National Cancer Institute (NCI)

Cancer is not rare; it is, in fact, the second leading cause of death in the United States. For 1999, there will be about 1,221,800 new cases of cancer (other than non-melanoma skin cancer), with about 563,100 deaths. These projected numbers for 1999 represent a decrease from those projected for 1998 reflecting the sustained decline in overall cancer incidence as well as in cancer death rates as reported last year. Cancer, however, is not one but many distinct diseases. Certain cancers including breast, prostate, lung and colon can no longer be classified as rare diseases because prevalence data indicate that these cancers have exceeded the 200,000 cases per year maximum for inclusion as a rare disease. While incidence and mortality rates for most cancers have dropped, rates of malignant melanoma (now one of the most common cancers in young adults), non-Hodgkin's lymphoma, and kidney and renal pelvis cancers have risen. Furthermore, the previous reported decline in uterine cancer rates has leveled off.

The NCI's mission is to develop the means to decrease the incidence, morbidity, and mortality of cancer. It does this through the conduct and support of research in cancer biology, cause, prevention, control, detection, diagnosis, treatment, rehabilitation, and continuing care. The NCI's section of this report discusses selected major research advances, research initiatives within the NCI intramural and extramural programs and other program activities in these areas.

Recent Scientific Advances in Rare Diseases Research: Cancer Biology and Etiology

Basic research studies exploring the mysteries of how cancer develops form the foundation of cancer research. Through these studies, scientists are identifying, at the molecular level, the fundamental processes that underlie a cell's transformation from normal to malignant. The implications of this research are profound: Identifying the processes and pathways that lead to cancer provides attractive targets for new prevention and treatment approaches.

Over the last 10 years, cancer researchers have discovered that the signaling pathways involved in cell division and cell death play an important role in the development of tumors. Signals are relayed via cellular pathways from outside the cell to the nucleus, deep in the cell's interior. These signals regulate events that ultimately control the decision to divide and usually involve oncogene/suppressor gene products. In malignant cells, the regulation of these events becomes disrupted due to mutations in these oncogenes or tumor suppressor genes. In this past year, NCI intramural researchers have identified a form of transforming growth factor-beta (TGF-b1) as a novel form of tumor suppresser gene showing true haploid insufficiency.

TGF-bs are potent inhibitors of epithelial cell growth, and they also regulate many other cellular and organismal functions that are relevant to tumorigenesis. Recent work by NCI Intramural scientists has shown that the TGF-b receptors or downstream signaling components are mutated or deleted in certain human tumors. This suggests that the TGF-b system may have important tumor suppressor functions. They used a mouse model in which one of the two copies of the TGF-b1 gene has been experimentally deleted in order to address the importance of TGF-b1 dosage in protection against tumorigenesis. Although these mice are superficially normal, they are more susceptible to tumorigenesis in the liver and lung when challenged with chemical carcinogens than are normal mice. This suggests that TGF-b1 has important tumor suppressor functions in these organs. Interestingly, the remaining copy of the TGF-b1 gene is still active in the tumors. This distinguishes TGF-b1 from the classical tumor suppressor genes such as the Retinoblastoma (Rb) gene, for which both copies of the gene must be deleted in order for a tumor to develop, and therefore identifies TGF-b1 as a novel form of tumor suppressor gene that shows true haploid insufficiency. That is to say, for TGF-b1 to be effective as a tumor suppressor, the TGF-b1 protein levels must exceed the dosage achieved by only one copy of the gene. This has important implications for human tumorigenesis, as current strategies for the identification of tumor suppressor genes presuppose loss of both copies of the gene in the tumor. The power of this type of animal study that was performed is that a single genetic alteration can be introduced into otherwise genetically identical animals and thus relatively weak susceptibility phenotypes can be identified. Since most of the tumor suppressor genes that give highly penetrant phenotypes have already been identified through genetic analysis of cancer prone pedigrees, the next phase of tumor suppressor research will require them to focus on milder predisposing phenotypes. Animal studies of this type will be critical for the identification of candidate human tumor suppressor genes that show relatively low penetrance, but which could have a major impact on cancer incidence if highly prevalent in the human population.

Mutations in a newly discovered tumor suppressor gene, PTEN (or MMAC1), have been linked to a variety of human cancers including glioblastoma, the most common and malignant brain tumor. Frequently, PTEN appears to be lost in advanced stage cancers, suggesting that its deletion may accelerate cellular functions necessary for tumor progression. The PTEN gene product functions as a phosphatase, an enzyme that removes phosphate groups from other molecules. Unlike other tumor suppressors which remove phosphates from proteins, PTEN removes phosphates from phosphatidylinositol-3,4,5-triphosphate or PIP3, a lipid that resides within cell membranes. This discovery has generated excitement because PIP3 is a key component of major cellular growth control pathways, acting both to stimulate cell growth and to block cell death. In addition, PIP3 acts as an internal messenger for certain cell-growth stimulators like insulin and epidermal growth factor. When PIP3's phosphate groups are removed by PTEN, cell proliferation is checked and cell death can proceed normally. Conversely, when PTEN is missing or mutated so that it can no longer remove phosphates from lipids, cells continue to divide. Consistent with this finding, glioma cells, whose malignant phenotype is linked to PTEN inactivation, have elevated levels of PIP3. Introduction of the normal PTEN gene, but not the mutated PTEN gene, into these cells inhibits continued cell growth. Although it is not yet clear if PTEN normally suppresses tumors by allowing cell death or by checking cell growth, it does provide an attractive new target for the development of new therapies.

