Carcinoembryonic antigen in serum in diseases of the liver and pancreas

Analyzing the high frequency of false-positive carcinoembryonic antigen elevations in postoperative pancreatic ductal adenocarcinoma

PURPOSE

The dynamics of postoperative carcinoembryonic antigen (CEA) in pancreatic ductal adenocarcinoma (PDAC) patients have not been well assessed. This study investigated the correlation between postoperative CEA elevations and tumor recurrence.

METHODS

Medical records were retrospectively analyzed for 84 patients who received curative resection for PDAC from January 2019 to December 2020. Postoperative CEA levels were monitored for a minimum of 12 months. False-positive CEA elevation was defined as a CEA level exceeding 5 ng/mL without evidence of recurrence in imaging studies.

RESULTS

Of the examined patients, 59 (70%) exhibited CEA > 5 ng/mL within the observation period. The sensitivity and specificity of elevated CEA levels for detecting recurrence were 84% and 41%, respectively. CEA elevations without tumor recurrence were observed in 27 patients, indicating a false-positive rate of 59%. More than half of these patients demonstrated peak CEA levels between 5 and 10 ng/mL, while only true-positive patients exhibited CEA levels exceeding 40.0 ng/mL.

CONCLUSION

CEA may rise in more than half of postoperative PDAC patients without recurrence. CEA alone is not a robust postoperative marker.

Improved diagnostic accuracy with three lung tumor markers compared to six-marker panel

Background

Combining multiple tumor markers increases sensitivity for lung cancer diagnosis in the cost of false positive. However, some would like to check as many as tumor markers in the fear of missing cancer. We though to propose a panel of fewer tumor markers for lung cancer diagnosis.

Methods

Patients with suspected lung cancer who simultaneously underwent all six tests [carcinoembryonic antigen (CEA), cytokeratin-19 fragment (CYFRA), squamous cell carcinoma-associated antigen (SCC), neuron-specific enolase (NSE), pro-gastrin-releasing peptide (ProGRP), and sialyl Lewis-X antigen (SLX)] were included. Tumor markers with significant impact on the lung cancer in a logistic regression model were included in our panel. Area under the curve (AUC) was compared between our panel and the panel of all six.

Results

We included 1,733 [median 72 years, 1,128 men, 605 women, 779 (45%) confirmed lung cancer]. Logistic regression analysis suggested CEA, CYFRA, and NSE were independently associated with the lung cancer diagnosis. The panel of these three tumor markers [AUC =0.656, 95% confidence interval (CI): 0.630-0.682, sensitivity 0.650, specificity 0.662] had better (P<0.001) diagnostic performance than six tumor markers (AUC =0.575, 95% CI: 0.548-0.602, sensitivity 0.829, specificity 0.321).

Conclusions

Compared to applying all six markers (at least one marker above the upper limit of normal), the panel with three markers (at least one marker above the upper limit of normal) led to a better predictive value by lowering the risk of false positives.

The Number of Positive Tumor Markers (NPTM) Achieves Higher Value in the Prognosis Prediction of Gastric Cancer

Purpose

The clinical application of combined tumor markers is still limited. We aim to explore the value of the combination of multiple tumor markers in gastric cancer (GC) prognosis.

Methods

The prognostic significance was evaluated using Kaplan-Meier log-rank survival analysis and multivariable Cox regression analysis. The estimated area under the curve (AUC) was compared to evaluate the discriminatory ability of different indicators. A nomogram was constructed based on the results of multivariable cox regression, and its performance was evaluated by Harrell's concordance index and calibration curve.

Results

NPTM (number of positive tumor markers) displayed independent prognostic significance whether in the whole cohort or in patients with different stages. Patients with the all-negative tumor markers had a worse prognosis after postoperative chemotherapy in all cohort (P = 0.023) or in age ≤60 subgroup (P = 0.012), while patients with positive tumor markers had a better prognosis after postoperative chemotherapy in stage III (P = 0.012). The AUC value of NPTM is higher than any individual tumor marker. The 1-, 3-, and 5-year AUC values of the CTNM (combination of NPTM and pTNM) increased by 5%, 4.8%, and 3.6%, respectively, compared with TNM staging system. The nomogram constructed including NPTM showed its high accuracy (C - index = 0.706) versus TNM staging system (C - index = 0.646) and CTNM (C - index = 0.681).

Conclusions

NPTM was an independent predictor of gastric cancer prognosis, showing more accurate prognostic performance than individual tumor markers. Especially its significance in guiding postoperative adjuvant chemotherapy regimens and predicting prognosis by combination with TNM staging system may have a better clinical application value.

Carbohydrate Antigen 19-9, Carcinoembryonic Antigen, and Carbohydrate Antigen 72-4 in Gastric Cancer: Is the Old Band Still Playing?

Background

Gastric cancer (GC) is characterized by aggressive behavior and a high mortality rate. The diagnosis of GC is challenging because the GC is often diagnosed in an advanced stage. The use of tumor markers is a putative way to improve the detection and treatment in patients with GC.

Summary

In this article, we review the significance of serum carbohydrate antigen (CA) 19-9, carcinoembryonic antigen (CEA), and CA 72-4 in GC. The results from different studies regarding the diagnostic and prognostic role of CA 19-9, CEA, and CA 72-4 in GC are encouraging, but inadequate sensitivity and specificity obstruct their use as standardized and unconditionally reliable markers in GC. New prospective clinical trials are mandatory for clarifying their value in GC.

