A prostate-specific antigen, distinct from acid phosphatase, was identified by immunologic procedures in prostate tissues (normal, benign hypertrophic, and cancerous) and seminal plasma, as well as in sera of patients with prostatic cancer and of nude mice bearing human prostatic tumor. This antigen was shown by immunoperoxidase staining to be confined to epithelial cells comprising the prostatic ductal elements. Prostate antigen was purified from prostatic tissue and seminal plasma, and it was shown to have a molecular weight of 33,000-34,000 with no subunit component. The isoelectric point of purified antigen was around 6.9, though several unpurified isomers with different isoelectric points also were observed. Serum-borne prostate antigen showed a molecular weight of 90,000-100,000 but it exhibited a molecular weight of 36,000 in the presence of sodium dodecyl sulfate. A sandwich-type, peroxidase-linked immunosorbent assay capable of detecting 0.1 ng of the antigen per milliliter of blood was developed. With this technique, serum level of the antigen was found to increase in patients with prostatic cancer as compared with normal males. The prostate-specific antigen can be a useful marker for detection of prostatic cancer.

Three monoclonal antibodies (MAbs) (DF3, F36/22, CU18) were used to monitor expression of distinct epitopes present within a family of mucin-like, breast carcinoma-associated molecules. Primary tumor specimens from more than 190 stage II breast cancer patients were evaluated for expression of the high molecular weight antigens. With a median follow-up of 6 years, patients whose tumors exhibited high immunoperoxidase staining scores (greater than 50% positive cells) with MAb DF3 had a superior disease-free survival ([DFS] 56% +/- 6% v 37% +/- 5% at 6 years; P = .0088) and overall survival ([OS] 72% +/- 5% v 59% +/- 5% at 6 years; P = .025). Staining scores with the other two antibodies did not correlate with improved prognosis. For MAbs DF3 and CU18, patients whose tumors exhibited predominantly apical cellular reactivity patterns had improved DFS, although differences reached conventional levels of statistical significance only with MAb CU18. In multivariate analyses, the prognostic value of MAb DF3 staining was independent of other identified prognostic factors. Furthermore, the concordance between primary and axillary lymph node metastases staining with each MAb was 73%, 80%, and 85% for MAbs DF3, F36/22, and CU18, respectively. These results suggest that staining with MAb DF3 identifies a group of node-positive women with a relatively favorable prognosis. Expression of the DF3 mucin-like glycoprotein is related to better differentiation, and staining with MAb DF3 provides an accurate and objective estimate of clinical outcome independent of histopathologic evaluation.

Monoclonal antibodies (MAbs) to human immunodeficiency virus type 1 were produced. Two antibodies reacted with the 17-kilodalton core protein (p17) of the virus and with its polyprotein precursor. To various degrees, each MAb neutralized infection by the cell-free virus. With a series of sequential overlapping hexapeptides which represent the p17 gene product, the epitopes identified by the MAbs were defined. The epitopes localize to overlapping regions near the amino terminus of the protein. Soluble synthetic peptides which span the antibody-binding sites of interest were demonstrated to competitively inhibit the reactivity of p17 MAbs, thus confirming the location of virus-neutralizing sites within the core protein.

