Association

of Oral and Maxillofacial

and Methods

Surgeons

0278-2391/97/5509-0011$3.00/O

964

KUROKAWA

ET AL

Table 1. Anatomic Oral SCC Anatomic

Sites of Primary

Region of Cases n (%I

Site

Table 3. Positive Oral SCC Tumor Oral

Tongue Maxillary gingiva Mandibular gingiva Buccal mucosa Floor of mouth

16 (38.1) 1.5 (35.7) 8 (19.0)

The positive rates of all tumor markers are listed in Table 3. As shown, the positive rates for CEA, SCCA, IAP, and Cyfra among patients with 0-SCC were all significantly higher than those in controls. The mean serum CEA level in patients with oral benign diseases (n = 12) was 1.58 t 0.55 ng/mL, whereas in the oral cancer patients (n = 42) it was 1.92 + 1.03 ng/mL, showing a significant difference between the two groups (P < .05). The serum SCCA level was 1.46 $

Stage Classification

1 II III IV

diseases

SCC

CEA

SCCA

IAP

Cyfra

8.3% (l/12)

8.3%

16.7% (202)

15.0%

(l/12)

31.0% ( 13142)

38.1% ( 16142)

52.4% (22142)

(3/20) 38.1% (16142)

1 (2.4)

Results

Stage

benign

in

2 (4.8)

limit of 2.0 ng/mL was considered to be normali The sensitivity of the test was 0.7 ng/mL. For the quantitative assay of IAP in sera, single radial immunodiffusion, as described by Mancini et a1,r6 was used with commercially available agarose plates. These plates, which contain rabbit ant-IAP, were purchased from Kayaku Antibiotics Research Co (Tokyo, Japan). An upper limit of 500 pg/mL was considered to be normal.*’ Serum levels of Cyfra were determined by using a CYFRA 21-1 IRMA kit (Centocor Diagnostic Division, Tokyo, Japan). CYFRA 21-l is a solid-phase immunoradiometric assay based on the two-site sandwich method.17 Using receiver operating characteristic curves (ROC) for Cyfra, a threshold of 2.0 ng/mL was chosen as the upper limit of normal values. This cutoff level corresponded to a sensitivity of 42% and a specificity of 90%. Statistical analysis was done using the arithmetic mean and Student’s t-test. A value of P < .OSwas considered to be significant. Differences in the positive rate for each tumor marker between the two groups were evaluated by the x2 test.

Table 2.

Oral

Markers

Rates of Tumor Markers

0.47 ng/mL in the patients with oral benign diseases (n = 12) and 1.81 t 2.08 ng/mL in the oral cancer patients (n = 42); this was significantly higher than in the controls (P < .05). The mean serum level of IAP in patients with oral cancer (n = 42; 523.3 t 216.9 pg/mL) also was significantly higher than in patients with benign disease (n = 12; 376.7 ? 107.0 pg/mL; P < .Ol). Concentrations of Cyfra were significantly higher in patients with 0-SCC (n = 42; 2.25 2 2.76 ng/mL) compared with those with oral benign disease (P < .Ol). These findings are summarized in Figures 1 and 2. The diagnostic sensitivity, specificity, and accuracy of CEA, SCCA, IAP, and Cyfra, either as single indicators or in various combinations, were determined according to the method of Kurokawa et al. The combination assays are presented in Table 4. The sensitivity and accuracy were relatively low when each marker was assessed separately; however, when the four tumor markers were used in combination, the sensitivity (81.0%) and accuracy (77.8%) of the assay were at their highest. High sensitivity and accuracy levels could be obtained when patients who tested positive for any of the four markers were included in the calculations; however, the specificity was low in this combination assay. Discussion Many efforts have been dedicated to establishing a specific marker for the diagnosis and follow-up of pa-

of Oral SCC Cases n (%I 9 (21.4)

12 (28.6) 9 (21.4) 12 (28.6)

FIGURE

1.

Serum

levels of CEA

and SCCA.