Key Message

CA 19-9, CEA, and CA 72-4 should not be used for screening and early diagnosis in GC, whereas they are beneficial in the detection of late GC. CA 19-9, CEA, and CA 72-4 could be used as prognostic and monitoring tools in GC, and their combined measurement in shorter periods of time is the best method to increase sensitivity and specificity.

Practical Implications

Serum CA 19-9, CEA, and CA 72-4 are useful diagnostic and prognostic tumor markers in GC.

The Value of the CEA Test in the Diagnosis of Metastases of Adenocarcinoma of the Gastroenteric Tract

Of 174 patients previously resected for gastric, colon or rectal adenocarcinoma, 75 had distant metastases at the time of the first examination and provided information on the specific characters of different types of tumors as regards CEA sensitivity. The remaining 99 cases, all without clinical evidence of disease, showed positive CEA levels in 26 and negative CEA levels in 73. Of the 26 CEA-positive cases, 24 showed clinical signs of metastases in a period of time varying from 1 to 11 months. The remaining 2 cases, although CEA positive at subsequent tests, did not develop clinical metastases and have been disease-free for 5 years after the first test. Of 73 CEA-negative patients, 30 developed metastases: in 17 the clinical appearance of recurrences was preceded by a positive CEA test, while in 13 the CEA test remained negative even with the presence of disease. The remaining 43 cases are still CEA negative and without metastases from 44 to 51 months from the first examination. Altoghether, in 22 cases CEA was negative, but the clinical evidence of recurrences was positive, and in these cases 63.6 % were undifferentiated carcinomas; 2 patients were CEA positive but without signs of evident metastases.

Evaluation of the Behavior of Carcinoembryonic Antigen in Cirrhotic Patients

Benign liver diseases are a cause of increased serum levels of CEA. We studied the behavior of CEA in 86 patients with liver cirrhosis who underwent extensive clinical and laboratory evaluation. We found abnormal CEA levels in 38.4% of the patients (28.6% Child's grade A, 40.6% Child's B, and 42.4% Child's C) with a mean of 4.75 ng/ml. Significant differences were found between patients and controls. There was a trend towards higher levels of CEA in more severe cirrhosis according to Child's classification, although this was not significant. We found significant correlations between CEA and some liver tests, including glycocholic acid (r = 0.264., p = 0.012), a marker of severity in liver diseases. The increase of CEA in these patients is probably due to alterations in its metabolic processing caused by hepatocellular dysfunction. Moderate elevations of serum CEA can be expected in cirrhotic patients independently of malignancy.

Organ Specific Tumor Markers: What's New?

Tumor markers are molecules produced in the body in response to cancer. An ideal tumor marker should have high sensitivity and specificity, should be cheap, and should be easily detected in body fluids. Identification of novel markers is important and it is expected that with the advent of newer technologies, more reliable markers will be discovered. This review discusses the currently available tumor markers for different malignancies.

Primary amyloidosis with increased plasma carcinoembryonic antigen concentration. A case report

A patient with suspected malignant disease had increased concentration of plasma carcinoembryonic antigen (CEA). Amyloidosis was demonstrated at autopsy. The amyloid fibril composition was characterized by immunohistochemical and immunochemical techniques and proved to be of the lambda light chain (AL) type. CEA was demonstrated in the liver parenchyma by using antihuman CEA antiserum. Increased plasma CEA concentration in a patient with primary amyloidosis has, to our knowledge, not been reported before.

An immunohistochemical study of hepatic atypical adenomatous hyperplasia, hepatocellular carcinoma, and cholangiocarcinoma with alpha-fetoprotein, carcinoembryonic antigen, CA19-9, epithelial membrane antigen, and cytokeratins 18 and 19

Eight hepatic atypical adenomatous hyperplasias (AH), 30 hepatocellular carcinomas (HCC) consisting of 11 well-, 13 moderately and six poorly differentiated HCC, and 10 intrahepatic cholangiocarcinomas (CC) were investigated immunohistochemically with anti-alpha-fetoprotein (AFP), carcinoembryonic antigen (CEA), CA19-9, epithelial membrane antigen (EMA), and cytokeratins (CK) 18 and 19 antibodies. Immunostaining was regarded as positive when more than 5% of cells were stained. Alpha-fetoprotein was positive, although focally, in five (17%) of 30 HCC but negative in all AH and CC. Carcinoembryonic antigen (polyclonal antibody) did not stain the cytoplasm of all AH and HCC, but stained two (25%) of eight AH and 10 (33%) of 30 HCC in a bile canalicular staining manner. Carcinoembryonic antigen showed intracytoplasmic or luminal border staining in six (60%) of 10 CC. CA19-9 was negative in all AH and HCC, while six (60%) of 10 CC were positive for CA19-9. Epithelial membrane antigen was positive in one (13%) of eight AH, seven (23%) of 30 HCC and in all 10 cases of CC. Cytokeratin 18 was positive in all AH, HCC and CC. Cytokeratin 19 was negative in both AH and HCC, whereas it stained the cytoplasm of tumor cells in all CC diffusely and intensely. These results suggest that immunostaining of AFP, CEA, CA19-9, EMA, CK18 and CK19 are not useful in the differential diagnosis between AH and well-differentiated HCC, and that CK19 is the most suitable reagent for the differential diagnosis between HCC and CC.