Chromosomal aberrations are detectable in all kidney cancers, both hereditary and sporadic (nonhereditary) forms. However the significance of these aberrations to the diseases state in most cases is not known. There are three known types of hereditary kidney cancer: von Hippel Lindau (VHL), Hereditary Papillary Renal Carcinoma (HPRC) and Hereditary Renal Carcinoma (HRC). Researchers have recently determined that the Met proto-oncogene, located on chromosome 7, is the HPRC gene. Defects in the Met gene have been found in the germline of HPRC families and these mutations appear to account for most of the cases of inherited papillary renal carcinoma. The Met gene codes for a cell surface receptor for a systemically circulating growth factor, hepatocyte growth factor (HGF). The germline mutations identified in the HPRC kindred's are located in the tyrosine kinase domain of the MET gene and are predicted to activate this receptor. This work is the first to identify a germline mutation of this gene and confirms its importance as a cancer gene. Associations between germline mutation of the Met gene and the appearance of other cancers, such as colon cancer, melanoma and stomach cancer are under investigation. The identification of germline Met mutations makes possible pre-symptomatic genetic testing for at-risk individuals in HPRC families and paves the way for additional studies to learn the pathology of HPRC and for the design of effective new therapies targeted to the specific defects brought about by mutation of the Met gene of kidney cancer.

NCI-funded researchers recently demonstrated the exceptional value of employing Nature's conservation of function in disparate organisms , flies and mice, to provide fundamental knowledge applicable to human cancer. A group of researchers sought to identify genes that control the proliferation of cells in mutant fruit flies. Theyisolated lats (large tumor suppressor) a novel gene that physically interacts with several proteins that are key to regulation of the cell cycle. Flies that have a mutant version of lats develop tumors in many tissues. The protein product of the fly lats gene is very similar in sequence to that of the human LATS gene, suggesting an analogous function in flies and mammals. Indeed, mice missing the lats gene develop tumors, particularly soft-tissue sarcomas and ovarian stromal tumors, and are hypersensitive to carcinogens. They also are defective in mammary gland development, infertile, and growth-retarded. Although the LATS gene is very highly expressed in human ovaries, it now remains to confirm whether it is important in the etiology of human ovarian stromal tumors, and whether it has a role in breast cancer.

Melanoma is a deadly cancer whose incidence has increased in the last decade to greater extent than that of all other cancers. Its propensity to metastasize and resist radiotherapy are major impediments in its treatment by radiation (e.g., x-rays) or anti-neoplastic drugs, and underscores the need to understand the basic mechanisms that determine the radiation and drug resistance phenotypes. Investigators have recently identified in human melanoma cells a UV-response genetic element and some of its regulatory proteins, including which proteins confer radiation resistance. Their results strongly suggest that ATF2 a transcription factor, is a key component of a heteropolymeric protein complex that modulates both radiation and drug resistance in melanoma cells. Thus, ATF2 modulators may be useful as potential radiation sensitizers in the treatment of this cancer.
Viruses have been shown to play a significant role in the causation of human cancer. Increasing evidence implicates specific viruses as etiologic agents for specific cancers. Chronic infections with the hepatitis C virus (HCV) appear to be responsible for a recently noted increase in the incidence of liver cancer in humans. This increase is particularly noteworthy in black Americans. A new development in this research area is the identification of an infectious clone of the hepatitis C virus. This clone will be important in elucidating the cancer causing activities of the hepatitis C virus and should aid in the development of vaccines.

Kaposi's sarcoma herpes virus (KSHV/HHV-8), the etiologic agent of Kaposi's sarcoma (KS) was discovered in 1994. The entire DNA sequence of the 160 KB viral genome was rapidly determined, and the sequence was quickly disseminated to the scientific community to assist other KSHV investigators. From DNA sequencing efforts, a number of KSHV specific viral proteins were identified that are homologs of cellular proteins involved in the regulation of the cell cycle and the immune system. It is hypothesized that these KSHV proteins may contribute to the initiation and/or progression of KSHV-associated neoplasms. The viral proteins may directly inhibit cellular responses to viral infection such as the shutdown of cell cycling and induction of cell death, thereby overcoming host intracellular defenses and contributing to cellular immortalization or transformation. Future research will focus on investigating the role of KSHV in AIDS-associated malignancies, the molecular mechanisms of viral oncogenesis, and preclinical research focused on the design and evaluation of novel therapeutics for these AIDS-associated cancers.