Two distinct monoclonal antibodies (mAbs) were effective in the therapy of breast carcinomas of human origin established and growing in nude mice. Passive administration of either of the antibodies produced very rapid (less than 1 week) and significant reduction of in vivo tumor volume. Each of the mAbs showed in vivo targeting of the tumors. Histological analysis of mAb-treated tumors revealed extensive cellular necrosis. Each of the antibodies in vitro was effective in complement-mediated cytolysis at a concentration <1 ng/ml. The tumoricidal responses show that this is a useful model for passive human immunotherapy using mAbs.The treatment of malignant tumors with heteroantisera has been investigated for many years with interesting but inconclusive results (1). Hybridoma technology has provided monoclonal antibody (mAb) reagents, which may circumvent difficulties associated with the use of heteroantisera (2). Tumor-associated mAbs have been developed in the last few years by many workers, and some have investigated therapeutic efficacy in both animal and human subjects (3-5). Investigators have reported on the ability of mAbs to inhibit tumor growth by the administration of a mAb concurrent with or within several days of the implantation of tumor cells (6,7). However, passive mAb therapy has not been reported to decrease the volume of well established progressively growing solid tumors.Recently, mAbs developed in this laboratory against breast cancer cells have been shown to bind the target cells with high specificity (8). Passive immunotherapy experiments were performed to assess the therapeutic potential of these reagents. MATERIALS AND METHODSAnimals. Four-to six-week-old female Swiss nude (nu/nu) athymic mice were obtained from Sprague-Dawley. The mice were maintained in a laminar-flow apparatus under pathogenlimited conditions. Cells. Human breast carcinomas (BT-20, MCF-7, and MDA-MB-157) were obtained from the Breast Cancer Task Force, National Cancer Institute (9-11). Other human nonmammary tumor cells used were obtained from staff at Roswell Park Memorial Institute. MOLT 4 (T-cell leukemia), (B-cell leukemia), Peer (T-cell leukemia), and U-937 (erythroleukemia) were obtained from J. Minowada; K-562 (monocytoid leukemia), from J. Pauly; Chago (lung carcinoma), from B. Schepart; Daudi (Burkitt's lymphoma) and Palarmo (melanoma), from S. Leong and J. Horoszewicz; and AsPC-1 (pancreas carcinoma), from M. Tan. CCRF/SB (B-cell leukemia) and CCRF/ CEM (T-cell leukemia) were obtained from the American Type Culture Collection.All cell lines were maintained in vitro in RPMI 1640 medium/10% fetal calf serum supplemented with glutamine/pyruvate/nonessential amino acids/insulin. Cultures were maintained at 370C in humidified 5% C02/95% air.Tumors. Cells were scraped, washed, and resuspended in RPMI 1640 medium. Tumor-cell viability was assessed by trypan blue dye exclusion, and only cell suspensions of >95% viability were used. The nude mice were injected with 5-10 x 106 cells subcutaneously on their dor...

No abstract

The extent of human T-cell leukemia/lymphoma virus type II (HTLV-II) infection and its rate of spread have been difficult to determine owing to the serological cross-reactivity between HTLV-I and HTLV-II. The present study overcame this problem by directly detecting type-specific proviral sequences by means of the polymerase chain reaction (PCR) and liquid hybridization. Screening was performed on a cohort of primarily white intravenous drug abusers (IVDAs), and individuals of other behaviorally defined risk groups from the New York City area. Eleven percent (19 of 169) of the individuals in these high-risk groups were determined by PCR to have HTLV-II proviral infections. One of these patients displayed an exfoliative erythrodermatitis. Thirteen of the 19 subjects were positive in an HTLV-II enzyme-linked immunosorbent assay (ELISA). The remaining six individuals, although negative in the HTLV- II ELISA, were confirmed as HTLV-II positive by analyzing their DNA with a second HTLV-II-specific primer detector system. Four additional individuals were reactive in the HTLV-II ELISA but were PCR-negative for HTLV-II. PCR analysis for HTLV-I revealed that all four were positive for that virus. Thirty-seven percent (seven of 19) of the HTLV- II PCR-positive subjects were also PCR-positive for HTLV-I, and 84% (16 of 19) of the HTLV-II positive individuals were infected with human immunodeficiency virus (HIV-1). Six individuals were triply infected with HTLV-I, HTLV-II, and HIV-1.

No abstract

The gene for the cell surface receptor for HTLV-I, the etiologic agent of adult T-cell leukemia and HTLV-I-associated myelopathy, has been localized to distal human chromosome 17q. A panel of somatic cell hybrids containing fragments of human 17q as the only human genetic component was mapped with a set of 10 chromosome 17 probes and utilized to regionally localize the gene. When compared to the murine fibroblast fusion partner, L-M(TK-), and a hybrid cell line containing human chromosome 20, human 17q-containing hybrid cells bound high levels of both HTLV-I virions and the monoclonal antibody, Mab 34-23, which may be directed against the putative HTLV-I receptor. Additional experiments revealed that the human 17q-containing hybrids could also be more efficiently infected by cell-free HTLV-I virions than could the control cell lines. Western blot analyses of cell lysates showed that recombinant HTLV-I envelope gp46 protein and Mab 34-23 both bound to proteins of approximate MW 30 and 31 kDa which were found only in the hybrid cell lines which contained human chromosome 17q. The data suggest that the gene for the HTLV-I receptor is located on the distal region of human chromosome 17q demarcated by the tk-1 locus (17q23.2-17q25.3).