Influence of smoking on serum carcinoembryonic antigen levels in subjects who underwent multiphasic health testing and services

We quantified the effect of smoking on serum carcinoembryonic antigen (CEA) levels in 1341 subjects who underwent the multiphasic health testing and services in our center. Four hundred and sixty seven of them were smokers and the rest were nonsmokers. In males subjects, serum CEA levels were significantly higher in smokers (3.11 +/- 1.8 ng/ml) than in nonsmokers (2.14 +/- 1.8 ng/ml) (mean +/- SD; p < 0.01). For females, however, the levels had no significant differences between smokers (2.11 +/- 0.91 ng/ml) and nonsmokers (1.87 +2- 1.3 ng/ml). The CEA-positive subjects were 44, of whom 32 were male custom heavy smokers, and only 2 of them had gastrointestinal cancer. We concluded that the serum CEA level was influenced by smoking especially in males and its clinical significance for detection of carcinoma was doubtful.

Elevation of carcinoembryonic antigen related to biliary malignancy in hepatolithiasis

Bile duct neoplasms occur relatively often in patients with hepatolithiasis in Taiwan. In this study, 49 patients with a positive immunoperoxidase stain for carcinoembryonic antigen (CEA) in the bile duct glands were examined in relation to the concentration of CEA in serum and bile. The bile concentrations of CEA in 15 control patients, 16 patients with hepatolithiasis only, 7 with hepatolithiasis and bile duct carcinoma, and 11 with bile duct carcinoma only were 4.09 +/- 4.12, 70.49 +/- 81.62, 137.73 +/- 66.15, and 144.3 +/- 117.31 ng/ml (mean +/- SD), respectively. A positive CEA stain from the bile duct was seen in 13.33%, 81.23%, 85.71%, and 90.91% of each group, respectively. No strongly positive staining, (2+ or 3+) was found in the control group, but it was found frequently in the hepatolithiasis and bile duct carcinoma groups. The findings of this study therefore suggest the probability of a combined neoplasm in patients with hepatolithiasis if the CEA value is unusually elevated in the serum or bile.

Usefulness of a new tumor marker, Span-1, for the diagnosis of pancreatic cancer

Levels of serum Span-1, a new tumor marker for pancreatic cancer, were assayed in 64 patients with pancreatic cancer, 90 with nonpancreatic cancer, and 254 with nonmalignancies, involving 55 healthy controls. Furthermore, Span-1 was compared with other tumor markers (CA19-9, carcinoembryonic antigen [CEA], and DU-PAN-2). Frequency of elevated Span-1 levels was 81.3% in pancreatic cancer. False-positive elevations of serum Span-1 levels were rather common in liver cirrhosis (53.8%) and chronic hepatitis (26.3%). The sensitivity, specificity, and efficiency of this assay for pancreatic cancer, was 81.3%, 75.6%, and 76.5% against all subjects without pancreatic cancer, respectively. In comparison with other markers, sensitivity of Span-1 tended to be highest with similar specificity to those of CA19-9 and CEA. The Span-1 assay has a high sensitivity and specificity for pancreatic cancer. It is almost equivalent to CA19-9 assay. However, this assay is not specific for chronic liver diseases.

Hepatocellular carcinoma producing carcinoembryonic antigen

A 78-year-old female with hepatocellular carcinoma and high serum levels of carcinoembryonic antigen is reported. Using a immunohistochemical technique, CEA was demonstrated within the cytoplasm of the tumor cells.

Evaluation of MSA as a serum marker in breast cancer: a comparison with CEA

In a blind study, 518 serum samples were assayed for serum levels of mammary serum antigen (MSA) by an enzyme immunoassay (EIA) using the 3E1.2 monoclonal antibody. Using 300 IU as the arbitrary cut off to distinguish normal from abnormal individuals, 75% of patients with primary Stage I carcinoma of the breast (n = 12), 89% of those with Stage II (n = 9) and 93% of those with Stage IV (n = 57) had elevated levels of MSA. A relationship was observed between the level of MSA and stage of disease, and therefore with the extent of tumour burden. Levels of MSA were also determined in a series of 19 patients undergoing chemotherapy for breast cancer. Over a 2-24 month period, the change of MSA levels corresponded with the clinical course of the disease in 17 (89%) cases. MSA levels were also raised in some patients with ovarian, colon, lung and kidney cancer, but the average level was lower than in patients with breast cancer. A comparison of CEA and MSA levels in these patients revealed that MSA was a substantially better marker for breast cancer than CEA. The results of this study demonstrate that MSA levels are elevated in patients with breast cancer and may provide a useful means of following the clinical course of patients with this disease.