Two lines of investigation into the biology of multiple myeloma have converged to produce a new strategy for improving therapy in the disease. Multiple myeloma can usually be treated to extend patient lifespan significantly, but cures have been infrequent. Hematopoietic stem-cell transplantation, following high-dose chemotherapy, can cure the disease in some cases, but this treatment is far from optimal. The stem cells can be derived either from bone marrow or from peripheral blood. When the stem cells come from another person, severe graft-versus-host disease can occur. When the stem cells are collected from the patient, however, there is a danger that they will be contaminated with tumor cells, leading to rapid relapse. Nonspecific methods for purifying stem cells have failed to ensure that tumor cells are not present. Recently, NCI supported investigators found that multiple myeloma cells express both a gene called DF3/MUC1 and receptors for human adenoviruses, a rare combination. As a result, adenoviral vectors can be used to selectively target genes to multiple myeloma cells by placing the genes under the control of the DF3/MUC1 promoter. In a test of the feasibility of this approach, a mixture containing human bone marrow cells and a known number of multiple myeloma cells was treated with an adenoviral vector containing the thymidine kinase gene under the control of the DF3/MUC1 promoter. The cells were then treated with gancyclovir to kill all cells expressing thymidine kinase. This reduced the number of myeloma cells by a factor of at least one million, without affecting the number of hematopoietic progenitor cells in the bone marrow. This procedure can now be developed for clinical application.

Recent Scientific Advances in Rare Diseases Research: Genetics, Detection and Diagnosis

Promotion of research to improve cancer genetic screening and early cancer detection and to develop more accurate diagnostic techniques is of major importance to the Institute. The importance of these areas of research is demonstrated by their presence as three of the four mentioned opportunities in cancer research in the NCI's Bypass Budget for 2000 and the issuance of special initiatives (see Rare Diseases Research Initiatives). NCI-supported research, conducted at multiple centers throughout the country and by intramural scientists, is leading to rapid advances in these areas.
Each cell has a signature--a unique, identifiable characteristic related to its role in the body. As a normal cell is transformed into a cancerous cell, the signature changes, and that change becomes a unique signal of the cell's presence and character. By reading such signals accurately, we will be better able to detect and diagnose individual cancers. As part of the Cancer Genome Anatomy Project (CGAP) the regulation of gene expression in normal and malignant human lymphocytes was used to test this signature concept. We know from previous work that diffuse large cell lymphoma is a malignancy of mature B lymphocytes that accounts for roughly 40 percent of cases of non-Hodgkin's lymphoma. The BCL-6 gene is translocated in approximately 32 percent of diffuse large cell lymphomas and 70-80 percent have mutations in a presumptive 5' regulatory region of the gene. BCL-6 is also frequently rearranged in other non-Hodgkin's lymphoma sub-types and in AIDS-associated diffuse lymphomas. The coding region of BCL-6 remains unmutated in the lymphomas suggesting that dysregulation of gene expression underlies the lymphomagenesis. To address this question, the cDNA microarrays technology has been used to simultaneously quantitate the expression of thousands of genes. In brief, this technique relies on robotic spotting of cDNAs for defined genes in an ordered microscopic array on a glass slide. Fluorescently labeled cDNA probes are then prepared from total cellular mRNA derived from the cell of interest and hybridized at high concentration to this microarray. The extent of hybridization of the probes to each cDNA on the microarray is then quantitated using a modified confocal microscope. Two different cell types can be directly compared with each other on the same microarray by labeling the cDNA from each cell with a different fluorochrome. Genomic-scale knowledge of gene expression in a malignant lymphoid cell should allow one to determine which normal lymphocyte subset gives rise to a particular subtype of lymphoid malignancy, which signaling pathways are altered in the malignant cells, and the molecular basis for responsiveness or resistance of lymphomas and leukemias to therapy.

Initial experiments with an array containing genes which are selectively expressed in lymphocytes, termed the "Lymphochip," have focused on a panel of cell lines and lymph node biopsies derived from patients with diffuse large cell lymphoma, a common and aggressive subtype of non-Hodgkin's lymphoma. This malignancy was chosen for study since it most likely encompasses a variety of molecularly distinct diseases that cannot be distinguished morphologically. Preliminary results show that subsets can be defined based on correlative gene expression. Lymphochip cDNA microarrays are also being used to assess gene expression changes in chronic lymphocytic leukemia cells during treatment of patients with fludarabine chemotherapy. Current efforts are aimed at identifying gene expression differences in CLL patients who are sensitive or resistant to a variety of currently available therapies.

