Biological markers for breast carcinoma.

During the last decade, there has been an intense search for biological markers for breast carcinoma. Many different types of materials have been found that may be elevated in the body fluids of patients with this disease. However, no markers specific for breast cancer have been discovered and those currently available lack the sensitivity and specificity for early detection of the disease or for determining when the tumor burden is low. Problems may also occur in the interpretation of marker data due to apparent biological variations in synthesis or secretion. Plasma carcinoembryonic antigen (CEA), elevated in 60%-70% of patients with metastases, has had the most extensive evaluation. For the latter patients with increased plasma CEA, the levels in general are proportional to tumor burden. Changes in level with therapy correlate with measurable clinical parameters of response or progression in the majority of these patients. Specific patterns of serial CEA measurements after mastectomy may be helpful for predicting those patients most likely to develop recurrent disease. More recent attention has focused on trials of combinations of markers and on tissue measurement. The search for a specific marker for breast cancer using monoclonal antibody techniques is a promising area of considerable research interest.

Evaluation of the clinical management of cancer patients. A pilot study.

The Johns Hopkins Oncology Center, Baltimore, has carried out a pilot study of a method of data collection and analysis to be used for planning and evaluation of continuing education programs conducted in the Maryland region. This study includes definition of techniques used to prepare a test instrument that will accurately reflect actual medical practice within a hospital. The system has been field tested in three diseases, cancer of the lung, breast, and uterine cervix, in two Maryland hospitals. The system appears to be an efficient and inexpensive method for definition of decision making in specific clinical settings and for measurement of change if serial reassessment is instituted.

Urinary polyamines for evaluating the course of disease for patients with small cell carcinoma of the lung.

Clinical correlates with urinary excretion of polyamines were evaluated for 29 newly diagnosed and 35 previously treated patients with small cell carcinoma of the lung (SCC). The frequencies of pretreatment abnormalities were 12 (41%) for putrescine, 18 (62%) for spermidine, and 20 (69%) for spermine. In assessing disease parameters, the combined use of the abnormalities of spermidine and spermine as a discriminant was more effective than that of all three polyamines; it correlated significantly with extent of limited and extensive disease (P less than 0.001), and also resulted in significant separation of survival curves, the median survival of 11 months for both elevated compared to 19 months for neither or only one elevated (P = 0.062). No significant difference was seen in the abnormalities between no metastasis and one metastasis, whereas the frequencies of the abnormalities was highly increased in two or more metastases. The distribution of polyamines determined at regular treatment intervals showed distinctively more elevated patterns in progressive disease than in stable disease or partial and complete responses (P less than 0.01). In order to evaluate therapeutic effects on the relationship between polyamine excretion and tumor regression, correlations between urinary putrescine and spermidine were determined. The values of the ratio of spermidine to putrescine were significantly smaller in responders than in nonresponders (P less than 0.01); and these may be related to smaller tumor mass and higher tumor proliferative activity in responders, and larger tumor mass and lower tumor proliferative activity in nonresponders.

Serum protein-bound carbohydrates and small cell carcinoma of the lung. Correlations with extent of disease, tumor burden, survival, and clinical response categories.

The levels for serum protein bound neutral carbohydrates (fucose, mannose, and galactose) were determined at specific intervals for 40 patients with small cell carcinoma of the lung and compared to the corresponding carcinoembryonic antigen (CEA) levels. In pretreatment samples, the frequency of elevation was 92.5% for fucose and 77.5% each for mannose and for galactose. CEA determined in these same samples was elevated (greater than 5 ng/ml) in 45.0%. One or more of the three carbohydrate levels were elevated in pretreatment serum of 95.0% of the patients. The individual frequency of elevation for each carbohydrate was significantly related to initial stage of disease (P less than 0.01). Median survival was significantly longer for patients based on a discriminant of less than 3 carbohydrates elevated in pretreatment samples (25 months) to all 3 elevated (11 months) with P = 0.0302. A single value, termed the biomarker index, was calculated to represent the summation of the individual carbohydrate levels per individual serum sample. The biomarker index was found to be directly correlated with extent of primary disease, number of metastic sites, tumor burden, and clinical response categories assessed at serial time points. For patients with both low Biomarker Index values and normal CEA levels in pretreatment samples, an initial rise in both determinations occurred frequently corresponding to partial or complete tumor response. The occurrence of such discordant results must be considered as a likely possibility for those patients with low or normal pretreatment biological marker levels and subsequent response to primary chemotherapy.

New applications of urinary nucleoside markers.