Tumour markers in diagnosis and management

The 20-year period since the discovery of AFP by Abelev has seen the introduction of a wide range of new tumour markers and it is now clear that PLC is biologically heterogeneous. Hepatoblastomas, fibrolamellar carcinomas, hepatocellular carcinomas and cholangiocarcinomas may secrete a variety of distinctive markers which are predominantly glycoproteins, and may resemble those found in placenta or fetal liver. Diagnostically, AFP remains the best marker for HCC, both in sensitivity and specificity; it is known to consist of isoforms. In patients with elevated serum AFP and filling defects on liver scan, Con A reactive AFP may differentiate PLC from hepatic metastases, whilst fucosylated AFP may distinguish PLC from benign disorders when AFP is non-diagnostically elevated. With this recognition of tumour heterogeneity the value of a multiple-marker approach has become apparent. The measurement of vitamin B12 binding protein and neurotensin should lead to the detection of most patients with the fibrolamellar variant of HCC and many of these should be resectable. In patients with normal serum AFP levels, HCC-associated GGTP is of major value whilst in low-incidence areas for HCC, patients should also be screened for H-ALP; using a multiple marker approach in high-risk groups, 90% of clinically diagnosed hepatocellular carcinomas are serologically positive. The Chinese and Alaskan studies, in which small, potentially resectable tumours were detected, suggest that it is now possible to achieve 5-year survival figures of up to 60% in HCC patients detected by screening. The value of such a strategy in low-incidence countries is currently under study. In patient monitoring, as in diagnosis, AFP remains the outstanding marker. In AFP-negative patients, other markers including vitamin B12-binding protein, neurotensin, HCC-specific isoenzymes, des-gamma-carboxy-prothrombin and alpha-fucosidase, are of undoubted diagnostic value, but their value as indicants of disease progression remains to be established. In monitoring the response of hepatic metastases, CEA remains the least unsatisfactory marker but should always be used in conjunction with serial ultrasound scans. Tumour markers now play an important role in the diagnosis and monitoring of PLC but a role is also emerging in tumour imaging and drug targeting. The next 20 years should see the introduction of tumour markers of high sensitivity and specificity which make a fundamental contribution not only to detection and monitoring, but also to the effective treatment of liver cancer.

Serum beta-N-acetyl hexosaminidase (beta-NAH) as a discriminant between malignant and benign extrahepatic biliary obstruction: comparison with carcinoembryonic antigen (CEA)

Fifty-one patients (16 with malignant extrahepatic biliary obstruction, ten with benign extrahepatic biliary obstruction, eight with alcoholic liver disease, five with viral hepatitis and 12 with liver metastases) and 19 adult healthy controls were studied with determinations of beta-N-acetyl hexosaminidase (a lysosomal enzyme which is cleared from the circulation by the Kupffer cells), carcinoembryonic antigen (CEA), serum bilirubin, alkaline-phosphatase and aspartate aminotransferase (AST). Both CEA and beta-NAH were elevated in each disease group. Elevated beta-NAH levels distinguished between benign and malignant extrahepatic biliary obstruction better than CEA levels. Beta-NAH levels for the malignant and the benign groups were 47.6 +/- 14.7 U/l and 23.0 +/- 4.7 U/l (mean +/- S.D.) respectively. The groups differed significantly (P less than 0.001). Plasma CEA levels for both groups were 18.7 +/- 38.9 and 7.2 +/- 3.3 ng/ml (mean +/- S.D.) respectively. Beta-NAH levels for the 19 normal controls were 15.8 +/- 3.5 U/l (mean +/- S.D.). Beta-NAH also was significantly elevated in patients with hepatic metastases (36.9 +/- 20.1 U/l). In 25 cancer patients with metastases other than in the liver beta-NAH levels (18.3 +/- 5.2) were not significantly elevated over the control group. It has potential value as a marker for non-CEA-producing liver metastases.

CEA tissue staining in colorectal cancer patients. A way to improve the usefulness of serial serum CEA evaluation

The evaluation of serial plasma carcinoembryonic antigen (CEA) levels is one of the most important parameter used to establish the prognosis of surgically cured colorectal cancer patients. Carcinoembryonic antigen is particularly useful in the identification of recurrences and metastasis. However, to improve the usefulness of this assay, it would be helpful to accurately determine, if possible, those patients whose cancers produce CEA. The evaluation of the presence of CEA in these cancer specimens by means of immunoperoxidase staining technique does seem to improve the sensitivity of the CEA test. Fifty-seven patients with colorectal cancer who underwent surgical treatment were studied. Tissue CEA evaluation was correlated with the plasma CEA levels, the pathologic stage and grade, and histologic type of the cancers. Results demonstrate that 66.6% of Dukes' B cancers, 78.9% of Dukes' C, and 77.7% of Dukes' D cancers stained positively for CEA by immunoperoxidase. Thirty of 57 patients with preoperative pathologic plasma CEA levels had positive tissue CEA, whereas 8/57 patients did not. Of patients with a well-differentiated cancer (G1), 81.4% had positive tissue CEA versus the 64% of G2 and 60% of G3 cancers. The authors conclude that the use of the immunoperoxidase stain to measure CEA in tissue, so that the CEA serum assay may be used in those patients known to produce CEA, results in a major increase in the sensitivity of the test.

The prognostic significance of preoperative carcinoembryonic antigen levels in colorectal cancer. Results from NSABP (National Surgical Adjuvant Breast and Bowel Project) clinical trials

This analysis explores the prognostic significance of preoperative carcinoembryonic antigen (CEA) levels in patients with colorectal cancer. The data were derived from 945 patients entered into two randomized prospective clinical trials of the National Surgical Adjuvant Breast and Bowel Project. A strong correlation was evident between preoperative CEA level and Dukes class. The mean CEA progressively increased with each Dukes category and the mean value for each of the four classes was significantly different. This relationship was prevalent whether the data were analyzed for all colorectal lesions regardless of location or specifically for right-sided colon tumors. The prognostic function of preoperative CEA level was independent of the number of positive histologic nodes. Preoperative CEA level correlated with the degree of lumen encirclement by tumor. Tumors that did not encircle more than one half the lumen were associated with significantly lower preoperative CEA levels than those that did. The presence or absence of lumen obstruction was unrelated to the preoperative CEA level. The relative risk of developing a treatment failure was associated with preoperative CEA, in both Dukes B and C patients, demonstrating that the prognostic significance of preoperative CEA was independent of Dukes class.