In addition to the above molecular genetic approach, classical surveillance and tracking approaches are continuing. Li-Fraumeni syndrome (LFS) is a dominantly inherited disorder characterized by early-onset breast cancer, sarcomas, and other cancers in children and young adults. Among members of LFS families, there is a high frequency of multiple primary cancers. In a recent study, 200 LFS family members originally diagnosed with cancer between 1968 and 1986 were prospectively followed through October 1995. Fifteen percent of these individuals developed a total of 42 additional tumors, a number of which appeared to be radiation-induced, suggesting a gene-environment interaction. The cumulative probability of the occurrence of a second cancer was 57 percent at 30 years after diagnosis of a first cancer. Survivors of childhood cancer were found to be at greatest risk for developing a second cancer. These findings suggest that cancer survivors in LFS families should be closely monitored for early manifestation of new cancers, and that chemoprevention regimens be considered.

Familial eosinophilia is an autosomal dominant disorder with idiopathic hypereosinophilia with or without organ involvement. A genome-wide search of one large family provided evidence for linkage on chromosome 5q31-q33. This region contains the cytokine gene cluster, including genes for interleukin 3, interleukin 5, and granulocyte/macrophage colony-stimulating factor, whose products play important roles in the development and proliferation of eosinophils. However, no mutations or polymorphisms were found in these cytokine genes, suggesting that the primary defect is caused by another gene locus in the area.

In the first systematic follow-up study of 4,500 women under age 30 who had been exposed to diethylstilbestrol (DES) in utero, three cases of clear cell adenocarcinoma occurred over the time period from 1978 to 1992. While this represents a rate that is 40 times greater than that of the general population, no increased risk for any other type of cancer was found. Although the rate for other cancers over this period was the same as that of women who had not been exposed to DES, continued follow-up of these women is warranted, since they have not yet entered the age group where breast and other hormonally-related cancers are more common.

The Children's Cancer Group and the NCI Pediatric Oncology Branch were surveyed to define the spectrum of malignancies in HIV-infected children and to determine differences in cancer types between adults and children with HIV infection. Of 65 tumors reported in children, 42 were non-Hodgkins lymphoma and 11 were leiomyosarcomas. Unlike the children, no increase in the incidence of leiomyosarcoma has been observed in adults with HIV infection.

A NCI-supported extramural investigator recently reported that circulating tumor associated antigen (TA-90) immune complex can serve as a predictor for subclinical metastatic disease and survival for patients with early stage melanoma. Archival sera were analyzed in an antigen specific enzyme linked immunosorbent assay (ELISA) for TA-90 immune complex in a blinded fashion, and the results were correlated with the patient's clinical course as determined by database and chart review. The TA-90 assay predicted subclinical metastatsis in 43 of 56 patients (P< 0.0001) with 14 false-positive and 13-negative results. Sensitivity and specificity for the detection of occult metastasis were 77 percent and 76 percent respectively. Fifteen of 18 tumor positive regional lymph node basins (83 percent) and 34 of 46 recurrences (74 percent) were accurately predicted when considered independently (P<0.004). Preoperative TA-90 status was also highly correlated with survival: 5 year overall and disease free survival rates were 63 percent and 46 percent respectively, for the TA-90 positive group, compared with 88 percent and 82 percent respectively for the TA90 negative group (P = 0.0001). A multivariate analysis with standard prognostic variables identified preoperative TA-90 status as a strong, independent prognostic factor for both overall and disease free survival.

Recent Scientific Advances in Rare Diseases Research: Cancer Prevention and Control

Monitoring emerging trends in the national cancer burden, and the factors that influence these measures, is extremely important in the effort to reduce the burden of cancer. To identify what cancer surveillance research is most needed and how best to advance knowledge of cancer based on the opportunities available, the Director of the NCI established the Surveillance Implementation Group (SIG). The SIG identified 12 cancer surveillance research opportunities, which were organized within five overarching priority areas. These research opportunities range from expansion of data collection to development of tools for analyzing surveillance data to establishment of linkages among cancer and other health-related databases. These recommendations provided the foundation for development of the Cancer Surveillance Research Implementation Plan, released in March 1999.

A recent analysis of brain cancer trends found that incidence has leveled off for all ages groups except those age 85 and over. During the last decade, use of computed tomography (CT) scans was relatively stable for those 65-74 years of age, but increased among those aged 85 years or more. Detailed evaluations suggest that improvements in diagnosis and changes in the diagnosis and treatment of elderly patients provide likely explanations of the observed patterns in brain cancer trends.

With the use of incidence and mortality data for thyroid cancer in selected areas of the country, as well as recent information on the levels of I-131 exposure by county across the U.S. from the Nevada nuclear bomb tests, risks were not found to increase with cumulative dose received at ages 1 to 15 years. Associations were suggested for persons exposed when under one year of age and for those in the 1950-59 birth cohort. However, the number of cases is small, and no causal relation could be inferred from these ecological data.

A new study using dietary histories and other interview data from 436 pancreatic cancer patients and 2,003 controls showed that obesity was associated with a 50 to 60 percent statistically significant increase in risk. Caloric intake also showed a statistically significant association with increased risk. Although relative risk did not differ between white and black subjects, the greater prevalence of obesity among blacks may explain their higher incidence of pancreatic cancer.