In order to extend the usefulness of the quantitation of urinary nucleoside markers, studies were undertaken to explore the adaptability of such determinations for early detection in cancer-prone populations such as asbestos workers. Another study was aimed at exploring the usefulness of therapy in individual patients. During these studies, two heretofore unknown phenomena serendipitously emerged which expand the versatility of the marker determinations: (a) radiation damage in animals and humans causes an excretion of urinary BAIB which from preliminary studies appears to be proportional to the irradiation burden, and (b) lead poisoning in the human also produces BAIB excretion. Some of the practical uses of these determinations are self-evident. Among 13 asbestos workers without clinical symptoms, eight were found to have significant elevations of the marker levels. Nine asbestos workers with diagnosed mesothelioma all excreted two or more markers at high levels. Some of the psi levels were the highest seen. Currently the diagnosis of mesothelioma is difficult and painful, requiring a rib resection; however, an asbestos worker with such elevations--provided small cell carcinoma of the lung is ruled out--can be seriously suspected of having mesothelioma. In a study of the usefulness of the markers in following therapy of trophoblastic disease, these markers were determined in women with incipient invasive hydatidiform mole. After curettage, the nucleoside markers indicated absence of residual disease but the usual marker, HCG, was still markedly elevated. The women were followed up for 2 years and were found to remain symptom-free. Therefore the source of the nucleoside markers is cleared more rapidly than that of HCG.

Tumor markers derived from nucleic acid components.

It was known for some time that cancer patients excrete in their urine elevated levels of modified nucleosides. From earlier work, we were able to show that most of these modified nucleosides originate from transfer RNA (tRNA). The modifications are achieved at the macromolecular level by enzymes after primary synthesis. Such modifications are highly specific and, therefore, when the modified nucleosides accumulate from the tRNA breakdown, they cannot be reinserted randomly by the polymerases and must be excreted. We found that the modifying enzymes are aberrantly hyperactive in every malignant tissue. We also found that there is abnormally high turnover of tRNA in malignant tissues, which is probably the source of the elevated levels of excretion products. Since these products originate from a cardinal component of the molecular biology of every cell, the determination of markers in the urine may be a universal indicator of malignancy. We are focusing on the use of these markers in syndromes whose diagnoses are otherwise difficult. Since the marker levels return to normal very soon after chemotherapy, such determinations can be used to monitor the effectiveness of therapy. Therefore, the clinical oncologists can adapt their protocols to the specific need of a patient.

Modified ribonucleosides as biological markers for patients with small cell carcinoma of the lung.

A variety of individual modified ribonucleosides may be elevated in the urine of cancer patients. They can be readily measured quantitatively in a single reversed-phase high-performance liquid chromatographic run. A total of 41 patients with small cell carcinoma of the lung were studied. For 5-ribonucleosides determined in the pretreatment urine of 28 patients, the respective frequency of elevation was directly related to stage of disease. One or more nucleosides were evaluated in the pretreatment urine of 27 out of 28 patients (96%). Included were 11 patients with limited disease and 10 (91%) had 2 or less than 2 nucleosides elevated, whereas 16 out of 17 (94%) with extensive disease had 3 or more elevated. Based on this same discriminant, median survival was significantly extended for patients with 2 or less nucleosides elevated (24 months) in contrast to 3 or more (10 months). Using a single number to represent the summation of equally weighted individual nucleoside values as a composite score, a direct relationship was found between increasing extent of disease or tumor burden. This was in contrast to more variable results for carcinoembryonic antigen analyzed in plasma samples obtained at the same time. When determined serially the composite score paralleled in general the clinical response categories for individual patients.

Biological markers and small cell carcinoma of the lung: a clinical evaluation of urinary ribonucleosides.

Five minor base ribonucleosides, primarily degradation products of transfer ribonucleic acid (tRNA), were evaluated as potential biological markers for patients with small cell carcinoma of the lung. The urinary concentration for pseudouridine, 1-methyladenosine, 1-methylinosine, N2-methylguanosine, and N2,N2-dimethylguanosine was determined by means of reversed-phase high performance liquid chromatography and quantitatively expressed as a function of creatinine excretion. Comparisons were made with carcinoembryonic antigen (CEA) plasma levels. The total frequency of elevated values for the five nucleosides in pretreatment urine samples was directly related to stage of disease with 24/60 (40%) determinations increased in 12 patients with limited disease and 69/85 (81%) in 17 patients with extensive disease. For these same patients, CEA levels were elevated respectively in 2/11 (18%) of the former and 9/17 (53%) of the latter group. The frequency and degree of elevation of the nucleoside/creatinine ratios in pretreatment samples from patients with extensive disease was correlated directly with increasing number of metastatic sites. Of the five nucleosides, the mean number elevated was two for limited disease, 3-4 for extensive disease with one metastatic site, 4 for two or three, and 5 for four or more sites of metastases. Based on a summation of pretreatment nucleoside/creatinine ratios, a discriminant for survival was derived giving curves separating patients (P = 0.086) similar to the discriminant based on stage of disease. Although discordant results were noted, an overall correlation of 75% agreement with clinical assessment was estimated in response categories when monitoring changes associated with therapy.