Hepatocellular carcinoma presenting extrahepatic biliary obstruction

An autopsy case of hepatocellular carcinoma presenting extrahepatic obstructive jaundice was reported. The patient was a 68-year-old man with 5-year history of chronic liver disease and markedly jaundiced, and died of gastrointestinal hemorrhage. Autopsy revealed hepatocellular carcinoma associated with liver cirrhosis and metastatic polypoid growth in the common bile duct and cystic duct which completely obstructed the duct lumen. Obstructive jaundice secondary to complete obstruction of the common bile duct was a rare complication of hepatocellular carcinoma. The other 85 cases of hepatocellular carcinoma presenting extrahepatic biliary obstruction are reviewed.

Carcinoembryonic antigen: physician attitudes, patterns of use, and impact upon patient care

Recent data would limit indications for serum CEA measurement primarily to follow-up of resected colonic malignancy, yet physician attitude and usage patterns may lag far behind current findings. This discrepancy was investigated at our institution, where more than 1100 CEAs costing $71,000 are ordered each year. Of 45 physicians (all MDs ordering a CEA test during a preselected month), over 50% believed the test to be worthwhile in initial detection of colonic cancer, and 69% thought an elevated CEA to be an adequate reason to begin an aggressive workup to rule out cancer of the colon in a nonsmoking, previously healthy patient. Impressions of cost were less than or equal to $30 (50% of true cost) in nearly half of MDs and less than 20% of true cost in a tenth of MDs. Analysis of the medical record revealed that indications of questionable validity (initial detection of cancer together with follow-up of noncolonic malignancy) accounted for the majority of requested CEAs and included the attempted detection or monitoring of 12 different tumor types in addition to its use as a "general cancer screen." Patient benefit was realized in none in a random sample of 106 cases (beta = 0.11, power = 0.89 for an assumed benefit of 2%), while management was altered in only one patient as a direct result of the CEA value. It is important that we continue to inform and educate our colleagues about relatively expensive tests that have only limited and specific application.

Circulating CEA levels in patients with fulminant hepatitis

CEA levels were determined in blood specimens from 14 persons suffering from fulminant hepatitis. Values ranged from 1.9 to 21 ng/ml with a mean of 7.0 ng/ml. All except one patient had a CEA level greater than 2.5 ng/ml, but only three exceeded 10 ng/ml. There was no correlation between single CEA levels and prognosis, routine liver chemistries, or histopathologic characterization of hepatic tissue obtained at postmortem examination. These data show that patients with fulminant hepatitis commonly had increased circulating levels of CEA, but these elevations were far less than may be found in patients with hepatic metastases. In severe hepatic dysfunction reduced hepatic clearance of CEA may be responsible for the increased levels.

Plasma carcinoembryonic antigen in the diagnosis and management of patients with hepatocellular carcinoma

The value of serial carcinoembryonic antigen (CEA) measurements as a marker of disease progression or in monitoring treatment was investigated in patients with hepatocellular carcinoma. Of 40 patients, including 16 with normal serum alpha-fetoprotein (AFP) concentrations, 29 (72.5%) had abnormal plasma CEA at presentation. Although this was more common in patients with pre-existing cirrhosis, the mean and range of plasma CEA were similar in patients with and without pre-existing hepatic disease. There was no correlation between plasma CEA and any biochemical parameter of hepatic function, although plasma CEA concentrations were significantly lower in patients with well-differentiated tumors. CEA concentrations increased in 71% of patients who had no response to cytotoxic drugs, but CEA also increased in 62.5% of those patients who did respond. Plasma CEA concentrations were elevated in 62.5% of patients with normal and 79% of patients with raised serum AFP on admission to the hospital. There was no correlation between individual AFP and CEA concentrations. Although elevated plasma CEA levels may be of diagnostic value in patients with hepatocellular carcinoma in the absence of pre-existing hepatic disease, and in those with normal serum AFP, our findings indicate that it does not behave as a true tumor marker.

The relationship of the possible hepatic toxicity of chemotherapeutic drugs and carcinoembryonic antigen elevation

The value of the serum level of carcinoembryonic antigen (CEA) as an indicator of recurrent colorectal carcinoma has been accepted, and the use of serial CEA levels to monitor postoperative chemotherapy has been suggested, However, elevated CEA levels may be associated with nonneoplastic conditions, of which the most difficult to evaluate is hepatic disease. The effect of chemotherapy on hepatic function and therefore on CEA level is not clear. We discuss a patient who, after a potentially curative resection for adenocarcinoma, demonstrated a rise in CEA level in the absence of recurrent carcinoma. This rise correlated with administration of intravenous 5-fluorouracil. Liver biopsy demonstrated severe fatty infiltration with no evidence of cirrhosis. The possible liver toxicity of chemotherapeutic drugs must be considered as a factor responsible for the rise in CEA. Simultaneous assessment of hepatic function is essential to the accurate interpretation of CEA levels, especially in patients receiving chemotherapy.