Recent results have provided some explanation for the contradictory results of the human population studies of the chemopreventive activity of b-carotene. While there are persuasive epidemiological data that dietary consumption of carotenoids, particularly b-carotene, reduces cancer at many sites, three large intervention studies of b-carotene indicated that high dose supplemental b-carotene either had no beneficial effects or increased the risk of lung cancer in smoking populations. To address this issue, investigators hypothesized that the free radical-rich atmosphere in the lungs of cigarette smokers alters the metabolism of b-carotene and its metabolites, and that these metabolites are responsible for the observed adverse effects of b-carotene in the clinical intervention trials. In recent studies using ferrets exposed to smoke and high doses of b-carotene, investigators have identified some of the hypothesized metabolites and observed the development of cell proliferation and squamous metaplasia in the ferrets' lungs. These metabolites interfere with retinoid signal transduction and thus accelerate malignant transformation. Additional support for the hypothesis comes from the observation that some of these recently identified b-carotene oxidative metabolites can facilitate the binding of the metabolites of benzo[a]pyrene (one of the most important cigarette smoke carcinogens) to DNA.

A population-based case-control study of adenocarcinoma of the esophagus showed a strong and probably causal relation between gastroesophageal reflux and esophageal adenocarcinoma. The more frequent, more severe and longer-lasting the symptoms of reflux, the greater the risk. The risks of esophageal squamous cell carcinoma and adenocarcinoma of the gastric cardia were weakly or not associated with reflux. Another large population-based case-control study of esophageal cancer has provided findings that add to the growing evidence that the risk of cancers of the esophagus is reduced in users of non-steroidal anti-inflammatory drugs (NSAIDs). After controlling for all major risk factors, current users of aspirin were found to be at decreased risk of esophageal adenocarcinoma, esophageal squamous cell carcinoma, and non-cardia gastric adenocarcinoma, but not of gastric cardia adenocarcinoma, when compared to never users. Risk was similarly reduced among current users of nonaspirin NSAIDs.

Recent Scientific Advances in Rare Diseases Research: Cancer Treatment

Fiscal Year 1998 was a banner year for important advances in cancer treatment by intramural NCI scientists and by extramural scientists supported by the NCI. Discovery and development efforts have lead to the production of a new generation of immunotoxins directed at various solid tumors (breast, ovary, prostate, lung, colon, mesothelioma) that will be available for clinical trials within the next year. During the next several years, the results from clinical studies using new generation recombinant single-chain immunotoxins should also help define the role of these novel agents in cancer therapy.

Drug Development

The NCI Drug Developmental Program continues screening new synthetic and natural compounds for anti-tumor activity using the automated cancer cell line screen. Approximately 72,000 defined chemical structures have been evaluated since the screen became operational in April, 1990. More than 7,100 compounds have demonstrated in vitro antitumor activity of which 3,900 agents were selected for in vivo evaluation for assessment of therapeutic activity. It is obvious that there are more compounds to test/develop than current resources would allow. Thus the staff of the Developmental Therapeutics Program had formed a working network of investigators to prioritize compounds for development. Including vaccines and other biologicals as well as chemotherapeutic agents, a total of 32 agents are in Decision Network level 2A (small animal testing), 3 agents are in Decision Network level 2B Large animal/primate testing) and 25 are in Decision Network level 3 (ready for human testing subject to obtaining IND). Table I provides a listing of the agents in the Decision Network process. As the agents move through the different levels of the decision process, the level of financial commitment by the NCI increases.

To further expedite the movement of academic discoveries from the laboratory to proof of principle clinical trials, the NCI initiated a program named Rapid Access to Intervention Development (RAID) in 1998. The RAID program makes available to the academic research community on a competitive basis any or all of the following activities: (1) the steps in preclinical development that are necessary to convert a new molecule into a drug candidate suitable for clinical testing and that are generally not otherwise available to academic investigators who lack a corporate partner. These steps include GMP synthesis, formulation, range-finding and IND directed toxicology and pharmacology; (2) planning clinical trials; (3) regulatory affairs, so that the requirements of the Food and Drug Administration may be satisfied by any investigator who seeks to put a new molecule into the clinic; and (4) filing of the IND and direct study sponsorship by the NCI, where indicated. As of February 1999, sixty seven applications have been received, twenty nine applications had been reviewed, twelve applications were successful and received NCI support and thirty eight applications are currently awaiting review. A description of the successful applicants and the projects can be found at http://dtp.nci.nih.gov/docs/raidwin1.html.

There are alternatives to using solely NCI resources for the development of agents. Two formal mechanisms are in place at NIH for the development of specific compounds and agents with industrial partners. One involves the use of a Cooperative Research and Development Agreement (CRADA) and the other involves the use of Clinical Trials Agreement (CTA) or Material Transfer Agreements (MTA). As of April 6,1999, 21 CRADAs and 45 CTAs are active and 5 CRADAs are being negotiated (CRADA-LOI) (see Table II).