A feasibility study in the development of biological markers for ovarian cancer.

As a primary feasibility study for the selection of biomarkers for more extended clinical evaluation, 17 potential biomarker candidates were measured in the body fluids of patients with ovarian carcinoma. For comparative purposes, patients were staged and separated into three groups: those considered completely free of disease, those with residual but minimal tumor, and those with advanced disease. Individual markers included plasma carcinoembryonic antigen, serum human chorionic gonadotrophin, urinary beta-aminoisobutyric acid, serum UDP-galactosyltransferase, and urinary hydroxyproline. Structurally related groups of biomarkers included six modified nucleosides and three polyamines analyzed in urine, and three serum protein-bound neutral carbohydrates. The respective chromatographic methods developed for these latter biochemical materials enabled all the individual compounds in each group to be quantitated simultaneously in one analytical run. The general frequency and degree of elevation for the total number of biomarkers was directly proportional to increasing tumor burden with specific exceptions, human chorionic gonadotropin and beta-aminoisobutyric acid. Galactosyltransferase was the most frequently elevated in the limited disease categories. Several of the biomarkers were elevated in the majority of patients with advanced disease and appeared potentially superior to carcinoembryonic antigen or human chorionic gonadotrophin.

Biological markers as an aid in the clinical management of patients with liver metastases.

Liver metastases due to the more common neoplastic diseases such as colorectal, breast, or bronchogenic carcinoma are a frequent occurrence and are associated with an ominous prognosis. Earlier detection followed by appropriate therapeutic interventions might have a decided effect on the subsequent course of disease. Controversy exists over the selection of tests with the greatest sensitivity, specificity, and potential utility. Preliminary evidence suggest that gamma-glutamyl transpeptidase and 5'-nucleotidase may be of particular significance. Four enzymes--gamma-glutamyl transpeptidase, 5'-nucleotidase, leucine aminopeptidase, and alkaline phosphatase plus carcinoembryonic antigen--were compared in the same blood samples from selected patients with breast and small cell carcinoma of the lung. Gamma-Glutamyl transpeptidase was the most sensitive test with 28/29 (97%) patients with hepatic metastases having elevated enzymatic activity in their sera. For patients with small cell carcinoma of the lung followed serially, gamma-glutamyl transpeptidase activity was increased an average of 5 months before liver metastases were detected by clinical means. Two factors are important in the interpretation of the results of gamma-glutamyl transpeptidase analysis: (1) Hepatic dysfunction due to diseases other than metastatic tumor involvement can cause a rise in enzyme levels as can (2) medications or ethanol which activate the hepatic microsomal drug metabolizing system. Of particular importance, however, is the fact that antitumor chemotherapy, even intensive and multiple agent, did not appear to effect the enzyme activity in the sera of patients with breast or small cell carcinoma of the lung. Gamma-glutamyl transpeptidase in combination with carcinoembryonic antigen may be of particular value in detecting liver metastases and in assessing subsequent response to therapy.

Multiple biological markers and breast carcinoma: a preliminary study in the detection of recurrent disease after primary therapy.

The use of a selected group of biological markers in breast cancer patients who are disease free but at risk of relapse after mastectomy has potential for detecting recurrent tumor before there is clinical evidence. In this preliminary study, multiple materials were serially analyzed in the body fluids of patients without overt tumor but receiving adjuvant chemotherapy because of positive axillary nodes at surgery. The total frequency of elevated levels was determined and compared for those patients who remained disease free, for those who subsequently relapsed, and for a third group of patients with proven metastases. Frequency of elevation was directly proportional to increasing disease. Although differences in the relative frequency of individual materials was observed, the same trends with increasing tumor burden were found. The results suggest that the serial measurement of biological markers has potential for indicating the presence of occult disease. A nucleus of biological markers to be considered should include carcinoembryonic antigen, urinary hydroxyproline/creatinine ratio for bone lesions, and gamma-glutamyltranspeptidase for liver involvement.

Regionalization of cancer control efforts in the u.s.a.

The Johns Hopkins Oncology Center, one of the first of the nation's twenty-one Comprehensive Cancer Centers, established a cancer control program to mobilize resources to improve patient care and prevent the occurrence of neoplastic diseases. Initial efforts were based on the results of a series of investigations conducted in four communities. Findings from these studies were then converted into a series of programmatic interventions. The strategy included integrated approaches to creating a community education program within the overall effort.