The use and limitations of serial plasma carcinoembryonic antigen (CEA) levels as a monitor of changing metastatic liver tumor volume in patients receiving chemotherapy

Nine patients with liver-predominant or liver-only metastatic adenocarcinoma, who received hepatic intra-arterial and peripheral intravenous chemotherapy, were evaluated to determine if changes in plasma CEA accurately indicated changes in liver tumor volume as estimated from serial liver scan defect sizes, and if the two were quantitatively related. When selected periods of tumor remission or progression were studied during chemotherapy, the index of change of serial CEA values correlated positively (r = 0.764, P = 0.01) with the index of change of liver tumor volume in all 9 cases. Changes in absolute CEA values, however, aid not correlate significantly with changes in absolute liver tumor volumes when assessed either during these selected periods or throughout individual patient courses. Biologic and methodologic factors that influence the quantitative relation between liver tumor volume and plasma CEA are discussed.

Value of serial CEA determinations in a surgical adjuvant trial of colorectal and gastric carcinoma

At the Cross Cancer Institute in Edmonton, a concurrently controlled, randomized, prospective surgical adjuvant trial involving Dukes' B2 and C colorectal carcinoma and gastric carcinoma (T1-4, No-2, Mo) has been activated for the last two years. To date, a total of 150 patients have been entered into the three arms of the trial (namely control, immunotherapy, and chemoimmunotherapy). Of these, 127 cases are colorectal carcinoma and 28 are gastric cancer. As part of the protocol, serial CEA determinations are obtained in all patients on a regular three-monthly basis. So far, 28 patients have confirmed recurrence (colorectal 20, gastric 8) demonstrated clinically or radiologically. Out of these 28 patients, 25 (89.2%) had CEA greater than 2.5 ng/ml months before actual demonstration of recurrence. However, if one uses a cutoff of 5.0 ng/ml as significant, that is, one standard deviation from the usual accepted value, then 21 out of 28 (75%) had persistent, elevated CEA's months before the recurrence manifested itself clinically or radiologically. The CEA profiles of three representative cases from each arm of the adjuvant trial are included. It is suggested that serial CEA determinations will be an invaluable adjunctive test in following patients in an adjuvant trial setting.

Effect of secretin and cholecystokinin-pancreozymin on plasma CEA concentration in patients with pancreatic carcinoma and pancreatitis

Plasma CEA concentration before and after administration of secretin and cholecystokinin-pancreozymin has been determined in 89 patients with neoplastic or inflammatory pancreatric disease and other neoplastic and nonneoplastic disorders. The purpose of the study was to expore the specificity and sensitivity of such a provocation tests. Some rise of the plasma CEA concentration after hormonal stimulation could be observed in several patients in the various groups. However, none of the 37 patients with nonpancreatic disease who had a basal CEA concentration of 6 micrograms/l or less had a maximal CEA concentration above 13 micrograms/l after the stimulation whereas 7 out of 31 patients with pancreatic disease (3 with pancreatic carcinoma and 4 with chronic pancreatitis) showed such an elevation of plasma CEA concentration. Thus, the provocation test showed a satisfactory specificity for pancreatic disease but a low sensitivity. It is suggested that the possibility of an effect of physiologically released gastrointestinal hormones should be considered when "unexplained" high CEA values are found in plasma samples from nonfasting patients.

Predictive value of serial carcinoembryonic antigen levels in long-term follow-up of ovarian cancer

The predictive value of serial levels of carcinoembryonic antigen (CEA) in tumor monitoring was examined in 213 patients with ovarian cancer; each patient had been followed-up at monthly intervals for at least 12 months. CEA was not detectable throughout the period of observation in 35% of the patients. In general. patterns showing a disappearance of CEA or persistently low levels were associated with a good prognosis, whereas those showing a reappearance or highly elevated and rising levels were associated with a poor prognosis. A transient reappearance of CEA was observed in 10 patients; this did not appear to be associated with tumor recurrence or progression. "False positive" results were obtained in 6 patients in whom no tumor has been clinically detectable to date. "False negative" results were obtained in 4 patients with obvious tumor progression. In terms of a good or poor prognosis, the use of CEA levels was highly accurate in patients with minimal or no residual disease (97% and 89%, respectively); the rate fell to 62% in patients with extensive disease. As the clinical significance and limitations become better known, serial CEA levels should contribute substantially to the monitoring of patients with ovarian cancer.

Elevated serum levels of a biliary glycoprotein (BGP I) in patients with liver or biliary tract disease

Human hepatic bile contains a glycoprotein (biliary glycoprotein I, BGP I) which cross-reacts with the carcinoembryonic antigen (CEA). A radioimmunoassay for BGP I was developed. The interference of CEA or 'non-specific cross-reacting antigen' (NCA) in the assay was small. The serum levels of BGP I were determined in healthy subjects, in patients with hepato-biliary diseases and in patients with various infectious or inflammatory disorders. Healthy individuals, including pregnant women, had a serum BGP I concentration of about 0.5-1 mg/l. Diseases of the liver or biliary tract (e.g. hepatitis A or B, cytomegalovirus hepatitis, obstructive jaundice or primary biliary cirrhosis) were associated with elevated serum levels of BGP I, as opposed to infectious diseases not affecting the liver mostly showing values within the normal range. Raised levels of serum BGP I activity may reflect biliary obstruction as a result of interference with normal BGP I secretion to the bile.

Pathology of primary liver cancer

After a brief survey of the factors that play a role in the etiopathogenesis of human hepatocellular carcinomas, a detailed description is given of the macroscopic and microscopic features of human liver cancers as well as their association with cirrhosis. The ultrastructural features of liver cancers of various degrees of differentiation are described. The mode of spread, metastasis formation of primary liver cancers, and most frequent causes of death of liver cancer patients are reviewed.