Another mechanism for drug development is the use of cooperative agreements to support drug discovery activities through the National Cooperative Drug Discovery Group Program. This Program supported the development of Topotecan (camptothecin analog), which received marketing approval during 1996 for the treatment of ovarian cancer and was later extended to breast cancer. Gliadel (BCNU in a biodegradable polymer), also approved in 1996, is the first new form of therapy for brain tumors to be approved by the FDA in 25 years. DAB389IL-2 received FDA approved recently in 1999 for the treatment of refractory cutaneous T-cell lymphomas. Eight other agents are in clinical trials; N1,N11-bis(ethyl)norspermine; anti-EGF receptor antibody - m225 which was recently humanized; O6benzylguanine; a pair of anti-transferrin receptor monoclonal antibodies (A27.15/E2.3), cordycepin plus plus deoxycoformycin, a cryptophycin analog, and a farnesyl transferase inhibitor.

Two monoclonal antibodies were approved by the FDA for the treatment of cancer. C2B8 (Rituxan) has been approved for the treatment of B cell non-Hodgkin's lymphoma. C2B8 works by attaching to a specific antigen present on the surface of a cancerous B cell. Once the monoclonal antibody attaches, it triggers the body's immune system, causing a response that leads to destruction of the cancerous cell. The other monoclonal antibody is Herceptin (Her2/Neu). Currently Herceptin has been approved for treatment of women with breast cancer. NCI is launching new trials with these antibodies respectively in chronic myelocytic leukemia, other lymphomas and in ovarian and peritoneal cancers.

Clinical Treatment Research

Targeted Therapies

Many cancer patients who have endured chemotherapy know that cytotoxic drugs are effective against some tumors, but can also damage the patient's organs and weaken the immune system by killing normal dividing cells. They also know that an additional complication to chemotherapy is that tumor cells can become resistant to the drugs. Recently, researchers have developed a new strategy that circumvents the problems of chemotherapy-induced toxicity and drug resistance. This strategy is named In Vivo Phase Display (IPD). Investigators have used IPD to isolate novel peptides that bind specifically to endothelial cells undergoing tumor-induced angiogenesis, and have linked these peptides to cytotoxic drugs, thereby specifically targeting the drugs to angiogenic vessels, and sparing other tissues. Recent experiments with mice indicate that it may be possible to develop targeted chemotherapies that are based on recognition of specific receptors in tumor angiogenic vessels. NCI-supported investigators have also shown that the vascular beds of a large number of normal tissues differ from one another, and that specific peptides are capable of homing selectively to the vasculature of the lung, the skin and the pancreas. These results suggest that novel peptides could be developed to carry drugs to specific tissues to treat metastatic foci. These targeting therapies have the potential to markedly improve cancer treatment.

Despite recent advances using conventional approaches for cancer treatment (surgery, chemotherapy, and radiotherapy), most adult solid tumors, once metastasized, are incurable. It is clear that new clinical approaches are urgently needed. NCI intramural scientists have developed a new therapeutic agent for cancer treatment by using genetic engineering to modify a powerful bacterial toxin, Pseudomonas exotoxin A (PE), so that instead of killing normal cells it selectively kills cancer cells. PE is a three domain protein composed of 613 amino acids. Anticancer agents are produced by deleting its binding domain (aa1-252) and replacing it with the Fv fragment of an antibody or with a growth factor that binds to cancer cells. These agents are termed recombinant toxins. Recombinant toxins have been made that target the Lewis Y antigen which is present on colon, breast, lung, and other epithelial cancers (B3(Fv)-PE38), the erbB2 oncogene product present on the breast, stomach, and lung cancers (e23(Fv) PE38), the normal EGF receptor overexpressed on glioblastomas and squamous cell carcinomas (TGFalpha-PE38), the mutant EGF receptors present on glioblastomas (MR1(Fv)PE38), the IL4 receptors on glioblastomas (IL4-PE38KDEL), the IL2 receptor present on many leukemias and lymphomas (anti-Tac(Fv)-PE38) and the CD22 also present on many lymphomas (BL22). When administered to mice with human cancer xenografts, these recombinant toxins produce complete tumor regressions.

The same scientists are currently conducting clinical trials in patients with solid epithelial tumors using immunotoxins LMB-7(B3Fv-PE38), LMB-9(B3(dsFvPE38), and erb38. They are also conducting a clinical trial with LMB-2 (anti-Tac (Fv)PE38) in patients with leukemias and lymphomas. Radiolabeled MAb B3 (90Y) is being used to treat solid tumors. In collaboration with the John Wayne Cancer Center they are conducting a trial in patients with glioblastoma. Over 150 patients have now been entered into clinical trials involving immunotoxins developed in the Intramural Program at NCI.