COMBINATION

tients with O-SCC.2-7,10,‘2 There are, however, only a few reports concerning the use of combination assays,7x11that is, the simultaneous determination of several tumor markers for diagnostic, monitoring, or treatment evaluation purposes in cancer patients. In recent years, Cyfra has been reported as a new marker in non-small cell carcinoma and squamous cell carcinoma of lung, and uterine cervix carcinoma.” Cytokeratins are intermediate filaments that are part of the cytoskeleton in both normal epithelia and their malignant counterparts, and they are a useful marker of epithelial differentiation.i7 Pujol et alI7 reported that 63% of subjects with squamous cell carcinoma of the lung have Cyfra levels above 3.6 ng/mL, and Doweck et al” found that 60% of 20 patients with squamous cell carcinoma of head and neck had Cyfra levels above 1.3 ng/mL. In this study, the threshold was set al 2.0 ng/mL because ROC analysis showed that this threshold allowed the best sensitivity-specificity relationship. Using a cut-off of 2.0 ng/mL, positive values were observed in 38.1% of untreated patients with 0-SCC. This is comparatively low considering a sensitivity of more than 50% in squamous cell carcinoma of the lung and head and neck. However, there were no data available describing a higher sensitivity for any other marker in these tumors. In this study, the positive rates were high for CEA, SCCA, IAP, and Cyfra (31.0%, 3&l%, 52.4%, and 3&l%, respectively) in patients with 0-SCC as compared with control patients. From these results, we consider these four tumor markers to be of diagnostic value in 0-SCC. Although the sensitivity and accuracy of each indicator were relatively low, as shown in Table 4, their specificity was high. When patients who tested positive for either of the four tumor markers were considered, the sensitivity and accuracy of the tests were at their highest, and specificity was 73.3%. Our results indicate that the determination of CEA, SCCA, IAP, and Cyfra serum levels in patients with

.

FIGURE

2.

Serum

levels

of IAP and Cyfra.

ASSAY

FOR

TUMOR

Table 4. Evaluation of Diagnostic Tumor Markers in Oral SCC Sensitivity Tumor

Maker

CEA SCCA IAP Cyfra Combination

assay

(%I 31.0 38.1 52.4 38.1 81.0

MARKERS

Usefulness

Specificity (%I 90.0 93.3 83.3 90.0 13.3

of

Accuracy (%I 55.6 61.1 65.3 59.7 77.8

0-SCC leads to a more accurate diagnosis. This combination assay may prove to be useful in the early diagnosis of metastases to other organs. References 1. Vincent RG, Chu TM, Fergen TB, et al: Carcinoembryonic antigen in 228 patients with carcinoma of the lung. Cancer 36:2069, 1975 2. Richter GW: Comparison of ferritins from neoplastic and nonneoplastic human cells. Nature 207:999, 1982 3. Holmgren J, Lindholm L, Persson B: Detection by monoclonal antibody of carbohydrate antigen CA 50 in the serum of patients with carcinoma. Br Med .I 2281479, 1984 4. Einhorn N, Bast RC, Knapp RC, et al: Preoperative evaluation of serum CA 125 levels in patients with primary epithelial ovarian cancer. Obstet Gynecol 67:414, 1986 5. Kew MC, Berger EL: The value of serum concentrations of tissue polypeptide antigen in diagnosis of hepatocellular carcinoma. Cancer 58:127, 1986 6. Mino N, Iio A, Hamamoto K: Availavility of tumor-antigen 4 as a marker of squamous cell carcinoma of lung and other organs. Cancer 62:730, 1988 7. Kurokawa H, Tsuru S, Okada M, et al: Evaluation of tumor markers in patients with squamous cell carcinoma in the oral cavity. Int J Oral Maxillofac Surg 22:35, 1993 8. Silverman NA, Alexander JC, Chretien PB: CEA levels in head and neck cancer. Cancer 37:2204, 1976 9. Fischbach N, Meyer T, Barthel K: Squamous cell carcinoma antigen in the diagnosis and treatment follow-up of oral and facial squamous cell carcinoma. Cancer 65:132, 1990 10. Doweck I, Barak M, Greenberg E, et al: Cyfra 21-1: A new potential tumor marker for squamous cell carcinoma of head and neck. Arch Otolaryngol Head Neck Surg 121:177, 1995 11. Paganuzzi M, Onetto M, Marroni P, et al: CA 19-9 and CA 50 in benign and malignant pancreatic and biliary diseases. Cancer 61:2100, 1988 12. Vincenz K, Schoenthal E, Zekert F, et al: Diagnosis of head and neck carcinomas by means of immunological tumor markers. J Craniomaxillofac Surg 15:270, 1987 13. Hermanek P, Sobin LH: UICC International Union Against Cancer. TNM Classification and Grading of Malignant Tumors. Berlin, Springer-Verlag, 1987 14. Thomson DMP, Krupey J, Freedman SO, et al: The radioimmunoassa of circulating carcinoembryonic antigen of the human digestive system. Proc Nat1 Acad Sci U S A 64:16, 1969 15. Meiyer W, Eiermann W, Stieber P, et al: Experiences with SCC antigen, a new tumor marker for cervical carcinoma. Eur J Cancer Clin Oncol 25:1555, 1989 16. Mancini G, Carbonara AP, Hermans JE: Immunochemical quantitation of antigens by single radio immunodiffusion. Immunochemistry 2:235, 1965 17. Pujol JL, Grenier J, Daures JP, et al: Serum fragment of cytokeratin subunit 19 measured by CYFRA 21-1 immunoradiometric assay as a marker of lung cancer. Cancer Res 53:61, 1993