A reappraisal of the value of carcinoembryonic antigen in the management of patients with various neoplasms

Eight hundred and eight patients with histologically proved malignant disease had carcinoembryonic antigen (CEA) estimations performed at the time of tissue diagnosis. An elevated level was found in 384 of 518 patients with gastrointestinal neoplasms (74 per cent) and in 162 of 290 patients with other neoplasms (56 per cent). No correlation was found between CEA elevations and tumour differentiation. There was a good correlation between tumour staging and CEA levels for patients with colorectal cancer; the more advanced the tumour, the higher the CEA. Several illustrative cases are presented and the role of CEA assay in the diagnosis and management of neoplasia is discussed. CEA assay is a poor screening test for neoplastic disease, but serial CEA monitoring is valuable in the detection of residual or recurrent cancer.

Carcinoembryonic antigen (CEA) in patients with alcoholic liver diseases

The mean plasma carcinoembryonic antigen (CEA) level was found to be significantly increased in 66 patients with alcoholic liver disease (4.4 micrograms/1) compared with the mean CEA level in 164 healthy blood donors (1.6 micrograms/1, p less than 0.001). Eighteen of the patients (27%) had values above our upper normal values of 5 micrograms/1, compared with 3 of 41 (7%) in a group of patients with non-alcoholic liver diseases. On the other hand, the mean CEA values in the two groups did not differ significantly. The CEA level remained unchanged in patients who continued drinking. However, in 20 patients who stopped drinking, the mean CEA levels dropped significantly from 6.6 micrograms/1 to 3.7 micrograms/1. Concomitant fluctuations were seen in CEA, gamma-glutamyltranspeptidase (gamma-GT), asparate aminotransferase (ASAT), and alkaline phosphatase, although a significant correlation was found only between CEA level and the level of gamma-GT (r = 0.32, p less than 0.02). The correlation between CEA and gamma-GT was best in 41 patients without complicating disorders (r = 0.55, p less than 0.001). CEA was significantly increased in patients with complicating disorders. Much of the increase of CEA in many patients with alcoholic liver disease might be secondary to complicating disorders in the respiratory and/or gastrointestinal tracts. This could possible explain why patients with alcoholic liver disease more often than other patients with benign diseases have elevated CEA.

Carcinoembryonic antigen concentrations in undiagnosed patients

Serum carcinoembryonic antigen (C.E.A.) levels were measured in 381 undiagnosed patients who presented with clinical problems commonly associated with gastrointestinal malignancy. The results were compared with the final diagnosis after follow-up for up to 5 years to see whether C.E.A.-testing added any useful information. Of 307 patients presenting with upper gastrointestinal symptoms, lower gastrointestinal symptoms, or irom deficiency anaemia, C.E.A. levels greater than 20 ng/ml indicated malignancy in 5 but in 3 of these malignancy was also diagnosed after routine investigation. Of 74 patients presenting with obstructive jaundice, hepatomegaly, or abnormal liver function, malignancy was diagnosed in 38. In 9 of these patients the diagnosis of malignancy could otherwise have been reached only by laparotomy. The serum-C.E.A. thus reached only by laparotomy. The serum-C.E.A. thus seems to be of value in the assessment of liver disease but not in patients with gastric or colonic symptoms or iron-deficiency anaemia.

Use of CEA as an indicator of early recurrence and as a guide to a selected second-look procedure in patients with colorectal cancer

The usefulness of the CEA as an indicator of recurrence and a guide to selected second-look surgery was evaluated from a retrospective analysis of 358 patients with colorectal cancer and from a prospective experience with 16 patients all of whom had been admitted for second-look surgery because of postoperative elevations of CEA only. Our previous experience had shown that after curative resection the CEA usually returned to normal levels (less than 5 ng/ml) within one month, but became elevated at time of clinically obvious recurrence being very high in patients with liver metastases, but only moderately elevated or normal in patients with local recurrence. All 16 patients had previously had curative resection of colorectal cancer; 13 in the rectum or rectosigmoid and three in the right colon. There were 13 Dukes' C and three Dukes' B cancers. All had been followed clinically and by CEA testing at three monthly intervals and were considered free of disease (NED) at time of CEA elevation. The median disease free interval was 13 months (range 4-57 months) and the median CEA prompting admission for second-look operation was 21 ng/ml (range 10-56 ng/ml). The sites of recurrence were liver in six, lung in two and localized disease in six. Two patients had negative exploration for recurrence and were found to have cholelithiasis only (one of these later died of metastases). Resection for cure was done in seven and palliative resection or biopsy only was done in nine patients. At this time, four patients are NED (12-37 months), five are living with disease (10-16 months) and seven have died of disease (2-12 months). The CEA test provides a method of early detection of recurrence and may permit surgical retrieval in selected patients and earlier initiation of palliation in other patients. The longterm effects in patient salvage remain to be defined.