Tumor infiltrating lymphocytes (TIL) obtained from patients with melanoma have been used to clone the genes encoding the antigens recognized by these TIL. TIL have been identified that can recognize unique cancer antigens on murine and human cancers including melanoma, breast cancer, colon cancer and lymphoma. In clinical trials of TIL administration, 36 percent of patients with metastatic melanoma underwent objective cancer remission. Several of the genes that encode antigens recognized by tumor-infiltrating lymphocytes (TIL's) have been identified, including MART-1, gp100, tyrosinase, p15, TRP-1, TRP-2, ESO-1 and b-catenin. Concentrating efforts on the MART-1 and gp100 melanoma antigens because these are expressed in patients who are HLA-A2 positive (a restriction element present in approximately 50 percent of humans), immunodominant peptides were identified which bind to HLA-A2 molecules with low to intermediate affinity. Since binding affinity is believed to be important to maximizing the immune response, modified peptide from MART-1 and gp100 were tested. One from MART-1, referred to as g209-2M, showed a 10-fold higher binding affinity for HLA-A2 than the native peptide. Clinical trials using the parent and modified peptides have been conducted in HLA-A2-positive patients with metastatic melanoma. Prior to immunization, lymphocytes from patients were exposed, in culture, to the modified peptide for 10-12 days. Two of 8 patients who received the native peptide were successfully immunized, though the degree of immunization was very weak. Of the 11 patients who received the modified peptide, 10 were successfully immunized. Despite the high levels of circulating precursors in the ten patients, they did not show an objective response (disappearance of old lesions and no appearance of new ones) unless they were treated in combination with IL-2 (42 percent response rate with IL-2). A randomized Phase III trial to determine whether the 42 percent response observed is due to the IL-2 alone or to the combination of immunization with the peptide and IL-2 is planned.

Significant progress in the treatment of children with high-risk neuroblastoma has been made. In a landmark clinical trial, researchers found that using autologoous bone marrow transplantation versus intensive conventional chemotherapy improved the children's three-year, event-free survival from 18 percent to 34 percent. The researchers also found that 47 percent of the children who received 13-cis-retinoic acid, a drug that arrests the growth of neuroblastoma cells, versus 25 percent of the children who did not receive the drug were event-free after three years. 13-cis-retinoic acid is a differentiating agent, or a drug which forces cancer cells to behave more like normally programmed cells; these important results mark the first time a differentiating agent has been shown to improve outcome for children with solid tumors.

Data obtained by the National Wilm's Tumor Group indicated that six months of therapy with pulse-intensive drug administration (single dose) instead of divided-dose courses given daily should be the new standard therapy for children with Wilm's tumor. Pulse-intensive drug administration was recommended based on demonstrated efficacy, greater administered dose-intensity, less severe hematologic toxicity, and the requirement for fewer physician and hospital encounters.

Immunotherapies

Multiple myeloma remains a largely incurable disease with current therapy. Allogeneic BMT provides an opportunity to add the potential antitumor effect of marrow grafts to those of high dose chemoradiotherapy. To enhance a graft vs. tumor effect without aggravating graft vs. host NCI scientists are attempting to selectively target an immune response against a defined tumor-specific antigen. Using a rearranged immunoglobulin gene product of a myeloma tumor as the unique tumor-specific antigen, researchers are currently testing the hypothesis that tumor antigen-specific immunity can be adoptively transferred to BMT recipients from immunized donors. Five donor-recipient HLA matched sibling pairs have been enrolled in the study and are undergoing immunologic analysis for transfer of tumor-specific immunity.

The idiotype of the immunoglobulin on a given B cell malignancy (Id) can serve as a clonal marker. A previous pilot study in lymphoma patients demonstrated that autologous Id protein can be formulated into an immunogenic, tumor specific antigen by conjugation to a carrier protein (KLH) administered with an emulsion-based adjuvant. The immunological adjuvant used is GM-CSF because of its ability to enhance the potency of the prototype Id-KLH vaccine by augmenting almost exclusively the cellular arm of the immune response. The objective of current studies is: 1) to evaluate cellular and humoral immune responses against the unique idiotype of the patient's lymphoma and 2) to evaluate the ability of the Id vaccine to clear the bone marrow of malignant cells detectable by pathologic examination or molecular examination (polymerase chain reaction amplification of the rearranged bcl-2 oncogene). Twenty patients who achieved complete remission by standard clinical criteria after ProMACE chemotherapy participated in the vaccine study. In 17 patients, lysis of lymphoma cells was evident after vaccination. Eight of 11 patients, whose tumors were not detectable by "standard clinical criteria" but were detectable by the more sensitive PCR analysis converted to PCR negative, providing the first systematic evidence for an anti-tumor effect of Id vaccination.

The NCI intramural program continues to make advances toward the development of vaccines to prevent infection by cervical cancer causing HPVs. A virus-like particle (VLP) based vaccine composed of only the major coat protein of the virus made in insect cells is the lead candidate. It lacks the potentially oncogenic viral genome yet induces high levels of antibodies in experimental animals that prevent infection of cultured cells. A clinical trial of a VLP-based vaccine for HPV16, the type found in almost 50 percent of cervical cancers, was begun in 1998 in collaboration with NIAID and Johns Hopkins University. The preliminary results of this trial have been very encouraging. No substantial side effects have been noted and all women inoculated with a moderate dose of VLPs generated high levels of serum antibodies to the virus. A larger phase II study to further substantiate safety and immunogenicity is planned for 1999. Assuming that positive results are obtained, the NCI hopes to begin an efficacy trial of the HPV16 VLP vaccine. Discussions are curently ongoing with investigators and Government officials in Costa Rica on the possibility of conducting this efficacy trial in that country, in parallel with a large ongoing DCEG, NCI study of HPV and its association with cancer.

In an attempt to increase the effectiveness of a VLP-based vaccine, scientists in the intramural program have generated VLPs that contain other proteins of the virus not naturally found in the virions. In mouse tumor models, these Achimeric@ VLPs were effective in inducing potent anti-tumor cell mediated immune responses directed against the additional viral protein. They also generate high levels of virion inactivating antibodies. Clinical grade chimeric VLPs for HPV16 will be generated in 1999 and human vaccine trials will be initiated in the later part of 1999 and 2000.

Rare Diseases Research Initiatives

The NCI uses Program Announcements and Requests for Applications to announce special initiatives. These initiatives ranged from soliciting for specialized networks and centers, encouraging research using molecular approaches in tumor/biomarker classification and identification, to encouraging and supporting clinicians and minorities in clinical research. The following descriptions are from a selected few initiatives.

Requests for Applications issued included (1) Director's Challenge: Toward a Molecular Classification of Tumors, (2) The Early Detection Research Network: Biomarkers Validation Laboratories, (3) The Early Detection Research Network: Clinical and Epidemiology Centers, and (4) In Vivo Cellular and Molecular Imaging Centers. These four initiatives are particularly relevant to rare diseases in that they support the development of techniques and centers for detection of rare tumors.

Many Program Announcements to stimulate translational research were issued. Some of the relevant ones include: Correlative Studies Using Specimens from Multi-Institutional Treatment Trials, Clinical Cancer Therapy Research, Models for HIV Disease and AIDS-Related Malignancies, Mechanisms Underlying Individual Variations In Drug Response, and Cancer Pharmacology and Treatment In Older Patients.

Rare Diseases Program Activities

Meetings

NCI jointly sponsored with the Office of Rare Diseases, NIH, the following scientific meetings since 1994:
Phacomatoses Revisited including Neurofibromatosis 1 and 2; Tuberours Sclerosis 1 and 2 (TSC1, 2); Von Hippel-Lindau Syndrome (VHL); Gorlin Syndrome (PTCh); Cowden's Disease (PTEN); Familial Adenomatous Polyposis (APC), Workshop On Adult Soft-Tissue Sarcomas, Conference On AIDS-Related Kaposi's Sarcoma, Workshop on Celiac Disease, Head And Neck Cancer Symposium, Workshop on Lymphohematopoietic Malignancies, Symposium on Burkitt's Lymphoma, Workshop on Ovarian Cancer Biology, and the 17th International Natural Killer Cell Workshop.

IND Submissions

The NCI currently maintains around 170 active INDs for both cytotoxic and biologic anti-cancer and anti-AIDS agents. In 1998, an Investigational New Drug Application (IND) was submitted to the Center for Drug Evaluation and Research, Food and Drug Administration (FDA), for each of the following compounds:

Agent

1. Gadolinium Texaphyrin
2. Immunotoxin LMB-9
3. MGI 114
4. Recombinant Vaccinia-Prostate Specific Antigen Vaccine
5. Plasmid DNA Vector (VCL-4951) Expressing Mutated gp100 Gene, w/ or w/o IL-2
6. E1B-Attenuated Adenovirus (ONYX-015)
7. Sarcosinamide chloroethylnitrosourea (SarCNU)
8. Recombinant Humanized Anti-Vascular Endothelial Growth Factor Monoclonal Antibody (rhuMAb VEGF)
9. Oxaliplatin (L-OHP)
10. Recombinant Human Tumor Necrosis Factor-Alpha (rhTNF alpha) (Boehringer Ingelheim)
11. Recombinant Vaccinia Vaccine Encoding the Human MUC1 Gene (rV-MUC1)
12. Immunotoxin RFB4(dsFv)PE-38 (BL22)
13. 4'-Iodo-4'deoxodoxorubicin
14. Tumor Specific Mutated VHL Peptides
15. NG-Monomethyl-L-Arginine

Investigational new anticancer agents in early clinical trials are listed in Table III. Many of these compounds came through the Decision Network.

Shown below is a representative list of those cancers with estimated prevalence rates below 200,000:

Acute lymphocytic leukemia
Acute myelocytic leukemia
Bone
Brain
Esophagus
Hodgkin's disease
Kaposi's sarcoma
Kidney and other urinary tract
Liver
Larynx
Lung, small-cell
Lung, non-small-cell
Melanoma
Neuroblastoma
Non-Hodgkin's lymphoma
Ovary
Pancreas
Prostate
Rectum
Stomach
Testis
Thyroid and parathyroid