Carcinoembryonic antigen (CEA) levels in benign gastrointestinal disease states

Elevated circulating CEA levels occur in patients with benign gastrointestinal and hepatic disorders. These are usually less than 10 ng/ml. Of clinical importance is the influence of liver disease on the interpretation of CEA. At least 50% of patients with severe benign hepatic disease have elevated CEA levels, most often active alcoholic cirrhosis, and also chronic active and viral hepatitis, and cryptogenic and biliary cirrhosis. Patients with benign extrahepatic biliary obstruction may have increased plasma CEA, the highest in patients with co-existent cholangitis and especially liver abscess. The liver appears to be essential for the metabolism and/or excretion of CEA. Hence, liver work-up is needed to assess any patient with an elevated CEA. A damaged liver may further augment elevated CEA levels due to cancer. The increased circulating CEA observed in some patients with active ulcerative colitis tends to correlate with severity and extent of disease and usually returns to normal with remission. CEA levels also may be mildly elevated in patients with pancreatitis and in adults with colonic polyps. Smoking may contribute to the increased CEA levels seen in patients with alcoholic liver disease and pancreatitis. Therefore, in interpreting mildy elevated circulating CEA levels in patients with GI tract diseases, one must consider benign as well as malignant etiologies.

Carcinoembryonic antigen in normal and diseased liver tissue

Many reports have demonstrated an elevation of circulating carcinoembryonic antigen (CEA) in the majority of patients with alcoholic liver disease and, less frequently, in patients with nonalcoholic liver disease. Several explanations for this finding have been proposed, eg, increased production or release of CEA by the damaged liver, decreased hepatic metabolism, or diminished excretion of CEA of extrahepatic origin. In an attempt to clarify the mechanism of CEA elevation in liver disease, we have compared the CEA plasma level as measured by radioimmunoassay with CEA demonstrable in liver tissue by the indirect fluorescent antibody technique in 7 patients without significant changes in the liver biopsy specimen, 23 patients with alcoholic liver disease, and 16 patients with miscellaneous liver diseases such as acute or chronic nonalcoholic hepatitis or extrahepatic biliary obstruction. The mean CEA plasma level in patients with alcoholic liver disease was significantly higher than in patients with nonalcoholic liver disease (8.8 +/- 9.5 vs 2.7 +/- 2.5 ng/ml; P less than 0.02). In normal liver tissue, CEA was observed in the apical cytoplasm and along the luminal surface of bile duct epithelial cells, suggesting that under normal conditions CEA accumulates in and is excreted by bile ducts. In patients with alcoholic hepatitis and/or cirrhosis there was marked bile ductular proliferation and prominent cytoplasmic CEA-specific staining and both were associated with elevated CEA plasma levels in more than 80% of cases. In the group of miscellaneous liver diseases, bile ductule counts and CEA-specific staining did not correlate with CEA plasma levels. These observations suggest that proliferating bile ductules contribute to elevated plasma CEA in alcoholic patients.

Extraction of CEA from tumour tissue, foetal colon and patients' sera, and the effect of perchloric acid

The use of perchloric acid and water for the extraction of CEA from tumour and foetal tissues has been investigated. In the case of tumour, lower recoveries of CEA were obtained from perchloric acid extracts than from aqueous extracts of the same tissue. CEA has also been extracted with 3M KCl solution from insoluble perchloric acid residues of tumour homogenates and cancer patients' serum. Whilst a large proportion of CEA activity recovered from tumour was associated with the perchloric acid residue, the corresponding amounts from serum were very small. CEA elution volumes for each extract, obtained by assay of Sephadex G-200 column fractions, showed significant heterogeneity in molecular size. The purified CEA pools also showed quantitative variations in the binding profiles on Con A-Sepharose. It has been shown that perchloric acid modifies the carbohydrate in CEA, thus altering its Con A-binding properties. Preliminary experiments with foetal colon have demonstrated that, unlike colorectal CEA, a significant proportion of foetal CEA was not bound to ConA. Comparative immunodiffusion showed immunological identity of CEA from the various extracts, although the purified aqueous extract produced an additional precipitin reaction, indicating a second antigen which is relatively unstable or less soluble in perchloric acid.

The clinical utility of plasma and urinary carcinoembryonic antigen in patients with genitourinary disease

Carcinoembryonic antigen was measured in the urine and plasma of 108 patients with several types and various stages of genitourinary cancer. The value of the carcinoembryonic antigen assay as an early indicator of neoplastic disease was evaluated and a correlation was made between the extent of disease and the concentration of urinary and plasma carcinoembryonic antigen. Patients were classified according to stage of tumor involvement as follows: no evidence of disease, non-malignant disease, non-invasive disease, no known metastasis, regional metastasis and disseminated metastasis. The urinary carcinoembryonic antigen levels more closely paralleled the extent of disease than did the plasma carcinoembryonic antigen levels in patients with bladder cancer. Neither urinary nor plasma carcinoembryonic antigen levels were useful in assessing the extent of disease in patients with prostatic or testicular cancer. Studies related to microbiological interference in the carcinoembryonic antigen assay indicated that bacterial counts up to 10(5) organisms per ml. did not interfere. Cytological studies indicated that the presence of white blood cells, atypical cells and malignant cells could result in elevated urinary carcinoembryonic antigen levels.

The CEA concentration in duodenal fluid in patients with pancreatic disease

The CEA concentration in duodenal fluid after secretin-CCK stimulation has been investigated in 16 patients with pancreatic disease (6 with pancreatic carcinoma and 10 with chronic pancreatitis), 9 with non-pancreatic disease, and 10 control subjects. The purpose was to study whether the determination of CEA in duodenal fluid during the secretin-CCK test can give any additional information for the diagnosis of pancreatic disease and for differentiation between pancreatitis and carcinoma. We found that high values of CEA in duodenal fluid do not necessarily indicate pancreatic carcinoma. Moreover, the level may be elevated in non-pancreatic disease.